Black & Tan 332iT

[heinzboehmer]

I was pondering torsional rigidity last night - do you think there would be anything at all to be gained by replacing the 2x transmission/exhaust tunnel brackets with a single torsion plate across those 8x mounting points? Not from the perspective of preventing the tunnel from spreading out /collapsing in but in terms of helping to prevent longitudinal twisting? In theory it would make a non-zero difference, but the two brackets are fairly close together so not sure if there would be much, if anything, to be gained in practice? The non-M3 convertible had a similar sort of thing (although I believe that was only 4x mounting bolts not 8). BMW didn't do that on the M3 vert though so 🤷♂️

In case you want to do some more pondering: https://wilhelmraceworks.com/blog/me...ssis-stiffness

I'm currently pondering about the best way to build a similar thing for myself to get data from my car. Would be awesome to be able to quantify these changes instead of trying to feel them out.

About the tunnel brace, I think a plate sandwiched between the stock parts would yield the most gains for the least effort. Boxing the stock parts in would be even better, but that requires welding. Non-M part is likely the best answer though.

[Obioban]

Ordered a couple

[Obioban]

Thanks!

[Bry5on]

3D file attached as well, by request

[Bry5on]

Here you can see more clearly how I got my bends incorrect. When bent correctly, the holes should be offset right and pretty centered around the subframe drain hole.

[Bry5on]

I really have no concept of what kind of loads the hardware is being subjected to here. Sheer point of a M6 bolt is ~2000 lbs, and we have two of them-- Do you think we actually need M8 hardware?

Semi related: CAD file or sendcutsend link?

Awesome idea! I'm increasingly convinced that torsional rigidity mods are second only to weight savings-- and they come with much less cost/trade off.

I think M6 is likely not going to fail honestly, but the bearing area under the fastener head is pretty small, so the joint could slip. I also don’t have a good sense of the loads, but I can tell you that if I don’t push hard on the x-brace toggle clamps to tension up that joint, I can hear the joint slip in the cabin.

I also totally agree with you on torsional rigidity being of high importance. My stock M3 springs feel surprisingly firm now after reducing the “chassis as a spring” effect. These low weight (lighter weight for Slon brace!) mods are netting massive benefits. I would prioritize these changes over aftermarket suspension all day any day for any car.

I’ve requested to share the cart from sendcutsend, but here’s the DXF for now. I’m going to need to re-do that sendcutsend cart as I forgot the bracket isn’t symmetrical and mine is bent upside down. No functional change, but the two subframe fasteners aren’t centered in my car for aesthetics

edit: all six bends are 90 and in the same direction

[Obioban]

I really have no concept of what kind of loads the hardware is being subjected to here. Sheer point of a M6 bolt is ~2000 lbs, and we have two of them-- Do you think we actually need M8 hardware?

Semi related: CAD file or sendcutsend link?

Awesome idea! I'm increasingly convinced that torsional rigidity mods are second only to weight savings-- and they come with much less cost/trade off.

[Bry5on]

I made it over to the lift today, drilled out those M6 rivet nuts and replaced them with the M8 I'd originally intended. There's not quite enough room for the rivet nut to insert all the way, so I had to nip 2mm off of them so that they'd fully seat before I riveted them in. Tight squeeze!

Also, today I took the Porsche anti-squeal pads off of the rear of the car in addition to removing the ones on the front. They were on top of the textar anti-squeal pads already bonded to the brake pads. Of course the brakes got noisy coming to a stop after I removed the front ones, so I gave all 4 pads the sil-glyde treatment today and so far so good. Pad dragging issues are FINALLY GONE and the pedal feel is a bit nicer. Whew, only a year and a half to solve that one.

Do NOT add the Porsche anti-squeal pads to your 996 textar brakes!!

[Bry5on]

I was pondering torsional rigidity last night - do you think there would be anything at all to be gained by replacing the 2x transmission/exhaust tunnel brackets with a single torsion plate across those 8x mounting points? Not from the perspective of preventing the tunnel from spreading out /collapsing in but in terms of helping to prevent longitudinal twisting? In theory it would make a non-zero difference, but the two brackets are fairly close together so not sure if there would be much, if anything, to be gained in practice? The non-M3 convertible had a similar sort of thing (although I believe that was only 4x mounting bolts not 8). BMW didn't do that on the M3 vert though so 🤷♂️

I was just talking with some friends about this. I think the cabrio part is the answer as in that case the four inner fasteners won’t be doing much. I’m keeping my eye out for one and will swap if I come across one. It’s also lighter which doesn’t hurt.

[karter16]

On top of that, it does seem to be even more torsionally rigid when exiting extreme driveway angles.

I was pondering torsional rigidity last night - do you think there would be anything at all to be gained by replacing the 2x transmission/exhaust tunnel brackets with a single torsion plate across those 8x mounting points? Not from the perspective of preventing the tunnel from spreading out /collapsing in but in terms of helping to prevent longitudinal twisting? In theory it would make a non-zero difference, but the two brackets are fairly close together so not sure if there would be much, if anything, to be gained in practice? The non-M3 convertible had a similar sort of thing (although I believe that was only 4x mounting bolts not 8). BMW didn't do that on the M3 vert though so 🤷♂️

[Bry5on]

As you know, I “just” got the M3 and have no cracks and yet to schedule my reinforcement. Your brace, even if I got the exact fit, doesn't fit the coupe, right? Just trying to understand where to throw my wallet.

I'm doing my front suspension refresh next weekend - been dealing with family stuff. I'll grab your address and ship it all to you

Sweet, thank you!

This v-subframe-brace fits the M3 only. Since my wagon is converted to all M3 underneath, it fits my wagon The top X-brace behind my fold down seats will not fit the coupe as it's substantially different back there though.

I'll get this cart saved for sharing so you, Heinz, Stas, and any of the other Bay Area folks can do a batch order to save on pricing.

[YoitsTmac]

If I had an M3 with no cracks in the floor I'd do this, solid subframe bushings and a rear-only vincebar install and call it done. That would be a huge upgrade in performance and stiffness with almost no downside (the slight NVH of solid subframe bushings). Then add a Slon wall if you're alright losing the fold down seats and making that compromise.

As you know, I “just” got the M3 and have no cracks and yet to schedule my reinforcement. Your brace, even if I got the exact fit, doesn't fit the coupe, right? Just trying to understand where to throw my wallet.

I'm doing my front suspension refresh next weekend - been dealing with family stuff. I'll grab your address and ship it all to you

[Bry5on]

Great stuff! loving following all this and seeing all these rigidity improvements.

(also amused by the growing number of intake scoops on your bench in every new photo)

You know, I've got to throw an easter egg in there for you!

Got the chance to put more miles on the setup today, and all of the above still holds except for #3 - after warming up the gear oil, the m-clunk is still there, although it's now a more muted m-thud without the distinct clunk noise there. Additionally, I've got a lot more shifts in and it's very noticeable there. The drivetrain doesn't really wind-up or wind-down so there's less active thinking and feathering of the throttle to make shifting smooth. Tip-in and tip-out are way more responsive but also way smoother. That slight delay on throttle lift before you can feel the deceleration seems to be very nearly gone, and starting from a stop is easier to be both smooth and quick at the same time, whereas before it was kind of one or the other. My wife and I also both thought the car was quieter on the highway.. bonus. I was skeptical that I might have just been experiencing the placebo effect after the change, but I'm no longer skeptical.

Also now that I've had more time with it, I'm super convinced that this will mitigate subframe carrier failure. Feeling the absence of the movement has convinced me just how much carrier movement there was and even still is with the vincebar alone. If I had an M3 with no cracks in the floor I'd do this, solid subframe bushings and a rear-only vincebar install and call it done. That would be a huge upgrade in performance and stiffness with almost no downside (the slight NVH of solid subframe bushings). Then add a Slon wall if you're alright losing the fold down seats and making that compromise.

[karter16]

Early days and not a whole lot of seat time but I can say reasonably confidently that the hypothesis has been validated, and this brace mitigates a lot of that flexibility.

Great stuff! loving following all this and seeing all these rigidity improvements.

(also amused by the growing number of intake scoops on your bench in every new photo)

[Bry5on]

Another win today.

This evening I decided to install the first version of the v-subframe-brace.

As a recap, the hypothesis is that:
1) The remaining rear bump oscillations I felt after installing the X-brace are related to a lack of fore-aft/bending stiffness
2) The subframe can rock forward aft, sort of 'around' the axle axis, under throttle application, related to subframe mount failure, and the stock v-brace is not enough to constrain this movement
3, unstated so far) The m-clunk is related to 2) and additionally, the subframe floor acts as a bit of a speaker, amplifying the clunk






I wasn't thinking for a bit and ended up installing M6 rivet nuts instead of M8 rivet nuts, probably because I decided to do this on jack stands instead of driving the 2 miles down the street and getting on the lift. At some point I may try to remove the M6 guys and go to M8 as the design intended, but we'll see how we get along for now.

Early days and not a whole lot of seat time but I can say reasonably confidently that the hypothesis has been validated, and this brace mitigates a lot of that flexibility.

Response and oscillations following bumps are much tighter, and throttle applications are more 'instant' and don't have that coil-up feeling that the e46 has. You can be a bit rougher with throttle application and removal and it doesn't feel as rough on the chassis, it just follows your foot.

I've only got a handful of miles on the car, attempting to provoke it of course, but so far I haven't been able to get it to m-clunk. You can for sure get it to cluck by going on/off/on/off throttle very quickly, but this is true even on the newer G80 cars. Shifting feels a little smoother? Not so much in the lever, but in the car itself. Definitely need more miles to validate this.

On top of that, it does seem to be even more torsionally rigid when exiting extreme driveway angles. I can tell again that the front is less rigid than the rear. I wonder if recreating the e30 cabrio front strut mount gusseting would be useful. I'm really going to have to do those control arms and a new set of bushings now.

Anyway, pleased with this! The car has really transformed a generation or two in ride feel in the past few weeks.