Differential Equations

[geordi]

I am digging this thread. Nice work & thanks CEK

[tig]

I am digging this thread. Nice work & thanks CEK

Me too! Thanks.

Spent some time in-between tearing apart the parts car on prep. Cleaned up bare metal and the races are ready to be put in the freezer tonight.



I can't really do anything more but put the races in until the new seals arrive from Blunt sometime this week...

[Jelmer538i]

Are you going to rebuild the limited slip part to? That's really easy now.

[tig]

Are you going to rebuild the limited slip part to? That's really easy now.

Nah, I really don't think it needs it. I could be wrong.

[Jelmer538i]

http://thayermotorsports.com/products/b ... efresh-kit
For this kind of money I would refresh it...

[tn535i]

Those LSD discs at Thayer were very expensive. If the condition looks good I would re-use them.

[Jelmer538i]

$214,99 is not expensive for 2 clutches, 2 dog rings and a bolt set.

I payed $190,- for my 210 diff refresh kit in the UK.

[wkohler]

Hell, I'd add the third clutch. 40% lockup.

[Coldswede]

40% lockup would be fantastic on the Autocross course! It is on my To-Do list.

[foolish]

I'm no diff expert by any measure but after discussing a similar rebuild with Wanganstyle (one of the resident diff experts on R3V) he suggested a stock 2 clutch rebuild for a daily driver as opposed to a dedicated auto x or track car. He has provided a lot of info on diff rebuilding and is worth speaking with.

[tig]

I got my seals and crush bushings from Blunt and got going today. Stymied:



I was able to get the pinon nut off by using a pry bar thorough a hole in the flange, wedged against the case, using the impact hammer.

However, the diameter of the flange holes are such that not even the hardest-steel drift I have is strong enough to withstand the torque required to crush the crush bushing.

I'm now in search of someone who has the special BMW tool designed for this. I have not been able to find any pictures of it, nor a part #. It is apparently called the "Final drive work fixture" but searching for that gave me no love anywhere. I'm going to try to reach out to Wanganstyle on r3v....

Anyone else got ideas?

[bk_856er]

Sounds like you need a simple fixture to hold the pinion flange and prevent it from turning.

I rebuilt the differential in another vehicle a while back. The 4x4 guys have lots of experience with differentials and some have developed good hacks that you might be able to take inspiration from. Peruse gearinstalls.com or similar for ideas - check out the "tools needed" link on that particular site and scroll down for an example of a simple homemade fixture. Zuk is a differential jedi master.

BK

[bk_856er]

From the book. Factory holder is 230020, used in conjunction with a diff bench stand. I'd use some angle iron...

[Mike W.]

I got my seals and crush bushings from Blunt and got going today. Stymied:



I was able to get the pinon nut off by using a pry bar thorough a hole in the flange, wedged against the case, using the impact hammer.

However, the diameter of the flange holes are such that not even the hardest-steel drift I have is strong enough to withstand the torque required to crush the crush bushing.

I'm now in search of someone who has the special BMW tool designed for this. I have not been able to find any pictures of it, nor a part #. It is apparently called the "Final drive work fixture" but searching for that gave me no love anywhere. I'm going to try to reach out to Wanganstyle on r3v....

Anyone else got ideas?

Not a problem. I made one for a challenge something like this, the front hub nut and seal on a M52. You need a piece of flatbar, say 3/16"X2" a couple of feet long. Or 1/8"X4" You've got options. Simply drill for the driveshaft holes, bolt it up tight, attach something as long or longer than your big cheater you're going to be using and away you go. Thick enough and you can use 2 holes, a little thinner and use 3 or 4 with a cutout. This is the easy part, don't let it slow you down.

[tig]

Not a problem. I made one for a challenge something like this, the front hub nut and seal on a M52. You need a piece of flatbar, say 3/16"X2" a couple of feet long. Or 1/8"X4" You've got options. Simply drill for the driveshaft holes, bolt it up tight, attach something as long or longer than your big cheater you're going to be using and away you go. Thick enough and you can use 2 holes, a little thinner and use 3 or 4 with a cutout. This is the easy part, don't let it slow you down.

Thanks Mike. I don't think I could be going any slower. Good thing I don't need to be doing this.

[tig]

Duh.

Why didn't I think of this before?



Nailed it the first time too. I had bought a 2nd crush bushing just in case, but I was just super careful getting to 25in/lbs.



Checked the torque on all the fasteners and marked them (made me feel like a pro).



I do believe that is just about right. Whew. Turns out that the spacers DO go with the case, just as others have indicated. I thought I might have had them reversed too, even though I labeled them right/left...I couldn't remember if I had done left/right looking forward or left/right looking back! (It was forward).



Off to a dinner party... will check it again, and then finish putting it all back together!!!!

[tig]

Done!

That was kinda fun. I'd do it again.

[Mike W.]

Done!

That was kinda fun. I'd do it again.

Wow, looks good! Granted what's important is inside, but it almost looks too good to hide underneath.

[wkohler]

Why did you paint the drive flanges?

[tig]

Why did you paint the drive flanges?

Because I figured they will rust otherwise. I plan on wire-wheeling off the faces; I meant to mask them but forgot.

[tig]

Out with the old...





In with the new...


Not a huge difference according to the tach, but feels much better at highway speeds.

Before:


After:

[leadphut]

interesting...you're revving less, but getting about 8 mpg less. were the driving conditions the same?

[tig]

interesting...you're revving less, but getting about 8 mpg less. were the driving conditions the same?

I was probably on or off the throttle in one or both photos. I was not steady cruising in either.

[tig]

Update: after over 4,000 miles this diff that I rebuilt has NOT esploded. Nor has it started making weird noises or leaking.

I think I've proven that anyone can rebuilt diffs. There's no magic involved. You have to take care and read the various instructions out there, but you too can do it.

Party on!

[tig]

Oops, I'm doing it again...

This time, the patient is a S3.07 type 110 (large case) for Minerva.

The diff has unknown provenance and although it looks mint on the inside I have decided to do a complete rebuild just because is sounds fun.

This time I will rebuild the clutch pack too (again, for funzies).

This is the thread that was so useful to me last time. Alas, all the photos have been photofucketed.

I am going to regenerate those instructions with my new photos here as I make progress...

Hat tip to @dsmith for an amazing packing job. THIS is how you safely and securely ship a diff (or two):





Pretty good illustration of the differences between the large-case (type 110) and medium-case (type 188) differentials and E28 covers:















The plan is to source all the replaceable parts as OE (from Blunt). Kits are available, but OE bearings can still be had, and I don't want to skimp. Here's a typical eBay kit I found:
https://www.ebay.com/itm/BMW-E46-M3-dif ... 4188691315


I will use the Racing Diff's clutch kit and keep it at 25%:


I haven't visited my powder coater in a while, and while I know painting works well, I am inclined to powder coat the case and other bits that get paint. This way I can also be sure the case is really cleaned up well because media blasting.

Follow along and watch me eff this up!

[tig]

I wasn't quite sure how to correctly measure the limited-slip differential clutch break away. I did some searches and watched some youtube videos, but everything I found showed purpose-built equipment. I found few posts that indicated you could just use a torque wrench, but nothing that indicated how you attached a torque wrench to the output flanges etc...

So I just invented and think I figured it out.

First, I made a tool that screws onto an output flange and has a 17mm bolt head welded in the center.





I considered making a second one to hold the opposite side, but then I realized with the amount of torque in question (~50 ft/lbs) I could just wedge something in place. E.g:



To get a baseline, I measured the Type 188 S3.25 medium-case diff that came off of Minerva's parts car. This was a relatively low mileage 535i (~160k miles) and the diff feels very smooth. So I figured chances were the clutch discs were in good shape.

I got just around 25-30 ft/lbs on each side, which is (apparently) borderline for a 25% (factory) setup. I'll rebuild this one if I ever use it...


On the diff that's the subject of this discussion (S3.07 Type 210 large case), I was surprised to find the clutch break away point was well above 30ft/lbs. In fact, when rotating the direction of wheel travel (counter-clockwise) I saw values between 40 and 50 ft/lbs. I don't think the mechanical design is such that break away should be different depending on the rotation direction. Should it?





The fact that the Type 210 has higher values than the Type 188 has me asking questions. FWIW, @DJM1986-5 knows the province of this diff and it's just a very low mileage (km) unit at 144km that was also well cared for.

If found this info on s14.net:


I also found this PDF on the IE site:


Minerva's S54B33 motor has max torque (100 octane unleaded) of 286 ft/lbs at 5900 RPM (257 at 8000 ). I'll never run 100 octane gas, so let's moderate that to 275 ft/lbs. 25% of that would be 68 ft/lbs.

Is my thinking correct?

[Mike W.]

Minerva's S54B33 motor has max torque (100 octane unleaded) of 286 ft/lbs at 5900 RPM (257 at 8000 ). I'll never run 100 octane gas, so let's moderate that to 275 ft/lbs. 25% of that would be 68 ft/lbs.

Is my thinking correct?

There is a logic to that, but... If it's a regular street driven vehicle perhaps the breakaway torque should be related to the load/weight, not the potential HP. Otherwise with a high HP to lb ratio it might act like a locked diff in the rain. I know SLC doesn't get much rain, but in say the Pacific Northwest where as you might know, rain is common, it might get squirrelly in the wet. Dunno, just thinking out loud.

But worst case look at it this way Charlie, if the whole computer/tech thing doesn't work out you could always get a job setting up diffs.

[tig]

But worst case look at it this way Charlie, if the whole computer/tech thing doesn't work out you could always get a job setting up diffs.

Life goals!

I *am* currently unemployed, you know...