Custom Clevite Rod Bearings

[vsix]

Originally Posted by aus

Aren't there a bunch of S85's that have blown or had bad bearings in Europe? Some guy has a nice side business changing out bearings for other people's S85.

Clarify about who's side business your talking about, please.

[LSx]

Originally Posted by regular guy

You can check the dyno database, look for entries that have an "S" in the "Data Logs" column. In most cases, the files will be available for download. It's actually the other way around on the log files. The street log will show more than the dyno log.

Dyno Database: http://www.s65dynos.com/

The spark logs I've seen on the dyno are the same that can be logged with the Bavarian Technique tool. The interface runs at a limited bandwidth. It operates by sending request commands to the ECU, and the ECU responds with the data. Depending on how many "log" items requested, the frequency of the ECU output decreases. Log six items, and you're already below 50 ms resolution.

The Bavarian Technique tool can log much more than the dyno can. So that's why I say it's the other way around. FogCityM3 says he's logged hours of driving with not only spark advance, but knock detection. On the dyno, we only log spark advance.

I hope this helps.

By "more" I meant more actual instances of knock due to the dyno loading the engine harder than it would be on the street.

Originally Posted by SenorFunkyPants

There is something here about it (half way down the page):
http://www.bpmsport.com/blog/blog/bp...nock-analysis/

Interesting....no offense to BMW's sophisticated misfire detection, but I'd love to see what readings are given by a traditional knock sensor.

[regular guy]

Originally Posted by vsix

Clarify about who's side business your talking about, please.

Troy Jeup.

[FogCityM3]

I have done quite a bit of open road logging, with the stock 220e software, stock 240e software and two major tunes on this board (ESS and BPM), using 91 gas, 100 race gas mixture (for 93), 100 Octane race gas, and E25 (91+E85 and 100 +E85). Under most conditions, tuned or untuned, timing will be pulled under WOT runs on 91 Octane. 93 Octane, 100 octane E35 mixtures and cold weather will allow to hit stock and tuned timing "target"s. You can tell when a target is "met" because the knock intervention using BT logging tool will not register and for example, if hits 32 degrees stock, and you see no knock sensor intervention reading, then it's probably likely that 32 degrees at that rpm level is the targeted timing. Tuners typically target a slightly higher timing WOT. The ECU will also log mis-fires that may or may not trigger a CEL depending on their frequency. In 30,000 miles of checking for codes (via BT and BMWhat) I have only witnessed one misfire on cylinder 7 recently (did not set off a CEL) and I'm just about due for new spark plugs.

The BT tool as previously mentioned the sampling rate is not fast enough to obtain a continuous reading on timing and knock control. However, we do know knock intervention above 7,500 rpm at WOT occurs very often on 91 octane. The question is what information the ECU uses to detects "misfires" and pre-and actual detonation and whether this is accurate (I have no idea and never got too deep into the issue by logging other potentially revealing parameters such as ionic voltage, conductance, spark dwell times as I've only only seen one misfire) and with detonation and pre-detonation, can timing be pulled to a sufficient (and fast enough) extent that none of this occurs. In this regard, I think your question in another post remains unanswered.

Also, as mentioned, this isn't a thread about causes, I'm still curious as to why BMW chose 10w-60 oil and actually moved from lower weight oils in older M engines (in particular at the point when the S54 was produced).

Originally Posted by LSx

My first instinct when looking at the pictures of the bearings in the various threads was to blame this problem on detonation...but more than a few people said that they're sure that's not the cause and I'm not too familiar with the control system that BMW uses for knock detection and timing adjustment should knock be detected. If the strategy is not very aggressive, and knock is a common issue, then it's perfectly plausible that a little bit of detonation every time the engine is run hard can stack up over time and cause the damage seen even if the computer is pulling timing in each individual instance.

Anybody have a lot of data logging time? Are the knock sensor readings available?

[LSx]

Originally Posted by FogCityM3

I have done quite a bit of open road logging, with the stock 220e software, stock 240e software and two major tunes on this board (ESS and BPM), using 91 gas, 100 race gas mixture (for 93), 100 Octane race gas, and E25 (91+E85 and 100 +E85). Under most conditions, tuned or untuned, timing will be pulled under WOT runs on 91 Octane. 93 Octane, 100 octane E35 mixtures and cold weather will allow to hit stock and tuned timing "target"s. You can tell when a target is "met" because the knock intervention using BT logging tool will not register and for example, if hits 32 degrees stock, and you see no knock sensor intervention reading, then it's probably likely that 32 degrees at that rpm level is the targeted timing. Tuners typically target a slightly higher timing WOT. The ECU will also log mis-fires that may or may not trigger a CEL depending on their frequency. In 30,000 miles of checking for codes (via BT and BMWhat) I have only witnessed one misfire on cylinder 7 recently (did not set off a CEL) and I'm just about due for new spark plugs.

The BT tool as previously mentioned the sampling rate is not fast enough to obtain a continuous reading on timing and knock control. However, we do know knock intervention above 7,500 rpm at WOT occurs very often on 91 octane. The question is what information the ECU uses to detects "misfires" and pre-and actual detonation and whether this is accurate (I have no idea and never got too deep into the issue by logging other potentially revealing parameters such as ionic voltage, conductance, spark dwell times as I've only only seen one misfire) and with detonation and pre-detonation, can timing be pulled to a sufficient (and fast enough) extent that none of this occurs. In this regard, I think your question in another post remains unanswered.

Also, as mentioned, this isn't a thread about causes, I'm still curious as to why BMW chose 10w-60 oil and actually moved from lower weight oils in older M engines (in particular at the point when the S54 was produced).

Well if the computer is pulling timing because it's seeing knock, it's pretty easily to believe that the bearing damage very well could be detonation damage adding up over a long period of time.

An engine is only properly tuned when it can operate to it's full potential under almost all conditions without the computer needing to pull timing. While I have not personally done tuning on BMWs, I can say that most modern combustion chambers are efficient enough that they don't want more than 24º of timing at WOT. Targeting 32º sounds like a mistake even before being told that you see knock retard on a regular basis....

So why are people no investigating the tuning aspect of this issue more? That's the first place I would have started.

[FogCityM3]

BMW OE software targets 32 degrees at high RPMs WOT IIRC, which as I said is rarely seen achieved in most of my sea level logging conditions with 91 octane. This suggests that its own tuning (with KR insufficient to deal with detonation) is a problem for the gas used or that 91 octane doesn't belong anywhere near this engine. May explain also why engines untouched by tuning have also spun bearings. Scary stuff indeed!

Originally Posted by LSx

Well if the computer is pulling timing because it's seeing knock, it's pretty easily to believe that the bearing damage very well could be detonation damage adding up over a long period of time.

An engine is only properly tuned when it can operate to it's full potential under almost all conditions without the computer needing to pull timing. While I have not personally done tuning on BMWs, I can say that most modern combustion chambers are efficient enough that they don't want more than 24º of timing at WOT. Targeting 32º sounds like a mistake even before being told that you see knock retard on a regular basis....

So why are people no investigating the tuning aspect of this issue more? That's the first place I would have started.

[regular guy]

[Obioban]

Originally Posted by LSx

Well if the computer is pulling timing because it's seeing knock, it's pretty easily to believe that the bearing damage very well could be detonation damage adding up over a long period of time.

An engine is only properly tuned when it can operate to it's full potential under almost all conditions without the computer needing to pull timing. While I have not personally done tuning on BMWs, I can say that most modern combustion chambers are efficient enough that they don't want more than 24º of timing at WOT. Targeting 32º sounds like a mistake even before being told that you see knock retard on a regular basis....

So why are people no investigating the tuning aspect of this issue more? That's the first place I would have started.

It's targeting up to 32º, when appropriate (temp/octane) so that it is operating at full potential for the given condition. If you targeted 24º then on a cool day with 93 octane you would not be operating at the engine's potential. On days that it can't reach 32º, it doesn't.

This has been gone over before. The way the bearings are wearing is not consistent with knock. Please catch up before derailing the thread.

[LSx]

Originally Posted by Obioban

It's targeting up to 32º, when appropriate (temp/octane) so that it is operating at full potential for the given condition. If you targeted 24º then on a cool day with 93 octane you would not be operating at the engine's potential. On days that it can't reach 32º, it doesn't.

This has been gone over before. The way the bearings are wearing is not consistent with knock. Please catch up before derailing the thread.

You obviously don't know what tuning tables look like, if the computer is pulling timing because knock was detected then it was not tuned properly. The tables are setup to take sensor inputs on IAT, airmass, etc and target a timing value for those readings, not always target a number and rely on knock retard to dial it back.

The bearing wear is perfectly consistent with knock, performance engines are what I do on a daily basis.

You can take your snarky attitude and shove it right up your ignorant ass.

[Obioban]

Originally Posted by LSx

You obviously don't know what tuning tables look like, if the computer is pulling timing because knock was detected then it was not tuned properly. The tables are setup to take sensor inputs on IAT, airmass, etc and target a timing value for those readings, not always target a number and rely on knock retard to dial it back.

The bearing wear is perfectly consistent with knock, performance engines are what I do on a daily basis.

You can take your snarky attitude and shove it right up your ignorant ass.

I assure you the ecu does not take any inputs from the airmass sensor, as there is no such sensor on the s65.

Bearing wear can be a results of knock, but not the type of bearing wear that the S65 is seeing.

I'm sure you are extremely knowledgable about LSX engines, but please catch up before posting here. Repeating ground that has already been covered is counter productive. If you read through what has already been covered and find fault with it, by all means point it out. But starting from scratch because you're new to this issue is a huge waste of time.

... and this is likely not the thread for it. This is a thread for the solution that has been reached. There are threads discussing/diagnosing the problem.

[regular guy]

I think there's some hope for common ground here. As Obiadan said, this has been covered before in the Bearing Wiki thread, all 100+ pages of it. Bearing samples were sent to various engine builders ranging from Indy-500 winning, NASCAR championship winning, BMW Rolex champtionship engine builders, and Clevite themselves. None of the engine builders I sent bearings to saw any evidence of detonation. Most saw too little clearance; one thought he saw signs of cavitation. Clevite said they saw no evidence of detonation on the NA bearings, and added a similar opinion to LSx that if it had occurred, then BMW would have really screwed up the ECU tuning. But, to emphasize it again, Clevite saw no signs of detonation.

With that said, I'll be happy to send LSx the same bearings I sent Clevite and the other engine builders. Multiple opinions are welcome; but as Obiodan said, this probably isn't the thread for it. I'd say the better thread is the bearing wiki thread, found here.

http://www.m3post.com/forums/showthread.php?t=892838

If you want to see what Clevite said, then here's a link.
http://www.m3post.com/forums/showthr...892838&page=90

[LSx]

Originally Posted by Obioban

I assure you the ecu does not take any inputs from the airmass sensor, as there is no such sensor on the s65.

Bearing wear can be a results of knock, but not the type of bearing wear that the S65 is seeing.

I'm sure you are extremely knowledgable about LSX engines, but please catch up before posting here. Repeating ground that has already been covered is counter productive. If you read through what has already been covered and find fault with it, by all means point it out. But starting from scratch because you're new to this issue is a huge waste of time.

... and this is likely not the thread for it. This is a thread for the solution that has been reached. There are threads discussing/diagnosing the problem.

I was stating generic sensors, I already said I don't know specifically what BMW is using, but it's obvious that whatever they're doing is wrong considering that Fogcity said that in all his datalogging, he sees knock retard. That absolutely will cause the damage seen on these bearings, and unless you are purposefully hoping to ignore that fact I don't see why it's being swept under the rug.

If you guys want to throw parts at a problem in attempts to band-aid the symptoms, that's your prerogative. Have fun wasting thousands of dollars I guess.

[Obioban]

Originally Posted by LSx

I was stating generic sensors, I already said I don't know specifically what BMW is using, but it's obvious that whatever they're doing is wrong considering that Fogcity said that in all his datalogging, he sees knock retard. That absolutely will cause the damage seen on these bearings, and unless you are purposefully hoping to ignore that fact I don't see why it's being swept under the rug.

If you guys want to throw parts at a problem in attempts to band-aid the symptoms, that's your prerogative. Have fun wasting thousands of dollars I guess.

Knock will cause bearing damage (though not the type of damage that is being seen on the failed s65 bearings). Pulling timing to prevent knock will not.

And, yes, I mentioned it doesn't have a MAF to point out how little you know about this engine/issue so far, to be giving so much advice as to how this is the wrong solution.

Read the bearing wiki. Reply there if you find fault with the logic presented. But, read it first, eh?

[LSx]

Originally Posted by Obioban

Knock will cause bearing damage (though not the type of damage that is being seen on the failed s65 bearings). Pulling timing to prevent knock will not.

And, yes, I mentioned it doesn't have a MAF to point out how little you know about this engine/issue so far, to be giving so much advice as to how this is the wrong solution.

Read the bearing wiki. Reply there if you find fault with the logic presented. But, read it first, eh?

You don't seem to be capable of grasping the very simple concept that if the computer is pulling timing because it detects knock, that there is damage being done and that damage adds up over a period of time. The very act of the computer pulling timing and not meeting targets means that knock was detected, not that the computer can see into the future and pulled timing before it had a chance to occur.

Upper bearing shell damage is absolutely the type of damage caused specifically by detonation. It's obvious by your repetition of this that you have no actual real world experience with engines, their control systems, and failure analysis...and are simply parroting information you're seeing posted elsewhere.

What sensors are used is not relevant to the point being made, it doesn't matter a lick that I don't know what sensors are in play as I already stated that fact. The end result is all that matters, that end result is that the computer is pulling timing on a regular basis because the engine is detonating. When detonation occurs it dramatically raises the cylinder pressure to a point where the oil's hydrostatic wedge cannot support the rod and it makes contact with the crank journal which progressively wipes material off the bearing. That is exactly the kind of damage detailed in the pictures.

I already read the bearing wiki weeks ago, and my first comment was that this was textbook detonation damage. I was told that was ruled out and so I dropped it. Now I'm told that the computer is actively pulling timing which means that detonation should not have been ignored and is likely the main culprit.

And that will be my last response, good luck with your engine issues.

[SenorFunkyPants]

Originally Posted by LSx

Upper bearing shell damage is absolutely the type of damage caused specifically by detonation.

As a point of information, this was the reply I received from a technical enquiry I made to Mahle Clevite enclosing a number of images of worn S65 rod bearings (taken from the epic bearing thread) showing the typical wear pattern.
My question:
"Any idea on the wear presented would be great. Could using low octane fuel be a factor in what looks like excessive wear...the S65 engine does come with a anti knock ignition system."
Answer:
"Well, generally speaking, when you see heavy, accelerated wear on the upper
rod bearings, the cause is detonation. You can still see the machining
marks on the lower shells and I see no signs of abrasive wear caused by
contaminated oil" [followed by a discussion on detonation].

[kawasaki00]

Take this discussion to another thread please.

[BMRLVR]

Originally Posted by LSx

You don't seem to be capable of grasping the very simple concept that if the computer is pulling timing because it detects knock, that there is damage being done and that damage adds up over a period of time. The very act of the computer pulling timing and not meeting targets means that knock was detected, not that the computer can see into the future and pulled timing before it had a chance to occur.

Upper bearing shell damage is absolutely the type of damage caused specifically by detonation. It's obvious by your repetition of this that you have no actual real world experience with engines, their control systems, and failure analysis...and are simply parroting information you're seeing posted elsewhere.

What sensors are used is not relevant to the point being made, it doesn't matter a lick that I don't know what sensors are in play as I already stated that fact. The end result is all that matters, that end result is that the computer is pulling timing on a regular basis because the engine is detonating. When detonation occurs it dramatically raises the cylinder pressure to a point where the oil's hydrostatic wedge cannot support the rod and it makes contact with the crank journal which progressively wipes material off the bearing. That is exactly the kind of damage detailed in the pictures.

I already read the bearing wiki weeks ago, and my first comment was that this was textbook detonation damage. I was told that was ruled out and so I dropped it. Now I'm told that the computer is actively pulling timing which means that detonation should not have been ignored and is likely the main culprit.

And that will be my last response, good luck with your engine issues.

The issue with the detonation theory is that certain cylinders show wear and others don't, and these cylinders are not usually the same from engine to engine. Generally to have detonation severe enough to cause the bearing wear we are seeing, there would be signs on the top of the piston & or broken ring lands.

Second, the ion sensing ignition system on the S65 doesn't listen for knock than pull timing in the traditional way. It is continuously measuring ionization current in the cylinder to calculate the Peak Pressure Position (PPP) in a closed loop fashion.

Here is some interesting reading on how an ionization sensing ignition system works. http://www.ieeecss.org/CSM/library/1...eFineTuner.pdf

Anyhow, based on all current research and testing with the S65 I really believe that the bearing clearances/clearance stack up on certain engines are the root cause of bearing wear. Hopefully after testing these new clevite bearing shells we can find the favourable results I am expecting!

[s85e90]

Originally Posted by BMRLVR

The issue with the detonation theory is that certain cylinders show wear and others don't, and these cylinders are not usually the same from engine to engine. Generally to have detonation severe enough to cause the bearing wear we are seeing, there would be signs on the top of the piston & or broken ring lands.

Second, the ion sensing ignition system on the S65 doesn't listen for knock than pull timing in the traditional way. It is continuously measuring ionization current in the cylinder to calculate the Peak Pressure Position (PPP) in a closed loop fashion.

Here is some interesting reading on how an ionization sensing ignition system works. http://www.ieeecss.org/CSM/library/1...eFineTuner.pdf

Anyhow, based on all current research and testing with the S65 I really believe that the bearing clearances/clearance stack up on certain engines are the root cause of bearing wear. Hopefully after testing these new clevite bearing shells we can find the favourable results I am expecting!

We have seen s65 piston tops that show evidence of knocking. I have first hand HEARD AUDIBLE knocking on a few diff cars that was attributed to bad gas (93 octane). You mention that cylinders are showing diff but I'm pretty sure the s65 is able to pull individual cylinder timing (correct me if I am wrong).

I agree 100% with LSX and don't know why people are so dismissive that it can't be knock when there's pages and pages of threads and this is all just specualtion anyways.

Combine a tight clearance already and add some knocking that is forcing the piston back down and pushing the oil film out of the already small area and you have the top bearing wear we are all seeing. I have said the same before and it was dismissed. I for one don't know how much I trust this advanced IONIC knock detection system.

And yes, if you're datalogging the car and you're seeing it pull timing that means it is registering knock sums and not just pulling timing because. Why do you think the car is pulling timing??


Think about this; If I am hearing audible pinging, that's pretty bad and obviously the car is not pulling timing as it should be.

[s85e90]

Originally Posted by SenorFunkyPants

As a point of information, this was the reply I received from a technical enquiry I made to Mahle Clevite enclosing a number of images of worn S65 rod bearings (taken from the epic bearing thread) showing the typical wear pattern.
My question:
"Any idea on the wear presented would be great. Could using low octane fuel be a factor in what looks like excessive wear...the S65 engine does come with a anti knock ignition system."
Answer:
"Well, generally speaking, when you see heavy, accelerated wear on the upper
rod bearings, the cause is detonation. You can still see the machining
marks on the lower shells and I see no signs of abrasive wear caused by
contaminated oil" [followed by a discussion on detonation].

Exactly, this has been dismissed by all the masses here, but IMO detonation is real.

[s85e90]

Originally Posted by Obioban

Knock will cause bearing damage (though not the type of damage that is being seen on the failed s65 bearings). Pulling timing to prevent knock will not.

And, yes, I mentioned it doesn't have a MAF to point out how little you know about this engine/issue so far, to be giving so much advice as to how this is the wrong solution.

Read the bearing wiki. Reply there if you find fault with the logic presented. But, read it first, eh?

Your post makes no sense. If a car is pulling timing it's seeing knock and hence why it's pulling timing. You want to see minimal knock sums while trying to advance all within an acceptable AFR. Basic simple tuning.

[regular guy]

Originally Posted by s85e90

Exactly, this has been dismissed by all the masses here, but IMO detonation is real.

It was also dismissed by the same Mahle-Clevite representatives once they saw the same and actual bearings in person.

I'm also not sure anybody can draw conclusions about factory tuning by analyzing data a logs of aftermarket tunes. Seems like an invalid comparison to me. But let's assume BMW really did screw up the tuning, why hasn't those same tuners fixed it by now or exposed the flaw in BMW's tuning?

[s85e90]

Originally Posted by regular guy

It was also dismissed by the same Mahle-Clevite representatives once they saw the same and actual bearings in person.

I'm also not sure anybody can draw conclusions about factory tuning by analyzing data a logs of aftermarket tunes. Seems like an invalid comparison to me. But let's assume BMW really did screw up the tuning, why hasn't those same tuners fixed it by now or exposed the flaw in BMW's tuning?

I said I can and have heard audible knock on these s65 cars. Am I wrong? I know what I hear.