Flat6 Innovations

I've had a handful of bearing failure reports that happened after a vehicle has crashed... But there's no rhyme or reason with it. It's too far out for speculation too..

The studies continue. Maybe one day we'll be able to put the whole puzzle together.

alpine003

As I knew it would happen in good ole RL fashion, let's get back to the original topic.

Jake has made some observations that may or may not affect most of the engines. It is unknown at this point.

His tool won't fix the issue but merely diagnose it. There will be a certain amount of pissed off owners that will be turned away when trying to get a retrofit done. The ones that don't get turned away by shops that either do not carry out the prequalification or does it but wants the extra business of installing IMS, could have a false sense of security and/or suffer the consequences later.

When is that "later" is the million dollar question.

If my doctor told me I could get a certain cancer when I'm 90yo wouldn't bother me as much as if he said 30 for example,

Also do we know if other Porsche engines have always had some off center tolerance issues too?

wildbilly32

Jake: I echo what gnat said above...Thank You for sharing your knowledge and experience. I have not read the "other guys" deal and don't plan to. There will always be professional jealousy but it is ashamed when it turns into attacks. I appreciate that you speak with facts and not emotion.

Gonzo911

I would think that a misaligned IMSB would fail pretty quickly. Cars that are driven often and "like you stole it" should have failed faster than low mile garage queens. This goes against the conventional "wisdom" that is often mentioned around here.

Jake...if the IMSB Retrofit is not a true 50k mile fix for some cars, is the IMS Solution, the solution?

Imo000

Since we are doing the guessing game, the bearing flange could just as easily not be at a right angle to the block as the block bore being out of alignment. Someone needs to make some careful measurements of everything before coming to a conclusion. Until then we are all just guessing. However, the original bearing lasting longer than the replacement is an interesting point.

Flat6 Innovations

It definitely doesn't effect all engines, or even most of them. It is noted enough to warrant attention, and to create concern. I first noted this when developing the IMS Solution, and it made me chase my tail for a while before I figured it out. I had one of the test engines that created localized wear that was ever so slight on the flange, but none of the others had this. One late night I just happened to take not where it was occurring, and decided to break out the measurement tools. Thats when I found the issue. Every since then we've been watching for this and taking measurements during pre-qual.

Trust me, they will be. I have had a guy ship me a car from Canada that failed the pre- qual inspection and we refused to do the work, because the car wasn't qualified. He was all sorts of pissy when I presented him with the data and explained it. That makes no sense, because in the end we saved him lots of money and we caught an issue that was repairable before the engine failed. What did he do? Called car max and told them to come pick his car up here, he basically threw it away.

When we do any retrofit there's no chance of failure, because unlike any other shop, we can't simply blame the part for the failure. Here we have to know the case is a solid one and that the engine will not have any issues moving forward, no matter what. Thats how we've done over 400 retrofits and haven't ever had one fail. We take no chances, and moreover we know how to pre- qual the engine, because I developed that procedure and the associated tools.

.
The real issue is the techs. Other shops pay people on commission (we don't) and those people don't get paid if they don't produce. All too often the tech will pull the sump plate and find debris or other issues and they wont's anything, because thats their paycheck at the end of the week. Its not their shop and its not their product, so they don't care. They have nothing to lose, except money at the end of the week if they turn the engine down.

Its easier to be a victim than a failure, so if the engine that was not pre- qualified ends up failing, the shop just says "sorry, not our problem". This is what happens today, and it comes more from the shops that have done A LOT of IMSR procedures as they have learned how to cut corners and do the job cheaper. Thats why you NEVER, EVER price shop for this procedure. If you do, you will get exactly what you pay for, plus a bonus that you won't like.

If you have precision measuring devices and if you know how to use them well enough, you'll find a gripe with every engine that you measure. Just like when I crewed aircraft, I could down any aircraft on the flightily if my pre- flight was extensive enough. Some days it was better to be in the air than on the ground, so we had to look at the gripe and know if we were going to be at risk of falling out of the sky, or not. There were lots of those days. I got lucky.

No matter what, it is.. I am totally biased on this, but of all my developments the invention, development and distribution of my IMS Solution is the one I am most proud of. The Solution is the most forgiving of debris laden oil, misalignment and collateral damages stemming from other issues. Removal of all the dynamic components is the key, each of those bring with it another wear point and another failure point.

JD ARTHUR

I sent my car all the way from Las Vegas to Jakes shop within days of owning it. I read this site right AFTER I got the car. I was not looking to buy a Porsche of any kind but I took the car on trade towards another car I was selling. I knew absolutely nothing about its value or the issues that I soon discovered. I did not know anything really about the different model Porsche's but I always liked the way they looked and was under the impression that they had similar motors to a Volkswagon. I didn't drive my car until AFTER I took it on trade. I was planning on selling it as it only had 18000 miles on it and looked and smelled new. I knew the owner and remember when he bought the car brand new. I also knew he had some Ferrari's and other vehicles and that the Porsche would be just another garage queen that would get the best of care but would driven very little. In any case I liked the way it looked but after driving it I really liked it even more and I thought I should keep this car instead of selling it. After reading this site and still making the decision to keep it and to have the issues dealt with for peace of mind I decided to contact Flat Six. I liked Jakes attitude on all his posts, he seemed to be the expert and he didn't ***** foot around about anything, he had done much work and was in my book a fanatic. The perfect guy to send your car to if what you were looking for was the peace of mind that your car would be as good as it can be in todays world. The price would surely be higher but so would my peace of mind. I was told by Jake that he never had a failure, he didn't care how little miles were on my car it would still spend a day getting pre qualified and only if it passed that would he do the work. I've never been anything but happy that I sent the car there. Now I realize there is a guy in So Cal named Tony that Jake speaks highly of that also would have given me the peace of mind that I have now if he had done my car. I still am a firm believer that price should be the least of the criteria if a decision is made to do the IMSB, the main thing is to get the job done the best it can be because like we learn here there is lots more to this than a bearing swap.

Barn996

Jake, out of your 400 installs to qualified engines, how many didn't pass your pre-qualification? Just curious.

sds911

+1 - I sleep a little better at night knowing that Tony Callas and his excellent mechanic (Ramon aka "Ray") at Callas Rennsport did my IMS Solution install. I played Russian roulette for about 1 year after my CPO warranty expired, and based on Jake's advice went with the Solution over the retrofit. The Solution is pricier but I did not want to worry about changing it out again (assuming I keep the car for a long time), or with the potential that some third variable would cause even a replacement bearing to fail.

I've got about a week's worth of driving on the Solution at this point and all seems good (including 100 MPH on the bank at Festival of Speed AAA Fontana Race Track during the track tour), and my daily commute of 220 round trip. The Solution seems to make the engine rev a bit higher (e.g., 3200 to 3500 in 5th gear at 70 to 75 MPH) compared to before the install -- or this could just be my imagination. But I like it so far. Now, let's just hope the chain and rod bolts hold.

aftCG

RE: The misalignment issue.

I read through this thread and the one on 986 last night (I'm flaps10 over there). I'd like to add to the discussion if I may with regards to the misalignment issue.

I'm unclear as to why slight variation in the distance between the centerlines of the crank and IMS would cause an issue, since there is a chain with a tensioner between them. Jake, you obviously have observed something there so I'll respectfully leave that to you.

As far as angular differences between the centerlines, that is a discovery with some meat on it. When I first read it I thought, "okay, so we will end up welding up the bores of the crankcase and then align boring them to an agreed upon tolerance. Problem solved.

However, it is way more complicated than the bores in the crank case. This is a situation that as a designer/machinist we refer to as "stack up". It is the collective imperfections of a pile of parts. If you have several parts to bolt together and they are machined to a tolerance of +/- .005" they would, in most cases, be considered pretty damn well machined. In fact if you want parts machined to +/- 0.005" day in and day out you should be prepared to pay some good money. (Not that 0.005" is anything like amazing, but please stick with my blathering for a bit.)

If each of those parts were manufactured to the low end of the tolerance, then after 5 parts you are off 0.025" and that (could) cause a huge problem. In the old days we would have done a lot of machining of assemblies, or shimming. If you are cranking out a car every 120 seconds you don't have time for hand building anything. You need to design intelligently so that tolerance stack ups "wash out" every few parts, but I'm already getting off the track. This is my wheelhouse and I get a little excited. Onward...

To help others visualize my input here, there are a pair of holes in the front and a pair in the back of the crank case. The top pair is for the crankshaft and the lower one is the IMS.

But. Those bores don't center either of those shafts.

Let's talk crankshaft first. The bores in the crankcase have the rear main seal in one end and the front main seal on the other. The seal is a wide gasket. To be sure, if there was significant misalignment it would cause early failure of the Rear Main Seal (RMS), but again, the crank isn't located by that bore. In my perspective an RMS failure is a zit and an IMS failure is *** cancer, so I'm not going to fret about RMS failures in this discussion.

The crankshaft is located by the bearing carrier, which is dowel pinned into the bank 1 crank case. Right there is issue number 1. The bores in the crank case are done in one machining set up, and the locations for the dowel pins are done in another. Any time you take a part off of a machine and put it in another (or unclamp it and rotate it in the same machine) you introduce an opportunity for variation. Before anyone says "but it's done on a CNC!", my response if "pfft". All CNC does is allow you to make the same mistake over and over with shocking repeatability. And if you are running a $70k FADAL it just isn't the same as a $2 million Mazak or Okuma. It just ain't.

Just one chip getting between your clamping surface and your part and you are off. One guy clamps the vise like this, and on second shift the guy does it like that. Journeyman machinist vs. button monkey. We can pretend that Porsche only hires Journeymen.

Now add the bearing carrier itself. That is one beefy block of two kinds of metal. That carrier and the crank appear to me to be the only two things about the M96 that are "hell for stout". That carrier controls the crankshaft centerline, and it has a dowel pin pattern which matches the bank 1 locations.

They pin together very nicely, or did in my case, but the dowel pin pattern can be machined very nicely "to itself", that is, the hole pattern is spot on, but if bank 1 crankcase half or the bearing carrier half was tipped just a tiny bit in the machine where those features were created then everything is going to look just fine.

Let's go to the other shaft, the IMS. What a friggin train wreck. There is a bore on the flywheel end where the famous IMSB lives. It is a bore that is on the centerline split of the crankcase. Not exactly ideal right from the get go. If the case halves aren't "married" and machined as a pair then you run the risk of the bore not being round (picture a football for the exaggerated view), or one perfect half circle could be shifted vertically, resulting in a mismatch. To allow for this the hole has to be oversize. Just a touch, but it is yet another place for error.

We aren't done with that end yet, because we need to add the IMS cover itself. The outside diameter that goes in the crankcase bore has to be undersize or it won't go in the hole. Granted, it is a pretty snug fit, but the IMS cover is held with three bolts. In my years as a tool designer I would use press fit dowel pins to locate two parts as perfectly as my machining processes would allow. Bolts are only used to clamp the parts together.

Now lets look at the other end. Wow. Just wow. The oil pump end of the IMS, you know the one that doesn't cause any problems? It doesn't run directly in the crankcase bore either. It runs in the oil pump housing.

The oil pump housing is a die cast contraption about the size of your face, maybe bigger. It has a "spud" on it that goes into that bore in the crankcase. It has a gap between its outer diameter and the bore in the crankcase that is big enough that there are two O-rings used. There is nothing precision located about the oil pump housing. It is held to the front of the engine (986, rear for you backwards guys) with an array of, what are they M5 bolts? Oh and one M10. Add to that, the chain tensioner for the non flywheel end of the engine spends its entire life trying to pull the nose of that shaft to one side.

The holes in the oil pump housing are a class III fit, that is pretty damn sloppy so that it always fits together. But wait, there's more! This is where they chose to bolt on the front engine mount. In fact one of the four front engine mount bolts goes directly into the oil pump housing. I don't know what they were smoking when they came up with that idea but they probably had to go to Denmark to get it.

There needs to be an added machining process where the oil pump housing and crank case half are modified so they can be pinned together with precision. I'm leaving that laying on the table right now.

My point of all this is that it would be nice to say we could measure the bores of a crankcase and accept or reject it based on what we find. But the only way to get meaningful measurements during and engine build would be the assembly consisting of:

Bank1 crankcase half
The bearing carrier, with main bearings and crank installed
Enough clamping force to hold the bearing carrier socked into place, just like we do for installing pistons 1-3
IMSB cover installed
Oil pump housing installed
IMS shaft installed

Now you have a fighting chance to measure the angularity between the axis of the two.

dgjks6

Just a thought, don't want to start anything but wondering if anyone else is thinking the same thing - flame suit on.

Fact. Most bearings don't fail. What if the problem is a misalignment and not the bearing. And you only replace good bearings on cars with out any misalignment issues or signs of failure. Then, chances are the new bearing (whether is be single, double, roller, pelican, etc.) will last because the first would not have failed. Therefore, a lot of people have replaced perfectly good bearings for no reason. And the failure rate for the replacement bearing will be really low. And lots of money spent on an unneeded repair.

I wonder if Porsche has the answer or whether they don't really care.

sds911

Good question.

Probably multiple causes of bearing failure, including the faulty seals allowing grease to be washed out, the load on the single row, and who knows what else. If it were misalignment only, then why would the 1999 failure rates be so much lower than the 2001 to 2004 (according the the class action settlement statistics 1% for dual row, 8% for single row as of the date of discovery/settlement) -- unless the 1999 MY had a different manufacturing process for the alignment issues....

Think it is multi-factorial and given that even some high mileage cars are now experiencing bearing failure I personally would still buy the relatively cheap insurance of a retrofit rather than a $12 to $20k engine replacement.

Also since Porsche got rid of the IMS bearing in 2009+ after repeated attempts at fixing the issue -- my guess is they really don't know. Since our cars are out of warranty, they also really don't care. A few disgruntled 996 owners can be brushed aside since sales of the new models are still going strong.

DrMEMS

Hey Gnat:
So does that make anything that I said untrue?
I have previously been posting on 6speedonline. Because rennlist concentrates on Porsches, I moved here.

I talked to several mechanics. Two, including one who is well-known for his development work on Porsche engines, suggested running another few hundred miles to see if the problem went away. I saw little downside: if the engine needed a rebuild, it would still need a rebuild, at the risk of the core being worth less. The upside was that if there was no more metal, the engine was probably fine. Gnat, you seem really smart. Tell me how this logic is wrong.

No one could give me a reason that getting rear-ended would cause metal in the oil. In an email from Jake in February, he said that he'd heard of it, but I never got any details. I really wanted the damage to be from the accident, because then the other driver's insurance would pay for a new engine. After my local shop, who removed the engine, and MB, who rebuilt it, looked at it and found nothing linking it to the accident, I dropped the reference. Good job to the detectives on using Google!

Yes, I think that the LNE IMSB is at fault. The races are clearly worn and I'm out over $14K. Sorry Gnat if the truth is incendiary.

Yes, you are totally wrong.
I asked posters to keep this discussion useful. I guess you're not on board with that.

sds911

DrMeMs:

How long after the accident did the metal shavings start to show up?

In the law, there is a concept called "res ipsa loquitor" meaning the thing speaks for itself. If your car was working fine and then an impact occurs and shortly thereafter things aren't working fine anymore -- one can impute the damage to the impact.

I'm no expert when it comes to how these cars are put together, but a hard impact that causes some type of frame damage could damage other vital components from the impact itself -- and your shop and Martin may not have seen any "visual link," to the accident, but what if the impact caused the IMS or stud to wobble or come loose, causing the bearing to go south? (again I'm a lay person -- so maybe this can't happen).

Again, it's hard to understand why your bearing failed if other LN bearings are in service and have not failed. Very hard to quantify since there are many unknowns (such as how many LN bearings are in service and how many miles do they have on those bearings etc.).

You may want to speak with an attorney to see if you have a claim against the driver's insurance.

I also hope you filed a claim against the driver's insurance for diminished value relating to the impact/accident. This could give you another $5 to 12K in funds from the accident, since an accident damaged 911 (even if perfectly repaired) is harder to sell and will bring less upon resale. PM me if you need additional information.

DrMEMS

I saw the metal in my oil at my next oil change, about 4 kmi after getting rear-ended.
My first thought was that it was related to the accident, but now I have nothing that proves it.
I do feel that I am owed for the diminished resale value of my car. However, the OD's insurance company is with Southern California Automobile Association. Their agent just stonewalled me for several weeks after the accident, so I ended up going through my own insurance company under my collision insurance and paying the deductible, which I finally got back many months later. It's really easy for an insurance company to just do nothing. Any ideas for extracting the diminished value from them?