dr kenneth e garchow

i have been researching the "smoke on start up" issue w/993 TT's and want to see if i understand what the different schools of thought are.

other than the oil filter and oil fill problems i think the following is a simple rundown of some fixes

if the turbo seals were perfect (and engine was fine) there would be no start-up smoke after long sitting periods. and nothing else would be needed. (one school)?

seals are not perfect (although kevins almost are), so one might need further prevention to stop the oil getting by the seals, so the 996 type anti bleed valves may help (another school)?

or there is a oil line kit that puts a high spot in the line to help stop oil running back to the turbos @ rest ,also helps to stop the smoke (yet another school)?

am i getting the idea?

is there any new information?

is there a consensus as to what way to go?

thanks

Kevin

In a perfect world we would build a engine that would address the 3 major issues that you described. I will give the details as to what is happening.

1) KKK is manufacturing turbochargers that have a very large ring gap. I have measured new bearing housings with new sealing rings in them and the gap typically is around .005" to .006" When you reach .007" you are out of spec. To give you an idea of the measurement, we adjust the 911-930 valvetrain at .004, you can rock the rocker arm back and forth and hear the lash.. So what I am eluding to is that the gap is near the wear limit and depending on driving conditions one can wear the seal out quicker than bearing wear. In fact on the K16 or K24's I rarely see bearings worn past specs. In order to fix this one has to machine his own sealing rings and install them at tighter specs.. I take them down to .002 on the turbine side... If I know of a cronic leaker, I may close it tighter. This is not simple magic, this is hand filing and hand fitting with feeler gauges.. It does take time.. for this process.. When gapping the rings like this, one can be almost certain to prevent any leaks.. However, there is one failure that causes me sleepless nights.. Being that the tolorance is so tight, the gap is placed at 12' o'clock. In rare instances, the ring can rotate and if it rolls past 3, we will see a leaking seal. This is very rare, it will happen to the person who does not break the turbo's in for a period of time. Standing on boost right after install is a good way of rotating the seals. Hot shut downs will kill the sealing ring due to the heat and build up of coke.. Carbon will cut the sealing ring dimensions.

2) Starting back with the 930, Porsche installed a check ball to prevent the oil from siphoning back into the engine and also back into the turbo. The 993TT does not have this check ball, instead it has a serpentine loop in the oil inlet line. This was designed to prevent the siphoning, like a upside down P-trap. Well this didn't quite cut it, and the 996TT came out with a full blown check valve which is screwed right into the bearing housing. This is used with the..

3) Humped scavenge lines that run from the oil sumps to the central oil scavenge pump. This also acts like a p-trap, with the addition of the check valve, it stops the movement through the turbo and keeps the oil from the engine case from backing up into the turbo. I think that in order to get the total fix, one must do both the mods.

For most of us, a turbo rebuild with tightened seals will keep things at bay. If you don't cool your turbo's, the seals will fail, regardless.

I need to describe a little bit why the engine oil level, causes problems with the turbo. We all know that the turbo's sit below the oil level line. Due to gravity, the oil will try to seek its level, this can be dramatically induced if the car is parked on an incline. The elevated oil level is now higher than the turbine wheel in the turbo, thus finding the sealing ring, even at the 12'o'clock position. Oil will start to seep out... The wider the gap, more oil will leak out.. Thus the puff on startup...!!!

Turbochargers will not heal themselves. Carbon will not be displaced. If the oil is present, it will be burnt off and carbon will form. The smoke might not be present, but the carbon is there and it will cut the seal like a diamond...

Jean

Kevin, writing a book is long overdue..

Thanks for the explanation.
Jean

dr kenneth e garchow

kevin

thanks for taking the time to put that in writing.

its good to have a factual summary

your info should be in a FAQ section for easy retreval

i think the smoke is finaly starting to clear for me!!

now back to my detail

then check out my turbos

hatchy

Great info, thanks Kevin. Given the design, umm, features of the K turbos, what is your opinion of turbos like the Garretts or Mitsu for the 993 or 996. Aside from the plumbing issue, do you think it is a good alternative?

Kevin

When built correctly the KKK's are very durable. However, these issues will not change even with Garretts and Mitsubishi's. Infact, most require H2O cooling for the bearing housings! What are you going to do for that.. Garrett's don't last on these cars unless you spend many hours modifying them. I'd spend the time and money dealing with the problems at hand and keeping the KKK's.