BimmerFest BMW Forum banner
1 - 20 of 83 Posts

·
19,585 Miles
Joined
·
11,141 Posts
Discussion Starter · #1 · (Edited)
The '99/'00 M-Z3 coupe with the S52 engine has a 3.23:1 differential.
The '01/'02 M-Z3 coupe with the S54 engine has a 3.15:1 differential.
Based on the 7600 rev limit on the S54 engine, the 3.15 differential is not a good match for the car in my opinion. Have no idea why BMW really did this.
Looking to change this, I first focused on the 3.38:1 based on the numbers, and settled on the 3.46:1 used in the Z3-3.0's with automatic.



Back then, I had no idea if this unit would fit or work with the M. At the time we knew that the rear covers were different. We also knew that the output flanges were different, but based on the parts data the M ones would swap right in. The driveshafts are different lengths too.
But based on both differentials using the same input flange, we assumed it would work.
Al Corey (took the chance) purchased one to see. The differential retails for $2200.00 with a $100.00 core charge. So, if you want, you can keep your stock differential for $100.00.
He installed it and it fit fine. When we took it out for a test drive there was no speedometer reading. Turns out the speed signal generator (spider) is different, diameter wise between the two differentials.
Since BMW no longer sells internal differential parts, there was no way to check this since the internal parts no longer show up on the parts CD. They show up if you know the part number, but that's another story.
Al modified his speed signal sensor so that it would go in more. It worked.
Randy Forbes swapped out the speed signal generator (spider) with the one from his M differential and it worked too.
I had my machinist modify a speed sensor based on what Al found out. He also made me a plate to accommodate the shortened sensor.




After three unsuccessful tries, the fourth differential I ordered showed up undamaged.
Time for me to finally 3.46 my '01 M coupe.
Last time I swapped out a differential was over forty years ago in my '61 3 speed Chrysler. Used to swap back and forth a 4.11 with the stock 3.23 as needed. Looking back, I have no idea how I did this.
I remember slam hammering the axles out and removing the rear cover.
Five years earlier my first car was a 40 Ford Deluxe coupe.
Pontiac V8 with a '39 Lasalle tranny and Ford rear end. No idea how I did this either. At fourteen, I never finished the car. The guy I sold it to did though. My older brother, being into cars, got me into cars.
Part 2 to come.
Preview: I got the stock differential out.

 

·
Registered
Joined
·
7 Posts
Going to ask a stupid question...

Will this make the car easier to drive when in 5th gear?

My brother has a 95 M3 and in 5th gear on the freeway it revs sort of high. I thought a 6th gear would make the car easier to live with on long drives.

Is this the same sort of thing what you are doing will do? Lower the rpms when in 5th gear....

Sorry for such a simplistic question.

By the way the bottom of your car is too clean...you need to drive it more.
 

·
19,585 Miles
Joined
·
11,141 Posts
Discussion Starter · #3 ·
RPM's

The 3.46 is roughly a 10% change which will increase the RPM's for any given speed. 70 MPH in fifth gear is 3250 RPM. Based on the S54 engines 7600 redline the 3.46 is a better match for the car.
With a six speed, sixth gear is an overdrive (.83) so it would make the 3.46 differential look like a 2.87 one.
 

·
Registered
Joined
·
595 Posts
ron, you're a badass and stubborn, too!

lesser men would have sensed fate against them after the first 3 damaged differentials!

please keep updates coming along.

i'll bet the car just flatout rips with the new rear-end.

cheers,
coky nguyen
 

·
19,585 Miles
Joined
·
11,141 Posts
Discussion Starter · #8 · (Edited)
Install finished

Finished the install Monday night. Have around 200 miles on it since then. No droning, vibrations or leaks. Still breaking it in.

From BMW:



I'm not sure what the above statement means. I assume 2/3 of maximum speed in 5th gear, which is not a problem over here.

Mileage appears better (based on the OBC) with the new differential cruising at 71 MPH. Also measuring differential temperatures etc. Will try to post part two of the how to Sunday night. Have a lot of pics to edit etc. Going to Limerock tomorrow so it won't happen then.
We have planned testing of Al's car with Shawn Fogg sometime soon.
As mentioned, the 3.46 is a nice match for the car.

Preview:
Coming out. Since the differential weighs 83 pounds, having a nice jack makes this much easier to do.

 

·
19,585 Miles
Joined
·
11,141 Posts
Discussion Starter · #10 · (Edited)
Speed sensor apart

As mentioned, we modified the stock M sensor to go in more so that the smaller diameter speed spider in the 3.46 differential would mate with it.
The covers and sensors are different between the M and the 3.46 differential. If you were to keep the 3.46 cover using the 3.46 sensor everything would work fine. If you want to use the M cover with the 3.46 differential you have to change out the speed spider or modify the sensor.
I modified the sensor and it appeared to work fine initially. The sensor Al modified when he did his car is still working fine.
Depending on the drive time the speedometer signal becomes intermittent. This has been driving me nuts for the last couple of weeks. Appears to be temperature related. Thought it might be wiring based on me not disconnecting the plug when letting the differential down. Thought it also may be related to the many mods I've done using the speed signal.
Have a 3.46 sensor coming tomorrow so that I can compare things dimension wise. I have both the M and 3.46 covers.
If you place a scope across the speed sensor you will see a 5 volt square wave referenced to ground when working. When its not working the square wave goes away and you see a line on the scope at ground.
Wasn't sure how the sensor worked. Took one apart tonight to see.



In one prong of the sensor is a switch. In the other a magnet.



The magnet closes the switch (ground). When the magnet is interrupted by the speed spider blade, the blade shunts the magnetic field and the switch opens (5 volts).



I have a thermocouple attached to the differential case to measure temperature, which goes to a meter in the passenger seat.
I also have a lead paralled with the sensor going to a scope in the passenger seat footwell.
Knowing how the sensor works, I can now place an ohmeter across the sensor leads when it fails and hopefully figure this out.
Come this Winter, I will be installing the M spider in the 3.46 differential.
 

·
Registered
Joined
·
48 Posts
Do 99/00 Euro cars have different ratios?

Ron,

I am going to go for a drive later to day and check out the RPM/Speed ratio, as I believe that my car revs higher at the speeds posted. I am not 100% sure about this, but I will "have to" :D to out now for a drive! Will post the results. Give the incredible traffic here, its bound to be a challenge to do so...

Gus
 

·
Registered
Joined
·
9 Posts
Hay Gus!

If you'd like me to check my RPM at a certain indicated speed, let me know what numbers you want checked. Just not anything over 5k in fifth!!! I have the 3.15 rear end in my 01 M///Roadster.

Have a great weekend, looking forward to seeing you a couple weeks! We'll see what we can do about adjusting your hood. I've been practicing "mi Espanol cada dia": "Tiene una cervezas muy fria, tenemos sed!!!" ;) ;)
 

·
Registered
Joined
·
48 Posts
Hola Steve

Keep practicing that Spanish! :thumb:

I am the one who is going to verify for Ron the RPM to speed ration on my car and see if its a 3.15 or what that I have in there.

I promise not to go over 6K rpms in 5th :angel: Maybe not?:lmao:

Z ya soon!

Gus
 

·
19,585 Miles
Joined
·
11,141 Posts
Discussion Starter · #14 ·
Speed sensor problem solved

Reference the picture below.
The Z3-3.0 speed sensor arrived today. Compared it dimensionally to the M-Z3 one.
From taking the M one apart last night, I now know how they work. Based on this I checked the trigger point on the
Z3-3.0 one. Although the 3.0 one is shorter, its trigger point is further out towards the tip of the prong than the M one. Bingo.
Looking at the pic the stock M one trigger point is .074" away from the end of the spider blade tip. By modifying the sensor to go in around .1" the tip just goes over the trigger point. On Al's car this has not been a problem, yet. On my car it works fine initially and stops working after the differential heats up.
On the 3.0 unmodified one the blade tip is past the trigger point by .328". More then enough margin.
Keep in mind that why this is necessary is that the speed spiders are a different diameter between the 3.46-3.0 and 3.15/3.23 M differentials, The 3.0 cover (where the sensor mounts) is thinner than the M one.
So now, instead of swapping spiders or modifying a sensor, you just have to replace the M sensor with the 3.0 one.
Finally!

 

·
Registered
Joined
·
5 Posts
THanks Ron...the diff SINGS!

We did it. Last week we installed a 3.46 diff (from a 2000 Z3 auto)in a 2001 M coupe. Thanks to Ron Stygar who did the R&D and showed us the way. The car drives as it was ment do drive. What was BMW thinking putting a 3.15 in that car.
We got the diff from Pasific BMW for $1750. Install is a straight bolt on. We even used the Mcoupe diff cover so everything looks stock. We did have to swap out the Mcoupe speed sensor with a 2000 Z3 3.0 speed sensor (as Ron recommended) to make everything work correctly.
This weekend the car was driven 1000 miles (600 miles per BMW recs) as part of the break in period for the diff.
Now it just ROCKS!
THANKS Ron you are the GURU! http://community.webshots.com/photo/32881575/32881676bvwevy
 

·
Investigative Tinkerer
Joined
·
5,011 Posts
I hope to get my car out to the strip for some timed runs sometime this summer.
I got out late last fall and was only able to make two passes on a very busy "test-n-tune" night.
The car only had around 3000 miles on it, and I've since learned that it isn't a "fast track" as far as times go.
My best that night was a 13.8 at maybe 104mph with the 3.15:1.
The plan is to make it over to Norwalk (?) where they have some kind of high traction surface. That and a 3.73:1 should make a world of difference. Plus mounting the Hoosiers on the rear only.
 

Attachments

·
Equine 1/4 miler ancestor
Joined
·
1,687 Posts
Re: Re: curious is this 3.46 a torsen type?

Ron Stygar said:
Torsen style. Torsen style only used in the non M-Z3's based on my research
Ron, to avoid confusion for all of us that are contemplating this now or in the future, what is the full part number on the Torsen 3.46 differential?

Thanks for your R&D efforts.:thumbup:
 
1 - 20 of 83 Posts
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top