Oil Pressure Instability

I posted this vid in my thread but thought the topic itself and follow on discussion deserving of its own thread. I have brought up what I saw through the factory gauge as oil pressure instability on track previously. Over the winter I put a data logger on the car to allow what I was seeing to be quantified and better evaluated. Now, as NickL points out, maybe it’s an issue with scavenge pumps or just more broadly, a mechanical issue specific to my car. Or, maybe it’s an inherent often discussed issue that many have when running the engines on track and is inherent to the design of the oil system. In any event, I think it’s good to put data into a context that’s easy to digest and understand and I hope the video does a good job of that. Driven XP 9 LN 2QT Deep Sump & Baffle LN Spin on Filter and Napa 1042 Both Oil Pressure and Temp taken from ports on an oil filer adapter using AIM sensors Data logged at 20HZ Oil P and 10HZ Oil T Oil is filled to 1/4 up the dip stick, about 10QTS Water temp is not shown but was 200F and steady Oil Pressure is the bar graph on the right, it is shown as TEMP - But is configured to show pressure. Oil Temperature is the bar graph on the left I thought may be interesting to the group. Porsche 996 Oil Pressure Instability on Track You Tube

Thanks for starting this thread and sharing the data!

So after watching the video im not sure whare the problem is. Looked like pressure was responding to throttle input and the temp stayed under 240*. Was there a partucular moment in a curve that i missed?

Multiple pressure drops to 15PSI above 3.5K RPM. And that’s with cool oil, under 250. I would much prefer to not see the pressure drops every time G force is introduced.

I'm not an expert here but from watching the video it seemed like right hand turns exhibited bigger drops in pressure, especially at the initial turn it when pulling more g forces but there were also some dips at left hand turns. Makes me wonder, if having the oil pick up a bit lower to ensure that it is more often submerged in oil during hard cornering be of any help...

Quotes source post: 17439952

Can't really go any lower with stock pickup and sump cover, basically touching the bottom of sump.

Cross posted from the main thread... The display is awesome. Great data. Having this video along with graphs from the data logger is quite powerful. It seems there are some pressure drops on left hand turns as well, just not as dramatic. What is an expected amount of drop? Let's say, straight line, WOT to no throttle for 2 seconds, then back to WOT, how much pressure drop would there be? But this is not really the issue. The issue is pressure drop on a right hand turn under power. Maybe a different parameter is needed. Something like Pressure/Load/G or Pressure/RPM/G Really nice work.

Quotes source posts: 17439952, 17440073, 17440079

Agree with George here. I would like to only see pressure drops in line with RPM.

To me, it seems the only way to lose pressure is for the oil pump pickup to become in contact with air. So there is only a couple causes, not enough oil in the system overall, or, oil is pooling somewhere and not returning to the sump fast enough. Likely the second issue would be my thoughts - which match the comments about scavenge pumps and such. Also, any drainback limitations that exist, thicker oil will just make it worse. -Tom

Watching your video literally ties my stomach in a knot !!! Very nice data, and great display (needs more G-force definition), shows the issue very well. But watching it is a love/hate thing..

Quotes source post: 17440660

I know... it�s like, you know it�s happening, and you�re only catching glimpses of it here and there, and the cars running strong, filters great so how bad could it be. Then, I look at the video and it�s like... ****. Im still working out some details in how I want to format the videos, the inner circle was set to 1.5G but something looks wonky in how it came out because it�s showing around there all the time, which I�m not.

At least your pressure is really good up top. Above 5k rpm you have good pressure up to 80psi. The instability is between 4-5K, with the major dips occurring at below 4K. It would be really telling to see the delta between RPM and Pressure (where it counts) as related to G-Force. (where the problem lies). I credit your driving style to surviving this instability !!! Someone who downshifts too early, goes into a turn too hot at too an high RPM/G-force, cornering at a lower gear ,will not be so luckly. Your smooth driving style and taking care of the equipment is paying off. If I were a Race Car Owner, you would be the kind of driver I wanted..

Quotes source post: 17440751

Skip... I have that data. Let me know and I can send you some screen shots of whatever combo you want to see. Or, If you or anyone else wants to download the race studio 3 software I can send out the full logs and you can view whatever you want. https://www.aim-sportline.com/en/sw-fw-download.htm Its really pretty intuitive to use and fun to play with. I ran lawn mowers and plow trucks for my youth Skip... taking care of the equipment is engrained in me.

Quotes source post: 17440318

I used to have this problem with oil pressure drops in my younger years. I learned how to fix it - drive slow in the corners.

OK i got the race studio downloaded, now i need that file to look at..

Quotes source post: 17441401

You got em brother...sent to your email.

I'm going to put another video up tomorrow from this same track day. The difference being it will be from the first session where the oil is considerably cooler and the affect that has on the level of pressure drop. The video in the OP is from the last session of the day when oil temps, although cool - were at their warmest for this particular day. Should do a good job showing the affect oil temperature has on pressure.

Here is the first lap, 2nd lap, and a mid-lap from your run at Palmer. The pressure change from oil temp is quite dramatic. The colors go together. The second lap at 200degrees is already starting to show the pressure drops. The first lap is not driven as hard as the second lap, see G graph below, but the overall pressure shows much higher regardless of G's. Both left hand and right hand turns show pressure drops,(Positive G is a right hand turn, Negative G is a left hand turn) but the left handers seem to bottom out at 40psi where the right handers drop to 20psi. Pressure will drop with RPM drop, which all the turns have. If we assume the drop to 40psi is due to RPM decrease, we can say the extra 20 drop is the right hand turn problem. There are some weird bobbles in the pressure that occur during acceleration that seem out of place. Could these be air bubbles working their way through? There's also a weird bobbly mess at the bottom of each pressure drop. I also added some calcs to see if they could bring out any severity that would explain the difference in left and right hand turns. But neither of these worked. The RPM in the denominator overwhelms the pressure change. It effectively mimics the oil pressure graph. And G's cross zero, so dividing by this gets stupid really fast. I added a constant to get away from zero, but it's still useless. It's the first I'm thinking about this kind of thing so I'm sure I'll have more useless ideas tomorrow. Forward comments and suggestions. Data is awesome!

Quotes source post: 17441668

Jason and Golock, Nice work. Golock, your charts are a work of art adding clarity to what we all have been thinking about. Looking forward to the thoughts from the engineers in the group. ​

Cool Paul - super happy you're enjoying chugging through the data, I have to say - pretty efficiently as well ! So much good stuff both for driver improvement and car vitals. And you can see why keeping the oil cool is high on my target list - it does not resolve the instability but it does limit how dramatic the drops are. Now - a piece of data you may want to include when you're reviewing, water temp.

Try running race oil next time and see if you have pressure drops. By the way, the engine is slowly dying here; only a matter of time before the engine spins a bearing and throws a rod.

Check the post you quoted... XP9

Ah, I missed that. I guess aeration is not the issue then.

The main oil pump have a by-pass valve, the scavenge pumps does not. When your main oil pump goes in to by-passing, the flow is constant but the scavenge pumps keep increasing their pumping with revs. Below by-pass pressure the balance between scavenge pumps and main pump is constant. Above by-pass pressure the balance goes in favour of the scavenge pumps. So, when the oil is cool most of the time your engine is running in by-pass and as such the balance is often in favour of the scavenge pumps. The hotter the oil gets the less they are in favour and your start seeing pooling and pressure drops.

I believe the reason for "right hander" problem is most likely because there is asymmetric wear on the pumps. When I took apart mine the wear was considerably worse on the driver side. On the driver side you have more chains and tensioner paddles that shed material which will damage the pump.

Quotes source post: 17441668

This data set is super awesome. The oil pump can push 7-10 PSI of air when it starts sucking air. Also, the sensors are not super accurate at low pressures; most sensors are rated at +/- 5% of full scale. I am rather concerned about the lows of 20-ish PSI at 4000+ RPM, that's going to be a real problem. The "rule of thumb" is 10 PSI per 1K RPM and 20 PSI - or possible as low as 5 PSI due to sensor accuracy - is far below that. The weird bobbly mess at the bottom of the pressure drop is probably the oil pump cavitating. I think an accusump set to 40 PSI will keep the engine living quite a bit longer. The time spent below 40 PSI is very small compared to the rest of the time, which would let the accusump recharge.

The reason for the large difference on the right handers compared to the left handers is the sump is not the same left to right. There is a large void on the bank 1 side (left front) that will fill with oil on right hand turns. This lowers the level at the pick-up moreso and will suck air. Bit is not the sole reason for instability ( since it does so on left handers also) I would be willing to donate to a "go fund me new scavenge pumps" if Z would install them just to eliminate that as a possibility. They are really not that hard to install and with this great data acquisition would be instantly reveling..

Quotes source post: 17442229

That's probably why the pick-up is offset to the left. Here is a picture I produced (calculations are volumetric) while looking in to the pressure drops. It's a printscreen from another thread: Add some chain tensioners to that fundraiser also, they are also likely contributing to the issue if leaking(as is any oil feed IMSB). I will let you guys continue this thread without me from here on.

Quotes source post: 17442229

You set up the "go fund me" account and I will help to fund it. Jason's work is extremely valuable to all of us.

Only chart before work: If we take the rule of 10PSI per K RPM, we can calculate and overlay the min pressure. This will clearly show where the rule is violated. Blue below red is bad. This is a mid lap with hot oil This is 1st lap with 'cool' oil The long down trend in the pressure is due to the oil temp rising as seen on previous graphs. The low pressure at the beginning is idle. This is a good reference for an absolute minimum pressure when no cavitation is occurring and RPM is significantly lower than when running RPM. The only problem with this value is the oil is not hot. The actual value may be a little lower. I briefly looked at engine momentum. What is this? It seems to be associated with MAF. I was looking for engine load, it seems that this is the closest. Some additional reference data that would be nice to have is: -pressure at idle with hot engine -pressure vs RPM straight line with smooth ramp up and down, say 300RPM/sec from 2K to 6K. This would show the pumping system is smooth and stable -pressure vs RPM straight line with smooth ramp up and release accelerator and catch at 2K. This would show any problems due to fast mechanical changes in the engine Water temp graphs next. OK, gotta go work so I can afford to play.

I seriously doubt the scavenge pumps have anything to do with this. Regardless the part# for the scavenge pump is superseded to 997-107-021-05 which shows as $203 each (2 req'd) from Delaware Porsche. Also needed are two O-rings 999-707-554-40 at $6 each. If we're throwing parts at it, it might be worthwhile to install the updated oil pressure relief valve plunger (997-107-125-01) and spring (996-107-127-53) with a new O-ring (996-107-123-50) for a total of about $20. I would also install the larger Cayman oil cooler as it's rather obvious that oil temperature is driving this bus. This is going to be a lot harder to do; access is terrible; either the transmission or the entire power unit likely needs to be removed. I think the Bank 1 timing chain is dragging oil out of the sump, or contributing to aerating the oil during right-handers. The lower side of the Bank 1 chain moves sump->head because the engine turns clockwise looking at the crank pulley. Combined with the extra space over there for the oil to pool up and lowered viscosity at higher temps making the oil easier to aerate it's a perfect storm. There is a little Y-shaped thing in the cylinder head to partially block off the chain passage. One wonders how effective it is.

"Throwing parts at it" is a terrible way to put it, but we all know that is a part of "testing". But needs to be done systematically, one thing at a time, which will be all on Z if he wants to. Would be a burden on him and may take the fun out of tracking, most would rather turn a blind eye and hope for the best, or do a shotgun approach, or just a new engine.. I have a Go-Fund-me set up, but won't release it untill I get a green light from Z..

Wow guys - pumped up with the response, thanks for all the contributions from smart ****ing guys ! Awesome. And, I�m happy to test parts and share the data. My data is for the group, will continue to make everything I collect available to anyone who would like to see it, doesn�t matter if the interest is for car vitals or driving data. Paul, I�m sure I have a log somewhere of hot idle and can run whatever scenarios people would like to see, remind me and if I don�t get you one of the hot idle tonight, I�ll run one for you over the weekend. Im back at Palmer on SAT running same config as is reflected in the data above, with the exception being that the weather should be high 70�s low 80� which is much warmer than the ambient when the data above was collected. Should be interesting. Thanks for all the support guys !

Quotes source post: 17442473

Yes, perhaps I was a bit too forceful with that statement. Although for a total of less than $500 a set of new scavenge pumps & the updated oil pressure relief parts are relatively inexpensive, as far as Porsche parts go. It's also important to remember that none of this information is available from the dash gauges. Without the datalogger that Z installed none of this would be available and it would continue to be all guesswork. Of course there's no oil temperature gauge and the oil pressure gauge appears to be smoothed over many seconds. The less than 1 second dips in oil pressure will never show on the dash gauges.

Quotes source post: 17442310

https://www.gofundme.com/f/new-scave...f+share-flow-1 To donate to the research for oil pressure instability... I'll also donate $25 for each UAOS bought in the next 14 days to the research. Golock911 I love what you can do with the data. much better than I can do, so ill just look at yours . theperf you are absolutely right, none of this "testing" would do much good without the great capture of the data as proof positive of the results.

Quotes source post: 17442524

Aaron... as you know, the pressure sensor is mounted at the filter instead of the head for the factory... I think this should be a better reflection of what we see where it counts... mains and rods. Aaron know this cause I wouldn’t have been able to get it installed without him machining parts to get everything in my system to play together. This is exactly what I wanted when I sold my beloved My02’s to fund the Aim... get clear data that could get people with a technical mind and an interest in problem solving talking and working on issues that we may have known about, but never really saw in a way that allowed an assessment of their criticality or if the supposed “improvements” are actually even beneficial. I really believe, we can make these engines more reliable and help people spend money on things that are PROVEN to work. I’m looking at you 2QT deep sump 🖐 ETA... Paul ( Golock911) is running the AIM SW better than I do too... I’m thinking about hiring him as a data guy for trackside support.

For what it�s worth, I just saw oil pressure drops on the street in a long sweeping right-hand on-ramp. I wasn�t even really pushing it, and I�m on all seasons! So there, this isn�t always just track-related. My engine temperature was probably right at 100C or so, and on the street, I cannot imaging oil temps would have been much past 220C. I really wasn�t driving very hard, and the RPMs were below 4,000 the whole time.

For those that want to see the difference cool oil makes in the context of video as opposed to the sick graphs golock911 put up, here's the first couple laps of the day - a huge difference. Sorry for the washed out view, had not corrected the GoPro settings before the first session. What is evident 1. Oil instability happens regardless of temperature. 2. The level and severity of the instability IS dependent on temperatures. When the oil is very cool, the pressure still drops, but not to the levels that I think I would call extremely concerning. Now, if the cooling system be it water or oil can be beefed up enough to be able to maintain the oil temps required to mitigate the severity of the drops, and do so when ambient temps are in the 80's is a definite question I have. If you couple the additional cooling with an Accusump it seems like it would be a reasonable path to making a big dent in mitigating the issue - it would entail a lot of money, and a lot of added system complexity. Water temp on the left, oil temp in the middle and oil pressure on the right. You Tube

Interesting. Oil pressure drop very evident during right hand turns as expected. Here is a million dollar question... Why are folks saying that an acusump doesn't work? Any idea?

Quotes source post: 17443509

Don't know Glen... I would not hesitate to try one though, unless someone could show me data that they don't work... I'm pretty much done listening to people who say things.... Here's what LN says about the 2QT deep sump... I like LN - but as I have shown, I would not call it a solution to oil starvation. It's possible that my situation is unique and I would not assume that everyone will have the same experience. I'm speaking for me, and my experience only. https://lnengineering.com/products/b...51-baffle.html

Any engineers have any evidence one way or the other on the plus and minuses of the acusump?

One thing that is not clear with the video is what your oil pressure would behave like in a bone stock 996 on the track without any of the enhancements you have done such as, deep sump, new driven oil that wasn't aeorated, etc. You never had non-existent oil pressure and while it did drop in the right handers, in many cases you were running under 3500 rpm. As soon as you got your revs up over 4000 it started to build. There may be a conclusion that is developing which is that you have to get the revs up to keep the oil pressure up in our engines. So those that take a long sweeping carousel at too low of rpm run a huge risk of starvation if oil pressure are too low. Which means don't baby the car on the track or it starves. Edit add: So if we improve the supply of oil and possibly replace old sump, lines internals, use the right oil that is fresh and replaced before each track event, keep the engine and oil cooler with a 3rd radiator, replace your aos and Ims with an upgrade, then possibly 90% of the battle has been won. If we want 100%, then it may be an acusump. Just thoughts out loud.

I've had an Accusump many years ago when they first came out. The issue with Accusump is recharging the tank. After the initial dump, they start robing oil pressure and flow volume to refill the tank. It becomes a rob Peter to pay Paul game . But they do work well for priming a Turbocharger before starts.

Alright, you guys are have already talked about what the graphs show. Similar to yesterdays graph, 3 laps, 1, 2, & 8? from the first set. Top group is water temp, Mid group is oil temp, bottom is oil pressure, colors match. The first lap is actually alot longer than 2 and 8. It doesn't show here as the X scale is distance, not time. The first lap is mostly idling with a slow temp rise. Water temp gets to 185 and slowly rises with each lap. 8th lap shows 205degrees. 20 degree rise over 6 laps for 3.6 degree increase per lap. Oil temps change faster but show the same incremental increase by lap. Overall, the car is heat soaking. Conclusions already listed above: Improve both water cooling capacity and oil cooling capacity. I totally agree with Jason comments 1 & 2 above. I would add that there are 2 types of drops; from RPM decrease and from sucking up air. It seems that cooler oil drops maybe be from RPM decreases and not sucking up air. This leads me to think that hotter oil is more prone to either aeration or viscosity loss (is moving much more easily away from the oil pickup)(oil pickup is not centered making right handers worse) or ??. A center radiator and bigger oil cooler will absolutely help. 10 degrees? It may be all that's needed. Here is a chart of the 1st lap referenced to time. Its the same as the red above. Most of the time in the lap is spent idling. But look at the oil pressure decrease! OMG! And the oil isn't even hot yet. In 7 minutes it drops from 58 to 40psi. Oil temp at the end is only 145. This is why I think we need some straight line oil pressure vs RPM data. We need to know what pressure should be with no air being pushed. A baseline for the pumping system. Here's why I worry about just trying to keep the revs high. This graph is a zoom in of the one giant pressure drop, by time in seconds. It has Lateral Force and RPM overlaid. Look at the purple circles. These must be air events in the pressure line. I'm not sure keeping revs up will eliminate these. The second one occurs during rev up. I guess it depends on the length of the drops. The main drop is much more concerning than small side bubbles. The resolution in the data is awesome. Previously mentioned, it would be cool to get a baseline of the pumping system. -Smooth acceleration up to 6K and smooth acceleration down. No fast changes in the mechanics. Looking for kinks in the pressure curve. -Smooth acceleration up to 6K and quick drop and catch at 2K and hold for a few seconds. Not sure what to expect. It will somehow test the momentum of the oil against the mechanics of the pump. Maybe useless. If the air being picked up is simply due to oil moving around in the case, we need to change where the oil pickup occurs. Better centered, center back of the case. Maybe some of the deep sump designs do this. I haven't looked at any of the designs. The big question I have is: Why does the cooler oil not allow the same type of apparent air bubbles? It must be the difference in viscosity between the 2 temperatures that changes how it moves in the sump. I think its unlikely to add enough cooling to keep the oil at 180, which means that we need to fix the oil pickup. Being the DAG at a race track would be cool. I worked with a guy that did this for a CART team many years ago. He was so stressed out when the races would get close. I'd send him home from work. He couldn't concentrate. And I need the frequent flyer miles... As always, don't be shy with comments and questions.

Quotes source post: 17443774

Love your analysis. You give a lot to think about. Flow and aeration seems to be key ingredients. Lower oil temp, better pickup and circulation will potentially facilitate less aeration, allowing higher rpms and better oil pressure. Is that right? If so, how do we accomplish it? If not, please set me straight.

No need to set you straight. The way I look at this is the whole problem is that there isn't a good supply of non-aerated oil for the pick up at extremes. Solve this problem and you don't need to worry about oil temp. Apparently this is difficult to do. I must believe that one of the deep sump kits has this figured out. Off the top I would say a deep sump with a pick up somewhat lower than exists today would go a long way to fixing it. Regardless, either it's a hard problem to solve or people aren't willing to show the data that shows their deep sumps work, so we don't believe them. Now we find in the data that we can get some benefit from cooler oil. So let's do that while designing or begging the racing teams that have figured out what a good sump/pickup would be to give us one. Cooler oil is good for engine longevity anyway. To design a sump and pickup, you would first need to understand how the oil is moving around in the case. Nick_L had published some thoughts and maybe conclusions on this a while back. Maybe there's some insight there. Ultimately, similar to Jason, I'd like to see some data to base the design on. This would mean we need to instrument a sump and watch the oil move around during time at the track. I'm not familiar with what types of instruments we might use in this environment. Aeration from chains would get in the way. Pooling or bad scavenging would get in the way. Somehow this would need to be accounted for in the design. OK, it's late, I'm blathering. Maybe we should get the designs for available sumps posted up for a good discussion. Maybe tomorrow.

Quotes source posts: 16670711, 17047339

You are a boss Paul - really enjoyed reading your analysis. Thanks for taking the time to share it. I will get you the data with the variables limited, I agree setting up some baseline info to use in isolating the expected RPM associated drops from the G influenced is needed. Looking at the various commercially available sump designs, the seem to really mimic each other to a large extent, I know CTS references that he tested various deep sumps and was unimpressed with the results, probably a good time to link to some of his stuff for reading by the folks really interested in this that may have not seen it. From my thread - with a data log of his dry sump system https://rennlist.com/forums/996-foru...l#post17047189 From NickL's thread https://rennlist.com/forums/996-foru...l#post16923646 Performance of his dry sump on a boxster. He covers the issues with both stock and the aftermarket sumps in here. Lots of good stuff. https://rennlist.com/forums/racing-a...mp-system.html A Description of his dry sump system And finally, some thoughts he relayed to me in a thread I started in the motorsports forum... not particularly encouraging.

Quotes source post: 17443717

Skip nailed it. Reduced oil pressure and volume during refilling is the biggest issue with the Accusump. Depending on the track, it might not even recover fully before a second event. The two other issues we've seen which include failure of their EPC valve (which you would have to check regularly) and second, if a manual valve is used instead, remembering to turn it on before going to the track.

Quotes source post: 17443545

Have you done used oil analysis on the oil coming out of the engine to see if you're seeing elevated bearing wear? How long are you running the fill of XP9? Since you are seeing higher temps, you might want to try running XP6 instead, but I would certainly in that case run a billet high flow filter like we recommend. Alternatively, the Napa 1258 looks like a suitable candidate for an oversized filter that won't sit lower than the sump (when the 2 quart deep sump is fitted):

Quotes source posts: 17444073, 17444073, 17444073

NO The oil in the logs had 50 miles street driving on the way to and was in the middle of its first track day. For anyone that curious my intervals have been every 2 track days (4HRS track time MAX) or 4WKS. ETA... sorry my last response cut short. Do you see anything in my data that looks to be an anomaly from what you would expect ? The temps were all generated during cool ambient temps and I expect as we get into the warmer months.... we’ll really see what the temp picture looks like. I also think my setup is pretty vanilla and standard... so if you see anything that looks unique to my data, that would be interesting. My view is if this is what I’m seeing on a very standard setup a lot of guys run, the other guys are experiencing some level of it to, they just don’t know it.

Quotes source post: 17443849

I think that solving the oil temperature problem first will assist in masking the oil pressure problem. This is what CTS did for the race Caymans, it's in one of the linked threads. The trouble with the deep sumps is that the pickup is still located in the middle of the pan. That is the best location for many reasons, it's the least affected by shifts in oil position. As I understand all the deep sump kits presently use an extended pickup. Jason and I discussed this and he has an extended pickup. I think that we can assume that the pickup is as close as reasonable to the bottom of the sump. What can get missed in these discussions is that the oil pump is above the oil level in the sump which means the oil pump has to "suck" the oil out of the sump. The tube from the pickup to the actual pump inlet has several turns and isn't very big, either. Neither of these are conducive to unrestricted oil flow. The reason is that the oil pump doesn't actually "suck" oil out of the sump. What it does is lower the pressure in the inlet tube below atmospheric pressure (is, creates a semi-vacuum). Since "nature abhors a vacuum" air pressure pushes the sump oil into the intake tube, and the slight difference in pressure/vacuum is all the we have to move oil from the sump to the pump inlet. The height of the oil column over the pickup is also important because a higher oil level adds the weight of the oil in the pan to the force pushing the oil into the intake. When the oil gets hot it gets thinner. This reduces the pump's ability to create a vacuum while making the oil flow easier into the pickup tube. When the lateral G's increase the oil sloshes to the side. There's more area on the left (bank 1) side of the engine so more oil can slosh over there which gives a lower oil level on the pickup tube for right hand turns. The oil pickup doesn't have to be totally unported for the oil pump to pump air. Once the forces shoving the oil into the intake of the pump can't meet the pump's demand it cavitates which super-aerates the oil. It's important to keep in mind how much flow we're looking at here. The gear type oil pump is a constant-displacement pump so it pumps (tries to pump) the same amount of oil each pump revolution. Typically the oil pump will flow something like 1 quart per second at redline RPM (I do not have the figures for the M96). That means the oil pickup needs to supply 1 quart per second or 15 gallons per minute to prevent pump cavitation. It's clear the pickup system can do this with the car straight and level and the oil temps relatively cool. It's equally clear the pickup cannot do this under lateral G loads and higher oil temps. This is also the reason why baffles in the sump don't do a whole lot as there is a fairly small limit to how much oil the baffles can contain. There was some talk of a flop tube oil pickup that used the lateral G loads to reposition the pickup in the sump to where the oil ought to be the deepest. This is not unprecedented as the Lycoming aerobatic airplane engines have this sort of system. Jason and I also talked about a system to forcefeed the M96 oil pump from an external dry sump tank with electric scavenge pumps. This might work and it sure would be complicated. It looks like a fitting could be added to the M96's oil pump casting on the intake side that would be externally accessible. This fitting would receive high flow, low pressure oil from an external tank via an electric pump. A check valve in the existing pickup line would prevent oil from backflowing into the sump. A set of electric scavenge pumps would take suction from center left and center right of the oil sump, flow through an air-oil heat exchanger, and into the external tank. Additional scavenge pumps could be added for the cylinder heads if needed Would this work? Perhaps. Would it fix the problem? Well, if it worked it would.

Awesome post Aaron. I think it is totally reasonable to attack the temp issue first, or at least think it to be the far easier of the 2 problems. 1. This is one you can throw parts at and work your way up in cost and how invasive or complexity adding the changes are.... start with stock turbo rads, then center rad, then maybe CSF turbo side rads, then there is the oil cooler which as you said, seems very likely to be a big pain in the ***. 2. By incrementally adding cooling capacity, if you can get to an acceptable point - you can reallocate any additional funds ear marked for the next attempts to something else...like tires 👍 3. The consequence for lack of progress is low... get a sump or pump modification right enough... then wrong at the wrong time could put you in a bad place. Put some new rads on hoping to lower oil temps and have it fail to accomplish your goal... you’re out the money and your time.

A buddy has an accusump he’s willing to let me borrow for testing. Im going to see what it will take to get it mounted before I take him up on it.

someone mentioned a swinging oil pump pickup and I contacted his mechanic who developed it but got no response. The whole issue is very disheartening as pretty much every "common" solution does very little. I was resigned to running the very expensive race oil with as much oil cooling as I could stand and hoping for the best. I've also elected to pursue racing a different platform (e36 bmw) for which this is less of an issue and if the engine does fail (which ours did), a fully beefed up rebuild is about $5k...not $20k.

Yeah... not a lot of impetus for any manufacturer to solve this problem unfortunately. my oil pressure is terrible Dude... you gotta buy a deep sump and use XP9 yeah.... I have a deep sump and XP9 Dude.... your probably running the level too high... or too low... whatever it is, you�re oil level probably ****ed up. yeah.... it�s halfway up the dip stick Dude... it�s probably fine, have you checked the pressure sender Yeah.... sender is accurate Dude... you need to run a deep sump and XP9 yeah... ok. I have a lot more faith in the few owners that actually give a **** about maybe addressing this, ya know, so it�s actually addressed than I do in the vendor base. Not enough margin Not enough customers Customers that are happy to just buy what already exists and tell themselves its good enough cause this wicked smart vendor told me so.

Quotes source post: 17444548

Frustrating for sure. The other interesting thing is that the stock gauge tends to run high. My buddy's stock gauge stayed generally above 4bar, except on very dramatic turns like the toe of boot at WGI or similar. Once he installed a more accurate gauge, pressure was close to 1.5 bar. I think this gives people false hope. For reference, I can come off the track after a 20 minute session and see pressure drops as i make low speed turns in the paddock. I run my oil fairly low, which doesn't help, but if I run it higher, then I get smoke bombs at start up. As noted, aside from good oil and some extra cooling (my oil never gets over 240 as is), I'm just gonna leave it as is and just prepare for a used motor every 3 years or so. easier to do with the boxster 3.2 for sure.

Well, I’d say this... anyone trying to gain a level of comfort in what the stock gauge is telling them should be aware of what the gauge is capable of relaying, and what you’re capable of receiving. I knew I had pressure drops from the stock gauge, but that’s it. It took going back through track videos to try to gain some perspective of the scope because I was incapable of receiving what the gauge was telling me on track. And even then, questions on accuracy left me wondering. Now I know, and I’m happy I can put it out there for others, it’s spurred a good conversation I think.

Quotes source post: 17444499

Someone out there has to have a background that knows how to do this... Just gotta get them to read the thread. ​​​Sure seems like a good business opportunity to me for someone to use their knowledge to build a swinging oil pump pick-up. I'll buy one. If it just gives me 10k more miles on the car before a rebuilt, it may just be worth it.

Glenn, the problem is the market believes the deep sump to be sufficient and anyone thinking of developing a competing product has to convince the market otherwise. This would entail the market to be sufficiently technical to understand the differences and open to trying something new, what if you�re not a big name like LN - you�re going to have an army of people tell you... the LN already fixes this problem. And all that is after the development cost of designing and testing has taken place, all that liability floating out there in the hopes you can convince the market that your solution is a superior method to address a problem, the market by and large does not think they have. I wish I was wrong, but some of what I�ve written is stuff that I have spoken about with people I would trust to come up with an actual solution... at the time I discussed with them, they had no interest in doing so and I don�t blame them for a minute. So.... it�s on the people who want to change the situation to change it.

Quotes source post: 17444695

I'm not convinced this solves the problem either. With my 2 quart deep sump, I have vertical and horizontal baffles. At some point, i can't see how the oil could be anywhere BUT the pick up area. I wish someone could make a clear acrylic sump plate and get a video camera focused on that for a couple laps to see whats happening.

wow! thanks for the links. Reading Chris's thread about his dry sump is pretty amazing. This is the graph we want! Oil pressure mimics RPM. It's perfect. His comments about the inevitability of the engines demise is disheartening. And unfortunately with my quick learning, I agree with him. Not to be defeated, yet, but not having the resources or expertise to reinvent a dry sump, the next discussion would be about why the deep sumps fail or all the baffling has failed. The issue turns into a time issue. The oil returning to the sump needs to be in a position to be pushed(!) into the pickup well before it's needed. So the design needs to be for collectors to direct the oil to a cylindrical baffle around the pickup. If this could be kept full at the rate conjectured above, the issue should go away. The idea is to decrease the time it takes to get the oil returning to the sump to the pickup and not let it lolligag it's way to happen across the pickup. I like the idea of being able to see into the sump as its running. Some thick polycarbonate (maybe not high enough temp) would allow some observation for how the oil is returning. Maybe its already obvious and I'm just a noob, only having seen the LN deep sump kit online. It's a time and flow control problem

Quotes source post: 17444776

well, id like to say that would solve the problem, but this is why a lot of the deep sump kits swap the 996 style swirl pot return tubes to the straight shot 997 returns. The issue is aeration. The swirl pots delay oil return but supposedly reduce aeration. The 997 returns are quicker, but don't de-aerate the oil. They return the oil outside of the stock baffle to supposedly give the oil time to dissipate the bubbles. there is a guy on youtube who makes clear covers for combustion chambers and such, so I'm sure he could do this, but would need a test car and a track. You Tube

This was from CTS, boxster with a deep sump... looks familiar. Again to highlight - if you are driving an M96 on track - is this happening to you ? It�s now got me curious to see what guys who only use their cars on the street should expect... Im going to log a spirited run on the road and see... Anyway... another data point as said above... boxster with unnamed deep sump.

The deep sump pans, by themselves, don�t always fix the issue because: 1.) Aeration is still present, no matter how much oil you have. Actually, more oil can lead to more aeration. Counter-intuitive, but it�s the windage effect. 2.) The baffles don�t help that much. Like Accu-sump, you could be robbing Peter to pay Paul. Having a baffle sounds good, until you realize that the compartment to hold that oil is much smaller than the overall amount, so it can be depleted faster. 3.) A pickup tube that can rotate the way some aircraft engines have rotating pickups are absolutely required to prevent starvation, or else a deep dry sump external tank that is so tall as to never lose oil around the pickup is required. Sorry, aerated oil that�s seeing a 0.9+g force is going to be held up on it�s side. Unless the pickup is at the centrifugal center of where the mass is, it will likely suck air. All of the above also requires something in place to de-aerate the oil in the first place. By the way, we don�t need high temp to have aerated oil. To test, just put a cup of oil in a blender and watch it turn white after 10 seconds.

Quotes source post: 17445037

1. Windage is a real thing, but with how the M96 is designed, I'm not seeing this to be an issue while the engine is running. Given how much oil is retained in the heads and how high the crank is vs. the oil pickup, I can't imagine that the crank is whipping the oil. 2. While it is depleted faster, when there are any G forces, oil should essentially always be flowing through the trap doors. Especially when you prevent oil from climbing the walls of the sump with a horizontal baffle. 3. It would certainly help, but i don't think this is a perfect solution. I would think if it was this easy, people would have built this. Engines like an S54 just use a 2 pickup solution with a fairly isolated deep well sump and don't appear to suffer from starvation.

Quotes source post: 17444769

Back in the day I was involved with a project on medium speed marine diesel engines to figure out where the oil was going. A "medium speed" marine diesel typically has a redline of about 600 RPM. The particular engine I was working on was a Pielstick 18V, 18 cylinder V type rated 9,000 HP @ 530 RPM. As you might think rebuilds are not cheap... we instrumented the crankcase with ultrasonic distance sensors made by Polaroid to measure the air gap above the oil level. This worked OK but not great as any oil mist in the crankcase would fool the sensors. I don't think this would work well in a high-speed application like a car engine as the crankcase is much smaller and I suspect the oil mist factor is much greater. Besides Polaroid doesn't make the sensors anymore as far as I know. The factory electronic oil level gauge uses resistors and temperature changes to measure the oil level, it's not particularly accurate and very slow responding. A TV camera would get covered in oil almost immediately. Perhaps an array of float switches would work.

Quotes source post: 17445059

I wonder how they organize a dual pickup to prevent one pickup from sucking air. Perhaps grafting an S54 oil pickup onto the M96 would solve some of the problems.

Quotes source post: 17444323

All this rings true. The oil to water heat exchanger is going to have a delta-T between the oil and the water. This should be constant unless the heat exchanger is getting overwhelmed. Essentially the oil should be "x" degrees hotter than the water under all conditions. I think the data shows the oil is 30 degrees hotter than the water at it's highest. Adding cooling capacity will help, if the water temp is high and especially if the water temp changes with ambient air temp. I seem to remember 220F water temperatures; I think those can be lowered with greater cooling capacity. An aside. For reference while at Lime Rock last week running the Turbo pretty hard (but not as hard as Z runs his car!) I had a highest water temp of 187, ambient air temp was about 75. This is with the oil cooler, automatic transmission, and engine adding heat to the water. The highest oil temp I saw was 227F and it stayed under 220 most of the time. The ATF hung out around 200F. (I have oil temp and ATF temp gauges). I suspect the Turbo engine has a higher capacity water pump. It certainly has a much bigger oil to water heat exchanger. The radiator cores are almost twice as thick. The airflow into the radiators is somewhat better with the larger openings. The exit slots in front of the wheels (which Z has added) probably help too. So there should be room to lower the water temp on the C2, which should have a knock-on effect on the oil temp - unless the oil to water heat exchanger is getting overwhelmed. Then the Cayman part could be the next step.

87 at Palmer tomorrow... Likely we'll see what the stock cooling capacity for both oil and water really look like, I would say for most of us, I have run in cool to downright cold ambients so far this year with the exception of the last couple sessions when we were at Palmer, even then - it was probably mid to low 70's. Other than that, and for all instances data has been posted, its been around 50 or under.

Well, it looks like the fundraiser for the oil instability research "go-fund-me scavenge pumps" is stalled at about $300. What we gonna do ? How are we gonna prove/disprove Nicks theory about scavenge pumps ? Anybody else interested in this issue? https://redirect.viglink.com/?format...2Bshare-flow-1

Same discussion about oil return lines in the sump vs aeration from some years ago. https://rennlist.com/forums/996-foru...wirl-pots.html -Discussed is why the returns aren't directed at the pickup. Consensus is that more time is needed to get rid of the air and foam. Foam and air will rise to the top as the oil travels across the pan to the pickup. Part of the baffles job is to block the foam from getting to the pickup. This is why the swirl pots in 996 and return tubes in 997 are pointed towards the outside wall of the sump. -There is one mention that LN deep sump with 997 return lines solved the pressure drops (only data is visual from the gauge) while others say it didn't help. Jake adds a comment that the current tolerances in the engine 'everywhere' is a big factor in how much pressure drop you will get -The 2 scavenge pumps per head is also discussed. This seems more rare than the CTS dry sump. Too bad I didn't read all this before all the talking. Sounding like the 3 year broken record... Apologies to those that have been through this before. The data in this discussion is way better. So, based on the discussions and data here's what I think the best setup is with available parts: -CSF or bigger radiators -Center radiator -Bigger oil heat exchanger -997 oil returns -LN deep sump kit If the pressure drops are still seen, take the engine apart and go find the 'loose' bearings... Otherwise, resurrect the CTS dry sump or spend some time to develop a clever de-airing system that would reduce the time for clear oil to the pickup. What size screen is needed to filter the air and foam out of the oil? (easy question to ask when you have no idea about the bubble sizes and flow rates and such)

In my opinion, the Go-Fund-Me account should continue for the very reason that we are all looking for oil starvation solutions to prolong engine life in our cars, as it relates to oil pressure drops. What we are looking to do is analyze in sequential form whether specific enhancements actually work. With that being said, we all know that eventually our engines will need to be rebuilt. It happens to every engine that has ever been built in the history of engines. We know it costs a lot of money to rebuild regardless of who or how you choose to do it. And we know why the engine fails and what causes it. (1) Ims, (2) bore scoring, (3) oil starvation. The purpose is to find affordable oil starvation solutions in a logical sequence that can be tested to see if they actually work. Remember, this is not an exercise to determine if our 996 engines are a better choice or not to a GT3 engine or a Chevy engine in creating a race car. CTS, and everyone else knows that there is always a better engine design for racing. None of us can argue with that. Our purpose is to find viable solutions to prolong engine life so we can enjoy our cars on the street and track until we decide it's time to rebuild. So with all this being said, my suggestion is that we come up with a logical sequence of parts/procedures that should be tested and funded. I suspect if we Frame this mission properly, we can increase the contributions. I would gladly do it with my car, but I have zero mechanical skills and rely on my shop to do these things and the costs will be astronomical. Jason was kind enough to agree to the pumps since he has the AIM system to analyze and measure whether the part or procedure works, and more importantly the mechanical skills to do it, and the communication skills to detail and share it. Maybe Jason would be open to doing more than pumps, or even changing the sequence of enhancements given the previous posts suggested in this thread, if it can be funded by Go-Fund-Me contributions. No harm if Jason doesn't want to do it. It's a big ask. Maybe it is someone else. Or maybe we just forget about it and regroup later. My thoughts and suggestions may be completely off base, but Skip's message created an opportunity to frame out a clear purpose. I am open to however everyone wants to proceed. Glenn

Can't contribute as much to this thread technically but I just donated! Love seeing the data shared and the collective brain power try to solve this. Z - I'm looking forward to seeing how this issue presents itself during a spirited back roads rip.

1.) Windage happens in pretty much all engines. It cannot be easily avoided in such a small package. That�s primarily how oil becomes aerated in the first place. As long as the pickup is pulling in oil, aeration is not as much of an issue. 2.) Should, I agree. But�there are countless posts on folks still having oil pressure drops with all these aftermarket pans installed. If it worked so well, bingo, problem solved, let�s move on. Yet the issue continues. Air is getting sucked in regardless. 3.) It�s not easy, and it�s not cheap. Thus, the reason Porsche decided to go away from dry sump to hybrid sump.

Zbomb, Is your oil pressure sensor plumbed in prior to the oil filter or after? I ask because your baseline oil pressure is quite high. Chris Cervelli Cervelli Technical Service

I'm interested in a showing of hands from those weighing in on this topic that have funded their own research and product development in the automotive aftermarket. Who here has been the person that wasn't just doing the work for someone else, but also paying the bills, and spending the time to do so? If you have not had these direct experiences, you have no idea what this entails, but if you try it, you will instantly feel the pain. Of those here the only one I know of who has done this is @Porschetech3 and I greatly respect his efforts, and successes in the arena. FYI the "swinging" oil pick up tube was done by Jim Pasha [RIP] around 2003 with these engines. He's the only person I know of that has rolled a Boxster in an AX. The swinging pick up works great on circle track applications, where the pool of oil is always anticipated to be, and you have durations of time for the pick up to follow the pool of oil with the turns being long and shallow. Testing at Roebling Road in 2008 at turns 4/5/6 and also "Madness" at Mid- Ohio proved that the changes in the pool of oil vary a ton as the oil temperature increases, and the column off oil shifts quicker from side to side. Jim also saw this, and we compared notes about this, since I had previously done this with our SCCA engines in 914s, trying to use a wet sump. Though I have solved these problems in many ways over the years, what everyone needs to focus on is keeping the oil cooler. If you can keep the oil cool, you can skimp on other things and the engine still lives. We did this in pro racing with these engines, and could make it a full season without a single bolted on component being fitted to the oil sump of the engine, while using factory scavenge pumps as well. The best investment you can make is a stand alone oil temperature gauge, and the things necessary to keep the OT at 230F, or less. I have seen stock engines on the track on a 90F day see an oil temperature of 280F, take your oil temp readings from the oil sump, using a temperature compensated gauge/ sender. The second thing to buy is a stand alone oil pressure gauge that does NOT take the oil pressure reading from the stock location. With the installation of this you will actually see the oil pressure readings that are the most critical. You may find that you were staring at the stock gauge when you really needed to be staring at the track, and paying attention to driving. When people call me with what I know is going to be an on- track failure, I ask them what track they were at. I will then commence to tell them that they probably had their failure in the 3rd session of the day, and at 15-20 minuets into that session, possibly on the cool down lap. I will then tell them where their car came to a rest on the track.. On the right side of the track, just past turn 17 at Sebring is an example. When I take the words out of their mouth like this, they can instantly tell that we've been dealing with these failures for more than 18 years, know what causes them, and know how to solve them. Only 2-3 times per year do we see an application where our dry sump system is a necessity on any M9X. Typically these guys who are DE instructors, and "hot lapping" the car session after session. Doing this without any breaks are the only times where I will offer this as an option. No one else has proven to need it- Since most of these instructors have moved to newer cars, they are experiencing a whole new set of complexities than they had with the M9X The accountants always win in modern business/ corporate world "things".

Quotes source post: 17446293

Oil filter adapter.

Just saw the Gofundme and contributed as I'll indirectly benefit by this. If we don't get there in the next few days I'll close the gap. @zbomb I'd be very curious about data from a spirited drive as that's what I primarily do although it's entirely possible that our canyon drives in Angeles forest are closer to track conditions than they should be.

Quotes source post: 17446303

Jake, thanks for the message. Don't think there is any interest in funding R&D for product development for new automotive parts. The goal here is to find out what works as a bolt on solution that is already manufactured by an automotive expert such as you, Charles and many other providers. I appreciate every time you respond to these topics with a firm recommendation on cooling. Obviously that is the key ingredient or you wouldn't mention it. So what "already produced" parts do you recommend for water and oil cooling? A bolt on third radiator kit, or something else? What about oil cooling? We keep hearing about it but what is available? Can't speak for the other guys, but I can say that the goal here is to prolong engine life on the track and street. Eventually I will be contacting you for a rebuild.

Here's another graph from the dataset, time based. The graph from CTS showing the dry sump results of nearly exact RPM and Oil Pressure change made me want to see if there was a fit in our data. Here, the RPM's(green) are rescaled to fit the oil pressure peaks(red). It fits the peaks quite well. Anything below the line is less than optimal. The blue graph below is lateral acceleration so that you can reference right vs left hand turns. Not sure what to do with this other than say, makes sense... If anyone has ideas for different plots, let me know, I'll give it a shot.

Quotes source post: 17446656

Inverse of the absolute value of the lateral acceleration and scaling against the pressure drops (similar to what you've done with rpm and peak pressure).

Jake, No, I've never funded or been involved in the development of anything automotive. I can drive, maintain, and fix them. I have been involved in the development of materials, optical and mechanical components, and automation to assembly lasers for communications. I understand the dollars and manpower needed to get it right. The market for any fixes for these cars won't justify anything new unless it's developed for something newer and is effectively free to implement on the older things, i.e. Sumebore.(excepting the UAOS that applies to all 996's, not just racing 996's) Agreed with GC996 that the goal here is to show what the best combination of off the shelf parts are to minimize this effect. RyanH., I'll go halvsies with you to make up the difference.

Quotes source post: 17446656

This is a good way to look at this. Maybe adjust the scales to make the two fit together better. On the other hand your graph gets the point across well enough. I don't have any rear engine data because all my data is from Boxsters. I suspected that rear engine M96's would not have the problem as severely. Looks like I was wrong about that. What happens next? Its spring. Oil temps are only going to climb from here. Chris Cervelli Cervelli Technical Service

Quotes source post: 17446704

You will have a new understanding and respect for those who do this everyday once its all behind you. That said, doing something like this in the manner you are doing it will make it instantly successful if something ever makes it to market. Why? Because people have learned to hate the "evil vendor" that makes money from the cars that they love; meanwhile they will support anything that is offered by other enthusiasts- even if it doesn't work. SUMEBore has already been a 6 figure development, and it will take YEARS of flawless performance, testimonials, races won, and high profile users for it to be trusted. We could have set up a GoFundme, and a couple of straw men to be the people on the front lines as enthusiasts, and it'd have a line 200 people deep on day #1. That's a fact.

Quotes source post: 17446725

LOL. I find your bitterness completely understandable. Chris Cervelli Cervelli Technical Service

Quotes source post: 17446732

Thats because you�ve seen it, too! We all have, and it�s getting worse.

Zbomb has the only access to the fundraiser funds, looks like it will make it to the finish line for the scavenge pump research !! Jason can buy the pumps and install them and show the results, good or no difference, that will put that issue to bed. I just hope he makes it though the day and has fun.. I doubt there will be enough interest for any cooling upgrades from the fundraiser to see the effect, but if there is I'm sure Jason will be glad to install them to help keep his engine alive and show the improvement . Thanks Jake the strong acknowledgement ! I like you have spent my whole life on automotive engines.

Quotes source post: 17446775

Allow me to contribute this way: The double scavenge pumps will not help. I spent the necessary $3000 to prove this in 2016. Setting fire to the money would have provided me with more satisfaction. Why does everyone think there is a shortcut here? Why does everyone think that hobbyists are going to solve this problem for less money? Chris Cervelli Cervelli Technical Service

Quotes source posts: 17446303, 17446725

You guys have been busy while I've been at the track - nice Jake, Don't hate you, but I think occasionally you talk too much to say too little. If you have data on what you have done, post it. You won't. If you want to contribute, beyond what had been established prior to your arrival - keep the oil cool, do it, we knew that already - with data, not just words BTW. Don't turn this thread into a sumebore anything, its not what its about - start your own thread.. Jason I'll be back after I get settled from the day - ran a PB today - I'm going to have a couple cold Budweisers and I'll be back.

Quotes source post: 17446798

Well, we weren't talking about the double scavenge pumps,(ie early X51), We were talking about worn-out scavenge pumps that couldn't keep up with the increased supply from worn bearings/loose internals. that has been said by an Engineer what says it fixed his from doing this, that you have acknowledged some of his findings on oil return. And I know it is a "possibility" and so when Jason brought up the subject, I seen the opportunity to put THAT issue to bed, once and for all.

Quotes source post: 17446798

Yep, because scavenge oil is not the problem. We cut windows in cam covers and installed sight glasses to prove this with a Go Pro, on top of data. I started playing with the scavenge pumps around 2008 and had given up on them completely by 2010, and haven�t looked back. Fluid flowrators installed in the external line was the icing on the cake. I still have a set of billet tandem pumps in my display case of bad ideas at the training facility for people to see. I think it�s a classic example of people looking at what the X51 engine had, and thinking it was what to do. The only part of an X51 worth a hoot is the intake manifold.

Quotes source post: 17446798

Steps aren't shortcuts Chris, they're steps. An effort isn't a guarantee of success - but it is usually required to get there. Help, or don't. With your documented experience, It would be appreciated but it's entirely up to you.

What exactly is the goal here? Chris Cervelli Cervelli Technical Service

OK, here's the graph to translate the lateral acceleration And here are 2 fits of the Inv Abs Lat Acc fitted to the oil pressure. One is scaled to fit the one of the peaks and valleys. The other is scaled to ride on the top of the pressure curve. I've left the RPM's in as well. Not sure which is the better or more appropriate fit.

Quotes source post: 17446825

To eliminate oil pressure drops to levels that present a risk to the engine. To do so in the least invasive and or complex manner and at the lowest cost to the operator. To be able to demonstrate with data the success or failure of attempts at resolution and if a solution is found - for that data to be available. Thats my goal.

Quotes source post: 17446833

Ok. I am claiming to have already achieved the first and third goals, and I don't think there is any disagreement with that. I suspect I have also achieved the 2nd goal if market realities are to be considered. If you agree with me here, then consider what needs to happen in order for my (or any) or system to be cost engineered. I think you'll come to the same conclusion I have, and each additional engine that gets destroyed will serve as further convincing. I will flesh this out further if necessary. Chris Cervelli Cervelli Technical Service

I did not think you had developed any system achieving those goals for a 996.

Ok, so taking from Jake�s posts about lowering oil temps, do you take the heat exchanger and make it larger in conjunction with larger radiators? Or do you add you own larger oil cooler separate from the existing system?

Quotes source posts: 17446868, 17444548

Just wondering how you think my quote from way back in the thread matches up to your view on market realities. As I said before and still believe - there is very little, to none at all - impetus for a vendor to step in and resolve the issue of oil not being where it needs to be. I also think, if you read the thread, a tiered approach to solving the easy problem first - cooling - along with a discussion on the hard problem, keeping the oil where you need it has been primarily contributed to by "hobbyists"... I like that word, I'm getting some stickers made, going to throw one on my car. It would be foolish to not seek guidance from those who have demonstrated success in the arena being discussed... thats why I PM'd you asking for you to take look in here and share your thoughts even though I had an idea what they would be based off our prior communications on this issue.

Quotes source post: 17446871

My system will work on a 996, but you will need to decide where the oil tank goes. There are two things to consider: What happens to the oil tank in a serious crash and does whatever happens present a hazard to the driver? The hose from the tank to the pressure pump inlet is near it's flow limit. I use a -16 hose for mid engine cars, and to be safe a -20 hose (which will make electric power steering necessary due to packaging a hose that big) will be necessary unless the oil tank is mounted in such a way that the hose is no longer than it is in a mid engine car. This is possible, but now the tank is in the cabin with you. It is also important that the hose and/or tank is not lower (with respect to the dry sump pump) than it is in a mid engine car. These are solvable problems, but it won't be trivial. On the other hand, the cost of replacement engines does justify spending the money. I won't be financing the R&D on this, because there is no business case to be made for doing do. I would help somebody do it in any way possible. Chris Cervelli Cervelli Technical Service

Quotes source post: 17446898

This is indeed all about market realities. My market so far has been almost 100% endurance racers who are charging for seat time. The system cost totally makes sense compared to the cost of a DNF. The other market has been Cayman guys with expensive modded engines. I have also sold a couple to people who just hate DNFs even though they can afford replacement engines. I don't think physics allows cooling to be a workaround. I am pretty sure it will be impossible to keep the oil temp at 200 in normal North American racing weather. Plus the actual problem does not get addressed. In 2018 I did three events in a row where the outside temp never got above 35. Those conditions allow a max oil temp of 200. I think you have to engineer everything to allow racing in 100 degree weather because that is what people often do. Chris Cervelli Cervelli Technical Service

Pressure and temps from today - this was my best lap taken at the end of the 4th 20min session of the day. Ambient at the time was in the mid 80's. Lot more data from the day incoming. For those that like to see it in the context of video, see below. You Tube

Quotes source post: 17446893

I guess this is what makes this data and thread valuable, Chris opinion is that physics will not allow cooling to solve the problem. From my experience, not ALL 996 Carrera or Boxster engines suffer from this affliction. But Jason's car does, as do many others , (and his has the equipment installed to show it in detail ), but is not quite to the level of instant death since the average oil pressure is still quite high. So I think is a very good specimen for research like the scavenge pump theory, and others if the Porsche Gods see fit.

Quotes source post: 17447132

Congrats! Shaving 3 seconds is epic. You can visually see the extra speed you are carrying in the footage. Everything was coming at you quicker. Awesome!

Quotes source posts: 17447219, 17446893

Agree! Regarding the scavenge pump, both Jake and Chris say it's not a worthwhile solution. So we have agreement here from two seasoned professionals. Regarding cooling, Jake has commented multiple times that cooling is the key. Chris says it's not. So we have disagreement on the topic. It certainly would be helpful to add clarity to nrp3's questions on cooling from Jake. Jake and Chris, if I misunderstood your opinions, please correct my statement. Lastly, for some reason, it keeps being referenced that the goal is to manufacture our own products. I got my own company to run, and just like you guys, I got the same headaches. I have absolutely no interest in becoming a manufacturer of auto industry parts. But I do have a love for my 996, and as such, just looking for a bolt on set of solutions that prolongs the life of our engines until the inevitable rebuild. Do not discount for a minute the appreciation and respect we all have for what you have accomplished in the industry. You guys happen to be the clear cut experts, so we ask you questions. No harm is intended.

Here are some first graphs from the day. 1st lap 1st heat, 7th lap 1st heat, and 8th lap 4th heat(assuming 4 heats because Jason sent me 4 files... Not many surprises. Engine starts cool with limited drops. The RPM doesn't fit pressure well because the engine is not at equilibrium. Engine temp changes from 187 to 221 through the lap. There are a couple of drops, but they are limited in depth to 64psi. By the 7th lap, the engine is stable hot, 251-254 degrees. RPM's fit the top of the pressure curve well. There are a few mismatches, but overall shows the pressure drops that we hate! Same types of pressure drops we saw in the last files. Magnitude of drop is not yet to 20psi which we saw on the file. Here is the last lap of the day. The engine is roasting at 260. Pressure drops are down to 20psi as seen in the previous file. The engine is 16 degrees hotter today than last time. Need some better cooling ASAP Same graph with the Min Oil Pressure curve added as described a few graphs ago. I haven't yet found a function in the program that will calculate time or distance of one trace under another. If we wanted, we could use the cursor to find the crossover points and calculation the time of pressure under the min. Yaz, I'm not sure what to do with the inv abs lat acc. Let me know if you are seeing what you wanted to see.

Quotes source post: 17447454

Looks like the correlation is there, but there's probably a better way to visualize it. Finally got Matlab on a home computer so I could do some analysis if you want to share the data set (and I find some spare time).

Quotes source posts: 17447371, 17446775

As we keep going back to - the real solution is to keep the oil where it needs to be, at that point temp management goes becomes less critical (CTS). The path of least resistance, and consequence that is likely have some positive results (although very likely not good enough), with products that are currently available on the market, either in kit form or as discrete components to be affixed, is that the cooler oil, despite not being where we need it, DROPS LESS and LESS IS BETTER. With regards to development of any new shiny widget - the 996 market, in my opinion, generally believes the available deep sump offerings work to mitigate, or negate entirely severe pressure drops on their engines, specifically while on track. As a result of that, the market does not require a solution to a problem that is thought to have been resolved with a bolt on solution that is generally thought of as affordable for the application, so people buy them. I found this to be FALSE, data has been posted to show, in my application, there is no doubt it is FASLE. The deep sump kit I have fitted to my car, running the deep sump MFG suggested oil DOES NOT mitigate or resolve pressure drops that could be considered a risk to the health of the engine. Until and unless the market responds by not purchasing products that DON'T work and lobbying the aftermarket manufacturers in a reasonable position to develop a product that DOES, there will be no reason for anyone with a financial stake to take on the liability of developing and producing a product that DOES work. This is why, I wanted this data to be available. To negate conjecture or opinion - to show, ON MY CAR that this product, 2 QT Deep Sump conclusively DOES NOT WORK to mitigate or eliminate oil pressure drops that are considered a risk to engine health. I'm one guy - this is one data point (My car), there are a ton of people running this deep sump kit that are far more happy to NOT have the data, to NOT be able to provide feedback on what works and what doesn't and to NOT be able to help force the vendor base to produce products that WORK by proving when the products they do produce DON'T. WE, the guys with a vested interest in keeping our cars running, and running FAST for as long as possible need to speak with our Dollars - as it always does, the market WILL respond. You know I'll keep fiddling on this Skip - by myself, with you guys... whatever. And I'll continue to share everything I collect, good, bad or indifferent. I'll throw parts at it as I can afford and if people want the pace of data acquisition and distribution to be sped up, I'm cool to test things that stand a good chance to having a positive affect for anyone that is interested - my 996 track dudes, vendors, whoever...

zbomb, In your car setup, you didn’t indicate what tires you are running (street, 200’s, slicks), other than looking at lateral acceleration it’s hard to know. Also, are you running the stock or aftermarket sump baffle to try to retain the oil in the center? Can you correlate lat acceleration to oil temp and oil pressure? I think this community should reach out beyond the US aftermarket suppliers, certainly the UK and European racecar engineering community is more advanced than in the US. Just read an interesting article from Hartech ( www.hartech.org ) on M96 issues, e.g., bore scoring that goes into much more detail than I seen elsewhere.

The oil temp is way too high for continued success. I believe that needs to be addressed first. Certainly reducing oil temp is not going to fix everything. Keeping the oil temp in check will improve the situation for sure. But first, the oil temp sensor in the sandwich plate is before the oil/water heat exchanger (I believe this to be the case) so what is getting measured is closer to the sump temperature. So I will pose a question, is the factory oil temp sensor after the heat exchanger?

Zbomb, I guess I missed it, or wasn't paying enough attention, but what engine oil have you been using when gathering these data points? Also, I see where your oil pressure sender is located, is the oil temperature sensor also located in the extension from the oil filter adapter? I am guessing so, I just want to confirm. Do you have a link to the particular adapter that you have fitted to the engine in the photo? There's a characteristic that I'd like to check regarding that particular component. I have found that the biggest benefit that the 2 quart deep sump has is allowing the engine to have more oil capacity, which increases the duration of track time that can be spent before the oil gets too hot. It seems that even 5 minutes makes a huge difference in keeping things alive, and those extra minutes come from added oil capacity. Now, even though I was a part of the development of almost all the oil sump related products (at some point) I will say that I have found internal engine modifications to be the single biggest factor when it comes to keeping things alive. People want something they can just bolt on to solve the problem, but I have yet to experience it over the last 18 years or so. Some things help more than others, but at the end of the day the issues lie internally. Some band- aids work better than others, but there's a reason why engines with naturally high mileage seem to live a lot longer on the track than others do. I do respect the way the forum has come together to support this initiative, and I hope you can all figure it out. I feel confident that with enough time you'll come to the same conclusion that CTS and I have, and I just hope you will share the negatives along with the positives.

Quotes source post: 17447756

NT01 LN baffle that is supplied with 2QT deep sump.

Jake, XP9 - may want to check the OP as I gave a breakdown of the setup. Both Oil P and T from the same adapter, you can see in the photo I posted prior. Adapter design below. And I’ll post all data I have, be happy to send to you to review as I have offered to do for anyone that wants to see it... All of it... you can see how bad I suck at driving in there too

Quotes source post: 17447768

The factory oil temp sensor is integrated into the oil level indicator. I have found this reading to be partially correct, and partially a calculate value when compared to stand alone instruments. Being in the oil level indicator puts the oil temperature sensor in the sump, so according to how one looks at the oil system, that could be considered either before or after the heat exchanger. When developing the IMS Solution I used a 16 channel Race- technology logger and 12 of those channels were split between oil temperature, and oil pressure data from all over the engine. Comparing this data vs oil temperature, RPM, coolant temp, and time were all revealing when compared to the factory oil temperature readings. At the end of the day I was looking for the most consistent location to collect data from. What mattered was seeing oil pressure directly at the main bearing feed, and seeing the oil temperature there was also interesting. To do this I had to drill and tap several access ports at various places in the engine. Seeing oil pressure after the oil pressure relief, and at the main bearings always resulted in wildly different pressures than anywhere else. All of these were behind the reasoning of the development of the SPOFA as an oil feed for the IMS Solution, because of the behavior of the oil at this location, and primarily how fast this provided oil pressure to the IMS Solution flange compared to any other location. This was done to help protect against start up wear, which was a major developmental concern.

Quotes source post: 17447796

Thank you. Data from that location will not add to our understanding of the efficiency of the oil to water heat exchanger. I had hoped for an in-line temperature sensor similar to what the GT cars use.

I would really like to add a sensor on the other side of the exchanger for delta... will require extender for the aim as I have dash inputs maxed, the sensor and mounting. Not sure it will happen anytime soon, if at all. As cool as it would be, that’s basically component level testing and of limited use for anything else. If the temp pre cooler is good, you’re good.

Quotes source post: 17447454

Could you make a graph of the 4th heat 8th lap including water temperature if that is available? Also, the rule of thumb 10 PSI per 1k RPM isn't as applicable at high RPM. If the oil pressure tops out at 45-50 PSI that should be fine even if the RPM goes up to 7k. The engine load factors into this too, if the engine is unloaded lower oil pressure should be OK. That situation probably won't occur on the track.

Quotes source post: 17447841

Let me explain my thinking here. Let's pretend for a minute the oil temperature into the main bearings is "acceptable" however we define that. The oil does it's job be removing the heat from the main & conrod bearings and gets a lot hotter in so doing. Then the oil in the sump may be far beyond what is considered "acceptable". With the AIM oil temp sensor located before the oil to water heat exchanger the temperature being measured is closer to sump temperature. With this data we cannot directly tell if the oil to water heat exchanger is doing a good enough job at cooling the oil before it gets to the main bearings and we have to go by secondary indications such as oil pressure and water temperature. Oil pressure and water temp are influenced by other factors and may not be telling the true story. Tl/DR, high sump temperature may not accurately reflect what the oil system is seeing. There is no easy place to measure post oil/water heat exchanger oil temperature, and retrofitting another sensor is going to be very difficult. I have an idea how to do it, essentially a sandwich plate for the oil cooler with taps for pre and post temperature, though it may not be practical. As already established access to the oil cooler component is terrible.

Quotes source post: 17447795

Based on the design of this adapter, you are seeing the oil pressure and temperature before the filter. Oil comes into the filter around the outer diameter, and then feeds up into the main gallery through the center of the adapter. You need to make sure that adapter does not allow for oil to bypass the filter because of a poor seal between the center and outer passages. The filter creates resistance to flow, which does impact readings. That said, better filters impact volume negatively. Crappy filtration usually means LESS resistance to flow, and higher oil volume. The trade- off between filtration, and volume must be a consideration in this data. The Wix filter is a great all around filter for filtration, and volume, working well all around. This oil filter adapter is a decent location to attain oil temperature, as it is post oil pump, and pre- heat exchanger, so the oil is at a fairly high temperature at this point, it is only hotter in the sump, where return oil is being delivered from the heads/ cylinders. Based on my experience, this is a poor location to gather a pressure reading for consistency. You should install another sensor at the heat exchanger and toggle between the two, then place another on the bank 1 cam cover port. Between these 3 you will see some of the differences that the oil system sees at various points. If you want to see how the oil pressure relief system impacts these readings install yet another sensor in the oil pressure relief housing to see when the relief opens, and closes. That said, if you have not done so already, update to the 997 oil pressure relief valve, and spring during this observation, and you will see how the relief impacts pressures as temperatures increase. I swap every engine to the 997 system, and usually shim the spring 1.5mm depending on the depth of the relief bore, which varies, and is also part of the problem. From what I have seen, aside from drilling and tapping a port, the oil discharge outlet of the heat exchanger is the best place to gather a better understanding of the oil pressure that the bearings are actually seeing. This is pretty much the last stop for the oil before it sees the main bearings. This is post relief and filter, so everything that can impact the pressure readings has already occurred. Based on your engine being factory, we have no ideas what your internal clearances were when the engine was new, and more importantly, what they are today. Based on this, I think that you should not limit your testing to only XP9. You should also explore XP5, XP6 &XP7. Seeing how the engine changes with these formulas will, help to determine what the internal clearances are like. After these are explored, then do a big formula change, and try Motul 300V, Redline and then Amsoil Dominator. Dominator has proven to be a very effective oil for @ltusler since prior to 2008. His crankshaft broke in half and the engine was sent to me, it had been running Dominator for years and everything inside of it looked awesome, but the lightweight flywheel did it's job ands he ended up with a 2 piece crank. Based on his experiences, I allowed him to continue running the Dominator after I built the engine, and it continued to excel, even after 200 race hours. I then reconfigured the engine around 2014 and it still looked awesome inside, other than what broke due to the AOS failure that hydrolocked 3 cylinders. That engine has been making 450HP since, continuing to use the Dominator with excellent results. You will have to perform a short service interval flush between each of these oils to reduce clash of the formulas, and to provide for better UOA results due to cross pollination. Between each oil type change you need to drop the sump plate, pay attention to the soot, and particles on the sump plate that have fallen from suspension (typically due to overheating of the oil). Allow the oil to drain from the sump over night, and also remove the B1 chain tensioner to remove everything possible from the chain well. On our test engines I drill and tap the B2 head chain well so I can add a ball valve/ removable plug there to allow for draining more of the residual oil from the system. A ton of oil can be removed from these chain wells, and it must be for testing to be accurate. Cross pollination is your biggest factor here, and what takes time and effort to reduce. I would test through all the XP series oils first, back to back. These all use the same characteristics, so cross pollination of these will make less of an impact. I would make sure to use Redline last, since it does not jive in back to back operation with XP series oils at all, and chemical clashes are probable. At the end of it all, go back to some Mobil 1, and see what you learn, and then swap back to the same XP9 you had drained out and started with, to make sure the engine hasn't "gone away" or changed since the test began. If it did change, you didn't really test any oil directly. I always did this, and I always saved ALL of my drained oils in clean containers, so it could be reused again to compare results at any time. Never throw away the oil until all the testing is done. You may find that YOUR engine likes a different oil than XP9. Back when my wife was Land Speed Racing, the car doubled as a DE/ test car for oil system developments. This was 11+ years ago, and XP9 was still a formula called "980" that had no other designation, and was still being used in circle track pro racing only. Back then, my wife's engine didn't like the XP9, it preferred Redline, and 3 of her 4 records were set on Redline oil. I only learned this because I tried different oils ion this engine over a 5 year period, on the street, on the big track, and in LSR. My wife drove the car daily, around 60 miles per day, and on the weekends it had many other roles. It had data cables, sensors, and crap all over it. Lots of people think I only use Driven oils, but that isn't the case. Different engines, applications, and environments call for different oils. I specify these to the owners of each engine, and at times someone will even get a Liqui- Moly recommendation. You have to make the engine happy, and to make it the happiest you have to feed it different forms of lifeblood, paying attention to the results of each. A lot of days of changing oil, sensor locations, and etc lie in front of you. It can be fun, but it starts to become a chore when the data and experiences don't do what you anticipate. My rule after 20+ years of development is to not carry any expectation, or anticipation into a development/ test. You will learn more if you do this, and you will learn faster. My other rule is to set no budget for development, and not to keep up with expenses of the development. I add it all up when I am done, or sometimes I don't even want to know what it cost. If you keep track of the dollars you end up letting economics drive the result, and then you've just done the same thing that led to the issue that these engines have had since day #1. I don't want to allow my experiences to impact what you guys will do/ learn/ apply, so I will check in every now and then, but don't expect any input from me. It may get in your way, and you may learn something I haven't IF I don't get in the way. This is actually why I hadn't even opened this thread till yesterday to do more than read past page #1. I hope you have a lot of time, extreme patience, and like to scratch your head. Unless you change locations of your sending units, you won't ever have the big picture to really understand what is going on, and where it is occurring within the engine. This is what takes all the time.

Quotes source post: 17447862

I just weld a bung into the top of the cooler and install the sensor with the probe protruding down into the "post cooler" oil passage. The next stop from there is the main bearings. Zbomb's install is indeed reading the pre-cooler temperature. I'd like to confirm that the oil temp data we are seeing here is coming off the sensor installed in the oil filter block, and not the CAN data. The stock oil temp sensor, as Mr Raby states, is not entirely accurate. I think it maxes out around 125C or 258 F. I think I saw somewhere that the water temp was 217 F. There is probably room for improvement there. Zbomb's car probably has the radiators it was delivered new with. Those are going to be caked with crap, and replacing them will drop the water and oil temps. It does not really matter though, because we see abnormal oil pressure on lap 1, when the oil temp is 207. That's why I claim that while cooler is better, you can't fix this with cooling mods. Have a look at the RPM data. AIM gives the average RPM on these lap charts. The data posted this morning shows average RPM around 5100 (up from 4750 on yesterday's data). This very low and is mitigating the oil pressure problems. The average RPM is a decent proxy for how much oil is NOT in the sump. I have seen as high as 6100 average rpm with a 2.7 engine and close ratio gears, and 5700 rpm has been typical, and in my case at least that was for races of 7 to 25 hrs duration. My point is, what we are seeing here only gets worse as the car and driver gets faster. theprf called the oil pressure rule of thumb into question and he is right to do so. 50 psi is enough under all conditions for the M96. Speaking of that, the pressure numbers presented here are very high. It may be that the engine has unusually tight clearances or just that the oil is thick. Can someone check the calibration of the sensor? I suspect the high end of the scale is not correct. Also make sure that the sensor reads 0 psi at 0 RPM. Zbomb is very correct in pointing out what has happened here with regard to the sale of "fixes" that don't work. Hundreds of deep sumps and Accusumps have been sold. Literally a million dollars worth (you can also add up the value of the all the destroyed engines), likely enough revenue to have designed, built, tested, and mass produced the system we all want. People on Rennlist have been the most effective marketing department for these ineffective fixes, which is why I reliably chime in on these threads and complain about what is happening, again. Thanks to Zbomb and golock911 for presenting this stuff in an engaging and understandable way. Chris Cervelli Cervelli Technical Service

Lot of great info Jake - thanks for sharing it. Im a hobbyist, see the sig All the data is from my setup - from my use. I don’t plan on changing Professional fields anytime soon... just driving fast, collecting and sharing data and driving fast. For those that run setups similar to mine, or want to run one similar to mine - hoping the info is valuable for them. The development work you’ve done is awesome, the hardware fruits of that labor are unattainable to many of us though. So, thanks for the info.

Quotes source post: 17447949

Maybe more enthusiasts will become involved.. Unfortunately, unless you try all of these different things, the things that you share will not be conclusive. That said, back when I was doing a ton of this development was in 2008. The person who was getting a ton of the data and doing his parts of the development wasn't a professional either. He was a pediatric cardio- thoracic surgeon that was a huge part of the development when we teamed up with him getting track time and us preparing the car/ engine/ sensors. Due to his background he was insane with data observations and attaining OFAT (one factor at a time) data. To really be able to learn what people need to reading here, you can't be an enthusiast when it is handy, you have to be all in, all the way, all the time.

Quotes source post: 17447937

Oil temp from AIM sensor Water temp from CAN Oil Pressure at 0 RPM

Jake... people can take my observations, or not... I, unlike you... don’t sell ****. Prople can contribute their thoughts, or not. As long as they’re not trying to sell me anything, I take it for what it’s worth based on my assessment. Now, if you’re trying to sell me something or convince me you have resolved these issues: Present the test case and scenario Present the data Get some orders for your parts or ridiculed - the way it should be in my opinion.

Going back to what I posted yesterday, where the evil vendor just wants to sell someone, something. Again, I will state this: I have NOTHING to sell anyone. I have not sold a single component to anyone since 2009 when I closed the retail store. ALL I have to sell are engines, and I refuse to sell anything else, to anyone, under any circumstance. At the end of the day what I have learned makes my engine better, and that is all. It also helps me share what I have learned in my classes, with enthusiasts and pros alike since 2007. And there comes the heart of it.. Its all about sales, and chatting the vendor, just like I said yesterday. When ddid I try to sell you anything? last time I checked I offered you priceless insight into what you need to do to collect OFAT data. I';ve been doing that for a decade before you were a member here. So I deserve things that I don't sell to be ridiculed? At the end of the day you just want vendors to be ridiculed because they are vendors. All of your actions have proven this. There's no store on my site, I only sell engines.. Nothing else. People only need me after the things they have bought, or done prove NOT to work, and the engine pops. That is all. Be nice to all vendors, like it or not we are what keeps your car on the road. I never saw the level of vendor hate from Aircooled Porsche owners, they always supported what we did. They still do today.

Jeesh, so sensitive... I don’t hate... If you look at where I buy parts from, that would be clear. If a vendor... any vendor made a part I thought could fix this problem and I could afford it... I’d buy it in a heartbeat. Am I a meanie because I posted data that a part a vendor made doesn’t work for me the way it’s supposed to in my application ? That I think vendor claims should be backed by data or called out for their validity ? Please point me to the data ( data logs, dyno graphs etc... )you have shared on this site, on any topic. I’ve missed and I would surely like to learn from them.

I can donate my old oil cooler and get a bung tig welded into it. Let me know what size u need for the sender. Maybe the next gofundme will be for the aim expansion.

Quotes source post: 17447937

Assuming that Z decides to go this direction, would it be possible to supply one of your upgraded Cayman oil coolers with this bung welded in place? I am not a good aluminum welder.

Quotes source post: 17448047

It's probably best not to reuse an oil cooler from an engine that had a bearing failure. There is too much risk that debris is trapped in the oil cooler.

Quotes source post: 17448049

The welding is not easy. I should have one hanging around. It will have a M12 X 1.5 thread for the Temp sensor (Bosch 0 280 130 026, AIM will have a calibration for this) and a dash 3 male to attach the pressure sensor hose to. Zbomb currently has the sensor hard mounted, which is bad practice because the sensor can fail due to engine vibration. Chris Cervelli Cervelli Technical Service

Quotes source post: 17448036

When it is convenient to be a regular old enthusiast without resources you are utilizing that.. Then things don�t work the way you expect them to, and you instantly regard yourself as an expert with all the data to support the narrative that the products don�t work. In the beginning I believed you had a different overall goal. Your actions/ statements have proven this to be incorrect. As far as sharing data, I stopped doing that over a decade ago. I realized that no good deed went unpunished and that people read data to reply- not to learn. At that point I realized how dumb I was to be giving away what I had sunk so much time and money into. If you want my data, attend a class. Nothing good will come from any professional trying to weigh in on this thread. I�m out-

Quotes source post: 17448068

Chris - I'll take a look at what I need to do to hook that Bosch sensor up to the aim from a wiring perspective - all the stuff on the car is plug and play which is important, cause I do best with easy. I have heard that the hard mounting was not ideal, but thought that was more an issue with the old VDO sensors...

Quotes source post: 17448055

i still have the old one that i replaced 2 winters ago when i did my coolant hoses.

Quotes source post: 17448010

Jake, Your depth of understanding and relentless curiosity has yielded some incredible benefits to us all. I completely understand your reluctance to share when the respect and reciprocity isn't there. Looking forward to 4.3L of Fla6Innovations perfection from my well maintained 3.2L 2006 Boxster this Fall.

Dudes! How about we turn the temperature down a bit. This is a forum we come to for information gathering and giving because we like our cars. People will always have differing opinions and ideas. That is a normal part of life. I see no reason to get into verbal battles with someone else on a forum. There is enough craziness in the world already...

Folks, Let's refocus on the topic and move forward. The problem with emails, texts, and blogs is that it's difficult to judge the tone and intent; as well as humor. We all have bad days and can misinterpret, get pissed and then everything escalates. My suggestion is that we shake it off, suck it up, and move forward. Too much good info to let this thread die an early death. The intent is this. Jason has installed the AIM software on his 996.2 and set it up to measure oil pressure stability in combination with oil and water temps, RPMs, g-forces, gears, speed, etc., to create a baseline of data that can be used to understand what's actually happening on the track as it pertains to the items listed; and then find bolt-on solutions that will address the oil stability issue by Jason bolting them on the car and then testing them on the track for effectiveness using his Aim software. Paul produced some incredible charts from the baseline data that were incredibly revealing and caught the attention of many. Skip came up with a great idea to set up a Go-Fund-Me account to help pay for various bolt on solutions to test to help with costs to speed the process of testing up. Many of us liked the idea because we all want to find ways to protect and prolong engine life in our cars, knowing damn well that each and every one of our engines will have to be rebuilt someday. But if we can prolong engine life just a little, we will be in a better position to prepare and decide on how and what engine rebuild we want to ultimately do, and hopefully control the when. So at this point, the forum was discussing what next steps made sense as to what should be done. Discussions seem to have taken the path of cooling first. So, let's keep this thing going.

Can we post the funding info on the first thread?

Quotes source post: 17449357

I updated the OP With the link Mike.

Thx!

You all hit the number... amazing - really cool what can be done collectively. We’re going to learn something new. Thank you to everyone supporting the idea that we can get smarter, faster with data. Skip - let me know what I’m buying and where to get it.

Quotes source post: 17442441

Well theprf has found the New 997 scavenge pumps and New 997 updated oil pump by-pass valve and spring with part numbers all for about $430. While I have my doubts also as to whether this will have any effect, it is the only thing "claimed" to have fixed this issue on another car ,we can't "assume" anything since this a particular problem on a particular car. Other cars with similar symptoms may have a different cause. With about $90 left over you could buy the hi flow oil filter and x96 oil also and do separate data set. Of course its your funds ( I just set it up ) and your car and you have to do the install, so you have the say so. The scavenge pumps and oil by-pass are not a hard job. BTW someone in VA bought a UAOS-AL today so I donated another $25..

I’ll get the scavenge pumps and bypass stuff ordered tonight. Let’s open the discussion on how any remaining funds should be spent - my plan would be to keep first change limited to the scavenge and bypass and based on assessment of those results, make a firm decision on the next best opportunity. Thoughts ?

I tend to agree with one change at a time. Otherwise, 2 changes can be made, but then you�d have to go back and undo a change to confirm the results.

Quotes source post: 17450001

makes sense!

Quotes source post: 17449960

What all needs to be done to pull the bank 1 pump? Just a matter of disconnecting engine and transmission mounts, and shifting everything rearward an inch or so?

Quotes source post: 17450059

yep,IIRC it only took less than 1/2" more clearance to wiggle the bank 1 scavenge pump out. Bank 2 is a piece of cake.. Also FWIW, the pumps must be clocked or oriented correctly. The is 1-3 with an arrow and 4-6 with an arrow on each pump. The corresponding number to cylinder bank and arrow must point toward the crankshaft. If installed "backward" the pump will try to pump oil from the sump back into the head, instead of the correct flow direction.

Gaudin Porsche has the scavenge pumps for $161. They are my go-to dealer. ECS Tuning = 196 Pelican Parts = 202, claims 25 in stock.

just get the pierberg ones from FCP. They are like $112. Ultimately, I'm pretty sure we'll end up in the same place....the pressure drops are going to happen regardless and all you can do is keep the oil as cool as you can.

Quotes source post: 17450283

This looks like it. It's a slightly earlier version, 997-107-021-04. https://www.fcpeuro.com/products/por...rg-99710702103 They're up to $140 now. O-ring for the pump: https://www.fcpeuro.com/products/por...er-22543132001

So - are we good with the Pierberg ones ?

99710702103, 99610702153, 99610702155, 99610702152, 99610702154, 99710702104 , 99710702105.. four outdated 996 numbers and three outdated 997 numbers that's interesting. Could be from vendor swaps or could be actual design/material changes..

if only $20 bucks difference id get the newest part number..

Gaudin is 3WKS LT from Germany FCP has https://www.fcpeuro.com/products/por...rg-99710702103 In stock

Quotes source post: 17450814

yes, that makes a difference, Pierburg is probably making the Porsche boxed ones anyway, may even have the updated Porsche number on it.

https://www.fcpeuro.com/cart?utm_sou...1060639413%231 We good ?

Okay, I just read “most” of this thread and have a question: What is the [EDITED BY MOD] purpose of this thread? Thanks in advance 👍

Quotes source posts: 17450890, 17446825, 17446833

Oh - hey. Thanks so much for joining us for this technical discussion on oil pressure instability in the Porsche 996, M96 engines during spirited driving. I apologize if the goal of the thread was not clear. For your convenience, I have added a couple quotes below that will hopefully provide you with greater clarity. Please do lt me know if you have any questions, and thanks for stopping in

Here is the remake of the session 4 lap 8. In truth, this is session 5. I missed one of the files. The program allows functions so I was able to put a cap on the min pressure of 50C. It's easier to interpret. The water temps is added to the bottom with the oil temp. Here is the temperature series. I've separated water and oil into separate charts for clarity. The pattern is the same on all sessions with each lap increasing in temperature. The max for each session increases with the sessions. I believe the data is clean enough that if the scavenge pump have an effect, we will see it. Good comment from Chris about avg RPM. I'll try to get that indicated in some way. Good comments from Jake about planning out next steps and tests. Some planning required to keep it clean. And a lot of laps. Yaz, can you PM me an email address to send this file to? The program claims to be able to convert the data to the matlab format. I plan to send 2 versions. .mat and .csv. Let me know if you would prefer .xls.

Quotes source post: 17451008

I must have missed that, sorry. Have you lost an M96 engine to low oil pressure? Are oil pressure drops hindering your lap times? Will solving this problem increase your lap times?

Quotes source posts: 17451063, 17451063, 17451063, 17451063

Hey ! No worries at all man, all good in the hood. Nope, still under the hood, I mean trunk... I mean, ya know the lid in the back... I'm not sure what the real Porsche guys call it, its back there though - I'm sure of it. No, I don't think so, the nut behind the wheel has a lot more to do with that - haha Only solving it will tell. ​​​​​​​BTW, check out the data Paul just posted... That dude is a maniac with RS3. This thread really doesn't have much to do about lap times - It's really about not throwing rods through cases I'd appreciate if you could help make sure we keep it on topic. ​​​​​​​Thanks ! ​​​​​​​

In case u havent seen what the cumulative effect of low oil pressure in a turn can do to a motor. Maybe there us a solution out there, maybe not but at least z is trying.

Quotes source post: 17451038

Thank you for the graphs. The caveat here is that water temp is read from CANBUS so it has been filtered/averaged/manipulated in ways that we don't know. Making the possibly rash assumption that it is accurate, we can see that water temps are well under control around 220-225F and do not change much with large changes in oil temperature. From this I will draw some conclusions: 1) Adding radiator capacity could reduce water temps somewhat. I would not expect a dramatic change. 2) Lowering the water temperature will have only a small effect on the oil temperature. A dramatic change in water temp will be required to effect a noticeable change in oil temp. 3) It appears that the oil to water heat exchanger is being overwhelmed. 4) There is capacity in the water cooling system to absorb more heat, if it were possible to get the heat into the water cooling system. 5) Even if the water cooling system was not able to absorb more heat there are well known upgrades that can increase the heat capacity.

Quotes source post: 17446908

Just rereading some of the posts... Chris, Would the space where the mufflers live in a 996 be sufficient for the oil tank that was used in your Boxster application ? That would keep hose length to a minimum, especially if mounted passenger side

Quotes source post: 17451855

It does seem as though a significant water cooling improvement will be required if we are to see appreciable results on the oil side - 200 or less water temps ? Ambients getting hotter... only 1 day of warm weather posted so far.

Quotes source post: 17451932

Probably not, but maybe. The current tank is 18 inches tall and 7 inches diameter. Worse yet, you need some access to the top of the tank to fill the oil and to service the oil filter. The other issue is how low the bottom of the tank would be if installed here. The hose routing would be tough (-16 or -20 hose is very large and not flexible) and the pump inlet would be a few inches higher than the tank outlet. This is worth looking into. The tank design can be changed. It looks like a short fat tank could fit there easily and the muffler bracket would make the mounting easy. I have a couple 996 Turbos in the shop right now so I will have a look. Chris Cervelli Cervelli Technical Service

Quotes source post: 17451038

Thanks once again for this excellent analysis. A few comments: The minimum oil pressure concept is not valid. The pressure drops because the pump has sucked up some air, and even if the pressure only drops a little, that air has to go somewhere, and if it goes to the main bearings then damage will be done. This is why Accusumps are not a solution for this problem. The data shows the temperatures almost reaching equilibrium on lap 8 (not sure about out laps and in laps). This winds up being very important because many cars that are used in amateur racing (especially DE's) spend very little time at their equilibrium temperatures. Zbomb's car for instance is hitting 220 on water temp in the weather conditions present when this data was taken. If the weather was 15 degrees warmer, he is still going to be able to run, but he'll be worried about temps by the end of the session. The same car on the same day is unusable in an endurance race. It is gauranteed to DNF, because it is going to be running 230 water and 275 oil for 8 hours. This becomes really important when you are designing something that is going to used by a large number of racers. The product has to be able to survive worst case scenarios or it can't claim to be a solution to a problem. The waste heat dealt with by the cooling system is proportional to the average HP generated by the engine during the lap. As Zbomb's driving improves, and as the car is developed, this average power is only going to go up, making the cooling problems worse. Also, in a race environment it is likely the car will do a few laps glued to another car's rear bumper, and it will need a cooling reserve to handle this. We have not even talked about transmission cooling. The transmission takes a long time to hit equilibrium so trans cooling is not usually needed for DE use, but it is always needed for endurance racing. If you want the trans cooler to actually work, it will need to be oil to water, which increases the load on the water cooling system. Chris Cervelli Cervelli Technical Service

Great thread and great project, zbomb! I'm lurking from the 997.1 side, but I think the issues are pretty much the same there, including the recommended fixes. A third radiator is a no brainer in the 997.1 (cheap and OE Porsche), and it does help with temperatures at the track. (i haven't found much difference in street driving.) I'm intrigued by the low expectations for the 4 scavenger pump set up. After all, this seems to be the main difference to the 9A1 engine, which apparently does not have oil pressure problems. Of course, the 9A1 engine also has a variable load oil pump, but I don't think it can deliver more oil flow/pressure than the always-on pump in the M96/97 engines. To my hobbyist mind, the symmetrical 4 pump set up must be superior to the asymmetrical 2 pump set up because the issue is clearly worse in right hand turns compared left hand turns! This makes sense for the 911 configuration, where the 2 scavenger pumps are fore-left and aft-right, so in a right hand turn there is the longest distance to cover for the oil return.

Quotes source post: 17452137

Chris I didn't copy your entire post. I agree that the rule of thumb oil pressure guideline is just that, a rule of thumb, and the oil pressure drops seen in these logs are not good. I also agree they are due primarily to oil pump cavitation. I was under the assumption that an Accusump would mitigate the aerated oil pumped or at least minimize the number and size of the air bubbles. Leaving aside the difficulties of recharging the Accusump, of course. On Z's post about locating the oil tank in the muffler space. Would it be possible to use a hard line from the oil tank to the pump? I am thinking of a setup like the 6 Turbo uses for the main scavenge pump return line. The hard line should reduce pressure drops vs. a hose and would be smaller in OD for the same ID. A suitable hose would be needed at the pump end to absorb vibration and engine movement, of course, as the tank would not be mounted to the engine. Also, the fill pipe could be located on the side or even the bottom, as long as the actual fill opening is above the top of the tank. That could give some mounting flexibility too. There is data on water temp for some cooler ambient temperature days. It is not directly comparable as Z is extracting more power for longer which increases the waste heat, though it should give some idea of the effect of ambient temp on water temps. I would expect the hottest days in this part of the country to be no more than 15F higher than what has been recorded. Track temperatures will likely be higher due to the pavement absorbing heat and it's not possible to discount the effect of disturbed intake air. Given that converting to a dry sump system is likely to be a long term project, I believe that the best available band-aid is going to be reducing oil temperatures. There do not seem to be any other possibilities immediately to hand. In the interim even more frequent engine oil changes may help, too. Regarding the gearbox perhaps fitting the Turbo undertray with the NACA ducts could be an interim step, if it will fit. I am not aware of anyone having added pumped cooling to the G96 transmission. Probably frequent gearbox oil changes would be indicated too.

Thanks for jumping in and the kind words - I do think both generations have a dog in the fight despite your better headlights and interior. On the face of it, I agree it seems that the additional scavenge pumps should make an improvement at a minimum in theory... based on Jake, Charles and Chris� input, they don�t or don�t.... enough. For the cost of that �upgrade� I think or it to be a realistic for most owners path forward, it would have to do a damn good job to resolving this problem in full. Now, if a large component of the issue is MORE oil has relocated away from the pickup than is trapped in the heads, while then the impact of improving the scavenging may still leave you with problem of... oil not where you need it to be. That would be mu uneducated guess, maybe Chris can provide data behind his testing.

This is all so true.

Can those collecting data also run on the street in a couple of on-ramps or off-ramps? I asked because tracking might not be required to see the pressure issues. Only do this if it makes your lives easier. Otherwise, continue on.

Quotes source post: 17452332

Indeed, maybe the cognoscenti can chime in and explain what the silver bullet was in the 9A1 that resolved the wet sump oil pressure problems.

I’m going to get some street data. Give me the weekend.

Quotes source post: 17452411

I am not familiar with the 9A1, but I build a lot of Mezger GT3s and 996 Turbo engines. These engines use a very small oil tank and do a perfect job of maintaining oil pressure, although there is not much spare oil in the oil tank so it's critical to keep the oil topped off. The obvious difference between a Mezger and an M96 is the amount of internal space inside the engine. In the Mezger, everything in the lower end of the engines misses everything else by a millimeter or less. The rods are that close to the oil pump, and the crankshaft counterweights almost touch the piston skirts when the pistons are down. There is no space around the outside of the cylinders at all, compared to the huge area in an M96. The heads and cam housings also have nearly no internal space. You can see a lot of attention is paid to minimizing the space in the cam housings and valve covers. All this extra space in the M96 hides the oil when the RPMs and G forces are high. My dry sumps work quite a bit better when I build the engine with restricted oil flow to the heads. I think they have that calibrated correctly in the Mezger (which have very short valve guide life) compared to the M96 (which never wear out valve guides, probably because they are submerged most of the time). Chris Cervelli Cervelli Technical Service

Quotes source post: 17452319

The Accusump is better than nothing. The oil discharged from the Accusump pursues the aerated oil thru the pressurized oil system, but it does nothing to remove that air, which is going to wind up on a lubricated surface that requires oil, not air. A hard line is a good idea. I could supply a Wiggins fitting on the pump inlet to connect the hard line to. The hard line would need to be about 1.25 inches ID if it has any serious bends, which it will. Chris Cervelli Cervelli Technical Service

Quotes source post: 17452411

I wondered if there was some evolutionary change in these engines.

Watching, will donate shortly if funds are still needed (I only read key points so far). I have the same problem plus smoke! Porschetech also pointed to scavenger pumps for my engine as well. I just haven't had time to tear into it so it sits on the lift in hover mode Shawn

Do 986/7s have similar issues but on left-hand turns? Does the 40 Jahre with the X51 have similar issues on track, or not enough data points to know?

Sorry for the delay dudes - I wanted to be sure I was getting the right parts, I know nothing about these pumps or what if any interactive changes occurred - but 2 smart dudes agreed, and I just got the stuff on order. I am really excited to be able to move on with testing these parts and sharing the data with all of you. This situation of crowdfunding car part testing is certainly unique to my experiences on a forum or any where else for that matter - I appreciate you guys having confidence in the data to donate your hard earned money to see what the data says. I have told some of you privately, and I'll relay it to the group watching, and participating - I feel super lucky to soak in some of the info coming from such smart people in well written, reasoned and articulate posts. I'll certainly keep that in mind when I'm swapping out the pumps and I hope, we learn something that can keep a fire lit on the discussion if need be. Especially - got to give a shout out to Paul golock911 - his formatting of the data and written analysis form a perfect foundation to allow the data to integrate into discussion, he makes it look easy - It's not. Thanks Paul.

Looking forward to it!

Thanks Jason and Chris for the kind words. Putting these together is fun and a bit of a distraction from the ' work from home ' boredom. A good part of my job in development and new product introduction is to look at alot of data trends to establish cause and effect and communicate issues to the team. With the ultimate goal of better yields.(and higher profits) I'm happy to help. The key for me is to feed the discussion and generate the next questions. Do some work, generate the next dataset. Peeling the onion... Here's the chart for the day. It doesn't really tell us anything new, but I now know that the RS2 program is not good at XY plots. It's oil temp vs water temp. Same color scheme as before, each color is a lap. All it really shows is that we are very near equilibrium, as mentioned by others. If we need further correlations, I'll take the data to excel or JMP. If anyone would like to see other plots, post it up and we'll give it a try. This is turning into a really nice piece of work due to everyone's contributions. Thanks! -P

Quotes source post: 17452831

My guess is that the X51 motor would also suffer from this. I have a service information document that I'm happy to share on this but it states that the engine modifications include: Cylinder head port and polish, different cam shafts with different valve timing, different valve springs and the difference around oil supply seems to be around the X51 baffle, which I think Zbomb already has. I think there were also some improvements around getting the oil to cylinder 6 but I don't believe this would do much to alleviate the issue being discussed here.

Quotes source post: 17453593

Wow, interesting to see the cold shower for the water temperature as the thermostat fully opens...

I would HIGHLY encourage everyone interested in this issue to send an email to Hartech requesting the Crankshaft failures Final Report they have published. Hartech contact is [email removed] I have also linked him to this thread for his contribution if he sees fit.

Preferably please contact [email removed] for that report (I am really too busy with technical stuff to handle a lot of enquiries - sorry). There are also other reports on the 4 main problems with the engines, bore scoring etc - but they are not a quick fix read but serious technical stuff and not for the faint hearted. We are trying to find the time to make a few videos (accept it is the new and best medium) but they take so long to create - think we will only have time for off the cuff - non rehearsed stuff - but should still be valuable. Baz

Quotes source post: 17454036

Edited my post to reflect preferred contact Baz, sorry about that.

Golock911, is there any fore/aft G data in there? If so, would it be possible to create a 360deg polar plot of the maximum Gs in the horizontal plane?

Quotes source post: 17453593

I'd like to see a series of X-Y plots, lap by lap, in the following format: X axis is Rpm scaled from 3500 to 7500 rpm Y axis is oil pressure scaled from 30 to 90 psi Color channel if available is the throttle or gas pedal position Thanks, Chris Cervelli Cervelli Technical Service

Instead of bombarding Hartech with emails, could you get permission to share the PDF here?

Quotes source post: 17454824

I�d like to SV, but Baz asked that I not reproduce it but that I could provide the info on how to obtain, so thats what I�m doing. As I said before, get the paper. Set aside a chunk of quiet time to read it and come back and post your thoughts.

Deleted - off topic

I gather LWFW means lightweight flywheel?

Quotes source post: 17454922

Correct

Discussions about broken crankshafts ought to get their own thread. I have not read the paper but in my experience broken crankshafts are not a concern. Chris Cervelli Cervelli Technical Service

I would like to try to maintain all focus in this thread on the resolution and or mitigation of G induced oil pressure drops. We all know that the consequence for inadequate oil pressure in an engine is eventual failure requiring rebuild or replacement. Break a rod, spin a bearing, break a crank... pick you�re poison... The Hartech paper is a good read for anyone interested in their experiences with these engines.

Quotes source post: 17454768

Paul sent me some data last night. I was thinking to start with a series of 3D surface plots (with and without notable oil pressure drops), and some splices of those to account for constant oil and water temps. How useful would it be to develop an estimate oil pressures based on all other sensors with the idea of identifying oil-air intake and cavitation conditions?

Attached is a .zip file with individual lap plots for oil pressure vs. RPM. Here is a plot with both lap 1 and lap 8 to show the effect. Chris, I'm not sure this is what you wanted. I wasn't sure what you wanted done with the throttle position. I couldn't find a way to color the trace with the throttle position. Let me know what I'm screwing up. I'll fix it. The volume of lower pressure is significantly greater on the 8th lap (we already knew this) So I put together another plot oil pressure vs throttle position. I'm not sure it tells us anything I didn't have a chance to get polar coordinate graphs figured out, so I put together this chart to get my head around how to do it. Nick, I'll try to get to this tomorrow. There is a ton of information on this plot that can be used to infer how the oil is being thrown around in the case. The upper graph is our oil pressure with RPM overlay. The middle graph is lateral and inline accelerations. The Y axis is in G. The scale is correct for both lines. Stay with me now: -The green trace is lateral acceleration. Positive values means the car is turning right, weight is on the left, Oil will get tossed to the left. Negative values mean the car is turning left, weight is on the right, Oil will get tossed to the right. -The blue trace is inline acceleration. Positive values mean the car is accelerating and the weight is in the back, oil will move to the back. Negative values mean the car is braking and the weight moves to the front, as does the oil in the engine. The bottom graph is both the braking pressure(blue), which aligns with the negative inline acc, and throttle position(light blue) which aligns with positive inline acceleration. The surprise in this graph is the first big pressure drop. It occurs on a left hand turn! The combination of heavy braking and turning either way show the pressure drops. The reason we don't see it as much on lefts is because the braking before a left is not heavy. It would be cool to get data from this track racing the opposite way. This last one was a bit of a curiosity and likely only tells us about the measuring system and is also likely of no consequence. The oil temperature plots show a very repeatable pattern across all the laps. The variation is quite small, maybe a few degrees, but it's interesting none-the-less. I looked at all the other parameters and the only one that shows anything close to a relation is the lateral acceleration. As a big right hander comes up, the temp decreases. Then it goes back up. I would argue that the last right turn is so fast the effect is not seen by the temp sensor. First thing I thought of was that the air bubbles are going by the temp sensor and it cools slightly. Maybe I'm just full of it on this one. Thanks for the indulgence.

Quotes source posts: 17455770, 17043699

I just sent you the data acquired when running the opposite direction, ran that way on my 1st track day this year, 5/10. BTW - That left hand turn, T1 - is the first time I ever noticed a pressure drop, last year, running the opposite way (Right hand turn). It was a major inspiration to get setup to collect actual data as I thought with the parts I had installed, I had done what I needed to do to prevent such drops from occurring. Now that we have that data - we see just how wrong I was about that, at the time - how wrong was I was still a question and it took the data to answer it.

[EDITED BY MODERATOR] There have been 2 or 3 other professionals that have had no problem contributing data and see value in this thread. It Jakes policy to keep his data close to his vest and thats fine but we will proceed without it. The data shows that there is deffinately an issue and the current solutions dont fully solve the problem. Hopefully u read the entire thread and come back with a contribution.

Cool data, this is nice. So for left turns after hard braking the oil pools in the fore-right position, where there is no scavenger pump and the distance to the oil pickup is maximal. For right turns it pools on the left, where the scavenger pump is again furthest away from the oil sump. In theory, pressure drops in right-hand turns might be made worse by positive acceleration, which pools the oil in the aft-left position where there is again no scavenger pump. Actually, this agrees with the worst pressure drop I've seen in my car, which occurred at Palmer clockwise at the sweeping uphill righthand turn T8. This was before I installed the deep sump, but I haven't been back to test whether that made a difference. To me right handers are still more likely to cause trouble because in any case then the scavenger pump is furthest away from the oil pickup. But significant fore-aft accelerations can change this picture. I guess we could draw danger zones on the old friction circle....

Our report about crankshaft failure mentioned above is extremely relevant to this discussion about loss of oil pump pressure because it is that which causes the crankshaft failure. By investigating the consequences we found it possible to find the full explanation and find ways to avoid it because the testing and research did show up failures in many of the products and solutions currently available and explains why - together with pictures. The report it much too long to reproduce here but it is free and available by E-mail from [email removed] . It explains how it was crankshaft failure when we changed from the original oiling system to a new more technical automated one on our Cayman S that we hit problems and the painstaking way we went about finding the cause and applied solutions. We expose our own mistakes in doing so - but found such good information we prefer to share that for the benefit of others than keep our own failings over this to ourselves - but it is a touchy subject for us as a result. We basically found that most sump extensions on the market allowed oil to escape in corners and under braking and their only benefit was possibly more oil in there to start with. We found the Accusump automated EPS switching at too low a pressure to respond quickly enough protect the engines if connected to lengthy pipework and exterior oil coolers and implemented a solution It covers the weaknesses in the crankshafts. It was as a result of this we came up with 2 ways to prevent the overhung end of the crankshafts from flexing (one of which can be fitted without stripping the engine) and getting wider big ends shells made from the leading racing supplier exclusively for us - to spread the loads (main bearings done need it). It comes with pictures and sketches but is a long read and not as customer friendly as a pretty video - although I think it contains more information and anyone who works o n these engines and is involved in racing should find it valuable and interesting. If there is a way to make the whole report available accessed directly from this forum I will be happy to consider approving it. Baz Technical Director Hartech Automotive UK

Hi everyone, Weighing in here as a moderator. Time to publicly blow the whistle here and step in. First off, hats off to everyone who has tried to solve the oil pressure issues in the 996. When we work together, collectively solving problems is one of the coolest things that can come out of the RL community. (Trust me…if I’m ever lucky enough to own a 996, I’ll be coming to all of you with questions.) However, multiple people have insulted each other in the process. That’s never okay. Clearly, even this post’s existence has already upset a lot of people. Apparently, this a controversial topic with regards to the data, technologies, and potential solutions to the problems involved. There are clearly a lot of strong opinions here. We must handle this like science, people -- because that's what it is. Whether or not one's data, methodology, analysis, or even technology is correct -- that's not the issue. A Rennlist poster can and should have the ability to post their findings without harassment. It's always okay to question things when it comes to science and tech. Spirited debate over such things is fine. Solving problems together is how progress is made. As there’s an old saying, “the truth will out.” One side certainly doesn’t have to agree with the other. But we have to respect each other in the process - especially here on Rennlist. Again, as a moderator, I'm not taking sides here per this particular issue. Meanwhile, you can always find me on the side of science and truth, let alone people being kind and reasonable to each other. *********** 1) Strictly keep it to the problem solving here. That’s it. No ifs, ands, or buts. Respect the fact that some people are apparently finding the information coming out of this thread as actually helpful. 2) Don’t slam individuals or vendors. Keep it not just civil, but actually go one step further and actually - be kind . We're all in the P-car family here. It has not been fun to have to issue multiple warnings to people on both sides. No more snarky remarks. How we treat each other is ALWAYS more important than any 996 out on the road. Got it? Don’t let there be a second time, people. 3) BE OBJECTIVE . Per our posted guidelines, please review the following: Sponsors are cautioned not to disparage other vendors, or the products/services they may offer. Promoting a Vendor: Praise and a recommendation of a vendor by users is encouraged as a service to the community. On the other hand, excessive promotion of vendor(s), i.e., cheer leading, will be viewed as advertising. In the same vein, systematic and sustained vendor bashing, i.e., carrying out a vendetta against a particular vendor will result in sanctions. No cheerleading. No peanut gallery comments. No slamming. Not providing something useful? Just listen and observe. Don’t agree with the data or methodology involved, and want to speak up? Fine. Do so objectively. Learning how the academics handle this stuff is a great start. Overall? BE OBJECTIVE. That's science for you. (Having been on two published papers before, that's how science is done, people.) 4) If an individual or a business doesn't want to share elements of their process, don’t badger them. Leave it alone. Walk away. Particularly when it comes to vendors and trade secrets, that’s understandably their right. In general, it’s reasonable to ask questions, once and directly. Take “no” for an answer. But in any case, take it off Rennlist. *********** As moderators, it’s not our job to police the underlying science and the technology involved. However, please keep in mind that it takes a great deal of time out of our day to deal with people behaving badly. That’s time that could be better used helping others IRL — people who really need it. Be reasonable. Be objective. Be kind. Now everyone -- go figure out how to make the 996 an even better car. Thanks. PS -- If I'm ever fortunate to meet one of you 996ers IRL, I expect one of you to take me for a spin so I can learn firsthand why you guys love the 996 so. That's awesome.

Foodbiker - we have never knowingly or intentionally indulged in the criticisms you mention and agree 100% with your position on this. I am aware that there can be heated disagreements on forums but we try and avoid it. Could I send you directly a copy of my report and let you judge if you think it is in any way as controversial as you imply from the above comments. If you agree I will just need an E-mail address to send it to. I will be happy to accept your interpretation of it - we don't need to sell anything and don't intentionally criticise others but you cannot get a truthful honest assessment of things without inadvertently exposing issues that some others might be getting wrong. If as you say there are clearly problems with some solutions - it is difficult to know how to point that out to protect owners without at the same time exposing it and if businesses that subscribe to this forum are angry that there mightbe a post that has alternative views to theirs and as a result they are kept from the public - it seems to me that this could prolong the problem rather than solve it. I appreciate it must be a difficult line to draw for Moderators and am happy to allow our report to be excluded if that is what either fits in with your policy or placates those that are complaining - as aggressively as you suggest. We are (and I am) just simply trying to help owners with problems - not interested in any commercial implications, sales or issues otherwise. Baz

@Barry Hart -- Thanks for your thoughtful post above. Apologies for any confusion. My reply was in fact response to multiple incidents over the past few days related to this thread; your post that immediately preceded mine certainly didn't trigger my intervention -- it just happened to post immediately right after yours! (In fact, I had taken great care to spend a great deal of time to carefully word all that -- and had walked away a few times in order to ensure I was being fair and reasonable to all.) You or anyone are always free to send me information; just PM me and I personally am happy to take a quick peek at whatever you want to show me (hopefully I can make some sense of it). However, please keep in mind, as moderators, we're in the business of being referees to ensure discourse remains open, albeit non-disparaging. We're not here to make rulings on whether or not one's vendor's product does what it's supposed to do, is flawed, etc. We are all passionate about our P-cars, let alone solving problems to keep them performing well. In fact, some of these solutions can and may in fact be profitable within a business context. Meanwhile, as moderators -- I can't be clear enough -- we don't moderate the science and technology involved or take sides in that. If anyone wants to make objective, albeit non-accusatory and non-defamatory statements here, that's fine. In research speak, it would be something like: " In our testing, we found that Item X from Vendor Y did not offer the adequate oil pressure protection that was needed. For reason Z, we felt it important to alert the P-car community of our findings." Yes, it's dry as hell, and not everyone writes like a researcher. Meanwhile, it's keeping to the facts. You're not being defamatory or insulting to Vendor Y. You're not trashing Item X. You're just stating your findings and concerns. You're alerting our community that you have concerns about Item X from Vendor Y, because of reason Z. You're clearly stating your objective point of view and getting the word out on such. You're being reasonable, without being either defamatory or insulting. There's always a way to carefully frame one's message to present one's objective opinion and get the word out on one's tech findings. Alternative views about P-cars and the underlying technology are always welcome here on Rennlist. Treating each other badly isn't. Hope this clarification helps. Cheers.

Sorry being 74 I am not as familiar with some on line computer stuff as younger members and cannot see where there is an E-mail address that is not connected to the forum rennlist and therefore please spell out simply what E-mail address I should send on our report that will only be able to be read by you and I will be happy to do so. Baz

Quotes source post: 17455770

Thanks again Golock911. The first graph, change from lines to small dots. That shows it better. The oil pressure dots should group along a single curve that increases with RPM. Also dots that fall below the main curve show where pressure is lost. This is a quick way to see if the system is working well. It obviously is not. The second may have some good info in it. Not sure on that. The third graph is very interesting. Zbomb sent me the data and I looked at this a little differently. There is a strong correlation between drops in temperature and drops in pressure. Send me a PM with your email and I will send you some Motec data that you will like. 200 Channels instead of just the 20 or so you have here. You'll need to download I2 standard from Motec's website. Chris Cervelli Cervelli Technical Service

Baz, First of all, thank you for the white paper on crankshaft failure as well as the whitepaper on engine rebuild. I plan on reading your buyers guide over the weekend. Appreciate your work and conclusions. I have a lot of questions for you, but am not a mechanical engineer, so I will hold off in hope that our experienced engineers on this thread kick it off with a more meaningful Q&A than I could provide.

I haven't gotten much further with the data or the opposite direction analysis. I wanted to back up a step and label the graph relative the turns on the map. It gives a better visual of the relationship between what you see on the graphs and what the turn really was. The next step is to categorize each turn in relation to direction, lat acc, inline acc, rpm, speed, This should start to describe conditions that can create aerated oil. Some trends are already apparent but need another chart(later today) to bring out. For example, a left hand turn requires hard braking to see the pressure drop alot. The right turns do not require hard braking. The fast left turns with no braking have some small pressure drops, but not severe. I guess we need to define severe. Is any pressure drop severe? Here is the time series for session 4. It's just an overview that I probably should've put out first, but hadn't yet learned the program. Better late than never. If oil pooling and scavenge pump age are contributing, all we'll need is this overview chart to see the difference. Stay tuned...

Thanks Paul - I�m going to send you all the Thompson data - it will be the track the scavenge pumps get tested at. Looking like 6/24 will be test day... the parts are due in on Monday and I�ll be on track Tuesday - as much as I�d like to get the scavenge pumps on - just cutting it too close. So, I�ll have Thompson data from earlier in the year (cold ambient) as well as fresh data from 6/1 (ambient looking like mid 70�s and dry) that we can look at and get an idea based on track layout and historical data - what do we hope to see from the new scavenge pumps - besides the obvious, OP is miraculously great which I don�t think many of us expect - but hopefully we will get some idea of improvement or bust. After the 24th - the track schedule (fun testing) 6/27 Thompson, 7/6 Lime Rock, 7/13 Thompson... I have an idea for maybe being able to test enhanced water cooling that will be controllable, on/off, which would allow for testing of the effects of keeping the coolant temps down over a longer period in warmer ambients... that said, I have no idea if it works and need to get a mock up run on the street for feasibility. Then, I could run 6/24 Thompson scavenge and 6/27 Thompson enhanced cooling and not have to thrash parts on the car in between. I give this a 50% chance of working and maybe a little less being able to get the parts I need so, no promises. Paul - I think where you�re going with the above is perfect. I wait on the edge of my seat for you to post data man - you�re a Boss !

Here is a cool open-engine video explaining and comparing the scavenging systems of the Mezger engine (GT3, GT2, turbo), the M96 (Carrera), and the X51 variant of the M96 engine. You Tube

That really illustrates the problem clearly. Looks like it just needs a drain at the front and another pump. Question is, does the X51 upgrade do the job? And is that something that can be fitted easily. Edited On second thought, that�s really to solve the issue of oil pooling upfront under braking. Maybe that starts the problem though. Braking traps a bunch of oil at the front of the motor then we turn and accelerate hiding more oil at the back and there isn�t enough to keep pressure up. Need to test on a round circuit or skid pad to see if you can do high G right and left turns without really braking and keep pressure up

I think it more shows what Chris Cervelli pointed out. There's a LOT of volume inside the M96 heads! It looks like it could hold over a gallon, maybe more.

I've got a good used set sat in storage??

FCP got the parts here today to my surprise... I�ll get the car on stands and the parts installed as soon as I�m off the track Tuesday.

Quotes source post: 17459041

That was a great video. You can see with the proper G Forces that oil can get stuck in the head on the opposite side of the scavenger.

Here are some close ups of left and right turns. Red is pressure, Purple is inline acc. Blue is lateral acc. Coming out of the long straight right going into left turn1, braking starts at the 4 second mark. About 1.5 seconds later, the left turn starts, the big pressure drop starts 2 seconds after braking starts, heavy braking stops at the same time the big drop occurs. Braking time was about 2 seconds. The big pressure drop is about 2 seconds. At 1 second into the pressure drop, the car is now accelerating. The left turn continues for 4 seconds. The big pressure drop gives way to a medium pressure drop(under acceleration) that continues for the rest of the left turn. From this we can say: -When braking heavily and then turning left, it takes 2 seconds for the sump to run dry and start pulling air -The big pressure drop follows the braking profile, it's not known how long it would take pressure to recover if heavy braking continues -The remaining pressure drop through the turn is likely due to the turn itself as the car is accelerating, or other aeration remaining in the oil, limiting the amount of oil getting back to the sump This is left turns 3 & 4 with limited braking, the turns are easily as severe as turn1, but the pressure drops are not close to the same and quite noisy. It clear the left turn itself creates some drops, but it doesn't seem capable of creating the big drops. Here's a right turn with heavy braking. It follows a similar pattern to left turn heavy braking. Braking starts, car is changing from left turn to right turn. Heavy braking stops in 2 seconds. Big pressure drop starts about 2.5 seconds after braking starts and subsides 2 seconds later to a medium pressure drop that remains for the rest of the right turn, 5.5 seconds. From this it seems the it takes the sump a little longer to run dry than the left turn. Geometry of the heads would seem to support this. The G load from braking would push all the oil already in the return channel towards the scavenge pump. Oil not in the channel will get locked above the channel. Here's the right hand with medium braking. There are no right turns with no braking. But this one shows the huge drop that last the length of the turn. Again, the drop starts about 2 second after braking starts. My take away from this is to just not use your brakes! Seriously, heavy braking creates big pressure drops, big right turns create big pressure drops, big left turns create medium pressure drops. I am now with the group that says replacing the scavenger pumps will have a limited effect. What's the line in 'vegas on this? Is there a plan to get a third rad on this car. We also talked about the bigger oil cooler.

Quotes source posts: 17459879, 17312335, 17313469

Thanks you Paul - not using the brakes would also cut down on the pads / rotors consumable cost 3rd rad - I may have an interim solution to bring temps down w/o the 3rd rad. Although my possible "solution" is not what I would consider the preferable way of bringing about temp reductions, it should be less costly than the 3rd rad and thus allow me to get implemented far sooner. Oil Cooler - There is not, that I am aware of a bolt on solution here. Maybe, the 997 cooler which from the thread I link to below is noted to be an upgrade from what is on the 996 - I am not sure if it is the scope of upgrade (BIG) that we are looking for and that will take additional research or input from those experience with the chassis/cooler, I have none with either. MAYBE - the 997 cooler with the additional water cooling is enough.... I just don't know. I guess, if I was upgrading the oil cooler - I would want the largest I could to prolong the point to, or eliminate the heatsoak of the unit when we see oil temps pull away from water. From the thread I started prior to having all this data in-hand on my debunked good plan for an oil cooler HERE https://rennlist.com/forums/996-foru...il-cooler.html - LN makes an adapter plate which can allow plumbing in of a larger Cayenne cooler that they use in a PDK cooling upgrade kit. The thread is a good read with input from Charles @ LN and dgmark who looks to race his 996 competitively in events up to 1.5 HRS long. The issue that concerns me with taking this project on is mounting and plumbing of the remote oil / water exchanger in the 911 chassis that does not have a neat storage space within close proximity like the Cayman in the setup shown below. And here is a quote from Charles from the same thread:

The third radiator for 996 can be done for much less that the kit price. I sourced the radiator, mount and rubber entry duct (as a take-off unit) from Specialized German on eBay at a great price. That left hoses and a few mounting bolts. Install was easy. I cut the hole in my existing plastic bumper support vs buying the 3rd radiator version.

Quotes source post: 17460013

I don't see being into a center rad setup for $400 or less, if I'm wrong - link me up and I will seriously consider it. Thats about half the price of the CSF center rad kit on Pelican and about what it would have to be for me to think a real value off of a new setup with the CSF rad. My view - If I'm going to start tearing into, and throwing money into the cooling system - I'm going to do it once and replace with the best parts I can. CSF center (MIN) and dual core sides, little under 2K TOT if the budget allowed. Maybe this winter. I think the likelihood those changes make a meaningful impact on our pressure problems for a 20 min session to be unlikely WITHOUT an oil cooler upgrade to go along with it. From the data, for the temperature management route to stand a chance of being successful of keeping the pressure drops to - hold you nose, we'll call it not terrible limits - water temps need to be no higher than 200 and oil temps no higher than 230. ETA - Preferably, from a return on investment standpoint. Starting with the upgraded oil cooler that has the capacity to do the job would be better. When the water temps do not allow the oil cooler to perform up to it's potential, you step up the water cooling upgrades - Center rad > upgraded side rads. This is a major reason I would like to be able to get an idea for what sustained lowered coolant temps will do on there own before spending all the money to get there. But like I said, happy to take a look at any links or ideas you guys have.

I'm not sure if the 997 kit can be retrofitted but you can get that for $220 from FCP - https://www.fcpeuro.com/products/por...vD_BwE#fitment Perhaps with proper hoses maybe?

Jason, forgot about the post from DGman and Charles. Very good info. Chris, Baz and Jake have provided their thoughts on solutions as to how they build engines. Different approaches but the end goal is the same. Charles and many of the other vendors provide already manufactured bolt on solutions, some of which need modifications, but are already in use in PCA Spec 996 club racing. DGman is a good example. There are a boatload of 996 Spec Racers that we can learn from as to what they did, why they did it and how it worked. If any of you know of any, let's get them involved.

Radiator assembly (complete) was $160.50. 4 mounting bolts were $10 or so (factory bolts). Hoses were about $60 shipped. Total $230 or so. So yes, under $400... -Tom

Something to think about... Many have undoubtedly come to the conclusion that the scavengers aren't the answer nor a viable solution. In addition, it has become very clear that we have to figure out how the 996 Spec racers have addressed the oil sump issue during braking and heavy turning. Lastly, we have to address the cooling issue of which we have many possible solutions, but none for certain. My suggestion is that Jason returns the scavengers and gets the money back, to use the money for a viable next step that doesn't waste the money.

Quotes source post: 17459041

Great video. It�s easy to see where there could be issues draining oil from the heads on the stock m96

Quotes source post: 17460452

Isn�t there a cup cars subgroup/subforum? I asked once about cooling on the 996 cup car, and learned that those cars don�t even have fans on the front radiators!

Quotes source post: 17460557

Spec 996 racers have not figured this out, or contributed any knowledge or data. Nor have Spec Cayman racers, and nor have Spec Boxster racers. They all rely on luck and accept (or perhaps deny) that engine failures are inevitable. Chris Cervelli Cervelli Technical Service

Quotes source post: 17460637

we don't want "cup" we want "spec" 996. Cup runs mezger. Spec runs our engine.

Quotes source post: 17460639

No question Chris, engine failures are inevitable. Completely agree with you, there seems to be no "standard" solution. Hence the attempt by this thread to aggregate some info and test them. But I would argue that the 996 spec guys are attempting in some way, successfully or not, to at least prolong engine life, or their would be no 996 spec class. DGman's post was a good example. Your insights have been invaluable. Please keep them coming. Challenging topic to say the least.

To all. If you have never had the chance to attend a PCA Club Race, I highly recommend that you do. It is free, and it's more fun than any middle aged man is allowed to have over a weekend. The reason you want to attend, is that you will see every class of racing, and realize that with the exception of Cup Cars, turbos and GT3s, the 996's are running our engines under heavy load. As you watch it, ask yourself the question, "how do they keep the engine running knowing damn well they have the same issues we have" . We are looking for that answer. Also, you will notice that every regional and in some cases national Porsche racing/indy shop is in attendance supporting their clients. This is a good opportunity to get to know great guys like Chris at CTS to ask questions and develop relationships. PCA racing is a whole different world than what the 996 forum typically experiences. However, they have the same issues as us. Edit Add: Most PCA Club Races also include DE in between the races for those that have an interest.

I think what you'll find is the spares / repairs / maint budget for those invested in competitive racing are not analogous to a DE guy. When you total up the expenses of running competitively, a host of alternatives may be viable when compared to the cost of a DNF. Annual rebuilds, spare engines etc... There is a reason why we are not all running Motec ECU's, Moton 3 ways, truck and trailer and a new set of slicks every track day - and that makes this problem particularly difficult to solve. I keep going back to if there was a market for the solution, a solution we would have. If the market is out there and laying dormant, an awakening from the slumber before I have a bunch of data and a window in the case would be great for me.

Quotes source post: 17460732

Agreed. The PCA Club racers by nature are a highly competitive group of people. Type B personalities do not get into a car at speed and go wheel to wheel. Further, because they are type A, they are looking for every edge, which means durability to finish the race. As Chris said, there is not a lot of sharing. It is understood why. Club racers and their Porsche Indy want to win. But the solutions are out there.

I'm rather busy at the moment but I have quickly read thru this thread since I feel I'm a bit responsible for the scavenge pump idea. Those of you proposing lowering the oil temperature(directly via oil cooler or indirectly via coolant) as a mean of reducing the pressure drops, how do you explain that except the fact that the data seams to correlate? Please explain what physically changes in the engine when the oil gets hot which in turn causes the pressure drops? I see the majority treat this problem as a "black box" problem but in reality we know rather well what the box looks like inside. I totally agree (for other reasons) that the oil should be keept cooler than what Zbomb sees on the track. Without modifying the car I see three ways(which haven't been discussed) to cool the engine: 1. Run Less coolant in your coolant mix and more distilled water. Read on your bottle what is the thinnest(the least coolant) mix you can use. Distilled water is better than coolant at transferring heat. 2. Run thinner oil (but not soo thin that you can't achieve 10psi/1k rpm). This will run oil faster thru the oil cooler(which is a standard plate heat exchanger, google to learn more) at high RPMs and this will cool better, that's just how they work. I understand this is VERY controversial. For reference, I'm currently running the standard A40 M1 0w40 ( 12.9 cst @ 100c) and this is too thick, I'm not even close to 10psi/1k rpm even after 7000km on the oil. 3. Run with full oil. Zbomb, why are you only running 1/4 full now? And, something that just popped up in my head. Maybe one way to get around the pooling during hard braking could be to idle the engine instead of being in gear, thus pumping less oil into the heads.

In these PCA events, how much modification from stock is allowed? Sure slicks and and these other things help, but the fancy trailers and such don�t solve this problem with temp, pressure and volume.

Video from @roadie13 was great! I found the following historical info (2008) at https://www.renntech.org/topic/24037...comment-125528 useful, but confirms NLA status. Looks like the X51 return is a good solution for heavy braking and left hand turns, but does it address right hand turns? Are these or a similar system installed on M96 �track� M96 rebuild configurations advertised by engine builders such as Flat6 or Hartech? FWIW, more experienced drivers use brakes less than less-experienced drivers, so that may be a factor in this discussion. Would be interesting to see oil pressure data on same car with experienced racing driver (no intent to critique abilities of anyone here). @Nick_L, read the white paper �Hartech Engine Rebuilds (Interim)� from Hartech.org, it explains how piston geometry and oil viscosity cause bore scoring.

Quotes source post: 17460872

Nick, I am responsible for making the suggestion. It's up to the Go-Fund-Me group and Jason to decide. With that being said, you may want to take the time to read all of Chris (CTS) posts on this thread as well as Baz (Hartech) posts and his whitepapers on engine rebuild (2012) and Cranshaft (2017). These guys address the issues beyond the scavengers. To your point and their points, there are a lot of ways to skin a cat. In this case, they have huge amounts of time spent in systematically addressing the root causes and then finding solutions. Chris and Baz hopefully will address your questions directly. Appreciate the post.

Quotes source post: 17460872

While none of these are inherently stupid ideas I can comment with authority on the use of Mobil 1 on the track. Don't do it. Chris Cervelli Cervelli Technical Service

Quotes source post: 17460880

Can't answer the up to date regulations as to mods. But you can find it on PCA, website. They produce the data every year. But as you can see, there is ample room to address cooling and oil starvation beyond trailers and Hoosiers. DGMan is a good example on this post. Jason has had many Spec 996 club racers respond to his posts with insight.

Quotes source post: 17460890

I hate to always be the guy saying this but no actual information about the efficacy of the X51 scavenge pump was presented in the video. The narrator made some claims and offered no data to back the claims up. It's just a guy talking. Regarding the faster vs slower driver theory: I have tested this to a certain extent. I co drove an enduro with my girlfriend many years ago when I was still running a wet sump. Her oil pressure data looked only slightly less terrible than mine, despite running about 4 seconds slower than me. There were some unusual circumstances with the car (related to doing insane things with the oil system) in the race, and normally her delta to me is less. That said, the faster driver is almost always going to have a higher average RPM and a higher final oil temp (assuming maximum effort) either of which could push the oil pressure situation into engine failure. Chris Cervelli Cervelli Technical Service

Quotes source posts: 17460872, 17460872, 17460891, 17460920

Nick - Read what you wrote... The reality is - that data does correlate that when the oil temps are lower - the pressure drops are not as severe. Simple as that. regardless of explanation - in multiple different data collection events, its not anomalous. To answer #3. 1/4 up dipstick WITH 2QT deep sump to keep AOS happy. Have run 1/2 and some of these logs probably reflect closer to that - the point is, it is ALWAYS under 1/2 on the dip stick. 11 QTS of oil - just an aside - anyone else find it a little funny, but not funny that a 3.6L engine with 11 + QTS of oil could have oil starvation issues.... 11 + QTS ! My opinion is we should test the parts. I don't think they are going to do anything, but I do think that testing and publishing the data is good for the community and this should at least provide data driven clarity to a claim which has been somewhat vigorously debated at times that worn factory NON X51 scavenge pumps play a role in the G induced pressure drops. I think anyone who really wants to take a stand on the validity of a claim be ready to back it up with data. Frankly, I wish Nick had done this already and then it's likely we wouldn't have to but I can understand the fact that data collection is not something everyone is interested in doing. Regardless - I think where claims can be tested - there is value in doing so for the community looking into these issues tomorrow. I am not fast - I am in my 2nd season ever driving a car on track and have around 10 events TOT driving experience, if I'm having these issues, people should really be asking themselves if they are as well. As I continue to get hopefully faster, maybe we can see deltas. For instance - 1st set of Palmer data CC best lap of 1:56 - 2nd set best 1:53... With the difference in ambients, maybe the data is too messy... I'll let Paul decide if he wants to run those to ground. I don't really think it matters much... the point I'm at is pushing the car to where it's at - anyone I trust to review the data on the mechanical side have ALL said the same thing, and this would include people that are not M96 specialist but specialist in Engine building - this motor is not long for this earth with the instability - The lows are low enough with enough load to be causing damage that would not surprise anyone to cause a catastrophic failure... So - I either get it fixed, or I don't - I think the idea that data collection is important for people to not only monitor the health of the machine, but also the vehicle / driver dynamic is something people that really enjoy using these cars should take a hard look at. If we had 20 cars running collection with the different parts combinations that would entail, what would we know ? It's the data that allowed this conversation to get to where its at and got the eyes on it it does. And it's the data which can give people some confidence of validity if and when a fix or progress is claimed to have been found. Don't take less.

Quotes source post: 17460895

Chris, your also the guy who told me that the scavenge pumps doesn't wear, it's a waste of time and money(among other odd statements)...which I proved to be wrong in the other thread. I'm a professional engineer, you need to try a bit harder to convince me. The SINGLE MOST IMPORTANT part in the head oiling system, to avoid pooling, is the scavenge pumps. After that it's the chain tensioners, they may leak significant amounts of oil adding to the pooling. In the other thread I've done the calculations showing that the pumps are not over-size and you WILL have problems when they wear. This is a common theme I found in the engine, very few things are over-size, the oil cooler is another of these things. In my other thread I wrote the measurments from my worn out pumps and also added pictures. I'm a bit surprised Zbomb(and no one else) suggested to first measure his current pumps before ordering new. That should be a first step. Cost you a tootbrush, some cleaner and 1,5h of work. In the other thread I also mentioned that I needed full oil to completely get rid of the pressure drops. This indicates that pooling is happening but there is enough oilbuffer in the sump to account for it. The other thread: https://rennlist.com/forums/996-foru...n-corners.html Edit: Zbomb, my data is in the other thread. The data consists of pictures, calculations and measurements of the pump including my test drive observations. I used a GoPro to film the pressure gauge(sorry, the footage is gone) and since I did the same test drive every time I knew which corners to observe it. I don't feel the need for a data logging system.

Quotes source post: 17461569

Nick - I'm running full oil from a factory volume standpoint. Actually, about a QT OVER factory fill requirements. For me, a GoPro on the gauge is not sufficient even if the footage was available for what we are trying to accomplish. Unless it showed absolutely no pressure drop, I would not trust the factory gauge to relay the scale of the problem. Regardless, we'll have data from my car that show what the new scavenge pumps do vs. the 85K pumps currently on the car soon. I hope they work ! But am planning for them to not.

Yes, I see your data, I don't question that (since it's what I saw also) . Read my second question, how do you explain that? How is the temperature affecting the pressure drops severity? If we assume the pressure drops are caused by the pick up sucking air, the submerged depth of the pick up is the interesting part, not the actual volume. Do you know if you have more/less than a standard sump filled full with oil? edit: It would be helpful if you could disassemble your old pumps, take some pictures and measure them.

Quotes source posts: 17461688, 17461688

I'd leave the explanation to someone smarter than me - if I were to theorize - cooler oil more resistant to movement in sump. The oil pump pickup spacer that comes with the LN kit moves the pickup in-line with the sump spacer.

Here are some catch up plots. First is the XY plots, oil temp vs. RPM colored by throttle position lap1 and lap8. The best case would be to see all the points in a line along the top edge of the traces. The lap1 top line is a bit of a mess because the oil temp is changing throughout the lap. Not sure that throttle position coloring tells us much. Chris had this same graph in a post 4.5 years ago. He was showing the effect of the dry sump he'd developed. The dry sump creates the line described. I'll go find it again and post it up. Here's a different way to look at the effect of G's on oil pressure. It's Oil Pressure vs. Lateral Acceleration colored with Inline Acceleration. It's an easier way to see that there are significant pressure drops on both right and left hand turns. The inline acceleration color of Blue is when heavy braking is occurring. Previous graph show that pressure drops due to braking start 2+ seconds after braking starts, at which point the braking has already stopped. So, don't expect to see InlineAcc Blue color in any of the pressure drops. Blah, blah, blah. Here is the overview graph from the CW run at Palmer in April. Indeed it shows the opposite turns from what we've been studying. Here is the detail graph showing the accelerations and braking/accelerating. I specifically picked a lap where braking was not as aggressive. By reducing the pressure loss from braking, we can clearly see that the right handers have much more pressure drop than the left handers. So now there are 2 ways to keep pressure from dropping: no braking and no turning. Before you know it, you'll be driving a '60's American muscle car!

Not had time to read or consider all the above posts and graphs - but look guys the loss of oil pressure is simply caused by the forces on the oil in the wide pan lowering the sump level below that of the pick up. An Accusump is IMHO the best solution as long as it is not triggered by a pressure switch but runs live and as long as the return goes directly to the oil pump housing. Cylinder head trapped oil we return via a fast flow electric pump connected via a relay tot he brake light switch. All works perfectly. More technical explanations available entirely FOC and (and free of any subsequent contact) by E-mail by requesting from [email removed] The topic is entitled "Crankshaft failures explained" - but is exactly about oil pump pressure loss in corners and braking zones (especially right handers). Frankly it doesn't matter what digital records and graphs you may have or follow - oil pressure loss occurs tracking with slicks and fast drivers and can be eliminated by practical means when it occurs. The biggest enemy is time - you only have a few milliseconds to recover oil pressure to avoid wear and eventual failure of the crankshaft bearings. You need a deep sump that doesn't simply allow the oil to slosh out again in a corner or braking zone and a recovery delivery system that works instantly and at full system pressure. The report explains mistakes we made and solutions we found and could allow others to avoid similar misfortunes (sometimes getting it wrong and finding out or working out why is more beneficial than theoretical opinions and arguments). I trust that anyone who has read our report will agree it is neither contentious not a blatant sales document (as it really doesn't include things we are trying to sell). It is intended to simply provide a different way of looking at the problems that result from the issues above and we are not bothered what the outcome of reading it is - good or bad - it is always better to follow other alternative opinions and views than get locked into arguments about specific issues and also valuable to read about solutions that were tried and failed (and why) and those that eventually worked (and why). Baz Direct feed into the oil pump housing like this from an Accusump - should eliminate your oil pressure loss problems. Not being offered for general sale - just examples.

Quotes source posts: 17463179, 17463179, 17463201, 17463201

Here's the CTS data (from 2017) - First one is wet sump... Funny - From the looks of it we've got 4 years of catching up to do. And this one from the dry sump system I really like the graph above showing the G influence and think it can provide a really good basis for comparison for viewing any possible improvements. I am curious how much track layout may impact that graph though. The papers were a great read Baz. Understood on the thread having a lot of input - much of which from the papers, you have already worked through. One thing though I do feel as though I should correct: I'm running 100TW tire, not slicks, and more importantly, I'm not an experienced (fast) driver - the car is pretty good, but I'm a beginner. From the data above - when I start approaching 1 lateral G, pressure falls off - I think a lot of folks can even just review their GoPro track footage with a G meter and see that even with 240 TW tires, those are very approachable limits and they should be concerned about what kind or pressure instability they are experiencing. I think the: Accusump direct to oil pump A sump with sufficient oil control Are great paths to be pursued further - as I don't believe either of these solutions are attainable to the market of 996 owners other than on a custom 1 off basis.

I can make it happen with all season tires on my car.

Quotes source post: 17463566

I owe everyone some street data - this weekend has been constant rain and 45 degrees, so no good data until weather normalizes. It's coming though.

Quotes source post: 17463475

And the answer is... Yes, track layout has a significant effect. Here is Palmer CW from the Session 4 Lap 4 shown above. Very few drops on left turns.

Quotes source post: 17463810

Although the dispersion is weighted more it still paints the picture. When compared to the data from CTS, it clearly shows what we are looking to achieve and the visual of tightening that grouping (Gun guys should be smiling) the goal. Below is a video from my last day at Thompson - much as before, cool ambients in the 40's. I think you probably have this data Paul from the first dump I sent over. I'll be back there tomorrow, temps call for the high 70's / low 80's. Again, I expect the data to look pretty ugly. You Tube

Thanks for your input, I can only reference tracks and events in the UK of course and track surfaces, lengths of tracks, shapes and grip in corners and their radius's may all the slightly different in the USA and of course my comments are about cars over here that have the sumps that we generally fit etc. However the pump has the capacity to deliver much more volume than the engine ever needs over say 3,500 rpm - a rev limit that over here racing is always revving the engine above - and so it is the pressure relief valve that controls the delivery pressure which is around 7 bar (reducing as oit passes through the engine by about 2 bar by the time it reaches the oil pressure sender in the cylinder head due to bleed off rates through all the bearings, tensioners, tappets etc). If the oil level in the sump is high enough for the pick up to reach it your oil pressure will be OK. If centrifugal or braking forces throw the oil forwards or sideways so the level drops below the pickup - you will have oil pressure instability. Oil does collect mainly in the RHS head under braking and this will of course contribute to the general dropping of oil levels in the sump. My point is that it makes no difference to the outcome what analysis you apply to the data from the loggers - unless you are going to change how fast you corner or change the circuit shape. If you want to go as fast as your ability and car will go and you are losing oil pressure then the solution does not lay in analysing the data but making a permanent physical mechanical change that protects the oil pressure. Unfortunately some racing sumps allow as much oil to escape from the central level of the pick up as a std sump - they are just deeper and hold more oil but so not help prevent the flow away from the pick up. Some systems using an Accusump feature a switch that opens the Accusump to recover oil pressure at too low a pressure - you would not expect to race with 3.5 bar oil pressure - so a switch that only opens at that pressure is wasting precious time. You need a sump that will retain as much oil in the centre under the pick up as possible, and Accusump delivery that responds to a loss below 7 bar immediately and a return system from the front of the RHS cylinder head. fix that and the oil pressure will not drop and you will not need to look at the data graphs any more. I feel the emphasis has gone away from the root cause towards technology and missed the obvious causes and therefore looked at the wrong solutions. It doesn't effect me personally what any of you in the USA decide to do about it but as I understand this so clearly and have gone through finding solutions, made mistakes and finally found reliable solutions that work (and fit with all the science etc and by the way we also do use data recorders but understand they wont change a physical problem - only reveal it and if it has been fixed or not) that I feel an obligation to point out what we discovered is important and what does and does not work - simply to help you guys make progress in finding answers.. Good luck over there. Baz

Hi Baz, Thank you very much for your insight. Would it be asking too much to ask for an opinion on the X-51 changes to the return of oil from Bank #2 to a dual oil pump? Cost-wise, it seems the Accusump upgrade would make a difference as explained in your text, despite other options that do not apparently take into account the activation system you describe. If possible, can you make available the design/machining required to adapt the oil pump to the Accusump system, or is that part available commercially? Given the discussion in this forum, do your race m96 engines also receive an upgraded oil cooler?

Every crossed your mind why they put such a big pump there, that's not very Porsche-like to waste energy just circulating oil thru the by-pass. Up until the M96, Porsche engines was air, but more importantly, also oil cooled. The designer of the M96 would most likely also been involved in the previous air cooled engine and draw inspiration/experience from these. Oil is not just for lubrication it's also very much for cooling. I would say that we are running waaayy too high pressures rendering the oil cooling part of the M96 very suboptimal. That's why we see coolant having OK temp but the oil running hot. Edit: Baz, do you know more exactly how much it pumps?

Quotes source post: 17464830

Hi Baz - The analysis from the logger can give us a very clear picture the circumstances with which the issue occurs as well as the severity of the occurrence. Equally important, it tells us when the issue DOES NOT occur. Due to no currently available market available solutions, it is likely that we will need to be able to asses incremental improvements, the target areas for incremental improvement that may have a positive benefit to oil pressure stability should be data driven. It has been well believed that the current market solutions here in the US were adequate, the data has been required to highlight, that is not always the case. Lastly - The data provides us with a basis for comparison to weigh any potential or claimed solution to this problem. While reading your paper - I was struck by the omission of data, particularly in regard to the response time of the accusump. As you touched on the importance of the the near instantaneous reaction time required, it did make me wonder - What is it that you found to be good enough ? With the data logs as presented, I believe the trigger event as well as the response time and adequacy would be clear. Also, as the ability for the accusump to recharge and be ready for the next trigger event is something that has been discussed previously in the thread - as we can see from the data in my case - we would be triggering the accusump on multiple occasions per lap - have you found instances where the accusump could not keep up with demand or that the recharging of the accusump caused any instabilities in oil pressure to the engine ? I have never used one of these systems before so please excuse if my question looks to be from someone unexperienced in the use of an accusump (it would be true ) ETA... I think when we look at the root cause - oil not where it needs to be - the solutions have been much oversimplified to the difficulty of the problem, a deep sump will fix that, and people in the US have by and large held that belief to be true - It passes the sniff test and has been recommended by trusted sources, it was not until technology came in that me, a hobbyist (thanks Chris) was able to show that maybe those claims need another look. Maybe the engineering of the current systems needs a closer look for efficacy in the intended role. I did not know your paper existed prior to this thread. In the US currently - My opinion, there is NO commercially available solution, yet alone solution(s) to the keep the oil where it needs to be problem. This is not a technology problem - it's a market problem - the solutions exist, they're just not available as off the shelf, attainable to the general customer option.

As I started to read info on the accumsump site, I was wondering where to put it, how much volume, and how to valve/control it. Bunch of blog material there to pour through. I�m waiting for a copy of the crankshaft failure paper to read along with the rest of the group. At least you have an idea of where to plumb it to.

I think we are not getting much further until Zbomb have tested the pumps, that data will speak for itself. Zbomb, I suggest you change the pumps and nothing else(to be sure we isolate what is tested), and run a few first laps. After some laps, check your accumulator(s) for oil. I predict there will be much less if any in them. At this point you will probably feel more comfortable to start adding oil until it's full. New pumps and full oil is what did it for me. And we will se how that works.

Here are some overviews and details from the Thompson run yesterday and some overlays with the previous run in April. Executive summary, 10 degrees hotter, similar lateral acceleration, lower overall RPM's , faster through the turns , bigger pressure drops This is the overview for session 1. I didn't look at session 2 or 3 yet. It looks similar to the other overviews. Oil is a little above 250 at the max. Here is a comparison of 1st session lap 7's between yesterday and Aprils runs. Red is the current run. Temps on the bottom show a 10 degree hotter engine. Lateral acceleration is similar, but pressure drops are much more pronounced. Since G's are about the same, it's either due to temp difference, oil level, or engine wear is opening up and not holding the pressure. i.e. flow is higher & sump runs out of oil faster. The graph is distance based for the comparison Here's the fun graph. Same lap, s1l7, in time. Red trace is current, brown is April. Speed is maybe a little higher, RPM's are overall lower, G's are about the same, but total time is 3 seconds faster. It means you're on a better line. Much more efficient use of the car. Awesome work! And it's not just one turn. Each one is incrementally better. Really nice. So the elephant in the room is why the engine puked out all that oil. Is it only blow-by? How does blow-by push oil into the AOS? Whatever it is, it seems clear that the collection tank and UAOS saved the engine to race another day.

Time to try a 5-50 oil on track, methinks.

I have one of these high mileage/ Extreme Track Duty Auto-Drain Banjo fittings and plumbing in transit to Zbomb to take care of the added oil load he is getting. I only have two of these, I had ordered 20 sets from Summit 2 weeks ago, but they were backordered/ drop shipped with no ETA, so I finally canceled the order and ordered from another source, hopefully I can get them. I have all the seals,washers, clamps,3/8 hose, pressure fitting, brass connectors, but the Banjo is the main part and is giving the trouble being sourced. They will be available on the UAOS website for $75 for all the fittings necessary, or $175 with a new oil fill pipe pre-fit with pressure fitting/connector, once I have secured enough in stock.

1. Ran a PB at Thompson 1:22.6 2. Caught 3/4 of a QT of oil in the catch can - went from a couple of ounces at Palmer... Both times I have run Thompson this year I have had this issue. 3. I saw 1.4-1.5 peak lat G and 1.1-1.2 peak braking G, I was getting a lot more out of the car than in previous outings. 4. Car was making a new noise under cornering loads, I did not notice it real time in the car but later under video review. I will get the car in the air in the next couple days and check what I can mechanically before starting the install of the scavenge pumps. Car ran and sounded fine on the way home from the track, the filter is not full of metal and looked normal. I'm starting to put money aside for the inevitable...

Quotes source post: 17468617

You should take a really close look at your wheel bearings, they take a serious beating on the track. Look for even the slightest bit of play or roughness. I have replaced many on "track cars". The Gt3's with center lock wheels have a "track maintenance schedule" includes replacing wheel bearings at 4k track miles, center lock hubs at 8k track miles, and complete carrier at 16k track miles..

Quotes source post: 17468657

#1 on the list Skip - Brakes got hot - like really hot.

Taking almost 1qt of oil out of circulation is worrisome. Why is it collecting so much from the UAOS?

Quotes source post: 17467996

Skip, let me know when you have them in stock. I will buy instantly.

Quotes source posts: 17468691, 17468709

Well it won't be with the high mileage/Extreme Track Duty Auto-drain on it.. My opinion is that there is more oil pooling than normal or more blow-by than normal, or more of both. Will do Kris !!

Quotes source post: 17468617

Our 996 increasingly had a sound/vibration like tires on rough pavement when turning right or left only with power applied, changed differential fluid with standard 75W-90 noise gone.

Quotes source post: 17468818

I agree, there's more RPM and more hot oil leading to more oil vapor. Combined with the connection from the crankcase to the AOS likely getting submerged (Z is pulling 1.5 G in the corners!) which may let slugs of oil directly into the AOS. The UAOS has certainly prevented hydrolocking. It's a fantastic invention.

+1

I find it incredible that you guys seem to think analysing the data will produce an answer - it's simple - there are times when the oil is not near the pick up tube and the oil pressure drops. It wouldn't matter to me where it took place or how I would still want to eliminate it and the only way is to stop the oil leaving the sump area where it is sucked into the pump or apply another feed under pressure immediately the oil pressure starts to drop (and not too late). If you want to read about our solutions request a copy from [email removed] and we will make future batches to cover orders for our valve'd deep sump (which has an internal captive area that oil is fed into in corners) (we just don't carry a lot of stock) and anyway the Accusump we recommend is available in the USA and we can provide a suitable adaptor for it to feed into the engine now (in stock) for all but the Cayman S which is due soon. but I wasn't trying to make this a slaes exercise - just stop a lot of people wasting a lot of time discussing temperatures and lfow areas etc when it will not find a solution to the problem. Baz

EDIT - email sent to [email removed]

Quotes source post: 17470042

I think a lot of folks here and on the 997 forum would be interested in a worked-out accusump set-up along the lines described in your report. Same for your sculpted racing deep sump. I just haven't seen such these things advertised for sale on either side of the Atlantic....

Baz, I find it amazing that you have a product (valved deep sump ) that picks up the oil better than the factory or available deep sumps that none of us know about. Please post pics/prices/ ect. This problem has been discussed for years/decades and in depth in this and many other posts, with no mention of your product. Me and another partner in fact have been working on designing a more efficient sump system for quite some time, many many months actually, but were hesitant to mention it yet, due to the design is yet completed and didn't want the excess chatter/descent involved when introducing a new product. It's quite clear there is a need , and the reason we have been working on this, but we have have infact been wasting our time, but we had no knowledge of this, and I am very aware of things that are available,( and have read your indepth articles and the 4 main causes for m96 failures) , but don't remember that product mentioned or have seen it on your website. Cheers !! Skip

Quotes source post: 17461688

I concur with Nick that it is relevant/necessary to know what leads to the observed behaviour in order to find an effective solution (that baz and jake already know as I understand from their posts). I have the following hypothesis as an answer to above question: Cooler oil has higher viscosity than hotter and as such the oil pressure relief valve opens at lower rpm (say at x rpm) and thus less pressurised oil flows into the engine than is the case with hotter oil. The bypassed oil is dumped back into the sump right next to the oil pickup and is of greater volume at lower oil temps than at higher as with the hotter oil the relief valve only opens at a higher rpm (of x + delta rpm). This greater volume (of delta rpm * cc of oil/rev [edit: per revolution, not per rpm as written originally and quoted below) will still cover the pickup under higher lat. and long. acceleration than a lesser volume would. Add to that the fact as mentioned before that the cooler oil with its higher viscosity tends to move around slower/'less' in the sump than hotter oil, then it is obvious why fewer and less severe pressure drops are observed. In my thinking Nick's observation that new (vs worn) scavenge pumps manage to pump enough oil back into the sump even at higher oil temps fits this hypothesis.

Quotes source post: 17470317

A couple of problems with your (and Nicks) hypothesis. First and foremost, the oil pump by-pass does not dump back into the sump, it just by-passes back to the inlet side of the pump. I do agree that if the scavenge pumps are not 100% doing their job, that there will be excess oil pooling in the heads that can/will lower the oil available in the sump and also cause more oil to be pushed into the UAOS/AOS vent system. Hence the reason I suggested the oil pump replacement on Zbomb's car (and started the go-fund-me account and contributed). The proof will be in the data if the oil pumps are not 100% doing thier job. BTW fixing to donate another $25 to fund the research, someone in Texas bought a UAOS yesterday..

Thanks for setting straight to where the by-passed oil goes. It going back into the intake is even better than being dumped back in the sump near the intake. My point is that with cooler oil less is pumped into the engine and more remains in the sump covering the intake snorkel than is the case with hot oil. Therefore I don't see my hypothesis disproven. You wrote problemS (plural), which other problem(s) do you know/think of that would disprove my hypothesis? Perhaps some concerning aeration of the oil?

Quotes source post: 17470433

I has been stated (wrongly) that the main oil pump by-pass dumps back into the sump(1) and that it dumping back into the sump causes excess aeration (2), that's why I said "couple of problems".

Thanks for your reply. I'll rephrase my hypothesis to reflect your correction wrt the by-passed oil: Cooler oil has higher viscosity than hotter and as such the oil pressure relief valve opens at lower rpm (say at x rpm) and thus less pressurised oil flows into the engine than is the case with hotter oil. The bypassed oil is dumped back into the sump right next to the oil pickup circulated back into the intake side of the pump and is of greater volume at lower oil temps than at higher as with the hotter oil the relief valve only opens at a higher rpm (of x + delta rpm). This greater volume (of delta rpm * cc of oil/rev) of cooler oil must not be sucked out of the sump and as such remains in there and will still cover the pickup under higher lat. and long. acceleration than a lesser volume would. Do you concur with the above?

Quotes source post: 17470630

Yes, I agree, and is also the reason I would not suggest an oil with a lower viscosity (5w30 that has been suggested), the lower viscosity/thinner oil would make the problem worse.

Thoroughly fed up with this now - got up at 4.30 this morning and wrote a long detailed reply to all current questions for an hour only to be blocked out and lost it all when proceeding to add a photo of our sump. Too much time wasted to write it all again now. SO if you want to find out more about the pumps (how they work) our sumps and other products (and how they solve the problems you describe) Please request our report on crankshaft failures from [email removed] which covers all the questions and includes pictures etc. I haven't even got anyone awake to take it out on so the guys at work will have to watch out this morning as I am not in the best of moods today. Baz.

Thank you Barry for your continuing thoughtful contributions to this thread. Your frustration with the discussion is quite clear. I would expect this is a problem you solved 15+ years ago. These cars are old enough that a new group of people are buying them and learning about the problems. The pros that figured it out and provided fixes found only a limited market for their products and therefore haven't marketed them. Knowledge of the existence of those products passed with the last group of owners. So the current eclectic group of 996 owners tries to figure it out. I expect we'll be at it for a while yet, even when we know the solution. Probably until we have your pump or a system from CTS installed on someone's car with new data published. The data from CTS website is pretty clear that these pumps works. The data from Jason's car has become quite predictable. The only thing left to learn from it is how to make him a better driver, which he's doing just fine at without the data. OK, alot of words. Thanks again for your patience with our learning curve.

I feel for you, Baz! After having experienced similar I now use notes or a similar app to write a reply and then copy/paste it into rennlist.

I usually do that too hardtailer - but getting up that early I just launched into it - to my regret. Porschetec 3 please request our report on crankshaft failure from [email removed] . There are two and the right one includes details of the deep valve'd sump etc - it is entitled Crankshaft failures Final report May 2021 - I am sure you will find it very interesting.Thanks for that golock 911 - it makes sense. I will have another go. One of the problems with providing information via rennlist is that when I have tried in the past to describe the technical benefits of products that will help people in America they�ve been blocked from the forum or I�ve been asked to pay a fee for the privilege of providing technical information free from the UK because otherwise it upsets other people in America who apparently pay for the same privilege - which leaves me open to question how independent and unbiased some of the technical information is and perhaps explains why we find so much of it misleading and often incorrect. You are quite right that we have a range of racing products that have solved all the problems in the UK for over 15 years and we find they�re not necessary for people doing track days on treaded tires in our short-circuits in the UK but are necessary for front running race cars running on slicks. The biggest problem is that most deep sums require holes for the oil to drain back into the sump yet those same holes then let the oil you are trying to preserve � to escape in a corner or under breaking and consequently although they hold more oil they still allow the oil level to drop eventually and cause fluctuating oil pressure. There isn�t enough time with 120 odd revs per second to allow any oil pressure loss whatsoever or the crankshaft bearings will suffer. That speed is so fast that it can defeat the human eye and take ages to find on high-speed data loggers � so the outcome usually simply is a damaged engine. Our solutions were work well because our deep sump has 3 chambers and the outer two incorporate flap valves that keep the oil in the outer chamber in a corner while the same sideways forces direct the oil in the inner chamber back into the central one where the oil pick up is. So as well is preventing the oil from escaping from one side of the sump it directs the oil on the other side that is trying to move across the sump directly into a central area where the pick up is whether you are going round a right hand or left hand corner. As a result the oil level is preserved for as long as possible although some does still escape under braking because it�s much more difficult to find a solution to. That�s why we also use an Accusump but that also can have problems because if it is automatically operated with a valve that relies on pressure drops the ones available are rated at too low an oil pressure and that millisecond delay can still damage the crankshaft so it�s better to use the system which keeps the Acusump permanently connected to the engine while the car is on track. If you use too large an Accusump it can result in problems with the sump overfilling on return to the pits because the pressure has lowered and the oil level can rise so high it overcomes the air oil separator diaphragm and allows oil into the inlet system. It�s better to use the system which keeps the Accusump permanently connected to the engine while the car is on track. It can be quite complicated to work out a way to achieve both objectives. We had everything nicely sorted out for our Boxster�s 996�s and 997�s until we decided to race the Cayman S where we couldn�t fit the adapter plate under the oil cooler for the Accusump so we had to find either another way to connect oil or another entry point and we made some mistakes that caused crankshaft failure. We finally realised the best place to deliver the oil is directly to the oil pump gears and the benefit of that is because we needed something to recover system pressure the millisecond the pressure drops and this way the oil is delivered back to the pump which then restores system pressure immediately and this is better than the oil from the Accusump going into the engine system somewhere else because that still leaves the oil pump pumping air behind it and the oil going into the system can still travel backwards towards the pump where the centre of the oil pump then delivers it back into the sump until the oil level is high enough to start pumping again. Feeding the Accusump oil directly to the pump means you never experience a drop in oil pressure throughout the whole time you are on track because the instant that the system oil pressure drops below that previously held in the Accusump � oil flows into the pump itself. Don�t forget that as oil leaves the Accusump the internal pressure it is being delivered at drops � so there is a short delay when oil pressure is returned in both restoring pressure and re-filling the Accusump. Feeding oil directly to the pump � it doesn�t matter if the pressure is dropping as it does so because the pump itself then delivers the oil into the system under full system pressure and restores system pressure immediately without any time delay. You�re right the oil can accumulate in the right-hand front camshaft housing and although there are various alternatives to return it to the sump we simply use a high delivery electric pump to deliver back to the central portion of our sump where the oil pick up is located and we connect that via a relay to the brake light switch because it only ever traps oil there under breaking. 997 models and later engines have an angle cast in on the camshaft housing that allows oil to return back to the engine normally under most circumstances but the earlier cars didn�t enjoy that facility and they can trap more oil. Even so we still fit this pump to our 997�s. Our main problem with advertising and supply to the USA is that we are absolutely flat out always in our precision machine shop fitting closed deck Nikasil alloy liners to customers engines for standard and oversize capacity conversions and stopping to make parts for racing engines and small batch quantities not only really messes up both production and customer deliveries but also disrupts work loading for the other part of the business that strips engines and rebuilds them and fits them into cars - so we tend to fit racing parts production in small batches whenever we can and supply them to our UK market - usually to those customers whose engines we are already building and this is partly why we�ve been so successful in winning Championships in the UK. Although the American market is around 5 times larger generally (and probably 10 or 20 times larger for track day and race cars) � the advertising could only be justified if we had relatively large volumes and we could only take that risk if we knew we had a retail outlet that was prepared to order in some quantity. Furthermore - to their credit - Americans seem very loyal to their own providers (nothing wrong with that) but it does mean it would be more expensive for an outside country to promote their products in the United States against a Nationalistic resistance with no volumes established in advance. If we had a bigger market we could manufacture more products/batch and they would then be more easily available Internationally to assess onward demand. That�s why when we are handicapped for trying to help by refusals to allow anything that could be interpreted as �sales� or �advertising� but want to assist those struggling to understand problems we solved years ago � we resorted to free reports containing all the relevant information which for some reason people are reluctant to request � so that leaves time to try and help with short brief replies on forums to difficult questions against volumes of rejections on points that are often incorrect, technically wrong and generally from people with no intimate knowledge of the engines or engineering in general � which can be very frustrating. We are just trying to help with good hearts and genuine intentions yet we come up against a barrage of alternatives almost as if other contributors are offended that we offer our opinion which they take as against theirs when they often don�t have the qualifications, experience or track record to back up their own opinions � only misguided trust in alternatives that are often not as good. It often seems that if something is presented on a video and can be filmed or recorded on a data logger � or read about somewhere else � that counts as far more important than the experiences of some of the most successful engineers in the World of racing vehicle specialisation for over 50 years with both proven race success, engine design and manufacture success at GP level, with their own IN HOUSE facilities, that have won more production Championships in the UK and manage and handle more road and racing engine work in the UK than anyone else and have successfully provided for an International market resulting in a brilliant reputation � except it seems for in the USA � where we suspect we would be wasting our time � and which many of the responses seem to reaffirm. People over there seem to be generally relying too much on technology and not enough on common sense and simple understanding of basic issues, testing and something that is proven to work unless it can be described on a video with data logging screen shots to accompany it. The whole issue of bore scoring is a typical example � such a lot of nonsense everywhere and believed by most � and all wrong � providing opportunities for some others to market oils, insist on the necessity for parts and solutions not required and generally mislead owners. The thing is we are not after World Domination, perfectly content with our businesses performance and happily managing the gradual increasing demand we are grateful to receive for all the hard work we put into our solutions. There are also already some very accomplished providers in the USA that seem to be doing a good job of marketing solutions to their own larger and naturally loyal market with good promotions and presentation that we would find difficult to compete with from a presentational point of view and expensive to market against � when we don�t need any additional demand anyway. So guys I will continue to try and help out with technical comments when I can find time but our reports really are worth asking for if you want another alternative technical explanation upon which to base your ongoing decisions. Baz with a bit of luck pictures to follow.

In all fairness I should also point out that we are not in the dark ages over here. We do use high speed video camera's looking at gauges etc (this is where we first saw the oil pressure drop to the EPS switch drop to set pressure of 3.5 bar before the valve to the Accusump opened and realised that was too low and delayed recovery too long). We also use data loggers and it was only after analysing that when we found small losses of pressure and realised we needed to completely eliminate pressure loss from long pipework and in any circumstances. So modern data systems are of course really helpful - but my point is that although they demonstrated that we still had a problem and proved when we had solved it - they didn't provide the explanation why it happened nor came up with the answers which were very basic engineering. Baz

Quotes source post: 17471360

Baz, your technical contributions to this topic are greatly appreciated. A big issue with forums is that great information is buried in several threads and hard to connect. Thank you for connecting many of those above.

Quotes source post: 17471298

Exactly! There is now a new generation of M96/97 owners and those of us who want to track their car face some uneasy decisions to mitigate the risks with the oiling problems. You can spend an extra $30K and get a car with a 9A1 engine, of course. Or you could put money away to have your engine rebuilt after it fails, which will be at least the same amount, plus the hassle and heartbreak of blowing up your car on the track etc... So if there was a solution like the one illustrated by Hartech (sculpted racing sump + always-on accusump for track action) available at a price for a suspension overhaul ($5-8K ?) that would be of interest to a lot of people over here. I thought the Hartech report on crankshaft was very clear and convincing and even offered further solutions to allow you to compete for podium positions, but that's a lot more than we 'hobbyists' need (Though I'd love an electric third scavenger pump activated by pushing the brake pedal, oh yes please !) Baz, I totally understand the problems of selling specialized top-engineering hardware to the US originating from a small foreign engineering firm in Europe. After all, not long ago that would have described Porsche GmbH as well....

Quotes source post: 17464903

If the Accusump is connected to say the sump itself - or some lengthy pipework or even into some part of the engine normally under oil pressure then you are right that as soon as the oil reaches the pump gears it has two jobs to do - refill the Accusump and the engine up to pressure again. The more the pressure has fallen the longer it takes the Accusump to recover the system pressure AND refill the Accusump = temporary system pressure drop. The further away from the pump the Accusump is and/or the lower the pressure of the area it is connected to then the longer it takes to recover. But the Accusump is just a container with air one end and oil in the other and the pressure inside the Accusump is always system pressure or lower. When it works the piston inside moves the oil back into the engine but as it does so the piston moves, the air end volume increases and so the pressure of delivery drops = again takes longer to recover. But if the oil goes straight to the pump then the instant system pressure drops even by a small amount - oil flows to the pump and the pump gears then raise the oil pressure to the set limit of the oil pressure relief valve - so instantly re-pressurise the Accusump and the engine without any noticeable drop in system pressure. I have to admit I didn't realise there were such problems over there - assuming they had all been solved years ago - an opportunity we have obviously missed out on. I will have to try and get some more sumps and pump bodies into production and find a suitable outlet for them over there. Baz

Hey Folks. Given all the great info on this thread, coupled with a great conversation that I just had with Charles Navarro from LN Engineering this week on my 996, it may be helpful to begin drawing up a framework or summary checklist of options for those that want to address the issues we have been discussing. These issues certainly affect the 996 on the track, but can also come into play with spirited street driving. These issues include: 1. Oil pressure stability 2. Oil starvation 3. Water and oil cooling 4. Bore scoring Given all the analytics provided by Jason and Paul, and the expertise in defining the issues/causes/solutions provided by LN, CTS, FSI, Hartech, Skip and others; we can break the solutions down into 3 primary options or choices that you have to address the issues. Obviously, there are other variations, but we are trying to keep this simple. 1. Bolt-On Solutions : You can use bolt-on components to address individual issues, recognizing that they are not comprehensive structural solutions and may need to be combined with other bolt-on components to increase your probability of success. These bolt-on parts follow a sequence of Basic, Basic DE, Advanced DE, and Advanced DE/Club Racing depending on your driver level on the track. See list below. 2. Structural Rebuild : You can rebuild your engine to structurally address the issues we have been discussing without going thru the exercise of attaching bolt-on solutions; or the option to add them later depending on how comprehensive the rebuild. See list below. 3. Timed Combo : You can use a combination of prolonging engine life with bolt-on solutions first, and then have a structural rebuild done to your specification at a later date. Below is an initial attempt at a framework/checklist, with the purpose of integrating preventative solutions based on a driver�s advancement levels, with a comprehensive engine rebuild as the final solution or step. Please let me know if I am missing something or just got it completely wrong. No worries, we can easily make adjustments. Hopefully, this list becomes even more comprehensive to be used by 996 owners to add direction and clarity. Step 1: Basics - Necessary with or without tracking  Bore Scoping and/or used oil analysis  Fresh ignition coils, spark plugs, fuel injectors, AOS, MAF, primary o2 sensors  IMS Replacement  Proper oil and more frequent oil changes with a better oil  Replace plastic/rubber components because they disintegrate with age such as oil fill and aos tubes  Maintain Porsche maintenance schedule and replacements. Step 2: Basic DE - You are just starting out but gotta do these things to protect your engine, but you are still running street tires.  Upgrade AOS  Add Deep Sump Kit  Use better oil or race oil preferably  Change your oil after each track session with regular used oil analysis. Step 3: Advanced DE - You are getting faster with stickier tires (treadwear <200) and putting more stress on your engine and generating more heat.  Add Low Temp Thermostat  Upgrade Radiators with 2 CSF radiators and better yet, add third CSF center radiator.  Upgrade Oil Cooling - Note: Nothing is yet on the market. However, LN engineering is developing a more powerful oil cooler of which they need help in testing the product. Contact LN Engineering for more information if you are in the Chicago area and are willing to provide access to your vehicle for product development.  Add datalogging to monitor oil temperature and pressure. Step 4: Advanced DE/Club Racing - You are very fast (on slicks or equivalent) and need more oil pressure stability.  Dry Oil Sump - Note: Nothing is yet available on the market. However, CTS has developed a dry oil sump system. Contact CTS for more information.  Accusump - Note: there are pros and cons to using one including whether you use a manual or electronic control valve and also concerns with regards to where you can mount it safely. LN does still offer this as a kit.  Tandem Scavenge Pumps - Note: Porsche included these on the x51 package to help with scavenging oil from the heads, but they were discontinued many years ago by Porsche. LN did offer these as well as did TTP but both have stopped making them because of lack of demand. Step 5: Engine Rebuild - Comprehensive engine rebuild that structurally addresses the inherent issues with 996 engines depending on your objectives.  FSI Program  Hartech Program  LN Engineering Nickies and RND Programs  Porsche Indy Rebuilds The bottom line is that there are many solutions that are currently available on the market, with more on the way. It all depends on if you want to prolong/prevent , rebuild or both . As 996 prices continue to rise, we will most likely see additional solutions come to the market as costs become more viable. For those of you on this thread that provide product solutions, now is probably a good time to include any other solutions that we are missing. At the end of the day, 996 owners are just looking for clarity on what to do and how to get it. Hope this helps. A big thanks to Zbomb (Jason) and all on this thread for getting to this point. POST ON STEPS (ENGINE CONTINUUM)

Admin sent the crankshaft paper to me a couple of days ago and I look forward to reading it. I think we�d be happy for a solution from anywhere. I agree this is probably a different group of owners.

Very happy to see the direction that this post is going and thank you to all who have contributed meaningful information. I really enjoy my 1999 C2 which I track several times a year. It has a factory reman engine with not a lot of miles, but it would be great if a solution for oil pressure drops and cooler oil temps could be attained to prolong the life of my engine as well as others who are concerned. Thanks again to all who respond and especially to Zbomb who initiated this search of the holy grail for the M96 engine.

No thanks needed for me - but it's appreciated I'm a results guy - I'm waiting with the rest of you for something to come of it .... I am not going to drive slower or drive around this issue - if it kills the motor before a solution is found, well, then it will kill the motor. Maybe, and I hope - there are solutions for some of you guys currently available, maybe some have been discussed in this thread, be it cooling, deep sump, rebuild - whatever. I can't know how your setup works in comparison to mine and I would not inject my opinion and CAN NOT tell you what will work FOR YOU . Do know this though - I spent MY money on data collection INSTEAD of go fast parts, tires, track fees - Data collection is an integral part of my real life and I have seen the value - It is not enough FOR ME for someone to tell me something works. I will keep collecting data on my setup, my use case and driving the car, If people continue to be interested or I have something interesting, I will certainly continue to share it. If that data says something works GREAT on my car - you'll know, if it says something doesn't work for me - well, you'll know that too, I said, way back - the RS3 software needed to review the data I capture is free and I am happy to provide logs to anyone interested. This offer is for everyone - track guy, component manufacturer or just someone who is curious. No strings, I do not withhold any information, I have no sponsors or agenda other than a car that works as good as it can - you get the full data set, ask Paul Paul has set the bar very, very high for his ability to paint a picture with the information and if anyone deserves thanks for traction this thread has got, it's him. This stuff is just a bunch of squiggly lines without his formatting and analysis. Paul - you are a boss. If anyone is interested in collecting data on their own, I would be happy to help in any way I can - I have nothing to gain, other than knowledge and have NO COMMERCIAL INTEREST. It may be that the idea of setting up a data collection system seems daunting, it certainly was for me - but now I've gone through it - I can offer assistance to others. More collectors and sharers means more opportunities for visibility into efficacy of the different ways we setup out cars, be it for the street or the track. I'll have the scavenge pumps installed in the next few days and will be ready to have results to share from my next track day on the 24th. I will also have the street data between then and now for those interested in what the data looks like on my car under spirited road use conditions - I'm sorry for the wait on that, I have not forgot about you Mike.

Don�t feel obligated to do this; You owe me nothing. I just thought it could help if the condition of low oil pressure can be seen on an on/off ramp, it could be useful and easier to test than on an actual track. Then of course, tracking is probably a lot more fun to go with the testing

Not trying to sell anything - just trying to advise and help. I will try and add a photo of the pump so you can see how the outer gear teeth have larger gaps when they are not meshed together and send a greater volume of oil upwards towards the engine than the smaller gaps in the centre that can send it back to the sump again. The RHS gear goes anti-clockwise and the LHS gear clockwise If the oil level falls below the pick-up - firstly the oil pressure already in the engine then sends oil backwards towards the pump (because there is no pressure from below to prevent it) then when the engine pressure has dropped enough and only air or foam can be picked up by the pump it sends it upwards towards the engine. This means that any type of supplimentary feed to the engine first travels both back towards the pump and into the engine itself before the pump can restore system pressure and while an Accusump is doing this its delivery pressure is falling. If instead the feed is directly to the p ump then even if the feed pressure initially drops it is immediately recovered by system pressure with the pump quickly restoring pressure and there is no quicker way to do it. Feeding oil back into the sump when the whole combination of axial, radial and braking forces have already sloshed the oil in there away from the pick up can only help if the feed can fill the whole area of the oil low down in the sump instantly - as if the only problem is ever that the oil level just minutely drops below the pick up - it doesn't! You therefore need to do a number of things at once. (1) prevent as much oil as possible sloshing out of the sump in the first place (which our valving helps with), (2) recover oil pressure directly from the pump itself (which our oil pump cover feed does) and return oil from the heads under braking (which is the only time it is needed and which our electric pump does), prevent the overhung crankshaft from flexing too much (which our two options of an 8th main bearing also helps with). We have found at competitive levels you need all four. It makes no difference to me personally if anyone wants to just try one of those elements and record the results but if they are racing at top level on slicks in the UK all they would record eventually would be the exact time and conditions under which their crank failed. Maybe then after repeating that with additional changes until the failures stop you will also be able to record which is needed for a particular level of racing - but the cost of one crank failure is far higher than any of the solutions we offer - (or the data systems many fit). Good luck, Baz

Ok, I got the green light from my partner to share some info of what we have been working on for many many months. ( since I had already kinda let the cat out of the bag in a recent post).This is the third version on the same concept that I believe will solve the oil pressure drops on lateral G-force turns that I have seen first hand on the street with street tires in moderate turns, and what Zbomb and others are seeing on the track with great data to follow. The Ultimate Integrated Dry Sump Version 2.2 (UIDS V2.2) takes a systematic redesign of the sump to address the shortfalls of the factory Integrated Dry Sump and of the currently available deep sumps that pick up oil from the center of the sump. The UIDS V2.2 uses a G-Force sensitive flow control valve to direct the oil pickup to the lateral side of the sump where the oil goes under G-Force. The "Heart" of the system is the "Novel G-Force sensitive flow valve" that is robust and simple in construction, but is highly sophisticated and technical in design. This is the part that make this concept possible. What sets this apart is the dual pickups that are located at the left and right of the sump and is where the oil will be on lateral turns, and can deliver oil even at G-Forces of over 2g. The G-force sensitive flow valve has rolling steel ***** (Tungsten) that follow the oil in the same lateral direction under the same G-force and changes the pick-ups.. We have done bench work and flow bench tests, 3D printed prototypes, but are still a month or more from actual in car tests. (sorry Zbomb), but have been working on this since well into last year.(this kind of stuff takes time)The UIDS V2.2 should need no Accusump to work. Before anyone asks, the cost analysis of the time on the 100k CNC machine to make UIDS V2.2 parts has not yet been done so price prediction is not available, but I think the product will be well worth the cost and multiple times less that a Dry Sump setup and much easier install and cleaner. UIDS V2.2 bottom UIDS V2.2 UIDS V2.2 cutaway

Quotes source post: 17475851

GIF Video when you need a tester, let me know.

Quotes source posts: 17475866, 17475866

Just send over a U-Haul full of it, I could use it !!! OK, I'm hoping to get it on Zbombs car with all that great data logging stuff and a before and after data.

Quotes source post: 17475851

I predict you are going to make a lot of M96 and M97 drivers happy.

Yeah, hopefully there will be a version for the 997.1 C2S with the M97 engine!

Looks like the UIDS is a simpler install than the UAOS My car is low, deep sump is not ideal, I like this a lot.

Skip, this is awesome. Appreciate the work you put into this. Game changer on many fronts!

Skip, I give you a lot of credit for sharing as much as you have for something that is still in development, especially on a design path as unique as you are taking - I wish you the best of luck on progressing from where you are to a final design you are happy with. Happy to help any way I can.

Quotes source posts: 17475974, 17476030, 17476088, 17476093, 17476173

Thanks, I hope so. M96 and M97 can use the same one, it will work, same architect Yea, the UIDS will be a piece of cake !! ( the UAOS really isn't that bad..lol) Thanks man !! Thanks man, hope to get it on your car soon to see the before and after, .Hopefully the new scavenge pumps will help some and then the UIDS will fix it all.

Thanks for the positive comments. I have been working with Skip on this and the UAOS. Please be patient with our progress, I am only working on this in my spare time. I am really looking forward to making a prototype and seeing if this thing works.

Man i wish my motor lasted. I wont be able to use it on the new one it already has a dry sump

Everyone meet George, he is a whizz on CAD and CNC machine !!

Quotes source post: 17477430

Yeah I can't use it either. Both of mine are dry sumps. It does look like it ought to do the trick though.

Charles, Can you share some info about your new oil cooler prototype? Realizing that it would be best to add a third radiator and the oil cooler: from a sequence standpoint, if you can only do one this year an another next year, which one would you do first? Thanks

I�d be down for it and will certainly wait for it.

I would caution anyone to put this (UIDS as pictured above )on their car before it’s a ready product shipped with a proper engine warranty. Let Porschetech3 develop it on his own car. It got good potential to cause serious engine damage. You have been warned. Edit: Google "Christen 802 oil valve". Looks familiar to the " Novel G-Force sensitive flow valve" ?

Seems a promising initial idea that requires a lot of real world testing, optimizations and indeed a proper warranty. I see a simple piece of variocam or silicone sealant jamming the valve one way and causing catastrophic pressure loss in low speed corners. All clearances inside that valve should be bigger than whatever debris screen is gonna be put before it. At this point unless the valve gets a proper fail-safe, it imo adds a potential failure mode to the engine. While the idea is elegant in clean-oil conditions, a heavily baffled sump with simpler and less prone-to-failure flaps would still be a safer approach at this stage.

Quotes source posts: 17478816, 17478842

Hey dudes - I agree, it's a cool idea. The path from idea to a functional, marketable reality is a tough one and I think that fact is understood by those that need to understand it. The guys developing it have clearly said work is ongoing, so, I think we should let them do their thing, provide input that may be beneficial when appropriate and see what comes out the other side. If it works it works, if it doesn't, it doesn't. BTW - there is no such market available "heavily baffled sump" in the USA that is proven to work, it's just one of those, if you think that would be best - you have to design it, test it, market it and sell it in the United States. And one more thing - that IMS bearing thread sure turned into a raging dumpster fire - that won't happen here

Quotes source post: 17478903

Maybe I should�ve added that performance-wise baffles would be inferior to such valve approach, but safer so long as the design does not take plastic or metallic debris in consideration. Regarding starting a dumpster-fire, that depends on whether my post can be accepted for the constructive criticism it is without being interpreted as a personal insult. While I don�t necessarily agree with Nick_L�s manner of addressing the topic (if you want to speak up about a potential design flaw you should just say it straight away instead of this), I see no reason why giving constructive criticism (at any stage in the development) should end up in dumpster fires. That said, I�m obviously interested to see where it goes eventually.

Quotes source posts: 17478955, 17478903

K- I choose my words carefully, please read them as such. I did not say you started anything - you and Nick were certainly contributing factors to getting the IMSB thread shut down among what appears to be many, many others - THAT CAN'T HAPPEN HERE. I also said

Kudos to anyone who takes the initiative to solve problems. That is what this thread is all about. Certainly it isn't easy or solutions would be everywhere. However, there is always a more efficient solution. Just gotta keep looking. Keep pushing forward folks. ​​

Quotes source posts: 17478903, 17478842

Thanks Jason.. Understand that Skip and I have spent massive amounts of time thinking about this and are aware of the shortcomings that you are bringing up. What oil pressure solution currently exists that will warranty an engine? (honest question) Yes, extensive testing will be done with as much data as possible. I understand your skepticism, I'm not even convinced that it works. Very rarely does anything go from paper (digital paper) to real world without unforeseen issues. Like I said, please be patient.

Quotes source post: 17479140

Keep pushing!

I'm happy to assist with the effort as well. Have to support the things that are invented, developed, and made in the USA for these engines! Whatever Skip and George need/ want from this end I'll help with. Warranty... That's the best joke I've heard all week, but its only Tuesday. Does the seller of a short shift kit guarantee that the driver won't miss a shift, and zing the engine? Track activities are a big boy sport, at the end of the day you have to enter the arena with the attitude that the entire vehicle could be a total loss, or you could end up doing "sheet time" in the quest of having fun burning up tires, and fuel. If you want a warranty for ANYTHING regarding track actives, don't go to the track.

Well I see the Negative Nancy's have come out to **** on the UIDS Parade, just they did for the UAOS. They have such high regard for themselves that anything that doesn't come from their brain has to be flawed. Jake I agree 100%, thank you for your support. The warranty for the UIDS will be 2 years unlimited mileage, same as Genuine Porsche Parts and UAOS. And is for material and Craftmanship of the product ONLY. No labor or anything else. The UIDS will include a "Novel Big Red LED Delta Warning Light" that will use a Microprocessor set to 12psi per 1k RPM Delta with a 4 wire hookup, Pos, Neg, Tach, Oil Pressure, to be dash mounted in plain view. It will alert the operator of ANY oil pressure condition (whether UIDS related or not) that would be unusual and be a notice to take ACTION. Taking action will prevent damage.

^^ exactly why my track performer and Numeric kit equipped car will never see a closed course

I thought OP said he didn’t want a ****-storm, so maybe temper your tone a bit Jake. No need for that personal and aggressive tone, certainly not after your poor showing yesterday. Warranties don’t apply on track but when a device can fail due to day-to-day wear during day-to-day driving, it should be ironed out before release and imo warranted that under non-track conditions such failure cannot occur. @Porschetech3 , glad to see you will warranty it. Wasn’t as big of a joke after all then. You can do with my suggestion whatever you want (though I thought you put me on ignore, guess not), but if oil pressure drops and needs human intervention, it is already too late, damage will occur. I suggest you attempt to R&D an internal instant fail-safe. Try to keep an open mind, especially in early design where the goal should be to identify and address potential flaws.

Happy to see FSI back with an offer to help in the development. I will again ask everyone to keep this thread sharply focused. Warranty’s and if you will or won’t take your car to the track are outside the scope of discussion.

Quotes source posts: 17479365, 17479434

Thanks Jake, We will take all the help we can get.. and we probably need it. Clearly you haven't read many of Jake's posts, That was Jake being nice...

user 63031 this is the design, the failsafe is the "Novel LED Delta Warning Light" that is superior to the "Porsche Factory" pressure guage and warning light that is the end all, be all failsafe from the factory for oil pressure issues. If you have a better design, then present it for critique, otherwise stop casting a shadow over my Invention, yes I take it PERSONAL.

Let�s not crap on these people (vendors). I keep my business out of social media so I don�t have to deal with this stuff. Like some of these guys, I have all the business I can handle and don�t really need any more. Don�t like it don�t buy it, but don�t discourage them from either coming here with personal experience to share or a new product that�s either in development or coming to market.

Quotes source post: 17479491

Whatever you want mate, if you insist on taking it personally and dismissing a possible mode of failure in early development just because it comes from me, then be my guest. Simple plastic debris could lodge your current design shut one way which means under normal use in a slow corner the invention might be actually trying to pick up oil from the oil-starved side and do the exact opposite of what it's intended for; cause catastrophic pressure loss. If this thing sticks either one side, you're in big trouble. Since the 'actuator' is driven by G-force only, it does not need a lot to stick. By all means blow it in the wind and put me on your ignore list (for real this time), I don't really care. I'm just speaking my mind on how I think this thing might pose more dangers to the engine than it solves. And I'll say it again; the idea itself is elegant. It's the first thing I said. Stop being so offended all the time, geez.

Quotes source post: 17479550

You have no "skin in the game" so it really doesn't matter to you. No one knows your background (because you want to be anonymous) but yet want to spread your "opinion" all over Rennlist. Given you screen name, you probably have a Boxster that has "no oil pressure guage" anyway, so you have no idea what your own oil pressure is, or whether is it droping in turns, or fluctuating, or is even low from the debris that was in your motor from the failed IMSB You will not be a customer of the product anyway, since you opted for a cheap IMSB for your own car( instead of the IMS Solution), so your opinion about these issues hold no value, just casts a shadow over a product you didn't think of or may cost more than you are willing to spend.

Suit yourself. You guys are on a roll with wrong assumptions about me. First I'm 'Stefan' and now I have a Boxter. I have a 996, oil pressure is fine but thanks for your concern. Have a nice one otherwise, mate!

So at this stage we have two very promising solutions that are in development and more importantly, will need to be tested. 1. Skip/George UIDS deep sump kit 2. Charles LN oil cooler We all know that TESTING will be done to work thru the issues. Let's please move forward. Charles, can you tell us about your oil cooler when ready. Edit Add: plus two more solutions not mentioned above. 3. Barry's Hartech acusump and deep sump kit. 4. CTS Dry oil sump system

Plus Barry's accusump and oil pump cover too?

Quotes source post: 17479635

You are right. Just added with edit.

Sending out a public warning he re. Clearly the whole IMS fix remains a very volatile and controversial issue. In fact, we have temporarily locked two other threads while we figure all this out. Although things were difficult in this particular thread last month, a few people (including the OP) have thoughtfully worked with me to keep things clean, to not escalate, and to stick to the facts. Threads like these allow for different viewpoints, so everyone can make up their own mind as how they want to deal with potential IMS issues. Gather and share your data and experiences. But let's keep things free of slamming of users and vendors, let alone dropping the antagonism. Thanks.

Skip - Do you have an idea what the ground clearance is going to look like ? The clearance issues brought about with deep sumps and the added weight of a skid plate (Yes, I know its not that heavy, and its very low on the car, but present none the less) would be nice to shed.

Quotes source post: 17480605

yep, ⬆︎ this. I have to run stock height b/c of the 2L deep sump and skid plate.

Quotes source post: 17480605

Yes, the UIDS V2.2 is 1 inch thick total, from the case mounting face to the bottom of the cover. I think most 2qt deep sumps are almost 2 inches from the case to the lowest part which is in the center of the sump near the drain plug. So it will save almost an inch of ground clearance.

Maybe a little distraction with more plots... Chris at CTS was quite generous to share his data from the 24 hour run at Thunderhill in 2017. Data above is from this run that he published on his blog. Here is a plot of one lap of which there were 680. From top to bottom: RPM, Oil Pressure, Oil Level, Lateral & In-line Acceleration. -It's immediately obvious that the pressure drop issue does not exist. Oil pressure follows RPM's very well. Lowest pressure seen is 3.6bar at 3500 RPM. There is no pressure loss due to braking or turning. There are, however, very fast spikes of pressure loss. A closer look shows these occur at around 5850 rpm when accelerating and 5600 when decelerating. The difference in RPM's is likely hysteresis of the mechanics. The drops are very fast, much less than a second, they only drop down to 4 bar(~60psi) or so, and are NOT due to air being pulled into the system. Certainly not a threat to the engine. Pointing this out to Chris, his quick reply was that this is from the variocam opening and closing. There is a brief blip in the oil pressure. -The oil level(green) is a gauge in the dry sump oil tank. It seems to vary quite a bit. Looking at G's, we see it closely follows the lateral acceleration. This is oil sloshing around in the sump tank due to turns. Clearly the level never gets to a point where the pressure is affected. The size of the tank has been made large enough to not allow a dry pickup condition If we zoom out to catch a rest stop for the car and driver, we see the oil level in the tank maximize. It tops out at around 330mm. The average level when running is about 230mm. If we know the size of the tank, we can easily calculate how much oil pumped into the engine and is just hanging around looking for an exit. I know the tank is cylindrical, but I don't know the diameter. The calculations for a 5" diameter tank and a 10" diameter tank would be 1.3quarts and 5.3quarts of oil lost in the engine, respectively. I need to check with Chris to get the diameter, if he's willing to share, to know the actual value. 5.3 quarts would be a lot of oil doing nothing and would explain why mediocre deep sumps don't work. It's all about the baffling and oil return getting to the pickup. An interesting point is how long it takes the tank to refill to equilibrium: 80 seconds... That's a really long time to have 'lost' oil. Going back to the Palmer plots, I wanted to see if we can see these fast pressure drop due to the variocam. Sure enough, they are there! Look to the blue arrows. The RPM's at which they occur are slightly different from Chris's car, no surprise. The yellow arrow shows one that seems to be missing. On some laps, there are none. Two things about this: data collection rate on the Aim is slower than the Motec, so it's possible to miss them, and other than diagnostics on the variocam, we really don't care about it relative our problem. So, it's interesting but of no consequence. The take away is that some of the sharp short not terrible low pressure drops we see are due to variocam. Noise.

Paul, incredible analysis. I am assuming this data is from a car with CTS's dry oil sump kit? Is that correct?

Paul - looks like the sampling rate can be turned up - WAY up, from where it’s set now, which I believe to be 50HZ 20HZ... the aim SW limits a pressure sensor to 50HZ with the standard config setup, but, I can custom configure the sensor for a higher sampling rate. The limiting I’m guessing is for data storage reasons. I can play with this if additional granularity is wanted / needed.

Correct. Yes, this is from one of Chris's Boxster's with his dry sump. Here's another interesting plot that shows the oil consumption across the run. The oil level drops across the entire run. Chris confirmed that this is the oil used by the engine, ~1.3 quarts. I guess I could use this to recalculate the 'lost oil'. Maybe after work today. Data will set you free!!

Quotes source post: 17481249

I don't know that we need to change the sample rate. Chris was running his at 100Hz. What we should find out is where his pressure sensor is relative your sensor, just to think through how his pressure reading might be different. Whatever the difference, it's clear he's not pushing any air. I keep going back to Jake's statement that he used 6(or 8? or ??) sensors in different locations to figure out what was happening. [edit out erroneous statements] Since your sensor is after before the oil filter, if we discard the fact that we know the rod bearings will fail first , all we can say right now is the entire engine is at risk. Additional pressure sensors would fine tune that knowledge.

Mine is pre filter Paul, at the housing. If I recall from earlier in the thread, Chris placed the sensor at the cooler which is post filter and the last stop before heading into the mains. Having a setup with sensors located pre and post consumption areas would give you a great picture of drop in those areas and also an idea if you could steal some excess capacity for other uses.

Paul, you are doing a great job with the data analysis !! Keep it up !! I will start a new thread for the UIDS so that George can post updates on the CNC progress with pics [EDITED BY MODERATOR]. This thread should be kept clean and about the data. Once in car testing begins on the UIDS the data can be integrated back into this thread.

Golock911, that is some interesting data, to see the running oil level. My calculations shows that when the running oil level is 1cm below the minimum indication on the oil stick (the small dot) you can have a steady 1g lateral load, but not more(see picture in post #28). This data makes me think we should figure out a more precis way to measure the oil level so that the running oil level is high enough. The way it's written in the manual might be ok for street driving but for track I think this can be improved.

Quotes source post: 17481312

Paul, there is a channel labeled "oil tank volume in quarts" or something similar. I carefully calibrated that (0 = dry tank) and it is accurate (on average or when the car is not moving). That will answer your question about how much oil the engine hides internally, which then sheds some light on my conclusion about the futility of trick sumps. Short answer, the engine is hiding about 3.5 quarts, even with the 3 additional scavenge points provided by the dry sump system. I don't know the exact amount hidden by a wet sump engine, I did test racing the engine with 15 quarts of oil in it, and it does not help much. Also I need to correct the statement "The rod bearings fail first." I have never seen a failed rod bearing without the main bearing that feeds it being damaged, so the main bearings fail first, and the rod bearings don't fail before they are sent debris by the failing main bearing. Please keep in mind that I don't work on street engines and others' experience with this may be different. And just to make the point: there is not a single event of abnormal oil pressure in this 25 hour race (also true for the sister car which also ran in this race). That is the standard I had to use in evaluating the system. Chris Cervelli Cervelli Technical Service

Thanks Chris for your corrections. I've lined out my erroneous statements above. I have no experience to make these kinds of statements. I only have knowledge from all my readings on the engines, and I misspoke. I will, in the future, pose these types of comments as questions unless I have direct experience with the issue.

Quotes source post: 17482360

Your test with adding oil, what was the state of that engine (stock, rebuild, mods etc) and more importantly, had you check/replaced the scavenge pumps? My experience from before changing the pumps is similar to yours, oil fill level seamed to not make any difference. It was one of the first thing I tried when first seeing the problem. After changing them just going from 1/2 full to full made a noticeable difference.

Chris, thanks for sharing the data on your dry sump with Paul. The data suggests that it looks like you have build a robust solution given that it survived 25 hours with solid oil pressure. Nice work! Realizing that it has been a few years since you developed the dry sump, and production innovation may have occurred to drive down costs, what would it take to produce and at what cost to make it sell-able?

Quotes source post: 17484968

My 2017 kit price was $8500. Since then I have got my machine shop up and running, which allows me to manufacture the crank pulley, the pump bracket, and the oil tank brackets (applicable to Boxster/Cayman) which saves a few hundred bucks total. I use a higher quality hose for the 9 hoses now, and that added a little to the cost. I have not priced out oil tanks from my supplier in a few years, and I suspect they have gone up. The tanks are custom and quite expensive. The pump assembly costs me over $2k alone. It is a very high quality part made by Dailey Engineering who makes Le Mans and NASCAR quality pumps. It just adds up to a lot of money and effort. I did more development in early 2018 which allowed me to delete the entire crankcase ventilation system and AOS, which is nice on a race engine. Later in 2018 I had a bunch of engine failures even with perfect oil pressure which was expensive to figure out. It turns out that even with good continuous oil flow, the main bearings run close to destruction when the oil temp is high. Those failed engines ended my endurance racing program. In 2019 I put all the developments together and that is the current spec of the system. It is about 8% more energy efficient and less dependent on oil level, which became a problem in customer cars that used non-rebuilt engines with high oil consumption. I found that the level of tech support necessary to get a normal Porsche shop to do a successful installation made the profit from the sale not worth my time. I did have a couple systems installed by very experienced race shops, and those went ok without much hassle. The end users routinely under or overfilled the oil tank despite detailed instructions on how to set the oil level. I did not include the electronic sensor in any system other than the prototype, because the sensor is about $1000. In hindsight I wish I had made it mandatory. By now all the systems have racked up something like 100000 racing miles, and there has never once been a loss in oil pressure, provided the oil level was correct and the system was plumbed correctly. There has been one pump failure caused by low oil level (the pressure pump cavitates and shakes the assembly apart) and of course that driver drove the engine to failure despite the Motec alarms (pay drivers reliably do this). There have been some other minor failures that led to redesigns or installation procedure changes, One system is on a car that is used on the street and it does several thousand road miles per year in addition to some track use. Chris Cervelli Cervelli Technical Service

Chris - I remember reading about internal mods, restrictors in the feed to the heads. Was this done when convenient (during rebuild) or for all installations ? I am curious how much those restrictors aided in keeping pressures up in the short block, also, if they played a part in minimizing the pooling in the heads. Did you use a 3 stage pump ? I am guessing placed where the AC compressor would normally be ? I think it's really cool you were able to actually design and implement a solution for such harsh conditions, paid seat drivers in endurance cars has got to be about as tough a combo as you'll find on the equipment and makes sense the parts selection was so high end. That said, I think the cost, even at 8.5K to be pretty reasonable for the intended audience. I worked out a system for an LS3 using really good parts - Daily pump / pan and am over $7K...in parts alone.

Quotes source post: 17486727

My pump is 4 stage plus a centrifugal oil/air separator so it looks similar to this 5 stage pump. 3 scavenge and 1 pressure stage and it is placed where the A/C compressor went. Credit to Dailey Engineering: I never would have gotten this done without them. They had an answer for every question I could come up with and gave me some great tips on the basic design. For instance, I was told that the M96 has the largest oil pump of any car engine. Knowing this was really important later. The M96 makes some unusual demands on the pump and I was told exactly what I could and could not get away with. The head restrictors were added later in hopes of fixing the main bearing failures I had in 2018, so most of the systems did not use the restrictors. They did not accomplish that, but they have a bunch of good side effects and in general improve everything. They are not strictly necessary though. The heads need to be removed to install them but I think I could redesign that and make the head removal unnecessary. The stock oil system sends more oil to the heads then one would expect. I think this happened due to the machining requirements in the cylinder heads. It is difficult to drill small diameter, long length holes so they just made the oil holes really big. The GT3 and air cooled engines have the flow to the heads cut to the minimum. The design case you mention above is indeed difficult. The system has to work under all conditions and you cannot expect the driver to ever do anything right. Chris Cervelli Cervelli Technical Service

Replacement scavenge pumps are in - Anyone want to tell me anything specifically they would like to see from the pull outs ? I have a bunch of other maint stuff I need to get done - unlikely a test drive will happen for at least a couple days. Bank 1 Bank 2

Nice work! It will be interesting to see if they add any improvement.

Quotes source post: 17486761

Thanks Chris - fix something, find the next weak link, I’d imagine that is especially true in endurance racing. I’ve been thinking about the part of your response specific to the M96 oil pump with regards to the largest oil pump statement and the unusual demands... Could you elaborate on those points at all ? I’ve tried to run through what I could interpret them to mean but have run out of knowledge real estate and am curious. On the idea of restricting oil volume - would you have any suggestions on minimally invasive locations where flow could be restricted to enhance pressure where it is critically needed. Thanks.

Quotes source post: 17490812

Well, the stock oil pump is just really big. Porsche built these engines to run Mobil 1 0W-40 which does not generate as much oil pressure as a thicker oil would. Probably more importantly, there are a large number of oil destinations in the engine: 1 Intermediate shaft journal 7 main bearings 6 rod bearings 6 piston squirters 5 chain tensioners (or 3) 24 hydraulic lifters 8 cam journals 2 variocam solenoids (or cam adjusters in 3 chain engines) I guess it takes a lot of oil to feed all these things, although the GT3 engine gets it done with a similarly sized pump (not sure on this, I have not looked closely at the GT3 pump) with quite a bit more oil pressure on the race engines at least. The Cup engines seem to target 7 bar OP. Nobody ever talks about the oil squirters. They appear to consume a lot of oil, although I have not measured this. They are there for piston cooling, which I believe allowed the use of cheap but pretty lightweight cast pistons (some of the 996 and Cayman 3.4s as well as 3.6's have nicer pistons). These engines are tuned to run .90 lambda or leaner at WOT which (probably) requires piston cooling for long term WOT durability, which these engines absolutely have. I wound up reducing the volume of oil sent to the heads. These engines have no trouble with valvetrain wear even on street engines which see long term hot idling (cam/lifter loads are highest at idle) at very low oil pressure. Race engines don't see much hot idling, so this oil reduction does not affect durability. The restrictors allowed me to slow the oil pump drive ratio and still have enough oil pressure at hot idle. They also reduce the amount of oil trapped in the engine at high rpm by a couple quarts. I sized the restrictors so the oil pressure at the cylinder heads in about 60% of the pressure at the main bearings. More pressure is not a help if the oil system is intermittently sending air to the main bearings. Whenever the pressure drops, this is happening. It is impossible to do any diagnosis or evaluation of the oil fed parts if you know that they intermittently lose oil supply. This is why an Accusump cannot be a cure, although it may shorten the amount of time the engine runs without oil. That air, once picked up by the oil pump suction tube, has to go somewhere. Chris Cervelli Cervelli Technical Service

Didn't the Mobil 1 0w-40 arrive around 2001? The engine is not designed to use this oil but the oil could be (pure speculation) formulated to work well with this engine. If the piston squirters consume a lot of oil this would be possible to see as a 'notch' (reduced slope)in the rpm vs oil pressure graph. Golock911, can you see this? You could achieve some(of corse not the same as a restrictor) favourable balance shift from crank/head by running thinner oil also. You will have more head loss -> less oil to the heads since they are downstream of the crank. I'm experimenting with this right now(basically how thin oil you need to achieve 3.5bar @ 5k rpm).... Today it's been 34c(93f) here and I've driven a mountain climb(700m vertical) to heat the engine properly and check oil pressure( yeah, it was fun also ). With a 12cts@100c (3.6 HTHS) oil the pressure is around 3.5bar @ 4200rpm, when coolant is ~108c. I can still go thinner.

Quotes source post: 17493698

Do you have an oil temperature gauge? You should do some testing with oil temperatures around 125-130C. The oil runs much hotter than the water.

Not yet. At the moment I am using idle oil pressure as a cold-medium-hot (very rough) relative gauge. Yes, the oil runs hotter in most cases. When you drive hard like in hill climbing or track driving the difference between oil and water will be bigger. When using thinner oil this difference will shrink due to various factors. So, at the moment I'm not that interested in absolut values, it's going to be more interesting to check that when I find an oil which deliver (what me and the user manual) thinks is the correct pressure. I'm pretty close on the pressure now, I'll see how it looks when the oil has gotten a few thousand km on it.

Quotes source post: 17493698

The oil squirters do not behave the way they are described in the service literature, so no knee will be visible in the graph. The main bearings are designed in such a way as to favor flow to the rod bearings. Porsche accomplished this goal, but made the carrying capacity of the main bearings marginal in the process. See http://www.epi-eng.com/piston_engine...e_bearings.htm for an excellent generic explanation of this. I have some what I think is rather shocking data on this as well. I also wanted thin oil to work, but it is not the right direction to go on an M96, at least for my use case. None of this matters until the oil is at 240 F (115.5 C) or higher and then it starts to matter very quickly. It is tough to get a 2.5 engine that hot even under racing conditions, where a 3.4 or 3.6 will, as we have seen in this thread, will get to those temps even in mild conditions. Of course the main bearing loads are proportional to torque, so the larger engines are worse in that regard. Chris Cervelli Cervelli Technical Service

Hey everyone, Kam here from LN. Just wanted to share a few things that I was able to dig up. I went back to the original test data from when the 2 qt deep sump was developed in 2012. This data below was from a World Challenge 996 at Sebring. Ambients for the weekend for 75-80F. A few details on the car: 3.6L crate engine from Porsche, 2 QT sump, Motorsports AOS, Miller Nanodrive CFS 5w-40 NT oil. No larger oil cooler or additional modifications. I've selected two session data sets. Each from a different day of the weekend. An early lap from each session and a late lap (roughly 15 laps between) for a total of 4 laps shown. This is important to show the consistency throughout the session. Parameters shown are RPM, Oil Temp, Lateral G, and Oil Pressure. The oil pressure is very consistent save for the few low speed corners where pressure hits 40 PSI (still acceptable pressure) when engine speeds drop near 1000 RPM. The team also ran a case study for Miller which I've also included a link to. https://performanceracingoils.com/co...case_study.pdf The pressure in this engine is arguably the most consistent we have seen, but the data mostly mirrors many other sets we have seen with the 2 QT deep sump installed. The data that was provided to us prior to the the deep sump showed pressure drop to 18 PSI with much more inconsistent pressures throughout the lap. If anyone is interested I can show it as well. Engine health plays a large role in how well it is able to maintain pressures, regardless of what parts are bolted onto it. Temperatures are also very important as shown in much of the data in this thread. We're in the process of developing a larger oil cooler for the M96/97 cars. I'll share some details in a separate post.

Quotes source post: 17505620

​​​​​Thanks for info Kam.

Quotes source post: 17505620

Why does the oil pressure not vary with rpm? Chris Cervelli Cervelli Technical Service

Looks like this engine holds pressure so well that everything above 4k rpm has an open pressure bypass valve at about 5 bar.

The main oil pump by-pass is factory set at 5 bar, so that's not unusual , but what I find incredible is the oil temp from 160F to 180F during race conditions. My 996 runs a lot hotter than that going to Dollar General..lol

Has that anything to do with airflow over the radiators at race speed and the water to oil heat exchanger? Then again we have electric fans too.

Quotes source post: 17506007

You find it plausible that the oil pressure at 3000 rpm and 160 F is exactly (within 1-2 psi) the same as the oil pressure at 7200 rpm and 185 F? Chris Cervelli Cervelli Technical Service

I've never driven Sebring, but dips down to 1K RPM ? That seems like it would be hard to do on track, especially 5 times a lap. Am I missing something here ? The temps are so far off what I have seen, I would like to see the water temps to try to gain some more perspective ? Kam - would you be willing to share the full data set for some additional context ? Thanks for posting BTW.

Quotes source post: 17505976

This is correct and our thinking as well. We have seen engines sit on bypass like this before. I can't really comment much on the dips in RPM. Yes, slightly strange, but could be normal for the specific drivers style, not sure. We weren't analyzing the driving itself so we never inquired. I've attached a screenshot with water temperature, and another screenshot pre-2 QT. Coolant temps also very stable and low. Track temp data shows it hovering around 105F. This is just another data point showing how important it is to keep temperatures in check, which is why were working on our larger water to oil heat exchanger. Pre-2 QT below. This is a used engine. No details on mileage specifics. Not the same create engine as the data above. Note, this pressure is in bar.

Thanks, cool data! Speaking of cool, the oil temperatures in your first plot are really very low unless there is some special cooling arrangement. I've never seen an M97 below 210F even at country lane cruising. Is that data calibrated differently, or is the temperature taken at a different location from the standard M97 measurement location? Probably is, so where is the temperature measured? Thanks again!

Here's where we are with our oil cooler, along with the gear cooler. We started with the 987.1 Cayman first. This is utilizing the same heat exchanger we use in our PDK cooler kits. Roughly 45% larger than the OE 996 oil cooler. We machined a plate with AN outlets for the oil/water to replace the oil cooler mounted on top of the engine. This will also work on the 996. This issue we are running into on the 996 chassis is placement. There really is not a good location as the cooler is just slightly too large to fit on the left side of the transmission. We want the system to work with the factory mufflers so we trying to avoid mounting it there, but we may have to do that. Another thought was deleting the secondary air, including the delete plates, and placing it there. Again, a bit more involved than we would like the install to be for the average customer.

Quotes source post: 17507628

Pressure and water temp used adapter fittings reading in the OE location with separate sensors. I do not have specifics on the oil temp placement. Perhaps a sensor into the side of the sump spacer. The car had no AC condensers and most likely had the fans removed as well.

The oil temp running cooler than water temp, with no cooling system mods is eye opening. I don't think I have ever seen that on my car. I have no AC condensers either, but I do still have the fans in. I only brought up the multiple 1K RPM dips because it seemed so unusual, I don't want to question the data, cause it is what it is but the temps, both oil and water, the car running on bypass the entire time and the strange RPM behavior all are a pretty big departure from what I have seen in my data. Is the team that produced the data still around ? would you mind sharing who they were ? Hope you get the cooler figured out - looks like you have some good tools at your disposal to get there.

Quotes source post: 17505620

Thanks for sharing the data, Kam, much appreciated. Seeing it was a crate engine I wonder how much of the solid oil pressure across the rpm-range and both longitudinal and lateral g's was due to the engine being basically brand new and thus with oil scavenge and pressure pumps as efficient and effective as they can be.

Quotes source post: 17512102

Hardtailer, You bring up a very interesting point which hopefully Kam can weigh in, but also could be confirmed by Nick's recommendation on the replacement of the scavenger pumps that Jason just completed.

Fascinating thread, really love the work you are doing here. I did the video comparing the different scavenge systems on the engines which was posted earlier in the thread, and have been workiing with my business partner to create a replacement for the NLA dual scavenge pump for the 3.4l engine. I have just tried it on a 3.6l head and it fits, the only issue is the fixing hole for the Porsche pipe, where the holes doesn't quite line up, so something would need to be done with this. We had speculated between us about the pump design. We can see exactly why it is needed on the 3.4l engine, and assume that it was designed at the time when Porsche were considering using the car for racing. As a result they had the parts designed and so could put them in to the x51 kit. When it came to redesigning for the 3.6l, the shape of the cam covers was changed and the dual pump removed. We did wonder if this was something of a cost saving exercise or if they REALLY felt it wasn't needed under heavy braking, since this engine was not being designed towards any intentions of racing? Either way, we now have a dual scavenge pump which has just started in production and if you are interested we would be happy to sell you one for 50% off on the basis that we love the work you are doing and are interested to see the results. Martin

Quotes source post: 17513814

Martin, this is a very exciting development. Where would we find more information on the dual scavenge pump?

There isn't a lot of information on it yet, I will be doing videos on the installation and also one showing the internals so people can see how it works. We are just waiting on a batch of pump gears to arrive at the moment. https://www.h2-engineering.co.uk/pro...gescavengepump Martin

Quotes source post: 17509217

Yea the shop is Racers Edge Inc out of TN. I spoke with them recently to get a bit more info on the car, but outside of what I've already listed they didn't make any monumental changes. The 10 hour crate engine is certainly a large variable here that we feel attributes to its ability to maintain pressures. However, as I mentioned, we have seen engines sit on the bypass before with the 2qt installed. Even with many more race miles on them. Also curious to see the results with your replaced scavenge pumps.

Quotes source posts: 17513814, 17515190

Hey Martin - I have been subscribed to your YT channel for a while and really enjoy the build you have documented and your videos in general, cool technical content. The dual scavenge looks pretty cool. I think the main open question is still, can the scavenge pump do enough to resolve the oil where it needs to be problem. First, there would have to be a sufficient volume of oil trapped in the head that if that oil was in the sump, it would bring the oil level up to a point the pickup would no longer suck up air. Next, it would need to be efficient enough to move that oil from the head to the sump fast enough to maintain the level required to not allow the pickup to be uncovered. I think some good data on the actual efficacy of the upgraded scavenge pumps to maintain an adequate oil level in the sump would be very interesting. Thanks for the info Kam - yeah - I don't know, whatever they were doing, not much analogous between there results and mine - From temps to pressure to RPM. I'd be interested in the opinion of some of the collectors of data on these engines if the results Racers Edge saw are in-line with what should be expected and my results are the outlier or vice-versa, I'm back on track tomorrow so we will have fresh data with the new scavenge pumps. Based on how little visible wear was on the ones I pulled out, I don't expect there will be much, if any difference. Unless a very small amount of visible wear contributed to a large degradation in pump performance, which I guess is possible.

Quotes source post: 17506189

Yes, very plausible at those oil temps... On my m96/01 it hits the by-pass on startup (5 bar @ 75F) and stays there untill the oil temp is 120F then starts to fall at idle speed,but will hit the by-pass at 140F at 2k,... at 3k it will stay on the by-pass (5bar) at 160F and even up to 180F... All m96/m97 engines that I have specifically looked at this particular characteristic, acted the same way.. And the oil temp always lags 20F behind the coolant temp untill coolant temp stabilizes, then oil temp goes beyond coolant temp

If the pump is on the bypass at 3k rpm, I assume that means the flow to the heads is the same at 3k as it is at 6k? However for scavenge side that isn't the case. Our dual scavenge moves 20.7lpm per section at 3k rpm, so assuming that both pumps on the right hand bank are in use it will be scavenging over 80lpm at 6k rpm when heading down the straight. If engine speed drops substantially under braking to 3k rpm and the oil is mainly at the front of the head you are down to 20.7lpm on the right hand side. Given this, to get the best scavenge performance it would make most sense to keep engine speed up under braking, assuming of course that the bypass oil is available for use again very quickly, I don't know how that side of the oil circuit works. Basic numbers for the above would seem to indicate that for a 2s braking event, at average engine speed of 4.5k rpm the extra scavenge would return about an extra litre to the sump. Hopefully with more effective scavenging at all times the effect would be larger. Martin

I don't even consider 160F-180F oil temp as fully warmed up, fully warmed up is usually 200F-220F. Under extreme track conditions oil temp will be much higher--240F-260F 3k rpm will not be by-passing any oil at oil temps above 200 F. The oil pump is a positive displacement so 6k will be twice the volume as 3k and makes those calculations not plausible. Fully warmed up, my engine runs slightly over 1 bar per 1k RPM up to 5 bar ( then by-passes)

I have instrumented one of my cars to look at coolant temperatures, with sensors on inlet and outlet for air and coolant on three radiators, plus thermostat. Any suggestions on where to additionally monitor oil pressure and temperature to add to this project? Martin

Apologies for dropping out for a couple weeks. My job is like this. Some weeks of sheer terror and overwork and then back to normal. Here are some plots from the sessions today to try to show the difference the scavenge pumps have made. The executive summary is that they have helped significantly. But have not eliminated the problem... Here's the overall plot showing Oil Pressure, RPM, Oil Temp, and Water Temp. All parms follow the similar pattern. Oil temp never really stabilizes and track with water temp. Oil temp is always hotter than the water temp, 10 degrees at idle and 40 degrees on the 7th lap. Oil drops don't seem to be quite as severe as the last run. Same as before, the oil pressure is highest with less severe dropouts at the beginning of the session. When we overlay a session from last time with this session, you can easily see the difference the scavenger pumps have made. The number and severity of the drops have been significantly reduced. I've shifted the curves to match the oil temperature. The red is today's session. The grey is last session. The grey drops below the red show the improvement from the pumps. Don't be fooled though, the drops do increase with oil temp. The overlapping 2nd half of this run is immediately below. Ignore the giant pressure drops to 15mph. I believe there was some issue on the track that held everyone up, idle time... Here is a look at one lap from todays sessions. The drops follow the same rules we've seen before, hard braking and turning. I also think I've done a better job of fitting the RPM to the oil pressure. The bypass for this engine seems to be around 4300 rpm. This is quite different from the LN data showing the bypass being triggered at 3000 rpm. As discussed, this may be indicating overall engine health Overlaying a lap from last session with the same oil temps shows the significant difference with oil pressure today. The accelerations are nearly the same, yet the oil pressure is much better this session. The orange line actually extends down to 18psi. I would propose the next step is to use a thicker oil as the next easier step. Certainly an oil cooler would help, but with the water temps up at 215-220, it won't solve the problem for this engine. Need a center rad. If a cooler can maintain a 20 degree offset, that would put this engine at 240. The current drops at 240 are down to 25psi. Maybe with a thicker oil these would only be down to 40 or so. WAGs. If any drop like this is truly pulling and pushing air, then all the pressure improvement does is decrease the time the bearings are without oil. Is there any different sump baffling to try?

Quotes source post: 17519177

I think Zbomb mentioned that he was going to run the oil level higher for this outing. Perhaps we should confirm or deny this before reaching this conclusion? Chris Cervelli Cervelli Technical Service

I'd also like to add that we had someone contact us a few weeks back through our ticketing system with low oil pressure on their street 996. They replaced the pressure relief and increased viscosity of oil, but that did not help. He reported back over the weekend that he replaced all three oil pumps on the engine and the pressure issue was resolved. I am curious if replacing the scavenge pumps will make a noticeable difference and also if perhaps there is an issue with the primary oil pump as well contributing to this issue.

Interesting re. oil pumps. Do we know what kind of wear they exhibit that could cause decrease in their efficiency which would in turn cause lower oil pressure?

Quotes source post: 17519859

You are right. We can't make that conclusion yet. There are 3 distinct differences between these runs: -new scavenge pumps -higher oil level -blow by return In the data, I plotted sessions that had a similar oil temp progression through the laps. This resulted in the 1st session vs the 3rd session from last time. We know the can had already collected a bunch of oil, so it was definitely lower. I'll try to pull some graphs of laps from early in the 01June session to sessions with similar oil temp to make a better comparison. Later tonight...

For yesterday’s track session I ran the level at 1/2 capacity per the dipstick. This is an increase from the 1/4 dip stick level I have run previously in an effort to keep oil out of the intake. As I am now dumping reservoir oil back into the full tube, I wanted to progressively run higher levels with the thinking any blow by will end up back in the sump anyway. This was successful as the car consumed no oil during the day. It is fair to say that with regards to resolving G induced pressure drops, neither the scavenge pumps or the increased level were successful. Yes, there was some improvement but not anything I would say is even near resolution. I have not installed the new pressure relief valve yet and am not sure I will get it done before my next day at Lime Rock on the 6th, so that is still TBD. Ambient temp was 85-90 and as you can again see from my water and oil temps, nothing even close to resembling LN’s data. As far as the primary oil pump being a concern, sure - that is a possibility but some dude saying he had a non quantified issue on his street car is just not enough to go on. I have some concerns about the engine making it through the season, which for my fun and funds, I really need it to - I am hoping I can end up with a recipe that people can point to as one that is proven to work ON MY CAR - still work to do. my thoughts on the scavenge pumps are as follows, they alone, and even coupled with an advantageous oil level were not enough, slightly better, but better is not good enough. ETA - Paul - something to note, both sessions 1 and 2 were littered with traffic and well down on time. Session 3 was closer to where I would like to be with the car...before I put it in the grass haha ETA X2 - THANK YOU PAUL. Your analysis posts are right up there with my favorite things to read on this forum. You Tube

Not selling anything - just offering advice from over in the UK. Such a lot depends on the track and the G forces. Here in the UK many circuits are relatively short with tight corners (and over here G is usually higher in slower corners). The initial drive into the corner sends oil up the chain tunnel and as the braking G starts it moves it forward entrapping it together with the oil from the camshaft housings, tappets and variocam adjusters where it all moves forward. Faster high speed corners tend not to be as much of a problem unless they go on for ever (as after the start at Donnington Park for example). Sorry I cannot keep up with all the posts but I did notice one that said 40 psi is OK because the engine is only revving at 1000 rpm. IF you are braking the G forces are higher than accelerating and will be the main cause of oil shift and if you are using engine braking as well (or just blipping the throttle on change down) despite the fact that your engine is not under power the forces on the big ends as the piston goes over TDC with no upper cylinder pressure on every other stroke - are the highest you will get on track and at 40 psi - not enough - you should keep those revs higher. i also notices something about worn oil pumps. At track or racing revs (that rarely fall below 4000 rpm under power) the pump will be working just fine and shortly after than the pressure relief valve opens and at higher revs the pump should be able to maintain enough pressure even when slightly worn. The pump pressure is about 7 bar (over 100 psi). There are two main problems as we see it over here. Many sumps allow the oil trying to move over to one side to lift out of the sump anyway and allow the rest of the oil in the sump to follow suit. It is essential that this oil does not leave the sump and a better option is a sump that holds as much oil as possible in the centre where the pick up is, and allows oil from the inside to be directed into the centre and prevents oil from the outside to run away - that way most of the oil in the sump remains there. However braking loses more oil from the sump than cornering and it is difficult to prevent that with any sump design so we favour an Accusump as an additional safeguard. having said that there is a danger because the valving available is more suited to a sum designed to re-fill the main sump than to act as a direct switch to feed oil to the bearings. This can be overcome by feeding the oil directly to the pump where it then immediately delivers it at system pressure. This usually is all we find necessary but for top level performers an added scavenge pump at the front of the cambox is also a good idea (especially with 996's where there is no slope in the cambox as on later models to allow it to drain better). However if that returns to the usual sides of the sump through the separators that delays recovery so that is why an electric pump feeding directly to the centre of the sump can be better. If oil pressure drops it seems to have two issues.- the lowering of oil pressure to the main bearings allows more movement from the rear main and allows the flywheel to move more out of alignment and can prematurely wear the number 6 or 7 main bearing which then lowers the oil level to the number 5 or 6 big end and causes it to fail. But also the gallery from the oil pump that feeds the engine empties (because the two gears work backwards with no oil being picked up) and this means that even when the oil level returns to normal - it takes precious milli-seconds to recover full pressure. Delivering oil directly to the pump itself reduces this time delay. We find that for track days over here a properly designed racing sump delivering oil to an extension under the original oil cooler is sufficient. For race cars on treaded tyres an additional Accusump is an advantage (as long as it is plumbed in correctly) while for front performance on slicks the scavenge pump is also advisable. It is very important not to fit too large an Accusump or if you are not careful you can over-fill the sump and damage the air-oil separator seals. When starting just open the valve to the Accusump as the engine turns over then shut it (as the Accusump should hold system pressure). At the end of the race increase revs to over 3000 and then shut the Accusump valve (both can be done electronically with pre-set revs and pressure switches). I do appreciate that recording information is very helpful and my initial comments about so much time being spent on looking at the results were because unless you actually fix the problem you will only ever be looking at an impending disaster. It seemed more important to me to point out the many weaknesses we have seen and experienced over here using other systems that clearly are not designed fit for purpose and my point was that if the system is built to work (and it can be) then your data loggers will never need to be used. I hope this post will be received with the intention to enable readers to consider the points made and not need to be removed as it is only meant to advise - we are not trying to sell solutions to the USA (we struggle to manufacture enough for Europe anyway). Fingers crossed this hasn't crossed any guidelines or upset too many other contributors. Baz

That is some significant improvement! Have a look at this picture where I have marked some key points: Some drops are gone and its generally heavily muted. This is great improvement. But remember, your only 1/3 of the way. Now you need to fill the oil full and possibly(seeing the wear influence from the scavenge pumps ) replace the chain tensioners. The oil system is a system, you can't just replace one worn out part and expect it to solve everything. Of course, putting the system back to its original state will not guarantee to solve all pressure drops issues, but it will improve. The question is how much. It's a good starting point before going to modifications. It's too bad you removed the AOS accumulators, it would have been interesting to see how much less oil was coming up that way. Btw, which dip stick do you have in your car? There are two revisions. One is longer, giving you something like 1/2 quart less. I have the old one without the tabs. I just checked my oil level when the car is cold and have been sitting for a few days. The level is about 1/10 inch over the maximum dot. When checked normally it's reading full(the grove).

^^^ No problem Baz, I consider this forum a Worldwide Forum (even though I think it's US based), someone correct me if I'm wrong. There are lots of small very tight turn tracks here as well, "Little Talladega Grand Prix" being one as it was originally designed more motorcycles as was "Barber Motorsports Track" which now hosts actual Indy Car races. Both are very close to me and are my favorites. To recap Baz advice:: We find that for track days over here a properly designed racing sump delivering oil to an extension under the original oil cooler is sufficient. For race cars on treaded tyres an additional Accusump is an advantage (as long as it is plumbed in correctly) while for front performance on slicks the scavenge pump is also advisable. 1 (for track days) Hartech sump with extension under original oil cooler ( need clarification on "extension hardware") 2 (for race cars with treaded tires) add Accusump(size?) with Hartech oil pump cover and controls 3 ( for race cars with slicks) add electric scavenge pump to bank 2 head For me though it would be a very needed to know when you are having drops in pressure on your particular engine/setup so you would know what hardware to install without doing the "shotgun approach" to do it all. The factory oil pressure "idiot light" is useless and the oil pressure gauge can't be monitored when doing turns on the edge. This is why I proposed the Idea/Invention of the "delta oil pressure warning" that I am working on that will use the Autometer Mini Pro Light and a Microprocessor 4 wire hookup to oil pressure and RPM to monitor the "Delta" and warn if the delta is not maintained. I pla to offer it a a stand alone device as well as included with the UIDS V2.2.. A "Smart OIL Pressure Light" if you will...

Quotes source post: 17519911

I think you stated before that NO oil pressure drops is the only reasonable standard. I had to use that standard in order to look my paying drivers in the eye. Furthermore, it seems pointless to look in more detail at main and rod bearing problems when the engine is known to lose oil pressure here and there. The relief valve is a very simple install compared to the scavenge pumps. It is not going to help though. Hopefully LN can send that data to Golock for further analysis. I strongly believe that the oil pressure sensor was not operating properly when that data was taken, and that will be easy to determine by looking at the in and out laps. A brand new perfect oil pump may do a better job than Zbomb's current pump, but I don't think anyone disputes that Zbomb's data shows that the oil pump is picking up air intermittently. It does not seem possible that a new pump will solve that. Your last sentence here is the thing, isn't it? The clock is ticking; you only have some finite number of laps with bad OP before the engine fails, and engine failures have a very significant impact on your ability to continue the test program. In my opinion, Zbomb's pace of testing is not fast enough to outrun the inevitable engine failure. This all leads me to believe that while there may be a wet sump solution that works, no one is ever going to find it, because the failures sustained during the testing process will consume all the resources needed to refine the design. This looks to me to be a dead end. Chris Cervelli Cervelli Technical Service

Quotes source posts: 17520427, 17520427

Don't be so sure, the same thing was said about the AOS for two decades, untill the UAOS, and the UAOS is still receiving improvements as we speak, some which have allowed more oil level to be used in the sump which have have actually help this oil pressure drops also. Since the lowering of the oil level was a "Porsche Factory" recommendation to the problem of oil ingestion on tight track turns, that suggestion (and redesigned dipstick for mid 2002 for this reason also) followed by most 996 owners also adds to the problem of oil pressure drops. I sure hope it is lasts long enough. I was really concerned every time Zbomb went to the track, but seeing improvements now gives hope that it will make it to recovery.

Quotes source post: 17520427

Yes - the goal should be a system that is immune to G influence - your dry sump data reflects its possible and I think many people who buy deep sumps currently think that’s what they’re getting. I am certainly trying to keep a measured approach of keeping the motor running while being able to produce beneficial data. Hence, playing with oil levels to try to see an affect even though it did muddy the test results on the scavenge pumps. I would say that muddying was minimal but present none the less. I need to keep the motor alive and I thought it likely the oil level may be more beneficial than the pumps based on the minimal wear I saw and with the lowered likelihood of oil in the intake due to the UAOS reservoir oil fill dump, it seemed like little risk for possible nice gains and I plan to continue to creep the oil level up until I have a reason not to. It is quite possible that before an oil cooler or wet sump that works is developed, my motor will not be around to test it on and highlights the importance of people with a vested interest, be it financial or personal figuring out if this is a problem they are interested in solving. I think your dead end premonition is premature Chris - you may be right, but, maybe not... The motor, for now, continues to make 120+ at the end of the Thompson straight. Eventually - it won’t and when that happens, we’ll know. Until then - I’ll do the best I can to allow the car to produce data that may be beneficial - or just a departure from the floor mats, side markers and other Nick knack threads that are so pervasive.

Quotes source post: 17520483

I think I am the only one who went far enough down this oil system road to get seriously bloodied. I wish you guys luck, but I don't think you are positioned any better than I was to absorb the financial devastation. I think that part of it winds up being more important than technical expertise or inventive inspiration. I blew up between 15 and 20 M96's in 3 years. That seriously interferes with the quality of a person's life and puts a lot of pressure on you to give up. I know I always sound like a pessimist, but I think my best value to you guys right now is letting you know what you are up against and helping you not waste money and time on stuff that has no chance of working. PorscheTech's swinging valve sump might work, but I don't see how it can be prototyped and tested without serious $$ behind it. Chris Cervelli Cervelli Technical Service

Great to see the data from the new pumps. Its surprising to see an interest in dual stage pumps. We manufactured some over 10 years back after the Porsche pumps became NLA. We had a large demand from the Cayman crowd, but when we manufactured the pumps people didn't seem to come running after them. Which was surprising to us with the advantages these pumps provide. We actually still have a few on the shelf. We made them for both banks. Admittedly, the fitment on bank 1-3 in a 996 was very tight. We never redesigned the part for better fitment in a 911 since they didn't sell very well. If the demand was there it is something we could certainly look into.

Can't emphasize enough that we all know our 996 engine is not an optimal track engine, and will eventually need a rebuild... especially if you track the car...or just don't properly service it. Knowing that it will need to be rebuilt, the Holy grail for many of us is to find ways to prolong engine life on and off the track until a rebuild is needed. Finding ways to reduce oil pressure instability, oil starvation, etc. Is a logical step along the continuum. Yes, it's a street engine, but maybee they are more durable than many would believe if you do the right things. No, not too run the car at LeMans for 24 hours straight, but to DE the car a number of times a year and drive it around town. My 2002 is a good example having tracked it from 2004-2018 very consistently for 14 years with only a deep sump, Motorsports AOS, and maniacal oil changes after each event, while running sport cups and Hoosiers in the fastest PCA run group for 7 of those 14 years. Have I been lucky? Sure. But what it shows is that the deep sump, better aos and maniacal oil changes and over- servicing certainly didn't hurt. So if I add a 3rd radiator, oil cooler and find a new or even better set of oil pumps or dry-sump set-up, maybe I can get a few more years out of the engine before a rebuild. Or maybe not. Ultimately, a comprehensive rebuild is the only way to solve the problems. But in the meantime, what everyone is showing here with the shared data, looks to me like incremental enhancements that may partially help with oil stability and as a result, may prolong engine life until the inevitable occurs. But let's be very clear, none of these "bolt-on" parts are a comprehensive fix. But they may provide potential incremental improvement. Hence the logging and measuring of data to see. I like where this thread is going. Nothing to loose and everything to gain for all of us. It creates solutions along the engine continuum from incremental "bolt-on" solutions to a comprehensive engine rebuild that addresses everything. All the 996 owner has to do is make a choice.

Quotes source posts: 17520610, 17520616

I hear you - and I will leave the development and test stuff budget to guys with a lot more experience than I have. For a few people I really trust, if they asked for my help testing, and by help I mean running the parts and collecting data, I'd take on a calculated risk to do that, especially if it helped defer some of the prohibitive cost to a MFG to get a working solution to market. I really appreciate the data you put out, for free, that could allow others a head start that you had to pay for - pessimist or whatever, that's all I need to know. It has, and would continue to be really beneficial to have your input and experience. I think the issue with the dual scavenge pumps is two fold: 1. Install - having to modify the valve cover really, to me, makes this a do while rebuilding mod. In that case, I think it makes a lot of sense to enhance the platform for what seems to be a generally agreed upon area in need of improvement. 2. Is it enough to resolve the issue in and of itself, is the cost and install worth it when there is no data that says - On this platform, with this specification - It did THIS. I'm a sucker for cool billet parts, so I dig it...both yours and Martin's look like well engineered pieces of kit, but for me, I think it would be more of a refinement mod rather than one I would expect to fix the oil where it needs to be problem. However, I could be totally proven wrong on that as it is only my opinion.

Quotes source post: 17520610

Chris, I feel ya, I understand your pessimism now .You have given the much sacrifice to the Porsche Gods. May the Porsche Gods now bless you !!

Quotes source post: 17520618

I want to address this post because it gives me an opportunity to sound optimistic. These engines appear to me to be indestructible in the racing environment if they don't fail due to oil starvation. I NEVER had any mechanical issue with an engine that was not main bearing related. No valve train failures (except one which was self inflicted), no IMSB failures, no dropped cylinders, no cracked heads, no rod or rod bolt failures, no broken chains, and no failed tensioners. I do not doubt that some of these occur on street engines, but they have not happened to me on race engines. My engines got over revved, ran for hours with the coolant at 230F, completed 24 hour races at an avg rpm of 5600, ran backwards when cars spun out, and never seemed to care about any of it. They are easy and cheap (in Porsche world) to build and require no maintenance. They also don't leak. I always started with the highest mileage and cheapest street engines I could find in junkyards. Without fail those engines were in near perfect condition when I disassembled them even with 200k miles. I don't consider them a POS engine. Chris Cervelli Cervelli Technical Service

Quotes source post: 17520831

Chris, appreciate the comments and all the wisdom you have contributed to this thread. In one way shape or form, there is a market for your dry sump.

Quotes source post: 17520831

I agree 100% with most of what you said, except that working for a Porsche Dealership I have seen ALL the mechanical failures you mentioned on m96 engines. With 45 years working in Automotive Dealerships I have seen EVERYTHING fail, even cupholders..lol

We have found that a "belt and braces" approach is less expensive in the long run but it is important that the money is spent on the right areas. We also have high speed data recorders and cameras and agree looking at the oil warning light is a complete waste of time when there are over 100 crank revolutions/second. It took us some time to identify the pressure drops and realised the only safe way to protect the crankshaft is to supply oil at system pressure before it drops by more than 1/10th of a psi (which a properly constructed Accusump does if connected directly to the oil pump). If it doesn't connect to the oil pump then there is an inevitable delay in oil pressure recovery. Among our own "belts" are also two types of 8th main bearing (to support the overhung end of the crankshaft). Both align the out of alignment RMS hole with the crankshaft but one machines into the RMS hole and fits a ball bearing while the other is a plain bearing in an eccentric ring and fits into the original RMS hole after measuring the eccentricity. Our braces are a wider better quality big end bearing that adds both width, and a better contact patch achieving around 25% better support. We regard our role in trying to solve these problems as needing us to run engines until they break (sad for reputations but ultimately the only way to prove solutions) but with all our modifications - so far (touch wood) we have only experienced one crankshaft failure in several seasons of race and Championship winning performances and that ironically resulted from a failure in the tappet housing which allowed the part with the oil supply hole to crack off and the oil pressure to drop in an engine that was due a rebuild after that final weekend anyway. Nice to see so much attention to the subject over there - we thought we were struggling when a few years a go we only managed half a season without requiring an engine rebuild but judging from the frequency and interest I guess we might have managed to do better over here than you guys have over there and that is why we provide the free information knowing that some clever engineers over there are likely to understand the benefits of what we do over here and provide their own solutions along similar lines.] Baz.

Here is the best comparison I can find between an early lap from the 01June 1st session and a lap from 27June that have the same oil temp. This early lap from 01June should represent the closest oil volume condition to what we used on 27June. We know there is a difference of 1/4 range on the dipstick. Additional difference will be whatever had already collected in the blow-by can by the 4th lap. Executive summary: It's not clear what the new scavenge pumps are adding to the oil pressure. There are a couple of peaks that may be showing the scavenge pumps are helping, but overall, the improvement is not as dramatic as previously shown. The 2nd lap, session 3, from 27June was slow due to traffic, 1:31, as seen by the RPM drop between 1000 and 1500(dark blue). The 4th lap, session 1, from 01June was faster, 1:25. The G forces in the 01June lap are the same or higher than the 27June lap. The oil pressure drops in the orange trace that are not explainable by increased G's are: -at 3000, the G's for both traces are about the same, The change in G's is quite different. The yellow trace shows a more gradual slope to the max G's for the turn. It's not clear why there would be such a large drop out. -at 4900 there is a another drop that cannot be explained by the G's, both laps show the same forces -at 5300 shows a similar problem, both laps seem to be applying the same forces, yet there is a drop all the other drops that are larger than the 27June lap correspond to higher G's, either lateral or in-line Apologies for the misrepresentation yesterday. Thanks to Chris for pulling this in.

Here's the companion video for the 27th lap Paul used above, I will see if I can add the lap from the 1st after work. You Tube

Quotes source post: 17520610

Just to add more context to my replay to this post, you can be assured that when'/if the UIDS V2.2 is available to purchase, it will be fully tested. We have already done 3d printed prototypes, done bench testing and flow testing, and have collectively spent over 300+ man-hours in the last 9 months and haven't put it on a car yet, but I can't wait to put it on mine !! The " Delta Warning Light" ( Smart OP Light) is all I need to know how well it's working. I'm not sure what is meant by " serious $$ " but the 300+ man-hours is pretty serious, and we are not doing to make a ton of money, in fact I hope we break even. We are doing it because we think we can and it needs to be done.. I didn't want to post about the idea/Invention 9 months ago because of what I experience with the UAOS, it was very DRAMATIC !! On one side I had people praising and ecstatic about the solution to a two decade old problem, and the other I had skeptics and nay-Sayers condemning it for lack of sufficient testing and wanting years/decades of "evidence" that is was OK... It was very distracting and disheartening when also dealing with the logistics of sourcing parts,materials, sublet of some tasks, Patents, shipping , costs, all at the same time.....almost made me stop caring. Just like with the UAOS, it won't make much profit all said and done. in fact we could make more $$ cutting grass/lawn care .... We do it for the love of the platform.

I remember. No good deed goes unpunished! Can't wait to see it.

I think we all share the love for the platform, that's why we keep going. I got the question earlier in the thread if I had a temperature sensor to read the oil temp. The answer was no at the time but I have this week built, calibrated and made a test run. It's a complete stand alone sensor which take the temp from the bottom of the dipstick tube. Accuracy is +-0.5c. The installation looks like this. The road I drove today(not my normal test road) looks like this. 305m(1000ft) climb, all in 2nd gear going between 3k-7k(mostly 4-6.5): I made the drive twice and this temp graph is from the second run, hence the oil temp is already above 105c when starting. X-axis is seconds. Red arrows indicate start/stop of the run and blue arrow indicate something interesting. Water temp reading on the A/C unit topped at 106c, ambient temp 27c at the bottom. First blue arrow I got stuck behind a uber-eat scooter and second I stopped to take the graph screen shot. So, low RPMs in both cases(around 2k behind the scooter). Why does the temp shoot up high when the RPM is low? My guess is that when the RPM is low oil level rise in the sump and hot oil coming from the short block flow out of the center baffle and heats the sensor(which is located on the outside of the center baffle). The oil that returns from the head is what the sensor is mainly reading. The heads don't heat the oil that much so it's a good indication of the oil's temp going in to the gallery. So what does all of this mean? I think the running oil level at high RPM situations is significantly lower than what we measure on the dipstick. It also indicates that hot oil does not leave the center baffle in high G turns(if it was we should see temp spikes in all right handers), indicating that the original baffling works well . (Edit: on second check there seams to be a correlation, need to look further at that). Any other guesses? Oil I'm using is Ravenol VSW, 0w-30. It a thick 30 weight, just a bit thinner than M1 0w40. When stopping at the top to take the screen shot the oil pressure warning lights up at idle, so maybe not for everyone. Running pressure is good tho.

My guess is that windage and aeration is occurring at running speed, but when idling, windage stops and the dipstick sensor is now fully covered in oil.

The purple lines indicate the same slope(heating). I think the sensor measures air temp between the lines and at the lines it's submerged and starts measuring oil temp. At low RPMs it becomes submerged and that's why the temp goes up.

Nick - I know you like this stuff, spend some time to piece together a robust data collection system. You’re smart enough to figure it out and probably to figure out how to do it in a minimally invasive way if that is important to you. It will give you all sorts of ways to geek out, like I do, at what your car is doing. And - it would be a great addition to the currently very small pool of data. I don’t know enough to comment on the validity of a sensor in your location but if there is any concern it’s picking up air, you’re not getting the picture you need.

I assume the best place to put the sensor would be near the bottom of the pickup tube, that way if there are any signs of it reading air temperature you know you have an issue anyway! I have heavily instrumented the coolant system on one of my cars, but not the oil system. I just rely on the built in sensor for that. Martin

Quotes source post: 17524585

I really like the coolant setup you�ve put together. If you�re thinking of doing something similar for the oil system temp and pressure - my picks for locations would be : Temp Pre cooler Post cooler Pressure Pump outlet Pre cooler Post cooler Head

With regards to fitting the dual scavenge pump I agree that having to take the cam cover off is something to be considered- I fitted mine when I changed the right hand bank VarioCam solenoid as it had become rather lazy, at which point it was a minor additional task.

Quotes source post: 17524567

I don't have infinit time to work on the car so what ever I do needs to be efficient, time wise. To put the temp sensor in this position have given me not only the temp but also indication of other issues which I was suspecting(as you can read in this thread where I wrote about the need for better oil level checking). It also shows that oil returning from the short block doesn't escape in any large quantity = baffling is sufficient. Data is only good if you can interpret it.

I find it a great shame that I didn’t realise that over in the USA you apparently had a severe problem racing the M96/7 engines with crankshaft failure caused by oil pressure drop off. If I had it might have been worth paying the Rennlist sponsors fee to allow me to explain to you all what we did to overcome it over 10 years ago and how reliable the outcome was – but if I do that now my post will be removed because I don’t pay the fee – which I don’t want to do because I have had no intention of selling our products into the USA anyway and don’t like the idea of paying to provide free advice! Although we have our own machine shops it is so busy re-manufacturing M96/7 engines we cannot find the time or stand the disruption to break down the set-ups to make a very small number of the special parts that resulted in us solving the problem for only a small number of race cars used over here (much less than race and track cars in the USA) – so only occasionally knock up a small batch for the UK market and then only for engines we build here. So I have nothing to sell but I will confirm that in our experience you need both a deeper sump (that restricts the flow of oil sideways or as in your new idea over there that collects oil via the pick up form the outer side) but we also found it needed an Accusump (we don’t sell those either!) because under braking too much oil flows forward to restrict it all and sideways control alone we found insufficient. We then found that there is a right and a wrong way to use the Accusump – the wrong way is almost worse than not bothering the right way works brilliantly. if you have a downer about them over there you are probably not using them correctly! The market over here is probably 1/10 to 1/20th of the size in the USA and we do not have the capacity to supply the USA which does not therefore justify advertising the products but the advice from our experiences and the solutions we found worked – it seems - could have saved a lot of people a lot of heartache. It would however be nice to help solve the problems for someone over there which would save the unnecessary cost and pain of continual failures. We have no stock to sell at the moment anyway but do have a free technical report that explains it all (and other issues like oil temperature control, 8th main bearing support etc)– and that is available from [email removed] . I do hope that in the interests of the owners over there this post is not removed – I am not trying to sell anything to you and I think there is an opportunity for someone to help with advice from over here that will help you guys over there without there being any ulterior sales motive and no attempt to harm any Americans or businesses who do pay the fee to advertise in their own Country. Baz

Quotes source post: 17506007

The main oil pump is set at 7 bar it is the losses in the passage through the engine that result in it reading 5 bar on the cylinder head where the oil pressure sender pick up is situated. Baz

Quotes source post: 17526226

I also found that the Rennlist Vendor fee structure made no sense for my business model. I'd like to see if I can clarify what Baz is saying here and then contrast that to the approach that I have promoted in my own threads on oil problems and what seems to be a very similar approach used by Zbomb in this thread, I think Baz is telling us that he made technical changes X, Y, and Z and after doing so the engines stopped blowing up. Instead of making a similar claim I posted oil pressure data to prove that I had solved the problem of oil pressure "instability" as it is called in this thread. I can't imagine using Baz's approach today, but at the same time I am certain that I had success with that approach throughout my racing career, especially in the area of chassis tuning. Furthermore, it looks to me like most of the progress made with machinery in general over the last 200 years was made using an approach closer to Baz's than to mine. There are many people stating that they used their 996 (or Boxster) on the track in a certain way for some number of miles without having an engine failure. This IS a kind of data, but I suppose we all recognize that it does not universalize. For instance, I tend to dismiss any claim that does not approach my particular use case (endurance racing with dumb pay drivers) in intensity and difficulty (including the financial penalty for unreliability). Also for instance, Baz's data is suspect for me due to the simple fact that is never hot in England, and I often ran cars in 100 degree heat at high altitude. I wish there were a way to capture Baz's and other's contribution here. I suppose that is mostly up to him, but it seems he already has satisfied his own criteria, and he doesn't owe Rennlist anything. I would call this other category of data I am talking about here metadata. It seems to me that the metadata, as presented on Rennlist, on M96 track engine failures has in fact led nowhere, in that to my knowledge no one had taken this seriously with any success until I did in 2016. I also know that without my Motec data I would have also gotten nowhere. From these I am going to conclude that for me, these problems get solved with the use of real data. However, I do not reject the idea that the metadata may have been useful. Chris Cervelli Cervelli Technical Service

Man this thread is one of my favorites of the past 5 years. Great stuff everyone.

IMO there is more publicly shared, data supported information in this thread than in any other thread on any website I have ever seen. I thank everyone who has had a hand in creating, analyzing, sharing, and critiquing this data.

Hi Chris – didn’t fully understand the term Metadata but anyway I think you have misunderstood my points but thanks for your contributions anyway – very interesting. Anyone interested in pure engineering of solutions should find the following worth plodding through especially as I got up at 4.30 on a Saturday morning to write it! lol. Can I clarify that just because we can figure some things out without using data graphics – it doesn’t mean we are in the dark ages over here. Although we had no data logging when we started racing 10 years ago - we currently have data loggers running without direct connections remotely to the Can bus, sending information via the cloud to computers analysing everything from power, torque, oil, pressure, etc etc and even showing maps of where it took place – time based etc. We have monitoring equipment on our test cars running direct to i-phones and of course our own diagnostics – we do use data - extensively. However - without taking this too seriously - I would like to take you to task over your claim that publishing data proves you have solved the problems whereas having no crankshaft failures doesn’t. Even racers with no protective systems managed some races before their engines failed. Building and Running engines in Championships that finish ALL the races and win the events year after year - does prove you have solved the problems, whereas a selected screen shot of a successful track performance does not (for all I know you could have been going slowly and not shown the data from when your engine failed). I am not for one minute suggesting that is the case – I trust your being honest about it – but in terms of semantics – I think consistent public performance should never be discredited. We didn’t start Hartech to grow continually nor as a race car or component supplier – but to enjoy working with and selling road going Porsche cars in a manageable sized business – for enjoyment – and have achieved that. However most of what I know I learned through the competitive nature of motor sport and so there came a point when we thought it would help staff development and provide a valuable test bed for proving engine re-designed parts – to have a go. We had never raced cars when we started 10 years ago and didn’t take the easy route of buying a successful car – but together – built one from scratch from first principles. We looked at all the available stuff for engine protection and found it wanting. The most obvious mistake we uncovered (still available today) was sump extensions allowing ALL the oil in the sump to escape outwards and upwards under centrifugal force in corners through the very holes put there to allow it to re-enter the sump, combined with the weakness of the original sump material – so we made our own. It has 3 chambers right, left and centre. Any oil travelling towards the outside in a corner is prevented from leaving the outer chamber by valves. Oil from an outer chamber on the inside in a corner - is directed and fed into the central chamber (where the oil already in there is prevented from escaping) and this provides a continuous feed to the oil pick up chamber. It would only fail if the drop in oil level lasted too long and demand exceeded the quantity of oil available. These have been available for 10 years and are race proven. However we also tested what would happen to the oil flow under braking and decided that a significant issue was forward motion that combined with the radial motion would still possibly empty the sump a bit – not side to side (we had fixed that in the design) but forwards under braking. As a belt and braces exercise we fitted a manual Accusump but not too re-fill the sump but made spacers to fit under the oil cooler and feed the oil under pressure directly into the main oil gallery closest to feeding the crankshaft. We didn’t know if this was necessary but it all worked and all our engines had superb crankshaft reliability (we assumed our first season would expose our lack of experience – but thought the learning experience would be worth the pain – but also totally unexpectedly won at our first race meeting despite building a car from scratch with absolutely no experience or copying anyone else) but the downside was that we were immediately exposing our ability and our cars to the highest level of competition and had no smooth progressive learning route through to the stresses, strains and set-up problems we had to overcome. But we managed it somehow and after three great seasons learning more about handling and preparation, we moved up to a 996 with the same lubrication system and it worked perfectly as well. Just like a dry sump – it had an external reservoir (the Accusump) providing a continual source of oil to the engine but under pressure instead of using the oil pump to create it. All this engine oil supply system was accomplished without any data just good engineering analysis – but don’t despair – your point will be fully supported shortly because despite building a beautifully handling Boxster S we got really stuck with the 996 handling like a pig and had to resort to fitting potentiometers connected to a data logger – which them clearly showed up our error and which helped us end up with probably the best handling 996 in the competition. We also provided the lubrication system to others who also went on to success. Then we took a year out while a customer with a Cayman S (who didn’t fit an Accusump as recommended) had crankshaft problems. The following season we decided to run a Cayman S instead to find out more and discovered that we couldn’t have fitted out oil cooler spacer for an Accusump anyway because of a different and lower inlet manifold and also the rules had changed and we could no longer open and close the manual connection to the Accusump from inside the cabin. While thinking up where to connect the Accusump to and how to run a shaft through the rear bulk head our team came across a system (strangely originating in the USA) to automate the whole Accusump manual valve system and change from a coolant/oil cooler to an air oil cooler – connecting the Accusump into the pipework and I was happy to let them get on with it – exciting new technology ahead of my old fashioned methods - but it didn’t fix the problem. Our evidence was rear main bearing and big end failure. We assumed the higher BMEP also meant that under braking the crank was flexing more due to the new variable cam timing and valve lift systems (with the existing long hang out to flywheel) and that was wearing the rear main bearings which lost oil pressure to the big ends that being more highly loaded – then failed. So we fitted an additional rear main bearing and that stopped the main bearings from wearing at all – but still we had a big end fail. Then we realised that it was the big ends that were wearing first and that lowered the oil pressure to both the big ends and main bearing and that it was the drop in oil pressure that was allowing the crankshaft to flex more than when the big ends were still surviving – it was the other way around to what we originally assumed. I don’t think there would have been a way to connect a data logger to have worked this out (perhaps I am wrong). Our driver hadn’t noticed a significant oil pressure loss on the gauge (but thought the oil light might have occasionally flickered) so we fitted a higher speed camera and reviewed it frame by frame then saw it was very brief but did still exist. So we then fitted a data logger and picked up the pressure drops. However we were not bothered where it occurred or on what track or speed or even what “G” – it didn’t make any difference to the pragmatic reality that it had to be solved mechanically/electrically permanently for any circuit anywhere. Our clue to the explanation was why was it OK with the old system before and not know? Then I was frankly astonished to discover this new system only opened the Accusump at about 2 bar (from memory) and that the highest rated pressure switch available was about 55 psi (3.5 bar). I would not expect a racing engine big end under engine braking up to 7,500 rpm to survive with such low oil pressure anyway (or the higher version) – that was the issue and the problem. Under engine braking you force the deceleration of the crankshaft to be many time faster than it accelerates and so the big end loads over TDC increase dramatically and I couldn’t see any logic whatsoever in a system that waited for the oil pressure to drop and drop before doing anything about recovering it – it had to be instantaneous to make any sense to me and this had not been. We then tested the relief valves and found they opened at 7 bar (over 100 psi) and that there was a 2 bar pressure drop between the pump and the sensor in the cylinder head (didn’t need a data logger for that either). We also ordered a large batch of wider racing quality big end shells to support the big ends better with the larger capacity engines our series was now allowing to enter the events. We also realised that we have added a huge length of pipework to the air oil cooler and volume inside the air/oil cooler and that also takes time to recover oil pressure (the larger the volume the longer it takes) so since it had not lowered the oil temperature any better than it was before – we decided to remove all that added volume and go back to the original coolant/oil cooler (already had a 3rd radiator so coolant temps OK) and to connect a switch to the Accusump EPS pressure valve so we could make it be permanently open when we were racing (but retain the pressure control to the switch working for non-racing situations to prevent over filling the sump). However we could not now connect the Accusump either to the original Boxster S 3.2 and 996/7 spacer under the oil cooler (because the inlet system which our regulations didn’t allow us to change) was in the way – and we no longer had a pipework system to connect to – so we needed a new entry point for the Accusump. We then realised the advantage we would have if we made a new oil pump cover to allow oil to enter directly to the pump itself. This now worked exactly the same as the old first reliable system – perhaps even better because we also worked out that when the oil pick-up fails briefly the oil gallery pressure falls and the Accusump has to re-pressurise it to return to system racing pressure – and that this takes precious time – and while it is doing so - as the Accusump delivers oil under pressure – its own pressure falls until enough oil returns to the pump to be pumped back up to system pressure. By delivering oil directly to the pump (even if the pressure falls slightly) it is immediately pumped up to system pressure by the pump and with no milliseconds of delays whatsoever. It works like a dry sump except the oil feeding the pump is already at system pressure for even faster system pressure recovery. All this thought out without any electronic data collection – just boring brain power. This was now working similar to but even better than the original reliable system – the Accusump holds oil at system pressure (say 7 bar) but if the system pressure falls by even the smallest fraction oil has to flow straight into the pump which immediately restores system pressure – i.e. there is no possible delay. We had finally solved a new problem with the Cayman S that we had originally solved without the use of any data for the older models and 997’s, but data did help us see that the problem was over very short periods of time but still relevant. After that it was good engineering experience and innovation that improved the whole system even further. Trying to solve the Cayman S problem meant we had now returned to a system that had an 8th main bearing to stop the crank flexing, an oil pump housing that immediately would result in system pressure being maintained and much higher quality big ends that will last longer. We may now have gone “over the top” as a result of our original mistake when we changed the system – but have no regrets – racing is expensive enough as it is and engine failures are the worst possible handicap that demotivates everyone and costs a small fortune to repair. If we didn’t have to pay to advise others about these technical improvements I think we would have revealed all this years ago and many Americans would have understood all the issues we uncovered, benefitted from our experiences and found their own ways to make something similar to work (or reject things we had now explained that were mistakes in other products and solutions on their market) but it was assumed (wrongly) that we were trying to discredit others and sell our own products over there or just trying to prove we were better engineers – none of which was true – but it did prevent our experiences being made available to a wider International audience – intended entirely for their benefit – not ours and for which we were not prepared to pay for the privilege. I think this new valve idea to move the oil pick up point to whichever is the outer chamber in a corner is a sound idea and a similar solution to our valved system (possibly more complicated and expensive?) - but if it maintains the same volume of oil being available to the pick up system as ours – will be just as good and if it contained more - might be better (by not needing an additional Accusump) - except that we found braking resulted in the highest “G” forces and oil also escaped forwards in the engine – which our Accusump added protection for. So I am not sure that – on its own - it will stop surges due to heavy braking for which the added comfort of an Accusump system (which is American anyway and we do not retail – so no benefit for us to say so) - providing the system is setup to use it properly and not make things worse (as described above). Our racing solutions were not designed for commercial profit (in fact racing always costs us time and money - including developing and making the parts) but for technical development, training and improving the quality of our customer racing engines - so we have no great interest in breaking down our engine re-manufacturing machinery set-ups (that are always flat out) to make small batches of parts or advertising and selling them. Furthermore – for me personally - racing was the one sport I originally found to be relatively pure because no amount of bull ****, advertising, promotion or manipulation would change the outcome of the best handling, fastest and most reliable vehicle and that took pure engineering. We do realise however that the American market is many times larger, seems more commercially based, that there also seems much more National loyalty to home products and services (which is commendable) and that there seems a reluctance for individual owners to work things out for themselves and rely instead on slick marketing (which Americans are very much better at than us and have the market size to justify). However this does result in us sometimes reading about glowing high tech solutions that we can see obvious flaws in that seems to go unchallenged – supporting sales over there often at the cost of the owners - irrelevant claims to have been the first to solve something or to have sold more than anyone else - seem to take priority over the quality of the engineering solutions (of which different IMS bearings is a typical example). We are different over here at Hartech. We don’t seek any commercial expansion and are happy to advise people to help them make better decisions regardless of the outcome influencing our sales (and our free technical reports and videos are not sales documents, don’t include prices or availability and are pure engineering technical explanations to help others understand things and make better choices – not necessarily from us). There also seems a lot of posturing over there about who is personally the best at anything rather than which product provides the best and most proven reliable solutions. You will notice that when we regard a competitors products as good we still have the integrity to say so because the problems these engines present to road users and racers are huge and Porsche enthusiasts deserve better than pure commercialism to support them out of the consequences they should never have had to suffer anyway. That is rarely reciprocated except by other good dedicated engineers like Jake Raby. As I have frequently tried to explain to forum controllers - we have never had any problem helping fellow racers Worldwide with explanations and solutions that we can provide or they can go off and sort out for themselves without it having any direct commercial benefit. It is for us a band of motorsport brothers Worldwide and we think we should all help each other but unfortunately when we try it seems to degenerate into posturing, defensiveness and a competition about who is the cleverest – in some quarters – which frankly we have no interest in. The lesson for me is perfectly clear – you don’t necessarily need data to solve problems – it helps but you need experienced people working out what is going on as well, innovating solutions and sharing their experiences in an open and not just competitive way. Let’s leave the competitiveness to the drivers and on track and help each other off it – that is our motto. The American track and racing market is probably big enough to be significant for business profit over there but for us over here it is not – just a cost – which we indulge in out of a natural competitiveness and a spirit of a challenge, staff development and adventure. I wish you all the best with any new oil pressure instability solution and hope revealing our own problems and solutions might have helped. Having had problems and finding “this” did or didn’t work, to do “this” or not to do “that”, to avoid “this” and make sure about “that” etc is something we have always been happy to share without personal, professional or commercial interests being involved in the belief it would help everyone (including competitors) and as a result will benefit Porsche owners everywhere. Perhaps we are just too old fashioned! Baz

Baz, First thanks for sharing your successes and failures. it�s frustrating to think many here on RL are struggling with issues you've been able to resolve or mitigate for a while in different ways, and due to forum fee structures, or vendors protest that we�ve been in the dark about until now. Second, since this isn�t really core to your business model, perhaps you can partner with someone in that business like LN or Skip @Porschetech3 to license your 3 chamber sump design, oil cooler, and accusump integration so they can manufacture and sell and you get to collect incremental revenue. Everybody wins. Edit. Thanks to @CTS as well for sharing his solutions for the 986/987.

Thanks Tom, your response tells me I managed to get my message over - which is nice to know. I agree totally with top specialists working together and have in fact bought products to test and retail from two of the most prominent American suppliers but the situation has not been reciprocated - even though it was implied it would be, I actually love what I see as a more colonial, adventurous and competitive overall general frame of mind and attitude to life that I feel back from America and that feeds through to access to investment capital that unfortunately is almost impossible to get at the right cost in the UK (if at all) and is behind our poor International industrial performance. There are plenty of innovative lively and industrious businesses over here but premises are incredibly costly and trying to expand becomes extremely difficult to the point where many (like us I admit) eventually give up and accept a modest living despite often ideas and inventions that could be much more commercially exploited. The only downside to your more forward thinking and National loyalty seems to be a reluctance to imagine anyone else from another Country could possibly design something worth taking note of (as in this case). I think that if we had access to funding that is available over there we would be a much bigger organisation with an American outlet that would actually help you guys over there - but this is not something it would be justifiable to go looking for. My book (not trying to sell it by the way) "The pursuit of Dreams" that follows my experiences from leaving Uni to becoming director of engine development for a British plc trying to rebuild our once World dominating motorcycle industry - reveals exactly what is wrong industrially over here and why these attitudes we now also develop (to invest in property etc) become more attractive than manufacturing - sorry - I could go on. We have a balanced business of 16 people and all the machinery, sections and factory space work perfectly and efficiently producing re-manufactured Porsche engines and components almost 1/day for largely the UK but also World markets - but to make it bigger we would have to move premises and at least double everything and the funds are simply not available so we limit what we do to what we have and have to be content with that. I must say in many areas of these engines issues I think we have many of the best solutions available and certainly have a reliability record second to none. There might be some light at the end of the tunnel with a patented engine combustion optimisation idea we have that we are currently perfecting and could have some bigger potential - but there are many others will have been down that path and found it too hard to convey the ideas to others or get the rewards for their efforts - we are not fooling ourselves about the prospects. Anyway your response was refreshing Tom and we can only just wait and see if anything comes of it. Meanwhile anyone can obtain free technical information on a variety if issues with the M96/7 non GT or Turbo engines and even the 9A1 Gen 2 versions as well by contacting [email removed] Thanks again, Baz

Thanks Baz. Great info. I know it's not in your business model, but would love to see you team up with US distribution to provide access to your solutions. Couldn't ask for a better time with 996 prices now at levels that make many of the enhancements we have discussed on this thread a "must have" versus a "maybee". To put it simply, no one wanted to spend any money on a $15k 996 Carerra a few years back. Today's rising values change everything. Seems to me we are experiencing a 996 renaissance in many forms.

Not to side track, and I don�t expect builders to reveal their solutions. But if opting for a an eventual rebuild, are these oil pressure problems solvable, or even with a properly built engine oil pressure could still be a problem.

Quotes source post: 17530397

I have seen no data that demonstrates improved oil pressure stability through practices taken during rebuild. I would be interested to see if via o-ring, sealant or some other method shoring up the interference fit oil passages that have been said to contribute to internal pressure leakage make a meaningful difference.

So this weekend in the Porsche Club Great Britain Championship at Oulton Park - there were 30 entries, 10 class 1 (996 3.4 997 3.8 and 3.4 Cayman S), 14 class 2 (Boxster S 3.2’s) 6 class 3 (Boxster S 3.2’s) all with different regulations from slick tyres for class 1 to almost standard road cars with safety features and treaded tyres in class 3 lapping 8 to 16 seconds slower than the leading class 1 cars (getting in the way when lapping and often causing safety cars or even red flags which ended both races yesterday. There were only 3 cars with our engines (a 997 3.8 and two Cayman S 3.4’s) with allowable modifications to the oiling systems (as described above). Generally engine specifications are standard except for issues that improve reliability but not power or performance controlled by dyno testing. Our results were 1st, 3rd and 4th in race 1 and 1st, 2nd and 3rd in race two with the cars involved leading the Championship after 5 rounds (and the winner being last year’s Champion with the same car). Our own car managed 2nd and 3rd and a fastest lap (wet first race and dry second) with a completely new race technician team from us this season doing very well indeed with the suspension - in the conditions – just disappointed that in the last race with two laps to go they had managed the tyres well enough to not only get the fastest lap by close right up to the winner when a red flag ended the race 2 laps too soon – that’s racing. Tj - You cannot achieve this level of reliability at the front of a very competitive Championship if your products do not work. This kind of reliable performance and Championship wins with our products has been common over here for 10 years but you can see from the small number of cars involved (and with parts we made for them lasting so only needing making once) it is not worth breaking down a machine on continuous production to use with often 8 to 10 different settings and/or jigs and fixtures – to manufacture a batch of perhaps 10 racing products that might last us 3 years. If we had batch sizes of 30 off or more it could justify it and we could probably plan to fit it in over here – but we would not sell enough in Europe to justify the numbers – and the American Market seems very problematic to get into in quantity – so we haven’t tried to. Hope this explains why we do not post to sell products but to help - the solutions are not rocket science – maybe we need a rich American agent? Baz

Quotes source posts: 17530397, 17530437

The complete crankshaft carrier on both sides and the mating surface is the inside case, are a machine fit from the factory and are (according to Porsche shop manual) to assembled dry (no sealant). This is a HUGE potential for internal leak of oil volume and/or pressure. Any slight mistake in machining tolerance, surface imperfection, warpage due to heat, ect. can cause a big problem in oil pressure or oil control, too much oil leaving the sump and into the "loft" area of the case/heads ect. And can even cause a "light rattling sound" at around 3-4k rpms on light load. There was a Porsche TSB (Technical Service Bulletin ) issued for this "noise" on 986 Boxster, test procedure was to install manual oil pressure gauge in the head port, warm engine to operating temperature and observe gauge reading. At 4k rpm oil pressure was required to be above 4 bar, if not, replace engine. Porsche is very serious about NOT applying sealer where they say not to. Back when we were having bokoos of RMS leaks,( and Porsche had re-designed the RMS 5-6 times) we had a tech use an RTV sealer on the outside of a RMS as a insurance again a leak o the outer perimeter, on a Porsche Driving School car that had already had 2 RMS/s replaced, and an Engineer from Porsche came down to investigate and install a 3rd new design. He came apart on the Tech that done the previous RMS for using an RTV . The moral of the story, Porsche doesn't use any sealer on the crankshaft carrier to engine case connection, but on my personal engines and I'm sure most all engine builders, it will get a sealer. Also the same situation apply's to both lifter carriers to head connection . full length of the engine,, machine fit dry , no sealer. Porsche... My personal engines get sealed. These internal leaks can vary engine to engine as they come from the factory, or as they age and heat cycle, or if they ever experience overheating. So each engine can have a varying degree of unwanted internal leak. A very exact test of the oil pressure at specific temperature would reveal if the internal leakage is excessive. Internal leaks can contribute to "oil pressure drops in corners" but IF oil pressure is good (exact), there are other factors at play. As Baz said even with his deep sump, it needs an Accusump. I am not a fan of Accusump, I bought one decades ago when they came out. Yes it's an "American Invention" that was first intended as a " oil primer" as everyone learned most engine bearing wear occurs during start-up. Then it was popularized again with the popularity of "Turbo Kits" to use as a primer for the turbo and also to use after engine shutdown, half capacity was used after shutdown and half reserved for before startup. It has also been used for Track oil pressure drops, but the success is mixed. The problem with it is the "recharge must take place", if 2 qt is used to fill a dip, then 2 qt must be extracted from the main oil system when the dip is not occurring. This effects the Delta of the rpm/oil pressure and volume. It acts like a big "damper" on the oil system, removing delta from the upper limit, while adding delta to the lower limit. Sounds like Baz has worked out a "compromise" of the Delta that allows the dips to be filled, and the lowering of the upper level effects are not enough to cause any damage, so is sucessfull, although not very gracefull, and the amount of product support would be huge. I Imagine there are quite a few race cars running an Accusump in America simular to Baz's setup,. Another way to use an Accusump would be to "recharge with an electric oil pump", this way no oil pressure/volume would be from the main oil supply would be effected at all. I don't think any one is using that idea that I know of and I just thought of it, but would be better than "robbing" oil pressure and volume from the main system. This again would not be very elegant or would have a mass-market appeal. Don't forget I am working on the UIDS v2.2 . George is gone to a wedding and will be machining a "real one" when he gets back next week. I can't wait to try it out !! Another issue with oil pressure drops is simply the oil level. It has been lowered from the original design by Porsche due to inadequacies of the crankcase vent system. (AOS). There are quite a few inadequacies of the factory AOS system ( I can't believe people are still using them, but they are). On the track, normal blow-by pushes oil up into the AOS in a amount that cannot drain out fast enough, therefor lowering the oil level helps prevent this, but also contributes to dips in oil pressure. The Motorsport AOS can make the dips worse due it holds more oil (lowers the oil level even more).. One problem is Frequent failure of the diaphragm/expense of labor to replace. Diaphragm failure would cause massive oil ingestion...UAOS cures this. UAOS is also receiving other updates that will allow more oil level to be run that will have a positive effect on oil pressure dips.

Quotes source post: 17531584

Skip, looking forward to hearing about your UAOS updates as well as the first UIDS prototype.

Oh, I haven't forgot Skip - I am really excited for a new product development aimed at solving this issue. I'm really appreciative for you, George and anyone else taking this problem on. I'll be at Lime Rock Tuesday so we will get some new data from a track I've never driven before.

Quotes source post: 17531584

[Image-only post; no author-written text. Original image references:] https://cimg9.ibsrv.net/gimg/rennlist.com-vbulletin/700x394/shut_up_and_take_my_money_a23e53910b12eb353402eb1decc80399fff34bb3.jpg

I am not trying to sell anything. I increasingly find on forums (particularly from the USA) that when someone has expressed an opinion about a problem that is later challenged by an alternative – if it is recognised that the alternative explanation might actually be right - instead of admitting it and getting on with helping people solve the problems there is a kind of blanket response repeating the original flawed explanation with added diversions about how many have been sold, fancy videos and repeat after repeat of the same flawed explanations – as if quantity overcomes quality and will persuade readers that they were right in the first place – which appears more important to those posting them (possibly those reading them as well) than the real causes - of which the causes of BORE SCORING are a prime example with continuous claims it is caused by oil , fuel, injectors etc ignores why the bank 2 failing first obvious question. Now it seems because we have a solution to racing oil pressure instability worked out the hard way over in the UK – just when someone else is ready to try out theirs in the USA – there seems to me to be a similar response (well off topic) when there is a technical explanation freely available for anyone to trawl through and form their own opinions. It seems people have become too lazy to read and try and understand the issues and instead prefer to see who can blow the most smoke – even when (and perhaps especially because) there is total irrefutable proof that our system works via years of successful racing results and reliability. Then despite having no plans to export our solutions we can be handicapped from helping those suffering engine failures by forums responding to pressure from supporters that pay for the privilege by banning the technical responses we offer because we don’t want to pay for the privilege of posting them when they have no commercial benefit for us. Sorry that this response of necessity will be long – I will try and be as brief as possible. I am not sure if the Rennlist system or management will accept such a long reply. First to answer some points made recently. ---Quote (Originally by TjAlmeida)--- Not to side track, and I don’t expect builders to reveal their solutions. But if opting for a an eventual rebuild, are these oil pressure problems solvable, or even with a properly built engine oil pressure could still be a problem. ---End Quote--- RESPONSE – YES THEY ARE SOLVABLE ---Quote (Originally by zbomb)--- I have seen no data that demonstrates improved oil pressure stability through practices taken during rebuild. RESPONSE – A REBUILD ON ITS OWN WILL NOT ELIMINATE THE PROBLEMS IF THE ENGINE AND CAR DO NOT HAVE SOLUTIONS fitted to them THAT WORK. I would be interested to see if via o-ring, sealant or some other method shoring up the interference fit oil passages that have been said to contribute to internal pressure leakage make a meaningful difference. RESPONSE – IT IS NOT THE FIT OF THE OIL PASSAGES THAT MAKE A SIGNIFICANT CONTRIBUTE TO INTERNAL PRESSURE LOSSES – THEY ARE CAUSED BY THE FACT THAT EVERY BEARING, TENSIONER AND CAMSHAFT CONTROL MECHANISM IS A TOTAL LOSS LEAK PATH. 7 MAIN BEARINGS, 6 BIG ENDS, 24 TAPPETS, CAMSHAFT BEARINGS, TENSIONERS, VARIOCAM SOLENOIDS ETC ALL HAVE RUNNING CLEARANCES THROUGH WHICH OIL LEAKS WHILE 6 OIL SPRAY JETS ARE DESIGNED TO LEAK OIL WHEN RUNNING PRESSURE IS NORMAL. IF IT DIDN’T THE OIL WOULD OVERHEAT AND THE WEAR PARTICLES WOULD REMAIN ENTRAPPED AND ACCELERATE WEAR (RATHER LIKE THE OLD IMS SEALED BEARING DID). OIL PRESSURE DROPS IN MILLISECONDS IF OIL SUPPLY IS RESTRICTED. ---End Quote--- The complete crankshaft carrier on both sides and the mating surface is the inside case, are a machine fit from the factory and are (according to Porsche shop manual) to assembled dry (no sealant). This is a HUGE potential for internal leak of oil volume and/or pressure. RESPONSE – NONSENSE – PORSCHE KNOW WHAT THEY ARE ADISING - THE MATING FACES ARE PULLED TOGETHER BY THE CYLINDER HEAD BOLTS – LEAKAGE IS MINUTE COMPARED TO ALL THE BEARINGS, tensioners, solenoid valves, variators etc WHERE 1 TO 3 THOU CLEARANCE IS A DESIGN NECESSITY. Any slight mistake in machining tolerance, surface imperfection, warpage due to heat, ect. can cause a big problem in oil pressure or oil control, too much oil leaving the sump and into the "loft" area of the case/heads ect. And can even cause a "light rattling sound" at around 3-4k rpms on light load. RESPONSE – DISCUSSING OIL LEAKAGE WHEN THE PROBLEM IS OIL PRESURE LOSS DUE TO CORNERING AND BRAKING FORCES LOSING THE OIL PICK UP TO THE PUMP IS DIVERSIONARY TO THE MAIN PROBLEM. There was a Porsche TSB (Technical Service Bulletin ) issued for this "noise" on 986 Boxster, test procedure was to install manual oil pressure gauge in the head port, warm engine to operating temperature and observe gauge reading. At 4k rpm oil pressure was required to be above 4 bar, if not, replace engine. RESPONSE – AGREE - IF THE OIL PRESSURE IS TOO LOW THE ENGINE IS PROBABLY WORN OUT (NOT ROCKET SCIENCE AND DIVERSIONARY). Porsche is very serious about NOT applying sealer where they say not to. Back when we were having bokoos of RMS leaks,( and Porsche had re-designed the RMS 5-6 times) we had a tech use an RTV sealer on the outside of a RMS as a insurance again a leak o the outer perimeter, on a Porsche Driving School car that had already had 2 RMS/s replaced, and an Engineer from Porsche came down to investigate and install a 3rd new design. He came apart on the Tech that done the previous RMS for using an RTV . RESPONSE – RMS OIL LEAKS ARE CAUSED BY THE HOLE THE SEAL FITS INTO USUALLY NOT BEING IN LINE WITH THE CRANKSHAFT JOURNAL BY UP TO 10 THOU ECCENTRICITY – SOMETHING WE CORRECT. The moral of the story, Porsche doesn't use any sealer on the crankshaft carrier to engine case connection, but on my personal engines and I'm sure most all engine builders, it will get a sealer. RESPONSE – ENGINE JOINT SEALER CAN SQUEEZE OUT INTO THE GALLERY FEEDING OIL TO THE CRANKSHAFT BEARINGS AND THEREFORE WHEN IT DRIES AND FALLS INTO THE GALLERY CAN BLOCK THE OIL SUPPLY – IF YOU WANT TO SEAL IT - USE AN ANAEROBIC SEALANT TO AVOID (WE DO). Also the same situation apply's to both lifter carriers to head connection . full length of the engine,, machine fit dry , no sealer. Porsche... My personal engines get sealed. RESPONSE – AS LONG AS IT IS ANAEROBIC BUT WILL NOT INFLUENCE OIL PRESSURE LOSS ON TRACK. These internal leaks can vary engine to engine as they come from the factory, or as they age and heat cycle, or if they ever experience overheating. So each engine can have a varying degree of unwanted internal leak. A very exact test of the oil pressure at specific temperature would reveal if the internal leakage is excessive. Internal leaks can contribute to "oil pressure drops in corners" but IF oil pressure is good (exact), there are other factors at play. RESPONSE – INTERNAL LEAKS HAVE NOTHING TO DO WITH LOSS OF OIL PRESSURE IN CORNERS DISCUSSING THIS IS DIVERSIONARY. As Baz said even with his deep sump, it needs an Accusump. RESPONSE – NOT FOR ROAD USE, TRACK DAYS AND RACING MID FIELD BUT YES WE HAVE FOUND THAT BUT ONLY VERY OCCASIONALLY AND THEN WITH SLICK TYRES ON SOME CORNERS ON PARTICULAR SHORT CIRCUITS IN THE UK (NOT EVERY CIRCUIT) AMD THEN ONLY RACING AT THE FRONT MAINLY BECAUSE OF BRAKING RATHER THAN CORNERING FORCES. I am not a fan of Accusump, I bought one decades ago when they came out. Yes it's an "American Invention" that was first intended as a " oil primer" as everyone learned most engine bearing wear occurs during start-up. Then it was popularized again with the popularity of "Turbo Kits" to use as a primer for the turbo and also to use after engine shutdown, half capacity was used after shutdown and half reserved for before startup. It has also been used for Track oil pressure drops, but the success is mixed. The problem with it is the "recharge must take place", if 2 qt is used to fill a dip, then 2 qt must be extracted from the main oil system when the dip is not occurring. This effects the Delta of the rpm/oil pressure and volume. It acts like a big "damper" on the oil system, removing delta from the upper limit, while adding delta to the lower limit. Sounds like Baz has worked out a "compromise" of the Delta that allows the dips to be filled, and the lowering of the upper level effects are not enough to cause any damage, so is sucessfull, although not very gracefull, and the amount of product support would be huge. I Imagine there are quite a few race cars running an Accusump in America simular to Baz's setup,. Another way to use an Accusump would be to "recharge with an electric oil pump", this way no oil pressure/volume would be from the main oil supply would be effected at all. I don't think any one is using that idea that I know of and I just thought of it, but would be better than "robbing" oil pressure and volume from the main system. This again would not be very elegant or would have a mass-market appeal. RESONSE – THE ABOVE IS WRONG IF APPLIED TO HOW WE USE THEM AND IS THEREFORE DIVERSIONARY AND MISLEADING AND I WILL EXPLAIN WHY AT THE END OF THIS POST Don't forget I am working on the UIDS v2.2 . George is gone to a wedding and will be machining a "real one" when he gets back next week. I can't wait to try it out !! RESPONSE – IF IT PRESERVES OIL IN A HIGHER CAPACITY SUMP IN CORNERS THEN IT WILL MAKE A BENEFICIAL CONTRIBUTION TO OIL PRESSURE STABILITY – BUT ON ITS OWN IN SOME TRACK SITUATIONS IN THE UK AT THE PACE AT THE FRONT OF THE FIELD MIGHT STILL BE INSUFFICIENT – I HOPE IT ALL WORKS OUT FOR YOU. IT SHOULD BE PERFECTLY OK USED IN ISOLATION FOR MOST USERS IN MOST OTHER SITUATIONS (AS OURS IS). Another issue with oil pressure drops is simply the oil level. It has been lowered from the original design by Porsche due to inadequacies of the crankcase vent system. (AOS). There are quite a few inadequacies of the factory AOS system ( I can't believe people are still using them, but they are). On the track, normal blow-by pushes oil up into the AOS in a amount that cannot drain out fast enough, therefor lowering the oil level helps prevent this, but also contributes to dips in oil pressure. The Motorsport AOS can make the dips worse due it holds more oil (lowers the oil level even more).. One problem is Frequent failure of the diaphragm/expense of labor to replace. Diaphragm failure would cause massive oil ingestion...UAOS cures this. UAOS is also receiving other updates that will allow more oil level to be run that will have a positive effect on oil pressure dips. RESONSE – WE DO NOT EXPERIENCE THIS PROBLEM WITH OUR SYSTEM. Our first Hartech racing sump and Accusump system worked perfectly – let me explain why. The oil pump has a pressure relief valve. When oil pressure is high enough to reach “system pressure” it maintains that pressure thereafter unless the revs fall low enough to a lower pressure or the oil supply to the pump is compromised. System pressure at the relief valve is about 7 bar or 105 psi and there is a system pressure loss to the pressure sender in the head of around 2 bar due to leak paths already explained. The Accusump is like a bicycle pump in that there is a sealed piston inside and air pressure one end and oil pressure the other end. Before you start the engine air pressure is delivered to the closed end when there is no engine oil pressure – so the Accusump is full of air under the set pressure. From a rebuild you prime the engine oil pressure before the engine is allowed to fire up and this also charges the Accusump (because it always has a slow recharge path from the main oil supply). This pushes some oil into the Accusump and forces the piston towards the air end until the air pressure equals the oil pressure. Accusump recommend about 1 bar initial air pressure (say 15 psi). Until the oil pressure exceeds 15 psi the Accusump is just full of air. As soon as the oil pressure exceeds the air pressure – the piston is pushed back into it and oil enters the Accusump. How far it pushes the piston towards the closed end depends on Boyles Law (p1*v1=p2*v2) so if the initial air pressure was 15 psi and the oil pressure was 30 psi then the piston would move half way along the housing and the Accusump would be half full of air and half full of oil because as the oil enters the Accusump it compresses the air while moving the central piston towards the closed end. The higher the initial air pressure the higher the oil pressure has to be to start filling the Accusump and the less volume of oil it ends up containing – but the as the Accusump empties back towards the engine (under oil pressure loss) the air pressure pushing the piston (and with it the oil) back into the engine also reduces so the higher the initial air pressure the faster it delivers the oil and the higher the pressure it maintains for the same volume replaced. The right initial air pressure is a personal choice (if you use 15 psi air pressure and connect your Accusump from somewhere in the gallery system -then at a likely 90 psi the Accusump will be over 60% full of oil. If you use 30 psi air pressure it is usually about 60% full of oil but maintains more pressure under delivery which is faster). Ours was fitted with a manual valve next to the driver’s seat – which you them shut and continue to warm up the engine and set the oil level. When it is hot you open the valve and fully charge the Accusump with system pressure and close it again. The Accusump now holds oil (how much depending on the initial air pressure) but has nothing to do – the valve is closed and it has full system pressure in it – so you now make sure the engine has the right oil level for whatever sump you are using. On the start line when the revs rise to start the race the manual valve was opened but nothing happened because the oil pressure in the engine was the same as the oil and air pressure already inside the Accusump (making the Accusump redundant) – and this remained the case throughout the race as long as the engine never runs at such low revs that the oil pressure reduces below the setting of the pressure relief valve (which when racing it shouldn’t) or unless there is a loss of oil pressure from the pump because the oil level has dropped too low. If for any reason the oil pressure in the engine drops below that in the Accusump the air pressure inside it is now higher than the oil pressure and the piston moves the oil inside into the engine while the air (and oil) pressure slightly reduces. How much oil moves into the engine until the normal pump oil pressure is re-established depends on how long it takes to recover the oil level in the sump. The Accusump is not being used to re-fill the sump (that would be pointless under the “G” forces) but to re-establish system pressure and oil at the pump. In our experience with racing sumps that are not valve controlled it can take a few seconds to recover pump oil pressure and in that time the small Accusump can get close to emptying while the engine pressure has fallen almost to the original “empty” air pressure. As soon as the oil level in the sump enables the pump to re-pressurise the engine it also re-charges the sump (which does slightly delay oil pressure recovery but from a start point really close to full system pressure anyway). With crankshafts rotating at close to 125 times/second your engine will not last long under oil pressure loss unless it recovers oil pressure immediately (within fractions of a second). If your racing sump allows the oil to escape easily you might have too long a delay to save the engine and even with an Accusump it may loose too much oil pressure (or empty) before the sump level recovers enough for the oil pump to work again. You need to have a racing sump that preserves oil captive within it for as long as possible to stand a chance of reliability. What else you need to preserve or re-fill the main sump depends on the level of “G” under cornering and braking you are racing at. At the end of the race you rev the engine to refill the Accusump and close the manual valve leaving it ready charged to prime the engine on the next start-up. We never had any problems with any of our engines using this system. Delivery to the engine from the Accusump was via a spacer we manufacture that fits under the oil cooler. Porsche also understood that braking is as much of a problem as cornering and that is why they introduced an additional scavenge pump on the RHS cylinder head to return the trapped oil in the front of the cam box. Later models incorporated a slope in the cambox to help oil return more quickly which works fine for general road use and track days. The oil is returned to the outer sump area through small long drilled holes in the crankcases. Although the Boxster and Cayman S have the engine the other way around it is still the RHS cambox that entraps the oil. Because the problem only exists under braking we use an electric pump connected to the brake light switch via a relay to evacuate the oil instead. But where it goes to is important. We redirect ours to the central sump chamber externally where the oil pick up is. If you are switching the oil pick up point to the RHS or LHS (to suit the corners) I am not sure what you do with this oil return system? Our problem emerged when regulations prevented the oil pipe and manual valve being in the drivers cabin (in case in an accident it spilt hot oil on the driver) so initially we looked to fit a shaft through the rear bulkhead to the valve in the boot (trunk). But we also couldn’t use or oil cooler spacer (not enough room with a Cayman S) so needed a new Accusump entry point. Then a new modern system emerged with pressure switches and solenoid valves that could be set to do what we did mechanically with a manual valve – automatically - and it included fitting an air/oil cooler in the path of air entering the rear side scoop. The team were keen on it so I approved fitting it (my mistake for assuming someone had worked it all out correctly and it would work). It ended up with lots of pipework, a pressure switch and solenoid valve and even a rev limiting switch (so it only worked at a given rev point). It also had a filter, one way valve, thermostat and we connected the Accusump delivery pipe into the system. It looked fantastic and everyone was pleased with it – until we then started to get unreliability after a few races. That is when I started to look into it and found to my horror the pressure switch was rated at 35 psi which was the equivalent of waiting to open the old manual valve only when the oil pressure had already dropped that low to almost zero by the time it was recovered – much too low to protect the engine and extending the pressure recovery time unnecessarily. I also realised that the new system overall had such volume that reduced recover time as well and we couldn’t afford the delays. We improved our data acquisition system to look at smaller time increments and found the shortest of pressure losses but obviously significant enough to cause damage. Meanwhile we commissioned the wider racing big ends to help the latest 3.8 engines to stand the higher piston weight and over-run compressions on braking by increasing the load/unit area rating. I found a higher rated pressure switch rated @ 55 psi and we tried that but by then I was so upset that we had been so stupid to allow such a ridiculous system to be on our car (and was fed up with the resulting rebuilds, failures and costs) that I ordered most of the system to be removed and simplified. However we retained the rev limiting OMEX switch and the EPC (pressure control switch) but over-rode the pressure switch with a manual electric switch in the cabin. This meant that while we could use the rev limiting switch to only open the valve when the engine was revving over 3,000 rpm we could switch the pressure switch permanently “ON” when racing (so there was never any delay in oil pressure recovery). But without the pipework we still needed to find somewhere to send the oil from the Accusump into the engine (remember no space for our oil cooler spacer now we had removed the air/oil cooler). The we had a brainwave. If we deliver the oil directly to the pump we don’t even have to wait while the Accusump oil delivery fills the engine galleries until it reaches back to the pump and it doesn’t matter if during the oil delivery the delivered oil pressure from the Accusump reduces (as the piston moves in the Accusump) because the pump itself will immediately recover full system oil pressure. This is how it now works. Start up etc as before (using the manual switch instead of the manual valve) go racing. When racing as long as there is system pressure from the oil pump in the engine the Accusump does nothing but just sits there as an emergency supply system if it is ever needed. If there is ever the slightest drop in system oil pressure (even for a few milliseconds and a few psi) the oil (and air) pressure in the Accusump is now higher than that in the engine (even if it has still only reduced from say 105 psi to 100 psi) and oil flows straight into the pump where it is immediately delivered into the engine at full 105 psi oil pressure. Even if for some reason the oil level in the sump does not recover for a second or two (not applicable to our sump) as long as there is still oil being delivered directly to the pump it will go back into the engine at system pressure while slowly recovering the full Accusump storage pressure. This will always maintain full system oil pressure unless the Accusump runs out of oil before the sump has re-filled and with our system we never get anywhere near that scenario. So we have ended up with a very reliable system once again but bitterly regretting just trusting the apparent benefits of a high tech electronic system that claimed to sort everything out but failed dismally and it was only resolved by seeing what the extent of the problem was using data and then finding ways to eliminate the issues – which we did. We would probably never have had any problems if we could have just used our original manual valve system with an oil cooler entry spacer but now had an electric cambox oil return system, wider better big ends and had reduced recovery times while increasing recovery pressure by feeding the Accusump back directly to the oil pump. Is it all necessary – honestly – I don’t know but am I going to take the chance and reduce the system – no chance. That still leaves one problem. These are still fragile engines and while we race in public we want to see inside the engine before the season ends to measure wear rates but then we will always rebuild it with new big ends and mains. Do we leave it until the end of the season (and keep our fingers crossed when if it still fails our opposition will no doubt be jumping up and down for joy) or check it and then never find out just how long it would have lasted? WHAT WOULD YOU DO? Baz Free technical reports with pictures available from [email removed] That complicated system that didn't work

In another event last weekend - the Endurance Britcar Trophy Class at Silverstone the only Hartech engine entry (a 996 3.4 capacity increase to 3.7) was 3rd in class 2 out of 35 entries in the first race (7th overall) and 1st in the second race (and 3rd overall) fitted with our sump and oil pump system. I think these are important results that do more to satisfy questions about the reliability and suitability of our designs and solutions than data recording on its own or trying to find fault with systems that have proven themselves for 10 years in comparison to something that as I understand it has yet to be made. Not trying to sell these to anyone just informing readers interested in solutions how we managed it. Baz Our racing sump Our oil pump covers for Accusump direct entry

FWIW- We also never use sealant between the lifter carrier assembly and the crankcase halves. Nothing good comes from doing this, and Baz is spot on with his points on this topic. I also learned to never trust the electronic valve in regard to the Accusump. I want to be in control of when the accusump is active, and when it isn't. The 996 in my avatar here is the car that taught me this back in 2008.

Quotes source post: 17532351

Jake, do you use an acusump in your track engines?

Quotes source post: 17532381

No. Outta here guys.. We are closed this week for summer shutdown, and I have things to do~

Quotes source post: 17532159

I'd love to see data on this. If I was working on this I would want OP (at main bearings or closest possible place), Accusump air pressure and temperature, Accusump valve control, OT, Throttle, Brake Pressure, 3 axis G, Rpm at least. Accusump Oil Volume would be a math channel calculated from Accusump air pressure and temp. I'd want a flow rate vs rpm vs temp graph of the stock oil pump so I could estimate the amount of time the Accusump would be able to deliver oil against a pressure drop. The flow rate of the valve would be important too. I think some effort would need to spent determining all the delay times as well: How long does it take for the oil discharged by the Accusump to start increasing the OP for instance. I don't think anyone has conclusively demonstrated what an Accusump actually does in practice. BTW I think Baz made a great point about high RPM coasting earlier. I have anecdotal evidence that doing this, coupled with an OP drop, is instantly destructive. Chris Cervelli Cervelli Technical Service

Baz, This is best and most thought out description of an Accusump system that I have seen anywhere. It's plausible to me that using the system the way you describe could shorten the duration of the pressure drops to the extent that an engine builder could have enough confidence to specify and warrantee a reasonable rebuild interval. A couple questions: My data shows that the most common OP loss situation is the scenario of a long high rpm straight, followed by braking, followed by a right turn. From this I initially concluded as you did that braking trapped oil in the right head, as there is no path for oil back to the sump when the Long G is negative. Later I began to question whether the braking lasts long enough to trap significant oil here, and now I no longer think it does. I am looking at data right now (which I will take the time to screenshot and post if someone asks for it) showing OP drops roughly equal in severity even though the first example has a 6 second brake zone (130 mph to 45 mph) vs a 2.8 second brake zone (96 mph to 60 mph). Now I think that it is the sustained high rpm that traps the oil in the right head. I think the cam chain carries oil out of the intermediate shaft area and defeats the ability of the scavenge pump and the right head is always full of oil, braking or not. Did you try running your electric right head scavenge pump continuously? My dry sump picks up oil at the front of the right head and I have never been sure whether this was strictly necessary. Just for clarity, you are feeding the Accusump output into the oil pump outlet, not the oil pump input? I don't ever see the action of the oil pressure relief valve under racing conditions. My dry sump pumps also have a relief valve, and I don't see it activating once the oil is warmed up. Can you see this on your data? Is most of your experience with 5 chain, early 360 groove main bearings or with 3 chain, later larger short groove main bearings? Did you see additional problems with the 3.8 engines? Did you retain the crankshaft damper and dual mass flywheel with the 3.8's? Have you seen rod bearing failures without a corresponding main bearing failure? And just an FYI for everyone on the oil pressure channel sealing question: I build all my race engines using Loctite 574 only (anaerobic sealer) and use no silicone anywhere. I do take the time to seal all those channels, but have no data that indicates that this helps anything. Chris Cervelli Cervelli Technical Service

Quotes source post: 17532697

I am in complete agreement with this and I think it's among the biggest issues. There is a little baffle in the chain box which I suspect does about nothing, as the chain moving at hundreds of feet per second has plenty of inner space to drag oil right out of the sump. For a bonus it's aerating the oil simultaneously.

I can't re'respond to 100+ paragraphs , I don't have the energy. maybe later, it mite just get longer and longer, let it settle..

Quotes source post: 17532697

Hi Chris I would love to answer this and have positive responses especially about the pick up point for the oil pump and why - but I think I have been banned from Rennlist for not paying for the privilege but I will respond to you directly if you send me an E-mail to [email removed]

This is the only website I've been on that throws reasons at you to cancel your membership. What a joke.

Thank you gentlemen for putting in the time and effort to post your experiences on understanding the mechanisms underlying the oil pressure drops and alleviating them. Baz, I appreciate your carefully chosen and thus precise wording in proper English. What a joy to read! Sad on the other hand to see you got banned. I hope it is 'just' because rennlist cannot make exceptions to their own rules in order to keep the paying contributors on board and their money coming in. Still, banning contributors to a civilised discussion is counterintuitive to what is the purpose of a forum. Don't you agree, 'rennlist operators in charge of banning'?

Quotes source post: 17534155

Baz, is there any possibility that you can respond to Chris and make your response available thru Hartech.org?

Mods - What articles of thread posting did Mr Hart actually violate? This was one of the best and most forthright threads, on any forum, I’ve had the pleasure of participating in. Sad the membership suffers.

Quotes source post: 17535032

I have an email into Baz. I will keep the thread updated when he gets back to me. Does anyone have confirmation that he was banned? Chris Cervelli Cervelli Technical Service

Says “Banned” under his user name.

Has anyone discussed the H2 X51 recovery system shown in this video? You Tube . It indicates the tube is available from Porsche. Would like to know if the X51-equipped cars suffer from oil starvation during braking and cornering.

Martin was on with a post or two but didn't discuss at length. Love to hear about it.

Quotes source post: 17535097

Don't know. Certainly isn't selling anything other than providing valuable insight to the forum. Damn shame. Moderators, you need to rethink your strategy on this one. RL is getting first hand knowledge for free. Can't figure you guys out.

Kind of funny, like not funny - funny - the guy is like - "HEY, I'm giving these ideas away for free, take them ! I'm not interested in selling stuff in the USA so just take my experience and ideas for these parts and if you like, use them yourselves" And he gets ****ing banned...

Quotes source post: 17535348

Sub to Martins channel, he's got lots of good stuff. He put another video out on the scavenge pump today. You Tube

Baz did reply to me. He took the time to create a word document complete with pictures so I would understand what he is talking about. I am taking what he sent me last night and the totality of what he wrote in the thread here as a 100% good faith effort to help the community not blow up engines. I need to confirm with him, but I expect he will not have a problem with me distributing the document. I don't think his solutions would have been suitable for my particular use case, but I am convinced they are more than sufficient for his, and Baz's use case is MUCH closer to what most of you guys are doing than the insanity I was trying to pull off. You can read whatever level of endorsement into that statement that you are comfortable with. I suppose that it is now obvious that the vendor payment structure in place here on Rennlist is going to prevent the further discussion of the technical realities in play here. Several vendors have and continue to profit from the sale of non solutions to this problem, and it appears that they and Rennlist are now making sure that the technical discussion and understanding does not reach a level here that can threaten future sales of those non solutions. Chris Cervelli Cervelli Technical Service

Moderators, it�s none of my business, but I, too, hope to learn why Baz may have been banned. He seems to have been a great resource on these forums. And a wealth of knowledge. If this is a revenue thing with vendors, just throwing an idea out there - is there a lower cost option for these vendors to pay where they don�t really benefit from sharing their information, but still generate revenue for the site? I would also entertain a different package option for regular users to pay a small premium - if they wanted it, optionally - so these ideas would be shared? In other words, a special sub forum whereby I�d pay a bit extra to be able to view.

Quotes source post: 17536710

Thanks Chris.

Quotes source post: 17536714

Sounds like a good idea. I am in.

This is juvenile. What is the difference between what CTS has shared for his dry sump, and what Baz has shared for his solutions, or Nick, or anyone contributing to this thread? Seems someone has an issue with Baz specifically. I�m tempted to create a single topic forum site with no advertising or fees so that knowledge can be shared FREEly.

I wish I knew the full story behind this “banned business “ then I will rethink some ‘’things”

Vote with your (virtual) feet. There are other forums available, just migrate the discussion. Rennlist’s sponsors and advertisers are the ones to force more open content when they see lost revenue. Or open a new account under a pseudonym.

Nick_L, You have studied the 996 oil system from every angle. I suspect you are thinking thru application of Martin's new H2 scavenge pump to build an X-51 like oil system. What are your thoughts, how would you use it in combination with other enhancements to help mitigate oil starvation and oil pressure stability?

I've been alerted to this thread, I am investigating what is going on with Barry's account.

Back to the main subject of this thread. I Googled “oil starvation” + “wet sump”. Not sure if anyone else has taken an a priori look at this issue. I found, among other things, that Vipers, Corvettes/Cameron’s, even the old Lotus engines all suffered engine failures due to the same problem of the 996. One interesting article suggest you all are looking at the wrong variable (oil pressure): https://www.enginebuildermag.com/201...il-starvation/ . In addition, you might be interested in the work published in the Boxster forum of this site written by Chris @ CTS. Apologies if this is historical to this audience, just thought I would share a bit of research on a well-studied topic. Acusump and dry sump may be effective, but are possibly not needed?

I have read about oil issues with those cars in the past. The article was a good one, but nothing the engine builders and designers don�t know about. It�s interesting that newer 911 engines and probably a lot of cars these days have variable speed oil pumps. These newer pumps can adjust for different temps (oil viscosities) and engine speeds, not to mention improve fuel economy by moving less oil than what is absolutely needed. My �99�s cold start ECU revs the engine to nearly 1800 or more RPMs to solve for the oil pump priming issue.

Never as good as it seems and never as bad either. Common problem with a number of solutions that we have seen materialize on this thread. I am very interested in learning about Martin's pump and if he has an enhanced version of the X51 kit. More importantly, the question is will Skip's new deep sump, coupled with Martin's new pump/X51 kit solve quite a bit of the oil starvation/oil pressure instability as a package? Martin, Skip, any thoughts?

Quotes source post: 17536710

I think it's highly likely that the next, and best (my opinion) developments to resolve this issue will be coming from people who are not paying vendors of RL. I hope we get to hear about its existence if that happens. If RL and IB sees fit to ban individuals who have stated no intent to use RL as a marketing vehicle and are instead informing RL members of their technical experience, and or developmental work for the enhanced knowledge of the member base - than the technical progression on problems relating to the members will be stifled and pace dictated only by it's paying vendors - this is a big loss for the community trying to keep their $20K engines out of the rebuild pile. If current vendors were responsible for lodging complaints about Baz's post leading to his banning on the grounds they were commercial in nature (despite Baz's statements to the contrary MULTIPLE times) - well, that would be very unfortunate - I have my own thoughts as to why vendors may feel his input should be quelled, and I'll keep them off RL. I really hope that Baz's membership is thoughtfully reviewed ( Robb M. ) by the powers that banned him for the benefit he can provide to the RL community and that his banning is reversed. I also hope he would be willing to continue posting and sharing his experience, there just are not many people with his M96 experience that are as informative as he has been and I think his posts have been of great benefit to the community. Chris, If Baz would be OK with you sharing the document, I'd love to take a look - you should have my email. As to keeping the conversation progressing - I sent some data from Lime Rock to Paul, not sure if / when he'll have an opportunity to do his thing with it. From what I saw, pressure was much improved there vs. what we've seen. I don't have anything particular to lend that to. Some thoughts - Oil level was about 1/2 capacity, it was my first time at the track and I battled some issues (ABS failing again) that kept me from pushing full tilt, this can be seen in the G data. I ran the day anyway to learn the track for next time and get the data (and not throw away the $$$ cost for the day ) Here's the video from my best lap of the day with the data overlaid - 7th lap of the first session You Tube

Quotes source post: 17542582

I have never understood why the "3.4 X51 with the dual stage scavenge pump" was never offered for the US market. Possibly because of emissions reasons with the camshaft profile or DME tune? OR because it had only 1 cam position sensor? Anyway , The 996.2 X51 did NOT have the Dual Scavenge Pump on bank 2 head ,contrary to what many people believe. Porsche did state in Technical Training Class P10W that the new design of the 996.2 head had better drain bank to the single stage scavenge pump and felt the dual stage scavenge pump wasn't necessary. ( This needs "real data" to support or deny) I have always felt the dual stage scavenge pump was a big help on the 3.4 996.1 and would even help on the 996.2 3.6 engines.( it certainly wouldn't hurt) The Porsche Aftermarket (especially in the US) is a lot tougher than most everyone thinks. Back in the early 80's on my racing cars,(setting World Records), I had many many innovations/Inventions that were not able to enjoy commercialism, even though they worked beautifully . Sometimes the market is limited, just not ready for the Innovation/Invention, or just doesn't like/support it , or think it necessary. LN Engineering had reproduced the Dual Stage Scavenge Pump for the market, as shown in previous posts here, but the market has not supported the product very well as sales were lacking. And Martins Dual Stage Scavenge Pump seems like a well made product also that I believe would help both the 996.1 3.4 and the 3.6 996.2 , but if the market doesn't support it, I'm not sure how long it will be available. The UAOS doesn't have a big profit margin , but is enough to sustain as sales are consistent. Even though it is a superior product to the Factory AOS sytem , it has a warning system to warn of diaphragm failure, and eliminates the labor intense procedure of the replacing the AOS every 3 years. It is Superior to the Motorsport AOS in that it cost about half of the 996.1 Motorsport AOS, comes with necessary modified hoses, moves the diaphragm to a very accessible location, and with the Extreme Track Duty Auto Drain accessory, provides an addition drain back to the sump to keep oil levels higher in the sump. But I keep seeing people buy the Factory design and installing them , so the "market " actually hurts itself by not supporting superior products. I am an Enthusiast at heart, with only 1 product to offer at this time, the UAOS, but I am on the verge of crossing the line to Vender status. Being a paying member I can only post 1 product for sale, but I'm not sure if I can continue on my current path to provide support, info, and products once I have other products for sale . Just an update on the UIDS V2.2 " THE VISION IS COMING TO LIFE !!! " I will start a new thread for the UIDS v2.2 (if allowed) to continue updating the progress.

Quotes source post: 17544022

Skip, this is very good info. Two things: 1. Sometimes the right product is developed that is incredible, but it is launched too early before a need and therefore solution is fully recognized. Been there and done that myself many, many times in the investment industry. In the case of dual pump, dry sump etc., they were the right products at the wrong time. With the exception of the racing community, very few folks knew at the time that they had an actual problem, nor cared and as a result, didn't want to pay big money for a solution on a car that hit rock bottom prices. It's a different mindset today. The people know they have an oil starvation issue and prices of the cars have doubled and most likely will continue their ascent. This makes the purchase of a solution more viable in their minds today. Simple behavioral science 101. BUT, it is important for the manufacturer of these solutions to walk the 996 World down a path of identifying the problem and properly lay a foundation for these viable solution. Simple sales and marketing 101. 2. With your experience with the 996 engine, how would you hypothetically use Martin's and LN Engineering Dual scavenge pump to create an X51 oil system in combination with your new sump prototype. Would that potentially address some of the oil starvation/pressure issues in your opinion? Martin and Charles, you have these pumps, do you have everything else needed to create an X51 oil system. If I understand what skip is saying, he is saying that these may be better pumps. Can you expand on this. Here is the bottom line: we may have an oil system here with an enhanced X51 oil system compliments of Martin and Charles coupled with a prototype sump system compliments of Skip that may more than partially address the oil starvation/pressure issues. Further, add on CSF radiators, low temp thermostat and LN Prototype oil cooler, we may have a winner. This system is not a replacement for a dry sump system such as what Chris has built nor what Hartech has engineered into his engine rebuilt. But it may be a logical next step. Skip, Martin, Charles, Chris and all need your thoughts and feedback so set me straight if I am not reading the the threads right. Thanks

[EDITED 21-08-06] The oil temp data used in the graphs below is incorrect for the analysis. The oil temp data is from an oil tank that is remote to the engine. Also, the temp sensor in the tank is not calibrated. The statements about the oil to water heat exchanger are incorrect. I have not changed any verbiage in the original post. Please see post 653 for an update to these charts. Apologies for misdirection. Thanks Chris for quality checks on my work. Chris Cervelli provided the data from which these graph were made [END EDIT] Before getting to the Lime Rock data, here are a couple of plots from Chris' 25 hour data. I've been meaning to put it up for a while now. Dumb work gets in the way. Just need to figure out how to cover healthcare for 7 years and I'll retire and be able to focus on this kind of stuff. One problem we have with session data is that the temps are never at equilibrium. We can use Chris' data to take a look at how the cooling systems work relative engine load This first plot shows running full out, but then transitioning to a yellow flag (my assumption) where speeds, RPM, engine load, and everything else drops. The yellow flag is long enough to get the system into the new equilibrium. From this we can judge how good the cooling system and oil cooler are. Laps 175 to 180 represent the high load condition. The average engine load at full track speed is 44% (this is an average over these laps) The max load in these laps is 72%. The oil temp is very steady at 248F, The water temp is 189F. Delta T is 59F. Laps 187 to 190 represent a yellow flag load condition. The average engine load is 21% with a max of 64%. The oil temp is has dropped to 210F. The water temp has dropped to 182F. Delta T is 28F. So what do we say about this: the oil cooler needs to be bigger if this delta T is intolerable. The water coolers seem to be OK as the temp drops from high load down to low load. Maybe it can be made better, but the bang for the buck for oil temp will be to work on the cooler This plot shows what happens when airflow is removed from the rads. The car is brought in for driver change or some rest interval. The water temp goes up quite dramatically, but the oil temp comes down to meet it. At a base level, with no engine load, the cooler is OK. Not recommended for racers though. It's possible this car didn't have any fans, hence the increase in temp during driver change out. We know that oil temp is keenly involved with OP loss in braking and turns. This data shows a strong relationship between the oil cooler capability and the oil temp. For an oil pressure system used in this car, Chris' endurance Boxter, which has the external oil tank and pressure controls, this temp delta is not an issue. The system assures the oil system sees zero loss of engine oil. For everyone else's stock machine, This seems significant. From this data, someone smarter than me could probably figure from heat loads how to design a cooler for this specific condition. 2 questions for Chris: What oil cooler were you running? Were you running a center rad? (apologies if already answered) Looking forward to a better oil cooler from LN folks. As always, forward comments and questions.

Quotes source post: 17544411

This data may be misleading. First, use the following channels for Coolant Temp and Oil Temp: Engine Temp ECU Engine Oil Temperature The Oil Tank Temperature is used for Diag of the oil level sensor only and is not accurate. The actual oil temp and delta is much lower. The Engine Inlet Coolant channel may also be misleading. Second, this race was run in very cold conditions. The high temp for the race was likely under 40 degrees, so the engine temp was running on the thermostat which almost never happens. This engine was running a Boxster S/996 oil cooler with no additional oil cooing. A 3rd radiator was in use. The condensors had been removed and the ducting optimized, and the fans removed (they are a negative to cooling at racing speed). None of this was needed for this race. The car ran with the normal cooling system which was set up to deal with 100 degree temps at 5000 foot elevation like we have at the local track here. I can send you data from a 24 hour race run at the local track earlier in 2017. That engine was a 2.7 instead of a 3.4. The weather conditions were much hotter. Chris Cervelli Cervelli Technical Service

What does the X51 dual scavenge pump actually fix? I've been asking myself that the last week. We know that it was only fitted to 3.4L engines. Lets have a look how the head design have evolved. The pictures show in order 3.4l, 3.6l and M97 3.8l. 3.4l 3.6l 3.8l If you look at the bottom part of the head you can see that they have reduced, and reduced more, the flow resistance. First by adding a slope and then by removing the oil 'shelfs'. I think Porsche identified the problem as too much flow resistance going from the head down to the chain channel(and thus the scavenge pump). The flow here is still reliant of gravity and any resistance will slow it down. The X51 dual pump would help somewhat in this but looking at the design iterations I think the pick-up for the pump is in the wrong location. Better to put it half way instead, at the bottom of the center cylinder. edit: Higher flow resistance will cause a bigger oil level drop in the sump directly proportional to RPM. I'm looking to make a RPM vs Oil level graph when I get the time.... When I started looking at the oil pressure issues I posted a question in the 997 board, asking if there is an oil pressure issue on the 997s (that is the 3.8L M97) on track. The answers was no. So what ever Porsche did seams to work well for the general user . I think it's time for this thread to differentiate between solutions for different users, not everyone needs a dry sump setup for reliable oiling. Could we make two cases: road tires and semi-slick/slick tires?

Quotes source posts: 17544143, 17544143

I'm not sure I agree with your behavioral 101 analysis as it pertains to this Porsche problem/products. You should start a new thread and "poll" the Rennlist members and ask the questions (1) Do you think you have oil pressure issues on your 996? (2) IF yes, what , if anything, do you plan on doing about it? As a VERY simplified overview of the M96 oiling system, I break it down into 4 parts . (1) pressurized oil (2) un-pressurized oil (3) sump area (Integrated Dry Sump) (4) loft area ( all area above the sump, including heads) (1) Having studied the pressurized oil channels for decades (but not measured the exact actual volume) a rough guess is that it holds about 2liters (2) In a 10 liter capacity, that means the un-pressurized oil volume is about 8 liters . 8 liters is way more than enough if managed properly. Managed properly means having it available at the oil pump pick-up. (3) sump area ( Integrated Dry Sump) this is my area of focus at the moment in the design of the UIDS v2.2 , to make the most efficient use of the available oil and correct the shortfalls of the sump shape and design. (4) loft area ( all area above the sump, including heads) oil remaining in this area does no good at all, it can get sucked up by the crankcase vent system, and is oil that needs to be drained/pumped back into the sump area ( Integrated Dry Sump). Any thing that can help get the oil back down to the sump, will improve sump function, but there will always be considerable oil in this area. When you start throwing in dynamic variables into play, mainly G-forces , it can get very complicated, but the simple dual outer pick-up design of the UIDS v2.2 has the best chance of a simple solution to this problem.

Quotes source post: 17545819

I think In the video it helped return the oil that would be delayed making it's way back to the sump during braking, which is why they placed it at the front of the motor.

Quotes source post: 17545819

Nick, good insight and a lot to unpack. Let me ask a question that hopefully you, Martin or Charles have an answer too. Do the X51 dual scavenge pumps that Charles and Martin are producing come with the full X51 kit including the "tube" or are they just the pump? Further, what is the plus and minus of having the "tube" ? Martin did a great video explaining how the X51 kit works and transfers oil to the other side with the tube. Seems to me that while the scavage pumps are nice, it's the whole kit with the "tube" that makes it a winner. The question for Skip is if there is any advantage to using the X51 with your prototype sump? Double the benefit? Assuming an X51 kit becomes available for sale for the 3.4 and 3.6. Your thoughts on street tires versus slicks and two different threads certainly makes sense. However, I would recommend at this time that we do not dilute this thread. Instead, think of the thread as an oil starvation/pressure stability continuum that is intertwined. Way back when after a conversation with Charles Navarro on my car, I posted summary points of the conversation broken down into prevetative Steps that should to be taken. I'll find it and forward. Step 1: Basic items: factory maintenance, ignition coils, spark plugs, fuel injectors, aos, sensors, water pump, IMS Step 2: Basic Drivers Ed: upgrade aos, Deep sump, better oil Step 3: Advanced DE: oil and water cooling, X51 kit?, Porschetech3 sump? Step 4: Advance DE/Club racing: Dry Sump Kit Step 5: Engine Rebuild Interestingly, the list is still valid after a month or two. The only thing that has changed is more clarity around the X51 package and Porschetech3's sump prototype. All steps are intertwined and IMHO should not be split apart but instead looked at comprehensively to eliminate confusion and clearly articulate what solutions are available at what stage of tracking your car. Interestingly, there isnt one solution that we have talked about, regardless of how advance it is such as a dry sump kit, that wouldn't benefit even a street car. Yes, it may not be needed, but everything we discussed is additive to longevity and durability of the engine. 19 years ago, I for one, would have loved a list of steps to take with my car and available solutions. None existed. Quite frankly, none exist today other than what Charles and I talked about. It's time we add some clarity of choices along the engine continuum. Hope that makes sense.

Quotes source post: 17545851

Skip, Thanks for response. Gotta read it a few more times to grasp all the nuances. You will see in a post to Nick I also included you, Charles and Martin with a question or two. Your suggestion for the poll is a good one. I'll figure out the system and get one out. Having dealt with money and people my entire 33 year career, one thing is certain. People don't fix small problems when in their mind they are not consequential, they fix big problems which have big consequences. Until the problem becomes big enough, people don't spend the money, especially if the underlying asset (996 prices) isn't worth enough to fix or if the problem is not completely understood. You can see this first hand with the aos. How many folks would rather do an exhaust, interior or body enhancement over replacing the aos with a UAOS. Most. Why, because they don't yet get the consequences. Just behavioral science 101.

Quotes source post: 17545819

Just FYI, 95% of data I show is on 200 Tread Wear tires. Slicks don't affect the situation as much as you think. 1G Lateral and the always present 1G down gives a 45 degree vector. 2G Lateral and 1G down gives a 60 degree vector, not too much worse. In the case of a cylindrical dry sump tank, having the tank over half full makes it immune to any combination of forces, as long as the 1G down remains. Chris Cervelli Cervelli Technical Service

That makes sense, as even my car with all seasons can easily brake at .9g, and probably turn a corner at .85 or better, and yet, I see the pressure drop occasionally on my car on a highway ramp. If it can happen there, it will happen on the track.

Quotes source post: 17545982

Chris, very interesting. I know you aren't involved with the x51 scavage pump development, but any thoughts you have on viability would be greatly appreciated. All ears if you have any thoughts from your communication with Baz.

I concluded that scavenge pumps, dual stage or not, were never going to work because they suck from the extreme right side of the head (using the right head as an example). So when the the car is turning left, the scavenge pumps are indeed picking up oil (as long as the cam chain does not make that impossible) but when in a right turn, the oil in the right head is stuck in the cam/lifter area, away from the suction points. This means that only one side of the engine is scavenging at all in every turn. My dry sump scavenges the intermediate shaft area, which helps solve this particular issue. Also the additional scavenge points on the heads are more centered laterally, instead of being in the valve covers. It would be easy to calculate the Longitudinal G required to overcome the various angles on the heads Nick pictured. Tricky sumps rely on something being the case that has never been determined: is there enough oil being returned to the sump at sustained 7200rpm, zero lateral G and longitudinal G operation, to replace what the oil pump is sucking out of there? My assumption is yes, because I don't believe Porsche could be stupid enough to allow the alternative to happen. But, how much "reserve" is there? 10%? 20%? Until you know this, and thus the actual oil level in the sump under 7200rpm, zero lateral and longitudinal G operation, sump design is guesswork. If there is 10mm of oil level in there (like I think there is) then no trick sump is going to work. PorscheTech, you should have your machinist make a sump that incorporates a Gill oil level sensor so you can answer this question before you get started. Chris Cervelli Cervelli Technical Service

Quotes source post: 17546028

Thanks Chris. Does your assessment include the use of the "tube" in the X51 design?

Quotes source post: 17546061

The X51 tube attaches to the very outboard area of the valve cover. I blew up an X51 pump-equipped engine in 2 hours in 2003. That was the end of my first go round with this. Once the car is handling well, and especially if you use a close ratio gearbox (average rpm goes up, sump level goes down), the engine life becomes a few hours. It's a highly non linear function. Chris Cervelli Cervelli Technical Service

Quotes source post: 17546072

thanks

Somebody like PorscheTech could experiment with this configuration pretty cheaply: Install cylinder head oil restrictors as I have described elsewhere. (5 chain engines only as I have not developed these for 3 chain engines yet) Use an X51 style pump to scavenge the intermediate shaft area instead of a cylinder head. My system is optimized for mid engine installation and this may actually be important. It does not look possible to relocate this scavenge point to the "back" of a 996 engine. Sump of your choice. I don't think any commonly available sump is measurably better than stock. There is a decent chance this will do the job. Chris Cervelli Cervelli Technical Service

Quotes source post: 17546103

Chris, I get cha, using the second stage of the dual stage pump, run the pipe to the intermediate shaft area instead of the bank 2 head front side..not a bad idea... Since the 3.2 Boxster is the same as the 3.4 996 oiling system, what do you calculate the pressurized oil flow volume at both with and without the head restrictors you mentioned?

Chris and Skip, This is good insight. So I hear you saying hypothetically, that if the X51 kit was reworked to connect into the intermediate shaft area, that this may provide better oil flow from the valves down to the oil pan?

Quotes source post: 17547112

Glen, yes exactly, but you must realize this is light conversation between two Technically experienced people. We may talk light about simply introducing a port into an oil location, but we understand it would need to be done with the engine disassembled to keep metal chips from being a problem. And the commercialization of such an idea would be a lot more than light technical talk. In other words, the idea has Technical merit, but to commercialize the idea for consumer benefit is a whole other ballgame.

Quotes source post: 17547433

I do have a way to do this on an assembled engine, but it requires more skill with tools and attention to detail than the hobbyist (heh) is going to have in most cases. My data shows that the head restrictors and resulting slower oil pump speed raise the oil level in the tank by around 2 quarts. If, on a wet sump engine, the same thing happens in the sump, then it's likely that the problem will be greatly mitigated if not solved. I don't have data on the contribution of the intermediate shaft area scavenge point. My mental model of the oil capture inside the engine has most of the oil here and in the chain boxes, and this scavenge point (mid engine) is effective under acceleration, or roughly 80% of a lap. The reverse will be true in a 996. If I am correct that this area is full of oil all the time, then this won't matter much. If it is true that only a small amount of oil is present there, then my above proposal will not work. If someone wants to try this on a 5 chain engine and publish the results with data, contact me and I will send you a set of restrictors and instructions on their installation and the installation of the scavenge point. The cylinder heads have to be removed to install the restrictors. I have some double scavenge pumps lying around too. Chris Cervelli Cervelli Technical Service

Quotes source post: 17547433

Skip, understood! Just trying to keep up with you guys and try to follow along. Any corrections to what I attempt to summarize are appreciated. More so, I appreciate the conversation that you and Chris bring to the thread.

Quotes source posts: 17547526, 17547526, 17547526, 17547532

I know what kind of procedure you speak of, it involves a high flow vacuum source and a special drilling/milling tool and yes "skillz"..The procedure could also be used to open up the 4 6mm drain back holes in the "loft area" to 10mm to help with drain-back from the loft. Easily done during rebuild, but could not be commercialized in situ for consumer benefit. Anyone want to calculate the increase in oil flow from 4 6mm holes to 4 10mm holes under gravity at 212F ? 2qts less in the "loft area" and 2qts more in the "Integral Dry Sump" is very substantial and I agree this may solve the whole issue. The cost to do this would probably be about 20 hours labor for consumers @ $175hr average =$3500 labor plus about $400 parts so for about $4k a consumer could have this done. But this would be a pretty "invasive" and costly procedure to eliminate the problem on a otherwise healthy engine Looking at the loft area, the oil in the Intermediate Shaft area has two ways to get to the Integral Dry Sump. On accelerating and turning left It can drain through the 4 6mm holes back to the sump, or it can go out the chain box to the bank 2 scavenge pump, on braking and turning right it can drain through the 4 6mm drain holes or go out the chain box to the bank 1 scavenge pump. The right turns are also where the AOS get overloaded with a rush of oil and blow-by that is too fast for the pump to handle all at one time. Increasing the drain holes to 10mm may provide enough drain back fast enough for substantial increase in sump function. [QUOTE=CTS;17547526If someone wants to try this on a 5 chain engine and publish the results with data, contact me and I will send you a set of restrictors and instructions on their installation and the installation of the scavenge point. The cylinder heads have to be removed to install the restrictors. I have some double scavenge pumps lying around too. Chris Cervelli Cervelli Technical Service[/QUOTE] Thanks Glen I hope you ( and everyone else) can keep up, the knowledge gap between Techs and consumers can cause so much confusion , thanks for your understanding. I am sure it would be like me trying to follow along with you and a colleague on high level Investments.

Quotes source post: 17547740

This all assumes that the atmospheric conditions inside the engine allow any oil to drain thru these holes when the engine is running. I think they were put there so the oil would all drain into the sump when the engine was stopped. Chris Cervelli Cervelli Technical Service

Quotes source post: 17547740

I could easily calculate the increase in area? Not sure if it is that simple.

Quotes source post: 17547821

I agree, using a "Factory AOS" means the engine would be running 5 inches of water less that Atmosphere pressure(3.5 inh20 on UAOS) and may not be the same all through the "loft" and Integral Dry Sump,. Maybe an even more simple solution would be a "balance tube" to equalize the pressures between the two Actually the high mileage/Extreme Track Duty Auto-Drain hose does this to an extent, it connects the Loft area to the crankshaft area that has direct flow to the sump.

Now that I think about it: The sump area has to be under high pressure because the blow-by gasses are directed there by the plastic guide around the intermediate shaft. Porsche even took the trouble to seal that plastic against the case halves. The intermediate shaft area, the chain areas, and the cam/lifter areas would all be under lower pressure due to the scavenge pumps sucking the air and/or oil out all the time. Furthermore, with an AOS in use, the pressure is dropped even further in these areas from the limited manifold vacuum applied thru the AOS. The blow by gasses's path out to the AOS is not obvious. I thought the AOS drained liquid oil thru the diagonal passage down to the sump but maybe that's how the blow-by gasses are supposed to escape. That would mean in times of high blow by (usually this is high rpm, Throttle >10% and <50%) the AOS cannot drain. That would explain the smoking tendency all by itself. In a dry sump system, having the tank under a little pressure makes sense. In the case of an M96, this seems counterproductive because the high pressure in the sump slows the drain back due to gravity. I always have a hard time thinking that Porsche screwed this up so bad. I must be missing something. Chris Cervelli Cervelli Technical Service

Quotes source post: 17547835

Would need to calculate in the viscosity of a specific oil at a the specific temp. but the increase in area would give a "percentage" of the increase.

Quotes source post: 17547870

The AOS is drained from the cyclone down through a funnel shape in the AOS into a 14mm drain hole to the sump under the oil level. The UAOS high mileage/Extreme Track Duty Auto-Drain connects the Accumulator to the oil fill tube which is directly over the crankshaft area and drains down through the plastic guide directly to the sump.( adding another drain path) The blow-by gasses and flow are directed out of the top of the crankcase above the crankshaft carrier up the chain boxes to the heads where the AOS vent pipes are located, directly above the scavenge pumps.

Quotes source post: 17547870

The oil goes down thru the plastic guide and blow-by gases goes upwards an escapes to the AOS ports on left/right head...I don't think it's air/oil tight on the top. The sump would be under pressure due to the scavenge pumps but the small holes let those gases escape....thus not much oil goes down these holes since air is flowing the opposite way. Edit: that can't be correct. Gases have to mainly escape thru the short block and up. There should be a small overpressure in the sump but it might be so small it's not possible to measure and can thus be ignored. The main task of the holes is thus to drain back oil when engine is off.

FWIW the entire "air space" in the engine is under Atmospheric pressure (14.7psi) minus 0.2psi ( 5inh2o for AOS ) except for turbulence and windage caused by the rotating of the crankshaft and pumping of the pistons (bottom side) and the respective cross windage pumpage. The pressure in the Integral Dry Sump is the same as everywhere else and can easily be verified by testing at the dipstick tube, which goes directly to the Integragted Dry Sump unobstructed. I just did it to confirm what I already knew.

Here's a look at Jason's run today. This is the overview. The second chart is the overview from 6/27 previously posted. Both have oil filled to 1/2 on the dipstick. Both have the blow by oil return from the UAOS. The huge differences are in temperatures and a wet track limiting speed. The difference is clear. We can say the obvious: a cooler engine has fewer pressure problems. Look at the difference in the red graph drops between the 2 sessions. Quite striking. Here's a look at the personal best lap from 06/27 and the best lap of this run. Green lines are from 6/27. Blue are 7/13. -The overall pressure is higher. This is a little surprising given the bypass valve. My guess is that this shows the difference in viscosity of the oil due to the lower oil temp. The interesting peak to look at is at 4800ft. The G's on the right turn are the same. The RPM's are the same. Because of this, we can assume the difference in the pressure change is only due to the oil temperature. The total time the hot lap is bobbling around under the bypass pressure is around 8 seconds(4700-5400ft). The time for the cold lap is around 5(4800-5300ft). My list of next steps would be: -center rad -bigger oil cooler (available?) -thicker oil Given the known blow by on this engine, the thicker oil will probably help more than just pressure drops. What issues does thicker oil present on the track?

I'd second trying out a 5-50 oil on the track. I am not aware of any drawbacks with a hot engine, but the viscosity is significantly higher (about 25%) at normal/low operating temperature.

LN Engineering Prototype Oil Cooler Charles and Cam, how is fitting and testing the oil cooler coming along?

Quotes source post: 17551037

The Boxster and Cayman engine oil cooler kits are completed. Just waiting on the fittings and lines to come in so we can finish kitting them up. If anyone stopped by our booth at Parade, we had the 987 engine oil cooler and gear cooler setups on display. The 996/997 is a work in progress. We've sourced several oil water to heat exchangers in more favorable shapes to see if we can make something fit, but with our existing cooler setup the only place it would appear you could mount the cooler would be where the SAI pump is which would require its removal which isn't an option. Since space is limited, we're thinking maybe of leaving the factory heat exchanger on the top of the engine and adding a smaller second cooler whereas on the Boxster/Cayman, out kit would eliminate the factory cooler. Just considering all our options. My other issue is that I don't have easy access to a 996 for this development work - I'd be willing to donate to the cause a cooler and all the adapter pieces if someone wants to try to make it fit in exchange for help with coming up with a plug and play bracket to make this easy.

Quotes source post: 17551126

I'm in IL and may be able to help. Pm me.

Quotes source post: 17548826

The oil will run hotter. Partly due to that is the nature of thicker oils(HTHS value is directly proportional to the heat created in the bearings) and partly because the oil cooler will work less efficient since flow will be reduced. In general the heat will also be less even thru the engine, more hotspots, due to less flow. I think it's the wrong way to go, thinner(but not too thin, HTHS should probably not be less than 3.5) would be a more logic option in this case. Have a look at my post with the temp graph(post #406), the difference between my coolant and oil is only about 10c, zbomb is roughly double that. Granted, he has 20hp more but it should not make that much of a difference.

Understood the oil will run hotter. If the whole problem with pressure drops is the pickup pulling in air, leaving bearings dry for moments, a thicker oil might solve the problem by not moving as fast. This would leave the pickup exposed for less time, or no time. The trade off would be worth it, unless the temp goes up higher than the system can manage.

I don�t think thicker oil makes that much of a difference or solves the problem. I tried M1 0-40 on the track and also M1 15-50 on the track, and either one had pressure drops. Sure, the thicker oil had overall higher pressure, but it didn�t really solve the problem. I can guess the increase in viscosity that helps keep the pickup submerged is offset by the longer time it may take to return back to the sump. Add in the additional heat of the oil, and I�m guessing it�s a wash.

This has all been addressed in other cars. Google searches for oil starvation, oil cloud, windage plate, crank scraper bring up posts in other forums e.g., Boxster, Corvette, Subaru, Porsche 944/928 that point to solutions used in track and race situations, very few solutions for 996�s but Mantis kit and LNE�s sump baffle have a sump plate that prevents oil shifting during turns/braking, the Mantis receives good reviews. So far, have not found any info on crank scrapers or true windage plates for 996s. From what these other sites teach, air dissolved in oil is a big problem that these try to solve, partly by chain drives (oil pump, cam shafts) and the crank that churn up hot oil; the UAOS may be a band-aid to the oil-air emulsion known as the oil cloud, need to control the formation of this air-oil emulsion. Maybe one reason Porsche used gear-drives on race engines. While the oil pickup may not be getting sufficient oil, it�s also sucking in an air-saturated oil that precludes lubricating the crank bearings. AOS systems for Subi race engines only cost $300, would they work better than the OEM AOS or UAOS? As indicated by Chris at CTS, blocking oil passages in the head is a known solution, also.

Quotes source post: 17551126

Charles, good to hear you have the oil cooler sorted out for the 987. Hope that George can help in some way for the fabrication for the 996. Certainly wish I had mechanical skills to help in the process. Either way, looking forward to your next steps. The cooler seems to be a necessary component within the 996 engine continuum we are all working towards on this thread.

Quotes source post: 17554368

The $300 Subi AOS system is for race cars or competition cars only and requires custom modification.. It just dumps blow-by gasses to the Atmosphere. It is not Legal for street use..Not much engineering needed for this, back in the old days we used large tomato soup cans for this. The UAOS is a whole different ball game is bolt on and is emission compliant. "dessolved" air in oil causes no problem, and the higher the oil pressure the more air the oil can dessolve. "entrained " air or "free air" (foam) is what causes problems in ICE or hydraulic components. Yes, the head restrictors would be a big help at a cost of at least $4k if not done during a rebuild.

Another anecdotal data point from a track day at Watkins Glen this weekend. My Car 99 with 112k, with unknown history. UAOS with drain LN 2qt sump LN spin-on filter 11.5 qts of XP9 Yokohama Advan A052 Ferodo race pads and slotted calipers from FCP Euro. Stock suspension Mild track alignment Conditions Track temps were in the 80s, mostly dry. Pace was moderate (first time driving this car on track) The oil level was filled to two bars below the top line on the gauge cluster. Results Zero issues with oil ingestion/smoke even with a full level. Pressure started out good 3.5-4.5 bar at 5k, but after about 4 laps it would drop to below 3, but not lower than 2.

Quotes source post: 17557348

That's that part I wanted to see !! Did you get any Go Pro footage ??

Quotes source post: 17557348

I'm really happy with the drain back, I'm getting no oil consumption or smoke which is a LONG way from where I was before. Thanks for sharing your observations - goes to show even with stock suspension, the deep sump just doesn't do a good enough job. If you check out some of my earlier posts on the topic with snapshots of the factory OP gauge, what you saw is similar to what I was seeing, so it's not a huge jump to think if you could pull data with the granularity I can, your results would be similar. I wonder how many other people are in the same boat we are...

There were a lot of 996 there this weekend. Setups varied from full race cars to modified street cars. I asked around the paddock and the results were inconclusive. Some cars had stock sumps, FVD, and LN 2 qt. One other driver with a basically stock car except for the LN sump was around 3 bar on the straight. Another car with the deep sump was at 4 bar @5.5k I think Kris and I sold a couple of UAOS for you Skip. Kris had the same results with the UAOS, and he can chime in but he said his pressure was improved from having more oil. Sorry, no video, though I wish there was. The big coolant hose on the bottom right of the motor decided it no longer wanted to be connected going into the toe of the boot when it let go and made an epic smoke screen. Fortunately the person behind me got out of the way and nobody went off track.

Quotes source post: 17557848

This happened to me twice. U need to get a good wom drive camp on there. Cost me $200 on oil dry at njmp. I think the water pump just pushes too much pressure after a high rpm de-cel.

Quotes source post: 17558318

It had a worm drive clamp, but the hose also looks to have been cut and was a little shorter than stock. That in conjunction with the motor mount led to the problem. I have not yet received my bill. Back to OP instability.

Porsche changed to the worm-drive clamps from the spring compression clamps at some point. As long as the worm-drive clamp is not a cheap part, but the real Porsche (or equivalent) clamp, you should be good. The cheap clamps do not supply equal pressure around the entire circumference, and instead, can crack high pressure points or leave gaps at other points. The cheap clamps also rust or strip easily.

Quotes source post: 17557848

Thank you Tom, word of mouth and actual results from places like Watkins Glen (that has a section known to cause severe oil ingestion) is the best way to spread product awareness and help out the platform !! Thanks to you and Kris, and I think Randy was there also with the UAOS w/high mileage/Extreme Track Duty Auto Drain.. Watkins Glen would be a good place to test out the UIDS v2.2...

Quotes source post: 17558720

Yea it would have been

Quotes source post: 17557848

Hmm. If only there were parts a person could install so the engine did not move around so much during high G maneuvers. It's almost like the stock engine mounts allow too much movement for track use. Too bad there is no acceptable substitute. Chris Cervelli Cervelli Technical Service

Quotes source post: 17558848

OMG you are going to start a war.

Quotes source post: 17558848

I actually think it was failing or failed at the time.

Quotes source post: 17558848

Chris, with what you are saying here, what are your findings and thoughts on using non-oem engine mounts. Are semi solid or solid viable without messing with the Ecu? Do you have to reprogram the Ecu to use them?

Quotes source post: 17557420

Are you getting any lifter tick after the session once the oil pressure drops below 3 bar?

Quotes source post: 17560122

I have not attempted to replicate what Mr Raby says is happening in the video posted in the thread about engine mounts. If I were forced to bet on it, I would bet that I could not replicate it. Mr Raby does not state a mechanism of action that could cause this. I have to assume that the noise within the engine has been changed by solidly mounting it to the car, and this is totally plausible. But it is also plausible that using pistons and cylinder liners of a completely different material would also change the noise within the engine, and this does not seem to be a problem. All Porsche factory race cars use solid engine mounts, and some of them use stock type ECUs which almost certainly use stock knock detection algorithms. Porsche occasionally makes mistakes, but that would be a 3 decade long continuous mistake. I have a fair amount of experience with modded 996 turbos, and can definitively state that I saw no difference in knock ignition retard logs between 964 RS mounts (in a 996 turbo) and solid mounts. I used semi solid urethane mounts extensively in my Boxsters (as tranny and engine mounts) because they were mandated by the SPB rules the cars were originally built to, and they (more or less) worked ok. I never saw any evidence of knock, fake or real. I can't say I really looked for it though. The stock ECU programming very aggressively detunes the engine (by retarding the ignition point, whether it is knocking or not) when the intake air temperature goes above around 90F, and also when the coolant temperature is high. This effect is almost impossible to avoid on the dyno, and is difficult to avoid on the track. If Mr Raby is right then this is a problem to be solved, since stock mounts can't be used successfully in a serious track car, as Tom Loder (and luckily not the people behind him) found out. As a point of interest, in a 911, most of the benefit is derived from stiffening the transmission mount, since the engine mounts cannot prevent fore/aft movement of the engine. Most of the time the transmission mount is not changed, and the engine mounts are. This does not help all that much, although it might prevent the coolant hoses pulling off a 996 engine. Unfortunately it is also true that a stiff transmission mount almost always increases interior NVH, sometimes drastically. Chris Cervelli Cervelli Technical Service

Quotes source post: 17560356

Thanks Chris. Your insight as always is exceptional.

I spent the weekend at the glen also (saw your car tom, glad it wasn't more serious) and as usual, had monster pressure drops all over the place. I run XP6 towards the bottom of the stick and the pressure drops almost everywhere. The majority of the time its low under maintenance throttle (like on turn 5) but going up the toe of the boot I'm hard on the throttle at 5000rpm and pressure starts at about 25psi before recovering. Combination lateral G, vertical G, and sustained full throttle has got to be rough on the bearings. Obviously the motor is not happy, but it so far has not resulted in any ill effects. Its coming, thats no question, but hasn't blown up yet. 99 boxster with a 3.2 from a boxster S catch can with UAOS style regulator NT01s black run group oil temp max was around 215-220 degrees.

Quotes source post: 17548826

What about also upgraded CSF Radiators including the CSF center.. Thoughts helping keep the temps down? or at least more stable with less fluctuation..

I think I remember saying that lower oil temperatures were the key about 3 dozen pages back. I learned that 15+ years ago. Glad to see that you guys learned that for yourselves.

Quotes source post: 17565177

Yep. We are waiting on next steps with Charles' new oil cooler prototype. Seems to be a no-brainer. Hoping that George can help with the bracket.

Quotes source post: 17566431

I am not a thermodynamics engineer but I think there are some hard limits on how cool the oil is going to get. For a 3.4 or larger engine, I suspect no amount of oil cooling using the coolant as the cooling medium is going to shrink the coolant temp to oil temp delta to less than 20 degrees F. I say this because I have never seen an example of this being true on any race engine. I don't think oil to air cooling is practical without reducing airflow to the radiators and therefore increasing the coolant temp, which puts you back to where you started. I am pretty sure Zbomb's oil pressure looks like crap even during the first few laps of a session where the oil temp is lower that what is even remotely achievable by improving the oil cooling. How is getting the oil temp lower going to fix this? What I am missing here? We ought to be talking about reducing the coolant temp, which will drop the oil temp along with it. BTW, modern Porsche race cars all run at 230 coolant and 290 oil on hot days. Porsche says this is ok. I don't think it's ok, but it is the reality of a 400-500hp engine on a 100 degree day. Chris Cervelli Cervelli Technical Service

I will say CSF radiators and...importantly...ditching at least one fan and shroud and the AC condensers has done wonders for my coolant temps. What used to be 220-225 is now about 205.

Quotes source post: 17566637

Quad, are you running 3 radiators?

Cutting the water temp ought to be doable. The last session at Lime Rock with Jack and Jason I had highest water temp of 203F after 20 minutes. I'm not running quite as hard as Jason, my laps are on average 2 seconds slower, but I have a much heavier 996, and the full A/C and fan installation, and 150 more HP, and an automatic transmission in the water cooling circuit.

Quotes source post: 17566637

I have not yet seen data that convinces me that aftermarket radiators cool better than stock, but new radiators cool better than old ones for sure. I had to treat radiators as consumables. I would replace them at some number of miles because the fins get packed with crap. As you found out, the A/C condensors and fans are a big negative. The minimum realistic coolant temp is 180, which puts the minimum realistic oil temp at 220 or more. theprf: I think you have a 996 Turbo? That's not a fair comparison because the radiators are much bigger and the aerodynamics are biased towards cooling instead of low drag. Porsche really got that right. I have a 996 Turbo customer at 775 rwhp running at Cup car pace with a pretty much stock (GT2) cooling system. I did add an oil cooler and the temps came down from 275+ to 260 or so. Not dramatic, but enough to make the oil temp not the limiting factor any more. Chris Cervelli Cervelli Technical Service

Chris, that's an interesting point about the radiators being larger on turbo cars. Is there anything additional besides the radiators and the aerodynamics (I'm assuming this means the turbo bumper) that makes the cooling system more efficient on the Turbo cars? I mainly ask this as it appears that Turbo models as well as 4S and 40th Anniversary cars seem to share the same radiators and bumpers. I took a quick look at Pelican Parts and all part numbers are the same between the 3, which had me wondering if there is anything else there. I believe the main difference could be with the Turbo cars having a center radiator but I'm not 100% sure on that one.

Quotes source post: 17566700

Chris I do have a Turbo. I also have new stock radiators installed this year. My thoughts were along the lines of plpete84 above, if the C4S runs the same radiators, bumper, and inner ducts as the Turbo why can't the C4S enjoy the same cooling system performance? I suspect the M96 does not flow as much coolant and that would be the limiting factor. This doesn't help zbomb or any other C2 driver; perhaps modifing the car to use C4S parts would be a possible route. We looked at putting Turbo radiators in his car and it did not seem feasible with the C2 bumper and ducts. Somewhere I have a data log at Lime Rock with the original 20 year old radiators and I'll try to get that out to see the difference the new rads made. Interestingly the GT3 runs C2 radiators and bumpers as I recall. Do GT3's have cooling issues? I'd think not.

Quotes source post: 17566722

All Turbos have a center radiator. The C4S with Tiptronic also have a center radiator, and I think it was an option on the 6-speed cars.

Quotes source post: 17566637

Quad are you running any inner fender liners at all, or did you delete those completely? The C2 inner liners look very restrictive. I agree that one fan could be removed without problems - GT3's only have one fan from the factory.

Quotes source post: 17566776

GT3's use the same 3 radiators as a Tiptronic C2. They have a cooler on the manual transmission. Despite this, they cool somewhat better, probably due to the front bumper shape. GT3 Cups use a 996 Turbo center radiator mounted upside down in a pretty half *** manner. They run very hot when its 100 out but they hang in there and the engine does not seem to care much. I normally use a 996 Turbo water pump on my Boxster builds. If nothing else it gets the water pump rpm back up to where it was prior to the installation of the underdrive pulley. That's a plug and play mod, except for the belt length. The PET verifies that 4S shares Turbo cooling. That would solve all the cooling problems on a 300hp 996, as long as there isn't something in the design of the engine that limits cooling. Somebody should get a wrecked 4S and experiment with that. My Boxsters use the old headlights so it is not applicable for me unless I change the fenders and lights too. Chris Cervelli Cervelli Technical Service

Changed out the oil pressure relief valve plunger and spring during my oil change this evening - We'll see next Thursday if it makes any difference. The stuff I pulled out looked fine. Also, Regarding temps - the pressure still drops with cool oil - so I would say that DOES NOT fit the bill as resolution to oil pressure instability corresponding to G forces... Mitigation, sure - Resolution, Not for me, my data shows pressure drops still occur regardless of where in the oil operating temp the car is being run, at least on my car. Now - maybe the retort to that is - the pressure instability at the cooler oil temps is non-consequential and as such, resolving the instability to levels seen at XXX degrees sufficiently addresses the problem. I'd be interested in hearing if anyone is of that opinion, how to quantify what the temp needs to be (I saw 230 OT mentioned back in the thread) and also what min pressure / RPM needs to be in order to not be considered a risk to the engine during track use. I think most would agree - the data reflected from Chris' dry sump is resolution and should be the target - Aim small, miss small.

Quotes source post: 17566637

I see the CSF radiators instantly drop CLT by 15 degrees back to back on the same day, same road, same speeds. Funny how this turned back into a thread that considered temperatures again. Duh.

Quotes source post: 17446798

^ What Chris said. Look, guys, if you need another hobby in addition to actually driving the car, then go for it. Otherwise there are only a few options: 1. Keep tinkering and blowing engines up. 2. Cool the water/oil with upgraded rads (number and quality), air ducting, use excellent quality oil and change it frequently, and.... if you're racing or really consistently at the limit of sticky tires... rebuild the engine with high quality coated rod bearings every 2-3 seasons. I say this with significant experience. 3. Get a track/race car that has a dry sump system. Problem solved. /end rant

Quotes source post: 17566637

^ THIS

For those asking about reduction of flow through the rads due to front fender liners, Chris Ruigrok at Carfab.ca makes these screens for the front half of the liners (Boxster/Cayman).

Quotes source post: 17568442

jdsteffa, What gets you involved in this thread?

Quotes source post: 17567271

Always was Jake. Everything has been and always is on the table. Been the best damn thread for perspective and application. Appreciate your comments.

Quotes source posts: 17568586, 17047239, 17047339, 17047411, 17048747, 17057397, 17057426, 17057847

Looks like another dead M96 from over in the racing forum... https://rennlist.com/forums/racing-a...onversion.html For some history on my experiences broaching this topic over there... here's a thread I started last year. https://rennlist.com/forums/racing-a...onversion.html And a couple quotes from the thread :

Thanks Zbomb. Certainly get it. To those that are jumping onto this thread midway and haven't read the whole thing, some clarification may be in order. This thread is about finding ways to facilitate Oil Pressure Stability. Oil starvation, water and oil cooling as well as other items discussed are all part of the equation. I may be wrong, but in no way shape or form is anyone looking to turn their 996 carerra into a full fledged racecar and race it at Daytona or LeMans for 24 hours. Nor has anyone discussed club racing their car. The discussions are about DE and street. After watching the kind of money my buddies that club race spend on an annual basis to keep their cars running, I gracefully said no many times. Many in this forum are belly aching on a 20-30k rebuild after 20 years. Figure every 2-4 years for club racing. Completely different animal in monetary and service requirements. Quite frankly, it doesn't matter whether you have a 993, 996 GT3, or just our 996. The costs regardless of the engine are astronomical in club racing in comparison to DE. So make no bones about it. No one on this thread to the best of my knowledge believe or expect our carerra engines to be race engines. Just looking to find ways to incrementally prolong engine life knowing damn well we will eventually have to rebuild the engine. By the way. This has been the best damn thread period for information. Thanks to all that participate.

Quotes source posts: 17566662, 17566785

Yes all 3 csf Yes i do have fender liners, but they have huge cuts with screens to try to help exit the air

I went ahead and removed the fan and shroud from one of the rads and mega vented the liner. Changes from last data collection : Pressure relief spring and piston Fan shroud and fan removed with vented liner Oil level just a little above half full on the dip stick Really hoping the weather is good and I can push on Thursday for some good times and data to review.

Quotes source post: 17513814

Martin, Two questions. 1. Any updates? 2. Are you producing just the dual scavenge pump or will you be producing "your" version of the X51 package?

The cam cover redesign with the 3.6L engine is due to the 3 chain timing system, which replaced the 5 chain system, removing the cam adjuster from the scenario. This change allowed the same [longer] timing chain to wrap around both the exhaust and intake cam sprockets simultaneously, rather than only driving the exhaust camshaft, which then drove the intake camshaft via a simplex chain, adjusted by the cam adjuster. The addition of the vane cell adjuster to the nose of the intake camshaft also created a change for the primary oiling system, which was not adjusted for by the oil pump, or any other primary oil system changes to support the additional oil flow that the vane cell adjuster would use to adjust timing a full 40 degrees on the intake camshaft only. There's a reason why we instantly saw oil pressure issues with .2 cars that we never saw with the .1 cars, and this vane cell adjuster is the biggest difference in that equation, yet it has been completely overlooked by everyone else~ The 3.6L 996.2 cylinder head was completely changed internally when the 3 chain system was incorporated. Lots of question marks about oil pressure and internal losses can be found within these heads when compared to the earlier engines. We very seldom hear of on track failures with the 3.4L engines (M96.01, M96.02, M96.04), the norm is the 996.2 (M96.03) and the 997.1(M96.05/ M97.01)as well as the Cayman S (M97.21) have the absolute highest on- track mortality rate. Ironically, all of these engines share the 3 chain timing system. I'll just leave that right here, to stimulate thought. Now you guys can dicuss it, and I'll go back and hide under my rock.

Quotes source post: 17571259

well that pretty much sums up my 996.2 going on track any time soon.. unless i can find a track where I only make left hand turns LOL.. joking .. maybe .. So my comment above about the CSF radiators is 100% legit and I will be doing that next.. 3 CSF radiators, LN thermostat, and new water pump happening next as I wait for the new sump and testing..

Quotes source post: 17571281

Just watch the pressures.. If anything cut the sessions short when it is crazy hot outside. Most of our failure calls come in July- August (2 this Monday so far) after DE events in the southeast, especially at Road Atlanta, Sebring, and Barber. COTA also has a tendency for this. Things really start to change after 15 minutes on the track in high ambient temps. Most of these engines fail on the last lap of a 20- 30 minute session, or show signs on the cool down lap. It didn't take long for me to answer the calls 18 years ago and put the puzzle together. Today I stop the caller respectfully in their tracks and tell them exactly what happened, rather than them explaining the sequence of events to me. Things get quiet, and they know that this isn't my first rodeo, and they are not alone in the pool of failures. I recommend a stand alone oil temperature and oil pressure gauge, and drive by these. With these you can know when you need to come off the track.

Quotes source post: 17571311

Thanks for the advice JR.. good news if something does happen I'm fairly close to you and would just get on the list..

Quotes source post: 17571311

How do you know when to come off track ? What measures are you looking for ? Temp is easy enough but the pressure fluctuation occurs and recovers quickly, often mid-corner when driver is eyes up. Trying to watch for the pressure drops may be more dangerous than the drops themselves.

I used to watch temps/ pressures on both straights at Road Atlanta, but no other time. Once the oil hit 260F I knew to watch the pressure more closely, but only on the straight. If I lost pressure on the straight due to loss of viscosity/ thermal breakdown, I knew I would lose even more in the corners. I didn't have to see the gauge in the corners to know this.

I'm sure this might've been covered already, but is there a way to get oil pressure and temp from the OBD rather than running separate gauges? - I found my own answer. There is not.

Buy an early 99, throttle cable car. Less issues with them. Good luck 🍀 everybody.

Quotes source post: 17579634

​​​​​​Same issues plus more prone to cracking sleeves.

Quotes source post: 17579641

4/98 and before . Before the factory fire, lokasil one. Ask Jake who�s the guru on these engines and he let you know.

Quotes source post: 17579730

Early 99 still has oil pressure instability, and if you don't have the UAOS and the IMS solution, you are still far from "safe" IMHO

Quotes source post: 17579730

Jake is the guru. Remember, we are on the oil pressure stability thread. Same issues with version 1 or 2. Oil pressure drops. Oil starvation occurs. 😁

Quotes source post: 17579740

Amen brother.

Quotes source post: 17579740

completely agree. I don�t track my car even thou I take it supersonic and bust the tower every time I take it out. We tested my aos on Tuesday and it is in great shape. Sometime in the future I�ll be fixing that.

Quotes source post: 17579775

I have come to the conclusion that 996 owners are the most procrastinating people on earth..lol

Quotes source post: 17565177

It would be very helpful if you could let the forum know how you configure your track/race engines to circumvent the high oil temps, would save a lot of people a lot of time!

OK! Everybody loves charts! First: Overview chart from 7/29. Plotted v distance instead of time. This is different from previous overview charts but spreads the data uniformly for pressure change comparisons Second: Overview chart from 7/13. Plotted the same way for comparison. Oil temp is a touch cooler, overall pressure is higher, revs are lower, number of pressure drops is significantly less Third: Focus on Personal Best lap from 7/29: more pressure drops than 7/13, but OMG, look at the G's!!! Fourth: Focus on Fastest 'wet' lap from 7/13 with same oil temp at 7/29 Fifth: Focus on Previous Personal Best lap from 6/27: Lap 3 is the personal best lap. Water temps around 199, cooler than other charts here. UAOS with Track Drain, 3/4 range on dipstick, no fan on 1 rad, opening cut in wheel well to help airflow through wheel wells. Higher G's in turns easily explain more pressure drops that 7/13 session. Rainy session. Slightly cooler oil temps to previous chart. Water temp ends at 205(max range on chart is 200, charting error) G's are lower due to wet track, and pressure drops are smaller as well. RPM's are not as high. Overall oil pressure is higher. The curious thing here is that water temp is higher but oil temp is lower than the 7/29 session. I believe this is due to engine load. This session was not run as hard due to the wet track. So even though the water temp is higher, the engine is not dumping as much heat into the oil. UOAS with Track Drain. 1/2 range oil on dipstick. Before removing one fan and opening wheel well. Third chart: Lap 3 from 7/29 session: engine temp ~232 Fourth chart: From 7/13 session, similar oil temp ~233 Fifth chart: PPB, hot ambient, hot engine 250-255 The story remains consistent. Cooler oil temps decrease pressure drops. Higher G's increase pressure drops. Not sure that adding oil from 1/2 to 3/4 on dipstick improved pressure drops. Higher G's may be obscuring this effect. My vote on next steps remains the same: -5w50 oil -3rd radiator -Bigger oil to water heat exchanger Forward comments and questions.

Quotes source post: 17580693

I'd like to focus on this lap for a moment. I'm going to put the track map in here so people who haven't been to this track can see the track layout. The oil pressure drops obviously coincide with the turns. The time from the graph is in square brackets. Turn 1 [1400]: after heavy braking it's a sharp right hander (120 deg) that's very slow, and back on the gas right away after turn in. Turn 2 [1600]: quick right hander (60 deg) that's probably run at WOT. Shift to third in there somewhere. This is really a continuation of Turn 1 and is essentially a single increasing radius turn with a little kink on the end. Turn 3 [2200]. A sharp left hander (90 deg) with good camber so it's pretty fast. Turn 4 [3100]: A hard right hairpin (180 deg) that's slightly off camber so it's fairly slow. Turn 5 & 6 [3600-4300]: A wide left hander (180 deg) that's flat. maintenance throttle is typical. It's doable as a double apex or as a long carousel. Turn 7 & 8 [4900-5100]: A wide fast right hander (180 deg) that's slightly banked. This is one end of the small banked oval. I consider this one a carousel too. Turn 9. [5800-6100] decreasing radius left hander (110 deg), very bumpy Turn 10 [6500]. off camber right hander (110 deg) Turn 11 [6900-7100] fast uphill right hander (90 deg) that sets up the main straight. I think it's plainly obvious that right hand turns are the issue, longer right hand turns being more of an issue! The fast right hand 7-8 complex run at relatively low rpm is of interest. The high G at turn in combined with low rpm causes the huge drop. The slight relaxation of G towards the end combined with increasing rpm brings the oil pressure right back up. The fast turn 10 also shows a big oil pressure drop, which again recovers after slight relaxation of G and increasing rpm. The other point of interest is the drop in turn 9, which is a left hander that's bumpy and under heavy braking. This is the only significant pressure drop in a left hander. A final observation, the oil pressure doesn't go below 30-ish PSI. Also the largest oil pressure drops are with the engine under maintenance throttle or under coasting or braking, with much less load than hard acceleration. It appears to me that the initial "slosh" of oil on right hand turn-in precipitates the pressure drop and once the G load is set up a steady state occurs with lower than desirable but still OK oil pressure. This appears in the turn 7-8 complex. There is not enough time spent turning in turn 4 to set up a steady state, although the beginning of a steady state condition can be seen.

Quotes source post: 17580693

Paul, nice work. Questions to all. 1. Does a CSF radiator actually add any more cooling than a stock radiator? Advertisements suggest 10-20% more cooling. 2. In addition, how much does a third radiator add towards cooling? 3. LN/Charles is working on the bracket for the oil cooler. George was offering his expertise. Any news from Charles or George? As food for thought, it may be time to open the Go-Fund-Me account again to test a 3rd radiator or test an oil cooler. I am in if Jason and readers are open too it.

Quotes source post: 17571281

just turn left. Then no issues.

Awesome work Paul - unfortunately I’m a little time strapped to fully soak it in at the moment and will review when time is adequate. Just to add some more context. Below is video of the PB lap for folks to pair with the data and Aaron’s track map. Driving starts about 4Min in. You Tube

I don't want to start a radiator war. If the csf radiators are the same size as the stockers they will be within a few % of the stock rad efficiency. The csf may last longer as they're all welded aluminum instead of crimped onto plastic end tanks. The biggest improvement will be to have clean radiators without clogged fins. This may negate the longevity of an all welded aluminum rad, as it will get clogged just like the stockers and need replacement on a similar schedule. If csf has managed to shoehorn a bigger rad into the same space then it's worth a look. The rad thickness may or may not improve efficiency depending on a host of design choices so I would not base a decision on that alone. Certainly making the length and width larger would improve efficiency. The aluminum end tanks will slightly improve efficiency too. It would be nice for this test if we had a real hot day. That has not happened so far this year.

Quotes source post: 17580799

Any benefit to using turbo, C4S or Turbo radiator assuming they fit?

Quotes source post: 17580806

They don't fit unless the entire front is changed to the C4S/Turbo setup. If this is done there ought to be a HUGE benefit. I don't think I have seen anyone take a C4S on the track. It'd be interesting to find out their water temp values. I know mine with the Turbo which are essentially irrelevant to this discussion. Another gem that may have been overlooked: the Turbo water pump fits the M96 and because of the size difference between pulleys runs at higher pump RPM for a given crank RPM. The pulley sizes can easily be verified - anyone with an M96 want to carefully measure the water pump pulley? I have a spare Turbo water pump here I can measure.

Quotes source post: 17580891

Interesting. Looks like I had a typo and had turbo listed 2x instead of GT3 radiators. So any idea if GT3 radiators actually would fit? What I am driving at is that there is a boatload of aftermarket cooling parts such as CSF radiators available that are at comparable prices to Oem. Our cars are pushing 20 years old. Most radiators probably need to be replaced. CSF advertises 10-20% additional cooling. So for the same price as Oem, 2 new CSF radiators may or may not be a better choice for cooling. Same may able to be said for the GT3 radiators if they fit. But if there really is no additional cooling benefit to CSF or GT3 radiators, then your only option for additional cooling is a center radiator. So what actually works with radiator cooling and what are the options. Seems to me a 3rd radiator is a no brainer. But can we actually get better cooling with the other 2 radiators or is it just advertising bs. Clearly Csf if well made and durability may be a selling point nevertheless.

Thought the GT3 used the same radiators (including the center). CSF part numbers are the same.

If that's the case, then unlike using the Turbo and C4S radiators which are bigger, there is no benefit to the GT3 two radiators over OEM 996. So is there truly a benefit to CSF or just a rumored benefit? And how much additional cooling do you actually get with a center radiator.

Can't remember if it was in this thread or another recent one where Chris said he didn't see any difference between CSF and OEM while Jake said he saw a 15 degree drops. I know there's a quite the difference between the air in Colorado and Georgia but gotta be some other factors there. I actually had some center radiator parts from DC Auto delivered today so I'll be tackling that project sometime this year, but I'm still on the fence about which radiators. Probably swap the original left and right while I'm at it.

Quotes source posts: 17580988, 17566700, 17567271

It was this one Yaz. For water cooling upgrades for my use and my money - I would like to swap the full C4S or Turbo setup on to my car including the bumper and shrouds, I don't really have a preference if the rads are newish stock or CSF. I would do the center CSF or stock rad in the interim, but I don't plan to allocate any of that planned C4S/Turbo setup money to upgraded C2 side rads.

Quotes source post: 17580988

Yeah, I remember reading it as well. Hopefully you can see where I am going with this as a byproduct of Paul's analysis. We know we need more water cooling and oil cooling. The question is what water cooling exactly? CSF and OEM radiators cost about the same. We can all probably benefit from new ones. If CSF radiators actually cool more, than its a no brainer. If not, and they are the same, then it's a matter of what's on sale. But, we still need more incremental water cooling, so our only choice then is a 3rd radiator which takes some more fabrication but obviuosly doable. Which IMHO, would be a great exercise to "go-fund-me" if Jason would be interested. If not, I understand completely. However, we keep coming to the point of discussing the importance of more cooling and now is our chance to measure it properly.

The car, and the bumper used, as well as if a 3rd radiator, or even if a front license plate (or an offset front plate) is ran on the car will impact the results to the post that blanket statements can't be made. 100% of my CSF Vs stock radiator testing was done on the street in the middle of July/ August, where its already 90 degrees ambient by 0900 in the morning.

Quotes source post: 17581029

Was your testing with 2 or 3 radiators?

Quotes source post: 17581049

Both, with every version of 996 and 997 also. I have done this test indirectly many times as we have completed cars, and the owner decides to upgrade the radiators while we still have the car.. Dozens of times this has been the case, not just a couple, and not just one generation of these vehicles. The front license plate impact was one of the most revealing tests that I did, and like most things that have taught us the most, it wasn't planned at all.

Quotes source post: 17580945

That's true, and confirmed by De Jeeper. I understand that the Cup cars run a Turbo center rad (bigger than the M96 center rad) with some bracket hackery and hose mixing and matching from the factory. Since I know Zbomb has a Cup car wreck perhaps if those parts survived it might be a way to add a big center rad.

Quotes source post: 17581063

Center rad and pretty much everything in the vicinity destroyed. The rad that came off the car was full aluminum, don't think it was original. Old wheel to wheel race car life...

Quotes source post: 17581012

So what I am hearing is that if you want the best damn water cooling, you find a way to use the turbo side radiators and the third turbo center rad. In this case you will need a turbo front end. If you don't want to mess with changing the front end, you can go with 3 CSF rads on the Oem front end with fabrication on the front bumper for air. And the center turbo rad would able to be used with either the turbo/c4s front end or the 996 NB front end? Is that an accurate assessment? Edit: reworded after Zbomb and Prf post.

Quotes source post: 17581066

The C4S uses Turbo side radiators and an M96 center radiator. Presumably a Turbo center rad could be used on a C4S. All the other hardware such as radiator mounts is the same between C4S and Turbo. I believe the Turbo center rad could be used on the NB M96 with suitable mods to the brackets and such. I haven't got a spare center rad, I threw out the junk center rad I replaced earlier this year.

Quotes source post: 17581057

Jake, if we were to add the 3 turbo radiators, would they be usable in your engine rebuild?

OK, I just went and rewatched some of my data. I ran Lime Rock in my 996 Turbo last September. It was fairly hot, probably 90 degrees. My car has ~450 HP and an automatic transmission adding to the heat. The highest coolant temperature I had was 206F and the average was 202F. Over the winter I replaced all three rads which were probably original 20-year old, 180k mile radiators with new Mahle OEM equivalents. I did nothing else to the cooling system. In May this year I again ran Lime Rock with Jnelso this time. Temperatures were similar (it was hot for May!). The highest coolant temp I ran was 190F and the average was 186F. Now, I was pretty happy with 206F after flogging the car around the track for 20 minutes. Changing to new, clean radiators cut that to 190F - a 15 degree improvement! I should point out that I'm datalogging coolant temp from the OBD2 port. I don't have a standalone datalogger.

Quotes source post: 17581113

Now that's a strong endorsement for turbo rads.

Quotes source post: 17580769

We are using oil pressure as a proxy for oil flow. This proxy is effective as long as the oil supply is continuous and the fraction of air in the oil remains constant. When the pickup is uncovered, the fraction of air in the oil goes up, and the oil pressure/oil flow proxy breaks down. Therefore it is incorrect to state that a certain oil pressure is ok or not ok. The oil pressure drop is indicating either or both a drop in oil flow or a higher fraction of air regardless of whether the minimum oil pressure reached is sufficient (given a normal fraction of air) to prevent bearing failure. Zbomb: replace the side radiators and install the center radiator. Then the coolant temps and oil temps will be reasonable (200 and 220) and we can stop wasting time talking about magic cooling solutions. CSF or stock will give the same temps. Chris Cervelli Cervelli Technical Service

Quotes source post: 17581157

Yes, I agree. Air entrained in the oil flow is almost certainly what is causing the oil pressure to be lower. This is bad in may ways as you have pointed out repeatedly.

Quotes source post: 17581157

Hah! Can't agree more, which is why I propose that we put an end to this discussion by proving the worth of a 3rd radiator. If you are sick of talking about it, let's do a go-fund-me account to remove this variable.

Quotes source post: 17581185

I will provide the side radiators at my cost, which is $500 for pair. It looks like my supplier no longer carries the center radiator. I think I have in stock the pair of hoses Zbomb needs to install the center radiator. I will throw those in. Chris Cervelli Cervelli Technical Service

Quotes source post: 17581185

I'd be willing to donate more of my hard earned cash to get the tests/data from Zbomb, while making his car better and more knowledge to the platform. But it may be hard to come up with enough funds from the peanut gallery to do it.

Quotes source post: 17581120

I'm not suggesting converting to Turbo rads, because that involves changing the bumper and air ducts and radiator brackets and the bumper beam and on and on. That is a huge investment and changes the look of the car completely! I wanted to point out the significant reduction in water temp with clean radiators on the same car. In case anyone is interested, here are the radiator sizes: C2 side rads are 187 square inches Turbo/C4S side rads are 220 square inches C2 front rad is 103 square inches Turbo front rad is 159 square inches Turbo with 3 radiators adds up to 599 square inches. C2 with 2 side radiators adds up to 374 square inches.

Quotes source post: 17582040

Good info thanks. It is good to know that it's a viable option if more significant cooling is wanted. And many may want it especially if they are looking at a front bumper change for more aero anyways. Why not go all the way. I for one am in this situation.

As I said before, my preference is to do the turbo setup, L, R and center, I'll need as much cooling as I can get for phase 2 of the car's life. If I have to mod the bumper until I can get my hands on a Turbo / C4S bumper setup - that's not an issue for me. Even if the effectiveness of the rads is somewhat reduced due to the bumper opening in the C2 - I'd still be willing to bet it would yield a good improvement over the C2 rads with such an increase in surface area. I'd be happy to test the C2 stuff if people are interested, but that setup is not where I plan to end up.

At this stage, waiting for Skip's UIDS and Charle's Oil Cooler, which are potential solutions to prolong engine life, oil stability and mitigate oil starvation, I propose again that we get the Go-Fund-Me account going again and test the center radiator once and for all. To all, remember, we do not have agreement between the "experts" on this thread regarding this topic. If you are sick and tired of going in circles without concrete data, now is the time to test it. Let's get this done. The question we need to settle is if we do it with a turbo center radiator or a C2 center one. My two cents.

My $.02 I think u guys r comparing apples to oranges. Im not saying that the bigger rads wont help but the 2 motors r completely different and comparing how they cool is worthless. The turbo and the gt3 r the same motor and the gt3 had zero problems cooling the motor in the track with the smaller rads and 400hp. My point is there should be plenty of cooling capcity if designed correctly. The issue is the oil is not shedding enough heat. I really dont think once the motor gets heat soaked, ie t-stat open, that the larger coolers r gonna help. The car will hit the same equalibrum temp but u may be able to delay that point so u can get through a 20min session. Chris has basically seen this in his enduro cars and shared that data. I think the real answer is the larger cooler that chris has already got to work in the cayman and what ln is working toward. One of u Chi peps need to get ln a car for testing and fitment.

Clean the normal radiators. Run 33 percent coolant mix. Add water wetter and you should see improvements. Not just vacuuming, used a chemical spray. Yesterday it felt like 109 here in Houston I took the car to 150 going fast with a Camar� and my water temp never went past the middle of the 8. if you run 33 percent af/coolant you have to constantly check for water evap.

Quotes source post: 17582214

De Jeeper, appreciate your comments and your contributions to all of us. I couldn't agree more and disagree more. 1. We all know the mezger engine is the better engine, but we are looking for solutions to the engine the vast majority of us have. With that being said, most of us will not go thru the exercise, nor have the skills to exchange engines like you, nor do we want to sell our cars and buy a GT3 or Turbo. Instead, we have to work with what we have. Which is what this thread is about. 2. We all know that the dry sump kit that Chris has designed is a very attractive solution that has addressed the oil starvation/stability issue. BUT...it is not yet available for the 996. I hope that Chris is finding a way to package and sell it for the 996. ​​​​​​3. Further, none of the "experts" on this thread agree on cooling. It's now time to see if a third cooler actually works on the track, not the street, and by how much. It's also time to collect the data and quantify. So we have two choices. Continue down the path of broadly discussing topics that are not quantified and spend stupid money on solutions that don't work. Or we we measure those solutions to actually see if they work. Don't k now about you guys, but the prospect of being a lemming is not very attractive to me.

I agree with the testing and research, but i wanted to make a clear distiction between the motors. Im all for finding a solution for the issue that most could install. The turbo rads will most likely be a $2-3k investment and i think the cooling stack is fine for the hp. I just see the money being better spent on directly cooling the oil through a larger capacity cooler.

Quotes source post: 16050132

With all the talk on cooling... I'm just going to leave this here.

Quotes source post: 17582311

You may be absolutely right about oil cooling as being more important. Believe me, i can't wait for Charles' OIL Cooler to be ready for bolt-on use. Not to mention Skips UIDS. I'll be the first one in line for both. And I will be happy to report my findings. Unfortunately, I do not have a sophisticated data gathering system to accurately report any findings. Because we are waiting on the UIDS and LN Oil Cooler to be in production, we do have an opportunity to quantify the benefit to a third radiator. It would make sense to me that there would be a benefit, Jake and Charles have started there is a benefit, but to your point, Chris' point, and Baz's, maybe not. The only way we will know is to test it. Hence my suggestion. No worries if we don't want to collectively do this with a go-fund-me account. I certainly understand. It's just a group suggestion. The cost may end up being prohibitive, but if not, it certainly will be a damn shame if Jason would be willing to use his car and analytical tools, that we don't quantify this issue once and for all. IMHO we waste too mush time on RL going in a circle. It's fun, but at the end of the day, don't we all want quantifiable solutions versus non-quantified points of view? ​​​​​​Enough of my rah rah speech. You guys get the idea.

Chris, What the... you are holding out on us...

Quotes source post: 17582374

Regarding the third radiator. I have the same set up as Zbomb (LN 2qt Deep Sump, screw on filter, UltimateAOS with drain back. I do not have new scavenger pumps, but I do have the 3rd radiator, and I still have the same pressure drops. I don't have the data logging to prove it, but I know what the gauge says.

I've always done air to oil cooling. I have not favored the water to oil because it is another intermix point. It does help the oil warm up, which is a pro. I have always treated the water-cooled cars like they were aircooled, and it works for me, with my clearances, coatings, and engine combinations that are designed to run cooler in the first place. Heat that isn't generated, doesn't have to be dissipated.

I saw an FSI badged car the other day - did not get to chat with him about his setup, hopefully I'll get the chance next time I'm there on the 16th. Nice looking .1 with RI plates, maybe you know him. ETA... You don't find big pressure drop with your air / oil, I'd be curious where you mount the heat exchanger. I think every track experience I have read about here using air / oil exchangers has been negative for the 996 application.

Quotes source post: 17582388

Haha. Yes, I have done a fair amount of experimentation I have not discussed here. The system pictured above was originally designed as a transmission cooler for the endurance racing Boxsters. There is a very serious problem with a powerful (300+hp) mid engine car's transmission temperatures and this was how I addressed that. It was successful. I have a nice kit for the 987.2 that is plumbed differently but works just as well. Sooner or later I will do a PDK version. Powdrhound's 996Turbo project has been ongoing for many years now, and it is still ongoing. He asked me to reduce the oil temp on his engine and this transmission cooler was easy to adapt into this role. I have also used this on a Boxster for oil temp reduction on a car that sprint races and therefore does not suffer so much with the transmission overheating. (The transmission has a great deal of mass and takes a long time to reach its equilibrium temperature, so in a sprint racing car the transmission winds up not spending much time at very high temp even with minimal cooling.) Using this as an oil cooler is not all that relevant to the M96 discussion because of the following reasons: 1. On the Powdrhound 996T, the high oil temps were cutting the sessions short. That is, those temps were a hard limiting factor preventing the car from running further. This is not the case with an M96. 2. The 996TT has a really good water cooling system, so dumping more heat into it made sense because it still had some capacity to dump the heat overboard. This is much less true in the case of a 996 or Boxster/Cayman. The Powdrhound engine is now making 775 rwhp (and soon to be considerably more) and it is getting close to what the cooling system can handle, even for short (>30 min) sessions. This car is not used for endurance racing. 3. The turbo engine had a larger delta between coolant and oil temps, so the cooler is automatically more effective at reducing the oil temp. 4. In testing the cooler knocked 10-15 degrees off the oil temp, which fixed the problem on this application. The same 10-15 degree reduction is not much of help with the M96. 5. The system is quite expensive which should not surprise anyone looking at those pics. I have not commented on LN's upcoming oil cooler upgrade because while it may allow the same incremental improvement as what I did here, it is not going to have a big effect on what we are calling oil pressure instability. As I said before, Zbomb should have installed new side radiators and a center radiator first thing so we can stop talking about temperatures. This is no doubt whatsoever that radiators will be money well spent and there is also no doubt that there will still be oil pressure problems. The GT3R and to a lesser extent the Cups get by with the small non turbo radiators because the front bumpers are shaped to maximize cooling at the expense of drag. This is also true of the 996 Turbo. So the correct upgrade would be to use the 996 C4S or Turbo bumper and all the radiators and ducts. The 996 GT2 handles the center radiator differently and adds an additional duct to help air flow thru that radiator. All this stuff is in use on the Powdrhound car. It also has a Ruf front bumper with larger inlet ducts. I have to add that GT2 parts and GT3R parts are brutally expensive and sometimes unavailable. I don't consider using that type of thing on a car that costs less than $100k. Chris Cervelli Cervelli Technical Service

Hypothetically speaking - I have a dry sump or Skips sump and they resolve all instances of oil pressure instability due to the pick up being exposed to air… Do I care then about oil temp of 260 and water 225 in a 20 Min DE session ? If the oil pressure is resolved, the value in bringing those temps down is… Im not trying to be a smart ***, I’m asking because I don’t know other than… when it comes to temps lower is better.

A club member here who I saw yesterday has a protomotive 3.9 twin turbo that has 920 hp at the wheels without the methanol injection. Anyway his car is basically a street legal in your face race car. He was at the cars and coffee yesterday, he actually parked his car in front of mine. Anyway he does tracks, autocrosses etc. Apparently he is a bad as? Driver because people ask him constantly to drive their cars. Anyway he works with a shop here in Houston Driversource and he himself designs, engineers some of the stuff. Anyway this car has more than 150k$ in it. I bet it has to have a bad *** cooling system. I�ll see him next Saturday and will ask what have they done. I know if he is running the methanol injection that cools the hell out of it.

Quotes source post: 17582539

If it does not exceed 230 coolant and 260 oil on the hottest conceivable day then there is little value in adding cooling. I would accept these as peak temperatures but would not want to run the entirety of an endurance race at these temps. Chris Cervelli Cervelli Technical Service

Quotes source post: 17582498

There's only two FSI badged cars that have ever left here with the badging. Both of them belonged to our family, one was mine, the other was my wife's. If the one in your pic was FSI badged it is likely my wife's old Land Speed car that she set her records with, it was a 99 C2 factory aero car. The last record was set ion 2010, and she sold the car in 2011. The car was sold with a stock displacement engine that we didn't build up internally, it was still a 3.4 and was used to set her last record in the 3.4L class. The last I heard of the car it was crashed sometime in 2013 or so, and was repaired, but I think it has had a couple of owners since then. That's the car that's in my avatar here with the red Fat 6 Innovations" windshield banner, that is the only car that we've ever fitted any banner to. I haven't heard anything about the car in years, so I have no idea who owns it or what it is running engine wise now, I doubt the old stock engine is still alive that the car left here with. Anyway, my air to oil system uses a hand built cooler element that is fitted to the factory 3rd radiator location, using the factory ducting, and mounting. You can't tell it is there, and because of that many people just think it is a factory 3rd radiator. On cars that have a factory 3rd radiator, I remove it and then bypass it. I used intercooler cores to build these in house and route the oil lines down the center of the car. I sometimes used primary oil to feed the cooler, but would often use secondary oil returning from the scavenge pumps (routed externally) to pass through the cooler with the aid of an auxiliary pump. I make a 35% larger primary pump to fit the stock location and always use that if using primary oil to feed the cooler. Most of these have been fitted to Caymans, but one forum member here (Randy Maxwell) has one in a 996 with my 4.0 built 7 years ago and is a dedicated track car. This has proven to not only keep the oil cooler, but also to keep the coolant cooler (even when only using the two primary radiators) since this partitioned cooling system removes the oil cooling load from the coolant. The location, and cooler efficiency is the key. I loathe oil coolers that mix oil to coolant and consider them a failure point ready to take out an engine. As a matter of fact one of my tracked military vehicles is currently down due to a failed oil to coolant heat exchanger fitted to the factory 6BT Cummins diesel engine that it uses. They have their place, and I believe they are essential in cold climates to help the oil warm up faster, that is what they are really called a "heat exchanger" rather than an "oil cooler". The main source of heat in these engines has been the valve springs, the only thing that cools them is the oil, and running high oil temps has also proven to kill the spring rates. We rate every spring and have seen stock springs on track cars lose 40% of their tension over 4 seasons of use. Having new springs wound with less interference and a more progressive spring rate made a big difference, especially when being treated to isotropic super finishing and cryogenic enhancement.

Quotes source posts: 17582539, 17582744

That's what I'm talking about "technical discussion" at it's finest. Hard Core Q&A ...

Quotes source posts: 17582744, 17582752, 17582752

Thanks Chris - for this setup, this is the reason I have not made real cooling system changes as of yet - next season will be a different story. I get that I'm on the edge on hot days as is now but am in no way expecting changes I can make to the cooling system to fix an oil pressure (Flow) problem that is cause by the pickup sucking in air. Here's a pic of the car from another event Very cool on the larger capacity pump to drive the oil to the front of the car, which seems like the only reasonable place to get enough airflow to make the air / oil exchanger work. I have heard some people use an external pump with a closed loop system around the sump to get away from using the factory oil pump to drive the additional lines - I like your idea better. When you say the revised valve spring made a big difference, do you mean to heat generation or spring longevity ?

Yeah, that's Beth's old car. All four of her records still stand. I don't know the current owner, and I am surprised that he left her banner on the car. I should have removed it when she sold it. There's mote than one way to make an oil cooling system work, lots of my experience with old 4 cylinder, wet sump air-cooled Porsche's came into play when i started playing with all this stuff. New springs both lead to better valve control, and less interference which drops temperatures. I have even taken some engines that had factory dual springs, and made a new conical single spring to eliminate the interference that elevates temps. Tuning also plays into this, and then we go back to EGT, and start to consider heat soak. What is best for power is usually not best for temperature long term, and heat soak changes everything. This is why a track tune should never be ran on the street, and vice versa. Again, the key is not generating heat in the first place. Do that, and you have instantly won.

Quotes source post: 17582948

Thanks for the info Jake, the valve control stuff is interesting and not something I had thought about - cool to read your experience. Veering back onto the oil control with a little veer back to the cooling side - cross posting this from the UIDS thread, Looking at the sump I was reminded of an oil cooler I saw for a power glide where the sump itself has a water passage running through it - built by Mezeire... Just something I thought was cool and another take on possible cooling solutions.

Jim Pasha and I talked about a jacketed oil sump plate years ago.. In 2008 when he came here to do the Excellence tech article on my program we discussed it. I made a prototype using some copper tubing that passed through an original LN deep sump. I think it is upstairs at the main shop, I'll try to find it next time I go there. Back when I tried it things were not nearly as advanced as they are now, might be something cool to bring back.

Quotes source post: 17583011

This is something I have been contemplating too along with a separate cooling circuit that uses the center radiator. It would still be oil to water but it would put the cooling right where it is needed in the oil pan. I have noted that when my oil is cold, oil pressure is never lost. This is why I think the entirety of the problem is with high oil temps. Plus, the data so nicely graphed above support seems to support that theory . (Fully acknowledge that my opinion is supported with N=1 data point,namely, my car.). This thread has been very good to read and many thanks to those devoting their time money and knowledge to it.

And the same holds true till around 15 minutes on the track. Once the higher RPM builds a ton of heat from the valve springs, and other friction within the engine the oil things out from heat. The other lesser discussed aspect of this is what happens to the clearances within this entirely aluminum engine. Some clearances open up for the better, to reduce shear, BUT some others that are critical open up and create a bigger issue. Machining an oil pump body from steel was eye opening, so was making inserts to fit into the oil pump body that were steel, and did not change dimension with heat. The double negative is open clearances with oil that gets too hot. Hell, I'm just glad that you guys have finally accepted that the biggest issue is oil temperature. When this thread started, and I mentioned that, I never expected that you'd listen, or even admit it. I learned this 15+ years ago and since then everything I have done has been to reduce the operating temperature of the oil. This isn't limited to just bolting on an oil cooler, but this comes from the internal configuration of the engine, clearances, coatings, and components. I "fix" this issue from the inside out. Most of my engines don't run deep sumps, X51 style oil control baffles, or external oil coolers. They don't need them. They are options, but most of the time I dissuade the customers from buying them and tell them that we can add them IF their driving style, car preparation and other variables prove that they need some added protection. The biggest benefit the 2 qt deep sump has is the 2 additional quarts of oil volume. This increases the amount of elapsed time that it takes for the volume of oil to reach the critical temperature on track that changes everything. This usually gives the driver 2-3 more laps or around 10 minutes more in the DE session to drive before the massive thermal change makes impacts. I will say that once the additional oil does get hot, and the engine becomes heat soaked, the extra volume of oil does take longer to cool down than the standard oil volume. Most failures happen in the last 5-8 minutes that a car is on the track, so the additional oil volume acts as a very good band aid. When our local PCA had a hard time filling DE slots they upped the track time from 20 minute sessions to 30 minute sessions. As soon as this happened the failure calls were received locally doubled instantly. This was especially true when the August DE at Road Atlanta came along, and the heat is the worst of the entire year. That extra 10 minutes was everything. DE instructors that get to spend several back to back sessions on the track are the worst with these engines, and failures. They are faster drivers, but they also can "hot lap" and go out several sessions in succession. This is murder on the oil temps and heat soak. When we were doing a lot of our R&D work in 2008 or so LN had a Boxster with my engine installed, it was a stroker built from a 2.5 and the way we punished it was letting every advanced driver at the track beat on it session after session. It logged 14,000 track miles in 2.5 years and never had a "starvation" issue. It only failed when the crankshaft snapped in 3 pieces at Road America. That car did see oil temps as high as 280-300F sustained, but maintained sufficient pressure to live. It was using different test oils at the time, one was " formula 980" which became Driven XP9 at a later date. With a factory engine that hasn't been disassembled and whipped into submission, you don't have much of a choice except bolting on things to help. When it finally pops, you do have choices, and should take advantage of the opportunity that the failure creates. Till then: - try to avoid the track in July/ August or when temps are blazing hot. We don't get failure calls in cooler months, except from Florida and Texas. The phone explodes in July and August. -Change oil just before going to the track, and choose whatever flavor you want. Fresh oil is better than anything you can do to the engine before a track event -After you come off the track try to leave the track and go for a 15 minute cruise on surface streets to allow the engine to cool more evenly and thoroughly -Keep in mind when you go back to the paddock and shut down a heat soaked engine that it will GET HOTTER before it starts to cool down! -Don't hot lap, always let the engine cool down for at least two sessions before you go back on the track -If you can run a little more octane, just because it keeps detonation down, and reduced the chances of the protective film of oil between the rod bearings and crankshaft from being "squeezed out" when viscosity drops, and shear is experienced. This also helps to keep the entire engine a little cooler. I always want to see my hardcore track guys running 100 octane at the track, with the appropriate tune loaded into their ECU to accommodate the 100 octane. This isn't for power, as much as it is adding a margin of error. I have two chapters in "All About M9X Performance Engines" that goes over these things, but the above synopsis will stimulate thought enough for most to avoid an issue.

Quotes source post: 17581157

I am in 100% disagreement with your statement oil temp being the biggest problem if the problem is maintaining oil pressure (flow). The data in the thread shows oil pressure drops across operating temps. In my opinion, the biggest problem with regards to maintaining oil pressure is keeping the pickup from sucking up air.

Quotes source post: 17583698

Nope: The biggest issue is that the main bearings are intermittently fed air because the oil isn't in the sump. Chris Cervelli Cervelli Technical Service

Quotes source post: 17583339

If you�re not seeing pressure drops on track, I would guess it has more to do with you just not seeing them than anything else, including cool oil temps. I don�t think people relying on factory or even aftermarket gauges that do not have the capability to log have much of an idea what is going on. That�s a huge part of why I put the data logging setup on my car, to give quantifiable visibility into what I was seeing through the gauges, which turned out to be a small percentage of what was actually happening which was much worse. Go back, look at the data logs in the thread - the data is there.

Ok, you guys can keep working on it.

Quotes source post: 17583838

OK, so are you saying that you can correct G induced oil pressure drops with cooling alone ? I�m not really sure if you�re saying the above or if you�re saying that at reduced temps the drops will still occur but they do not possess a risk to the engine.

Quotes source post: 17583838

I agree with Jake about time, the technical term is residence time. Oil releases air and it cools. If you add an oil cooler you add oil capacity. Now the tank reservoir sump has to be designed in a way that when oil flows it releases air and it gets more time to cool.proper baffling.

Did you mean to say �oil releases air as it cools?� Could be that increased temps decrease the viscosity of the oil, and increases aeration due to increased velocity of the oil. So it does make sense that oil temps are a contributing factor, even if it doesn�t show on the data.

Does oil hold more dissolved air at a higher or lower temperature? I know pressure holds more, but in terms of the low pressure sump area, that�s my question.

Quotes source post: 17584126

a properly formulated oil has additives that help it release entrapped air. Is silicon that encapsulates the air bubble and makes it float to the surface and then it pops. Also when you design a reservoir for oil, you guide the flow of oil to curves, channels, etc, so that it cools, and helps it release entrapped air. The worst situation is with a hydraulic system due to the oil function, but if you have a general understanding of this you know what we are talking about. There are industry tests that you can see the oils ability to dissipate entrapped air, unfortunately most of it is in hydraulic systems and not motor oils. Anyway, Jakes mentions cooling cycle and the driver, yes I take my car supersonic and has been as high as 160. When I cool it down I take it easy for a while allow the oil temp to stabilize, to the point that I had squirrel 🐿 test the fan in my engine compartment because it never came on. When You shut it down temps are going to rise, but if you stabilize it, in my case the auxiliary fan was never coming on. A big variable is the one behind the wheel and driving practices.

Quotes source post: 17584167

Have you driven a DE session on a track? A single blast to 160 on the highway (in addition to being unsafe) has no connection to a 20 minute session on a track. To replicate track conditions, you'll need to accelerate at full throttle in 3rd and 4th to 120 MPH, brake as hard as possible to 30 MPH, and repeat this for 20 minutes while cornering at 1.2 to 1.6 G's.

Quotes source post: 17584167

These are not variables in endurance racing. The engine isn't going to be shut off for the duration of the race unless the rules specify that it has to be shut down for pit stops. There is no option for treating the engine better if you want to win. If the driver is asked to reduce the maximum rpm or otherwise change his driving due to external conditions, then the team has failed to prepare the car properly. This is relevant because an engine designed to survive in these conditions will also survive in all less difficult conditions. All machinery, Porsche engines included, is designed like this: envision the worst case scenarios, and design to survive those. Because of this I don't think discussion of driving or post-session practices adds anything to this thread. Chris Cervelli Cervelli Technical Services

Quotes source post: 17584209

Yes sir I have done plenty of track. From 2006 to 2013 I had a 2004 VWR32 HPA twin turbo 655 hp at the wheels. That�s faster than any 996 here and probably faster than most turbos. Anyway been there done that. As someone that formulated oils and was involved in NASCAR, SCCA, Indy, drag racing and a bunch of Motorsports I understand a bunch of the issues and have seen them before. My 3/98 996 is never going to have the power or be as fast as my R32. Now the R32 looked like a brick in front of the wind at 150 mph, the 996 has a drag coefficient of .30, Nice and cool. Just wanted to let you know I know there are a lot of differences and I know what they are. The reason I let HPA built my car was that it was a total package reliable and well matched up. I�m coming from the old days when you would see cars with huge amounts of power and they couldn�t put it down to the ground, be reliable and would break. Once again my fast and furious days are past, I want a reliable super sonic car, where I can take it to 150 for a bit and then come down to 65 mph and be happy.

3/98 - ON TOPIC please.

I�m sure it�s complicated, and lots of variables, but I was asking about one. Either oil holds more air at higher temps or lower temp, or neither.

Quotes source post: 17584376

mike in theory when oil is hot and it thins out the encapsulated air particle, bubble should go to the surface and burst faster. In the old days because you were concern about cavitation a simple test that was used , was basically to introduce an aerating stone (Like in a fish tank) create a certain volume of foam and then time it for it to dissipate at certain volume levels. I did a quick search and found the below article. STLE( Society of Tribologist and Lubricantion engineers) is a great source. I�m Not certain this data would be available and I will check with my contacts . If I recall correctly at Valvoline the VR1 racing engine oil had additional additives to combat aeration because racing engines turned higher Rpms, on a race your average rpms are much higher than daily driving. Whoever has a contact with Driven should check with them to see if the address this in their formulation. It is definitely possible to formulate an oil to aereate less.

Good stuff. I found a source that suggests that air decreases with increased oil temp, so that�s not a contributing factor. Oil with impurities, however, is a factor, so clean oil will aerate less, all things equal. https://en.oelcheck.com/wiki/air-and-foam-in-oil/

Quotes source post: 17584806

not a lot about motor oil but a bunch of about hydraulics and turbine oil. I�ll talk to Valvoline and Mobil.

Google is your friend: https://www.advancedta.com/blog/2016...should-we-care I would think one main advantage of the dry sump system is having a low-pressure reservoir where dissolved air can escape and foam can dissipate over a longer residence time. Wet sumps don’t provide the residence time.

Quotes source post: 17584883

Sure they can... Residence time is a function of sump capacity. The standard dry sump tank is 3 gallon, and is ran at 2/3 capacity. The 3.4 m96 engine has a capacity of 9.3qts + 2qts for deep sump = 11.3qts during oil change, a fresh dry motor will take even more to fill first time.(over 3 gallon) ..

Quotes source post: 17439914

Jake - Hope you can answer my question above but am wondering if it may have came from a post from user ltusler who is running your track performer engine. It seems like your motor, despite the pressure "droops" still maintains good reliability, would you attribute that to maintaining your preferred oil temp ? Are the droops due to the pickup becoming uncovered or some other cause ?

I originally built this engine in 2008. It has been freshened up once since then(2012), and made larger at the same time. It has been making 440HP for close to a decade. The first time around it saw over 200 hours of dedicated track time. That said, you guys can keep looking at all the squiggly lines on plots and I'll do things my way. As long as we all reach a result that is acceptable. I have too much going on with developments for cam adjusters, cylinder heads, and etc to state things that add to this discussion any further, only to be told that the squiggly line doesn't prove that its right. You guys will learn a lot more if you continue doing what you are doing, gaining experience, comparing squiggly lines, and sharing info. Maybe you won't blow anything up, or terminally injure anything in the process. I wasn't that lucky, and failure is the only way that I learn. I didn't have anyone to share info with, or to ask questions to. No life line. Everyone in those days thought I was a fool for even opening these engines up, or doing anything to them. You guys will figure it out. I'll just get in your way and we'll argue in the mean time.

If by asking if the pressure drops in your engine are acceptable due to the temps limiting their scope and if the drops are due to the pickup being uncovered or some other cause are, in your view, argumentative, or meant to incite an argument� that�s certainly not my intent. You are the most experienced and respected engine builder for M96�s - in a forum where asking questions in an attempt to learn is pretty normal - who would be a better person to ask ? I�m definitely going to keep sharing the squiggly lines ( actually, sending them to Paul for hopefully him to continue his awesome analysis ), and the videos for context of my setup - it takes my opinion and pre conceived notions out of the equation and who knows, maybe if some other people generate some squiggly lines of their own - we could have some interesting comparative data, for me - it�s fun. As always, anyone should feel free to contribute, or not to the topic at hand.

To be honest with anybody here, if I would have read half of this thread and everything else in this forum I would have stuck with vws. It is honestly scary to read all of this. Hopefully 10 years down the road I will be able to tell you guys my engine has 175,000 miles and it hasn�t blown up,

Great thread! I�m way more knowledgeable about this topic than when it started. Hopefully JR continues, as well as the others, to keep adding to the conversation. zbomb, nice job keeping it on point with all the personalities and egos in here.!😎

Quotes source post: 17586622

After 20 years of bull**** sump mods, fancy oil, and Accusumps, why would any M96 owner take someone's word for it on anything? With a now 20 year long record of failure, it's not much of stretch to call the claims made by these products fraud, and their continued sale theft. Chris Cervelli Cervelli Technical Service

Quotes source post: 17586973

Add to that the IPD plenum, The kN intake . Now I do feel different about skip stuff. My engine was ticking and a video was posted here. Most people said bore scoring. Skip said no that�s probably carbon deposits or the cats. Anyway after using CRC GDI. Ticking is gone. Skip has a world of knowledge and I can really see how this sump is going to be for real.

Welp i think most know that the deep sump and good oil didnt help my motor from blowing up on street tires. That being said ill have a used 2 qt pan for sale soon.

I was watching videos reading a lot on the singer DLS. Anyway the cooling system is pretty awesome, high capacity for a 4.0 air cool that is putting over 500 hp and revs to 9,000. If it was designing I would do something similar to that square box in the front bumper with 4, 5 fans, off course the bumper would have to be modified. I�m just convinced by what I saw on my car with the clean rads that if you lower the temp, car runs cooler, you�ll have better oil pressure.

Quotes source post: 17587411

Thanks for that, glad your ticking was due to carbon build-up on top of the pistons, I have seen this misdiagnosed before as bore-scoring/piston slap, the sound is the same... For 40 years I have led a dual life, from 8-5 as a Top Tech in dealerships, doing things by the book, the way the manufacture wants it/expects it. Making a good living,raising a family ect. Then from 5-till ? , building race cars/hot rods from the ground up, modifying everything , throwing the book away, doing things that have never been done before, setting records, creating/inventing in all types of Motorsports, WaterSports, Motorcycles, Dirt Bikes, just as a passion and not for the money. That's why I have a good understanding of anything mechanical/electronic. My latest Inventions (the UOAS and UIDS v2.2) actually came easy as Porsche missed the target by a mile.( as they did with the IMSB) But if you bought a 996 just to have a Porsche or the $$$$, or to look at, then you probably don't care..

Folks, No need to start jumping from the highest building just because a large number of products being sold are not as effective as advertised. Intuitively we all know this. ​​​​​​Hence this thread from Zbomb to measure the effects of various bolt on parts as it pertains to oil pressure stability which also includes oil starvation and cooling. Let's not lose the fact that we are talking about a 20 year old production street car here with the gen 1 and 2 996. There is NOT a production car known to man that doesn't need modifications if you are going to facilitate more longevity in engine life on and off the track. Certainly if it's 20 years old. So what have we learned so far... 1. CTS and Hartech address the oil stability issue by creating their versions of a dry sump as well as more effective cooling. Chris and Baz please correct if wrong. 2. FSI addresses the issue with oil cooling not with a dry sump. Jake please correct if wrong. Everyone of us can call up Chris tomorrow for his dry sump and cooling mods, and if you are in the UK, you can call Baz tomorrow and have Baz rebuild you engine of which both Chris and Baz detailed at great lengths the what, how and why they do it on this thread. The same can be said for going to FSI and having your engine rebuilt using their oil cooling approach, and Jake has provided countless videos, and narrative on what, how and why of what he does. BUT, what if you just want to improve the current oil starvation, cooling and ultimately oil pressure stability of you engine before going all the way with a rebuild? There are plenty of viable solutions that are available and can only help your engine, with many more on the way. They may not be the "everything solution" as is a comprehensive rebuild, but they are incremental and can be measured as to their effectiveness. 1. Water Cooling 1: add 2 or even a third CSF radiators. 2. Water Cooling 2: convert your front end to a turbo front end and add both turbo radiators or even a 3rd. 3. Oil Starvation & Oil pressure Stability: UIDS coming soon, pending testing by skip (Porschetech3). Can't wait. 4. Oil Cooling: LN Engineering Oil Cooler coming soon pending fabrication of attachment pieces. Can't wait. All these items can only help. Let's regroup and continue to measure their success. I will continue to sound like a broken record, but we can very easily begin measuring the benefits of CSF side radiators or a 3rd radiator anytime the group is ready... NOTE: To CTS, Hartech, FSI, LN and Porschetech, if any errors were made in the descriptions above, please set us straight. The last thing we want is inaccurate descriptions. Jason, big thanks to you and everyone involved in this thread, it has been incredibly helpful, let's keep it going with identifying and measuring solutions.

One item I missed is the dual scavenge pump. Both LN and Martin Hill have developed their version of the X51 pump. This may also be an incremental solution to oil starvation and oil stability, especially in combination with the UIDS. This is something we should discuss and ultimately test.

Quotes source post: 17587629

When mine comes off the car... I'm chucking it in the woods.

Quotes source post: 17589263

and that was a $1500 improvement that didn�t do Didly squat. Hopefully the manufacturers of all these after market parts are reading this stuff and voluntary withdraw their good for nothing products.

Quotes source post: 17591093

$1,120.37 Sump and Skid plate

Finally! Here are updated graphs from post 466 along with updated comments. Thanks to Chris for running quality checks on the work. I've also added an EDIT comment to that post to hopefully redirect future readers here. Here is the overview graph. In the middle are the temp traces. The ORANGE trace is the oil tank temp. The RED is the engine temp. The BLUE is the water temp. The oil tank temp sensor is NOT calibrated, please do not draw conclusion about the differences between the oil temps. Observations we can make with the additional data are the same as previous. All temps follow engine load. Engine oil temp is only 16 degrees hotter than the water temp at high engine load. During the 'yellow flag', or lower power section, the difference reduces to almost zero. This chart changes what was previous concluded that the oil to water heat exchanger was the weak link. This conclusion was not accurate. Based on this new graph, I'm not sure how to judge it. Here's the new commentary: The car is brought to pit for driver change or other. No air is moving through the radiators. Water temp starts to rise as expected. Engine oil temp starts to rise. Water temp rises faster than the oil temp. The rest stop isn't long enough for the oil temp to get to the water temp. I'm puzzled over the tank temperature though. Chris mentioned that it is for diagnostics for tank level, so I'm not going to guess at what's going on with it. So, thanks for reading a bit of a wasted post. Maybe the thing I need to go look at again are the relative deltas for significantly different load conditions on different temperature days. This would better show the oil coolers effect. The problem with Jason's session data is there is almost no equilibrium. Every session shows temps getting to equilibrium at the end of the run. I'll try. Chris, can you send the data from the higher temperature day? It would be interesting to see the delta temps during high load and lower load conditions. The issue with that data is it is not the same engine. So the relative cooling capacities of the rads and oil/water exchanger are likely different. Also, let me know if you would like me to send graphs to you before posting for quality review.

Quotes source post: 17583698

^^^^ This is a really good reason to not get rid of the deep sump for session driving. If it adds a couple of laps with slightly lower oil temps, we should take it. Maybe to help cool the oil after the session, we need to add fans in the pit to blow air on the sump. One could even make a finned heatsink to press on the bottom of the sump to pull heat out as the air goes through the fins. The problem with this is you're dragging more stuff to the track!

Quotes source post: 17592029

UIDS only small increase over stock capacity - when it�s tested, then we can see how much the extra oil helps prolong the time it takes for the oil to reach higher temps quantitatively.

Quotes source post: 17592035

Totally agreed that if we can keep the pickup full of oil the oil temp problem completely changes. The oil temp issue then turns to how hot is too hot. Less than 230 coolant and 260 oil are the current numbers. I think we had one session where oil temps got close to this. In this case, a 3rd rad would be all that was needed. So, just like my management pressures me for better yields and more margin: UIDS, UIDS, UIDS... no pressure.

Yeah, if the main function of the deep sump is in fact, a method of keeping oil temps controlled - it really begs the question: is it the best part for the job ?

Paul, nice work. No question the extra oil from a deep sump, better aos, Ims retro, constant oil changes and getting off the track early with a cool down lap around the paddock had helped to prolong engine life for me. Didnt know it at the time, nor did I pay much attention to it like we are now. While they frustratingly may not solve the oil stability issue, I gotta believe that they helped to prolong engine life or my engine would have gone a long time ago running Hoosiers, in the fastest run group, without enhanced water or oil cooling. Just Oem radiators. Go figure. Never thought to look at the gauges while on track because I was driving, but never had any lights flashing on the dash either. After my cool down laps I kept the car running for about 10 min and never had any issues with over heating. Our cars, if well taken car of, are more durable than we think for street and DE. Any solutions we find on this thread regarding oil starvation, oil stability and cooling, will only further help our cause. Let's not lose sight that we should continue to find these solutions. UIDS - Potential solution for oil starvation and oil stability LN Oil Cooler - Potential solution for oil cooling beyond Oem. CSF Radiators - two or 3 provide solution for more water cooling. Same can be said for turbo radiators which will provide even more cooling but will require changing the front end to a turbo. CTS Dry Sump Kit and Cooling - comprehensive fabrication to address all issues without rebuild. Rebuild - to address issues comprehensively or partially depending on your objectives from FSI, LN RND, Hartech I sound like a broken record, but there are solution here and now and in the works. None are compressive like a rebuild, or fabrication, but all are incremental. This thread while painful to read from time to time, is incredibly important to push innovation and application in keeping us all on the road. Let's keep pushing, innovating and identifying applications to solutions.

Found this thread from 2005 with another link to a 2003 (I believe) oil pressure thread. Interesting read and 16+ year old attitude & opinions. https://rennlist.com/forums/996-foru...the-track.html

What I am really liking is that Skip/George are going to give us the option for an extra 2 Qts. There is something to be said for more oil. The UAOS handles the extra oil with ease. If I am not mistaken, for most events I ran the oil level at 3/4 with the extra 2 Qt deep sump. The logic was the more oil the better, but don't push your luck with full oil to push the motorsports aos to its limit. From a length of time on the track, I only did a few back to back 30 min sessions at Road America which totaled 45-50 min in continuous track time to dial in something on the car. Even in that scenario, didn't have any engine lights or overheating. So something was preventing failure while on Hoosiers. Certainly a better AOS helped, but more oil didn't hurt either. Even though I didn't measure oil pressure, nor even looked at it, I am certain it had to have dropped. But not enough to cause engine failure. As a side note, we are talking about 2-3 days straight of tracking at Road America for 2 PCA Chicago region events a year. Not a one day and done deal and then go home and wax your car.

BUT, let me make this perfectly clear to everyone. What I did to my car over the years in prep for DE worked. But it's not the end-all, especially since our engines are now 20 years old. We still need to find more solutions for oil starvation, more oil stability, and better cooling. Keep pushing.

I would say it has worked so far. U r still damaging the car with every de day. It will eventually pop just like mine but thats the risk we take. This is one reason i decided to abandon the m96 platform. Had a solution been easily available for building the m96 that would cure the pressure drops i would have just kept with it. The ecconomics r just not there to rebuild the m96 and to keep tracking it. With any luck an available solution will be produced and many more can enjoy the track without worry.

Quotes source post: 17592516

I love what you did with you car and can't wait to see it in action. One of these days we will be at the same track having some fun. Agreed. Doesn't matter whether it's a mezger engine or m96, the second you start it, you are damaging it. And they all will eventually need a rebuild. And they all cost lots of money. Hence my interest in this thread. I have the unique advantage of owning and servicing my car since 2002. Which means that I didn't inherit the baggage from a number of previous owners, which I do believe is a big cause of the problems we all talk about on RL. As an example, I was going thru a bunch of records and counted that I have had over 70 oil changes since 2002. Still have one more box of 996 stuff to go thru and I suspect I have had another 10-20 or so. Think about that. Most folks have changed it 8 times over the past 20 years, every 15k miles. My car has 57k miles. Just sent in my oil sample which took forever because I had to put 1000 miles on it which I never do between oil changes. Now I am waiting for the analysis to start a baseline. Engine is strong, no visual or audio signs of bore scoring. Regardless of the test results my next step is a bore scope. If everything checks out, then I will be adding CSF radiators, UIDS and OIL cooler when they become available in prep for more DE. If the results are less favorable, then I have to prepare for a rebuild, of which this thread has been invaluable in understanding strengths and weaknesses of the engine as well as ways to address them with bolt-on parts to prolong engine life before the inevitable rebuild. Gotta thank Jason because I never got involved with RL until somehow I saw his original thread on building out his car and it got me interested because I was going thru the enhancement process for the second time. One thing is certain, these cars are awesome, and I am not going to stop tracking the car. While I have taken the last 4 years off, most of it has been spent understanding what is new surrounding our cars. But more importantly, I feel pretty good about the 996 renaissance and new solutions that are coming down the pike. He'll, this thread alone is probably going to be responsible for a number of new solutions that we haven't even heard of yet.

If anyone is bored this weekend I will be driving in the WRL race at HPR on both days. The races are 8 hours each day. The car will have live video streaming. I don't know the details yet but I think the stream will be found on Youtube. The car is a 2.5 Boxster, #211, and we'll be running in GP2. The team name is Racher Rennstall 2. Based on practice today, I don't expect much of a battle for the GP2 win, so we'll be trying to place as high as possible overall. I will have two co drivers, both of whom are very experienced in Boxsters. I am hoping the live stream will serve to highlight the intensity these races are run at. Chris Cervelli

Quotes source post: 17592841

I'm in... Is the car running a dry sump ? Can you share any additional detail on how the car is setup for both water and oil cooling ? Can you share some data after the race ? Looks like 88 tomorrow and 95 Sunday ! Should keep the cooling system honest. Best of luck.

Quotes source post: 17592396

Brings back old memories! We were all flying blind back then.

Quotes source post: 17592841

​​​​​​Good luck Chris! If you end up finding the link please forward it!

Is this the YouTube feed? You Tube

Quotes source post: 17592858

This car is still wet sump. The current plan is to dry sump it prior to the 24 hour Sebring race this fall. The 2.5 does reasonably well without a dry sump and HPR does not show the pressure drops as much as some other tracks. I have the data from today and it is not terrible. Still a few drops here and there. The cooling system is stock with 3 radiators. It does have some fancy duct work done to all three radiators. It is a small help but worth doing for conditions like we'll have this weekend. I think this engine runs the stock 2.5 oil cooler which is half the size of the 996 cooler. Temps today were 212 coolant and 242 oil in 95 degree ambient. Keep in mind that this engine is at least 70-80hp down on a 3.6. I should be able to share the data after the race. Searching WRL on Youtube should find us. You can also follow timing and scoring on WRL's website. Chris Cervelli

Quotes source post: 17592887

Yes that's us.

Quotes source posts: 17592887, 17592080

^^^ I'm in. For some of it anyway. Good luck!!! Definitely NOT the best part for the job. For oil cooling, I could imagine going back to the regular sump, Adding a set of cooling fins as deep as the deep sump is, adding some air ducts to make sure air runs across the fins, would do a better job. But, the fins are at risk of smashing on the ground. Have you ever bounced on your skid plate? Of course this is a ridiculous idea likely attempted by someone 15 years ago...

If possible I will update on the status of the race tomorrow in a new thread. Chris Cervelli

I drove my 99 today, and it�s hot out (90F). I experienced a slight drop in pressure on a long on-ramp right-hander (from 5-bar to 4 bar). I keep my oil level about 1/2-way up the dipstick, and I may have worn tensioners contributing to the issue. And I was just street driving� ​​

TESTED MY OIL Looks like we got quite a bit of inspiration over the weekend with Chris and team's two 8 hour stints of racing the boxster. What it showed all of us is that our "wet-sump" water cooled engines are more durable than we may believe. My car is a good example. After talking to Charles (LN) a few months back, he recommended I test my oil. After putting on 1k miles on fresh oil, I had it tested last week using Speediagnostix for the first time and just got back my results. The good news is that the test shows that the engine is in good shape and everything came back in the GREEN. Lake was helpful to talk to and the process was easy as he'll. I'll post a copy of the analysis when I get in front of my computer later tonight. You bet I am relieved. RL threads scare the he'll out of all of us. Didn't see or hear anything that would suggest that I had a problem, but wanted to know just in case. My car has 57k miles on it. I drive it around town as well as actively DE'd the car from 2004-2018 of which the last 7-10 years I ran in the fastest run group on Hoosiers. The only oil starvation/oil stability items I had bolted on where (1) a motorsports aos in 2006, way before the UAOS was in existence, (2) 2Qt deep sump kit in 2006, (3) IMS Retrofit in 2011 or so. ​​​​​All of these items certainly helped to prolong engine durability. But I do believe that maniacal oil changes have been a deciding factor in the positive oil test and my engine durability. My rough calculations for oil changes come out to approximately 5 a year thru 2018. Pre-season after the car came out of storage, and after every track event. It calculates out to an oil change every 500-600 miles. I suspect that none of the parts listed above did anything to solve oil pressure stability one bit. But I do believe these parts and constant oil changes did help to address oil starvation, oil cooling and overall just good lubrication, which we all know is the lifeblood of our engines. IMHO, the addition of more cooling with new radiators and LN's oil cooler that is in development, coupled with Skip's UIDS will be an added bonus towards further engine longevity. And maybe a partial step towards oil pressure stability.

Not solving the pressure drops, but possibly further mitigating them: the comment I'm interested in pursuing was from the WRL thread. Chris mentions that he runs his engine oil level overfull. Maybe up to 2 quarts on an engine with an AOS open to air. With the UAOS Track Drain Back, it seems we should be able to do the same thing and not lose any oil. It also has a built in 'overfull' detector. If you put too much in and it can't drain back fast enough, the original can with the detector will fill up and the coolant light(?) will blink. Take out 1/8 quart and try again. It would find the max overfill level for any individual engine, given the variance in blow-by. Stupid idea? This could decrease those pressure drops a bit more. Add a few more laps of life...

Quotes source post: 17600174

Not a stupid idea at all.. With the Extreme Duty Track Drain, the oil level can be higher due to the dual drains. It can return oil much faster than a single drain so overwhelming would be less possible. Jason has been slowly upping the oil level on his car each outing after the Track Drain has been installed. I'm thinking a quart overfull will be possible. I do caution about going about this too fast, the warning light has a Factory Delay in it of 19 seconds. If the coolant becomes low, or the oil accumulation becomes high , either sensor has to remain tripped for 19 seconds before the warning will flash. This is done by the Factory to keep from sloshing set off the alarm early/falsely. Yes, the Porsche Factory had everyone lowering the oil level ( and on the 996.2 the dipstick/dash level is even lower) due to inadequacies of the Factory AOS system, but with the UAOS, we are going the other way !!!

A few words about deep sump. The deep sump is a simple and effective way to keep the pick-up submerged, if there is any oil in the sump that is. Running oil level low, as I have pointed out several times, will do no good for the pressure issue. Zbomb, I read you are running an oil feed IMS. That might explain why you haven’t had the same success with it as others. You are effectively pumping oil away from the sump, to the internal space, with that. It’s the opposite of CTS’s head restrictors. That might explain why you have oil coming up the AOS also. Regarding the AOS (and as Baz said just before he was banned), I haven’t either had any issues with the original AOS. I’ve cleaned the port on the back side of the throttle valve and check it with my finger from time to time, it’s still dry (= no oil film). I always drive with full oil. Granted, I don’t track my car but I have some fine roads to push it on(legaly and safe). A small feedback from my test with 0w-30 oil(12cts @ 100c….so just a tad thinner than M1 0w40), Ravenol VSW. The oil pressure is pretty much the same as M1 0w40 except when the oil starts to get really hot, then you can see a noticeable lower pressure. Pressure is never an issue except idling when the oil is really hot, the no-pressure light comes on. A light foot on the gas to bring the idle up 100rpm solves that. It’s bit of a pain in the *** so I will probably go back to M1 0w-40. Second issue is that the oil seals around the variocam solenoids are “sweating” . Not to the point of dripping but it doesn’t look nice, it used to be dry. Just for info: NEVER use an oil that have seal swellers(some high milage oils have that) in this engine, there is a risk that you kill the IMS bearing seals with it.

Nick - FWIW, when I was still beating the **** out of the car on the road - I NEVER had an issue with the AOS, no smoke, no consumption, no nothing. That is something that has ONLY ever occurred during track conditions. And ETA... Until I see data against a stock sump - I am not willing to say a deep sump does anything.

Paul, Skip and Nick, I like where you are going with this. We can measure the effects of oil levels. FWIW, think I ran 2 qt deep sump at 3/4 full pretty consistently, but will check with my shop on their standard operating procedure. I will say that the Oem aos is a weak link. I experienced exactly the same thing Jason did with it failing on the track in 2006, which was my 2nd year of DE when the aos was about 4 years old. That's when my shop put on the motorsports aos and added the 2Qt deep sump. Hope this info helps as you guys think thru the issues.

Quotes source post: 17600332

Right, the Factory AOS should cause no smoking or oil consumption on the street as long as it hasn't failed. But they fail so often that all the top notch shops started recommending it be replaced every 3 years.(UAOS solves this) The smoking at PCA events on DE track days is where the this takes place, and it was from Porsche to lower the oil level to half on the level on 996.1. Then on 996.2 mid year 2002 they changed the dipstick and re-calibrated the dash level to be half quart lower. This was all due to the AOS couldn't handle the oil and drain it back fast enough. The Motorsport AOS was a Porsche attempt to solve this problem on the 996.2, but instead of drain back more oil faster, it held more oil capacity and effectively lowered the level.

Quotes source post: 17600381

Skip, A couple of quick questions. Assuming you are running the UAOS or the Motorsports AOS, as well as a 2QT deep sump, what is the drawback to running full oil or a quart extra of oil on the track? Any damage that can occur? It sounds like the UAOS/MSAOS handles the extra oil without a problem or am I interpreting this incorrectly? In addition, what if you add the UIDS into the above equation? Appreciate your insight. ​​​​​​

From 8/16 & 7/29 sessions: We may be at parity with oil level... Instead of an overview of the whole session, I'm just starting with session 3 lap 6. Same layout as previous. Each session I try to fit the RPM to the pressure and each time I come up with what I think is a better fit. Then I use this fit on the other session in comparison. Looking at the next chart, the fit seems to be not so good. I believe this difference is due to the overall pressure level, which I believe is defined by the temperature of the day. The primary change to the car for this session is oil level full on the dipstick. One tick down on the dash gauge. The 7/29 session was 3/4 full Oil temp: 240-242 ish Water temp 199 ish Lap time: 1:21.9 Ambient air temp: 95 ish (last graph) Here is the PB from last time: Using RPM as a reference set by the current session OP level, The OP on this session is higher Oil temp: 230-235 Water temp: 193-196 Lap time: 1:20.8 Ambient air temp: 90 ish (last graph) And on the overlay graph we see that the 8/16 pressures are indeed lower than the 7/29 pressures. RPM's are nearly the same. Drops are slightly bigger. Orange & dk.blue is the current lap, Red & lt.blue is the PB. Braking is noticeably faster this session (dk.blue) and this creates some early pressure drops than in the last lap. Look to 5700ft, the pressure drop follows the harder braking before the left turn sooner than the previous lap. The ambient temp was higher, the oil & water temps were higher and pressure changes were nearly a mimic of the PB lap with a slight decrease. I think we are at parity with this oil and the level. Need more cooling! And maybe a heavier oil. Very consistent laps. Nice driving! ETA: The interesting point about braking is that it only seems to preclude oil pressure loss when going into a left turn. The same thing does not happen when going into a right turn. At 5700ft, the harder braking into the left turn starts the pressure drop. At 4550ft, the earlier braking into the right hand turn does not start the pressure drop. The pressure drops mainly from the right turn G's. We've talked about this before, but that was some time ago and this graph has some clear points to show it.

Quotes source post: 17614607

By oil level parity do you mean the point of diminishing returns on raising the oil level? Will AIM allow you to create an on/off math channel? If so, create this channel and overlay on the graph. Channel true when: Throttle >75% and RPM > 4500rpm That should have some predictive power about when the engine failure is likely to occur. Chris

Quotes source post: 17614989

Yes, in this session adding 1/4 dipstick range (I believe 1/4qt) did not decrease the pressure drops. The entire pressure level dropped with the drop outs slightly increasing in size from the last cooler session. Either the 8-10 degree increase in the oil temp reduced the viscosity enough to leave the pickup uncovered longer, or the additional oil is lost moving through the UOAS circuit due to blow by. The problem with my parity statement is the temperature change. I assumed the 8-10 degree difference wouldn't decrease the viscosity very much. Maybe wrong. Yes, the program has a function builder. The top row is RPM with the binary function set to 4500. The bottom row is the throttle position with the binary set to 75%. In the middle is oil pressure with the AND function showing engine risk zones. Assuming the next step is to evaluate the pressure levels in the risk zone, we can add additional binary functions to check for OP at certain levels within the engine risk zones. Each row below the 1st row analyzes whether the OP within a risk zone is below a certain level. The second row is 45psi. There is no time in the risk zone below 45psi. The 3rd row is 50psi. There is one short time. The 4th is 55 and the 5th is 60psi. It seems that if we are above 45psi, everything is OK. What is minimum pressure expected in the risk zone? And for giggles, here's the entire session with the risk zones and pressure analyses. There is no time where the OP is lower than 45 within a risk zone. 2 occurrences at 50. and more above that. I guess I'm not worried...

Interesting...

Paul - what you can get out of the SW is amazing. Really cool idea Chris and an interesting way to look at the data from a new perspective. I’m going to keep moving the oil level up and see if we can get a little more solidification of the results. Next track day on the 31st.

This is all trying to tell us something but we just don't know what we need to know to go any further. Please read and understand: http://www.epi-eng.com/piston_engine...e_bearings.htm We don't know what the oil pressure drops really mean with regard to air entrapment in the oil at the main bearings. Hence: We don't know the actual effective viscosity at the main bearing. We don't know where we are on the BOC curve during normal operation. We don't know the magnitude of the bearing load. I think I have data which proves that under racing conditions with Mobil 1 0W-40 oil the engine operates near the 35 point on the BOC curve and is therefore in danger of sliding up into the danger zone of mixed lubrication even without pressure drops, but I cannot be sure. If Zbomb's engine supplied us with it's BOC data I suspect we would see that it spends very little time with a BOC < 35, because most of the oil pressure loss occurs when the bearings loads are very low, ie during braking. I also suspect that sooner or later he is going to to encounter conditions which cause the engine to accumulate some time with BOC < 35, and the engine will fail. I think Baz's Accusump and scavenge pump install does reduce the amount of time spent at BOC < 35 by shortening the durations of the pressure drops. Mr Raby's cool and thick oil (among other things) pushes the the BOC to the right, away from the mixed lubrication area, at a cost of additional friction losses. Chris

Quotes source post: 17611967

Glen , sorry I missed this post, I have been concentrating on the UIDS V2.2 ( George is done with his part and is shipping it to me), and have only been jumping on RL spotingly.. I would not try to run the Motorsport AOS "overfilled". It still uses the single Factory drain hole, so it can't drain oil back to the pan/sump any faster than the Factory AOS, it only helps by "holding" more oil in its body, that can drain back over time when not in a high g-force turn. And it has no warning devise. The UAOS with hm/XTDD has "two" drains (plus extra capacity ) to drain back more oil MUCH faster and therefore can be ran at full or overfilled( by how much still TBD). The Factory Integrated Dry Sump and all existing single central pick-up sumps with baffle actually divide the sump into three sections of oil. No matter which way you turn 1 section of oil cannot be used by the central pick-up. The UIDS v2.2 doesn't divide the sump into sections, therefore can utilize ALL the oil in the sump, giving it an advantage of approximately 33% more efficiency. With a 33% more efficiency from the UIDS v2.2 , running a full oil level or overfilled will be much less important. Stamped oil screens, still needs a frame bottom covers done... stamp to make screens for pick-ups

If it works as good as it looks... we'll be made in the shade.

Quotes source post: 17617727

It will, or I'll " eat my hat " ....

Quotes source post: 17617632

Skip, No worries. Glad you guys are focused on the next steps with the UIDS. Appreciate the additional thoughts on Motorsports AOS in relation to running over filled. Kinda what I had thought but you confirmed. It did its job for my car when the UAOS wasn't yet a product. But I will be replacing it with the UAOS hm/xtdd and adding the UIDS for next season. You have built a better and more flexible solution. The concept Chris introduced of being able to run the oil "over filled" and the UAOS hm/xtdd being able to handle it, is an attractive topic but yet foreign to what we all have been taught by the Porsche world. Looking forward to hearing more about it, measuring it and understanding how to apply it. Keep pushing!

Quotes source post: 17617632

These pictures are pure M96/97 pornography, Sir. 👍😊

Revisiting some old data and applying the 'Engine Risk' model to show how much better off this engine is now than when we started: Here's the hottest lap from Palmer back in May. The ambient temp was high, oil temp was 258-260 making the condition worse. G loads were lower than current making the condition better. The engine did not have the drain back yet. The G's & Oil temp are at the bottom for reference The lowest pressure during an engine risk period is between 25 and 30psi. The latest run was between 45 and 50. That seems like significant improvement. How much of the difference is due to temperature only? Oil level? The only way to know is to make a bunch of runs at different temps with all the other conditions the same. Not easy. From the article, the specific condition from one engine to the next is easily seen. 2 engines with the same external prep, like this engine has, may still have quite different outcomes based on specific internal difference. The key, as mentioned previously, many times, is to keep the entrained air out of the oil. Oil pressure is being used as a proxy for entrained air. Higher pressure indicates there are fewer air bubbles or they are further apart. We could also be at the mercy of the oil pressure sensor. How much entrained air can move past the sensor before a pressure drop is seen? Will that amount of air damage bearings? To 'know' we would need some type of new sensor. A through window section in the main flow path with a light on one side and a sensor on the other. The light would be disrupted by air bubbles passing through the stream. Interesting to think about but impractical to implement and correlate to bearing damage without the same amount of trial and error. I'll stop thinking about this for a moment and go for a drive!

Paul, So if I hear you, oil pressure troughs improved? Whether it was from more oil or something else, oil pressure stability improved?

I want to believe they improved. Certainly pressure stability on 8/16 was much better than on 5/22, using the engine risk filter at 75%throttle and 4500rpm. The problem with all the data is the crosstalk. -Viscosity changes with engine temp. Higher temps, Larger instability. ( believed to be less time covering the pick-up ) -Higher G forces create higher instability, same issue In these 2 sessions, 1 factor contributes to instability and 1 contributes to better stability. Which of these has a bigger effect? Do they cancel out? The 'engine risk' mask (or model) is a simplified version of using the Bearing Operating Condition to define when good pressure is needed. The problem using the mask to represent BOC is that BOC uses the oil viscosity in the equation, which we know is affected by temperature. Because of this, the engine risk mask needs to change from run to run based on oil temp in order to be able to make conclusive statements. Because we don't know how the viscosity changes with temp, we can't calculate BOC directly. So I will make the bold statement: I want to believe it is better! Chris gets veto power if my interpretation of BOC is wrong. We need more trials! And the issue with more trials is that as Jason gets faster, the G loads will get higher. So every dataset will have some variance that can be argued over. I'm pretty sure we will not be able to convince him to run the next session at Thompson keeping G's at 1 to compare to older data. But, over time, even with imperfect masks and improving driver skill, eventually we should be able to make a stronger statement. Maybe just before the engine gives up...

Jason and Skip, Seems to be good news with what Paul's analysis may be showing. What are you guys thinking about oil levels for Jason's 8/31 outing?

@Jason. I’ve been reading up on aeration a and foaming. Increasing oil level may just be placing the oil intake below the oil foam and reducing the amount of air resulting into in loss of oil pressure. Therefore, one variable you are not testing is foaming, or level of anti-foaming agents. What oil are you using, Gibbs XP with a noted Si level of “6”. Would you be willing to try a different racing oil at the same viscosity, say Motul 300V, with a different level of anti-foaming agent? Another variable I found uncontrolled in these runs is bearing clearances, which determines oil viscosity. For 0.0015” main bearing clearance, I found 5W-40 is optimum across the oil temperature ranges observed.

Quotes source post: 17621526

Jason will be getting the "54" dipstick (996.1) and running it at the "full" mark, which is 1/2qt "overfull" according to the 996.2 "55" dipstick and dash gauge. This will be a good test for the UAOS with hm/XTDD ..

Quotes source post: 17621809

Never thought I would be this excited to see the results of running an engine 1/2 quart over. My wife tells me I need to expand my interests beyond oil starvation and stability... This is like being a kid again waiting for Christmas. 😁

Quotes source posts: 17621201, 17621526, 17621809, 17621726

Paul - your analysis has added so much - I wish I could get you data that would be more suitable for the high level analysis you can levy on it - but... I want to go faster and learn be a better driver... so you'll continue to get my imperfect data I hope you'll keep doing your thing with it. The Consiglieri has spoken. Running XP9 and I do not intend to change oils before the end of the season. Based on what I can see - clean filters, sump and oil the smells good and looks clean after my usage - I am not going to wander into the oil testing business.

You need to know film thickness necessary to get to full film lubrication to get the oil viscosity correct. As you see from the chart below a sae 40 oil can go as low as 12.51as high as 16.29 at 100. Anyway also the lubricant vi ( viscosity index) will give you an idea how the viscosity behaves in relationship to temp. A final comment I like the fact the you guys are testing running full oil or a little over. I personally run full On all my cars. Jason I have seen almost all your driving videos you are getting better, just keep at it.

Quotes source post: 17623410

The recommendation to run the oil level at half the min-max has been happening for about 20 years. The Official word came from Porsche due to customer complaints of smoking on turns during PCA DE events to lower the oil level., then a Porsche Technical Service Bulletin was issued and a new dipstick and dash calibration. ...... TSB 1/02 1716 Blue Smoke from the Engine During High Transverse Acceleration -- dated May 10, 2002 This was due to a POS AOS design., and has been going on for 20 years......

Quotes source post: 17623533

Tech I was told by Porsche Germany long time ago that the recommendation wanted to take care of the people that would overfill, so kiss keep it simple stupid, that was while getting A40 approval for a Chilean blended synthetic oil. Off course they collected the fee. Funny thing, in some markets countries it was good enough to write in the label meet the performance level of Whatever spec. In other markets, specially since Mobil was pretty strong, you have to have copies of the original Porsche approval letter. If not you were not going to sell a single quart.

3/98 Porsche customers will definitely overfill the oil, then drain too much out, then overfill, then drain too much out, then overfill , then drain too much out. Kind of like when I try to make my biscuits and gravy run out at the same time, I have too much biscuits left over or too muck gravy left over, can't seem to make them run out at the same time. . I know you can relate to that...

Zbomb, can you do a short summary what you have done to the oiling system since the beginning of this thread?

Quotes source post: 17623533

Good thing your UAOS eliminates the need for this recommendation.

Quotes source post: 17624001

Yes thank you. I firmly believe that following that recommendation has caused a lot of bearing damage over the years. Even though Porsche says the oiling system was good as long as tires were "N' spec, we know from experience that certain engines would experience oil pressure drops even on freeway ramps on "N" spec (or lesser) tires. Of course it is presumed that there are internal issues that are a contributor to this, they have never been identified.. The goal in testing the limits of the UAOS with hm/XTDD and in the developement UIDS v2.2 is that through better design sub-systems, the issue will be completely resolved no matter the internal inadequacies.

Quotes source post: 17624799

Amen brother.

Alright, here we go... place your bets... The current engine risk model is based on bearings being at risk when the engine is over 4500rpm and throttle is greater than 75%(engine load). The last session: full on the old dipstick, 240 engine oil temp This session: full on the new dipstick = 1/2 qt over on the old dipstick, 248 engine oil temp Last session showed the lowest pressure seen during the engine risk periods was 50psi. This session, the lowest pressure seen during the engine risk period is 60psi. The drop occurs at exactly the same location as the last session. Other than oil level and oil temp, the other primary difference to check is G's. This run, tires were showing their age and not able to hold as well, so we were concerned with a difference in lateral acceleration. Here is the current lap with the previous session lap overlaid. Red is the current lap. Lateral acceleration is the same... Ignoring the engine risk mask and looking at the overall trace, the red trace(current lap) shows much better oil pressure than the green trace(previous) so even if the engine is not at risk because RPM or load is low, the pressure is still better. All good. It seems clear that the increased oil quantity pushed the oil pressure up. More oil, more oil pickup coverage. Even though we think we may be happy, this oil level on a 260 degree oil temp day will show lower oil pressure. It will be much better than the oil level at 1/2 on the old dipstick. What we don't know remains the same, what min pressure is OK under load. So we need to keep pushing the basics, lower the water temp, lower the oil temp. And now with the UAOS doing a lot of work moving oil around, understand where it's limit is. I expect this will be quite different with each engine. Need to be careful with this, as the good Dr. has pointed out. 3rd rad, better oil cooler, instrument UAOS to understand how it's loaded. Data is king!

While not a perfect solution, the increased oil level does seem to show a measurable improvement based on this sample, I’m going to confirm the oil consumption tonight after letting a good drain back to the sump occur and if all looks good on that front, I plan to run at the same level, or slightly higher for Lime Rock next Tuesday. There was no smoking or otherwise noticeable detriments to the elevated oil level. Long live the UAOS drain back.

cross posted in zbomb confused thread. Here's a zoom in on that low pressure point in the table from lap 7. It occurs at low RPM, low engine load, but high G's, in a right turn. And is outside the engine risk mask. We see a drop here on every lap. This is where we should argue about how much damage a bearing can pick up outside the risk zone. Oil pressure does not follow RPM closely, so there must be air in there... Risk, no risk??

Paul - I’m wondering why the dip was so low there compared to the other laps I looked at - laps before and after. The isolation is strange.

Great insight on the data Paul, I may be having a brain fart today, but what is the data marked PPS? And yes, I think there is more positive results available from the UAOS with hm/XTDD, but must be cautious here... I think the outlier in lap 7 can be called a rogue wave..a perfect storm of waves of oil at the surface.

Quotes source post: 17641994

I think it will be worth it to figure this out. The answer is likely in the longitudinal G and/or brake pressure data. It could also be that the avg rpm on the previous straight was higher/different. 3400 rpm in 3rd is too low. You will find a few tenths if you downshift there. Chris

Quotes source post: 17642387

I was thinking the same thing on RPM - I need to go back and see if that�s where I always am at that point. Didn�t get the results I wanted from a time perspective yesterday and even though I felt like I was pushing, it wasn�t reflected in my times (first time that has happened) and the car was not as compliant to me ham fisting my way around the track. I think for an isolated data set, pretty good results but as Paul indicates and I agree, more data, different conditions and as you say - closer interrogation of available data all need to really happen to increase confidence in the result. Maybe even some corroborating data - wouldn�t that be sweet.

Quotes source post: 17642203

PPS is throttle position. I'm not sure why it's labeled this way, maybe pulses per second? Maybe Jason knows. Regardless, it's being used as an indicator for engine load. The lowest pressure turn is the result of the longest turn. The big drop is from turns 7 & 8. Here's the track. Turns 7 & 8 take the longest to get through. Turn 7 is just the warm up taking 5 seconds below 1G. This warm-up moves the oil to one side of the case but doesn't leave the pickup dry. Then turn 8 shows up with G's above 1, the oil moves further and leaves the pickup uncovered for longer than the other turns because of it's starting point. Turn 8 takes 5 seconds. Pressure starts to recover when the car accelerates and the oil moves to the back of the case and the lat G's subside. Now look at Turn 4. The G's are nearly as high, but it doesn't have the total time or the warm-up that Turn 8 has, so the pressure drop is not as severe. This pattern is quite consistent from session to session. Another interesting thing to note is that the pressure deviates from RPM right around 1G every time. With the exception of the weird drops at 1:11 and 1:13. Left turn drops always follow and nearly require braking to create a drop. I think we already knew this. And, just for giggles, here's lap 9 from back in June. Oil temp is the same as the lap above. Just try to tell me things have not improved... So the low pressure question really is, how much air is in the oil, and how severe is the wear if the bearing is not loaded... It's still spinning pretty fast. This issue probably occurs at the moment load is reapplied. If the pressure is low or aeration is high, the load will push the oil aside and do bad things. Chris previously provided the link: http://www.epi-eng.com/piston_engine...e_bearings.htm

Nice work Paul! Data shows quite a bit. In particular it shows the benefit of running more oil over full, to enhance oil pressure stability. It also speaks of the successful functionality of the UOAS w/XTDD to be able to handle the extra oil with ease. Without it, these tests would not be able to be done. The data also provides a baseline to test the potential incremental improvement of (1) using even more oil, (2) more water cooling, (3) more oil cooling when LN has completed their oil cooling set-up, as well as (4) the UIDS once it's ready for Primetime. (5) the final question may be if there is any benefit to running the UIDS with a 2QT extension or not. Jason and team, can't emphasize enough the positive impact that this thread, and all of the auxiliary threads as a result, are having on the the 996 community in finding viable solutions to oil starvation and oil pressure instability.

Quotes source post: 17643647

The turn 7/8 complex is one decreasing radius turn which explains the buildup of G's. The end of 8 has a little tight kink on it and the track transitions from helpful camber to zero camber right at the exit of 8. All this depends on the line through the turn. In my experiences there are as many different lines as there are drivers and all the lines are bad.

That complex, T7/8 is 27 seconds into the video below for some context - thanks theprf for getting the follow view. Chris - in addition to being faster going through there in 2nd vs. 3rd - do you think the additional RPM (it will be singing through there in 2nd, maybe 5.5K RPM) will help to minimize pressure drop ? You Tube

by request, here's an overview of 6/01 and 8/31. Fun. I hope we never thought this was OK. Much improved. Which one of these glasses depicts the state of oil aeration when on the track? The further the oil moves due to G's, the closer you get to picking up the foam.

That’s what I’m talking about Brother ! That to me looks like progress.

So the LN Engineering 2QT Deep Sump Oil Pan Kit does help!

Go spend THIS Which is supposed to do THIS And then post about how in your eyes IT WORKS when it does THIS. You don�t give a **** about people throwing their money away on parts that don�t work. If anyone listens to anything you post here of a technical nature - well, that would really be too bad for them. Let me clue you in on something. You are part of the problem. You are the type of person who has prevented this problem from being solved. By saying something that works when it doesn�t because you don�t require any type of evidence to form your uneducated opinion - if your guy says it works, it works. You have been presented time after time with evidence that your guy is wrong in this instance, yet you persist. You and people like you are why problems persist, and vendors that make products that don�t work keep selling widgets that don�t work.

I didn�t buy the deep sump, YOU did. I don�t post data from a car that has the deep sump installed. YOU do. If you don�t think a deep sump helps, remove it and post the data. I�m sure there�s plenty of people who would like to see that information. I�ve offered to pay for the oil to make that happen. I�d never claim that a product works or doesn�t work without the proof to back it up.

Quotes source post: 17644915

Go buy the stuff you like and test it dude. I’m using my car, and trying to make it better and it seems like some people have found some of what I’m doing useful for them. I’ve been entirely transparent in everything I’ve done to the car, be it engine data, driving data, suspension setup. Had the guy with FSI old car come up to me and ask me why my car is so much faster than his - I sent him my full mod list and suspension setup data, even though I do take enjoyment blowing the doors off an FSI car with Motons or JRZ’s or whatever he said he’s running. Trailering his car to the track - mine is driven and I have installed every part, done every service. Cause I like helping 996 guys, especially those 996 guys that like to shred at the track. If that ain’t enough to satisfy you - too bad for you. If you don’t like what I’m doing, how I’m doing it - come with some data of your own that contradicts mine… YOU WONT. You’ll talk instead. ETA BTW - for what feels like the millionth time - I have not said it doesn’t help. I have, and will continue to say, it doesn’t work.

Quotes source post: 17644915

+1. as a Ph.D. cancer researcher, I�m struggling to see how these �experiments� prove anything. No statistical relevance from this or what others have indicated. Would you take a medication based on results of one or two people? Before/after with relevant data is the usual benchmark.

Quotes source post: 17645483

Dude ! You have a PHD... Why didn't you say so, you've been holding out on us. - you're just the guy we need to improve oil pressure in the M96 ! We're saved ! GTFO here... If you don't like the data - go make your own... Anyone that +1 Philberts... I have serious concerns man... Serious. This is what we put on the frontline of the most important research of our lifetimes.

Quotes source posts: 17644915, 17645483

so your biggest hang up is he wont pull the pan to show that the already poor performance may get worse or stay the same? At this point that baseline point is moot. The oiling sucks with their pan and no where near reflects their bold claims. His improvements on oiling have been made with other midigation efforts. who in a forum board has the resourses for this? Z has repeatedly said that this pan does not work for him based in the data he can provided. That is an accurate statement. And to yor apples and oranges comparison of advertising of medicine and car parts, pharma has to present data and prove it works, but for cars u just have to write a good description and have some nice pics. U just realy on others to prove that it doesent work. I get that they say it "helps" and maybe it does to a certain degree but for the cost and the hype it doesnt do nearly enough. if u need a data point u can add my engine to the chart. ​​​​​​

Quotes source post: 17645483

Now might not be the best time to make an appeal to authority using your own status in the medical industry as that authority. I'm not sayin'. I'm just sayin'.

To me, this is probably THE MOST IMPORTANT thread on the 996 M96 to have ever been posted in the 996 forum. It addresses the heart of the M96 engine (the oiling system), and why many people deny the problem , or don't drive their car to enjoy its full potential, in fear of a failure. To me it's much more important than the "current value threads" of the 996 that seem so popular here at this time, or any other thread that has had a through discussion over the years ,and has improved parts and or processes to address the deficiencies on the 996. e.g. IMSB ( IMSS ) , D-chunk, oval cylinders , scoring ( Nickies) , ect..( thanks to FSI, LN). Zbomb has been totally transparent in his data collection and sharing, and should be appreciated by ANYONE who is interested in this topic. I would even like to see more people posting there data collections to further the "experiments'... If you don't appreciate this thread for what it is and what it's trying to accomplish/improve, ( and is making "real measurable progress" ), then you don't have the M96 best interest in mind, or you have a problem yourself.

Thanks Skip - makes sense that as progress is made towards a solution to this problem, people who failed to solve it previously, and / or their apologists may come out of the woodwork to raise doubts... They won't bring any data and the will rely on who they know, who they are or who they want to be. I don't really give a flying **** about any of them - I want to see more 996's on track, more people less afraid to use them, more pissed off corvette and camarro owners. Faster and more reliable - thats the goal... step by step, brick by brick. ETA... Completely agree - more people posting data would be great, of all the good things this thread has done, if I had to be greedy and say...give me more - that's what I would ask.

Your new signature is hilarious !!!

Quotes source post: 17647426

This would be nice. While at the pca event at vir last weekend im pretty sure i was the only non mezger(origally) 996 at there.

I wonder how much @NuttyProfessor donated to the cause of trying to understand the cause and effect of replacing scavenge pumps and the relief valve using data instead of what someone tells you... If any of the guys who donated feel like I didn't hold up my end of the bargain - I'll give them their money back.

Quotes source post: 17647452

My guess would be $0.. Data collection and sharing is not easy or cheap ,else everyone would be doing it . Some would rather just donate to the cause because IT IS so difficult and costly. I certainly think you held up your end and I don't want my money back. Also having the data and being able to translate/use it for analysis takes another skill. (thanks to Paul)..

While data collection in my case is expensive - it can be done more economically if the scope in narrowed, I would be happy to help anyone looking to collect their own get a system going that will work for them. Here's what is cost me for the data in the thread (thank GOD Paul is free cause I could NEVER afford someone of his talents to do what he does) And the money for the above came from selling the below, dropping $2400 on data stuff is not something I could just do out of pocket - I still miss them

They are good looking rims...

Back OT - The new full oil level using the shorter .1 dip stick compared to the 3/4 capacity on the old, longer dip stick accounted for an average LOW oil pressure increase from 26.37 to 32.33 - an improvement of 5.96 PSI, I removed the first and last laps data from inclusion to not factor in any idle, paddock influence. I really like this comparison because the other data factors, from temps to G and lap times are so close, really illustrates the oil fill level difference IMO. 26.37 average low - .2 longer dip stick, 3/4 capacity 32.33 average low - Full capacity on the .1 shorter dip stick

Even though you can compute to 0.00, your gauge/sensor error is probably +/- 2-3 psi (+/- 5%?), so going to hundredths of psi is irrelevant and the difference of 1-2 psi is probably within the error of measurement.

Quotes source post: 17649021

https://aim-sportline.com/en/product...ressuresensors

Google precision vs. accuracy. 0.5% of 150 psi is +/- 0.75 psi. I would round up your psi numbers to integral psi, e.g. 51.45 is 52 psi and temp the same

Quotes source post: 17649036

I don't care what you would do.

Quotes source post: 17647471

Agreed, any data collection is good because it can create a baseline that can be further tested, questioned, retested, etc. You all know my position, I want to keep the Go-Fund-Me account open and test cooling, before or after the UIDS is ready to go. Unfortunately, all we hear is crickets. Chris was kind enough to offer the CSF radiators for his cost. This work benefits every 996 owner. Even though there seems to be fist fights every week on these threads, the work that is being done by Jason and everyone that is contributing to this thread is invaluable. Great thing about guys and fist fights, is that after its over, we can all grab a beer and it's forgotten. But the data and accomplishments will live on.

The most fascinating thing about this thread is that two huge breakthroughs on oil starvation and stability have come into focus that will have widespread positive impact on the 996 community. 1. UAOS and it's ability to run oil overfilled. 2. UIDS All from the same inventors. All being able to be tested and refined on this thread. Wow...

Quotes source post: 17649094

Thank you !! A 5.96psi improvement in the bottom number (the one that is most scary) is an improvement of , what, about 20 % ?? HUGE

Quotes source post: 17649275

HUGE!

We will find out the next chapter on Tuesday at Lime Rock, which is a track of mostly long right hand sweepers. My understanding is that long right hand turns are the worst case. The weather is predicted to be dry and zbomb's ABS is working properly this time.

Quotes source post: 17649484

Jason please video the session as you have done before. I even turn on my Bose sound bar with base when I�m watching your stuff on you tube at 3-4 am in the morning.

Quotes source posts: 17649484, 17649518

Looking over some of the data from last time - I was going slower than I would have liked, pressure looked pretty good compared to what we have seen at Thompson and Palmer, not really sure how much the pace contributed. I expect to push the car harder on Tuesday and we'll see what we see from there - excited. Yup - video will be up after with the data overlay... think I have finally got the content of the track videos with regards to data where I want. Glad you like the videos, its an easier context to visualize the data for me and add some context to the squiggly lines, I watched one on TV for the first time Yesterday and it was a lot cooler than on my little laptop.

Quotes source post: 17649689

Dude i like how you drive and i can honestly see how you are getting better. Almost 8 lb brisket point angus choice on smoker, wish you were closer you would be welcome to drop by eat and drink, off course talk Porsche stuff. Using post oak and mesquite, this thing is going to be done between 1 and 2 am. Anyway I am also doing country style pork ribs, with my wacko BBQ sauce and jalape�o cheddar smoke sausages. Yeah, I made the potato salad in the morning. Took my wife to a Mexican place for breakfast and told her I wasn�t going to drive like a bat out of hell, anyway 110 in a straightaway, I told her , are we really going that fast she said, yes ac and stereo playing Boston.

I see that Hartech engines in the UK had another great weekend racing at Brands Hatch during the festival or Porsche event - 1,3,4,5,6 Race 1 and 1,2,3 race 2 - all fitted with their valved racing sump, leading the Championship with 100% engine reliability and one event to go.

Quotes source post: 17647450

Come to a Riesent�ter PCA event. There are usually around 10-12 996 in various stages of development and driver experience.

Quotes source post: 17650904

just curious, you guys up north??

Quotes source post: 17650918

Northish. DE Jeeper = Delaware not spanish . Zbomb is a Mass-hole .

Quotes source post: 17650772

Good new for Baz. Sure wish this forum would allow Baz to come back. Never heard about resolution from the administrator. RL Admin, what's the story dude?

Quotes source post: 17650904

It looks tempting to go to summit straight from the NJMP event I�m doing on October 14-15.

Continue to dig this thread and I don't even have a 996 (though my 997.1 C2S is close enough).... Was at Watkins Glen last weekend and someone brought a 2003 996 40 Jahre edition car, which has the X51 package. That's a rare beast, they only made 1963 of them I think. We compared our cars and drove in each of them together and then switched seats, which was fun. I was impressed by the power of the 996, just a smidgen below the power of my car, we thought. I also liked the suspension. The brakes however were clearly inferior to the brakes on the 997 S, which I think are the same as the brakes that were on the 996 turbo in that year. Anyhow, reason I bring this up because I of course checked the oil pressure gauge as we were driving. In my experience WG is not very hard on the pressure issue, ie my gauge stays pegged at 5 bar throughout on that track, but not on others. The gauge on the X51 however was all over the place and dropped wildly in the right-hand turns. . So I think it just demonstrates what is already the consensus here: traditional sump & scavenging modifications (LN or Mantis deep sump, X51) can somewhat mitigate but not eliminate the oil pressure issue. {PS: someone raised the issue of statistical data significance, and that our attempts here would not clear that threshold in medical research. That's true, but even the complicated M96/97 oiling system is utterly trivial compared to even the simplest functions of the human body. In medical research cause and effect are terribly difficult to disentangle, hence the need for large, statistically significant tests. But our little oiling problem would be utterly trivial to figure out if we could only look inside the engine as it goes on . Jason's sensors are the best thing we have for looking inside the engine at the moment. And opinion in good men is but knowledge in the making.. }

Here are some simple charts from today's work at Lime Rock 1st is the overview. The lowest pressure is 34ish during a dip in RPM. Oil temp ends up around 248. Water temp is 201ish. Here is the detail of the PB lap. The largest pressure drop occurs towards the end of the longggg turn 1. Adding the engine risk map and putting it at 45PSI, you can see there is no time where the pressure is below 45 during the risk periods. Not horrible! The drop at 38 sec is interesting. It doesn't follow any of the rules we've learned. I believe this is a drop from the variocam kicking in. It's around 5800RPM. We saw this in early in this thread in Chris' data but haven't seen it much here due to sampling rate. The only weird thing about it is the length. If the drop from the variocam is 0.1 seconds, we should see all of them. I'm not really worried as the drop doesn't come close to touching the risk mask. Current BKM seems to be UAOS with Track Drainback, Short dipstick(996.1?), run full, LN 2qt deep sump. Or, get a Mezger.

Thanks Paul. As always great work. Jason and Skip, any extra oil in relation to last week's session, or the same amount?

Same as last week, full on .1 dip stick. Will review current data against previous LR data and circumstances soon.

Sounds great. Looks like you guys had a good time. 👍

Some notes to consolidate: The BKM will not guaranty every engine will survive this kind of track duty. This engine with this method has moved away from danger provided the risk mask is correct. Being the first detail data like this on the forum, it provides a baseline for others but no guaranties. For example, if your engine has larger clearances at all the bearings, we would expect the pressure drops to be greater. Certainly this list will decrease pressure drops by increasing the time the oil pickup is covered, but the number and magnitude of drops may be greater putting the engine within the risk zone. If you have a fresh build with good bearings, and less blow by, the drops should be much less, keeping the engine further from the risk zone. There's no way we can know unless others start instrumenting their cars. Recall: (also need to add XP9 oil as part of the BKM package as no other oil has been tested) -The LN 2qt deep sump was already on the car before instrumentation. The belief is that it helps pressure drops, but we have no data to quantify it. -This engine has large blow by and started out with huge blue clouds through turns, so much blow by the AOS couldn't keep up. -UAOS (before track drainbrack) got rid of the blue cloud problem by managing the large blow by, but, the blow by was still large enough to fill up the street use reservoir -New scavenge pumps were added but the data was mixed on whether they helped -A larger catch can was added which enabled increasing the oil level -The track drainback became available and it was added allowing further oil level increases -The dipstick was changed to the .1 version to reference and use the original oil level, currently running full All through this, driver ability has improved. Braking is much faster, turning is much sharper. Both of these work against pressure stability. Is there anyway to mitigate aeration? Seems like an engine design task.

Paul, great summary and question on aeration. ​​​We should also take into consideration the potential advantages of the below listed items which certainly could help going forward. 1. UIDS - coming soon 2. More water cooling - available now 3. More oil cooling - coming soon 4. X51 dual pump - available now 5. Hartech's internal engine machining procedures for opening up oil flow down to sump and water cooling flow. Just thinking out loud.

Quotes source post: 17650963

Massholes are OK, married one (not Z bomb) so at least I have some experience.

The comparative data to last time at Lime Rock kind of blows, Here are some of the variables to consider when viewing the below. - It was hotter in July, using the temp sensor being logged from the car's CAN 7/6 102 F 9/7 91 F - I was going a lot slower in July fighting ABS issues. - Oil capacity in July was 1/2 full on the shorter .2 dip stick compared to FULL capacity on the shorter .1 dip stick 9/7. That said, slower would be advantageous to better pressure, but the higher temps are conducive to larger drops... I'll leave it to the reader to establish an opinion on the contribution from these factors. 7/6 average LOW pressure 37.4 PSI 9/7 average LOW pressure 43.5 PSI A 6.1 PSI LOW pressure increase between the 2 days. Just taking the numbers for what they are, the other day at Thompson was a 5.9 PSI increase coupled with a 6.1 PSI increase at Lime Rock... That looks pretty good so far. Again, I have omitted the out and in laps from the equation. 7/6 9/7

For sure looks like 6psi trend to me

That�s promising. After a few more consistent increases to get a reliable baseline, if it were my car, I would replace the sump with the stock one to measure the change if any, and if that showed a negative trend, try the FVD sump. It would be really interesting to see the differences in the three sump designs, before you add the UIDS. For science.

Tom makes an interesting point. Question for Skip and Jason, are you planning on trying higher oil levels than what you have run the last two track days to see its effects. In essence attempting to replicate what Chris and team did with the Boxter to some degree? The enhanced capabilities of the UAOS have certainly been an incredibly positive development on mitigating oil starvation and pressure stability. Just like Chris has demonstrated, maybe there is more starvation/pressure mitigation to be had with even higher oil levels with Jason's current set-up before testing the UIDS. Fast forward a bit to when the UIDS is ready for sale and in combination with the UAOS with return pipe, my questions would wrap around what is a more effective set-up for the track wrapped around oil levels and the use of 2qt extension. 1. UIDS/UAOS return pipe without 2QT extension at full oil. 2. UIDS/UAOS return pipe without 2QT extension at overfill oil (level TBD). 3. UIDS/UAOS return pipe with 2QT extension at full oil. 4. UIDS/UAOS return pipe with 2QT extension at overfill oil (level TBD). You get the idea. While the deep sump doesn't solve the oil pressure stability issue, there is a benefit to extra oil in the sump. It may help with oil starvation and it may help with oil cooling in a 20-25 min session, but after that it will work against you as the oil heats up and swamps the stock oil and water cooling capabilities of the car. Add edit: So at this stage, without data to support the comments I am about to make, here is the set-up I would go to in 2022 with all that we know at this point as well as providing flexibility in how extra oil can be used in the engine. 1. UAOS with return pipe to run overfill if needed. 2. UIDS (in development) 3. 2QT extension for more oil. May not be needed depending on ability to run overfill and results of UIDS tests on track at full levels. 4. LN Oil cooler (in development) 5. CSF 3 radiators

Quotes source post: 17657144

Tom - I would highly suggest you run the tests, it would be interesting. I will not be testing Sumps I do not feel (my opinion) will FIX oil pressure instability and here’s why: 1. Everything I do to my car is to try to make it my version of better, faster and more reliable at the track and on the road. I am not interested in going backwards for science. 2. Every data set posted here has cost me : entry fees, oil, gas, brake pads, rotors, tires, misc repairs and the list goes on. I run track days for fun and it is an expensive hobby - I simply cannot afford to throw away a track day in the name of curiosity and science. I really wish I could, but I can’t. 3. I do not have any motivation to go and find every part that doesn’t work as advertised - that’s up to the consumers to demand more from the MFGs we buy parts from to validate their claims. It is a bummer that I started all this with the LN sump on the car because now with the UAOS based improvements, we all want to know what it would be like without the deep sump - the reality is - if I started this thread and DID NOT have a deep sump the majority of responses were likely to be along the lines of “you need to get the deep sump, what did you expect” obviously those responses would not contribute to RESOLVING the problem but most people did not know any better, myself included. I bought the LN part because I thought it would prevent exactly the issue we see documented so thoroughly here - now we know and we have a couple data sets with documented IMPROVEMENT. My opinion is the good money is not on one of the currently available deep sumps but on the UAOS, drainback, .1 dip stick and full oil level, again - that’s my opinion and what I would do if I knew everything I know now and had to do it again. You guys can do whatever you want. Hopefully, when Skips sump gets some miles, we’ll have a solution paired with the UAOS that is bulletproof - we’ll just have to see what we see there.

Quotes source post: 17657190

Jason, I think you nailed it. At this stage, everything revolves around finding the best set-up with UAOS and UIDS at this point with oil levels and cooling.

Quotes source post: 17655447

I�d like to add a couple more variables to test, based on posts in other RL forums e.g., Boxster, from over a decade ago. I�m not saying this is for science, it just makes sense to compare products if we�re spending significant sums on them. - using a different oil, especially Motul 300V, to see how it handles the aeration issues at these elevated temps. - the indy tech I use runs a Mantis sump with the LN .5 L spacer with no issues. I was surprised at the engineering of the Mantis sump plate, shown below. The design and, especially the windage plate to limit sloshing around. Lastly, I apologize if my previous posts have been off-putting to some.

Quotes source post: 17658618

Turkis, All good suggestions. However we have to get a few more guys involved. I will certainly volunteer my car for next season after I figure out how to get my AIM system up to snuff. But we have to get a few more of us in the game with Jason. No way he can do it all. Who wants to help?

Quotes source posts: 17645483, 17649021, 17649036, 17658618

You say and and and then you post this... What do you mean by no issue, how have you quantified it ? Kinda hard for me to square up your posts on my data when you post a straight unsubstantiated claim subsequent... See, it's statements like " I ran the LN/Mantis/FVD etc... deep sump with no issue" that got us to the point we're at now, where people like me spend $900 on something that does not resolve the problem its built, marketed and parroted on the forums its supposed to resolve. You have people like BAZ and Chris who say.... they're all junk, none of them work. The only data you have seen - cause its the only data that is available confirms - junk, don't work. So, if you want to make a statement in this thread about the efficacy of something, it should be based on your professional opinion, like BAZ or Chris or based on the data you are prepared to share. There is no one who wants to broaden the horizons of data that's collected, collectors and the sharing of that data in an open forum more than me. This thread is for that - NOT "someone told me it's good"

Quotes source post: 17658618

you should go test this yourself and provide the results. In my previous life I was extremely involved in Motorsports at a professional level and at an amateur level. Bunch of people will sell a lot os stuff that doesn�t work, but they advertise these many hp improved, etc, etc. Z bomb has done an awesome job, I�m with him when he says that some of this stuff doesn�t work and it�s a lot money that people are spending in bs. If you are taken away by the way something looks, put your money where your mouth is and go ahead buy it and test it.

Hello im currently rebuilding my 2005 997 3.6. Ive been reading this post about low oil pressure. I am at the point of installing crankcase carrier into bank 1. I know the crankcase carrier does typically not have sealant between it and the crankcase case halves, but I�m wondering if the two different metals and expansion rates cause a leak when it�s hot causing a loss of oil pressure. It�s a big surface. And the alloy should expand at a greater value then the steel case. Plus at high temps the oil is very thin less viscous. what do you think should I put a tiny bit of sealant around the carrier to crankcase half interface on each side?

Here is data from Thompson run on 9/16. It's pretty incredible in that the previous personal best was 1:20.8, and in this session 6 of the 11 laps are better than this. The best is lap 11 at 1:19.3. But I digress, this thread has nothing to do with speed or driving efficiency until it affect oil pressure... Here is the overview of session 3. The pressure drops are larger than we've seen in the last sessions. But... (go to next graph) Oil Temp ~ 245 Water Temp ~ 200 When we apply the engine risk mask, we find out that there is no time when the pressure drops below 45psi during the mask periods. This is the personal best lap. So the question moves to why the drops seem larger even if we are not in the 'risk' zone. Here's a comparison of the current personal best and the previous personal best from 7/31. Top group: Red = Oil pressure 9/16, Blue = OP 7/31, Orange = RPM 9/16, Purple = RPM 7/31, light Blue = engine risk mask (4500rpm with 75% throttle) set to 45psi 2nd group: Greens = 9/16, Blues = 7/31 3rd group: Black = 9/16 Oil Temp 4th group: Time difference (7/31 - 9/16) The key thing to look for here is why the OP has higher drops during this session. The 2 primary effectors are Oil Temperature and Accelerations (both Lateral and Inline). Looking closely at the overall accelerations plots, the values through the turns are very close to the same. The oil temperature shows a difference of 7 or 8 degrees. It's tempting to say that this lap is 1.5 seconds faster and that will explain it, but the G's don't really bare that out. The time difference is due to driver improvement as can be seen by the RPM's and Inline acceleration. It's clear to see that braking is occurring much later into a turn, indicating carrying more speed into the turn and shaving time. I'll try to get some better plots up in the confused thread to show this and the effect on speed. Discussion required.

Oil temperature being 7-8 degrees higher would be my theory as to why the drops are bigger on this session. (provided the same oil at the same level was used).. 40 weight oil starts out at 100-125 viscosity at 70F, and by 212F is down to 13-17 viscosity, and by 245F is probably down to 8-10 ( no specific data for oil viscosity above 212F but it is obvious)..

You should put back the math-channel for 10psi/1k RPM. Makes it much easier to see where to pressure drops unrelated to RPM.

I was a **** show getting ready for the track - i did not spend my normal time preparing the car and the oil level was not interrogated closely, that's due to the fact that I was relatively certain it was equal to the previous days settings, after looking at the data, I am not so sure. I am sorry I can't give more definition there. The oil temps were analogous to my previous track day, slightly less even so I would not say the drops are temp related. New data above the black line, old below.

Quotes source post: 17675588

Jason are you still running the .1 dip stick?

Quotes source post: 17675594

Yes.

What happened to this thread? Did the party move elsewhere?

Maybe winter hiatus. For me, I've been pulled overseas to help fix some production line issues, other than the fact we are paying 10X for 60% of the electronics we need for our parts. The product is ramping 12X the previous one due to now opening to new markets. Yield is always under pressure during ramps and this one is on steroids. This gives me very little time to spend on the boards. The discussion here is awesome and I'll be back, but not until mid-late December. Things are bad enough here that our VP ops has threatened to take my passport away(jokingly, I hope) until yield improves. For me, relative this thread, the next step is to create a good summary of all the data Jason collected at Thompson and create some graphs that show how oil pressure drops have changed relative the changes made to the car. Not easy given how the curves shift a lot from ambient temperature. Maybe I can have something done by Christmas. Jason, can you send me the last data set? I missed it in the mail. Or it got stuck in the pacific ocean bit bucket. If anyone else wants to take a stab at the data in the meantime, contact Jason. Don't worry, the party will continue! ETA. I guess it's Fall hiatus... time here is weird.

Quotes source post: 17740915

End of season lull, as we wait for the release of UIDS and LN Oil cooler.

Baz shared pics of his sump design on FB and plans to start selling this year. https://www.facebook.com/groups/5918...0583780971107/

May have miss read it when Baz was posting here. But along with that sump, didnt he recommend an accusump plumbed to a modified oil pump for it to all come together. Could be wrong here, it�s been awhile since I read his posts.

You may be right.

Quotes source post: 17761626

U r correct. There is more then 1 piece to this puzzle. Im glad baz is getting into the us market

IMHO, we may have three phases of oil pressure stability before us. The first will be to fine tune the use of the UIDS and Oil cooler in 2022 for optimal oil pressure stability to prolong engine life until we need a rebuild. The second will be to identify internal machining solutions during a rebuild to improve oil and water flow to improve durability in the M96 engine when we bore it out for an extra 100+ HP. In particular, who does it, how they do it, etc. Hartech outlines their procedures. FSI broadly discusses their procedures. So the procedures exist for a rebuild. Along the way, sandwiched between 1 and 2 above, there may be an additional external hybrid solutions of using the X51 pump and Oil tube available thru LN and H2 externally for oil flow improvement in combination with the UIDS and Oil cooler. Just thinking out loud.

You guys are doing the lord's work. My dream is to get an M96 as close to Mezger engine robustness as possible. It will take alot of work as has been outlined in this thread but more and more it is looking like a possibility.

Quotes source post: 17763224

He may own a 996! He is just testing us to see if men with free will could find a solution to oil starvation and oil pressure stability in the M96. 😁

Wow. What a thread. The collaborative community effort to use real data to work towards solving real problems has been phenomenal. I appreciate Jason for pointing me in this direction and all of the work he's put towards moving this effort forward. Sounds like fun. Count me in as another test vehicle. That puts us at three doesn't it? A little background on myself. I've done both DE and racing and fall somewhere in the middle. While my racecraft is almost non-existent (more like ragecraft), I have had a few local podiums and typically outqualified 400+ HP cars with a 265rwhp 99' C2. I had 60+ track days with that car using a stock AOS, stock IMS (dual row), AESCO LWFW, and FVD baffle. I also had a center rad, ran distilled water/water wetter, changed oil every 7 hours with XP-9 or M1 15w50 oil, and got a used oil analysis(UOA) after most oil changes to monitor engine wear and oil performance. And I never had an engine issue or worrisome UOA. It's been said multiple times in this thread by engine builders and racers both, and I will echo the same, if you want to put real track days on a Porsche 911, and don't have an unlimited budget, go buy an early '99 911. It is the best value in price, speed, and reliability for tracking a Porsche by a long shot. And if you are worried about the 3.4's HP, go check out how many of Spec996 track records are set by a '99. One other thing I haven't seen mentioned in this thread, is that X51 3.4l 996 factory shortblocks, which include the last factory updates, is $8500 list. This is the least expensive rebuild option with a motor that has proven to take a pounding. But based on posts from Chris, Baz and Jake's, it seems to be the 3-chain, M96/M97, non-DFI engines that have oil starvation issues and that can gain the most value from this thread. Which brings me to my current project, an '05 Boxster S. Why did I choose it you ask? I didn't, it chose me. I took it in on trade for a 03 996TT track car to teach my kids to drive an MT and then fell in love with it and just couldn't justify getting something else when I decided to do a race build. While I enjoy my days at the track, my real passion is all of the time I spend in the garage getting to work on my car. I used to be an avid PC gamer, but working on Porsche's seems to scratch that same itch, only IRL. And the 987 Boxster is light, easy for me to get around in and work on, has cheap parts, full CANbus, multiple engine upgrade paths, extra frame reinforcement etc� Downsides include it doesn't have a good PCA class which means limited aftermarket 987 Boxster specific race parts. And I really dislike the feel(or lack of) of the stock 987 ABS. I've included a link to my full build sheet below, but parts of relevance to this effort include � '05 M96.26 3.2L 3-chain motor (stock internals), 60k miles, 5 track days Stock Sump Stock AOS (5 track days since replacement) Original stock rads OEM Center RAD Deleted AC, Condensers, All fans (+ more air flow) Deleted Heater core and lines (- less water volume) 50/50 Coolant with Water Wetter Amsoil Dominator 15W50 Stock Oil filter housing AIM MXP Data system 5 Track Days, Good UOA On the data system design, I think it makes sense to mimic Jason's pancake adapter method of OP/OT collection for data consistency but I'd also be interested in one of Chris's oil coolers with bungs for additional OP/OT sensor locations. From there, we can get a baseline set of data, make sure everything looks right, and then start choosing variables(one at a time) that you all want to test. And while I'm open to loaner parts or discounts people are in a position to extend, I have no issues buying and installing radiators, sumps, fluids etc if you all are willing to continue crunching the data and providing the analysis. I still have some work to do to finish this winter's build list, but then I should be able to test a variable every month or so on the track with additional data from a few Autocrosses as well. Buildsheet: https://docs.google.com/spreadsheets...it?usp=sharing Brake fluid, pads, tires, and oil and off to the track for our first day. Opening up the wiring harness before the holidays to trace and clip every wire that doesn't make it go, stop or turn.

Quotes source post: 17871807

Slakker! Welcome back! I am still using many of your old parts on my �99 996 track car. Thanks for all the wisdom over the years 👏🙏

Having Brandon @Slakker volunteer to actively devote his time and resources to deepening the data pool should be greatly appreciated by the community, it is certainly greatly appreciated by me. After getting to know him and his depth of experience with using these engines hard, he has a lot to offer from experience, use and the data system his car is equipped with. People want to see more than just my use case, I don�t think many more than me in reality. My resources are constantly strapped and I simply do not have the bandwidth from a financial, track time or ability standpoint to give the people who want to see this solved all that they require or need. There are a lot of times in this thread where people want me to test this or that, run 6 different oils, run sensors all over the engine all while keeping variables to nil. That is not something I am able to do to the degree of someone who does this for a living, and frankly - for the products on the market that have been marketed to address this issue, should have been done by them. The fact little data is available on what the professionals have done is because us, the consumers, have not determined it is important for us to see it. I hope @Slakker volunteering to take part in part testing and data publishing may help more decide it�s something they want to take part in as well and the pool of data on the parts we spend our hard earned money on can gain real clarity. Slakker mentioned something in one of our chats that in reflection, although we were not in agreement in reasoning - I think no harm in stating publicly, although I have previously, not in this thread (I�m not even sure anyone really cares my plans forward going). It is not my intent to run an M96 long term and my plan currently is to get one more season out of the motor in the car and hopefully be able to pull it next winter and recoup some value from it. I plan to put a dry sumped LS3 in the car to reach the power goals I have made. My contributions here are of little to no personal gain and entirely motivated by getting parts developed to give people more confidence to get their cars on track where their performance is really pretty amazing. I have the most legal fun I have ever had shredding on track with my 996 bros and would love to see the party get bigger.

Thanks guys. @zbomb , I certainly understand your choice and we each have to build our own build, Your contributions to the public knowledge pool on these engines has been tremendous. It will be fun to see what all we can learn this year.

Welcome back Slakker !! Glad you will be contributing again !!!!!

Quotes source post: 17873940

Thanks. Great work on the AOS and Sump issues! Got any interest in a 987 beta tester?

Quotes source post: 17871987

[Image-only post; no author-written text. Original image references:] https://cimg9.ibsrv.net/gimg/rennlist.com-vbulletin/643x539/10821dd8_f810_4684_8b90_58be58e4ec5c_30d1562c08e76aa51e0ef2a7cfa0ae3d57de9d1b.jpeg

Quotes source post: 17874345

Thanks !!! I have already done the Alfa and Beta testing on the UIDS, but I welcome any tests on the UIDS you want to do ,especially if you can show where your present set-up is unstable, then show how stable oil pressure is after the UIDS.

Quotes source post: 17881483

That sounds even better. I�ll put it first on the list after grabbing my baseline data with a stock sump.

Look at this: possible fuel starvation in more modern 911s! https://rennlist.com/forums/992-gt3-...-g-forces.html

Ha! My friend who tracks his 370Z regularly has the same exact issue. Pretty surprised this to be happening on a GT product though that is build for the track first and road second.

New technology always equals the discovery of new flaws. Glad we know all the flaws in our 996 engines and solutions. Pretty easy to sleep at night.

I thought I'd post these videos on Zbomb's thread also. I have found that a lot of threads can be missed by even people who come on RL frequently ..I hear it alot the times " I just found this thread"....... Any here are two videos of a Boxster M96 at Road Atlanta and Sebring ..... You Tube You Tube This thread has had a litle rest but I think people will be sending me more oil stability video's and I'll post them here also along with any data logs..

This issue is certainly not unique to our cars and not the genesis of the thread. The issue I wanted to bring to light was that the predominant aftermarket remedy available did not work for me and to further the discussion of what what could be done to mitigate or resolve apart from the boiler plate answer of put a deep sump (LN or otherwise) and baffle on and run a certain oil. Unless there was some level of acknowledgment that the legacy systems may not work as well as people expected, it seemed highly unlikely that there would be any further push to make something better.

New car, well, not new. New data! This is from Geoff.T's car from Buttonwillow in February. Here is the car: (it looks fantastic, I want one) 2002 996 M96 LN sleeved 4.0L, compression increased (100 octane) Oil pressure taken from stock location (cylinder head, bank 2). oil temp sensor located on cylinder head, bank 1, water temp measured at the water pump housing. Exhaust: headers, no cats, 996 center exit cup style exhaust Sump mods: Bilt racing 2qt deep sump and baffle setup Oil: Driven XP9 10w40 Cooling: CSF radiators with CSF center center radiator, no thermostat, no fans. Weight: Approx 2,500lbs (CF hood, roof, doors, engine lid, completely stripped and caged) Tires: 245/40r18 & 315/30r18 Nitto NT01s Suspension: 996 GT3 coilovers increased spring rates Here's the overview of the last session. Only laps 2 and 4 show the car moving at speed. The other laps are so slow, something must've happened. I've tried to mimic the colors and style used in the previous overviews for comparison. -The first standout is oil temp. It is nearly identical to water temp. Need to check what's happening there. I can't believe this is possible. Look at the purple and black traces -The overall pattern is the same between this and Zbomb plots. The first lap we see the water temp low and the oil pressure high, and through the session, oil pressure comes down to some equilibrium -The elephant in the room is the lack of oil pressure drops. But, there are 2 in lap 4. All the others are easily associated with engine RPM. This pressure sensor is on the head. May be different from Jason's point of measure. Need to check Here's the track map made with the GPS recorded during the session with some of the turns labeled. (sorry, I'm either lazy or running out of time, you get to pick) Here's lap 4 with accelerations added. The shaded areas are showing the turns as colored on the map. It's a nice function with the software, however, it does create clutter in the graph, noise. Somehow I never found this when putting together all the previous graphs. (again, either lazy or rushed) The 2 drops are quite apparent. Very small in magnitude. Not occurring at a variocam trigger point. And both occur on left turns!!! There are no drops on the right sweepers. I didn't run the 'danger zone' test because I think we are well away from it based on how small these drops are. The Inline acceleration doesn't get above zero very often. This seems a little strange. Need to confirm any differences in Oil Pressure measurement location between cars before we can really say anything. I'll look at it some more tomorrow night after work and get the Danger Zone plotted. Please publish questions. -How many miles on the engine since rebuild?

Appreciate @Geoff.T adding data points to the thread and, as always Paul @golock911 formatting the data in a palatable way ! As a reminder, both my oil P and oil T are measured at the oil filter. The data from Geoff makes me very interested in placing a second sensor in the head - I'm wondering if, as a result of the data being captured in the head (end system) as opposed to the filter (near beginning of system) what we may see is some level of smoothing happening, where my sensor sees drops right as they exit the oil pump, a measurement taken at the end of the system has had the opportunity to be influenced by the various loss points where the system has bled some pressure out. Hoping this makes sense.... Geoff had mentioned the ambients at the time the data was collected to be around 55F, still not cool enough to make it clear to me the temp data we are seeing, and the inversion of water / oil temps. This car has both water and oil temps topping out at about 175F with the oil running slightly cooler. An interesting note, this car has both a discrete water temp sensor in the pump housing and is reporting ECU water temp and the values are basically identical, this makes me feel good about continuing to rely on the stock system for monitoring water temps. This data covers the 2 "at speed" laps recorded during the session. Same base data Paul posted above, just a different way to view it.

Here's a thread from 986forum that contains some measurements of oil pressure before the oil filter, after the oil filter, and then in the valve cover. I'm on my phone and can't really zoom in, and also it is an accusump thread, but I think he ran a few laps with the accusump disabled and should provide some realtime comparison of various oil pressure sensor location readings and whether right/left turns show a difference in the valve cover location. https://986forum.com/forums/boxster-...some-data.html

Some data from the almost new car. Also, an updated version of the software that will help the user pull stuff together, but I'm not sure it helps report out. Regardless: First is the track layout broken up into sections by color for sections: Instead of plotting the whole session, this plot overlays the best lap of last year with the best lap of this session. Sections of the map are shown across the top of the chart. 1st drive in an extensively modified car. The green traces are from the best lap last year. The brownish trace is the best of this session. The primary difference seems to be braking point. The bottom trace shows the time difference between the laps accumulate. The second from the bottom shows the oil temp on this lap is hotter than the PB lap. Lateral acceleration seems similar, inline acceleration is a little less. Overall, the oil pressure is better in this lap than the PB lap. The 2 odd points on the current lap occurs towards the end of segment 9. I can't explain it. The rpms are increasing out of a right hand turn off the peak lateral acceleration. Odd. Here is the oil pressure vs lateral acceleration for both laps overlayed. It shows the overall pressure was better this session. Here is the current lap with coloring to show the RPM at the given oil pressure point. It shows that at the highest accelerations, the RPMs are lowest. Not an issue. The only thing bugging me here are those 2 drops towards the end of segment 9. I can't explain them. All the others fit into the pressure drops we saw on the PB lap. Overall, the engine seems the same as before. We'll need to watch it closely as I expect the accelerations will be going up significantly in the next sessions as Jason gets the car sorted. Let me know if there is anything you wish to see. I'll scrabble something together.

Is this most recent lap data from the lap where the back got loose in Turn 9?

Quotes source post: 18230656

I don't believe so, that little ride was a lap killer as far as time goes.

BTW, oil fill for the data above was 11.75 QTS or XP9. Still running the LN sump and baffle and the UAOS with the track drain back. I attribute much of the slowness to getting familiar and gaining confidence in the new brake setup. The car also needs some chassis optimization to reap the benefits of the new dampers, bigger front splitter and the 275 front tires. That all said, I don�t really have an answer for why the pressure looked better, the temps certainly seemed to be playing against that outcome. Paul, thanks for picking up where you left off with the fabulous data summary ! The couple dips you mentioned, were they consistent every lap or anomalous to the lap we�re looking at ?

Jason, did you make any changes with water cooling system this year? Or same as last year?

Quotes source post: 18231109

Cooling system hardware unchanged. The new bumper may have an affect on airflow to rads but not something I can quantify.

I haven't studied the old data since last season in relation to new data, but oil pressure looked good, water temps were good, but oil temps moved from the 240s last season to low 250s on the footage but held. Maybe one more late shift on the straights may help to contain oil temp? But may hurt speed due to aero and meaty tires? Dunno, just thinking out loud. It's always a balancing act.

Quotes source post: 18231055

Sorry I didn't look at this before blathering about it. The dips do occur on other laps but to varying degrees. Here are the 1st and last laps of the session with the worst dip lap. The overall change in pressure is the oil temp increase across the session. RPM's are at 4700. Lat G's are less than 1. Even with the dip down to 35psi, it's still well above the lows from last year. Overall G's are lower. Here's the equivalent plot from last year. It shows the same drops. Not all laps have it, but like some other weird coincidences, it is amazingly consistent in lap position and car dynamics. I got nothin. Last burp of aerated oil? But how could that be so consistent?

If I was a betting man, this thread may evolve into a conversation on cooling this season as it pertains to oil pressure instability. It's the last piece of the puzzle that hasn't been fully quantified and solved for.

New dampers? Is there maybe less body roll that can explain the better oil pressure.

Sure, anything is possible. But my best guess is that because Jason has done nothing to his engine and cooling, with the only variables being his new aero, brakes and suspension that would create a question on anybody's mind as to where is the "new" limit. As such, you don't attack the turn and braking areas as you did with your past set-up until you figure out where the limit is. As a result, your Gs are lower and you compensate by pushing the throttle harder and longer on the straights which creates more heating. We have all been there until we get the set-up to our liking. Happened to me many times as I changed things. Even as simple as moving to a different set of tires that you never used before.

G`dday all from down here in NZ. First off thanks for the awesome R&D work being done by people on here, as well as the contributions from the like of Jake and Chris - it`s been very informative. I`ve just read through the 55 (and counting) pages of this thread and figured I'd chime in with some observations and what i`m doing on our 2001 Carrera 4. My car has around 150k kms, Pilot Sport PS2 tyres (so not especially grippy), and I have observed mild oil pressure instability on tarmac rally touring stages (limited to 160km/hr) and track days. Shameless pre-Targa pic I know this, as I film in-car, and car review the oil pressure gauge against track geography and RPM / engine temperature. Never below 2.5 bar (~35 psi?) and generally on left hand bends following heavy braking. This has been observed with the current setup of an FVD deep sump, FVD low temperature thermostat, and a new genuine waterpump, with street oil (the Porsche Classic 996 oil or Motul). Anecdotally, the deep sump and slightly better cooling did improve observed minimum oil pressure by around .5 of a bar, but it`s still affected by the resultant G forces from braking and cornering. I`d have to concur with Jake`s observations about heat build-up in the oil, as it mirrors my experiences. Following a delivery from Rosepassion, I now have the OEM return line to go with one of Martin`s (H2 Engineering) Dual Scavenge oil pumps, together with the OEM third radiator kit, sitting in the car to be fitted. Rear face - really well machined H2 Engineering scavenge pump This is all being installed next week by my awesome mechanic, in advance of a PCNZ DE day at the start of August, and the Targa NZ South Island Tour in October. I know there has been a lot of conjecture about the dual scavenge pumps, so i figured I'd put my money where my mouth is and get one, fit it up, and see how it goes. I`ll circle back with the outcomes, which may be useful to the wider M96 community.

the show so far: I've gone through the Thompson races and pulled together the best laps in each outing to visually show how the oil pressure drops have changed over time. The oil quantity used and date of the race is at the bottom of the plot. All plots have the same Y-axis. The new Race Studio version doesn't manage Y axis labeling as well as the last version. It takes up alot of space. This is close to the beginning: 1/4 full on the 996.2 dipstick, immediately after installing the UAOS due to blue smoke belching on the previous track. Blue smoke belching solved Here are the next 2 races as oil level is being increased. Clearly the pressure drops are becoming less. The hard part is interpreting the effect of oil temperature on the drops. The black line is the oil temp. The 7/29 drops are clearly smaller than the 6/27 drops. The oil temp is quite a bit cooler as well. The improvement is likely due to both. Significantly better than the first, 8/31 drops seem to be similar in magnitude, but look at the width of each drop, they are much narrower This shows the personal best from last year and the first plot of this year. The newest plot is pretty amazing even with the hotter oil temperature. We could claim rusty driver, given it's the first of the season, but the time on this lap is the same or better than the laps shown(except PB), so the pressure drops are absolutely less. Tomorrow I'll try to get the engine risk summary up. The hope is to show how the total time the engine is at risk has gone to zero. (until the new suspension and aerodynamics are worked out...) Stay tuned!

Engine Risk. Recall that first we create a mask that identifies any time where RPM's are higher than 4500 and throttle is more than 75%. Then we look for oil pressure Less Than certain values within those times. This plot overlays 2 laps. The green is the fastest lap from June'21. The brownish(sorry bad color) is the fastest lap June'22. This lap is faster by 0.7s. The oil temp in the current lap is a little hotter. Each of the rows below the track data show the time/distance that the oil pressure is below the test value. The first time blips we see are below 35psi for June'21 lap. We know it is more than 25 but less than 35. There are 3 periods that violate the rule. The first risk for the current lap is between 40 and 45. It's funny that this location is the same as that weird drop out we discussed above in turn 9. In any case, the current lap shows a second blip show up on the 50 line. While more time gets added to the lap from June'21. No question the engine is much lower risk today. What pressure did we say we were concerned with? My brain isn't coming up with a number and I couldn't find one in the thread. The primary thing that's been done to push this value up is to add oil, as allowed by the UAOS Track Drain Back. I think the next step is to add another 1/4 to 1/2 quart. We don't know the flow limit for the drain back, and the UAOS has a warning chime when it is full. Maybe it's not necessary if 40 to 45psi for the worst case drop is actually not a risk. There is no question how valuable the Drain Back has been.

What a cool way to look at it ! Very impressive Paul - A million thanks for taking the time to go through and create such an interesting way to review the data. I think for my comfort level, I'll prob shoot for another increase for the next day I have a week from today, I haven't checked consumption yet from the previous day but if it looks as though there was no meaningful oil consumption, I will add in the 1/4 QT that will make for 12 and see what that gets us. And to your point on the drainback, it is the singular mod that has kept this motor in the car, with the blow by I have, and how keeping oil out of the intake got harder and harder as I got faster, without it I would have either got fed up and went in another direction or it would have blown up if I ran it too low.

Paul, I think you need to read the thread over again. The primary things for getting where Zbomb is today is: Deep sump, scavenge pumps and more oil. It works like this: The deep sump lowers the pick up x inch below the oil level in the sump, creating extra buffer before it runs dry when the oil sloshes around. Fresh scavenge pumps makes sure that maximum oil is returned to the sump for critical situations, preventing/slowing down the sump from running dry. More oil gives more buffer in situations where the engine is pooling oil outside of the sump. The question now is, can he safely keep filling oil until the pressure drops are gone or will this damage something? If you look at my investigation in post #49( https://rennlist.com/forums/996-foru...corners-4.html ) , the running oil level is below minimum level if pressure drops happen below 1G lateral force.

Quotes source post: 18241871

The rule of thumb is 10 PSI per 1000 RPM. At 45 PSI as a minimum there should be no additional problems, provided the oil temp stays tolerable. The concern that I have is how much - if any - engine wear occurred before the oil pressure was secured with these mods. The other concern I have is, if we have secured a suitable height of oil above the pump inlet why are there oil pressure drops at all? It would seem to me that if there is sufficient oil for the pump to ingest there ought to be no oil pressure drops. I suspect that the oil pump is still cavitating to an extent because there is not quite enough oil column to overcome the long suction tube. This would be very difficult to prove or disprove. Regardless the improvement is substantial.

@Nick_L Thanks for the summary of the build and description of the oiling system and your input on the possible causes for oil pressure drops and statements that I haven't been paying attention. You left out the changes to the AOS system that allow it to manage significantly more oil saturated air. The original AOS lack of capacity is why oil levels need to be run so low given the design of the oiling system. The UAOS track drain back has increased the AOS system capacity and allowed the oil level to be increased, which overcomes whatever oil return obstructions exist within oiling system. I absolutely agree with your question, " The question now is, can he safely keep filling oil until the pressure drops are gone or will this damage something?". What we don't know is the capacity of the new UAOS. In this case, we know this engine has significant blow by which overwhelmed the AOS at only 1/4 on the .2 stick. With the UAOS, the engine is now running 1.25-1.5 quarts above that initial low level. With the integrated warning system, we can slowly increase the amount of oil without worrying about overwhelming the new UAOS. We are still 1.5 quarts less than the 2 quarts overfull with an open AOS that Chris spoke about in one of his posts. I think we have room to move the oil level up further.

I would imagjne that u would want to stay safely below the crank to keep the airation to a minimum. Not sure if anyone has tested thar limit yet?

Just to clairify a few things any deviation of the max oil pressure achived when no external forces are encoutered such as G-force and only relative to RPM and oil temp/viscosity would be less than ideal. These less than ideal circumstances are only due to air or areated oil being less than ideal for ideal pressures. The 10psi per 1k RPM rule of thumb is valid if the oil is of good quality and not full of air. A 45psi with good oil is better protection than 45 psi of aerated oil/air, and the pressure differential from ideal is the key to determine how much the oil is aerated. The rod bearings and main bearings and also the hydraulic lifters can withstand a certain amount of aerated oil , but less is better... The improvements seen here are solely due to the increase in oil level due to UAOS with hm/XTDD having more "good" oil delivered to the lubrication system. The new scavenge pumps made no difference and the old scavenge pumps were in good condition.( I have them right beside me).I measured the volume of the scavenge pumps and 1 revolution of 1 scavenge pump will return 72cc of oil, at 4000 RPM it will return 38 gallons of oil per min per pump. Even an inefficient/worn scavenge pump will deliver way more oil back to the sump than can leave the sump.. The maximun the lubrication system flow is about 10 gal per min.. BTW there is no danger of the oil level being so high it will contact the crankshaft on the m96, even filled with 12 qts oil and ALL of it in the sump, the level if about halfway up the IMS bearing, still inches away from the crankshaft.. Instead of pushing the safety limits of the UAOS system, a better test/experiment would be a XP6 50wt oil. This would slow the "turn-over rate" of the oil and increase the "residence time" in the sump and therby reduce the overall aeration and have more "good oil" availible to the bearings..

Quotes source post: 18243692

Wow, that�s interesting.

I think the higher operating oil temp is due to oil volume. More oil volume requires more oil cooling. Oil cooling control is key to longevity at the track. Jake points this out in #76. More volume within a closed system means more oil in the sump where it is just not shedding heat. Sure, it�s a brief enough period of sump time, but this increasingly hotter oil continues to circulate throughout the engine. There must be a point of diminishing return for max volume for our sump system design. Once the aeration issue is contained and the pickups are not sucking air, like it appears to be in zbomb�s case, the next hurdle is cooling. Hopefully digging through the data will reveal areas with �ah ha� potential answer, with regard to an optimum oil volume to cooling capacity range. I wish I was smart enough to compare the dips from both years and plot the temp/pressure/rpm for those troughs and come up with an answer. There has to be a ratio for lubrication system volume/flow to cooling system efficiency with respect to oil temp optimum range. I just can�t find these equations anymore (ha, or solve them).

Quotes source post: 18243910

Yeah, I just don't agree with this. While cooler oil would be better, oil temps in the 240 range are not out of what I would consider relatively normal track operating conditions for a modern engine and oil combo. The key, in my opinion is keeping the pickup fully submerged in oil AT ALL TIMES through whatever means are required to accomplish that goal. I do not believe cooling can get you there, ever. We would be talking about oil temps in the 180's if I remember from back in the thread, If someone can develop a solution to keeping oil temps in that range, even when ambients are in the high 80's or 90's, than I will be glad to eat all these words. I know there are some examples in this thread of cars that accomplished this but from my results on my car, I don't see it happening but maybe I am the nail that sticks out...

zbomb - I acknowledge in my post that non-aerated oil has to constantly make it to the pickup…period…or game over. I wrote that once the pickup part of the system is good, then temp control is next. All things equal, If the pumps are working and there are no mechanical weaknesses for pressurizing the system, temp control is the next piece.

Quotes source post: 18243985

If the oil pressure never dropped unrelated to RPM, why would you need to spend money and likely add complexity to bring the oil temps below 240 or 250 if they held there with relative stability ? What would you say the targeted temp would be and why that temp and not some other ? All I'm saying is if you bring them down to 220 or 230, that's not solving the pressure problem and if its not solving the pressure problem, is the 20 degree decrease in temp actually doing anything you NEED done. I don't know the answer.

Good topic. Big difference between a 25-30 min session when oil temps hit 250 and an Enduro race when oil temps hit much higher levels. With 3 quarts of extra oil with a deep sump and UAOS, oil temps will be contained for 20-25 min, then the fun is said to start with oil temps swamping water cooling. We actually see this in Jason's video when oil temps were above 250 and water temps were under 220. So what are we solving for? Does the temp of oil influence oil pressure instability? Does a third radiator even matter when controlling oil temps, or is it a waste of $1k? How high do oil temps go after 30, 40, 50, 60 min on the track and how much does oil pressure drop ?

Quotes source post: 18243800

Yes, its one of the cool things about the "Integrated Dry Sump" of the flat six design. The crankshaft is mounted really high inbetween the cylinders of the flat six, but still has a very low center of gravity unlike typical "V" type engines. Some "V" type engines may even be lighter than the flat six, but the higher center of gravity makes then less able to corner as effectivly.. But Porsche didn't finish the design of the "Intergrated Dry Sump" , the shape of the sump is terrible and is what leads to all the oil pressure instability issues. The Twin Tunnel Port/UIDS design is the only way to correct this, you will see. Here is a schematic of the M9X oil level in green. You can see it is WAY away from the crankshaft when full level oil at rest, and even further when the engine is running..

After thinking out loud, did more research. Saab Scania did research in the mid-2000�s on oil cooling and pressure relationships. They deduced that temperature alone does not play a statistically significant role in oil system pressure function. Horizontally opposed wet sump or hybrid sumps all have system pressure vulnerabilitie s. Doesn�t matter if it�s on a truck, motorcycle, airplane or car. Some have issues with high pressure, some with low. Chris Cervelli (CTS) did lots of research on pressure drops. He concluded the decrease was due to uncovering pickup/aeration. Granted this is on a 2.7, but orientation and location of engine isn�t going to skew the results. His mid-engine dry sump solution stopped pressure drops. Unfortunately, the application is not suitable for our engine location. By the way, Cervelli even made a transparent sump pan to see how much oil is actually in the sump. Answer was very, very little at high rpm. Every type of M96/97 engine experiences oil pressure drops under nearly identical dynamic ranges. No combination of sumps, baffles, scavenge pump, brand or viscosity of oil has provided a solution. Porsche completely redesigned the oil system for the 9A series, doing some very cool bench testing to come up with an answer for the follow on to the M96/97. Skip is telling us the scavenge flow rate is good enough? If so, the pump system on our engines should be sufficient as-is. Jason�s data tells us many things. Props for his efforts. It appears his data shows that these pressure drops can occur independent of rpm. As Jason points out, the pickup needs oil coverage 100% of the time. To me, the only way for eliminating oil pressure instability in our engines is an engineering solution keeping the pickup below the aeration line, no matter what. A UAOS w/drain back and conventional deep sump with baffles are not fixing the issue. Does the UIDS do that? Seems that it should, by design. Also, the UIDS sump has cooling fins and can take an off-track excursion or concrete smack without destructing. The survivability of an off-track excursion with previous deep sumps hanging low on the car was a risk Cervelli was not taking for his race cars. So shouldn�t the combination of UAOS and UIDS provide the answer for keeping the engine alive and healthy through that third track session on Day 2? And yes, oil cooling is important to prevent critical temp spikes adversely affecting oil lubrication (not oil pressure) performance when the engine is pushed to 10/10 for 20 to 30 minute track sessions. The fins on the UIDS should help some for sump cooling between sessions. The LN oil cooler will be nice for a layer of temp control for the system if it becomes available. So, long story short: Theoretically, a properly baselined engine with a UIDS, UAOS with drain back, CFS/3rd radiator setup combo, should be as close to bulletproof for DE or racing as possible. �Critical� oil pressure drops and instability should become a thing of the past. Has anyone gathered data with this equipment package in 996 or mid-engine worlds? When that data becomes available, we�ll know. That could be golden goose stuff.

Quotes source post: 18245281

Please don't stretch what I'm saying about the scavenge pumps too far. I still see plenty of improvements that can be made to the scavenge/return system. But unfortunatly they are not improvements that can be easily or cost effectivly made. Yes, the Scavenge pumps are capable of returning 38 gallons of oil each per min at 4k RPM, whick means one scavenge pump is capable of returning almost 4 times the oil that is leaving the sump, but there can still be a lot of oil "hiding up top" that doesn't make it to the scavenge pumps ( under certain curcumstances ) and therefore doesn't make it to the sump. That is why managing the oil in the sump with it's terrible shape is very critical, and can't be done by a single center mounted pick-up..

Hatchet, good summary. Gotta agree, that the 2 qt deep sump and UAOS seem to mitigate quite a bit (but not all) of the oil starvation and oil cooling issue for a 20-30 min DE session. But the UAOS plus UIDS with an oil cooler should ultimately win the prize. Quite frankly I look at the oil pressure/starvation issue pretty much solved with the UAOS and UIDS. It's just a matter of seeing the numbers on the track to confirm. However, oil cooling has not been solved yet. Once LN rolls out the oil cooling kit, the question will be were it fits on the car. How strong is it, and if we will need a 3rd radiator or not. But it's only a matter of time until cooling is solved. And, it's only a matter of time until we will have nothing to talk about on rennlist except bore scoring and what radio to put in the car.

curious - what is the solution implemented in the 9A* series engines?

Quotes source post: 18245374

Which is demonstrated by Cervelli’s experiment with the transparent sump plate. If there’s no oil in the sump, there’s nothing to pick up.

Quotes source post: 18245406

Quite simply put the 9a1 is a square sump instead of a rectangle ( UIDS makes the m96 sump shape into two squares left and right and switches between them via Twin Tunnel Ports, with other design enhancements). The 9a1 square sump was baffled using this 1 million dollar machine.. You Tube

Quotes source post: 18245420

I paid close attention to Chris's comments, there was some very valuable info that he shared. He was very uncertain to how much oil was in the sump even with the transpatent window. There is so much splashing/movement that is was hard to determine exactly how much was "good oil" and how much was just splash/foam. He did suspect it wasn't very much. I also belive there is only about 2 quarts of "good oil" in the sump at high RPM (and the rest is just areated foam) based on all the data that clearly shows that the instability starts at around 1G normaly. Calculating backwards from what it would take to keep pressures ideal below 1G , the 2 quart figure seems to be accurate and why I chose that for the bench tests in the slosh tank... This is why it is important to manage what's there effectivly, and why the UAOS with hm/XTDD makes a big imrpovement. The higher oil level/capacity allows the Scavange pumps to have more oil to return and gives the rest of the oil less turn-over rate and more residence time...

I just wanted to add an update as to consumption - in 1 HR of track time the engine consumed just under .25 QTS, pretty normal from what I have seen with the drainback. My plan for Sunday is to up the level to 12.25 QTS approx and we will see what we see. And for those that like to tie data to video, here is the video from the day with which the last data Paul posted was taken. You Tube

I also don�t understand the temp issue. Not the same thing, but I saw a heat exchanger on a 29-cabinet, multi-megawatt supercomputer that was only about 14� x 14� x 8� in size. Water to water, not that different from an oil to water heat exchanger except the fluids and different of course. In any event, you�d think the stock heat exchanger on this engine is �good enough� as long as you remove heat from the water adequately. Sure, hotter oil may aerate more, and it�s going to be thinner, but still, it seems to me, the main issue is at the pickup. Solve that problem first. And yes, seems to me, all you need is UIDS and UAOS. And then to bring down oil temps, I�d do some mods at the radiators to bring down coolant temps.

Mike, good points. My understanding thru the Chicago club racing community is that a third radiator helps but it's not uncommon for oil temps to hit 290 during a hot summer day race and engine failure becomes a reality. Club racing community is dying for a m96 oil cooler. LN rolled one out for the cayman/boxster last year. Tried to fit it in the 996 and there was not enough room. It can possibly be run if you remove the mufflers on the 996 with a little fabrication. The oil cooling system is the last piece of this puzzle. LN will eventually get it. Maybe someone on this thread has configured the cayman oil cooler on a mufferless 996 and has oil temp and oil pressure data. Be good to see. Same can be said for those that have data on the cayman with the ln cooler, uaos, deep sump, etc. For us DE guys that don't run enduros but instead run 30 min sessions, my wag is that once the m96 oil cooler for the 996 is ready for sale, we won't need a 3rd radiator. Just a UAOS Track version, uids and oil cooler. And we will all be happy as hell.