Ed, thanks for commenting..

I hate to post another thread on this topic, but this is one that covers territory I haven't read before--sort of a review of what the condition was of the original bearing when it was replaced and some info on the origins/Porsche use of the bearing itself.

One poster claims that the bearing mfg.--NSK--sold the the bearing to Porsche unlubed and with one side of the plastic bearing shield installed.
Porsche then supposedly filled it with their own lube and added the second shield. There's also some discussion the amount of lube in the bearing was not meant to completely fill the cavity and that the installation process of the second seal (not tearing the seal when installing it) might have been critical to prolonging the life of the bearing and explaining why some fail at 30K and others go 150K+.

Any way..
[forums.pelicanparts.com]

and there's something funny going on if someone needs a new clutch at 30k.

In any case, a few anecdotes are interesting but not worthy of any broad sweeping conclusion.

Agreed--these are just a few more "tea leaves" to look at on this is issue, but it struck me a bit odd that Porsche would buy a bearing assembly from a supplier, then add lube and fit the other side shield in house, if the report is accurate.

Wouldn't it make more sense to have NKG do the entire process in their shop and then ship the finished assembly to Porsche? I would think--if they were comfortable with this setup--they could furnish the entire assembly, with what ever special lube Porsche wanted inside, and the second shield added as a finished product. They'd be responsible for the product start to finish and Porsche has a finished product in inventory and invests no more time and money to get it ready for installation.

Hydrogen in the bearing exploding and damaging the seal? I think not. Hydrogen is very reactive and would combine with the oil/grease. Bearing seals have a steel insert and do not come out of a bearing easily. As for high pressure in the crankcase, modern engine crankcases are designed to have a small vacuum. Think air-oil separator. If you have a pressurized crankcase you have other problems.
No bearing seal is perfect. Too high a seal pressure and the seal lip will overheat and wear. Too little pressure and they will leak. We are talking tolerances in the thousandths of an inch. End play in the bearing only makes it worse.

If Porsche is doing their own lube, the tools for inserting seals are simple and don't damage the seals.

1/3 to 1/2 full of grease is the standard for grease filled bearings. Full pack is OK for some bearings, (wheel bearings) but death to higher speed bearings. They will overheat. (So will some wheel bearings)
Ball bearings such as the IMS bearing need very little lubrication. It's as much for cooling as for lubrication. There should be enough oil spray and oil vapor in the crankcase for the IMS bearing. Grease wears out.If you can't replace it, it's all over. Oil gets contaminated; that's why we change it.

Of all the IMS bearings I have analyzed, The remaining lube is black with worn seal particles and overheated, (burned) lube. Unless the bearing has disintegrated, the seals are intact.

Jake Raby has seen oil temperatures as high as 270F. The average commercial bearing is OK to about 250F. The steel will begin to soften and wear will increase with repeated temperatures over 250F. Grease doesn't help.
A constant spray of clean oil will help.

Have I made my point?

Ed B

Bearing engineer

I was hoping you'd see the thread and offer some observations based on your expertise.

From your comments, it would seem that the Porsche IMS assembly (race bearing packed in grease of some sort, mostly sealed) wasn't exactly an ideal solution for its intended purpose, given the operating environment that it had to function in. The failure of this bearing due to heat/wear from improper lubrication makes sense.

Having said that, I'm still puzzled when I read about how this same setup has been able to function in cars with high miles and track use (like Pedro's, which IIRC, was intact with it's seals but I can't remember if it was full of oil or not when removed) but fail prematurely on low mileage cars, driven sedately. Do you think the high mileage IMS bearings are preserved by high revs, pumping more engine oil into the bearing assembly or something else?

Finally, I've read where the bearing on the other end of the IMS is a traditional shell bearing, rather than a race. Is that true, does it rely on crankcase oil for lubrication and why wouldn't that be also used to support the flywheel side of the shaft?

thanks in advance for any additional help.

Higher speeds do seem to help the life of the IMS bearing. The additional oil spray can't hurt either.

Bearing life can be predicted statistically based on degree of risk. Tolerances, finishes, hardness, lubrication, etc. All have an effect on life.

While the IMS bearing is not a high speed bearing, centrifugal force on the balls will take some of the load off the inner ring, which is the more highly stressed ring. Also, as the speed increases, the balls will tend to ride up on the oil film, (think hydroplaning), limiting the metal/metal contact between the balls and ball grooves. This is called EHD. (Elastohydrodynamic lubrication)
This is a two way street. Too much EHD and the balls will stop turning and skid, resulting in wear and debris. A little slippage is OK. There are complex computer programs that will analyze these conditions.
The oil film is very thin so the peaks of the surface finish can pierce the film and make contact with the balls producing local high temps and small particles. (Think a grinding wheel) This is another reason the oil in a failed bearing is dark and smells burned

Other things to consider are stress and deflection. Ball bearings can operate at up to 350000 psi Hertz stress. The steel of the balls and ball grooves will be flexing and will eventually crack. If the stresses are low enough, and with good lubrication the bearing can last indefinitely.

The IMS bearing sees varying temps, loads, speeds, accelerations and marginal lubrication. A difficult application. There is always the cost factor.

The other end of the IMS is supported by a shell bearing and would have to be pressure lubricated by the engine oil pump.
There is very little room to put a shell bearing on the flywheel end of the IMS. I have a crankcase half and an IMS. I'll try to post some photos.

Ed B