e46 M3 torque spec thread-- every bolt on the car on diagrams

[NE///M3SIS]

Hey guys, thought I would add my 2 cents.

Just recently finished a complete rebuild on my S54. Video to come in the next week.

I torqued the Chain Tensioner to the recommended torque setting in this guide (50nm/37ft-lb).

I let the engine idle for a couple minutes after reinstalling it into the car and discovered oil leaking from the chain tensioner. YES, a brand new washer from BMW was used. See images below.

Buildjournal says to torque to 70nm. I then torqued the chain tensioner to 70nm/52ft-lb and now there is no more oil leak. So perhaps this file needs to be updated to reflect that?

Cheers.

[heinzboehmer]

Hmm, yeah that's a weird one. TIS doesn't actually list the S54 torque value:




There is that blank line at the bottom with the 50 Nm value, but not sure what that's supposed to be for. Considering that the tensioner is the same part as the one on the S52, I think the 70 Nm value makes more sense.

I've never been able to fit a torque wrench on it though, so I kinda just guess

Anyway, updated the file with the 70 Nm spec: (see first post)

[bmwfnatic]

I think another change should be made to this document.
Page 82 where there is a picture of the diff, there is this odd spec of 13.5 Nm (10 ft-lb) + 40-degree angle.

This seemed specific and odd to me, so I dug a bit deeper to figure out where it came from.

I entered E36 M3 into TIS, since E46 M3 has no instructions really in regard to the internals of the diff.



It is quite easily found where this data comes from, there seems to be a slight mix-up between Nm and ft-lb, but regardless.

However, the diff found in the E46 M3 is not a 'Type K' diff, it qualifies as a 'Type G'
According to TIS, this is determined simply by counting bolts.


This would make the torque spec in the PDF for a totally different type of diff, not good.
I believe the correct spec should therefore be 22 Nm.

Furthermore TIS also states Loctite 270 is required, which is a high strength thread locker, while the PDF currently has 242/243 which is medium strength listed as what should be used.



Since it seems heinzboehmer is the active maintainer of these edits, I think this should be corrected.

[heinzboehmer]

Hmm, BMW definitely made this one confusing. Looks like you're right that the side covers are type G, but the diff case is a 210N:



Confusing cause there's "K"s in both. Plus there's no E46 M3 info about the side covers.

Anyway, I think you're correct, so I updated the doc (see first post). Thanks for bringing this up!

[JudahJD904]

thanks again to the author! this saved my biscuit today.

[Seano666]

Serious question, how does the factory figure out how tight fasteners should be torqued?

[Seano666]

Do they perform tests to figure out how many lb/ft is too tight and what's too loose?

[Shonky]

Do they perform tests to figure out how many lb/ft is too tight and what's too loose?

Some would largely be just based on fastener size and material and in that case the main thing is usually getting even and consistent clamping force across a number of fasteners like an oil pan or simply adequate clamping without stripping e.g. spark plugs.

There are lots where based on feel is just fine hence plenty of "clicks" from YouTube mechanics.

Very critical ones like head gaskets, big end bolts, main bearings etc are based on calculated clamping force. Torquing to specific torques results in fairly precise clamping forces. Torque to yield bolts where the bolt is in the elastic region provides even better consistency in the clamping forces generated.

[Feffman]

Thanks to Flow, back on M3forum:

Original Flow PDF: http://sawashie.free.fr/M3/TorqueSpe...SPECS%20LR.pdf
Updated PDF with suggestions from this thread: https://drive.google.com/file/d/1NMI...ew?usp=sharing

Thank you OH so very much!

Feff

[chris.erdei]

OMG the PDF's in first post are gold !!!

[heinzboehmer]

Okay, so, the spec for the seat rails to chassis in the pdf aligns with what TIS says:




However, I'm fairly convinced that this is a typo in TIS (as evidenced by the three fasteners that I destroyed a couple weeks ago trying to stick to the spec). I dug through TIS and couldn't find a suitable alternative spec, so I'm just going to remove it from the doc and leave a note.

File in first post updated.

[Shonky]

I'd prefer to leave the TIS value in but note it seems to be wrong.

Bentley says 45Nm without the angle torque which seems reasonable.

[heinzboehmer]

I'd prefer to leave the TIS value in but note it seems to be wrong.

Bentley says 45Nm without the angle torque which seems reasonable.

Fair enough! Looking at this again, I agree with you. Updated.

[ac427]

This is a great work and i have been following the updates to the PDF since Flow made it.

I was wondering, could the pages be re-ordered though, to match the list as shown on RealOEM and the EDK ?

Then you could step through both without a lot of paging up and down to find the page in the PDF and the same on RealOEM ?

[Nate047]

Maybe this is a dumb question, but does this doc include, or is there another doc, that mentions which fasteners need to be torqued under load vs which ones can be torqued with the car off the ground?