Help me decide on a UOA plan for my bearing-eating BMW

Status
Not open for further replies.
Joined
Oct 20, 2005
Messages
11,956
Location
PA
I want to use oil analysis to get as much warning as possible of any developing bearing failure in my '08 M3. These engines are known to be hard on their rod bearings. Mine is going to have its rod bearings replaced preventatively very soon, so I'll be starting fresh. The new bearings are a tin/aluminum alloy.

I plan to sample at every drain to build up a trend and watch for deviations from baseline.

This engine does not have a dipstick tube, so samples have to be pulled from the bottom -- but the drain plugs point downward, so a Fumoto valve would be risky.


Questions:

1. Assuming I have other ways to catch fuel dilution and tune-related problems, I'm thinking I don't need much in a UOA beyond an elemental analysis. Does that make sense?

2. Polaris Labs has nice bulk ordering options: ~$150 + shipping for a 10-pack of barebones tests, or ~$240 for a 10-pack of tests that include particle quantifier (PQ). Is that the best pricing from a reputable lab, or is there another option I should consider?

3. Any other recommended tests, sampling ideas, etc.?
 
This might be the perfect time to try the old shop-vac method! I've never had the chance to try it out, but I've seen it work before. It does require a shop-vac and a second set of hands, but if it works then it would be an easy way to sample without having to do a complete drain.

Remove the oil fill cap, turn on a shop-vac, place the hose over the oil fill hole and then remove the drain plug. The vacuum created by the shop-vac should, in theory, be enough to suspend the oil in the sump and keep it from draining out. Have someone gently create a vacuum leak at the end of the hose and collect what drains out. Re-seal the hose and then insert the drain plug.
 
Looked at a few forum posts on the issue, and there are opinions all over the place. No surprise.
Switching to tin/aluminum alloy bearing material may make it more difficult to detect bearing wear, as the OEM bearings have lead.
Bought some popcorn for this one.
 
Q: Have you thought about trying to snake a tube down into the sump via the oil filter housing?


I always thought that the bearing wear was the result of having to use the spec'd 10w60 in all driving conditions whereas a 0/5w40 might be more appropriate for daily driving.
 
Just do an OA at your oil change. At 66,000 miles my 2002 M Coupe (S54) had 3 times the average lead and copper in my OA. Did rod bearings and saved the engine.
 
With appropriately sized and correctly installed aftermarket bearings, I wouldn't bother with UOAs.

The problem will be fixed and you can (and should!) rest easy.
 
Originally Posted by rooflessVW
With appropriately sized and correctly installed aftermarket bearings, I wouldn't bother with UOAs.

The problem will be fixed and you can (and should!) rest easy.


Agreed. The vehicle is already 10 yrs old. I wonder if the OP is going to keep it for another 5 or more?
 
As far as I understand issue with rod bearings is thick oil and tight tolerances in conjunction with people slamming on gas accelerator.
Originally Posted by rooflessVW
With appropriately sized and correctly installed aftermarket bearings, I wouldn't bother with UOAs.

The problem will be fixed and you can (and should!) rest easy.

As far as I understand issue with rod bearings is thick oil at cold temps. and tight tolerances in conjunction with people slamming on gas accelerator.
I personally would go your route. Change it, and maybe do UOA sometimes down the road.
 
Thanks, guys.

Yes, my plan is to sample mid-stream but only at drains. I mentioned the difficulty of sampling to describe why it'd be tough to sample mid-OCI, which I had hoped to be able to do.

JustN89, the Shop-Vac method is an interesting one. I don't have one but I'd be interested to give it a shot...

Danno, I think the reason people are saying the tin/alu bearings are harder to track with oil analysis is that people are trying to use one-off UOAs. Tin/alu bearings might be wearing in a way that's less likely to show clear signs on a single UOA. But the way you're supposed to use UOAs to track wear is by establishing a trend with many analyses and then looking for deviations. With that method, there should be no reason why lead-free bearings couldn't be tracked.

I don't want to get into the whole debate about why the bearings fail, which bearings are best, etc. I have my opinions but there just isn't enough hard evidence out there to support any given hypothesis about what the problem is or what's the best solution. The important point for me is that they are a failure mode that I will likely have to address at some point while I own this car, and doing it now is better than postponing it in the hopes that I can catch the problem before it causes other damage. No matter which bearings I have installed, I want to keep tabs on them; hence UOAs.

Barring those points, it seems like there's not much disagreement with the tentative plan I laid out, correct?
 
Originally Posted by d00df00d
Thanks, guys.

Yes, my plan is to sample mid-stream but only at drains. I mentioned the difficulty of sampling to describe why it'd be tough to sample mid-OCI, which I had hoped to be able to do.

JustN89, the Shop-Vac method is an interesting one. I don't have one but I'd be interested to give it a shot...

Danno, I think the reason people are saying the tin/alu bearings are harder to track with oil analysis is that people are trying to use one-off UOAs. Tin/alu bearings might be wearing in a way that's less likely to show clear signs on a single UOA. But the way you're supposed to use UOAs to track wear is by establishing a trend with many analyses and then looking for deviations. With that method, there should be no reason why lead-free bearings couldn't be tracked.

I don't want to get into the whole debate about why the bearings fail, which bearings are best, etc. I have my opinions but there just isn't enough hard evidence out there to support any given hypothesis about what the problem is or what's the best solution. The important point for me is that they are a failure mode that I will likely have to address at some point while I own this car, and doing it now is better than postponing it in the hopes that I can catch the problem before it causes other damage. No matter which bearings I have installed, I want to keep tabs on them; hence UOAs.

Barring those points, it seems like there's not much disagreement with the tentative plan I laid out, correct?


Are you still using Dyson for your UOA's?
 
Tin/Aluminum bearings have lower unit load capability than leaded bronze bearings.
Seems to me like you're going backwards. Or are they some of the new-tech bearings with the Ten-Rox polymer coating? The polymer coating is supposed to help maintain lubrication during periods of marginal oil supply.
Are you polishing the crank journals during the bearing change? Just installing new bearings and not polishing the crank would be, ummm,,,, less than optimum.
 
Last edited:
Originally Posted by BMWTurboDzl
Are you still using Dyson for your UOA's?

Haven't in a while. I'd love to but... man, the prices....

Was thinking of going Polaris per the first post, and looking for other economical suggestions.
 
Originally Posted by A_Harman
Tin/Aluminum bearings have lower unit load capability than leaded bronze bearings.
Seems to me like you're going backwards. Or are they some of the new-tech bearings with the Ten-Rox polymer coating? The polymer coating is supposed to help maintain lubrication during periods of marginal oil supply.
Are you polishing the crank journals during the bearing change? Just installing new bearings and not polishing the crank would be, ummm,,,, less than optimum.

Don't want to derail this thread, but I'm happy to talk about this elsewhere or by PM if you're curious.
 
Originally Posted by Danno
I'd like to see bearing materials discussion, thread seems to have died out.

+1
 
Well, alright then.

So, quick background. The engine in my 2008 BMW M3 is a 4L naturally aspirated V8 called the S65. 414 horsepower, 8400 RPM. Seems to like to wear rod bearings much faster than normal.

There's been a typical flood of aftermarket "fixes." Many are just factory bearings with extra anti-friction/anti-wear coatings (e.g. WPC).

Some people noticed that BMW's rod bearing clearances are much tighter than certain industry-standard guidelines would suggest for the high loads and thick oil (10W-60) involved. It's not hard to imagine that those tight clearances might be causing the excess wear. Google BE Bearings for more info; the guys behind it did a bunch of research to validate this idea, and they've made it probably the most widely believed explanation out there right now. They also sell their own bearings, made by Clevite with extra clearance and an anti-friction coating.

Not everyone finds it easy to believe the problem is so straightforward as that. It seems a bit strange to imagine that BMW's best engineers ignored such an easy fix for no good reason, even years after the problems became evident. Not impossible by any means, but not likely either.

There are also some alternative hypotheses for the bearing failures that haven't yet been ruled out, e.g. people running cold engines too hard.

A couple of years after the S65 was released, BMW issued updated bearings with a new design: tin/aluminum (the originals were lead/copper), slightly more clearance (not as much as BE uses), different eccentricity, and tighter tolerances. Of note: there's no way they were still ignorant of the bearing wear issues at this point, and yet they still didn't increase clearances as radically as BE and other companies did.

BE alleges that tin/aluminum was a step backward because it's much harder, which can cause more damage to the crank if a bearing spins. They and some other companies stuck with lead/copper.

Of all the bearings out there, the BMW bearings are the only ones that have seen enough miles and teardowns to give a good idea of how they run. The new tin/aluminum bearings do not seem to have fixed the wear issue entirely, if at all. There isn't remotely enough data on any of the aftermarket alternatives to know how they compare.

For my part, I've learned to never ever screw with OE unless you KNOW aftermarket is going to be better. There's no such evidence for aftermarket bearings for this engine, so I'm going with BMW bearings. A lot of people feel the same way, and a lot of people would disagree (often strongly). It'll be at least a few years before there's enough evidence for a good verdict one way or another.
 
My suggestion is to send the bearing you are going to be using to http://www.arrowlab.com/ labs and have them do an analysis on the elemental content.

Without knowing the metallurgy of your engine components, UOA's will tell you nothing.
 
Originally Posted by d00df00d
Well, alright then.

So, quick background. The engine in my 2008 BMW M3 is a 4L naturally aspirated V8 called the S65. 414 horsepower, 8400 RPM. Seems to like to wear rod bearings much faster than normal.

There's been a typical flood of aftermarket "fixes." Many are just factory bearings with extra anti-friction/anti-wear coatings (e.g. WPC).

Some people noticed that BMW's rod bearing clearances are much tighter than certain industry-standard guidelines would suggest for the high loads and thick oil (10W-60) involved. It's not hard to imagine that those tight clearances might be causing the excess wear. Google BE Bearings for more info; the guys behind it did a bunch of research to validate this idea, and they've made it probably the most widely believed explanation out there right now. They also sell their own bearings, made by Clevite with extra clearance and an anti-friction coating.

Not everyone finds it easy to believe the problem is so straightforward as that. It seems a bit strange to imagine that BMW's best engineers ignored such an easy fix for no good reason, even years after the problems became evident. Not impossible by any means, but not likely either.

There are also some alternative hypotheses for the bearing failures that haven't yet been ruled out, e.g. people running cold engines too hard.

A couple of years after the S65 was released, BMW issued updated bearings with a new design: tin/aluminum (the originals were lead/copper), slightly more clearance (not as much as BE uses), different eccentricity, and tighter tolerances. Of note: there's no way they were still ignorant of the bearing wear issues at this point, and yet they still didn't increase clearances as radically as BE and other companies did.

BE alleges that tin/aluminum was a step backward because it's much harder, which can cause more damage to the crank if a bearing spins. They and some other companies stuck with lead/copper.

Of all the bearings out there, the BMW bearings are the only ones that have seen enough miles and teardowns to give a good idea of how they run. The new tin/aluminum bearings do not seem to have fixed the wear issue entirely, if at all. There isn't remotely enough data on any of the aftermarket alternatives to know how they compare.

For my part, I've learned to never ever screw with OE unless you KNOW aftermarket is going to be better. There's no such evidence for aftermarket bearings for this engine, so I'm going with BMW bearings. A lot of people feel the same way, and a lot of people would disagree (often strongly). It'll be at least a few years before there's enough evidence for a good verdict one way or another.

I am with you on OE. Unless there is absolute evidence that aftermarket is better, OE is safest bet, especially when it comes to such critical part.
I never owned M models, but owned BMW's, and my $0.02 goes to cold starts and heavy foot while oil is still cold. Could be 100% wrong!
 
Status
Not open for further replies.
Back
Top