AOS modification.

Hello all! I have had a 996 for about 7k miles now requiring an AOS replacement. I do not know how long this AOS was in use for, but I have read threads covering the issue and talking about how frequently it happens. I wanted to know how the AOS works. Does the oil in the chamber fill up to a point where it floods the membrane? Even if the oil fills up to the bell-mouth, it affects the airflow and hence how pressure is applied to the membrane. I have been thinking about adding a one-way flow valve to the midsection of the AOS, connected to a small canister that can be drained after some use, and I would like to know if it would help the situation. I would like the community to chime in and tell me their thoughts on this. Thanks! I have attached a link to a 997.1 thread that has a cut section of the AOS. I have assumed that the 996.1 AOS would be of similar construction. 997.1 AOS

The AOS is really just a polite and eco-friendly replacement for a draft tube, vent, or catch can as a sealed system. If you never want another AOS failure, plug the vacuum side and run the oil suction side into a catch can that then vents to atmo. Or get the motorsports AOS.

You can reroute the AOS vent line to the air filter box. No more excessieve vacuum (at zero throttle or idle) applied to the AOS diaphragm. The AOS will stay always in open position and never fail. In fact in theory this is also how it is done on the gt3 models. There the vent line is not subjected to huge intake vacuum as in the NA m96/97 engines. No more greasy throttle body, no more oil and gunk in your intake. Only drawback is the air filter gets dirty a bit more at the spot where you let the rerouted line enter the air filter box. Further advantage is a colder air intake temperature as hotter AOS fumes are rerouted and have more time to cool down. Makes for some extra noticeable bhp. Engine runs smoother at idle as no unmetered air gets into the engine. As said before the main advantage is no more oil in your intake. In case of an internal problem which causes the AOS lines flooding with oil and the positive crankcase pressure pushing the oil out, it ends up in the filter box rather then the intake to possibly cause a hydrolock. You could put an additional catch-tank to filter out any oil getting past the AOS. Saves you some cleaning of the air filter box every time you change the air filter. I have this set up and it saved my engine half a year ago when my head 1 scavenging pump gradually lost its pumping capacity. At some point the AOS spitted out oil every right hand turn I made. In the beginning just a tiny bit. But got worse and worse and the car started smoking of the oil ending up in the air box and starting to drip on the hot exhaust, I did not have an additional catch tank at that time. At some point I connected a small bottle to the line and only one right hand curve at moderate 35mph resulted in 200ml of oil in the bottle. Imagine that going into the intake. Hydrolock festival It turned out the scavenging pump in head 1 was not able to pump the oil back fast enough and the AOS vent line in bank one got flooded with oil which found its way out via the AOS due to the overpressure in the crankcase. To prevent the engine from this overpressured cranckcase scenario. There is now a second connection from the oil cap to the additional catch tank. So any overpressure in the cranckcase when the AOS gets flooded can escape via there. if you analyze the setup of the 997.2 and it�s oil filler tube you will see it is exactly the same what Porsche has implemented in its later models. As well as the rerouting of the AOS line to a �lower vacuum at zero throttle�-position on the intake. It makes sense, not reinventing the wheel again.

This is a really useful post. I always thought there should be something akin to a catch-can setup that would prevent the AOS from being an issue, but I never took the time to track down the vacuum / vent line routing to see how it was plumbed together. Thanks for the post.

Quotes source post: 15569721

Very interesting. Is there a write up on doing this mod with more detail and pictures?

Quotes source post: 15570151

Also interested in what, if any, impact this would have on emissions and if you could get away with it on a car that is subject to yearly inspections.

I�ve read something about long term fuel trim being an issue with a catch can on these cars. Something to look into.

Quotes source post: 15570203

Can you post a link?

Here it is http://986forum.com/forums/boxster-g...ket-aos-4.html

I always wonder if the AOS can be moved to make it easier to replace in the future.

Quotes source post: 15570203

The AOS under driving circumstances (diaphragm open) is in fact a catch can, so I fail to see the influence on it during normal operation. Adding an additional open flow catch tank just helps to catch and condensate any oil dispersed in the crankcase fumes which gets past the AOS under extreme driving circumstances. Remember with the setup as explained above you completely take away the excessive vacuum created by a closed throttle body in combination with high rpm. The diaphragm is always open as it should be ideally and it will never fail. Unless there is another problem within the AOS you do not need to replace it. I understand the probability that in the U.S. you might not pass vehicle inspection. In theory it should be okay as the air is recirculatid to the intake only via an alternative route.

Quotes source post: 15570358

Thanks! 3 years later now, I wonder if the issue has been fixed?

Quotes source post: 15570447

Seems like a catch can plumbed between would be any easy fix to catch excess oil but that sounds to simple.

Quotes source post: 15570429

Did you read the thread in the link I posted?

Thanks all for chiming in. And thank you Joey for sharing the link to Stefans work. I was thinking of integrating a removable catch can design to the OEM part. Stefan mentions that the bellows deteriorate due to exposure to acidic oil. I think adding a removable catch can to the existing separator chamber via a one-way vacuum shut-off valve will ensure that the AOS is always under the intended design condition, but will simply drain off the collected oil after each drive when there is no vacuum in the separator chamber. As the AOS works under intended design use, the operation conditions do not change at all, and hence there is no need to worry about the fuel trim. The device will keep functioning as stock, but gives the flexibility of draining out the oil after use. Draining the oil will ensure there is no change to flow characteristics inside the chamber, or oil churning up to turn acidic. Good oil maintenance practice, which is generally expected, will ensure that the bellows are not exposed to acidic oil. The one-way valve will also ensure that the oil from the additional catch is not sucked back in. I certainly have not had a lot of time with the vehicle, and I have not spent enough time understanding the situation. But let me know what you guys think about it.

Silk your post was fascinating, hopefully an aftermarket company will make a kit for it one day!

All you need is a hose. Disconnect the AOS line from the plenum and reroute it around through the engine compartment to the air filter box under the air filter. During driving it will get enough suction from the air box in addition to the positive crank case pressure under driving to breath the crankcase fumes. The extra catch tank is not necessary. It is just a neat solution to not dirty the air filter box too much or catch some oil if you have another internal engine problem which pushes oil out through the AOS. The connection on the oil cap is also in case such scenario happens and is there to let any overpressure escape. Even Porsche understood this and implemented it on its later models. As you can see above the oil filler neck has a connection to the AOS (the big ribboned tube with the 90degr connection ont the oil filler neck)

Will be looking more in to this.

Quotes source post: 15569721

Has this ever been long term tested on a 996 to see if the theory works in the real world?

I have it 45.000 miles on my car. Saved my engine from massive hydrolock when my head 1 scavenge pump deteriorated and flooded the AOS. So in that respect it works. My indie who developed it has installed it on more then 250 cars by now.

Quotes source post: 15575865

Cool, that's a pretty good sample size.

Quotes source post: 15575865

Would he do an instructional write up with detail/pictures for us DYI Yanks?

Was there ever an instructional write up for this idea? Any one done it yet?

Subscribed

I hope I'm understanding this correctly since I need to get my AOS replaced, could I do this instead with no downside ? Seems like it would be cheaper and a better option.

I am thinking the same thing. Is it an easy DIY or just instruct my Indy how?

The air box idea sounds great. I think a non vented, inline oil catch would still be needed. Depending on how much oil people are getting, they may start fouling MAF sensors.

Is there a way to just move the AOS to a more convenient spot? That would make the whole thing a non-issue

Would assume that the AOS developed for the 986 engine referred to in an above link was a fail? He made 10 units. His last post on that thread was 11/16. Seems like the fuel trim issue was insurmountable? Sounds like Silk's method to reroute the vent to air box is a good idea.

Quotes source post: 15771635

Yes this would be great, especially if it was a DIY.

Quotes source post: 15771813

Plumb it in after the MAF?

Quotes source post: 15574910

Silk, Do you know how the hose attached to the air box? Is there an existing port to tie into or does he drill the airbox and install a vacuum port (something like this, below) to connect the hose?

Quotes source post: 15777431

Yes, you got it. You'd want a bigger nipple though. The two mounting pins under the airbox is where they are added.

Quotes source post: 15777803

Do you have inside information on this? I have not seen this. I'd have to look more closely, but I think those mounts are boxed-in. Might be difficult to run hoses to them. I just ordered one of these to take the place of the nipple I installed last week. Image does not quite match the description - it is 3/8" NPT to thread into the air box and 5/8" barb for the hose (image is smaller - probably 1/2") This will allow more flow. $9 on Amazon; material is aluminum. Got my order from US Plastics yesterday, when I ordered a bunch of stuff to replace the brass fittings in my interim fix. Turns out they are much too restrictive. Found a similar (HDPE) piece as below, for about $0.58 each. Shipping is $11.... In case you haven't seen it, I attached the outlet of the catch can as shown below. Uses 3/8 hose. Will be replaced with 5/8 hose when I get the next batch of parts.

Yes, Silk sent me the details and the picture of his set up that I posted earlier. Sorry, it should have read between the two mounting pins. I assume horizontally...

You need 3/4” fitting from the AOS vent line. I think anyting smaller is too restricted. I fitted my on the left side of the airbox just under the filter element. This was the only space I found usable with a 3/4” hose. the I used a 1/2” 90 degrees brass fitting on the oil filler cap and connected a 1/2” hose to the filterbox air duct. The clearance between the cap and engine room lid is very restricted.

Quotes source post: 15778125

Can you post a pic of your set up when you have a chance?

Quotes source post: 15773809

That defeats the entire purpose of rerouting it. If you route the aos line in the airbox, oil will settle at the bottom of the box. If you put it after the MAF, it's just going to get sucked right into the engine like it does in the stock location....

Quotes source post: 15778191

I was thinking with a catch can inline.

Quotes source post: 15778274

Again...what you're suggesting just defeats the purpose of rerouting. Might as well leave it in the original location with a catch can. Doesn't make sense to drill a hole somewhere in the intake tube to create the risk of an unmetered vacuum leak.

Quotes source post: 15778332

Understood. I look forward to hearing what you come up with. Seems to work for some...

The real issue with the AOS is the �massive� vacuum it gets exposed to when going of the throttle and the throttle body closes. To prevent this vacuum sucking out the oil fumes without the cyclonal separation doing its work it has this diaphragm. The diaphragm is not perfect, to some extent in extreme situations some oil gets past, furthermore it wears out and eventually fails gradually decreasing the AOS filtration capacity. The clue of the mod is to move the AOS exit to a less high vacuum position. The most suitable is somewhere before the throttle body. During driving there will be light vacuum to evacuate the cranckcase fumes via the AOS out. As the fumes are still dirty it is best that they do not directly come into contact with the MAF. It is very important that the AOS vent line is unobstructed by small diameter bottlenecks. By rerouting the vent line you move it to a lower vacuum position. You will need a big enough diameter for compensating for that. Otherwise you risk to have higher crank case pressure. I saw some pictures above of converter pieces. The diameter is way too small to allow proper venting. I would say the min diameter is half an inch or so.

Related to the picture posted above. A line is still going from the catch tank to the airbox under the filter to create enough under pressure during driving. Furthermore a catch tank is not necessary. You can leave it out to simplify the mod. It is just a more clean solution as without it eventually some oily fumes end up in the airbox and could dirty a bit the airbox inside. Eventually in extreme cases a bit the engine bay.

Quotes source post: 15778365

Not sure if you realize it, but Silk is routing it to the air box, which is before the MAF. If you reroute after the MAF like you suggested, you're only a few inches away from the original location. Hope that brings some clarity why routing it after the MAF like you suggested is a wasted effort.

I brought this up in Bill's blown (but not actually blown ) motor thread, but would like to discuss it more here in this dedicated AOS thread. Before I do anything with my AOS, I would like to toy with the idea of rerouting the AOS line from the intake to the exhaust. The unknown factor here is if the exhaust from a single bank will be able to generate enough vacuum to keep the AOS working properly. I'd still want to run a catch can, and ideally have the bung welded after the second 02 sensor. Oil vapor (not to be confused with oil mist) would then just be vented into the atmosphere. I have the equipment to measure and monitor the vacuum at both locations (plenum and exhaust), but the dang contractors renovating my house have their dumpster in front of my detached garage that the 996 lives in. It's the only place the dumpster will fit at the moment. Grrrr.

Quotes source post: 15778452

Yes, I got that. My brainstorming obviously went without much thought...Fouling the MAF has been on the back of my mind with this set up, but maybe it's a non issue? I like your idea and look forward to seeing if it works.

Then the only possibility after MAF will be just before the TB. Otherwise you have the original situation. You will dirty your MAF with �unfiltered� cranckcase fumes in that case.

Quotes source post: 15778474

It very well could be a non issue. The previous owner ran a k&n intake, and in his service history had to change the MAF two times after the "upgrade." I've since had to replace it once during my ownership, and then I swapped in the 997 3.8 intake. No MAF issues since. That being said, an oily k&n filter is completely different than this application, but it still leaves a bad taste in my mouth when it comes to putting something oily upstream from the MAF. A new MAF is heck of a lot cheaper than a hydrolocked engine, so I will move it there if there are no other viable alternatives.

Quotes source post: 15777431

there is a full size hole in my airbox and the rubber hose is squeezed through. This has too small all diameter to keep the airflow going

Quotes source post: 15778499

The air filter keeps most greasy dirt out. But you are right. Eventually it could dirty the MAF. I noticed some degradation on mine after 30.000 miles. So I just replaced it. As it is one of the most crucial sensors in the whole system, nothing beats the performance kick, back to original, as a new MAF. I see it anyway as a wear item. Cleaning MAF is just temporary effect. They all eventually degrade. You should be surprised how much stronger all of your cars wil pull with a new MAF (given it was a long time ago replaced). A new MAF is sometimes one of the cheapest �performance upgrades� you can do.

Quotes source posts: 15778382, 15778389, 15778519

I will assume you're referring to the picture I posted. My plan is to go with larger fittings and hose on the outlet side. Your later suggestions (below) are excellent. I like this idea. Simplifies everything. One hose directly from the AOS vent to the air box is brilliant. The condensate will simply roll to the other end of the box and out the drain holes. Great information. Thank you for revisiting this thread and providing your valuable insight.

If this is a solution why do we need an AOS?

Quotes source post: 15777803

I would not use anything with a smaller diameter then 0.5inch. you really need a free flow without bottlenecks to prevent pressure build up by the naturally blow by of the engine. In the originaliteit situation with the butterfly valve in the TB open the AOS in full operation gets a higher vacuum/suction (also due to the venturi effect of the incoming airco�s at the vent line location at the plenum) for the cyclonal separation to do its job then at the relocated place in the airbox. That is why the diameter of the tube should be not too small and also not too big. As the air flow will need to bridge a longer distance to a somewhat lower average vacuum position. You will want the air to be moving freely and but still fast enough. Here lies the inventive part of the modification. To get the airflow right there to keep the AOS in its operating window. I did not design the modification so I am not aware what the special requirements are. The extra venting on the oil cap is just a safety in case something gets blocked. At least it will prevent for some too high pressure built-up in the cranckcase.

Quotes source post: 15779342

You will need the overpressure of the cranckcase out and the AOS keeps functioning. It will separate the oil from the outgoing airflow. In fact you need to look at the AOS as a smart catch tank with a built-in shut of valve to close it when subjected to too high suction force from the engine intake. With this modification you will take this excessive vacuum situations away. The AOS wil stay open all time. No more problem with a failing diaphragm. I am not saying that the modification is the ideal solution. But for me it saved my engine at least already once from massive hydrolock. The previous owner of my car needed to do a rebuilt due to a cilinder scraper and a small twist in the cranckshaft. Probably also a hydrolock issue... Besides oil the AOS could spit out a lot of dirt up till solid pieces which collect in the plenum. So there is more to it then hydro-locking alone.

Silk, Sorry if this is mentioned already but what is used to block the intake port in your setup?

Leave a piece of tube connected to the plenum and cap it off. Used once a screw which fitted neatly in the tube and used a clamp to secure it properly. But feel free to use your imagination.

Quotes source posts: 15779627, 15779686

It looks to be capped off (inside yellow ring). This is very similar to what I did with mine. Will now change mine slightly, to get the AOS out directly to the air box.

thought this would be appropriate to post... You Tube I will probably add a catch can on my new engine build if I can fit it in there.

The diameters of the catch can connections is of utmost importance. Those shown on the pictures are too small. No free flow secured. You probably will need to modify an existenti�le catch tank, drill wider entries and fit custom imkers and outlet with appropriate diameter. All my catch tank catches is condensation.

Quotes source post: 15780222

I bought that same "$100" catch can for $18 on Amazon. Agree with Silk's comments - it needs larger ports.

This radium engineering catch can looks like it should fit the bill. http://www.radiumauto.com/Universal-...-Can-P178.aspx

Or just search for "0.5L baffled brushed aluminium oil catch tank with 19mm or 3/4" barb. Magnetic drain plug " on Flebay - under $40

MANN Provent. Have used this on my 928.

Quotes source post: 15784481

Would you be willing to share the part number?

Amazon Amazon what is inside? You Tube Similar on Ebay for 1/2 price.

https://www.ebay.com/sch/i.html?_nkw...&ul_noapp=true

Hmmm. Very tempting. Smog inspector wouldn't even second guess it.

Can someone summarize this for the masses? Does this replace the AOS, mitigate a failed AOS, or provide a safeguard against a future AOS failure?

The challenge seems to be to find a catch can that fits the confined space, has a dipstick and an accessible drain valve and has 3/4" ports. It also needs to look stock so it passes the visual test for Smog. To satisfy most of the criteria on a different engine I used a stock can from a 2000 M62 4.4i BMW X5 .It is unusually small in diameter and very long. It is black plastic and cost about #20 on Flebay Maybe others can find a better alternative: The BMW part # is : 11151705237.

Quotes source post: 15786421

I am trying to piece this together myself but as I understand it, it mitigates the effects of a failed AOS and is insurance against a catastrophic failure. When the AOS fails during operation as it did to 808Bill, it can dump a large amount of oil into the intake. When this happens your first instinct is to pull over and kill the engine. This can leave a large amount of oil in - usually cylinder 4. The next time the car is started that oil can have the cylinder hydrolocked. As Jake pointed out in Bill's blown motor thread, just cranking with the starter can bend the rod on cylinder 4. Jake also mentioned he has seen a bent rod then cause the piston to perforate the cylinder wall. The mods discussed here will send the output of the AOS to either a catch can, and/or into the air box on the other side of the air filter, far upstream from the intake. I am understanding that this prevents the amount of oil being dumped into a single cylinder that would be required to hydrolock it. Therefore, you would possibly get a soaked air filter and/or relatively harmless smoke. Can someone verify that I understand this correctly or please correct me? Thanks.

Dana, that sums it up really well from what I understood. Fix/replace your AOS and ad a safeguard to the system!

Quotes source post: 15786480

Can you post a pic of your install and describe how you plumbed it?

Quotes source post: 15786666

Thanks. AOS replacement is an engine-out procedure for my car, so it's not something I care to have done very often. This sounds like a good bit of insurance between replacements. It would great if someone could put together a DIY with photos and parts description/numbers.

Quotes source post: 15786967

It's pretty much all here in this thread... Very simple, you have to move the AOS vent line out of the high vacuum trajectory from after the throttle body to before. I did not do it myself as I had it installed as a package with IMS replacement and a cooling modification (relocated Low temp thermostat, GT3 waterpump and reprogrammed high fan settings with an extra temp sensor on the cooling water leaving the rads) But the DIY is as simple as this: 1. Disconnect the AOS vent line from the intake. 2. Cap off the open intake connection. 3. Reroute the AOS vent line accross the engine bay to the right side 4. on the right side of the air filter box max a hole so the rerouted AOS vent line fits tighly through and sticks in the airbox under the air filter 5. make a hole in the oil filler cap, or use a 997 oil filler cap (more sturdy). secure an elbow fitting in it and connect a vent line which ends up the same way in the airbox under the air filter. Step 5 is not necessary. It is an extra safeguard against overpressure in the crankcase for whatever reason. If you check the later 997 DFI setup you will see such a vent connection on the oil filler tube Additionaly you can mount a catch tank in the right side of the engine bay and pass the two lines through it. It gives some additional filtering if dirty oily air gets past the AOS. When using a catch tank and rerouting the vent line the only thing to take into account to not choose a too small diameter, the engine needs to be able to get the blow by out without any bottle necks. All the above is kind of straight forward. choosing the right diamter pipes is the inventive part. I have no idea where the sweet spot is.

Quotes source post: 15786666

This is a great summary, I can never articulate as clearly. I either am too technical , or too general , or jump back and forth from too technical to too general and end up sounding like gibberish. Good job !!

Previous attempts at catch cans failed because of Air/Fuel ratio problems created by the venting. It may be wise to plot A/F ratios before and after installation to ensure you haven't fixed one problem but created another. It may be significant that despite all the expensive AOS problems for all these years there is no commonly accepted ,simple,economical fix yet. On other engines the catch can fix is successful and has been for widely used for many years. There has to be a reason it has not been successfully implemented and widely adopted mod for our engine.Let's hope the proposed fix works.It will be a very welcome discovery for many of us. For these reasons it may be better to fit a catch can in addition to the stock AOS .Previous failures were trying to replace the AOS with a catch can.

Quotes source post: 15787099

Sound advice!

Quotes source post: 15787099

Just for clarification Schell, were previous attempts done with a catch can that vents to atmosphere, or vented to intake vacuum?

The whole project with drawings should be in Search under the author "Stelan"

Got a link? I searched and found nothing.

the previous attempts being referred to are from 986forum where a member was trying to put an aftermarket aos/catch can in place of the oem unit. jake noted in the thread that his efforts to do the same thing were unsuccessful because of fuel trim issues. i don't think you can extrapolate that to 'catch cans cause fuel trim issues'; gotta compare apples to apples. the oem aos vents the crankcase through an air/fluid separator into the intake. it goes to the intake for emissions purposes - to try and combust the nasty vaopurs coming out of the crank case. further, porsche uses the vacuum present at the intake to help evacuate these vapours. speculating, but presume that the vacuum also helps maintain airflow to ensure that any remaining fluids in the crank case vapours remain suspended in air so as to not deposit on the intake tract and combust better. there may also be the consideration that the low tension piston rings used by porsche require a vacuum to function properly, and that a vacuum helps limit aeration of the oil in the sump. otherwise, porsche could have introduced the aos output at a number of locations that didn't have vacuum; the charcoal cannister for fuel vapours immediately comes to mind. stop for a second and think about the pcv valve on the crank case; without vacuum facilitating flow through this valve, will crank case pressure be adequate to keep the oil and gas-contaminated vapours moving through that thing; is that sufficient to keep it from gumming up? a running motor can produce a lot of vacuum, and porsche wanted to limit the amount of vacuum applied to the crank case so the aos has a diaphragm - as vacuum increases, the diaphragm closes so as to reduce flow. this diaphragm is the failure point in our aos and why a failed aos is identified as a very high vacuum on the sump - so high that it is difficult to open the oil fill cap. again, porsche could have avoided this failure mode by venting the aos someplace different but they didn't, which again leads me to believe that vacuum is required. they could have also removed the diaphragm all together but they didn't, which is an acknowledgement that the amount of vacuum applied to the crank case has to be managed in some manner. replacing the oem aos with an aftermarket aos, a catch can, or an aos with a higher flow diaphragm will cause fueling issues. situation: you come off wot, throttle closes, engine doesn't want more fuel but is still spinning quickly and crankcase pressure is high. further, vacuum at idle is high and fuel/oil vapours get dumped into the intake without a diaphragm to regulate flow. fuel trims go rich. this is not the situation being discussed here. in this instance the aos is not being replaced, but a catch can/aftermarket aos is added after the oem aos as backup. so, diaphragm still in place, and second catch can serving as a 'canary in a coal mine' to aos failure. the concern with this approach (as jake notes in the previously mentioned thread) is catch can volume; too small and it is useless. further, some consideration as to where you vent the secondary catch can; i'd suggest back to the original location so as to maintain the vacuum as dr ing intended. concern here is getting a big enough catch can to be of any use. you could also vent to the air box as is being discussed here - here you have enough volume in the air box to mitigate any disasterous aos failure and might not even need the secondary catch can, but concerns are (a) no vacuum, and (b) oily vapours fouling your mass air flow sensor. if an oiled k&n filter can foul a maf (it did mine) then crank case vapours certainly can even with secondary separation. hey, those who have done this successfully; perhaps you are not fouling your maf because you pcv valves are plugged! ultimately, porsche is aware of the issue and has been slowly and quietly upgrading the aos over the years (like the oil pump piston and chain tensioners for example). get the latest part number (ie, not nos from 2001) make sure it is an oem part and replace it. perhaps consider it a maintenance item that porsche didn't identify in their original service manuals, like water pump and ims. and speaking about posche upgrades, look what they did for the aos on the m97.2 (dfi) engines; same as on m96 and m97.1. that's a tough place to try and engineer an improvement (serviceable diaphragm would be nice ...). or i could be wrong. you get what you pay for on the internet.

Quotes source post: 15790992

post #9 in this thread.

just google "Stelan AOS"

The damn "rabbit hole" strikes again!

If you vent the aos rerouted line under the air filter oily fumes are less of an issue as the filter takes most of it out. But sure, keep into account that you might have to change the MAF at some point, probably a bit earlier. But i consider a MAF as a wear item. It slowly degrades. You are right about the right amount of vacuum. Not only for the cranck-case and rings. But also to get proper flow for the cyclonal functioning/separation of the AOS. The idea of the rerouting with proper big enough diameter tubes is to get as much necessary vacuum to the AOS as possible. I did not design my current setup. But I know at least some sort of research has been done towards the issue of pressure and vacuum. At least we agree that there is a purpose for the vacuum. But the diaphragm is also indicating that the position at the intake is not the ideal. Too much vacuum. Even Porsche did not like it. And I can understand that they did not want a lot of extra tubing in the engine bay at odd locations being prone to wear and failure. For Us as enthousiasts we take care and baby our cars. Much different then the first time owners of our cars!

and i think that is a great solution. i would watch my air filter like a hawk, as once it saturates, even in one spot, fluids will pass right through it. are you comfortable that you are getting enough vacuum there? i think you'll see pressure drop there (air constricts at the entrance to the air box, expands within the air box and pressure drops, constricts again as it enters intake pipe). you note in your initial post that the longer hose acts to cool the intake charge; i note that porsche actually heats the intake charge (that's the two other connections at the diaphragm). oddly the boxster doesn't. are you worried about condensing oil/fuel if you cool it too much?

Quotes source post: 15791059

The only challenge with this is all the reported failures of brand new, OEM AOS units shortly after installation. I seem to remember (and I may be mistaken) even Jake mentioning experiencing these units failing from brand new himself. I'd gladly replace mine with new, but presently it is working and I worry about actually be worse off with new.

Quotes source post: 15792014

Yup, I actually considered pulling the one off of my other motor (only 15K on the AOS) and using it. But ordered OEM and hope to have it by weeks end. I really need to get this fixed and behind me...

Quotes source post: 15791850

Running already two years and a lot lot of miles and going strong. Also more then 250 modified cars driving around here. The long term effects however are not predictable. That is true. But once again it saved my ***, euhhh engine, already once (at least that i am aware of). You are right, dont forget that at the intake you have also the venturi effect, which I think due to the perpendicular fitting on the intake and with throttle body opened is the main vacuum exponent working on the AOS. I did not do any measurements. Maybe someone can do it to see what kind of magnitude of vacuum we have there under the air filter . At least we agree that too big vacuum is not good. Is it confirmed that under driving conditions the engine is under such a significant vacuum? It sounds really strange to me. A healthy +3.0L engine has a considerable blow by. Would the tiny diameter of the AOS line be enough to compensate for this, and on top of that create a vacuum? My intuition says it is just to create a sufficient flow get the crankcase fumes out and get enough airspeed for the cyclonal separation the be efficient. And yes to lower the average crankcase pressure caused by the natural blow by. Okokok ... related to the statement right above ... I admit that my practical technical knowledge about engines is rather limited ... and I did not research all the previous relevant threads here on this forum. But for me there remain a lot of questions unanswered. How long does the AOS stays open? Does the diaphragm closes also under other driving circumstances. What level of vacuum is required to close the AOS? The diaphragm will also close under (wide) open throttle and in high rpm range when the air speeds are at max? Heating the AOS serves another purpose. With really cold weather the AOS will not heat up enough due to its location exposed to driving winds. If it stays too cold it loses its cyclonal filtration capacity as fumes (water / fuel?) will condensate too fast and disturb the effective functioning of the AOS. The Boxster engine doesn't need that active heating as its AOS is located at a spot which is not exposed to the (cold) driving wind. It will come up to temperature by itself due to the surrounding engine heat. I checked my fuel trims, they are perfectly in line with what to be expected.

Quotes source post: 15792731

Silk, so you didn't actually develop and install your "system", but rather someone over there is doing these mods for people? Is this correct?

Yes correct.

I wonder if this individual has ever considered packaging the mod into a kit that he/she could sell to DIYers?

He has his own shop, specialises in M96/97 engine . He has considered it but you know how it goes with liability, bad installation and possible unfair bad publicity. I understand his point of view completely.

For those having issues finding the Stelan thread... STELAN Aftermarket AOS

Quotes source post: 15792781

I understand his view completely, the 996 market is such a niche and fickle market that it doesn't make sense, let alone the liability. FSI and L&N have managed to successfully do it, but have had to go through hell and high water. An AOS redesign also has the burden of the EPA to be a "monkey on the back"..You have to take your knowledge from this thread and DIY (take your own responsibility) or leave it alone and OEM (and take your chances)..IMHO

I can ask if he wants to make some kits and ship it to you overseas. No idea about pricing. He uses a modified catch can. And makes his own mounts. This extra labour and steps adds up to price. So I would not be surprised it will cost double the amount of the parts only. Without the catch tank it is plain simple.

I'm still on the fence as to putting a catch can between my new AOS (next week) or just leaving it stock. All my driving is short distance daily commuting...

I’m wondering if any of these mods has a chance of passing emissions testing, for those of us living in areas where it is required. In my area it is an annual inspection, and the first step is a visual check that all emissions components are in place. Even if there is no CEL illuminated, the capped off plenum hose is a dead giveaway that the system has been altered. Maybe a rookie inspector wouldn’t notice, but any visual check by a knowledgeable one would discover it.

If the mod affects the fuel trim you'll probably get a CEL anyway. The BMW X5 'catch can' looks stock and is so slender it is less likely to be noticed. Maybe do the cheapest/simplest/least conspicuous version of the mod and check fuel trims with a scanner/Durametric. The mod would be easily reversible. assuming this is an addition to the AOS and not a replacement. The really compelling application is that your AOS fails and you want a temporary fix until you have time to replace the defective AOS. If that is feasible w/o fuel trim issues it would help some folks?

Quotes source post: 15793700

Easy enough to put back for inspections if needed. I don't need the inspection here.

Quotes source post: 15792781

Yeah, I get it. Shame though

What's tied into the AOS tube to the intake manifold?

Quotes source post: 15795875

That's a fuel vapor vent line (I assume from the charcoal canister). There's a thread (that you have visited) that describes this stuff....

Quotes source post: 15795881

Yes, but I can't remember it and have been driving to work (6 mile each way) with it all disconnected (waiting on my AOS)...I've been overloaded with info these past two weeks (with the flu) Am I causing more damage?

Quotes source post: 15796013

I get the overloaded thing. Sometimes life just keeps pouring it on. Hope you're feeling better and get some time to relax.... I doubt you are causing damage. And by this, I mean the car and the environment. You will be releasing small amounts of fuel vapor and crankcase vapors into the air. If you can't get your new AOS installed soon, you may want to try something similar to my temporary fix (125,000 mile thread). Don't sweat the little things.

Quotes source post: 15796025

I am feeling better after 2 weeks of misery, I had an allergic (first ever) reaction over the weekend on top of the worst flu (and had to work) I've ever had. I also wanted to wait on hacking my AOS line until I replaced the AOS and was committed to the catch can.

Quotes source post: 15793233

Anybody running this on the 996 with positive results? Mike

Quotes source post: 15798272

Have you read through the post and the links to others?

^^^^ No. Its far too long, lol. Cliff notes? Mike

Quotes source post: 15799247

Lol. yes it is. Just look for the key posters and skip the rest. Lots of very good information. The kit was for the smaller Boxter motors and less than 20 units where made and sold from what it looks like with not much follow up. Fuel trims seem to be the concern, but some have been successful...

^^^^ I very much appreciate the summarization. Good to know. That's too bad regarding the fuel trims because the system appears promising. Mike

Certainly was an interesting thread which seemed to have hope of finding an alternative solution to replacing the AOS but unfortunately it didn't seem like this was a well proven solution. With mine being a daily driver, I have to bite the bullet and get the AOS replaced since my CEL is back on.

Quotes source post: 15801228

What is(are) the code(s)?

Here is the diagnosis and was told the AOS needed to be replaced "Has DME/engine control system codes 27 (cylinder 1-3, oxygen sensing range), 35 (cylinder 4/6, oxygen sensing range), P1124 (system too lean at idle, cyl 1-3) and P1126 (system too lean at idle, cyl 4- 6). Most likely the crankcase vent valve/air oil separator is bad (it is also leaking some oil). "

Concerning fuel trims with relocation of the AOS outlet to the airbox. These are mine.

This was not a fun project!

Quotes source post: 15812964

What are those numbers telling us?

I am no fuel trim expert. But what I know is that: RKAT is the adaptation value near idle. These number you want to be 0 or close to 0. In my case: 0,14% and 0,33% deviation, considered as 0. FRA is the adaptation value under load and further divided into FRAU (lower load range / U = unter ) and FRAO (higher load range / O = oben) ranges. Ideally these values should be 1 or very close to 1. 0.95 till 1.05 is generally considered as ok A value of 1.03 means the ECU is adapting injection timing 3% to inject more fuel to correct a lean condition. A value of 0.97 would mean the opposite.

Quotes source post: 15813399

OK. Thanks for the updated edit. That helps a lot. Similar to my assumptions (below). My assumption is that percentages should be small (which yours are) and the FRA numbers should be close to 1 (which yours are, also). Are my assumptions correct? Finally the value 1 - is that the amount of "deviation" needed to maintain the stoichiometric ratio? I.e. 14.7:1 = 1? Maybe there's a link that would explain how this works (specific to the Porsche system)? I have more questions - but I'd rather not expose my ignorance on the matter more than I have already... Did a little searching: https://www.callasrennsport.com/blog...im-information

rkat is idle trims fra is low load trims frao and frau will always read the same as fra unless in a turbo with wideband o2 sensors the car does not trim at wot/open loop (no feedback from oem narrow-band o2 sensors to establish trims as ecu richens mixture when engine working hard) for rkat, values are given as +/- % fuel added, and you will get cell at +/-4% for fra, values are given as +/- fuel added from a baseline of 1, and you will get a cell outside of 0.7 to 1.32 your trims are fine

That is a very helpful,concise explanation.Thanks RK

I need to brush up on my Greek!

you mean geek .here's the less concise version ... the role of the ecu/dme/car computer, is to provide spark and fuel. for fuel, the ecu measures the velocity (mass/second, actually, which allows for changes in elevation) of air into the engine with the mass air flow (maf) sensor. the velocity is multiplied by the area of the intake tube to get flow of air (this is why the diameter of the maf holder is important). the ecu has charts where it looks up the air volume and adds an appropriate amount of fuel. there are four main types of charts that the ecu uses: - limp (maf is broken, so ecu uses throttle position sensor - tps - to determine fuel); - idle; - low load/cruise/closed loop/'taking it easy' mode; and - high load/wide open throttle (wot)/open loop/'giving her heck' mode. in low load the engine is not working too hard; it is cool, rpms are in the mid-range. in this mode of operation the ecu tries to add 1 part of fuel to every 14.7 parts of air. this is considered the ratio of air to fuel (afr) that is (a) most efficient, and (b) makes most power. in 'science' is called the stoichiometric ratio. the o2 sensors in the exhaust monitor the products of combustion and if the results stray from the 14.7:1 afr (ie, there is uncombusted o2 in the exhaust) then the ecu 'trims' the amount of fuel added. initially these are short term fuel trims (stft). if they persist then they become long term fuel trims (ltft). this the the fra number that your obd tool reads. the act of the o2 sensors modifying the fuel mixture is called feedback or closed loop. the o2 sensors used the m96 are called 'narrow band'; this means that they only operate in a small range around the 14.7:1 afr (as opposed to wide band which operate within a ... wide band). this system was designed to adapt to aging components; small vacuum leaks, drifting maf values, leaky or clogged injectors, drifting fuel pressures, etc. in high load the engine is working hard; it is hot, rpms are high, or throttle is wide open but not a lot of air moving (ie, going up a steep hill). as such, the ecu decides to switch between low and high load maps based on engine temp, intake air temp (iat) tps, rpms. in this case the ecu wishes to 'richen' the afr - add more fuel. this additional fuel serves to quench/cool the combustion chamber, otherwise it can get hot; detonation can occur and holes get knocked in pistons. typical high load afr is 12.5: 1 or so. the narrow band o2 sensors do not operate in this range of afr, so there is no feedback; this is open loop operation.

What a great forum this is with some fantastic members! Thanks for taking the time to explain all that...I learn something new everyday here.

^^ ++ Thanks for taking the time to explain all that.

Is the following correct ? The significance of these Fuel trim results is that 1. the :Silk" system has no fuel trim problems 2. the "Stelan" system may have a fuel trim problem. If this is correct, we can focus on SIlk's mod to the AOS system to make it as practical as possible?

Quotes source post: 15814245

It sure looks that way. Notice the tubing size between the two. I wonder if that's the difference?

Maybe just as simple as - Larger diameter tubing(5/8"?) with reinforced walls so it does not collapse under vacuum/looks stock/does not deteriorate with oil vapor/heat. Use stock-looking hose clamps.(Norma?)

the difference is the diaphragm and how the aos responds to vacuum. on decel the engine wants to idle (throttle closed) however rpms are high and lots of splashing happening in the crank case. with throttle closed not much fuel is required. further, with throttle closed there is a lot of vacuum in the intake. on an oem aos, this vacuum would pull the diaphragm closed and limit the amount of cleg (air with combustible fuel and oil vapour) drawn from the crankcase. with the aftermarket catch can, even with a third party diaphragm (i think stelan used an vw/audi unit) too much of this stuff gets past and throws off the trims. all silk is doing is attaching the aos to the intake at a different location. completely apples and oranges.

Quotes source post: 15813773

Great stuff! Thanks RK. What a community. I made a decision to investigate this further after today's conversation. Bank #1 looks good. Bank #2 is a bit iffy on this trial. Had a lot of trouble getting Durametric to work for me - at least the way I wanted it to work for me.... Had a few minor starts and stops and the engine didn't fire-up in its normal willing way the last go-round. Not sure what to make of the 0.80 for idle compensation. I'll try to test again and see if it changes. I've got a lot of other things to look at as well.

sooo, first things first. rkat (idle) is supposed to be % at +/-4%, while fra is an absolute number centered on 1 (0.7 to 1.32). durametric shows the opposite; your pics show fra as the % and rkat as an absolute number. who's right? the clue is in the negative rkat. negative % makes sense (add or remove a % of fuel) but a negative value doesn't unless centred on zero (but zero would mean no fuel added ...?). i think durametric has their scales wrong? if so, an rkat trim of -0.8% where a cel trips at 4% is not a big deal. if you are worried about variation from bank to bank (your other rkat is -0.94%; even higher but well within acceptable limits) the variable is probably one o2 sensor getting a bit tired. note that the ecu is pulling a bit of fuel which means less air than anticipated so no vacuum leak; perhaps a lazy injector (sending them to rce for cleaning is always good for an engine spa). note your fra's are pretty much bang on and equal on both sides (nobody gets a 1 and you are well away from .7 limit).

Quotes source post: 15814519

RK - thanks for the lesson. The units are swapped, aren't they? I was having a little trouble interpreting the 0.80, but with your explanation, the other bank may be more suspect. This is potentially related to another set of data I collected. Cly#3 is substantially rougher than the other cylinders. This is just a comparison between #1 and #3. The injector does come into question.

leaky injector or mebbe bad spark resulting in incomplete burn. perhaps pull the #3 plug?

Porsche's explanation of mixture control and adaption trims where RKAT is adding and subtracting Ti, and FRA is multiplying Ti (with a value of 1= no change)

Durametric seems to be displaying a different value system. I would run the adaptions to each limit to see where they are by artificially enrichen (propane) and lean (vacuum leak) too see what the limits are.

Quotes source post: 15814245

once more ... the credits are not mine

Quotes source post: 15814289

My installed mod got some months ago an update. The rubber hoses were replaced by PVC ones. Exactly to avoid bending and collapse of the rubber hoses in a smaller radius curve. The heat of the engine does make the rubber soft and makes it to �click� inward when into a tighter bend. The PVC tubing is also not completely the final stage of the development. It turns out to be quite fragile. I pierced recently the line coming from the oil cap when closing the deck lid. It is a ribboned PVC line. Too tight curves and the plastic will crack at the outer side, cracks in the inner side of such ribbon, as it gets hot. Typical iteration progress with something new developed. When it is put to the real test the flats will appear.

Quotes source post: 15815336

Was it the 997 that has the vent line into the oil fill tube? Would that fit the 996?

WOn the 997.2 it looks like this On the 997.1 and 996 it can be easily made like this:

Does the 997.2 run under the intake or is it plumbed into it like 996? Hard to tell. Thanks for the pics!

No, the pictures solely show the overpresuure protection, which is integrated in the 997.2 oil filler tube. And the mod to put it onto the 996/997.1 you should look at the diagrams to check where it exactly leads to. for the mod it goes to the air filter box. With/without the catch tank in between. how exactly the crankcase ventilation / AOS is connected in the dfi engines I further do not know.

[Image-only post; no author-written text. Original image references:] https://cimg1.ibsrv.net/gimg/rennlist.com-vbulletin/2000x1504/00000portrait_00000_burst20190430192043545_2_df7e2492de6896553c9aa10adead66f101fec66d.jpg

Thanks! I was looking at my old one last night trying to figure out how to get it open as I was curious to see what's inside. You just saved me some time.

Sledge hammer would have been my tool of choice!

The only weakest part is the diaphragm, which has to be air tight. I just don't understand why it has to be so flimsy. I'm sure it will last a lot longer if it were thicker. I also measured the two springs inside controlling a reduced orifice (to reduce the vacuum flow to the throttle body). It's a 2-stage design. At 11" water, the vacuum intake flow is reduced and at 17" water, the diaphragm shuts off the vacuum completely. The trouble is when the diaphragm has tears, the valve doesn't work anymore and it allows full intake vacuum to be applied to the crankcase.

Also, you can see the flows from bank 1 and bank 2 are merged on the left first, then it goes up and enter the center "cyclone cone" to generate a swirl action not too different than to the bagless vacuum cleaners. The motorsports AOS has two such cyclone cones each dedicated to a bank so it's > two stock AOSs because it's taller and bigger overall too. Photo found on the Internet.

One more thought. 997 uses a smaller AOS placed in the middle of the engine but on each bank, it has a pre-AOS. I think it's just a swirl pot with no mechanical parts inside. Perhaps if we can installl this pre-AOS on our bank 2, it will increase our effective AOS capacity. Kinda a poor mans motoraport AOS. Not sure if our stock aos bank 2 hose can mate to it. It definitely needs an adaptor plate to be installed on the head. If my engine was off the car, I would have experimented with this idea. https://www.autohausaz.com/pn/141913...BoCIb0QAvD_BwE

Good stuff, thanks!

Quotes source post: 15814332

This makes me wonder. Would a Stelan option, or simply a catch can instead of the AOS, with the rerouting suggested by Silks setup, be a good option? This would eliminate the high vacuum problem causing fuel trim issues with the wrong diaphragm since vacuum is seriously reduced by the fact that its connected to the air box instead of close to the plenum. In short: 1. Eliminate AOS (once it fails). 2. Install catch can instead of it, with large fittings. 1/2 or 3/4. 3. Use air box as outlet to intake instead of close to plenum. Catch can wouldn't fill up as fast as some race cars have noticed because of lower vacuum when off throttle. In my case I'm talking about a 986, however it applies to 996 too I would assume.

My understanding is that a certain amount of vacuum is needed to relieve crankcase pressure and eliminating the AOS is not a good idea.

You would get the suction from the intake as in Silks setup. Silks setup is already capping off the connection close to the plenum. What I'm suggesting does the exact same thing, except it eliminates the AOS. Unless I missed something? I did read the whole post and mentioned threads. Any thoughts?

Understood, but will you get the same amount or enough vacuum from the airbox? Silk seems very happy with his set up.

I would assume Id ge the same amount with or without the AOS, if I hook it up to the airbox like Silk. Ordered a catch can on Amazon. Will start looking at fittings and hose needed soon. My AOS is failing and I need to address this.

Removing the AOS is silly. With the modification the AOS stays open and functions perfectly fine as a cyclonal separator. It gets heated by the coolant when necessary. No need to take it out. It will not fail. At least not the typical problem with the diaphragm not closing as there is no more excessive vacuum. Every healthy engine has a certain blow-by creating an overpressure in the crankcase which gets pushed to and ventilated via the AOS. The moving air in the air-box creates a (slight) under-pressure so it helps evacuate the aos and crankcase. In case the AOS gets blocked for some reason there is the oil cap vent line to get the overpressure out. Porsche integrated this safety in the newer engines, see the extra line on the oil filler tube in the 997.2.

Cant do the oil cap vent on the 986. But maybe i can tap i. Somewhere else. My AOS needs replacing, trying to avoid having to buy a new one AND do this....

My indy normally installs it without oil cap vent. if you dont track your car or dont do very aggressive cornering you can easily go without

I will be tracking the car. But i'll figure something out for that.

dunno the year of your car, but porsche has been regularly updating the oem part. just make sure you get the latest version from an oem manufacturer that hasn't been sitting on a shelf for an eternity drying out and you should be fine. typically in porsche part nomenclature the last two numbers are the 'version' of that part; i think they are on 04 for the aos at this point?

Its a 2001 3.2.

http://www.autoatlanta.com/porsche-p...section=104-10

Quotes source post: 15880350

986 is less problematic for overwhelming the cilinder heads on track which can cause overpressure and pushing out oil via the AOS.

Quotes source post: 15880111

Is the oil fill tube on the 996 interchangeable with the 997?

I have a 996 4S tiptronic 2002 with approx 115.000km and had the same issues with a failing AOS. Have read a lot about it and how this AOS works and my biggest help to find out was 'SILK' who had already done a AOS-bypass. About 12 months and 10.000km's ago I had a failing AOS, had a new one installed and now again it failed! I have had it with this system and am happy there is a solution with the bypass because the AOS is simply not a good standard. At least not the standard I would expect from Porsche, not to say about their solution on the 996's... With this I want to show you how I did it because in my case the set-up is a bit different because i have changed my original airbox to a Fabspeed airintake, (in combination with a 82mm gas throttle body and IPD plenum) Hope this helps somebody here. 996 4S Tiptronic AOS bypass without original airbox Steps taken from beginning of the bypass: -1- Cut off the large AOS inlet to the throttle body -2- Large diameter tube to oil catch can, (tube leading over the throttle body / I used clear tubes to have possibility to check fluids going through them) I had to modify the in- and outlet to oil catch can also to a large diameter -3- Outlet tube oil catch can to oil filler cap, (tube leading under airintake) With only this set-up the inlet to the throttle body is still open and here the difference to what Silk did starts with my solution because of the Fabspeed Air intake. Main issue is to have a bit of airflow (vacuum) same as when you have it when you use a 3rd outlet on the oil catchcan leading into the original airbox as how it is done at Silk's. This is to prevent an overpressure in the crankcase because of 'blow-by'. -4- On the oil catch can I used 1 of the 2 unused original in-outlets on top of the oil catchcan, (in photo to the right of the vacuummeter) which has a small diameter, (approx 4 mm) and used with this an airtube, (with a flexibe small diameter tube to connect) and lead this blue tube to the cut inlet to the throttle body. Also this last connection is done with a piece of the larger flexibele tube so it fits better onto the harder original black tube. This blue tube (return) is long and in the photo it lays below the air filters. I made it extra long to have the air in it cool down as much as possible to have a cooler air-intake. In the photo you can find in the middle -under the T-bone of the airintake- this bue tube has a connection piece. Easy 'click-on' system when you want to remove... same at the intake to the throttle body. The other original in-outlet on the oil catch can I blocked off with a fitting bolt. So this works fine (for me and -more important- for the engine) I was lucky the small original diameter was just prooving it brings in that bit of vaccum needed But.... you see more on the photo This is about this next piece Connection on 'connection piece' Vent line This part has been changed to new by me when I did the new AOS last year. It is originally going to the AOS and answers some of the questions I read in this thread. This is venting the crankcase when something goes wrong with the AOS and/or when there is too much (over)pressure in the crankcase. I tested this. For example when -in my bypass- I did not use the vacuum from the throttle body, (so, creating a closed system) it created an overpressure and this vent blew oil out. So, I decided to secure the also with a 2nd oil catch can! This is the plastic can I made doing some 'McGiver-work'... and will be re-done in near future with a new small oil cath can. Steps taken: -1- connected a flexibele tube to the vent (see 2nd photo and when you check it out thoroughy you can see it still has some oil in it from the testing... no harm there because this vent is only for 'going out') I check this spot after every drive to see if the level has gone up -2- Connected the flexibele tube to the plastic oil catch can (on top) -3- outlet on plastic oil catch can leads to a larger 'catch can' just in case, because 'when it goes'... it goes with a lot of oil!) So, in future it will be a bit more 'tidy' ;-) Now also looking for the blue tubes in a larger diameter which will be connected using a small piece of the clear larger tubes. With this small piece of clear tube you will be able to check if some oil-flow is happening before it will be visible in the oil catch can Below an example of a connection like this This was during testing: AOS going to oil catch can with small diameter. Did not work. In the back you can see the blocked off inlet to the throttle body: did not work. So just to show how it looks like to have possibility to check flow and the connection is much better because the xlex tube fits better to the hrader original tube... Hope what I showed you here helped you in any way!

As I see it you are using the original intake vacuum behind the throttle body and “just” fitted a cath can in between? Then you still have “high” vacuum at idle. the idea with the “Silk” modification is it eliminate this by moving the line from the AOS to the other side of the throttle body.

Hi 'Soren.991', My vacuum is low, lower then before... 'Silk's modification' to the other side off the throttle body? At Silk's it is going to the air intake box...

gentlemen, I've been pondering this a lot. Instead of venting to the airbox, how about just installing an electric vacuum pump instead. So basically, aos> catchcan> electric vac pump. Accomplishes the same thing, but you have steady vacuum of approximately 10in/hg and no dirty air filters. Might still have fuel trim issues as the pumps vent to the atmosphere, but I'm not sure. I want to give this a shot.

Quotes source post: 16055468

Then give it a shot and report back.

Is there any further development or experience to share on this subject? I have only owned my car for a few months and would like to look into this further.

Been reading thru this thread and I'm surprised no one seams concerned about where the AOS 'catch' ends up. Yes, dirty oil/fuel/water mix is being put back in the sump to lubricant your engine. It's a closed system and all the blow-by product is re-introduced in the engine oil. My AOS also needs replacing and I want to replace the AOS completely with a industrial catch can(google MANN ProVent) where the 'catch' is being accumulated in some kind of container and will need emptying from time to time. There is two things with the current AOS system I haven't figured out which I would need some help with. Look at the picture below. There are two thin lines going to the AOS, item 13 and 12. Can someone tell me where they are connected? And maybe their purpose?

Those are the coolant supply and return lines. Heats up the AOS to better the oil mist removal.

Thank you! Now when you write it I remember reading it was coolant heated. I was afraid it was a balanced pressure equilibrium system, that would have been tricky to replace. Ok, so I will build a system to replace the AOS with a catch can. I have a build thread and I will put all the details there: https://rennlist.com/forums/996-foru...ew-life-3.html

Quotes source post: 16231777

The Mann Provent has been discussed previously. See Schnell Gelb 04-19-2019. https://rennlist.com/forums/928-foru...ation-kit.html It has also been discussed on the 928 Section of this Forum. https://rennlist.com/forums/928-foru...ation-kit.html The Mann Provent does not seem to solve the problem of controlling the vacuum. It seems like this would be a simple problem to resolve until you read how many smart enthusiasts and vendors have failed. Which just makes the 'prize' for success even greater :-) It is a worthy technical challenge .and many of us would be most grateful for a solution. Good Luck.

Quotes source post: 16231987

Why are you trying to reinvent the wheel. Just use the stock unit and be done with it. It works well.

This is not re-inventing the wheel. It makes the following improvement (as mention on the top of my build thread I’m re-engineering some of the bad solutions on the car): +No risk of hydrolock in case the AOS break +Cleaner air gets re-circulated (less oil in it) +No bad oil re-circulation in the engine +Easier to service On the negative -More labour, one more filter(every 40000km) to change and oil(5000km) to empty on regular times -Not original (but for me that is not important) And then we have the discussion of engine problems with the M96 engine. Having bad oil to lubricate the internals doesn’t help. And using it as an oil-burner also doesn't help. For me this is the perfect solution as the oil in the engine is kept as clean as possible.

What bad oil are you talking about?

Before we start a new attempt to improve the AOS it seems prudent to carefully review why previous attempts by all others failed. Most were much smarter engineers than me no doubt, Some are pros from the Porsche aftermarket.They all failed to improve(commercially) on the original AOS (except the Motorsport alternative). Why ? We couldn't find a better way to regulate the vacuum within the limits required for M96. That is way more important an obstacle than the 'dirty oil' problem. Just change the oil plus filter more often if that is the concern? And avoid 'swamping' the AOS with excess oil or wild turns? It would be wonderful if someone could improve on the AOS system but lets not spin any reinvented wheels. We've done lots of that already and documented it for all to read.

I spent a lot of time reading all things AOS/Catch can and came to the same conclusion...Just replace with a new Porsche unit, monitor (manometer) and drive with clean oil changed at 3000k or six months!!

I've asked this a few times but no one ever answered. Is it possible to move the AOS to a more accessible spot so that it's easier to replace? For Tips, it's an engine out procedure.

It would be nice to have an alternative to removing the engine to replace a cheap plastic part. I have read a DIY that demonstrated that it can be done without engine removal, but it looked like quite a bit of pain and heartache might be included in using that method.

The obvious relocation is to the trunk ,behind the rear wheel arch.Plenty of space.Use flexible hoses+grommets. First guinea pig should be a Track car ?

Wyovino: The current AOS have a oil return to the sump which is located at the back of the engine, it bolts directly to this hole. If you relocate it you will most likely need to drill a new oil return hole somewhere in the lower part of the engine and pull a hose there. If you decide that this system is good for you then just replace and keep it in place. It can be done with the engine in the car (what I've read online), but a little tricky. I personally find this system a bad compromise and will replace it with a catch can solution which can be easily serviced.

FWIW,, the same vintage BMW's have the same issue and more with the AOS system. Theirs is just as goofy as the Porsche one and is almost as expensive. Theirs didn't have the hot water heating and used to freeze up in the winter, unless you bought the 300 dollar piece of insulation that covered the AOS lines.. The lines also age badly and fail dumping oil all over the place.

Quotes source post: 16234214

Nick - Please start a new thread when you install this on your car. If your idea works, there will be others looking to follow your lead. This thread has become a bit diluted from its original intent - starting a new one, for a new idea only makes sense. Sorry - I missed the link to your other thread. It still think it makes sense to split off from that one with your AOS specific plan.

Sure thing. I will create the thread after I have a system that I feel is solid and fully functional.

Quotes source post: 16233257

The "motorsports" units are bulletproof, however I've always wondered myself if you could use a Mann/Hummel Provent instead. Granted it's big, bulky, and mounting it high enough in the engine bay in some installations won't be feasible, but I'm going to try putting one on my aircooled 911 in the spring: https://oe-products.mann-hummel.com/...ProVent_en.pdf

Quotes source post: 16237321

Hi Charles, do you still sell the Motorsport AOS adapter for the 1st gen engines ?

Quotes source post: 16237445

Yes. We are going to be releasing a how to for making new hoses when using our motorsports aos adapter on a 5 chain engine.

Quotes source post: 16237449

I want to try that.

Quotes source post: 16237321

If the problem is too high a vacuum gradually destroying a diaphragm inside the AOS unit, would the easiest remedy be an inline check valve that when vacuum reaches a certain level then the valve closes and protects the diaphragm? That would seem to allow the factory AOS to last indefinitely. The valve could be small and easy to mount where the AOS vacuum line connects to the throttle body, and quickly removed if needed for a visual smog check. It could be calibrated to be open most of the time to allow normal operation of the AOS, and only close under high vacuum conditions.

996.2: the function that you describe for the check valve is the exact function that the diaphragm is doing, closing when a certain vacuum is reached. From pictures I've seen I suspect acid oil is the culprit 'eating away' at the diaphragm. The original unit have no good way of protecting the diaphragm from the oils.

Quotes source post: 16237964

Correct!

Part of the problem is the oil return from the AOS ends in the sump and sits close or under the oil level. With a failing AOS oil gets sucked like a straw via this return line back into the AOS and partly into the intake. Part of the return line is integrated in the engine block, running through a small ribbon along the sump side wall. One could drill a hole through the sump wall at a height well above the oil level and ending in this block integrated oil return line to provide a sort of �venting� line for the oil return line from the AOS. If high vacuum gets applied the AOS the oil will not get sucked up. The return line will draw air from the channel made. This could give some safety for the engine as the AOS will not directly be flooded with oil.

Can one of the experts on these things explain why running a vacuum pump or tandem pump (on the head) wouldn't produce a more consistent vacuum source without any connection to the actual intake?

Quotes source post: 16239711

Running an engine driven vacuum pump is less efficient and has a simple problem that would need to addressed. Since the internal combustion engine is it's self an air pump, using it's self to evacuate/vacuum the crankcase is the most efficient way. If running an engine driven vacuum pump, the exhaust from the vacuum pump would contain oil/vapor mist and would need to be collected/disposed of in some fashion. This is actually what is being done with DRY-sump multistage aftermarket systems. The engine driven oil pump has several "stages", some are used to create a vacuum on the crankcase, and the exhaust is just expelled in the oil tank and recirculated. I have an idea for correcting the misgivings of the factory AOS system. That will maintain the factory design but take care of all the misgivings and still function as designed.. Stay tuned !!!! I will make a new thread when done....

Quotes source post: 16240917

Interesting! if you can place a link here so I don�t miss it.

Quotes source post: 16244245

Link to the "Ultimate AOS Solution" so you don't miss it...

At the moment +400 cars modified without any problems! only benefits: no more crap in intake, cooler intake temp (+bhp), smoother idle, never hydrolock due to oil dumped in intake, AOS always open (diaphragm never �fails�), easy to temporary revert to OEM for inspection.

This system is a good "low cost, temporary down and dirty DIY Mod", but the Type 3 pcv system was abandoned in the US decades ago. It may be ok to use in other countries, because it will stop the risk of hydrolocking, but not as efficient at removing the blow-by that can cause sludge in the oil or create enough negative crankcase pressure to help stop oil leaks. Would not be a "legal fix" and definitely cross the line in the US, unlike the " Ultimate AOS" which is a slightly better design than the OEM all the way around.

I can agree on your arguments. Legality of it is different for every country or regio. I just wanted to share this modification works very well. Notwithstanding all the hypothetical technical drawbacks posed. AOS is still in place, evacuated crankcase gases still go to the intake only now before the throttle body instead after. Fresh air get mixed. Would this means it is still a type 3 system? How It would not be good to remove the blow by gases? Even a throttle body in small open position already creates enough vacuum to evacuate the crankcase gases back into the engine. As the throttle body is closed there is no or very little evacuation of gases in the original situation, just the same as in the modified situation. As already said I in the US it will not pass any technical inspection I suppose. I hope your ultimate solution works well too!

Quotes source post: 16379248

Yes your system is a typical Type 3 system. I have used the Type 3 system on my 996 and I know it does work, it's just "messy" with the oil vapor getting on the air filter/MAF sensor, and doesn't produce as much negative crankcase pressure or a constant negative pressure as the original design, therefore is less efficient and hence it was abandoned in the US..My engine was a lot happier and peppier when I went from the Type 3 to the Ultimate AOS design. More negative crankcase pressure means more power. The original design and the Ultimate AOS keeps a constant negative pressure on the crankcase at all times even during high load or low load due to the balancing of pressures by the flow regulator. The flow regulator uses atmospheric pressure, spring pressure, and blow-by pressures to control the negative crankcase pressure provided by the intake vacuum. More negative pressure on the crankcase is better for emissions, ring seal, and also overall power (windage losses from high piston speeds)..

True. Also it gets a bit more dirty quicker. The dirt stays in the airbox though, so it is a very easy clean. And who does not change is airfilter regularly? It is a fair trade off between safety and workaroundability. The MAF getting dirty is all within reasonably expectations, cannot complain about that. I relaxed it after 30.000 miles. I consider it anyway as a �regular� wear item. So I don�t mind the extra dirt. Hot film voltage values stay perfectly within the +- 0.03V bracket for a long time. For what it is worth the MAF design is also highly flawed. Highly sensitive to heat soak. When spoiler is down and low speeds, intake temps rise high fast, but It takes very long for the temperature sensor to read the accurate intake temperature when driving at higher speeds again. I really admire the engineering behind your mod. Hope it works good in practice with the IR sensor.

Quotes source post: 16380246

Thanks man, The infra-red sensor is a really nice piece, very accurate, durable, not effected by ambient light, bubbles ect. It sends an Infra-red LED light into a prism which reflects the light back to a photosensor. When the liquid level reaches the prism tip, the Infra-red light is refracted into the the liquid and away from the photosensor, causing less reaction of the photosensor which is picked up by the microprocessor and switches the output. It's very fast, accurate, and precise. I know it's hard for you to go away from the type 3 system since it is a low cost, easy mod, and you are having success, but you may want to think about going to the Ultimate AOS for the efficiency, better ring seal, lower oil leaks, more power, and of course emission friendly. BTW I like your drain vent hole idea in the lower case, that's makes since, and is easy to do "while your in there"..

The drain hole vent works. Tested on car with a fragged AOS refitted. It will just give you that safety when the diaphragm breaks your engine does not convert into a direct oil injected engine. Coincidentally after making the vent hole in my car and reverting back the AOS modification to stock the AOS broke after 400miles. Probably it was already terminal. I got none of the usual symptoms. Except the puff of smoke at startup and symptoms of high vacuum under idle when removing the oil filler cap. I got it replaced and all good now ahhh and for oil consumption ... I have had a stable oil consumption of 1 quart every 1250-1500 miles. I was curious to see if reverting my car to stock AOS setup would make any difference. As I was convinced the non ideal vacuum would influence ring efficiency. In my case no difference. I would love to try your solution if it gets available. On the other hand I am one of those that drive their car, hard, no mercy, ... and in those circumstances the AOS cannot cope always well with the amount of oil mist to separate. I really don�t like all that gunk going directly into the intake. I notice it now with my stock setup that I have the occasional smoke on startup. Which I did not have with the mod. So I prefer the extra safety over the little extra power and efficiency. i will make up the loss of uuumpf by getting rid of 9k unsprung mass and rotational inertia at the wheels and putting finally my recaro seats. Would excessive hard driving and occasionly overwhelming the AOS trigger your IR sensor?

I'm in the middle of an AOS rebuild right now, replacing the AOS unit with a Provent 200. Everything is put in place but have not yet started the engine due to other works needing done before. In a few weeks. Something which have not been mentioned is the tubes in the system. Mine where very brittle, basically shattered when trying to bend a bit. They also had beginning cracks in bends and connectors. I think replacing these should be high on the list of to-do when sorting out the AOS. The connectors can be reused if 19mm I.D hose is used.

Quotes source post: 16389343

Please do post photos of your Provent install. I have always thought this would be a great alternative and have a Provent in the office for this exact reason. AOS vent lines and oil fill tubes are a must on almost every engine we see come in for an IMS bearing. They are all brittle and when they crack, it leads to vacuum leaks that cause enrichment and wash down cylinder bores. Definitely a maintenance item.

Quotes source post: 16389534

Sounds like motor out maintenance is a good idea for these close to 20 something cars. I had the air cooled engine out routine mastered. Now gotta figure out this water cooled lump.

Quotes source post: 16389619

Lately we've been dropping the engine and trans on most IMS jobs as we're doing so much preventative maintenance, it's faster that way. The engines more or less fall out and in without much fuss. Porsche did a good job making them plug and play.

Quotes source post: 16389619

While it's not "difficult" to drop the motor, it is a lot of work - especially the first time around. Even though I'm using jack stands, for doing something like the AOS or the oil cooler/heat exchanger, I'd do it again in a heart beat (and I'm sure it would go quicker the second time around). It really makes these jobs so much easier (and safer - there are lots of brittle plastic things to break or knock loose if you're fumbling around blind in there). It also gives you the benefit of being able to look at everything on top of and behind the areas you can see with just the engine lid open, as well as ding a good cleaning while you're in there (99% you've got some rodent nest/droppings on top of your engine).

I have used silicone hoses. But you need to spend time with a catalog to find the best shapes and adapters.I used only the best quality USA made hose and thick wall[ or wire reinforced to prevent it collapsing under vacuum. https://www.siliconehose.com/blog/si...bsolute-best-/ Use black so it looks stock -not Ricer red ! I found some import silicone hoses became stiff after heat cycling. The best connectors imho are barbed brass. Another rabbit hole of hours of nerdy research.They are cheaper online than the official Pcar plastic parts. https://fastfittings.com/ Hope this helps QUOTE=Nick_L;16389343]I'm in the middle of an AOS rebuild right now, replacing the AOS unit with a Provent 200. Everything is put in place but have not yet started the engine due to other works needing done before. In a few weeks. Something which have not been mentioned is the tubes in the system. Mine where very brittle, basically shattered when trying to bend a bit. They also had beginning cracks in bends and connectors. I think replacing these should be high on the list of to-do when sorting out the AOS. The connectors can be reused if 19mm I.D hose is used.[/QUOTE]

Quotes source post: 16389846

Dropping the combined engine and transmission is much safer and easier with a scissor lift table on casters .We've discussed this at length with videos in other threads. Some have used a motorcycle lift on casters. There are some vital concealed connections and obstructions to deal with. Also covered in the past. Then there is the need for an engine yoke to fit a standard engine stand.....

Quotes source post: 16389946

[/QUOTE] Be cautious when using silicone hose for oil or fuel, silicone works well for everything from baby food to coolant ,except oil or fuel. If useing for oil or fuel it needs to have a fluorocarbon or fluorosilicone liner to last a long time. I found this out while researching hose for the Ultimate AOS..

Update about Provent 200. After testing, this unit will not work. It have a regulator which functions for a usable pressure BUT it needs a lot more air flow to function. When the valve closes it really just limits the flow to 'minimum' which is still so high that vacuum increases in the crank case. The flow is just too small on the M96 3.4L, leading to very high crank case vacuum. I have spent the last week trying to modify the unit(restrictors and/or new valve seat) to function better at low flows but to no vain. Maybe the Provent 100 is a better option. I've had enough of experimenting with this now, so reverting to original for now. I did a test of the filter, filling it with oil, to see how well it would block in event of the system sucking oil. Well, the oil passes in rather fast pace thru the filter. It doesn't do a good job preventing oil from going to the intake in this case. Charles, a question came to mind while doing this that you might be able to answer. Why are the vacuum ports on the engine on the heads? why didn't porsche just suck the vacuum from a more center point (say the oil filler tube) to avoid oil going up in the system?

Quotes source post: 16424503

I wouldn't be too concerned about high crankcase vacuum. It's going to be different than the stock AOS. We've actually maxed out our manometers when using Total Seal rings on these engines, so it's just a number, not a constant. You won't hurt anything. I would imagine the AOS wouldn't function as well for evacuating crankcase windage if it just pulled from the oil fill tube. Pulling from the furthest points in the crankcase most likely ensure a more thorough evacuation of the crankcase. That's my thoughts anyway. Just a guess!

Quotes source post: 16424614

Charles, using the Total Seal rings on my rebuild. What should I be expecting for manometer readings?

Quotes source post: 16424936

We have had 3.8 (which have a different AOS setup than all the other engines) with Total Seal rings pull normal manometer readings, however, we have had many other engines max out the manometer and even with the engine off, they will hold vacuum for quite some time. Long story short, I can't give you a firm number to look.

Be cautious when using silicone hose for oil or fuel, silicone works well for everything from baby food to coolant ,except oil or fuel. If useing for oil or fuel it needs to have a fluorocarbon or fluorosilicone liner to last a long time. I found this out while researching hose for the Ultimate AOS..[/QUOTE] I found this out when I used silicone hoses for my PVC/catch can mod on my VW Live a little, learn a little.