VAC Coated Rod Bearings, A FULL engine teardown after 39,500 Miles. Look inside!

[avusm3]

My belief is that the techs/robots assembling the engine at the factory followed the prescribed torque procedure, but that either that torque procedure and/or forces from high rpm use caused the rod bolts to be pushed past their yield point and they permanently stretched (lengthened) a small amount. Once they were in a permanently stretched state they lost some/most of their clamping force on the bearing. Lower clamping force on the bearing led to higher bearing wear and finally failure. This permanent stretching would also have led to a reduced measured torque value for the bolts on dis-assembly as Malek is reporting.

I'd be very interested to hear exactly what was measured in terms of reduced torque on 4 & 5.

I just lost a motor in very similar circumstances to Malek's (2011, 43k, no mods) and this is the only explanation that makes any sense to me.

[pbonsalb]

When did BMW go to doing the 3 stage torque sequence 3 times for new rod bolts? Did it start prior to the S65/85? I had not seen it specified for the S54 or S52. but have not seen everything. If it came out sometime recently, maybe it is an indication BMW knew there were issues with new rod bolts. I did the 3 stages 3 times for each bolt when I changed bearings and bolts this summer and thought it was a pain in the rear and was wondering why it was necessary. Seemed like I was pre-stretching the bolts.

[Rupes]

Malek, in your own personal situation, what ultimately will happen to the engine? Can it be repaired or is it toast? We now know this is an issue, I'm just curious how often engines can be repaired after failure and at what cost.

[bigjae1976]

Originally Posted by pbonsalb

When did BMW go to doing the 3 stage torque sequence 3 times for new rod bolts? Did it start prior to the S65/85? I had not seen it specified for the S54 or S52. but have not seen everything. If it came out sometime recently, maybe it is an indication BMW knew there were issues with new rod bolts. I did the 3 stages 3 times for each bolt when I changed bearings and bolts this summer and thought it was a pain in the rear and was wondering why it was necessary. Seemed like I was pre-stretching the bolts.

I believe the S54 had a similar procedure...which is why a lot of people go with ARP bolts after their bearing swap.

[Dave07997S]

Originally Posted by JEllis

Interesting you mentioned airline crash. I was thinking the same thing. In the aviator world we use the analogy of swiss cheese. Normally the holes don't all line up. Something like an out of spec torque wrench, for example, is caught at the QA level. But sometimes a few seemingly simple mistakes/holes line up to provide a catastrophic failure.

In this case, if we were to speculate, you could say that BMWs use of tight tolerances, combined with something as simple as an out of calibration torque wrench or a bad batch of bolts, add in less than stellar quality assurance and you have a failure.

Of course, this is just fun speculation and should not be taken as anything more.

Yes it is speculation, but one that probably has legs.

Dave

[Dave07997S]

Originally Posted by Rupes

Malek, in your own personal situation, what ultimately will happen to the engine? Can it be repaired or is it toast? We now know this is an issue, I'm just curious how often engines can be repaired after failure and at what cost.

With the cracked block its probably toast, he will probably be able to salvage some parts from the block. At this point though its probably better to find a used motor and build that.

Dave

[Dave07997S]

Originally Posted by Rupes

Fair enough, but then why perfect a example here after 39k miles using bearings with tighter clearances?

I get the sense that you have some thoughts of your own on this matter, but are making us pull it out of you one post at a time. What is your explanation?

My feeling is because the new bearings were put in their with the correct procedure and torque sequence.

Dave

[kawasaki00]

Will explain more later but there are no signs of the bolts being loose.
Look at the bearing wiki and I posted pictures of bearings that are moving around in the saddle from bolts that are not good enough or too little crush. The back side of many bearings do not show they have been moving in the saddle.
It cant happen folks. If the stretch is lost the bolt will fall out. I have yet to see a bearing that is moving in the saddle of the rod and yet to see a rod that is fretted at the parting line.
Have many experiences with bolts that are inadequate even with doweled rods. If the bolts do not have enough clamp load it will shear the dowel pins in half. Have seen this happen on many occasions. There is no evidence that the broken caps are fretting together.
Also, it is impossible to measure break away torque on a TTY bolt. One cannot say a bolt was loose from factory if it is a TTY bolt

Can the s65 benefit from the better bolts... of course, bolts cant come loose and continue to run. If there were enough force to stretch the bolt beyond its original state and lose clamp load the bolt will eventually come out because every time it stretched it becomes looser and looser.

[JEllis]

Originally Posted by kawasaki00

Will explain more later but there are no signs of the bolts being loose.
Look at the bearing wiki and I posted pictures of bearings that are moving around in the saddle from bolts that are not good enough or too little crush. The back side of many bearings do not show they have been moving in the saddle.
It cant happen folks. If the stretch is lost the bolt will fall out. I have yet to see a bearing that is moving in the saddle of the rod and yet to see a rod that is fretted at the parting line.
Have many experiences with bolts that are inadequate even with doweled rods. If the bolts do not have enough clamp load it will shear the dowel pins in half. Have seen this happen on many occasions. There is no evidence that the broken caps are fretting together.
Also, it is impossible to measure break away torque on a TTY bolt. One cannot say a bolt was loose from factory if it is a TTY bolt

Can the s65 benefit from the better bolts... of course, bolts cant come loose and continue to run. If there were enough force to stretch the bolt beyond its original state and lose clamp load the bolt will eventually come out because every time it stretched it becomes looser and looser.

Interesting. Do you have a theory?

[Hujan]

Originally Posted by JEllis

Interesting. Do you have a theory?

I assume he would say inadequate clearances. It's a good theory but doesn't explain Malek's finding with the VAC bearings. VAC and Harrop are not idiots and they employ thicker bearings apparently with some success. This suggests the clearances theory is not a 100% lock (or that there are other factors at play in addition to clearances).

[OM VT3]

Originally Posted by Hujan

I assume he (and Pencil Geek/Regular Guy if he weren't banned for using multiple aliases and flaming board sponsors) would say inadequate clearances. It's a good theory but doesn't explain Malek's finding with the VAC bearings. VAC and Harrop are not idiots and they employ thicker bearings apparently with some success. This suggests the clearances theory is not a 100% lock (or that there are other factors at play in addition to clearances).

With out checking clearances on that motor there has been 0 evidence given so no one can say it's bolts or bearings also id love a reply to know how under torque was checked on TTY bolts

[pbonsalb]

Originally Posted by bigjae1976

I believe the S54 had a similar procedure...which is why a lot of people go with ARP bolts after their bearing swap.

When I looked up the S54 torque, I saw just one 3 stage sequence. No requirement it be repeated 3 times for new bolts. No big deal doing a 3 stage versus the arp single stage, but it is a pain to do a 3 stage 3 times for each bolt.

[Malek@MRF]

Originally Posted by kawasaki00

Will explain more later but there are no signs of the bolts being loose.
Look at the bearing wiki and I posted pictures of bearings that are moving around in the saddle from bolts that are not good enough or too little crush. The back side of many bearings do not show they have been moving in the saddle.
It cant happen folks. If the stretch is lost the bolt will fall out. I have yet to see a bearing that is moving in the saddle of the rod and yet to see a rod that is fretted at the parting line.
Have many experiences with bolts that are inadequate even with doweled rods. If the bolts do not have enough clamp load it will shear the dowel pins in half. Have seen this happen on many occasions. There is no evidence that the broken caps are fretting together.
Also, it is impossible to measure break away torque on a TTY bolt. One cannot say a bolt was loose from factory if it is a TTY bolt

Can the s65 benefit from the better bolts... of course, bolts cant come loose and continue to run. If there were enough force to stretch the bolt beyond its original state and lose clamp load the bolt will eventually come out because every time it stretched it becomes looser and looser.

As I have made clear in previous posts, I think all speculation and any type of accusations should be set aside until I take the time to fully spec both of these engines out in different configurations and check clearances to be sure. I have also made it clear that what I find is not a "ploy" as I have been accused of privately to dismiss any other findings with bearing clearances and tolerances. Long before the bearing Wiki thread those of us with experience with BMW engines (mainly the S54, S65 and S85) know the clearances are too tight from the factory to begin with. Of course it's not possible that all these engines suffer from this, but the clearances are on the tight side.

I have been asked many times how one can measure bolt torque on the BMW stretch hardware (Torque to Yield), definitively you cannot, this is an obvious fact. However, a crude way of checking it is still possible in the hands of the assembler, and I am sure you know this better than many as this is what you do for a living. With the Snap-On Tech-Angle torque wrenches, when set in reverse, one can simply observe how many degrees it takes while releasing the fastener until it breaks loose. Cylinders 5+6 on my own engine with respect to connecting rod bolts broke free much sooner and easier than the other connecting rods.

Please don't take this that my findings definitively indicate the hardware being the end all problem and that upgraded hardware is the solution.

Interestingly enough, no thoughts have expressed any on how the bearings from the first engine posted are BETTER with the Calico coated bearings (which have been known to be worse for bearing clearances, with ARP hardware after 33,xxx miles vs. the OE bearings + OE hardware from the same engine with the same rods and same crankshaft at 6000 miles.

Things we know 100% from the first engine:

-Same crankshaft
-Same connecting rods
-First 6000 miles used with OE Bearings+Bolts show excessive and worrisome bearing wear.
-Remaining 33,xxx miles are on Calico coated OE bearings+ ARP rod bolts showing no bearing wear.

Things that will be done with first engine:

-Crankshaft will be mic'd for specifications on all 8 journals.
-Connecting rod clearances calculated with rods loaded with the original OE bearings and torqued with OE hardware.
-Bearing shell thickness will be mic'd for both Calico coated bearings and the OE bearing shells.

The point of this thread wasn't to engage in some sort of contest as to who is right and who is wrong; the point is to share information. I am not sure why this thread is being sidetracked into anything else but that.

Originally Posted by leigh

With out checking clearances on that motor there has been 0 evidence given so no one can say it's bolts or bearings also id love a reply to know how under torque was checked on TTY bolts

Please see above. I will take the time to spec these engines out. It's the only way to be sure of anything.

[OM VT3]

Originally Posted by Malek@MRF

As I have made clear in previous posts, I think all speculation and any type of accusations should be set aside until I take the time to fully spec both of these engines out in different configurations and check clearances to be sure. I have also made it clear that what I find is not a "ploy" as I have been accused of privately to dismiss any other findings with bearing clearances and tolerances. Long before the bearing Wiki thread those of us with experience with BMW engines (mainly the S54, S65 and S85) know the clearances are too tight from the factory to begin with. Of course it's not possible that all these engines suffer from this, but the clearances are on the tight side.

I have been asked many times how one can measure bolt torque on the BMW stretch hardware (Torque to Yield), definitively you cannot, this is an obvious fact. However, a crude way of checking it is still possible in the hands of the assembler, and I am sure you know this better than many as this is what you do for a living. With the Snap-On Tech-Angle torque wrenches, when set in reverse, one can simply observe how many degrees it takes while releasing the fastener until it breaks loose. Cylinders 5+6 on my own engine with respect to connecting rod bolts broke free much sooner and easier than the other connecting rods.

Please don't take this that my findings definitively indicate the hardware being the end all problem and that upgraded hardware is the solution.

Interestingly enough, no thoughts have expressed any on how the bearings from the first engine posted are BETTER with the Calico coated bearings (which have been known to be worse for bearing clearances, with ARP hardware after 33,xxx miles vs. the OE bearings + OE hardware from the same engine with the same rods and same crankshaft at 6000 miles.

Things we know 100% from the first engine:

-Same crankshaft
-Same connecting rods
-First 6000 miles used with OE Bearings+Bolts show excessive and worrisome bearing wear.
-Remaining 33,xxx miles are on Calico coated OE bearings+ ARP rod bolts showing no bearing wear.

Things that will be done with first engine:

-Crankshaft will be mic'd for specifications on all 8 journals.
-Connecting rod clearances calculated with rods loaded with the original OE bearings and torqued with OE hardware.
-Bearing shell thickness will be mic'd for both Calico coated bearings and the OE bearing shells.

The point of this thread wasn't to engage in some sort of contest as to who is right and who is wrong; the point is to share information. I am not sure why this thread is being sidetracked into anything else but that.



Please see above. I will take the time to spec these engines out. It's the only way to be sure of anything.

It might have been better to check it with torque in reverse but I don't know if that's possable

[Malek@MRF]

Originally Posted by Rupes

Malek, in your own personal situation, what ultimately will happen to the engine? Can it be repaired or is it toast? We now know this is an issue, I'm just curious how often engines can be repaired after failure and at what cost.

The second engine posted is officially unusable and trash. Bed plate is damaged, so is the block. The bed plate and block come as one assembly when purchased new as well, so the parts cannot be mix and matched either. The sides are stamped with corresponding manufacturing numbers to indicate the bed plate and main block match one another.

Originally Posted by Dave07997S

With the cracked block its probably toast, he will probably be able to salvage some parts from the block. At this point though its probably better to find a used motor and build that.

Dave

At this point the only thing that can be re-used from the original engine are the accessories and the cylinder heads and associated parts not related to the block and rotating assembly. Cylinder 1+2 have some scratching on the cylinder walls as well as cylinders 5+6. The Pistons themselves on those corresponding cylinders have some scratching on them as well. If pictures are wanted to depict the damage more clearly, I can post those as well tomorrow when I am back at the shop.

[StealYourFace]

Wow, nasty stuff.

Malek, if this car was stock other than exhaust, and only has 39k miles, isn't it still under warranty, if so could you get bmw to do the right thing and pay for a new motor?

I know you are tearing it down for the sake of research and sharing info, but it sucks that you have to eat the cost of a motor because of that.

[avusm3]

Originally Posted by JEllis

Interesting. Do you have a theory?

+1

[Hujan]

Originally Posted by leigh

With out checking clearances on that motor there has been 0 evidence given so no one can say it's bolts or bearings also id love a reply to know how under torque was checked on TTY bolts

I think you're missing the point:

The same engine that showed severe bearing wear at 6,000 miles with OEM bearings and bolts, magically showed no wear after 30,000 with VAC coated bearings (i.e., thicker than OEM bearings) and ARP bolts.

Unless the clearances magically got significantly looser after 6,000 miles, you have to admit this is a bit of a head-scratcher if the root cause of the initial bearing wear was excessively tight clearances.

The only variable between the two was a supercharger. Do you believe that actually helps ameliorate tight rod-bearing clearances? If so, I would appreciate understanding the physics involved.

[SenorFunkyPants]

Originally Posted by Hujan

I think you're missing the point:

The same engine that showed severe bearing wear at 6,000 miles with OEM bearings and bolts, magically showed no wear after 30,000 with VAC coated bearings (i.e., thicker than OEM bearings) and ARP bolts.

Unless the clearances magically got significantly looser after 6,000 miles, you have to admit this is a bit of a head-scratcher if the root cause of the initial bearing wear was excessively tight clearances.

The only variable between the two was a supercharger. Do you believe that actually helps ameliorate tight rod-bearing clearances? If so, I would appreciate understanding the physics involved.

The only thing that occurs to me would be the associated tune that accompanies adding a supercharger changed the fueling and ignition. Its vaguely possible that the engine was originally suffering from some level of detonation that was eliminated by the tune...but it doesn't seem especially convincing given that the boost itself significantly raises the upper cylinder pressures.

[MFKN3]

I've been told by my engine builder that the BMW torque specs are wrong

[Dave07997S]

Originally Posted by Malek@MRF

The second engine posted is officially unusable and trash. Bed plate is damaged, so is the block. The bed plate and block come as one assembly when purchased new as well, so the parts cannot be mix and matched either. The sides are stamped with corresponding manufacturing numbers to indicate the bed plate and main block match one another.



At this point the only thing that can be re-used from the original engine are the accessories and the cylinder heads and associated parts not related to the block and rotating assembly. Cylinder 1+2 have some scratching on the cylinder walls as well as cylinders 5+6. The Pistons themselves on those corresponding cylinders have some scratching on them as well. If pictures are wanted to depict the damage more clearly, I can post those as well tomorrow when I am back at the shop.

Don't sweat it Malek, I take your word for it. Sounds like you are busy enough big guy!

[Hujan]

Originally Posted by SenorFunkyPants

The only thing that occurs to me would be the associated tune that accompanies adding a supercharger changed the fueling and ignition. Its vaguely possible that the engine was originally suffering from some level of detonation that was eliminated by the tune...but it doesn't seem especially convincing given that the boost itself significantly raises the upper cylinder pressures.

Good point. I forgot about the tune. That, too, may be a variable. (As might other transient things like gas quality, etc.) But I always assumed detonation would result in damage to the cylinders not rod-bearing wear but I'll freely admit that is a very speculative assumption on my part.