2003 911 C2 Cab - IMS and RMS experience

Hello All, I wanted to share my experience with my entry into a P-Car. I purchased a 2003 911 Cab in Dallas in Feb of this year. It was a one owner vehicle with 87k miles. I purchased from a dealer. With some help from the local shop who took care of this car for over 12 years, I was able to trace the connect with the previous owner. He was a true gentleman and helped me understand the maintenance history of the vehicle. I skipped the PPI as the car looked and sounded super good. (Worried be a bit later). The car has a Porsche Sport Exhaust (makes me smile everytime I fire it up). I went to a local shop to get a post purchase inspection. They quoted $9k for Clutch + IMS bearing replacement. The internet convinced me to replace the IMS bearing even though it seems perfectly healthy. I did not check the engine oil for any metallic parts. The Decision After staring at the $9k quote for a few days, I went down the rabbit hole — Rennlist, Pelican forums, Youtube videos. The 3.6L M96.03 has a much lower failure rate than the earlier 3.4L cars (roughly 1-3% vs 8-10%), and a factory bearing surviving 87k in a well-maintained manual put me in a statistically safer bucket. Purely on risk, I probably didn't need to do this. But the trans had to come out for a clutch eventually anyway, and the incremental cost of the IMS while in there is small. So I decided to to get help from a friend who is a 30 year veteran diesel tech. Part Selection Lots of debate on bearings. The forums push hard toward LN Engineering's Single Row Pro at $800-1500. After more reading, I went the cheapest OEM-spec path: a sealed NSK 6204DDU (the factory bearing IS an NSK 6204 — modern double-rubber-seal version is the same bearing with better seals). Plenty of guys on Rennlist and 6speedonline have run NSK retrofits 50-80k miles with no issues. Cost DIY parts and tools: ~$1100 IMS bearing - $18 NSK 6404 DDU - Exact OEM replacement. Flywheel $690 from Europarts Clutch kit $370 from AutohausAZ Shop quote: $9,000 Delta: $8,000 $35 - One Gallon - Gear Oil - TriaX GL4 - 75W90 $55 - 2 Gallon bottles of Valvoline 25W50 VR1 racing oil. + Liquimoly Ceratec additive. Time 3 weekends, 2 people, home garage with jack stands and a rented trans jack. The Real Eye-Opener — What I Found Inside the Old Bearing The old bearing looked fine externally. No grinding, no roughness, turned smoothly by hand. By all visual checks, healthy. Out of curiosity, that night I pried the seals off to see what was inside. It was completely filled with engine oil. No grease left at all. The original Porsche grease had washed away at some point in its 87k-mile life, and engine oil had migrated past the seals and was the only thing lubricating that bearing. This was a real "huh, the lore is actually true" moment. The conventional Rennlist wisdom — change your oil often, and drive the car hard — turns out to be empirically correct: The factory bearing seals are not bulletproof. Eventually engine oil gets in. Mine had clearly been running on oil-lubrication for a while. Once oil is the lubricant, oil quality is bearing life. Old, dirty, acidic oil = bearing damage. Frequent changes = clean lubricant whether you want it there or not. Driving it hard matters because splash lubrication only works at RPM. A car driven gently in stop-and-go traffic doesn't splash enough oil up to keep that bearing wet. Forum stories of failed bearings in low-mileage, gentle-driver cars suddenly make a lot more sense. The good news for me: no metal flakes, no scoring on the races. The previous owner clearly drove and maintained it well. The bad news: with 87k miles of oil-only "lubrication," that bearing was on borrowed time even though it sounded and felt fine. If you've got an early 996 with an unknown maintenance history and you're on the fence, this is the data point I wish someone had shown me — the factory bearing seals don't last forever, and once they fail you're betting your engine on your previous owner's oil change discipline. Lessons Learned The 3.6L IMS is NOT the same risk as the 3.4L. Don't panic. But the factory seals don't last forever. Once they fail, engine oil IS the bearing lubricant. Change oil often and drive the car. The old wisdom is correct. Do the IMS with the clutch. Incremental cost is small once the trans is out. Standalone, it's hard to justify. My clutch was almost new when I opened it and replaced it anyway. OEM-spec parts are enough. A sealed NSK 6204DDU is what the car had for 87k miles. Old bearing - with seal removed No grease - looks like its washed NSK 6204 DDU

Yes that bearing is fine but I would have removed the flywheel side seal. If oil gets in eventually why not make it easier? Conventional wisdom is that the earlier dual row was the lower failure rate. Never seen is reported to be the other way around. Also why wouldn't you wait to look at the clutch to see if it needed replacing, seems like an unnecessary expense.

$18 IMS bearing? One of two things are true: 1) The after-market IMS bearing at $800 each is a huge rip-off playing on Porsche owner fear. or.. 2) You are a very high-risk individual. I am not sure which camp I would fall in.

Quotes source post: 20581471

Hmmm, never heard of this 2 5 -50 VR1 oil. Sure you don't mean 20w-50 VR1 full synthetic oil...? Are you using the conventional non synthetic VR1 racing oil ? as its the only VR1 to come in gallon bottles? The non synthetic VR1 racing oil is not a great modern day oil for our m96 cars. In our m96 engines it breaks down quick and the viscosity loss is substantial. Anyhow, VR1 20-50 racing synthetic is what to use and a great oil, it needs no moly additive as it comes with similar moly levels as Driven DI50 ,,, and other great m96 oils with moly type additives. Using moly additives lowers viscosity so if not needed, don't use it as we need the most stable viscosity to combat m96 oil pressure instability problems.

Oh.. Shucks.... Indeed, I should have removed the flywheel side seal. Looks like eventually Oil will make its way into the seal, I might as well should have removed the seal. I did kick myself for jumping the gun on the flywheel and clutch. Both in really good condition. Completely unnecessary expense of $1100. I did feel that way and should have waited until it's open to order parts.

Oh.. Shucks.... Indeed, I should have removed the flywheel side seal. Looks like eventually Oil will make its way into the seal, I might as well should have removed the seal. I did kick myself for jumping the gun on the flywheel and clutch. Both in really good condition. Completely unnecessary expense of $1100. I did feel that way and should have waited until it's open to order parts

Quotes source post: 20582041

This was not an easy decision. I did hours of research into and reached out to my friends who is an authorized distributor of NSK/***/Timken for a large continent. We consulted with a few automotive engineers, and eventually came down with the following logic. "IF we replace the exact OEM spec to spec part, it should at least run another 80k miles, as it did the first 80k miles. or at least until the next clutch job. " After removing the original bearing I checked data sheets exact spec to spec. The matter of fact is $18 NSK bearing is what Porsche used and if its good for porsche engineers it should be good enough for me.

I used Valvoline VR1 Racing 20W-50 Motor Oil, Its a conventional oil. I might have gone overboard with the additive. Will change the oil in a couple of months, I already put 1000 miles on it after the oil change.

Quotes source posts: 20582041, 20582121

I used Valvoline VR1 Racing 20W-50 Motor Oil, Its a conventional oil. I might have gone overboard with the additive. Will change the oil in a couple of months, I already put 1000 miles on it after the oil change.

The important part! �FULL SYNTHETIC �

I think that you have the bearing failure statistics reversed. Early cars (99, 00 and some 01) have the double row bearing w the failure rate less than 1%. Later cars w the single row bearing the failure rate is usually stated as 8%. I understand that these stats come from Porsche in the class action lawsuit.

Quotes source post: 20582334

You may want to read up on the Eisen vs Porsche class action lawsuit. No way in hell would I put in another OEM single row in my 3.6 for $18 and expect to get 80k miles. As you discussed, you were at the tail end. Sure, maybe you do get 80k, but maybe you dont. You were very lucky with timing on yours. You are looking at $30k for a reputable rebuild today. Just saying. The 3.6 is so damn easy to keep it running strong and long. But you gotta... 1. Upgrade 2 key parts (IMSB, UAOS) 2. Replace the OEM water pump every 5-7 years with another OEM water pump 3. Run high quality synthetic oil with lots of Moly/Viscosity and change it frequently ...and you have easily addressed the big engine issues.

Quotes source post: 20581471

I feel the need to correct you on this. Failure rates are approximately as follows on 911s: -dual-row IMS bearing (3.4L until ~2000): 1% -single-row bearing (later 3.4L + all 3.6L M96s + early 3.6L M97s): 8% -large shaft, single row bearing (later M97s): 1%

Quotes source post: 20583009

Agree. Imho. 1% is statistically way to high, but 8% is certifiably nuts.

Quotes source post: 20583009

Yes Sir, I stand corrected. The 3.6L with single row bearing has higher probability of failure. Looks like the 8% stat is within the warranty period. So realistically out of warranty almost most of them might fail.

Quotes source post: 20582638

I did end up using conventional oil this time. Will do a few miles and then change over to synthetic in the next change. I plan to change every 3000 miles.,

Quotes source post: 20582799

Indeed, this is a gamble I guess. My car has a new AOS and recently a water pump swap. I will keep up with the water pumps changes every few years. 50 grade oil seems to be the way to go,.

Quotes source post: 20582041

Pelican parts is selling the KIT with the same exact bearing kit for $400. If you look closely it's the same exact NSK 6204 bearing that can be bough in Amazon for $13-$20 . So there is a bit of profiteering of porsche owners fears

Quotes source post: 20585299

Yeah, they can give me the kit for free and offer to put it on my car for free and I still wouldn't put it on my car. Unless they offered to fully pay for a rebuild of my choice from a builder of my choice.

Quotes source post: 20585917

Agreed. An 8% chance of catastrophe resulting in a $30K rebuild is worth 0.08*$30,000=$2,400. If the known way to avoid the catastrophe is less than $2,400, then it pays for itself. Indeed, if it were my running LN Engineering, I would price the bearing right where it is and use this formula to set the upper limit. There is no reason to take on the added risk of a stock $18 bearing when there is a proven alternative under $2,400.

Quotes source post: 20586136

Good analysis. Its just a no brainer to get a better IMSB on the car. Just like its a no brainer to upgrade the AOS and replace your water pump every 5-7 years. But sometimes the attractiveness of saving money over the near term clouds ones judgements over the long term. Happens to all of us.

Quotes source post: 20583009

I would put very little weight on those published numbers. They are from Porsche's lawsuit iirr. A lot of this failure % # balances on what the definition is of a 'failed imsb'. Is a failed imsb only 1 that has come apart completely, or is it also a bearing that has just started to shed metal but still spins good..? Big difference in the failure rates depending on what 'IS' is.... If a imsb blown engine car even gets to the Porsche dealership is questionable. Once out of warranty most/many owners never go back to expensive Porsche dealerships. Once at a indy shop that tells owner his engine is bad, its doubtful the failure ever gets back to the Porsche published % numbers. Or even get diagnosed correctly, back in the day. Without a costly total engine tear down and failure analysis most/many engine failures never get correctly diagnosed. The engines that have double row bearings that start to fail, but still spin ok, shed debris into engine oil for much longer periods of time. Constantly Shedding small flakes or even microscopic metal debris in the oil/engine takes its toll on all reciprocating parts ,especially crank/rod bearings/cylinder walls/piston skirts. So if an engine comes into shop with spun bearings is that the correct diagnosis,? even though the spun bearings were caused by a failing double row ims bearing that had not come apart/completely failed yet, but had the chance to spew hardened metal bearing particles/debris into engine maybe for months/thousands of miles. Double the amount a single row can shed in its quicker time frame failure mode.... So how many of these engine failures were not listed as imsb failure when in truth the imsb was the initial cause of many engines doomday no matter what the end outcome was..? IMHO, the published imsb failure rate is woefully inaccurate. Especially in the early days before much was known about this type of failure. I'd take more heed in reading Porsche forums and see what owners have gone thru with the m96. IMO its almost impossible to get an accurate number. More than likely if all the m96 failures were combined back in the beginning of all this, it would reflect a more accurate imsb failure rate, even though obviously some m96 failure would not be imsb related it would still be a more accurate # imo. That would be the few back in the day. Nowadays with m96 knowledge and preventative measures most owners have correctly been scared into taking preventitive measures. What we've learned over the decades about imsb failures have drastically reduced this type failure mode. From what Slakker says about them getting mostly spun bearing engines in for rebuilds from oil pressure instability, and iirr LN has said very similar along with bore score rebuilds... jmho

Quotes source post: 20586387

If this is true and I suspect it is, it makes the calculation for buying the LN Engineering bearing/solution even more compelling. If the real failure related number is more like 10%, then the value of the IMS bearing replacement would be $3,000.

Quotes source post: 20586387

Agree with your comments. The 996 world was flying blind for years with the brand new M96 engine. But after 25 years, we clearly know what can break, how it breaks, and how too prevent it with upgraded parts and better procedures. I don't loose any sleep. Your comments on what Slakker has seen with mostly spun bearings is interesting. No question, this is a real issue for track related cars from heat and oil starvation. Not as big of an issue with street cars unless a current owner bought a 996 that was extensively tracked by a previous owner(s) who never did the "right" things to protect the engine for the track. All one has to do is read thru Rennlist and Facebook, as well as watch Youtube and you will find all of the knuckleheads who openly refuse to address the issues properly for whatever reason and then quietly disappear after engine failure. Nevertheless, I don't loose any sleep because we have an abundance of solutions available and procedures to easily prevent oil pressure stability and starvation. It sure is damn easy to test your oil to see what's going on inside your engine.

Yes we have seen it ALL here on Rennlist in the 25 years the M96/7 has been produced... Some shops have never seen an IMSB fail and claim it is all overblown..lol, some see nothing but failed engines from Spun Rod Bearing like Slakker, some see mainly bore scoring like FSI ...then others see a smathering .. Make no mistake ... a 1% catastrophic failure rate on engines is very very poor, an 8% failure rate is ridiculous ... and this is just on IMSB ( 6204 dual row 1% , 6204 single row 8% , 6305 single row 1% ).. And then we have Spun Rod bearings and Bore-Scoring and AOS related issues..... But as we know all this can be fixed or avoided with upgraded parts and procedures...

Quotes source post: 20583009

Those numbers are from the lawsuit, decades old. I am sure, if available, those percentages would be far higher now, given time and mileage on our cars. When I bought my C4, with 55K miles on her several years ago, I immediately had the bearing replaced (along with base lining all the maintenance) and glad I did. Seal was gone and the bearing was in the process of failing.

Sure would be a fun exercise to put a new short block in a 996, run the factory AOS, run Mobil 1 40 and change it every 15k miles, never change the fuel injectors, and never change the water pump. Along the way do some canyon carving, maybe a couple of track days and alot of grocery runs and daily driving. Then see what kills the engine first and when. Will it be... 1. Bore scoring from bad oil and bad fuel injectors? 2. Failed IMSB? 3. Failed waterpump leading to cracked heads? 4. Failed AOS leading to hydrolock and cylinders cracking? 5. Failed rod bearings from bad oil lubrication from old oil, oil pressure instability and starvation? 6. Something else? Add: We all know that all of these failure points are easily mitigated with the right parts and procedures. But it would be a fun experiment if we didnt mitigate the risks and see what happens.

Quotes source post: 20582334

And here is another perfect example of why the stock bearing isn't good enough. Because it isn't.: https://rennlist.com/forums/996-foru...ild-motor.html It's what most of us already knew.

� The matter of fact is $18 NSK bearing is what Porsche used and if its good for porsche engineers it should be good enough for me.� The fact is this; The choice of that bearing (or any ball bearing in that spot) was made by accountants - not engineers� Ruminate on that for a while and then ask yourself why you didn�t use � The Solution� instead of an $18 ball bearing�

Quotes source post: 20595486

I understand the rationale and sentiment, It was clearly a design with little though for longevity and may be they testing it on a testing rig and simulated driving. I still believe that replacing the OEM spec bearing every clutch change and ensuring the oil flows is not a bad solution. In fact if anything I do regret not taking the grease seal off the bearing to allow oil circulation.

Another thing to note here is that, according to Lake Speed Jr, he showed, with testing, that you dont really want to use additives with oil, you want to use the right oil. The only exception is that LM Ceratec is good, but only to use it with LiquiMoly oil. Combining additives with certain oils can counteract the effectiveness of the oil. So he wouldnt mix Ceratec with any VR1. Id change your oil to either the VR1 full synth or Driven.

Just saw this from LN and figured i would post it on dual row IMSB failure. Even though we are talking a 3.6 on this thread, the dual and single row will all eventually fail. Just the way mechanical physics work. You Tube

Quotes source post: 20596033

At this point .. I think most consider this as a wear item and swap it every chance to change it along with the clutch or sooner. I also wonder why go for LN eng. solution when they also warranty it for 24k or 2 years. Might as well use OEM solution when both are equally less durable.

Quotes source post: 20596054

Can't speak to the warranty on the Solution. But i can say that if you have been around the M96/M97 long enough, you begin to understand what's going to fail, why it will eventually fail and how to prevent it from failing. To your point, I originally went with the LN Retro when it originally came out. At that time the Solution wasn't yet on the market but was arriving soon. Charles and I were both volunteering at the local Chicago PCA region HPDE event as tech inspectors and he told me about both LN IMSBs and made it pretty straight forward. If you want a permanent solution that you never have to replace again then go for the Solution. If you want a much stronger IMSB than the factory IMSB then go with the Retro as long as you know you will eventually have to replace it again when it times out. For me, the Retro was a better choice because i am a track guy and am at my indy many times a year having them service and refresh parts. So it was no big deal to replace the Retro next time my Indy dropped the engine or trans. But one key point that always stuck with me was the different in strength between the LN Retro and factory IMSB. You can find that info on LN's website. To me, it was a no brainer to go for more strength versus the factory IMSB, even if i was going to replace the factory IMSB like i would the Retro. However, in the next few years I plan on doing an elective rebuild to take advantage of all of the M96 internal part enhancements that have been developed over the past 25 years, not to mention power boost. At that time i will add the LN Solution because it's the most durable IMSB solution on the market. So the best advice i can give you if you use the OEM IMSB is to change it often. Just like the factory AOS and water pump, its a huge "weak link" in an otherwise extremely strong engine. Since I vote with my wallet, i always vote for the strongest parts to put on my car. Best of luck on your journey.

One point I haven’t seen made (apologies if it has) is the chance the NSK bearing from Amazon isn’t genuine. That would be down to a bad merchant getting knockoff items in the mix with genuine - only if it’s being fulfilled by Amazon and not shipped from the merchant directly. If the latter is the case, I wouldn’t be concerned on that topic.

Quotes source post: 20596054

The LN �Solution� is a permanent fix. Once installed there�s no need to mess with it again. Since it�s a plain bearing (just like in earlier air cooled and Metzger water cooled engines - and the other end of the intermediate shaft) there�s nothing to wear out like a ball or roller bearing. It pays for itself the first time you have to replace the other bearing (6 yrs/50k mi.)�It just seems like putting anything other than �The Solution� in an early M96 is foolish.

Quotes source post: 20596466

This. The "Porsche tax" is heavy, but not that heavy. To me the riskiest choice in this scenario, and there are a few, is the choice of an Amazon sourced $18 bearing for a very critical application. NSK doesn't appear to have an Amazon webstore and that appears to be a very counterfeited part as a cursory, brief search brings up a lot of Chinese bearing manufacturers and wholesalers offering that part complete with NSK packaging...down to $.30/ea. "OEM Quality" The 6404 bearing is ~$110 from an authorized, legitimate NSK distributor. If it were me, I would replace whatever bearing you received with something known to be genuine. Actually, I would do the IMS Retro. Good luck.