DIY: S65 M3 Rod Bearing

[Pig Farmer]

Originally Posted by Helmsman

Nice, and what clearance do you expect with the mixed set?

When measuring and selecting the ACL Bearings, I was basically looking to identify eight H bearings with the nominal thickness of .07870” and eight HX bearings with a nominal thickness of .07815” I was able to accomplish this, but I should point out that with the 16 bearings in each set, 2 of the bearings were thinner than nominal, 8 were exactly nominal, and 6 were thicker than nominal.

My engine has a build date of 9/08 so the bearing are the 088/089 version. I believe the nominal thickness of the 088/089 bearings are .07885” (top) and .07865” (bottom). I measured my old bearings as well. I’ll crunch the numbers a bit later and post the results.

[pbonsalb]

So you were going for original stock clearance on replacement, not added clearance?

[Pig Farmer]

Originally Posted by pbonsalb

So you were going for original stock clearance on replacement, not added clearance?

No, when comparing the nominal ACL mixed set to the nominal 088/089 bearings, the additional clearance would be .00065”. Unfortunately, all bearing manufactures have acceptable tolerances, so there are variances in bearing thickness as well as variances in the rod bore and journal. I don’t have the tools, nor could I accurately measure the rod bore and journal diameter without tearing down the motor. The only thing I can be reasonably assured is that the ACL mixed set has increased clearance at the point 90 degrees from the parting line. I found the nominal thickness of the 088/089 bearing at the link below, and this would suggest that the theoretical nominal clearance with the ACL mixed set is .0020” compared to .00135” with the 088/089 bearings.

https://www.m3post.com/forums/showth...t=Rod+bearings

[Pig Farmer]

So hear is the measured thickness of the 088/089 bearing removed from the motor (at the 90 degree point). The bearings had ~51k miles but looked better than most of the pictures posted here. Unfortunately, the deepest wear is typically not at the point 90 degrees from the parting line. The most wear is around 60 degrees from the parting line.

In order of cylinder 1 to 8:

Top/Bottom
.07865/.07870
.07865/.07870
.07840/.07880
.07860/.07865
.07855/.07870
.07865/.07860
.07845/.07860
.07840/.07870

Consistent with the measurements, the top bearing on cylinder 3,7, and 8 showed the most wear.

[Helmsman]

Originally Posted by Pig Farmer

When measuring and selecting the ACL Bearings, I was basically looking to identify eight H bearings with the nominal thickness of .07870” and eight HX bearings with a nominal thickness of .07815” I was able to accomplish this, but I should point out that with the 16 bearings in each set, 2 of the bearings were thinner than nominal, 8 were exactly nominal, and 6 were thicker than nominal.

My engine has a build date of 9/08 so the bearing are the 088/089 version. I believe the nominal thickness of the 088/089 bearings are .07885” (top) and .07865” (bottom). I measured my old bearings as well. I’ll crunch the numbers a bit later and post the results.

Thanks Farmer, very interesting. Curious how you decide the Nominal dimension?

ACL committ to a 0.008mm/0.000315" thickness variation (btw BE has 0.012mm due to coating which they compensate nicely with binning/matching), their own H spec says 0.0783-0.0787 (with 0.0785 "nominal"), and HX 0.0778-0.0781 (0.0780 "nominal").

Also curious on what minimum thickness you measured, to see if it correlate with stated 0.008mm tolerance.

Thanks

[Pig Farmer]

Originally Posted by Helmsman

Thanks Farmer, very interesting. Curious how you decide the Nominal dimension?

ACL committ to a 0.008mm/0.000315" thickness variation (btw BE has 0.012mm due to coating which they compensate nicely with binning/matching), their own H spec says 0.0783-0.0787 (with 0.0785 "nominal"), and HX 0.0778-0.0781 (0.0780 "nominal").

Also curious on what minimum thickness you measured, to see if it correlate with stated 0.008mm tolerance.

Thanks

I found the ACL nominal thickness posted here on M3Post and someone (Bert?) measured the H-STD and HX-STD. I will try to find the post. When I measured the sets I purchased, in both instances 2 bearings were .00005” thinner and 6 were .00005” thicker. As you have probably seen, the surface of the ACL Bearings do not have a completely uniform appearance. They appear dark gray and blue. While I would never rely solely on appearance, the thinner bearings tend to have more blue and the thicker bearings tend to appear a more uniform gray color. I can only assume that the variance in overall thickness is a result of slight variances in the lead overlay.

[Helmsman]

Originally Posted by Pig Farmer

I found the ACL nominal thickness posted here on M3Post and someone (Bert?) measured the H-STD and HX-STD. I will try to find the post. When I measured the sets I purchased, in both instances 2 bearings were .00005” thinner and 6 were .00005” thicker. As you have probably seen, the surface of the ACL Bearings do not have a completely uniform appearance. They appear dark gray and blue. While I would never rely solely on appearance, the thinner bearings tend to have more blue and the thicker bearings tend to appear a more uniform gray color. I can only assume that the variance in overall thickness is a result of slight variances in the lead overlay.

Thanks again Farmer. Go it, yep seen BE's page. Although they clearly know what they're doing I wouldn't go as far as take their suggested spec's of ACL over ACL's own. BE's stated nominal appear to be ACL's max thickness dimensions. Let me know if you're interested in their spec and I'll PM you.

Do I get you right, you measured a 0.00005x2 => 0.0001 variation in respective H and HX set? If so not bad, less then a third of an already pretty tight 0.00031 spec.

[Pig Farmer]

Originally Posted by Helmsman

Thanks again Farmer. Go it, yep seen BE's page. Although they clearly know what they're doing I wouldn't go as far as take their suggested spec's of ACL over ACL's own. BE's stated nominal appear to be ACL's max thickness dimensions. Let me know if you're interested in their spec and I'll PM you.

Do I get you right, you measured a 0.00005x2 => 0.0001 variation in respective H and HX set? If so not bad, less then a third of an already pretty tight 0.00031 spec.

Yes, that was my experience but that’s just two sets. I would expect the variances to follow a bell curve where extreme variances would be less likely. My experience suggests the bell curve is pretty tight but skewed slightly to the right (thicker).

[Green-Eggs]

Originally Posted by deansbimmer

A word of caution to those considering mixing ACL H and HX shells to mimic BE clearances. Be aware that ACL, like most manufacturers, measure and bin their shells to achieve sets that fall within their specified combined clearance. As a result, we've seen a significant amount of HX shells measure the same as H shells, and vice versa.

As a result, one cannot simply buy the two sets, blindly mix the shells, and assume that they've just created the ideal "extra clearance" set. They need to be individually measured, and the typical micrometers are not precise or consistent enough for that task. A proper drop mic and stand are required (shown below), and is similar to the setup used by BE and other qualified engine builders. We match them this way, and the set sold by Lang would likely fall into the category of properly matched sets.

Yes, the drop indicator shown looks very familiar. That looks like the same model Bert has -- minus the data logging cables that Bert uses.

[Green-Eggs]

Originally Posted by Pig Farmer

So hear is the measured thickness of the 088/089 bearing removed from the motor (at the 90 degree point). The bearings had ~51k miles but looked better than most of the pictures posted here. Unfortunately, the deepest wear is typically not at the point 90 degrees from the parting line. The most wear is around 60 degrees from the parting line.

In order of cylinder 1 to 8:

Top/Bottom
.07865/.07870
.07865/.07870
.07840/.07880
.07860/.07865
.07855/.07870
.07865/.07860
.07845/.07860
.07840/.07870

Consistent with the measurements, the top bearing on cylinder 3,7, and 8 showed the most wear.

What equipment did you use to measure? Anvil micrometer, drop indicator, ???

[Green-Eggs]

Originally Posted by Helmsman

Thanks again Farmer. Go it, yep seen BE's page. Although they clearly know what they're doing I wouldn't go as far as take their suggested spec's of ACL over ACL's own. BE's stated nominal appear to be ACL's max thickness dimensions. Let me know if you're interested in their spec and I'll PM you.

This is because of the difference between measure specs vs. manufacturer's theoretical specs. The specs at the BE site are measured specs, not manufacturer's theoretical specs. And to be honest, measured specs are better than manufacturer theoretical specs anyways -- because that's what actually shows up in the field. But the problem with the measured specs -- as I pointed out in another thread a few moments ago -- is that you'll need to measure 1000-2000 shells to get the full variance spread. Whereas the data at the BE site only represents the few sets they've actually measured.

[Green-Eggs]

Originally Posted by Pig Farmer

Yes, that was my experience but that’s just two sets. I would expect the variances to follow a bell curve where extreme variances would be less likely. My experience suggests the bell curve is pretty tight but skewed slightly to the right (thicker).

Yes, it's a bell curve. I've seen Bert's graphs that he keeps for all BE measurements. Bert throws away anything that's farther than +/- 5-sigma. But withing 5-sigma, it's a very normal bell curve as you suspect.

[Pig Farmer]

Originally Posted by Green-Eggs

What equipment did you use to measure? Anvil micrometer, drop indicator, ???

A digital anvil... QuantuMike 293-180-30 with a spherical spindle attachment.

[328i]

Just did this job on a friends car. My suggestion is to remove the subframe entirely, it's only 6 nuts and maybe added about 15 minutes to the job between removal and reinstall. I feel like you will more than make up for that time in how much easier and quicker you'll be able to get to the rods and change the bearings themselves. The last thing I want to be doing when trying to remove the bearing is fighting a dirty subframe. YMMV.

Fan removal also seems unnecessary as was mentioned before, and you risk breaking the thin delicate coolant hose going across the radiator.

Besides that the worst part was the endless amounts of oil that keep dripping on you, especially when you rotate the engine. Seemed like way more than any S54 I've done.

[SYT_Shadow]

Originally Posted by 328i

Just did this job on a friends car. My suggestion is to remove the subframe entirely, it's only 6 nuts and maybe added about 15 minutes to the job between removal and reinstall. I feel like you will more than make up for that time in how much easier and quicker you'll be able to get to the rods and change the bearings themselves. The last thing I want to be doing when trying to remove the bearing is fighting a dirty subframe. YMMV.

Fan removal also seems unnecessary as was mentioned before, and you risk breaking the thin delicate coolant hose going across the radiator.

Besides that the worst part was the endless amounts of oil that keep dripping on you, especially when you rotate the engine. Seemed like way more than any S54 I've done.

I prefer to remove the fewest components possible. Every single thing one touches includes a risk of screwing up.

You think more oil drips down on this engine vs the S54? I did another S54 not long ago and it took longer than the S65.

[pbonsalb]

If you have the luxury of pulling the pan and pump/pickup and then letting it sit overnight, the drips might be fewer. A lift is nice because it gives you room to stand to the side at times to reduce drippage on you.

[Johal E32]

I did this job on my M3 over the last few weekends and I wanted to thank the OP. Your guide is very helpful and I was able to learn a lot about my M3 by doing this job at home.

I used ACL HX Bearings, Corteco motor mounts, genuine BMW gaskets and seals, and ARP fasteners for the rod bolts.

[SYT_Shadow]

Originally Posted by Johal E32

I did this job on my M3 over the last few weekends and I wanted to thank the OP. Your guide is very helpful and I was able to learn a lot about my M3 by doing this job at home.

I used ACL HX Bearings, Corteco motor mounts, genuine BMW gaskets and seals, and ARP fasteners for the rod bolts.

You're welcome, glad it was helpful!

[iamthestig_91]

Absolutely Impeccable DIY. Had some hiccups but it was pretty smooth. Used jackstands, was annoying but not impossible.

I mistakenly disconnected my battery at first and then reconnected it when it said to put the car into neutral. Now my mirror controls are not correlating to the button press and the driver window button controls the passenger window and vice versa. Also, rear park distance control doesn't work either.

Other than that, car runs great! Used BE bearings with ARP bolts.

[pbonsalb]

A real man, doing it on jack stands!

[Helmsman]

Originally Posted by pbonsalb

A real man, doing it on jack stands!

Now with pain in the arms and ears full of oil...

[iamthestig_91]

Haha im still in my 20s so it wasnt terrible. Shoulders were on fire the whole time. Basically was bench pressing everything to its location.

Electric Ratchet Wrench was invaluable