When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
Diego - dumb suggestion, do you have the right dipstick? Could your actual level be too high even though it reads right?
It's a valid question. Yes is the factory dipstick and always I set it at 1/4 (below half machined section) at 9 PM or 945? oil temp needle, also cluster oil level is always showing "low or empty". I'm under some other tests and re worked a piece, seems like it corrected the issue in the left turbocharger but then after some tests, the same problem moved in to the right side turbocharger, so now I'm doing the same mod in that side and will test it.
It's a valid question. Yes is the factory dipstick and always I set it at 1/4 (below half machined section) at 9 PM or 945? oil temp needle, also cluster oil level is always showing "low or empty". I'm under some other tests and re worked a piece, seems like it corrected the issue in the left turbocharger but then after some tests, the same problem moved in to the right side turbocharger, so now I'm doing the same mod in that side and will test it.
Kinda difficult to explain, but still the problem is present, although seems better. It's very random the leak of oil pass trhough the turbine wheel with the reworked check valves. However, and this is not the first time happend in the car, sometimes, you will find a chunk of oil in the inlet tube (elbow) of the turbocharger, SOMETIMES, so when you remove the intake elbow of the turbo (left side mostly) you can trace the oil comming from the "Y" intake pipe...then you ask yourself, how is that even possible? I'm not sure what cause this , but all I can think is that, under certain circunstances, the turbo oil pan vent hose, seems to flow oil back all the way up into the "Y" intake pipe while the car is resting (but then again, very very randomly) and flood the intake of the turbocharger of oil.
For example I was testing the car for probably 2 weeks and all was perfect, no turbo (turbine) leak or intake flowback of oil at all, and one day, after starting the engine, in the next morning, showed oil between the turbine housing and turbo exhaust "downpipe" on the right side turbocharger...it burned quickly. Then, after 3 days, in the morning showed this flood of oil in the elbow (left turbo)...so yes, very confusing and frustrating.
All I know this residual oil pressure happening randomly in the engine, cause to push the oil through the turbo shaft into the exhaust side or making the oil pan vent hose of the turbo, relief oil into the "Y" pipe, making the oil flow back into the mouth of turbo.
had my 993TT for just about 15 years now and smokes like James Bond pushed a button in his Austin Martin. Just must be parked greater than 24 hours prior and guaranteed smoke screen every time.
i have done NOTHING about it.....EXCEPT park it on ramps. I've said this a few times before. And I think it makes sense to some people. I'm just mentioning it again to contribute albeit on very basic level..
I think all of the people in this thread smarter than me might benefit from that information and make sense of it to confirm their theories.
I read about this along time ago when Porsche as a company formally endorsed putting the rear end up on ramps between races in the 90's..
I could leave it for six months+ (and I have) and it still will not smoke at all (not even the slightest puff) after rear on ramps. To reiterate it just takes 24 hours parked to smoke w/o ramps.
this absolutely works.... 100% every time... I just have no idea why.....ha
-joe
Last edited by Ridin Dirty; 07-26-2021 at 02:42 PM.
This thread got me working towards figuring out and trying to solve my smoking, dripping 993tt. Some history. I have owned a ’97 993tt for just about two decades. When I first bought her, she was totally stock and running the standard K16’s. She did though come with the factory 430hp option (different ECU, additional oil cooler etc.) She was in good condition and well maintained. I upgraded to K24’s (along with some other upgrades) within the first years of ownership to customise the car to my driving needs and style.
Around 15 years ago or thereabouts, I started getting some oil leaks as well as smoke from the Turbos. I had the Turbo’s checked and all clearances were good so they weren’t touched. The engine though was stripped down to a point where all seals and gaskets were replaced. Valve and timing chain covers etc etc. The dealer put a dent in my wallet for that work but the leaks were reduced. I still got oil drips off the Turbos though. And she still smoked like a banger on startup if left standing for more than a few days.
After a few years of her startup smoking (that got progressively worse), I decided to replace the Turbos with brand new (not rebuilt) K24s. Did this stop the smoking? No. So I went down the route of installing P trap oil lines (even as I questioned their logic). That didn’t help either.
Anyway, I had time on my hands so, I had a good look and think about the problem and the oiling system on our cars. Some clues I had to work with;
a). I had started to drain the turbo oil reservoirs on my car If I left her standing for long to reduce the startup smoke (which helped). And always found almost a cupful of oil drain out immediately. If I left the drain plugs out, oil would slowly continue to drain for hours after. Similar to DiegoR's test.
b). When I did my oil changes, the crankcase would contain quite a few liters of oil. And here too, oil would continue to drain for hours after if the drain plug were left off (with the oil tank full).
c). I'd heard of raising the rear of the car on ramps. I tried that and it worked. My car stopped smoking and dripping. But the only reason I thought that could have worked was that the oil return lines to the turbo scavenge pump were now at a higher level than the front of the crankcase leading to oil pooling away from the return lines.
d). The experiences of s14kev and his tests.
I figured the cause had to be oil flowing back into the crankcase from somewhere and from there flooding the heads, turbos etc. With resultant smoke and oil leaks.
Conventional wisdom was that the small oil filter was the culprit. The fault lying with either a Non-original filter (mahle was named as a brand not to touch) or an original filter screwed on too tight thus damaging the check valve in the filter. But I couldn’t see how this could be given the volume of oil I was seeing back flowing. This couldn’t have been residual oil in the upper areas of the engine and if it was a faulty check valve/filter, oil would have flowed back into the oil tank through the high pressure side of the oil pump, not into the crankcase itself. It had to be flowing from the oil tank, maybe even oil cooler, back through the oil thermostat, back through the low pressure or scavenge side of the main oil pump directly into the crankcase; the most direct path. The oil level there would then rise and then siphon back to the Turbo sumps etc. Why was this problem not present when our cars were new? New seals in the thermostat, tighter clearances in the oil pump perhaps; I’m not sure but one other clue was that those who used thicker oils seemed to suffer this problem less. P-traps, check valves in the turbo oil return lines all would help somewhat but did not address the root cause, oil returning to the crankcase from the oil tank.
That was my guess. And research backed this out as the earlier gen air cooled cars all had oil check valve solutions to prevent oil from migrating back into the engine from the oil tank.
I went ahead and ordered a brass non return valve rated for oil at high temps. Some of these valves came with plastic plungers, I made sure I ordered one that was fully constructed in brass. I also ordered the appropriate connectors/fittings then took a deep breath and…. Drained the oil from my car and cut the oil return pipe leading from the scavenge pump back to the oil tank. Noting yet again the large volume of oil present in the crankcase and piping. I then plumbed in the fitting with the connectors and high pressure/high temp oil resistant hose. Making absolutely sure of the orientation of the valve (to the point of paranoia!) Pics attached.
I then installed fresh filters (topped off with a bit of oil for the small filter) refilled my engine with fresh oil (Mobil 1 V Twin 20W-50 for those who wish to know) then turned the engine over for about 30 seconds with the DME relay pulled. I then stuck the DME relay back in, held my breath, said my prayers and fired her up. The oil pressure gauge immediately peaked at over 5 bar (eliciting a huge sigh of relief) and she settled down into a nice idle. I let her idle and had a good look around. No leaks and all looked good. Took her out for a test spin and she ran fine with oil pressure readings all good. Once fully warmed I was getting around 1.5 bar at idle (I’m in the tropics). So all good. Topped off oil till the halfway mark on the gauge.
The above was done over 3 months ago. I drive her even less now because of our lockdown but even leaving her standing for over week, she fires up without a hint of smoke when it used to be smoke screens after just a few days. No oil drips/wet turbos and it even appears the valve covers appear drier (they weren’t that wet before). And another thing I have noted which lends credence to my theory? When I turn my ignition on, without firing her up, I actually get an oil level reading on my gauge. Below the true level (it rises once running and fully warmed up) but a level is present. Meaning the tank hasn’t drained back to the point where there was no reading.
So there you have it. My experience and solution. If you intend to explore this on your own, do proceed at your own risk etc. Purchase good quality components, fittings etc and make sure everything is installed securely. While the line the valve goes into is on the scavenge/low pressure side, there is some pressure there and the last thing you want is to blow out a connection. When you cut the oil line, use a rolling pipe cutter not a saw so you don’t get swarf. Flare the ends. Clean everything meticulously before fitting. Oh, and do make sure you fit the check valve the right way around.
Thanks for an informative and in-depth write up. I really like your caveats and things to be aware of/careful of during the process. It's interesting that the check valve has appeared to solve the problem. I've talked to more than one mechanic/tuner that has said that the check valve won't solve the problem. It appears that they may be wrong.
On my TT, I was getting a lot of oil in the intercooler, although no smoking on start up. I installed a check valve, although in a different location and it appears to have gotten to the root of the problem in the intercooler. Last time I took it off, the drainage out of the intercooler was significantly reduced from several ounces down to about 1/4 ounce. (estimates).
As usual, just my opinion....I could be wrong and ymmv.
No worries JC. Thought I'd put this out there. Interesting that some mechs and tuners said a check valve in the crankcase/oil tank return line wouldn't solve the problem. But looking at how this problem has been raised for over 20 years and the solutions that were offered, unsurprising. The check valves in the oil feed and return lines for instance. The oil feed lines would not be feeding the turbos once the engine was shut off. Whatever came through would be residual oil in the line and upper engine areas. But my experience (and DiegoR's test) showed almost half a liter of oil come off the turbo's when parked. That could not be residual oil.
Oil return lines from the turbo to the turbo scavenge pump? Here too, the volume of oil should be minimal, not the volumes we were seeing. It would have to be oil siphoning back into the turbos from the crankcase meaning a high oil level in the crankcase (which others have noted is solved if you raised the rear of our cars such that the oil in the case pools away from those return lines). We should not be having high oil levels inside our engines as they are dry sumped with a twin stage pump both scavenging the case back to the oil tank and delivering pressurised oil from the tank back to the engine. Faulty, miss installed small oil filter? By my estimates, oil could not be siphoning from the oil tank up through the small oil filter and then down into the crankcase as that oil filter was at a higher level than the oil level in the oil tank. If it flowed in reverse through a faulty check valve in the filter, that oil would return to the tank, not the engine.
The slug of oil that others have noted in the turbo sump breather lines? If the turbo sumps are full of oil (and they shouldn't be), a vacuum in the breathers would suck oil up for sure. I think that breather is there to ensure that there is a vacuum there on startup so oil can be fed into the turbos better (to assist the turbo scavenge pumps create the negative pressure/vacuum needed to allow fresh oil in). That's my theory. But that vent line is not meant to sit below the oil level in that sump (it is located at a high point). It's meant to draw air and create a vacuum there.
The final clue was that there is a check valve solution for that oil return line from the engine to the oil tank to prevent this very problem on earlier gen oil cooled P cars, that of back flow of oil from the tank back into the crankcase. Non turbo cars but, high oil levels in the crankcase would lead to oil migrating past rings into cylinders, heads etc. as oil pools at the bottom of our parked and cooling boxer engines. Resulting in those cars burning that oil and smoking on startup as well. Which would explain why having check valves/p trap oil lines in the turbo oil return lines on turbo cars have a somewhat hit and miss success rate for many. You solve the problem of oil back flowing into the turbos but not oil rising in your crankcase and burning off elsewhere.
I will keep monitoring my car and report back the instant I start seeing smoke and/or weeping turbos. Maybe drain the turbo sumps and note the volume of oil there but so far all good!
That's a unique solution Krish and one I had not seen before. Thanks for the write up. Where did you source the check valve? Do you have specs for it, most importantly opening pressure and flow coefficient since this is a low pressure line? Most of the ones I've used have Buna-N or PTFE seals. Buna-N seems good to 250 degrees F and PTFE to a little over 300.
More interesting is that both Krish and Joe report that their cars have been chronic smokers/drippers for decades. The theory of rapid turbo seal damage from leaking turbos doesn't seem to hold true. Krish's turbos have been leaking for 15+ yrs and now are fine with the recent addition of a scavenge return line check valve. Previous theories would have suggested that the leaking oil would turn to diamonds (?) and damage the sealing surfaces.
No worries Kev. I sourced a valve similar to this from a local supplier https://tameson.com/valves/check-val...58-363psi.html . I made sure the spring loaded plate is all brass as stated earlier. The sealing ring is NBR which is oil resistant.
Quick correction on your observation of my post, I did note my Turbo's were smoking and dripping around 15 years ago but they were checked and found to be good. I only replaced them 3 years ago however. In those 3 years though they did smoke pretty much frequently on startup if left standing for long and got to the point that if I drove her to work, she'd even smoke on startup before the drive home. Parked for just 8 hours. Maddeningly random as others have noted but fairly frequent. After my mod, I left her for a week parked (despite wanting to drive her) to see how effective the mod was. No smoke, no oil drips.
One other thing I forgot to add. With the backflow of oil into the crankcase from the oil tank and thence to the turbo sumps, once the sumps and breathers are full of oil, guess what happens on startup? You get fresh high pressure oil entering the CHRA (core). That oil has nowhere to go as the sump is full and the vent lines are blocked. Oil is incompressible so it'll just blow out past the seals and enter both the compressor and turbine chambers.
An update. I hadn't driven her for over a week. Not a drop of oil coming from the Turbos, not a hint of smoke from the exhausts. She used to smoke like a chimney before. I decided to make the fix more permanent as the hoses were only temporary. I got a pair of threaded brass compression fittings (3/4" again) with a new check valve. Fitted everything together with some Hylomar blue for insurance (on the threads only). Took the fitted check valve assembly out and installed the new assembly onto the piping. Gave it a once over then started her up and checked to make sure no oil leaks. Will update this forum again if she ever starts to drip/smoke again.
Would be interesting to try and replicate this on another car to test it out. krish is this something you would be able to sell a few of us the parts you used and let some folks try it out?
You could probably find similar in your local hardware if you looked. Sizing should be for 3/4 inch tubing (18+mm was what I digitally calipered) which is what's on the car.
Update on my fix, she hasn’t smoked a day since (and it’s been three months). Remarkable as she was a frequent smoker before. No oil drips at the turbos either. When I turn my ignition on and before turning her over, I actually note an oil level on my gauge indicating oil in the tank and not elsewhere. I had the intercooler off last week and it was clean as was all the plumbing. Emptying the turbo sumps yielded no more than a couple of tablespoons of oil. I honestly think I found the Rosetta stone here in terms of a solution for our cars burning oil.
The usual cares and caveats apply. Make sure everything is meticulously clean as you don’t want dirt and debris getting into the oil lines. I suspect oil from that line does a first pass through the large oil filter but still, keep it clean. Tighten everything well on assembly but resist over tightening. They are brass fittings. A feel thing. Make sure the check valve is orientated CORRECTLY. This cannot be stressed enough as doing this incorrectly could be catastrophic. Finally, when you do a first start, check that you have almost instant oil pressure. Do not drive your car immediately but check for oil leaks and give your car a once over. Only when sure do you then take her out for spin. And of course proceed at your own risk on this fix.
I totally forgot to come back at this thread to update it. The check valves in the oil feed of the turbos, we re-worked them with a custom higher pressure spring inside. If memeory serve me right, we ended up to about 6-8 PSI of oil pressure to free up the check valves and flow into the turbos. Witht his trick, the problem was solved for life (or so far so good, no more oil floods).
So basically my ex turbo has check valves for the feed and for the return to the scavenage pump. Problem solved.
04-19-2021, 09:57 PM
