E46 / E92 M3 advice
- Thread starter Robbo
- Start date
Soldato
- Joined
- 19 Oct 2002
- Posts
- 18,050
- Location
- Shakespeare’s County
you ain’t got the car kid.
Associate
- Joined
- 26 Oct 2002
- Posts
- 577
- Location
- UK, Surrey, Guildford
You don't need to own the car to know about this problem! lol
(I do have 1 JFYI though).
You're right their is some doubt about the mains, they also have tight clearances, but their have been far fewer failures of those, and a large chunk of those are supercharged cars. Also another reason you don't hear much about them being done is that because it's a mammoth specialist job, so really expensive to do! The occasional mains issue has been known about for more than a few years (at least at m3post).
You've ignored or not read or forgotten what I said about oil flow and pressure for the S65 engine, it has a variable volume oil pump and it is capable of compensating for the extra clearances, as I said BE have measured the oil pressure with the bigger bearings in, see here.
Yes bad workmanship has screwed a small number of engines, you can only try to find a reputable garage and trust them, or do nothing, but going that route would mean you would have nothing done for anything
.
You're also right that it is random, or variable at least, that in part at least will be due to the different tolerances that each engine has.
Gaygle
Re bearings failing due to it be high revving, how come other high revving engines don't have large amounts of failure? e.g Honda S2000 and the Audi V10. I doubt they're treated any better?
I don't think being careful when the engine's cold would stop the problem, but it would delay it. Either way the OEM clearances can be too tight, and from the sounds of your old shells they were, and so OEM shells would likely have worn excessively again (unless you got lucky and got OEM shells at the upper end of the tolerance range).
(I do have 1 JFYI though).
No they are not the right clearances when combined with the oil used, that's the whole point of using non OEM bearings. The OEM bearing clearances are too tight at the lower end of tolerance range (that's a design flaw!) as stipulated by the manufacturers who make the bearings, specifically Mahl(Clevite), King bearings and ACL, if you read their pdf tech files they all state at least 0.00075" per 1" journal, and then advise to add 0.0005" for high hp/rpm engines. Even with the LCI bearings, at the lower end of their clearance they fall below that requirement. Many other engines do run tight clearances but they'll tend to use 0w20 or 5w/30, and don't tend to be high revving engines.
You're right their is some doubt about the mains, they also have tight clearances, but their have been far fewer failures of those, and a large chunk of those are supercharged cars. Also another reason you don't hear much about them being done is that because it's a mammoth specialist job, so really expensive to do! The occasional mains issue has been known about for more than a few years (at least at m3post).
You've ignored or not read or forgotten what I said about oil flow and pressure for the S65 engine, it has a variable volume oil pump and it is capable of compensating for the extra clearances, as I said BE have measured the oil pressure with the bigger bearings in, see here.
Yes bad workmanship has screwed a small number of engines, you can only try to find a reputable garage and trust them, or do nothing, but going that route would mean you would have nothing done for anything
.You're also right that it is random, or variable at least, that in part at least will be due to the different tolerances that each engine has.
Gaygle
Re bearings failing due to it be high revving, how come other high revving engines don't have large amounts of failure? e.g Honda S2000 and the Audi V10. I doubt they're treated any better?
I don't think being careful when the engine's cold would stop the problem, but it would delay it. Either way the OEM clearances can be too tight, and from the sounds of your old shells they were, and so OEM shells would likely have worn excessively again (unless you got lucky and got OEM shells at the upper end of the tolerance range).
Last edited:
Associate
- Joined
- 26 Oct 2002
- Posts
- 577
- Location
- UK, Surrey, Guildford
Forgot to say, somewhere in BE's wiki they talk about measuring a dozen or so cranks, and use the mean of those IIRC! I think the figure they use for the S65 is this - nominal RB journal 51.9811mm, 2.0465"
Oh, thought I'd add this in again as it kinda got drowned in my previous mega-post of a list of bearing photos. 1 of the photos of VAC (extra clearance) rod bearings (S65) that had done 52k miles with a supercharger, this is how shells should look when removed! (hopefully I can get it to show below. Here's the link to the relevant post - https://www.m3post.com/forums/showpost.php?p=25681122&postcount=1386 ). The engine was stripped due to water ingestion.
Oh, thought I'd add this in again as it kinda got drowned in my previous mega-post of a list of bearing photos. 1 of the photos of VAC (extra clearance) rod bearings (S65) that had done 52k miles with a supercharger, this is how shells should look when removed! (hopefully I can get it to show below. Here's the link to the relevant post - https://www.m3post.com/forums/showpost.php?p=25681122&postcount=1386 ). The engine was stripped due to water ingestion.
Oil pressure is measured before the crankshaft. Don't confuse that with oil flow at the bearing surfaces which is then how oil is entrained to create the hydrodynamic wedge itself, tight clearances are an issue here when oil entrainment speed (viscosity) is too low meaning the inherent load that a journal bearing can support is too low for the load it is under from the conrod, the S65 runs a relatively low BMEP so its not like it has massive demands from normal use (wear is more apparent on the top side. This will happen with one scenario, thick (cold oil) and high load...
When the system is set up for specific clearances and people then go and change the rod bearing clearance only (which oil removal from the journal is aided at high speed by 8400rpm). I saw that link years ago and its not really saying anything other than they got lots of data points. I mean some of the stuff they saw they were surprised at and claimed a benefit. How can bigger clearance cause higher oil pressure and then claim a benefit of the upgraded bearing?
Also high oil pressure is good, why would people also say 10W-60 is to thick (that provides oil pressure via resistance to flow). Oil pressure should be lower with bigger clearance, yes the pump will work harder for to maintain the same pre crank pressure, but then you are just wasting power pumping excess oil through the system. Hence the reason went 10W-60, tight clearance and an oil system specified for that. Protection without the power losses normally associated with thicker oils. The original 10W-60 has some nice properties often ignored in the discussion too, its base viscosity with additive and its cold performance is more like a 7.5W-60 if you were able to call an oil that within J300. BMW did testing on this oil in 2010 after developing the S65 using the previous iteration of the, at the time, supplier's oil.
This topic has come up so many times in the 4 years, and i have read a lot. Yes it is choice if there is doubt of history but i challenge the design flaw. If you want to spent money for peace of mind due to unknown history that is fair enough. but don't feed the scaremonger machine. This engine is highly tuned and needs care to warm up like any motorsport derived engine, comparing to generic clevite charts for some 1980s oldmobile is not relevant. Of course i have looked into it. I spent significant money on my M3 to get a mint one and obviously haven't got my finger in my ears about this issue and I have the ability to investigate this issue more than most people so just wanted to bring some measured response to it.
When the system is set up for specific clearances and people then go and change the rod bearing clearance only (which oil removal from the journal is aided at high speed by 8400rpm). I saw that link years ago and its not really saying anything other than they got lots of data points. I mean some of the stuff they saw they were surprised at and claimed a benefit. How can bigger clearance cause higher oil pressure and then claim a benefit of the upgraded bearing?
Also high oil pressure is good, why would people also say 10W-60 is to thick (that provides oil pressure via resistance to flow). Oil pressure should be lower with bigger clearance, yes the pump will work harder for to maintain the same pre crank pressure, but then you are just wasting power pumping excess oil through the system. Hence the reason went 10W-60, tight clearance and an oil system specified for that. Protection without the power losses normally associated with thicker oils. The original 10W-60 has some nice properties often ignored in the discussion too, its base viscosity with additive and its cold performance is more like a 7.5W-60 if you were able to call an oil that within J300. BMW did testing on this oil in 2010 after developing the S65 using the previous iteration of the, at the time, supplier's oil.
This topic has come up so many times in the 4 years, and i have read a lot. Yes it is choice if there is doubt of history but i challenge the design flaw. If you want to spent money for peace of mind due to unknown history that is fair enough. but don't feed the scaremonger machine. This engine is highly tuned and needs care to warm up like any motorsport derived engine, comparing to generic clevite charts for some 1980s oldmobile is not relevant. Of course i have looked into it. I spent significant money on my M3 to get a mint one and obviously haven't got my finger in my ears about this issue and I have the ability to investigate this issue more than most people so just wanted to bring some measured response to it.
Associate
- Joined
- 26 Oct 2002
- Posts
- 577
- Location
- UK, Surrey, Guildford
I don't have the figures for the clearances for those VAC bearings, but they were stated as being the latter extra clearance ones. Those bearings are still available, although the VAC's store website gave no useful information on that!
As for the oil pressure, another factor for oil flow for the oil wedge is clearance, if insufficient it won't allow enough oil flow for the oil wedge to be effective. Agreed cold oil will make that worse.
RE the S65's BMEP, IIRC that essentially means the power of each combustion event (I only remember for sure that it stands for Brake Mean Effective Pressure), so more related to torque output? I know it's relatively low compared to a turbo diesel or turbo/SC petrol engine, but I wouldn't describe it as low, anyway, semantics .
As for BE's oil flow test, I'm not sure which part you are referring to mentioning increased oil pressure with BE bearings, I can only guess you mean a not insignificant amount (so ignoring a few PSI variations here and there), so increased pressure IIRC was only significant at startup, for 1st ~2s. The oil pressure increased to operating pressures more quickly due to the higher flow of oil due to the bigger clearances, the oil got to the sensor (and bearings) faster, and the variable flow/pressure oil pump was mostly able to make up the pressure despite the extra oil losses, see here (I'm baffled too as to why they didn't realise that! lol I'm sure they say somewhere else that's why). So it's not higher oil pressure overall, but it gets up to (sometimes very slightly lower) pressure faster. If you're not referring to that then you'll have to be more specific.
People say 10w/60 is too thick because of the tight clearances that the S65s have, it's the combination that's the problem.
As for the extra power lost due to higher flow rate, that's going to be a tiny amount, if any at all, as that will be countered somewhat by having less resistance to pump it out in the 1st place. I'd imagine the 2 would near enough cancel out?
Afraid you're point about using tight clearances with thick oil to reduce power losses isn't right, thicker oil uses more power to pump (and causes more drag), most modern engine use 0w/20 (or thinner!) or 5w/30 with tight clearances, they have to be tight for the thinner oil, the thinner oil reduces power losses and improves MPG. Too thick an oil through too tight clearances will cause oil starvation.
Interesting about 10w/60 being nearer a 7.5w (I tried to convert the 40C kinematic viscosity of it to a 'w' rating but was unable to find an appropriate chart), that probably explains why people using 5w/50 tentatively don't seem to fair much better (although their is precious little data for that). It would imply that 0w/40 ought to be used, but then you have the problem that the 40 might not be thick enough for high temp high load scenarios! (although some people do use 0w/40, more in cold countries, IIRC).
I never directly referenced an old 1980s clevite bearing,and I doubt BE would, where did you see that? The files I referenced are of this century at least! (my Mahle one is from 2014, they own Clevite now).
There is a design flaw, it's clearly there in the clearance numbers, but even if you ignore that, the photos of pulled bearings says it all.
Of the 266 cases with good enough photos I have looked at (and have linked in my data file, I can post it here if you like, I haven't yet as I'm still adding to it), 250 of them are excessively worn, 105 of those are critically worn (only a few are down to debris scoring the shells), i.e they are nearly all the lead/copper type which are showing copper (once copper is visible they could suffer imminent failure, as per the pdfs I linked earlier). Only 12 are moderately worn, and you could argue they are ok, and just 4 are in the good condition which they should be! Of which, 1 of those is the VAC bearings I posted above, and another one was suspected of being non OEM bearings.
So that's 94% of the shells pulled are in a bad way, even if you dismissed the ones which are not critically worn, the remaining 105 account for 39.5% of pulled bearings.
Although it appears as if the LCI bearings fare somewhat better, the trouble with those bearings is that when the embedded layer is worn through on those to copper (it turns out the tin/alu LCI bearings do have a very thin copper layer), it doesn't stand out like it does on the lead/copper ones. I think I have a few photos with LCI bearings giving a slightly goldish colour, which I think means it's worn through the embedding (babbitt) layer, but it really isn't certain, so they're not off the hook either unfortunately.
So all we can say with a good degree of certainty is that a very high proportion of shells are excessively worn.
Are they ALL imminently about to fail? Probably not, but unless you know they've already been changed, you don't know which end of the spectrum your engine lies, nor how much life is left in your bearings.
In addition to that, we can't rely on oil analysis to catch it, it sometimes misses it, either due to tests not being done from the start &/or it not covering the size of particles produced by failing bearings, oh and it can't monitor the tin/alu bearings either.
Anyway, my car's a 2008 model with 75k miles, when I found out earlier in the year that oil analysis is of limited use for bearing monitoring I garaged it, I had planned to change the bearings in April/May but the lockdown hit, and finding measured non BE shells is easier said than done, although I think I have finally found another person who has! (incase you're wondering why I haven't bought BE bearings, it's because they're 3x the cost of ACL ones!! Even allowing for their research that's too much! ).
As for the oil pressure, another factor for oil flow for the oil wedge is clearance, if insufficient it won't allow enough oil flow for the oil wedge to be effective. Agreed cold oil will make that worse.
RE the S65's BMEP, IIRC that essentially means the power of each combustion event (I only remember for sure that it stands for Brake Mean Effective Pressure
), so more related to torque output? I know it's relatively low compared to a turbo diesel or turbo/SC petrol engine, but I wouldn't describe it as low, anyway, semantics .As for BE's oil flow test, I'm not sure which part you are referring to mentioning increased oil pressure with BE bearings, I can only guess you mean a not insignificant amount (so ignoring a few PSI variations here and there), so increased pressure IIRC was only significant at startup, for 1st ~2s. The oil pressure increased to operating pressures more quickly due to the higher flow of oil due to the bigger clearances, the oil got to the sensor (and bearings) faster, and the variable flow/pressure oil pump was mostly able to make up the pressure despite the extra oil losses, see here (I'm baffled too as to why they didn't realise that! lol I'm sure they say somewhere else that's why). So it's not higher oil pressure overall, but it gets up to (sometimes very slightly lower) pressure faster. If you're not referring to that then you'll have to be more specific.
People say 10w/60 is too thick because of the tight clearances that the S65s have, it's the combination that's the problem.
As for the extra power lost due to higher flow rate, that's going to be a tiny amount, if any at all, as that will be countered somewhat by having less resistance to pump it out in the 1st place. I'd imagine the 2 would near enough cancel out?
Afraid you're point about using tight clearances with thick oil to reduce power losses isn't right, thicker oil uses more power to pump (and causes more drag), most modern engine use 0w/20 (or thinner!) or 5w/30 with tight clearances, they have to be tight for the thinner oil, the thinner oil reduces power losses and improves MPG. Too thick an oil through too tight clearances will cause oil starvation.
Interesting about 10w/60 being nearer a 7.5w (I tried to convert the 40C kinematic viscosity of it to a 'w' rating but was unable to find an appropriate chart), that probably explains why people using 5w/50 tentatively don't seem to fair much better (although their is precious little data for that). It would imply that 0w/40 ought to be used, but then you have the problem that the 40 might not be thick enough for high temp high load scenarios! (although some people do use 0w/40, more in cold countries, IIRC).
I never directly referenced an old 1980s clevite bearing,and I doubt BE would, where did you see that? The files I referenced are of this century at least!
(my Mahle one is from 2014, they own Clevite now).There is a design flaw, it's clearly there in the clearance numbers, but even if you ignore that, the photos of pulled bearings says it all.
Of the 266 cases with good enough photos I have looked at (and have linked in my data file, I can post it here if you like, I haven't yet as I'm still adding to it), 250 of them are excessively worn, 105 of those are critically worn (only a few are down to debris scoring the shells), i.e they are nearly all the lead/copper type which are showing copper (once copper is visible they could suffer imminent failure, as per the pdfs I linked earlier). Only 12 are moderately worn, and you could argue they are ok, and just 4 are in the good condition which they should be! Of which, 1 of those is the VAC bearings I posted above, and another one was suspected of being non OEM bearings.
So that's 94% of the shells pulled are in a bad way, even if you dismissed the ones which are not critically worn, the remaining 105 account for 39.5% of pulled bearings.
Although it appears as if the LCI bearings fare somewhat better, the trouble with those bearings is that when the embedded layer is worn through on those to copper (it turns out the tin/alu LCI bearings do have a very thin copper layer), it doesn't stand out like it does on the lead/copper ones. I think I have a few photos with LCI bearings giving a slightly goldish colour, which I think means it's worn through the embedding (babbitt) layer, but it really isn't certain, so they're not off the hook either unfortunately.
So all we can say with a good degree of certainty is that a very high proportion of shells are excessively worn.
Are they ALL imminently about to fail? Probably not, but unless you know they've already been changed, you don't know which end of the spectrum your engine lies, nor how much life is left in your bearings.
In addition to that, we can't rely on oil analysis to catch it, it sometimes misses it, either due to tests not being done from the start &/or it not covering the size of particles produced by failing bearings, oh and it can't monitor the tin/alu bearings either.
Anyway, my car's a 2008 model with 75k miles, when I found out earlier in the year that oil analysis is of limited use for bearing monitoring I garaged it, I had planned to change the bearings in April/May but the lockdown hit, and finding measured non BE shells is easier said than done, although I think I have finally found another person who has!
(incase you're wondering why I haven't bought BE bearings, it's because they're 3x the cost of ACL ones!! Even allowing for their research that's too much! ).
Last edited:
You don't need to own the car to know about the design flaw. He also agrees that it is a design flaw, you are the only one adamant that it isn't, which is baffling to be honest.
The fact of the matter is, I said something, you poo poo'd it. Someone else said the exact same thing, you lapped it up. Happens all the time on here. Acme said it so the automatic stance is to assume he's wrong and oppose it.
Thanks @Assimilator1 for adding such a high level of detail and amassing a portfolio of examples to back up the point. Something I was unable (and unwilling...) to do.
Last edited:
I made that post almost a year and a half ago, and had forgotten what I said. It is a little OTT I grant you!
Surely you can't think it isn't a design flaw? Just that the criticality is exaggerated?
Associate
- Joined
- 26 Oct 2002
- Posts
- 577
- Location
- UK, Surrey, Guildford
No worries Acme, that post on p3 is just a portion of the data file I've collated (mostly based on the rod bearing condition thread at m3post), the full list has over 320 entries now and I've not quite finished (I could post it here, but it would be a ginormous post! lol. Maybe I could attach it? .....don't think I can here?). Only a few more pages of that thread to go through! lol.
Although I grabbed 30 odd from that M3 UK FB group I joined, once I got to over about 3mths old photos FB was painfully slow to load, by 6 mths it was taking 4-6 attempts to get a picture to load! Anyway, I think I'll have enough from the m3post thread.
As a general request to anyone, if you have photos of M3 bearings in a good state (as in a similar condition to the pic I posted above), I'd be happy to add them to my list. I'm also after bearings pulled in a good state from any high mileage engine, I'll be searching for them, but it would be nice if I got some offered to me!
Although I grabbed 30 odd from that M3 UK FB group I joined, once I got to over about 3mths old photos FB was painfully slow to load, by 6 mths it was taking 4-6 attempts to get a picture to load! Anyway, I think I'll have enough from the m3post thread.
As a general request to anyone, if you have photos of M3 bearings in a good state (as in a similar condition to the pic I posted above), I'd be happy to add them to my list. I'm also after bearings pulled in a good state from any high mileage engine, I'll be searching for them, but it would be nice if I got some offered to me!
Last edited:
The Audi R8 V10 is likely to escape these issues for the following reasons:
- It uses a 5w30 oil which is much thinner giving better cold start protection. Also, likely to only be used in only fair weather. I remember leaving some of my 10w60 oil outside on a winters night and it being like treacle, I used to half wince when cranking it up at -10c! The 5w30 used in my girlfriends car was visibly way less viscose
- An Audi R8 is likely to be babied much more
- They sold massively less units than any M car so instances will be much less heard of
- They generally have much lower mileages
- They may well use bigger clearances also
The S2000:
- Again, uses a 5w40 oil giving better cold protection
- May well have bigger clearances
- Anecdotally, car with a V8/V10 is more likely to be revved from cold to show off to your neighbour/co-worker/prospective girlfriend/mechanic/brother/sister/random car enthusiast?
Like I said, the problem with the S65 and S85 (and S54 to some degree) is that it runs tight clearances in combination with a very thick oil AND is one of the only mass produced cars to use this oil so is likely to be the only one with known issues. The other cars that use is oil are Bugatti, Ferrari, , Koenigsegg, Aston Martin etc. I agree that BMW should have run some slightly bigger clearances on these engines (or ditched the 10w60 oil and designed the engine to run something thinner, or a combination of them both), but not because it was a "design flaw", but because they should have anticipated how these cars would have been used by customers and as such, they would give issues down the line.
I bet you that the well known RB issues that have been around in M cars since 2000 won't be present on S55 and S63 engines which use the 5w30 oil now and don't rev out as highly.
Last edited:
Associate
- Joined
- 26 Oct 2002
- Posts
- 577
- Location
- UK, Surrey, Guildford
All good points, except I think the last one about S2000, very dubious .
R8 uses 5w/30 a? Interesting, goes to show you don't need the thicker oil for a high hp high revving engine (when designed for it).
Any idea how many V10s R8s were sold? (I don't know if the V8 versions were a high revver?).
.R8 uses 5w/30 a? Interesting, goes to show you don't need the thicker oil for a high hp high revving engine (when designed for it).
Any idea how many V10s R8s were sold? (I don't know if the V8 versions were a high revver?).
Soldato
- Joined
- 27 Mar 2013
- Posts
- 9,439
On a similar vein, wonder if anyone has run an me or 5 on the thinner oil?All good points, except I think the last one about S2000, very dubious
R8 uses 5w/30 a? Interesting, goes to show you don't need the thicker oil for a high hp high revving engine (when designed for it).
Any idea how many V10s R8s were sold? (I don't know if the V8 versions were a high revver?).
Associate
- Joined
- 26 Oct 2002
- Posts
- 577
- Location
- UK, Surrey, Guildford
Yea some guys in the m3post forum have run 5w/50, 0w or 5w40, I know theirs a thread for the former, their might be one for the latter too.
The few people who report that oil grade and post their pictures in the rod bearing condition thread, seem to fair no better than anyone else though, which along with a small number of people having oil reports going all the way back to nearly the beginning of the car (but not yet one for the running in service), makes some of us wonder if most of the wear happens very early on in the engine's life.
The few people who report that oil grade and post their pictures in the rod bearing condition thread, seem to fair no better than anyone else though, which along with a small number of people having oil reports going all the way back to nearly the beginning of the car (but not yet one for the running in service), makes some of us wonder if most of the wear happens very early on in the engine's life.
Last edited: