alexb76

Oh really? So, yours is a 2006? I thought the 3.6s had the old design, my bad then.

X51

M96 accessible without splitting the block - had mine replaced on my '02 with little effort.

utkinpol

M96 is just a name, it went over all same internal mods as M97. both M97 and M96 in pre-'06 had same replicable design. after 2006 both changed. it is same motor pretty much, only parts different are those that are dependent of displacement diffs. there were fair share of '05 M97 reported that went kaboom due to IMS. so - ...

Palmbeacher

A lot of what Bill Ryan said corroborates what I've been thinking and speculating for some time. For one, that the large ('06-'08) single-race bearing, though it may have the same load rating as the dual-race, introduces the issue of centrifugal force as a potential detractor. Any out-of-true will be magnified compared to a smaller-diameter bearing.

Another thing where he corroborates my thinking is that the failures LN has experienced with their small-single-row bearing is probably the result of fitment variables (either bodged installation, or pre-existing out-of-tolerance specs in the IMS shaft itself, either or both of which would contribute to less-than-ideal loading). Clearly to me, the small single-row bearing would be more sensitive to that than the dual-row which has a higher load-rating.

What I found the most gratifying about his discussion was that he backs up his hypotheses with engineering science.

utkinpol

it was not exactly a new info but it was well summarized.
all i can do for it is running 15w50 oil or 50/50% mix of 15w50 with 5w50 and always add a qt can of ZDDP additive mix. and change oil often enough during DE season.

Palmbeacher

From what I've gleaned from numerous people in the racing industry, the notion that someone would track his daily-driver on weekends and expect it to last 200K with nothing more than "severe duty maintenance", is utterly daft. Race engines, they say, get torn down and rebuilt proactively with some regularity.

Neotorque

Thanks for posting this. If accurate, there are two interesting aspects of this analysis that would appear to support the anecdotal reports that tiptronics are far less susceptible to IMS failure than manuals. First, there's the indication that sustained hard driving can lead to IMS damage. And since there's no risk to red-lining in a tip because the computer won't let you, there's less risk of sustained hard driving. Second, because a tip is more likely to be the choice of someone looking for a daily driver, it's less likely to be a garage queen that sits around for weeks/months at a time. And of course, these two things probably work in concert. It's hard to imagine a worse program for engine health than sitting around for months at a time and then being driven hard.

I've also read that it has something to do with the way the engine is mounted differently for an automatic transmission, but not being an engineer, I have no idea whether that is even true, let alone could make a difference. But the first two points seem very plausible.

Palmbeacher

Bill Ryan mentions that only in a comparative reference between the older, smaller-diameter bearings and the final, larger-diameter bearing in the 06-08. Anecdotal reports indicate that higher RPMs correlate to less IMS failure in the dual and small-single race bearings.

How to you get to that conclusion? You can drive a tip all day long with the revs right under redline. If there's anything protective in the tip, it has to be the absence of risk of lugging the engine.

I'd like to see some real data that supports that assertion.

My own belief, which I'll hold until someone shows me solid data to disprove it, is that the smaller number of IMS failures in tips relates mainly to the smaller number of tips sold in relation to manuals. Until PDK, 6-speeds out-sold tips by a huge margin.

Neotorque

Sorry, I mispoke (before coffee). I meant to refer to the risk of lugging.

And, you're right, I don't have any data to support the assertion that a tip is more likely to be a daily driver. It's a hunch, one that may be incorrect, based on my guess that no one buys a tip to be a track car. Pretty much the only advantage of it that would justify the added cost is for people who have to sit in traffic all day.

Palmbeacher

The PDK is clearly a better transmission for manual shifting than Tiptronic was, so you're probably right about the 6-speed being preferential for tracking. But away from PCA gatherings and enthusiast forums like this, I daresay the majority of 997 buyers you would encounter most likely do not track their car, regardless of transmission type.

Neotorque

Agreed. But let's say 99% of manual owners don't track their cars, and 99.9% of tip owners don't track their cars. When we're talking about an overall fail rate that's likely below 1%, the difference could be meaningful.

Again, all rank speculation, I admit, based on the highly speculative conclusion that anyone who bought a tip did it because the car will be their DD and they want it for ease of use in traffic.

pissedpuppy

does casperlabs post here? they should

Palmbeacher

I happen to think it's absolutely preposterous to believe that meaningful, reliable quantitative conclusions on failure rates can be extrapolated from any internet data currently available on these cars, regardless of the source. The sampling is much too small even if there were no biases, agendas or other variables to take into account...and there are a ton of them.

The best we can say is that even if we can't accurately guage the probability of IMS failure, we should agree it is a significant possibility, the consequences of which are severe enough to warrant addressing the issue. That is why well-informed discussion such as on the Renntech thread I linked are so valuable. We've heard a lot of theorising and speculation from mechanics and entrepreneurs, most of which are conclusions drawn from anecdotal experience and the application of intuition to a cursory understanding of engineering concepts. Even the world's best construction foreman telling us why he thinks the towers collapsed on 9/11 would carry less weight in my estimation, than hearing it from an architect. And I know nothing of building engineering. I do know of mechanical, automotive engineering and the explanations and even the speculations Mr. Ryan offers in regards to IMS failure are supported by the science.

Neotorque

I have no background in mechanical engineering (or any other sort of engineering) and don't have any reason to doubt him or you. My suppositions about the fail rates are not based on anecdotal internet data, but rather on the prices charged by extended warranty companies, who presumably do have a sufficient sample of data upon which to base their judgments. Based upon the going rates for powertrain warranties for M96/M97 engines vs the price of a replacement engine, it seems reasonable to surmise that the fail rate must be extremely small, and almost certainly below 1%.

Is it possible that the extended warranty companies are underestimating the risk of failure? Sure. But that's generally not how those companies have managed to stay in business.

I don't dispute for a second that, even if the possibility of an IMS failure is very low, the consequences may justify taking some preventative action, even if solely for peace of mind.

DDL24

Porsche sold anywhere from 9000 to 12,000 911s and 4-11k boxster/caymans annually in north american market between 2003-2006. see sources below. total about 75k, if including 99-2002, number can easily exceed 100,000 combined. of course you have to take out turbros and GTs but i don't know the sales mix but shouldn't be a huge number. http://en.wikipedia.org/wiki/Porsche#cite_note-51

even 1-2% is couple of thousand of engines/kits. what number does LN need to breakeven? and this doesn't count int'l market at all.