Cockroach Individual M3 5/15/2026 Floating rear rotors, steering coupler, dyno

[samthejam]

Howdy from Georgia!

Untitled by Sam DaJam, on Flickr
Untitled by Sam DaJam, on Flickr
Untitled by Sam DaJam, on Flickr

So we did a group Dyno here in Atlanta, with Epic Motorsports.

I just wanted to double check the AFRs, all looks good.

Untitled by Sam DaJam, on Flickr

Untitled by Sam DaJam, on Flickr

I also did the steering coupler, and finally got some brand used floating rear rotors in there with new PFC 11 pads.

I can now say all 4 of the cheap drilled and slotted rotors I got for free cracked and failed in the same way. Glad I replaced this before this weekend. One day I'll find ZCP calipers.


Untitled by Sam DaJam, on Flickr

Untitled by Sam DaJam, on Flickr

Untitled by Sam DaJam, on Flickr

[Bry5on]

Would a caliper work? One end needs to be flat against the edge of the rim.

Yes but vernier calipers will give you the .1mm accuracy you need without needing a digital or dial tool if that’s what you’re after. I figure that’s probably more reliable to toss in your track setup

[bigjae46]

Vernier calipers

Would a caliper work? One end needs to be flat against the edge of the rim.

[Bry5on]

Using the steering angle reading makes life much easier and, IMO, the secret to my outstanding alignments. I would love to get something that can read tenths of a millimeter from the string to the wheel...I'm sure something exists.

Vernier calipers

[bigjae46]

Using the steering angle reading makes life much easier and, IMO, the secret to my outstanding alignments. I would love to get something that can read tenths of a millimeter from the string to the wheel...I'm sure something exists.

[samthejam]

Leaving to Atlanta this evening

[samthejam]

Relevant docs for Turner Monoballs installation:

https://nam3forum.com/forums/filedata/fetch?id=38673

http://bd8ba3c866c8cbc330ab-7b26c6f3e01bf511d4da3315c66902d6.r6.cf1.rackcdn.co m/013276TMS~A_INS.pdf


I found the best method ended up being painting the control arm for the high spot. Ugly but it was much more efficient for finding the interference. Did another string alignment and like using the steering angle sensor output to keep it more consistent per bigjae46 advice


Untitled by Sam DaJam, on Flickr

Untitled by Sam DaJam, on Flickr

Untitled by Sam DaJam, on Flickr

Untitled by Sam DaJam, on Flickr

[samthejam]

I went ahead and did the Turner monoballs. Man that sucked to install on the driver side. I had to file the control arm to make sure the press fit was too severe and cause binding. Probably 3-4 hours on that side of filing, and hammering it on and off.

Definitely less NVH than the Bimmerworld TrackCABs and just seems to track the road better, reminds me of my E28 now which is all on sphericals everywhere for front and rear.

[samthejam]

They key to alignments is repeatability. The car has to be rolled to settle the suspension so if the car is sitting on wheel cribs, the cribs must be long enough to roll the tire one full rotation. Must do that even when using turn plates. That will be why you get a significantly different reading.

The wheel cribs should be laser leveled to be close. The closer to level...the alignment will be more repeatable.

I find that zeroing the SZL in a scan tool or INPA massively improves repeatability. Even if the wheel looks straight, probably is off enough to skew the toe reading. AND...when most center the wheel they are sticking their head through the window and not sitting in the driver's seat which should have ballast in it anyway using the BMW spec. Ballast in both front seats.

Great point on using the steering angle sensor. I'll add that to the routine. I love that. I make it a point to crawl into the car to put it straight head, but it still has human error.

The tables we use our laser leveled, and I trust the person helping me as our rear measurement was perfect. We always raise and lower the car handbrake off in neutral on the stands and repeat the measurements twice and it is usually very consistent but we are not getting the roll. The stands/tables have large casters that allow it to settle as needed, but it isn't as good as a full roll.

Car is now tracking straight again like when I headed to track Saturday.

[bigjae46]

They key to alignments is repeatability. The car has to be rolled to settle the suspension so if the car is sitting on wheel cribs, the cribs must be long enough to roll the tire one full rotation. Must do that even when using turn plates. That will be why you get a significantly different reading.

The wheel cribs should be laser leveled to be close. The closer to level...the alignment will be more repeatable.

I find that zeroing the SZL in a scan tool or INPA massively improves repeatability. Even if the wheel looks straight, probably is off enough to skew the toe reading. AND...when most center the wheel they are sticking their head through the window and not sitting in the driver's seat which should have ballast in it anyway using the BMW spec. Ballast in both front seats.

[samthejam]

Untitled by Sam DaJam, on Flickr
Untitled by Sam DaJam, on Flickr
Untitled by Sam DaJam, on Flickr


Car has not moved too much, I am going to Road Atlanta in 2 weeks? But we are also in process of buying a house that has a detached shop, so hopefully I can finally have a space of my own to wrench in instead of borrowing someone elses.

I checked the alignment, I did not it felt like it shifted at last PCA event. It did. It went from 4mm toe out to 5mm toe in. Why? idk nothing is obviously loose or broken. Everything is paint marked, there is no smoking gun.

I hate those bimmerworld trackcabs.

I will be switching to Turner Monoballs and hopefully some new Lemforder arms. I have some takeoff ones but would like to do new, but we will see how the home goes. I have a dyno scheduled in Atlanta just to check it over, wanted to do new tires and pads but these will have to do.

[samthejam]

"Ill send u all the data in a bit but basically had no tangible wear metals present comparing virgin to used. Really clean stuff, roughly the same amt of wear I see with my 530 over 5k miles. It did shear 20% but bc its 10w60 theres still plenty of room, and the visc was still heavier than a new 40 or even 50 weight oil. Could you get something more shear resistant like redline or motul? Yes, but with how frequently you change it I dont think it’s going to matter."

VOA Result:



UOA Result:


This oil had one dyno day, one full weekend. Since this was the first weekend with thermal control of the coolant. I did back to back sessions see how hot I could get oil. First time to 3rd dot, after heat soaked for sometime in paddock before going back out on track. Maybe 1000 street miles.

I am not sure of the OE bearing new construction. I believe they are Alu and Tin but need to do more research. From my friend and ChatGPT nothing seems concerning, so I'll keep to my current schedule. Do more test and hopefully can begin a trend.

[Bry5on]

The other one I can think of that might have a non-negligible effect is that the AlphaN table on the standard M3 software is smaller (18x12) than that CSL AlphaN table (20x24). In particular the y-axis has twice as many interpolation points. AlphaN on the CSL software will get you a better end result I'd have thought.

Plus EGT enrichment compensation for low temp/load

[karter16]

FYSA there are other differences in the CSL software than just the MAP sensor that help with drivability, but the MAP is the main one.

The other one I can think of that might have a non-negligible effect is that the AlphaN table on the standard M3 software is smaller (18x12) than that CSL AlphaN table (20x24). In particular the y-axis has twice as many interpolation points. AlphaN on the CSL software will get you a better end result I'd have thought.

[Bry5on]

It's on a euro 1501 software as a base file with Epic's Alpha-N software changes ontop. IAT is in a similar placement to stock, so should be fine? So should be following Euro M3 logic, and sensor is sort've in same place as the factory MAF, but thats a great point and something to consider if I switch airboxes or go to a MAP sensor setup at some point on underlying DME logic for that

Ah yeah, can't say for sure either way then but I expect you're right and you're fine with your current 1501 Alpha-N. And also yes on if you switch to a CSL based and/or MAP setup. FYSA there are other differences in the CSL software than just the MAP sensor that help with drivability, but the MAP is the main one.