Fuel Dilution Causes?

What are the typical causes for elevated fuel dilution numbers in UOAs? Obviously leaky/old fuel injectors and vacuum leaks could cause it. But what other culprits could there be? Weak fuel pump? Failing AOS? My numbers previously hovered in the high 1�s prior to installing new 996.2 injectors, and they�ve been climbing ever since that install. It now appears to be affecting the additive package in my oil so it�s time to act. The engine has been smoked multiple times with no visible leaks. New Bosch O2 sensors, new Beru plugs, new Bosch coils, new Bosch MAF.

First thing I would do is put those original injectors back in there and see what the subsequent UOA's tell you.

Quotes source post: 18389078

Unfortubately those were tossed months ago. They appeared to be original to the car (1998/1999) and I figured they weren�t worth keeping around.

That's a shame. In that case I would go back in there and remove the new injectors and test them or have them tested. Most probably they'll perfectly fine but re-doing the whole r&r of them will require you to re-visit your/the mechanic's work and that in itself will show what went wrong and is causing the fuel dilution of the oil. Hindsight is 20/20 but for the future you could consider, also for 20y old parts -never change a running system -parts quality is never higher than during the actual manufacturing period of the vehicle they are used in

Were the fuel injectors new OEM Porsche?

Quotes source post: 18389734

Bosch OEM. Bought during the pandemic and I struggled to find them in stock at that time. I�m honestly suspecting the AOS at this point. I�ve been considering upgrading to Skip�s UAOS soon as my AOS is 7 years old and has 35-40k miles on it.

Quotes source post: 18389693

It seems like a strange suggestion to me to never replace 20 year old injectors, that largely, have been recommended here to be replaced. Plus, we're not comparing parts quality to when brand new. The comparison is between new injectors in 2022, and 20 year old injectors with many many hours and heat cycles on them. If new injectors are worse than that, then there is a significant new part production problem! It's pretty cynical to expect that all new (Porsche or OEM) parts are automatically even worse now, than the 20 year old worn parts currently on the car. Practically, you might also double check for vacuum leaks. Do a smoke test. Replacing injectors requires reaching around the top of the engine, it could be that something was left loose, or a vacuum line broken. Did you buy the Bosch injectors from a reputable dealer? I think I've read cases of counterfeit parts from ebay and Amazon

Quotes source post: 18389787

Read closely what I wrote. Both wording and meaning is different from what you took it for...

Quotes source post: 18389693

Uhh, not sure about that statement. For example, Porsche implemented a 6305 IMS bearing after the 996 generation had ended thus unofficially admitting that it was a superior choice over the 6204 single and even the 6204 dual row. All short block replacements did not include the older bearings. I could list a bunch of other parts that were improved because of poor quality and/or performance after the manufacturing period. Personally, I think the problem with OEM injectors is the amount of knock-offs that seem to be genuine BOSCH but end of being copies and subpar to say the least. That's why it's important to purchase replacement parts from reputable companies like LN Engineering, FCP Euro, Pelican Parts, etc. And when in doubt, purchase genuine OEM from your local Porsche parts department.

Quotes source post: 18389889

Comparing part number (incl. its revision) per p/n is how I mean it as well as quality in the sense of how well the original spec is met by each and every individual item of said p/n produced by the parts manufacturer. And it's not about finding the exceptions to this statement. The value is in taking it into consideration when replacing parts. Nothing more nothing less. This is not contradicting what you pointed out BTW.

Quotes source post: 18389988

I think it's important to be specific to owners especially new owners that don't have experience with the good, bad, and ugly with parts associated with these engines. For some other makes, it's a standard practice that owners should seek out NOS parts. However, that's not a good practice with these engines. Another example, Porsche recognized problems with coolant expansion tank caps. They leaked under pressure. After a few revisions, they finally got it right. But if someone were to purchase the exact caps that were sold during the manufacturing period, it would be a problem that could lead an owner in the wrong direction thinking something else is causing the leak with the cooling system. They had similar problems with earlier versions of the RMS. In 2005, Porsche moved to Polytetrafluoroethylene and that stopped the oil leaks. I think you get my point and I'm not trying to be argumentative. Just think it's a better message to tell owners to purchase parts that are proven to work and in some cases, that may mean aftermarket.

You have been considering installing the UAOS for some time... JUST DO IT !! ..lol There is no downside to installing the UAOS, in fact it should have been designed that way from Porsche... I recall a while back you had reported that a manometer reading was nearly -9inwc on start-up that slowly went down to -5inwc after some seconds. That is a sign the AOS diaphragm has a small tear ( failing). The possible scenario from failing AOS to fuel dilution in UOA.... .is as follows....The failing AOS will cause a lean condition while cruising which will cause the fuel trim to adapt to more injector open time to compensate ( which produces the correct mixture) , but after the trim has adapted this increased injector time is reflected also in cold starting which the increase in injector open time will cause a VERY rich mixture when cold..( like an old school choke that is over-chokeing).. this can lead to fuel dilution after a while of doing so..

Quotes source post: 18390291

Yessir I�ve just been putting off the inevitable purchase for various reasons. Really I was just hoping I could make it to Black Friday for a holiday deal. Order submitted. Thanks Skip.

Quotes source post: 18390393

Got it !! Thanks !! It will take a couple of weeks though, I am sold out at the moment...George has been very busy and his small CNC machine that usually does the UAOS's has been tied up on a large-long running job...That little machine has been so busey that he actually bought a new machine to help out on the work load...He is suppose to start on them this week and then send them to be Annodized...so will take a little time... Please be patient..

Usually poor fuel injector performance, a fouled/ failing MAF, aftermarket intake or exhaust are responsible for high fuel dilution. In the cases where these have been rules out, the driving factor is usually mechanical in nature, with combustion byproducts and raw fuel being allowed to bypass the critical seal between piston rings and cylinder walls. If the above conditions are present, they can create a lack of ring seal that can further drive the issue.

Quotes source post: 18391490

Thanks for that info. The injectors are less than a year old, the MAF is less than two years old, the intake is factory and the only exhaust mod is a set of TechArt mufflers which have been on the car for around 10 years, maybe longer. 🤷

I'll throw out another idea, maybe bores are ovalized at the top and causing leak down and some blow by, allowing more fuel into the oil than normal?

Quotes source post: 18391556

That thought had occurred to me but wouldn�t there also be some oil consumption associated with that? I�ve owned the car for almost 8 years and she�s never had a noticeable use of oil.

Quotes source post: 18391621

The bores ovalize more at the top than the bottom. So the oil rings see little deformation, so in theory should still control the oil consumption

Quotes source post: 18392751

80% of oil control comes from the 2nd ring. The oil control ring measures the amount of oil delivered to the bore, but it doesn't determine oil consumption. BTW- A MAF that is 2 years old isn't considered new.. They can last a shorter amount of time than this, or become fouled. What injectors were used a year ago? Were they new? Rebuilt/ cleaned?

Quotes source post: 18393184

The injectors were brand new Bosch 996.2-style. I�ll get some Durametric readings on the fuel trims and MAF and post them here.

The plot thickens. Today my Durametric showed a P1123 code though I�ve not had any CEL occurrences. I cleared it and went for a drive to log some data. Lo and behold, the code returned after I got back from the drive.

Here are some data logs from the Durametric. It seems like my fuel trims at idle are way out of whack, yes? Hot idle with the AC On: Hot idle with the AC Off: Hit engine, lurking around my subdivision in 2nd gear: Warm engine, shifting through some gears on the highway: Warm engine, accelerating down the highway:

https://rennlist.com/forums/996-foru...l#post15990479 https://rennlist.com/forums/996-foru...l#post15990810 https://www.callasrennsport.com/blog...im-information I just post these as reference to you, or others who come cross this, as to what the TRA/FRA numbers mean. Maybe a yours is running little rich? You might also look at what your MAF air mass numbers are (kg/h).

Quotes source posts: 18400010, 15990479, 15990810

I have read through those before but thanks for posting them to read again. Here�s my kg/h at hot idle today with the AC Off. MAF appears to be ok since my 720 rpm idle is about 20 rpm above spec (680 +-20).

Every time I have seen this scenario it has been a faulty MAF sensor...( but this could be the exception so test fuel pressure first ) Test the fuel pressure and the pressure regulator, and make sure the vacuum hose to the regulator is hooked up and functioning...it should lower the pressure when hooked up. The CEL set-point for code p1123 is -0.36 ms for the TRA fuel trim, you are right at it on one bank with -0.35ms.. and not to far away on the other at -0.30ms Your injection time is 2.4ms .. then is lowered by .35ms on bank 1 and by .30ms on bank 2 Also your "specified air mass" is 12kg/hr with a "specified air adaption" of -3kg/hr for tank venting so the expected air mass of 15 kg/hr but your MAF is sending a signal of 18.5 kg/hr.......15kg/hr is the norm for the 3.4.. I would test the fuel pressure/regulator to confirm they are good, then substitute a known good MAF to confirm the fuel trims go back to near 0...

Quotes source post: 18404498

The car threw more codes today after a cold start. P1123, just as before, and now P1319 and P1318. Thr car was audibly rough running this time under load while at low rpm�s for a few various moments until it fully warmed up. Still no CEL has presented itself�.yet. I�m working on procuring a fuel pressure tester but I did remove the vacuum line from the fuel pressure regulator. The rubber connector that attaches to the regulator was slightly moist on the inside with fuel. Wouldn�t this indicate that there is possibly unmetered fuel entering the engine via this vacuum line? We kept the vacuum line removed (and plugged) and then ran the fuel pump: once using the Durametric activation and once with the engine running. We expected some fuel to spurt out of the top of the regular but none did. Maybe it has a leak that only presents itself under vacuum (and not simply with the vacuum line removed)? Interestingly, when we removed the vacuum line from the fuel pressure regular, the high MAF kg/h readings didn�t correct themselves. The car still idled between 680rpm and 720rpm and the kg/h varied from 18.75 to 22.0. If there was an unmetered fuel leak, wouldn�t those number have immediately corrected themselves after removing the vacuum line to the regulator (and therefore removing any unmetered fuel from the equation)?

Possibly the membrane inside the fpr is pinholed and fuel is entering the vacuum line but I would be very surprised if it only happens with the line connected to fpr and plenum as (at idle) this effectively only lowers the pressure with which the fuel is being pushed against the pinholed membrane from 47 psi to around 44. IOW if it is leaking with 44psi behind it, it would do so with 47 too. Perhaps something else inside the fpr is amiss but the fuel pressure tester will hopefully reveal this. Good call on observing the fuel wetted hose nonetheless! You could put some length of clear hose onto the fpr, connect it to the plenum and watch whether fuel actually comes out of the fpr. Wrt the MAF reading: it measures air mass not fuel mass so it is not surprising that the numbers are unaffected by any fuel from the fpr coming in through the vacuum line or no longer does so with the line removed from the fpr and plugged.

How would a weak fuel pump factor into all of this? Purely coincidence in regards to the MAF readings? Additionally, I disconnected the MAF while the car was idling and it immediately died. I don’t recall that ever happening before.

I have a better update now and having some issues with my fuel pressure testers yesterday. Fuel pump On with Durametric, engine Off: 54 psi Engine On: 45 psi Engine On with vacuum line removed from the FPR: 54 psi Engine Off after 30 minutes: 42 psi Engine Off after 60 minutes: 40 psi Engine Off after nearly 4 hours: 24 psi

Deleted double post

With those results your fpr is fine and pressures are within spec too. In order to find out the source of the pressure drop you could clamp the pressure hose shut between pump and fuel rail when pressure is 54psi. If pressure still drops, then one or more injectors are leaking which could cause fuel dilution. Clamping the pressure hose that goes from one fuel rail to the other at the same time allows you to identify which rail or whether both rails are affected by leaky injector(s) fitted to them. If not, then the check valve within the fuel pump itself is bad or you have a (albeit tiny) leak in the pressure hose inside the tank. Neither of those causes fuel dilution.

Quotes source post: 18409943

I�ll work on that but I�m still going to replace the FPR since it appears to be slightly weeping gas from the vacuum port. But isn�t some amount of system leak down acceptable? Isn�t the system only designed to hold pressure just long enough to allow the fuel time to cool and not boil in the line?

The weeping of fuel will at least contribute or cause it in its entirety depending on how much fuel weeps out so replacing it is a good idea. Perhaps the valve inside the fpr between inlet and return is compromised too and a new fpr would solve that in one go. I don't know tbh if that is actually a failure mode of an fpr in real life but you never know. Rerun the test after you have fitted the new one to keep a tab on what action resulted in which change to the symptoms. It will help in keeping on overview on the problem solving as chances are you'll be doing more than just replacing one component. True that some pressure loss is acceptable as long as the fuel bleeds to the tank and nowhere else. Worthwhile reading in this thread too wrt fuel pressure and the fpr https://rennlist.com/forums/996-foru...d-start-2.html

Lack of cylinder wall/ piston ring seal is a mechanical reason for fuel dilution. If the rings don't seal, unburned fuel and carbon goes straight into the oil. This is why mechanical inspection is required prior to carrying out other corrective measures to engine subsystems. All too often we are seeing engines with new injectors, FPRs and other items that were replaced rather than proper diagnosis being carried out first. This is the absolute most incorrect way of attacking these issues. Rule out mechanicals first, then you can start chasing the issues by firing the parts cannon at it. People often say "the engine runs fine" so they avoid the proper chain of diagnosis. Don't do that.

^^^BRS is correct and is why I posted the "blow-by Test for Engine Health" thread.....too much blow-by will cause fuel dilution, BUT takes quite a few miles to show up or effect "fuel trims".. Your fuel trims are out of spec... It's like the "chicken or the egg" but we KNOW which one comes first.........with fresh oil too much blow-by will not cause fuel trims to be out of spec, once the oil gets diluted so badly that "fuel fumes" are being drawn off the oil and through the crankcase ventilation system can it have any effect on fuel trims... So if fuel trims remain out of spec with fresh oil, then blow-by is not the cause...... Also your fuel pressure is in spec , so fuel pressure is not the cause....( the wetness you saw in the vacuum hose is probably due to oil vapor condensation..) My money is on the MAF Sensor, it takes 45 seconds to install a known good one...and a few minutes to re-test to see if the fuel trims go back to near 0 ....I would just borrow one from someone/somewhere since it is so easy to install before spending the $$$$.....go to a local PCA Cars and Coffee and make a friend with a 99 996,,,, or I can loan you one, I have one sitting not being used....

Quotes source post: 18411855

You�re probably right about the FPR but it had a very faint smell of gas to the liquid so I figure it�s a cheap and easy item to knock out. The oil was changed in mid-August (that oil sample is the one that came back with a fuel dilution warning). The fresh oil only has about 500 miles on it at the moment. Yesterday I disconnected the battery for 30+ minutes and connected the positive and negative battery cables to each other in order to reset the fuel trims. I drove to work today with no issues or CEL, though the idle began getting a little rougher the longer I drove. I will drive home this evening and check the trims again. Fwiw, apparently Bosch has a lifetime warranty on MAFs and they are working to get a new one sent to me. Hopefully I can get this all resolved before my shiny new UAOS arrives.

Here�s the results after resetting the fuel trims, driving 30 miles to work, the car sitting all day, and then driving 30 miles home:

Yea, your MAF reading is high.....at 680 rpm should be 15.5kg/h ..........

I think that table is just an "random" example and to show that it's important to group those certain values when doing diagnosis. 0.96 in that table is below acceptable range for example for fuel trims (0.98) and MAF is just at the low level. In that example it's easy to see that adding air affects only on TRA and not FRA because that's is measured above 1700rpm+ range. That 18.00 at 680rpm is actually within acceptable range as it's at 680rpm 17.00 +-2.5 However I agree with Porschetech3 that the issue is likely a faulty MAF since both of those trims are way above accepted range and if fuel pressure is ok.

Quotes source post: 18409943

@hardtailer This may be a dumb question but how do I clamp off the fuel line when they all appear to be made of hard plastic?

I have an interesting development (and I�m still somewhat skeptical of the results below). I replaced the FPR yesterday, took the car for a 14 mile drive to get her warmed up, and then hooked up the Durametric to see if anything had changed. It�s important to note that I did NOT reset the fuel trims before doing this nor did I make any other changes to the car. My MAF readings are still problematic but take a look at my short term fuel trims. One is now zeroed out (as it was previously -.38) and the other has ever-so-slightly improved from -.30 down to -.29. 🤷🤔🤷🤔

Quotes source post: 18415162

Small correction - both TRA and FRA are long term fuel trims - Porsche/Durametric doesn't publish real short term fuel trim values.

My 3.4 MAF is fine and reads 15 kg/h at 680 rpm for 80k miles and hasn't changed in 15 years... P9C I'm curious to where you got that 17kg/h +- 2.5 kg/h spec and was it specific to 3.4 engines or both 3.4 and 3.6 engines? Also the table looks typical to me for a 3.4 of unknown mileage....

Quotes source post: 18415788

@Porschetech3 Here:

PC9 ,yeah , that looks like 3.6 specs.......notice the specified air mass is 17 kg/h whereas the OP has 12-13 kg/h specified air mass.....

Quotes source post: 18414055

On my 99 they are rubber hoses going to the fuel rail as well as between both rails not the black PA12 ('nylon') plastic hardline The hard plastic ones I would not clamp as it will damage them and as such know no method how to safety clamp that type of line shut.

Quotes source post: 18416208

Thanks. I�ll look again to check but it seemed like mine were rubber with a hard liner on the inside but I could be mistaken.

I�ll. Be. Damned. The new MAF came today (free warranty replacement by Bosch). I reset the fuel trims and this is the latest scan after a 15 mile drive. @Porschetech3 and @BRS-LN nailed it.

Good stuff! Congrats and thanks for reporting back!

The fuel trims look really nice now, that should cure your oil dilution issue.. Thanks for the quick closure on the outcome !! I hate it when a repair thread drags on and on and then the OP just disappears...lol

Circling back to this ongoing topic for me. I installed my UAOS a few months ago and the car idles nicer than it has in the previous 5+ years. Unfortunately it looks like my TRA fuel trims are headed in the wrong direction again. Zero air leaks detected when I smoke the car via the throttle body and via the oil tube (which is also new). I also feel like my MAF air mass reading is trending toward the wrong direction again too. Starting to run out of ideas as the car has had the following replaced within the past 12-18 months: O2 sensors coil packs plugs MAF UAOS injectors (996.2 version) fuel evap/regen valve oil fill tube Hopefully my fuel dilutions will be finally coming down due to the new UAOS but I won�t have that oil analysis for another month when I do my next 6 month change. The car burns little if any oil; maybe 1/2 quart every 2500-3500 miles. No CELs (logged or active) either. 107,xxx miles. Any thoughts from the hive mind? EDIT: it was pretty hot the day that I took these readings. Probably 90�F.

R your precat o2 sensors responding when u graph it? They should go up and down with throttle. To have both of them at .97 at idle is pretty strange, maybe? Edit Never mind.....i see the voltage numbers lower and they look fine. The rest of the numbers look decent to me.

Quotes source post: 18958511

I noticed the lower voltage number as well. I wonder if my starter Y-cable is dying again?

OP - Check your O2 sensor graphs for proper sine wave. Have you read the older threads where ahsai and others went over expected sensor resistance and voltage? Unrelated to fuel, but as De Jeeper mentions voltage, you could check the nuts to the cable ground points for tightness. Have you done the secondary ground cable mod?

Quotes source post: 18958717

What is the �secondary ground cable mod�? Is there a second ground besides the bank 2 cable to chassis?

You fuel trims look great/fantastic/marvelous/beautiful/excellent ......( your are obsessing ..stop it!! .. ) Maybe you need to re-polish your "ugly" headlights to stay busey...I would love to have your lack of problems.. lol

Quotes source post: 18958842

PM�d you to not hijack.

Quotes source posts: 18958717, 18958985

Good tip on the graph. I�m triggered to turn them off because I�m running my Durametric off a 12 year old Dell laptop and it really bogs down sometimes. Thanks for the reassurances. I�m just paranoid now after everything I�ve been through to get them corrected.