Official S65 Bearing Specification/Clearance Wiki

[regular guy]

Originally Posted by hughs65e90

This thread is a bunch of (what I suspect) grown men going back and forth w/ each other comparing resumes.

Anymore cars spin bearings yet??

We're at page 57

Yesterday my local dealer told me of four recent rebuilds with bearing changes and no other information. Does that count? Probably not.

[regular guy]

Originally Posted by Yellow Snow

Very sorry to have assumed it was your car when it was actually only the engine.

Anyway just in case you carry on accusing me of falsifying figures, I've quoted your sizes which you wrote.

That's a maximum journal size of 52.011mm. And you have the nerve to accuse BMW of getting cranks wrong. (BMW grind to 51.981mm)

I think we can close the aftermarket crank, Carrillo discussion like this. After page upon page of misdirection, the only point you've made is that you don't agree with using aftermarket cranks and rods in this article. I disagree. Even without it, there's still have a mountain of other OEM data that measures and shows the same thing.

[swamp2]

Originally Posted by regular guy

As long as the stroker cranks and connecting rods are made to the same exact specifications as the factory parts, then they remain good analogs to compare for bearing clearance issues.

For reasons discussed earlier I do not at all agree. Different designs, different mass and moments of inertia, possibly different materials, different manufacturing process, different tolerances, different fasteners. One key difference these things contribute to is a different overall stiffness in the "hoop" that contains the bearing.

Just too many variables.

Perhaps in a perfect world if one was positive that this is a clearance issue, one might be able to also observe a similar phenomena is a non-OEM rod if it had very similar dimensions and tolerances to the OEM unit. Really the only way to gurantee that would be if the vendor had full access to the original factory print of the part and I can guarantee that hasn't happened here.

Originally Posted by regular guy

I know I've said that I wouldn't want to guess because I don't like to guess. But it seems that you want a guess. Ok, I'll guess.

I will guess 200-500 completely bone stock, factory motors, without any mods, who have religiously serviced their cars at BMW dealerships in the USA alone. I only have six or seven contacts who share information over the past four years. But when they share, they often tell of multiple cars per month, and sometimes multiple cars at a time. So if you extrapolate those estimates to the rest of the USA, then I'll say 200-500 is my guess.

There, you happy for a guess that nobody can prove or stand behind as anything but a guess? Well, that's it.

Thanks for taking a stab at this. Let me try to see if I understand how you estimate was made.

1. You have access to 6-7 BMW service departments and they share information with you about an potentially serious recall-ish type of flaw.
2. They gave you the quantities of motors they have observed excessive bearing wear.
3. You tried to scale this information up to the entire world based on some rough estimate of what percent of the total E9X M3 population was represented by this select number of dealers.

I'm sure I sound quite critical but overall I think the numbers and process (assuming I got it right) sound about right.

With this ROM (rough order of magnitude) estimate and using the worst case 500 number is brings the percentage of cars affected to a whopping 0.9%. It doubles my previous (very "generous") estimate of 0.5%

[swamp2]

Originally Posted by oldmanstyle

Hey guy, I'm a practicing structural engineer with BS and MS degrees from a school that was #2 in the nation in the field at the time of my graduation. I graduated both times with highest honors and have been practicing in this highly technical field for over a decade now. Be careful what you assume about my thought processes.

Right back at ya. My three degrees are in Math, Physics and Physics again, also from top national uni's. I've also worked as Mechanical Engineer in multiple industries. However, that really isn't so relevant here. Do you have formal training or experience in reciprocating machine dynamics, tribology, IC engine design, bearing fault analysis, etc? I completely stand by my previous post as a reply to yours about the "intellectual dishonesty" present by many of use here practicing some healthy skepticism.

Perhaps you want to reply to my post rather than just throw up an appeal to expertise and some lame "warning". All I can go on is your posts, I'm not a mind reader and again I completely stand by my post below (of course with the very minor update that regular_guy has since made a ROM estimate of the number of cars affected by this).

Originally Posted by swamp2

If you don't recognize proper debate and proper evidence I guess too bad. Personally I've been courteous and open and have asked more questions myself than making any conclusions. Most others who have shown some level of healthy skepticism have shown a similar level of maturity. This thread overall has been pretty tame and largely free of ad hominem attacks and other distractions.

Your friendly local BMW service adviser works "in the field" and even for this particular company. I can't count the number of times I have heard utter 100% false crap about my car and BMWs in general from a BMW SA. There are also countless second hand stories of similar levels of utter incompetency. This is just one example of a plethora of others that exists that demonstrates that "working the field" is vague and truly does not mean squat. I've asked the only real potential expert we have here in this thread to share his credentials and he has consistently refused to do so. The OP freely admitted on multiple occasions that this is not at all an area he has expertise in.

A fairly clear amount of intellectual dishonesty has however appeared from the parties claiming to have all of the answers. No one on that side can even give a ROUGH COUNT of how many motors that are entirely or largely stock and have exhibited this excessive wear. I've only asked for this about 3 times... Similarly there have been many post here on this thread and in threads referred to be this one with motors with superchargers and on non OEM BMW internals showing the dreaded wear. Again such cases are nearly 100% irrelevant to whether or not BMW screwed up their clearances and/or oil recommendation in a big way.

Sorry but you are way off base here.

[regular guy]

Originally Posted by swamp2

For reasons discussed earlier I do not at all agree. Different designs, different mass and moments of inertia, possibly different materials, different manufacturing process, different tolerances, different fasteners. One key difference these things contribute to is a different overall stiffness in the "hoop" that contains the bearing.

Just too many variables.

You make some very good points there. That's a very persuassive argument that I may have overlooked earlier. I will give that some more thought. I'm still inclined to leave it because I don't look at it as "proof" but as "evidence" and I'm more inclined to let the reader filter out specific cases. I would be perfectly fine putting a caveat description on the bearing photo page for this entry to reflect what you just said, if you want to word one up.

Perhaps in a perfect world if one was positive that this is a clearance issue, one might be able to also observe a similar phenomena is a non-OEM rod if it had very similar dimensions and tolerances to the OEM unit. Really the only way to gurantee that would be if the vendor had full access to the original factory print of the part and I can guarantee that hasn't happened here.

That implies we'll never be able to use any of the disassembled Dinan strokers with proper clearance as proof. Since Dinan is only a mile down the road and I kinda know the guy who builds the strokers, I was expecting that would be our best hope in about 24-months time.

Thanks for taking a stab at this. Let me try to see if I understand how you estimate was made.

1. You have access to 6-7 BMW service departments and they share information with you about an potentially serious recall-ish type of flaw.
2. They gave you the quantities of motors they have observed excessive bearing wear.
3. You tried to scale this information up to the entire world based on some rough estimate of what percent of the total E9X M3 population was represented by this select number of dealers.

I'm sure I sound quite critical but overall I think the numbers and process (assuming I got it right) sound about right.

With this ROM (rough order of magnitude) estimate and using the worst case 500 number is brings the percentage of cars affected to a whopping 0.9%. It doubles my previous (very "generous") estimate of 0.5%

Yes, 99% correct of what I'm saying and did. The only difference is I was keeping it to the US not World. I figured 1% failure rate with those numbers, so we're on par with each other. I don't know what a normal failure rate is, but to me anything more than 0.1% sounds like a large failure rate. But maybe that's my ignorance of manufacturing mechanical things like an IC engine. Maybe 1% is OK for IC engines. It just sounds big to me but maybe it's not.

[swamp2]

Originally Posted by regular guy

Detonation vs. Bearing Wear: Could this be the cause?
By Kawasaki00
...
The purpose of this post is to address the detonation. What was found is that there is no significant detonation leading to bearing wear on this engine. Not saying some engines may not have it but this engine does not.

Right so we have ruled out severe detonation, so severe that is caused fretting on the back side of the bearing shell, in one particular engine where there was bearing wear from a presumed too little bearing clearance.

Is detonation always accompanied by fretting on the backside?

Unfortunately, the other 10's (or hundreds) of other cases where the bearing wear has been noted have not been examined in a similar fashion. It seems obvious that such an examination is the only thing that could entirely rule out detonation.

[swamp2]

Originally Posted by regular guy

You make some very good points there. That's a very persuassive argument that I may have overlooked earlier. I will give that some more thought. I'm still inclined to leave it because I don't look at it as "proof" but as "evidence" and I'm more inclined to let the reader filter out specific cases. I would be perfectly fine putting a caveat description on the bearing photo page for this entry to reflect what you just said, if you want to word one up.

I'm happy with your acknowledgement of this. Feel free to change any of your material in the opening posts as you see fit.

Originally Posted by regular guy

That implies we'll never be able to use any of the disassembled Dinan strokers with proper clearance as proof. Since Dinan is only a mile down the road and I kinda know the guy who builds the strokers, I was expecting that would be our best hope in about 24-months time.

Great idea. Good luck getting the data required...

Although it would not be "perfect proof", a marked absence of bearing wear in some reasonable number of Dinan builds, along with their actual clearances or clearance ratios would still be some good evidence. Unfortunately, it would leave the possibility of a faulty OEM rod design "on the table" rather than too small of bearing clearance ratios. My understanding though is that connecting rods are one of the most loaded and intensely analyzed (FEA simulated) components in an engine, a high performance one especially but even a regular passenger car engine. That being said I still can't quite say that a botched rod design is any more or less likely than a botched clearance ratio.

As for the Dinan build it would be great to know the oil they are using as well. My guess would be that they stuck with the TWS 10w60.

As discussed prior we'll likely never have perfect proof but we are not and can not do "academic quality" science here. We'll have to make good engineering judgements and rely on inductive reasoning.

Originally Posted by regular guy

Yes, 99% correct of what I'm saying and did. The only difference is I was keeping it to the US not World. I figured 1% failure rate with those numbers, so we're on par with each other. I don't know what a normal failure rate is, but to me anything more than 0.1% sounds like a large failure rate. But maybe that's my ignorance of manufacturing mechanical things like an IC engine. Maybe 1% is OK for IC engines. It just sounds big to me but maybe it's not.

The worldwide production count for E9X M3 (not counting perhaps ~100 or less E93 "stragglers" at the tail end of product) is 58k. 500 out of 58k worldwide is just under 1%.

[regular guy]

Originally Posted by swamp2

Right so we have ruled out severe detonation, so severe that is caused fretting on the back side of the bearing shell, in one particular engine where there was bearing wear from a presumed too little bearing clearance.

Is detonation always accompanied by fretting on the backside?

Unfortunately, the other 10's (or hundreds) of other cases where the bearing wear has been noted have not been examined in a similar fashion. It seems obvious that such an examination is the only thing that could entirely rule out detonation.

Good idea, I will add this of the list of things to track. I'll set the word out for the one or two shops I know doing this service to request taking photos of the bearing backs and rod bores. We'll be lucky if we get 50% of them to do it. But every little bit will help.

[swamp2]

Originally Posted by regular guy

That implies we'll never be able to use any of the disassembled Dinan strokers with proper clearance as proof. Since Dinan is only a mile down the road and I kinda know the guy who builds the strokers, I was expecting that would be our best hope in about 24-months time.

Are Dinan universally using the standard Clevite bearing clearance ratios for a "high performance engine" (the extra 0.0005) and are they achieving this with ground crank journals. That is the only way to get such clearances. Are they changing the clearance ratios for the mains and connecting rod big bore ends both?

[regular guy]

Originally Posted by swamp2

I happy with your acknowledgement of this. Feel free to change any of your material in the opening posts as you see fit.

I added the following note to the page-1 bearing description of these parts. Please review to see if it captures the points correctly.

NOTE: This engine has aftermarket crank, rods, and pistons. They are different designs from stock, have different mass and moments of inertia, possibly different materials, different manufacturing process, different tolerances, and different fasteners. Each of these key difference contribute to the overall stiffness in the connecting rod bore holding the bearing and may affect the observed bearing wear in the following photos.
http://www.m3post.com/forums/showpos...73&postcount=8

[regular guy]

Originally Posted by swamp2

Are Dinan universally using the standard Clevite bearing clearance ratios for a "high performance engine" (the extra 0.0005) and are they achieving this with ground crank journals. That is the only way to get such clearances. Are they changing the clearance ratios for the mains and connecting rod big bore ends both?

BMRLVR knows the exact clearances Dinan is using. I never wrote them down. In this post, he mensions they 0.0019 for mains and 0.0025 for rods. That's 0.00080" per journal-inch on the mains, and 0.00122" per journal-inch on the rods. It does not appear that Dinan is using the extra 0.00050" margin. Dinan has billet cranks and can have the journals manufactured to any size they want. I just helped a local guy do this. He plans to build a stroker. I helped him order the crank and we specified our own journal sizes for extra clearance. I have no plans to share the sizes he used.

EDIT: After doing a few more calculations, here's what I think Dinan is doing. Mains are 0.00080" per journal-inch diameter without the extra 0.00050" margin. Rods appear to be exactly 0.00100" per journal-inch diameter PLUS the extra 0.00050" margin. Using this formula on rods, I get the following: 2.04655 * 0.001 = 0.00204 + 0.00050 = 0.00254 inch.

[Yellow Snow]

So..... Does Dinan have a bigger R&D dept than BMW ?

[CSL_E36]

Dinan recommends Total 10w-60. Total was formally "Elf". Its also a few bucks cheaper per liter then TWS.

I think Dinan is great company, however the "weak" 4.6L in the dyno shootout thread proved something was off and Alekshop has yet to update that thread with a follow up dyno after the Dinan car was "sorted out".

For those that follow Motorsport and saw the second half of the Rolex series remember that midway through the season the Ganassi racing team finally ditched the 5.0L S62 in favor of the new Dinan 4.5L S65 racing motor. They blew 3 motors in a handful of races. IIRC 2 failures were due to "valve-train" failures.

When it comes to my money, Dinan wins simply due to their racing experience and 50k warranty. If you do ever have a problem, they will take care of it.

Getting specs on stroker builds? I dunno about that one...

[Yellow Snow]

Originally Posted by regular guy

Aftermarket rod journals: 2.04720 +/- 0.00005"

Originally Posted by regular guy

Why did you chose the smallest variance of official values and compare them to the largest PLUS tolerance of the aftermarket crank? Isn't that the most misleading way the comparison could be done?


Instead of making up more numbers of your own, go look up the actual (measured) clearance numbers...they're posted on this site. Send me a PM if you want help finding them.

I am not making anything up as you like to put it. They are your figures. How can you possibly accuse me of falsifying the figures that are quoted directly from your post?

If it were my crank, I would be more concerned with the top limit size, but in fairness to your cause I will take the smallest tolerance which makes your aftermarket crank 2.04715" (51.998mm) that still makes your crank .00065" bigger than oem which seems to have been fine for 24k miles.

I don't get it. Are these figures a mistake?

[SenorFunkyPants]

Rod bearing wear patterns.
One of the issues that is a poor fit in the tight bearing theory is the lack of visual wear on the bottom bearing shell relative to the high wear rate on the top shell seen across both the S85 and S65 OEM engines. The pattern is almost always high top bearing wear, very low bottom bearing wear.
In an engine with tight rod bearings, elevated upper bearing wear is predicted by the much higher (combustion) loads that they receive relative to the lower bearings, however some lower bearing wear is to be expected given that it too is subjected to considerable loads.
Looking at RG’s stroker motor (post#2) which can reasonably be taken as having tight bearings – “they measured as little as 0.0011 inch rod bearing clearance” when Van Dyne was re-assembling the motor with the same (non OEM) crank with new rod bearings and (non OEM) rods. As the upper bearings only show early stage wear then following the pattern seen in OEM engines, the prediction for lower bearing wear is essentially none. Pictures of the lower bearings are not included on page #1 but are in the build log noted before on m3post.
On the bottom bearing shells the wear is clearly visible. Obviously this supports the principle that tight bearing clearance will increase bearing wear, but it also illustrates a significant discrepancy between the wear pattern of rod bearings in this Stroker engine and the wear patterns seen in bearings from an OEM environment where a tight clearance is still not fully proven.



Stroker motor at 24,000 miles





For comparison
Pictures from a stock engine at 30,000 miles with higher wear
These are also the pictures I sent to Clevite which cited detonation as a cause

[SenorFunkyPants]

Originally Posted by regular guy

Given the amount of time you spent researching that build on other web sites and then appearing here as an expert on the build specifics, I would have expected you to know its history a little better.

You were chiding Snow for not knowing the history of your engine when all he had to go on was the information you gave on page #1.
I simply quoted the relevant post to show that he was correct in restating facts as you had presented them. That they turned out to be incorrect was your oversight not his.

Originally Posted by regular guy

Given the amount of time you spent researching[...] and then appearing here as an expert...

Wow that takes Pot Kettle to a whole new level.

[SenorFunkyPants]

Originally Posted by swamp2

The worldwide production count for E9X M3 (not counting perhaps ~100 or less E93 "stragglers" at the tail end of product) is 58k. 500 out of 58k worldwide is just under 1%.

If you take the low end of RGs guess of 250 USA failures against 25000 USA cars sold....for the UK market of 6000 cars would equate to 60 engine failures.
I can find one and that was due to a piston failure.

[CSBM5]

Originally Posted by 96LTWM3

For those that follow Motorsport and saw the second half of the Rolex series remember that midway through the season the Ganassi racing team finally ditched the 5.0L S62 in favor of the new Dinan 4.5L S65 racing motor. They blew 3 motors in a handful of races. IIRC 2 failures were due to "valve-train" failures.

When it comes to my money, Dinan wins simply due to their racing experience and 50k warranty. If you do ever have a problem, they will take care of it.

Not sure "ditched" is the right word here. Three years back Steve gave a talk where he said they were probably going to be forced to switch to the S65 by the organizer during 2011 (note it ended up being 2.5 years later). I made a post about his talk on the M5board back then, so I'll just quote the snippet I posted back then (lest my memory fail me with revisions, lol):

Originally Posted by CSBM5 Nov2010 on M5board

Next year they have to use the S65, and he didn't sound too excited about it since the largest displacement they can get out of the S65 is 4.5L. He thought they could get close to the same peak HP out of the S65 as the S62, but they'll have to rev it about 1500-1700rpm higher resulting in more pumping and frictional losses. The only pros were that it is slightly lighter and smaller, so it can be placed a bit lower in the chassis, and they'll have some small amount of ballast they can place appropriately.

Regards,
Chuck

[regular guy]

Originally Posted by SenorFunkyPants

Rod bearing wear patterns.
One of the issues that is a poor fit in the tight bearing theory is the lack of visual wear on the bottom bearing shell relative to the high wear rate on the top shell seen across both the S85 and S65 OEM engines. The pattern is almost always high top bearing wear, very low bottom bearing wear.
In an engine with tight rod bearings, elevated upper bearing wear is predicted by the much higher (combustion) loads that they receive relative to the lower bearings, however some lower bearing wear is to be expected given that it too is subjected to considerable loads.
Looking at RG’s stroker motor (post#2) which can reasonably be taken as having tight bearings – “they measured as little as 0.0011 inch rod bearing clearance” when Van Dyne was re-assembling the motor with the same (non OEM) crank with new rod bearings and (non OEM) rods. As the upper bearings only show early stage wear then following the pattern seen in OEM engines, the prediction for lower bearing wear is essentially none. Pictures of the lower bearings are not included on page #1 but are in the build log noted before on m3post.
On the bottom bearing shells the wear is clearly visible. Obviously this supports the principle that tight bearing clearance will increase bearing wear, but it also illustrates a significant discrepancy between the wear pattern of rod bearings in this Stroker engine and the wear patterns seen in bearings from an OEM environment where a tight clearance is still not fully proven.

These are very valid observations. The stroker is a mere data point in a much bigger picture. Take it or discard it as you wish.

For comparison
Pictures from a stock engine at 30,000 miles with higher wear
These are also the pictures I sent to Clevite which cited detonation as a cause

I had a very different experience emailing Clevite than yours. I sent them different pictures of the same bearings to make it a little less obvious I was checking your story. But instead of asking a leading question and getting the response I wanted (telling Clevite I suspected detonation and asking them to confirm), I simply said they were the experts and tell me what they think from the photos. I gave them no bias in my question, and they gave no bias in their response. And instead of ending the conversation with the answer they gave me (which in your case was the answer you wanted), I continued the conversation with many challenge questions. This went on for about 14 emails, and some of the responses I received were so detailed they contained multiple typed pages.

I received a response a Senior Clevite representative that listed three possible causes. NONE of them were detonation. In fact, the word "detonation" wasn't even in the response email. I then followed it up with numerous challenge emails as well. Eventually the challenge emails included a discussion of detonation. Then I issued challenge questions to those answers as well. I specifically asked why detonation wasn't included in the very first email of possible causes. Answer: Due to the advanced detonation detection and correction in the M3, detonation was not considered a possibility of the wear pattern seen in the bearing photos.

That answer lead to even more challenge questions which consistently gave the following answer (no matter how I asked it): Unless something in the M3 wasn't working correctly (ECU?), detonation was not a likely cause of the bearing wear pattern in the photos. But to confirm this, Clevite asked me to send the actual bearings. Next week, Clevite will receive these actual bearings along with two or three other sets. They will inspect each set for causes to the wear patterns. I don't know yet if they will want me to post their findings on a public forum. I will seek their permission to do so.

[regular guy]

Originally Posted by SenorFunkyPants

You were chiding Snow for not knowing the history of your engine when all he had to go on was the information you gave on page #1.
I simply quoted the relevant post to show that he was correct in restating facts as you had presented them. That they turned out to be incorrect was your oversight not his.

Wow that takes Pot Kettle to a whole new level.

YS brought up the stroker build into this thread two months ago on September 27th in this post. That's a lot of research time. I would have expected more historical knowledge from two guys presenting themselves in this thread as experts on that build.

As for your primary concern, I corrected the description to indicate 22000 miles NA and 2000 supercharged.

[regular guy]

Originally Posted by Yellow Snow

I am not making anything up as you like to put it. They are your figures. How can you possibly accuse me of falsifying the figures that are quoted directly from your post?

History my friend...history. I don't think an honest person would have ever compared the smallest journal to the largest + tolerance (with or without the magical decimal point error) and tried to declare one out of spec with respect to the other. That's something you just can't put back into the box and explain as an honest mistake. I also wouldn't have tried to make the comparison with two different sets of instruments, on three different days, with three different temperatures and unknown instrument calibration. Don't you think that's also a very misleading way to compare data? As an experienced engine builder, certainly you know this. What's the reason for comparing this way in such a misleading manner?

I also don't think an honest person would continue to claim reduced clearance by 0.00080" after he's been informed the actual measurements are somewhere on this site and been repeatedly offered a link to go find them.

But I guess we just have different ideas for presenting data.

If it were my crank, I would be more concerned with the top limit size, but in fairness to your cause I will take the smallest tolerance which makes your aftermarket crank 2.04715" (51.998mm) that still makes your crank .00065" bigger than oem which seems to have been fine for 24k miles.

I don't get it. Are these figures a mistake?

You're pretty far off topic now to this thread. If you want to discuss that build and it's measurements, go to one of the threads you've linked already and ask your questions.

[MFL]

I'm having a chat with my dealer on Tuesday and will be mentioning my concerns with the rod bearings. I'm also hearing a lot of unexpected chatter of some kind between 3500-4000 RPMs, which they will be addressing.