Flat6 Innovations

Nope, I've seen it many times, first around 2007. We see it, I'd guess 3-5 times per year.

Ahsai

Jake, is there an easy way to tell (if you could share) whether the tensioner failed first or the bearing failed first? A failed bearing will allow the IMS to wobble and that would cause the IMS chain to stretch, and in turn allowed the tensioner to over extend?

Flat6 Innovations

Now that it's all apart, no way. This has to be determined during disassembly, through very careful inspection of where debris ends up, and other clues.

Sometimes it takes me two days to determine things like this, and sometimes it's damn near impossible to conclusively determine what happened first, the chicken or the egg.

Ahsai

That makes perfect sense. Thanks!

docmirror

If these run-out numbers are taken from shafts that came off the assembly line, there's something wildly wrong with the mfg process for this shaft. Or, the QA isn't being done. I can't see any way that normal wear on the engine would cause this kind of run-out within 100k miles in an otherwise normal engine. I would say that the run-out for the shaft should be +/-0.0005" for that size shaft when new, and at end-of-life, the shaft should not exceed +/-0.002" in worst case.

I would like to see some tests from brand new Porsche part IMS shafts, as you said it's hard to quantify if this is a chicken or egg issue, but if these run-out numbers are for real, I can see where all the catastrophic damage is coming from.

Flat6 Innovations

The brand new, M97 shafts are the best for runout, even after use.

These results are very real. Some came from shafts that had lost one or more IMS bearings, others were just good shafts that had not seen a failed bearing, and were being evaluated for use in an engine reconstruction.

With this, we've quantified yet another contributing factor to IMS Bearing failure. This one is the biggest of all; because, it can't be measured without engine disassembly.

So, this one is most worrisome with a single row bearing, that has much less load carrying capability than any other, and will fall prey to the runout monster most. With runout the bearing is working to control not only longitudinal and lateral thrust, but also to control an elliptical condition, where the shaft is not turning true. This creates vibration as well, which helps to weaken the cage.

I first noted this when developing the IMS Solution, as one engine always had a small polished spot on the face of the IMS Solution flange, and we worked for the better part of a year to eliminate that. We changed hardness, changed the coating on the flange, and even changed where the oil holes were positioned in the flange, but nothing ever omitted it completely. Finally I disassembled the engine and found the shaft had runout. We swapped the shaft and that polished spot never came back. Nothing else was changed, we just wasted a year of our lives.

That said, the IMS Solution handles this runout with no problem, and through that development we learned that the wear will only lead to one area of the flange that sees a polished surface in the coating, no measurable wear existed.EVER! The IMS Solution is an over built, super heavy duty part that can carry this load, and control a shaft with moderate runout. We've already proven that, even though we had no idea that we were doing that, at the time.

Lots more could be posted about this, but we'll save that for a TSD later down the road. This further proves that "lubrication is not the problem". Throw as many oiling devices that you want to at the engine, and if the shaft has runout, it'll still kill the bearing. Mechanically a single row bearing cannot live through it.

Also, note where some of the runout is... The main timing chain drive for bank 1 camshafts.... So, we also learned why we see bank 1 main timing chain failures. Ding...Ding...Ding.

Knowledge and experience like this is what you receive inside my engine. Been there, solved that.

WalterRohrl

Jake, do the factory remanufactured M96 engines that were made and installed as replacements into the 996.2 during the 997 era use the M97 shaft or just the larger bearing? I realize that these engines (and the M97's) seem to have less issues but is that a large part of the reason why? Thanks in advance for any insight.

Flat6 Innovations

Only if remanufactured post 2006. After 2006 ALL engines were remanufactured with the M97 larger diameter IMS Bearing.

TAGPorsche

So it's really a bad shaft not a bad bearing.
Would this runout also cause premature failure of the timing chain tensioner?
Seen more posts about that then the bearing actually.

Flat6 Innovations

We can't generalize. This is just ONE of over two dozen potential contributing factors that add up to equal IMS Bearing failures.

This finding was the key to understanding why some engines lose more than one IMS Bearing.

Schnell Gelb

Perhaps there is a wide variation in run-out of the shaft and at different locations along it's length ? We have just discussed the outliers - the majority of shafts that are acceptable run-out cause no short-term failures .The extreme ones Jake had kindly mentioned do. But what about all the in-betweens ?
For the in-betweens - the vibration may not be severe enough to be immediately catastrophic but it must surely shorten the life span of the IMS bearing and perhaps other components in the IMS system ?Once the IMSB does wear ,that too will cause even more vibration.
So this vibration increases the LOAD way beyond the design limits of the single-row OEM bearing(not a lubrication problem).
So a bearing with a much higher load capacity may be a kind of insurance against run-out? If so, doesn't this favor using an RND roller bearing instead of a Classic Single row or even Single(staggered 2 row) Row Pro ?
I am not trying to start an IMSB flame war.It just seems logical ?

docmirror

Jeez,,,, this is sad. And, again - thank you for the research and documentation of this stuff. I know it takes time, and effort but the results must be satisfying to find so many root cause issues that combine to cause failures.

Scary to think that back in the 80s Porsche built aircraft engines(not many!). I'm sure the tolerances were much tighter, but the overall picture of this failure type sure makes it look like all those who wail about the bean counter take over in late 80s caused a lot of heartburn.

Things I now know, that I can't unlearn about my new car. Gulp.

Flat6 Innovations

While anything has more load carrying capability than an OEM single row bearing, until you remove all the wear parts from the equation, nothing "comes to the rescue".

The IMS Solution doesn't care about these types of vibrations or even moderate runout. That's why it is, and will always remain, The IMS Solution.

dgjks6

Compliments to all. This last discussion has been very informative without all of the usual drama. I learned a lot reading these last few posts.

docmirror

I used to do a lot of QA for massive disk drives built by IBM back in the 70s and 80s. These platters were ~26" in diameter, so as you can imagine, we could not tolerate any run-out or suffer constant head crashes. Well, after a few years, even though we specced a real good set of bearings, and very true shafts - we started to have HDI problems and discovered that yes, run-out was the problem. Led to a huge, massive costly recall or field replacement of many bearings.

The takeway from this little story is that NONE of those shafts should be in customer hands, or engines. Evar. End of story. If there's a IMS out there with .004" run-out it's junk. Even .002" run-out on that length of shaft is marginal. Anything over .001" would be a reject.

Which leads to the conclusion that Porsche wasn't doing any cont process improvement(sustaining engineering) once the part was on the mill, that was the end of that and whatever came off the mill(machining IMS), was put in an engine. There's just no excuse for a part variation allowance like this from a major maker of high perf autos. No excuse at all, and yet we have examples in that photo array of I would say 5-7 parts that are in the field and are going to blow up an engine if installed. No excuse(did I already mention that?)