Write Up: Building a True(er) Trac

True Trac owners... you may not have noticed it yet, but your diff may be loosing it's ability to transfer torque due to one design flaw. The primary worm gears are wearing agains't the housing ends, slowly wearing a groove into them as a result. As this wear occurs, the center bevel washers drasticly loose their preload and you eventually end up with a normal open diff. :tear:

I have seen a couple with the same problem and intend to "propose" a fix to Eaton. This write up covers this fix for those of you who would like to do this as well. I had picked up a used True Trac a while back with the intention of rebuilding it to test this potential fix. While this may seem like a lot of work to improve on a limited slip, the True Trac and other helical gear LSD's do work great and offer a stiffer housings than compareable clutch type LSD's. Also the center cam bearing and bevel washer design work great, will last a very long time, and can offer tuneable amounts of load transfer. By changing the thickness of the bevel washers or adding shims to the cam bearings you can change the amount of preload. As you'll see later in the write up, I added two .008" stainless shims to increase torque transfer.

Parts List (McMaster Carr):
5909K42 - Bearing Cage (2 needed)
5909K55 - Bearing Washer (4 needed)
98126A433 - .008" Stainless shims (up to you)
6 - 1/2-13x1.25" Grade 8 flange bolts
Loctite 243 (oil resistant)
Some moly lube or grease for reassembly

Tear Down:

I started with disassembling the diff by unbolting one side of the case, and then removing the components making sure to note the correct orientation to help reassembly. Once the inner components are removed, the other side of the case can be unbolted and the remaining worm gears removed. All parts were thrown in a parts washer for a couple hours and then inspected for wear.

worm gears:


one of the two primary gears:


Empty case:


One of the cam bearings that presses against the primary gear


The two bevel washers at the center of the diff (note the orientation):


This is what the assembly looks like inside an assembled diff:


One of the case ends. You can easily see the area where the primary gears is starting to wear into the housing.



A close up of the wear. Notice how much material has been removed simply by the operation of the diff.

The Fix:

I machined the housing ends to allow for a thrust bearing to fit between the case end and primary gear. This allows the diff to function like it should with the friction load being shared between the gear and cam bearing and NOT the housing.

Dimensions machined into the housing:
2.500" in Dia
0.1500" deep

After machining:


5,600lb rated thrust bearing:


cage:


The hardened washers prevent galling the housing






Both washers and bearing installed. The top washer ends up being just below flush with the surface of the housing.


Gear installed:


It works!

Final Assembly:

Reassembly was just as easy as taking it apart. I started with one housing end by first installing the worm gears, primary gear and thrust bearing. I also lubed them up with some moly lube just to be safe. I bolted on the housing end with some new grade 8 flange bolts, loctite and torqued them down to 80ft/lbs.





Once one side is bolted and torqued, the remaining guts can be installed starting with the worm gears and then the cam bearing.



Then I installed a stainless .008" shim:


Next goes in the bevel washers:




Followed by another stainless shim, and remaining cam bearing:




Finally, the last primary gear and thrust bearing can be installed:






Lastly, the housing end is bolted back on and torqued down.

DONE!



It's getting installed within the next couple weeks, and I'll be able to give a ful report on it's performance. Ideally I'd like to test it to find it's slip load value but I'm not exactly sure how accurate or relevant the info would be.

Total Cost: About $26 :cheers:

Enjoi.

Nice! I'm a big fan of the TT. That 26 bucks,.. 'That the cost of the bearings? 'Estimate of shop time? 'Bout how long did it take you to do the machine work?

Nice work and write-up!

Great work

Super repair. In particular, I liked the photo showing the correct installation of the belleville (spring) washers. Bellevilles should always be installed with the small opening touching as you show it in the picture. This gives maximum support for the spring.

BTW, small bellevilles (up to .5" ID) make dandy lock washers... Guess I'll be looking for an el-cheepo, worn out TT. They came in the 29 spline 8.25 didn't they?

Negative, the Truetrac was never an OEM item, you're thinking of the Traclock

I just bought a F/R pair of new truetrac's for my jeep..any idea on what type of mileage these TT's start to show age? I have never run true tracs before..

Nice work you have done there, it makes sence to me..im just currious if its something you would do to a fresh unit also?

very nice write up and great idea. i bought a used TT for the D30 instead of a full locker (using a 242 transfer case) but wasnt planning to take it apart. now you may have changed my mind. at least to inspect for damage, but i may copy your technique for the fix. please give us an update as soon as you can. very interested to see how well it all worked out, and if the added (2) shims gave it enough, or too much slip load.

Brilliant fix dude. My only concern would be that the belleville washers don't get fully compressed as this would increase the load on the thrust bearings exponentially as the case bolts are tightened. I am sure you accounted for that but just a thought.

Thanks guys.

To answer your questions:

The $26 is just the parts cost, but the machine work shouldn't cost much at all. I'd try to find a freind of a freind that will accept a 6 pack No more than 10 minutes on any lathe is all it takes.

It's hard to say when the wear becomes significant, but this one was said to have 30K miles on it when I picked it up. If your buying used, I would definetly crack them open and at least inspect the parts. Just keep in mind that very small amounts of wear will have large effects on preload. It only takes approximately 0.050" to fully compress the two spring washers. Based on the dimensions of the factory washers, the estimated flat load (both washers fully compressed) is a little more than 2,000lbs. Factoring in the scoring on the housings was almost .015" deep (total), the preload was substantially reduced. I can say without a doubt the preload was reduced by over 30%, but more likely closer to 50% due to the exponential loading factor of the springs.

So, to be the Engineer that I am, here is a link to an online calculator whereby you can calculate the spring rate for a belleville.

http://www.tribology-abc.com/calculators/t14_6.htm

Plug in the variables and you will be surprised at how much is lost with the compression travel.

Just a couple of .01s...

Besides, I like data.

Awesome. Ill try to plug the numbers in later tonight. Although im more curious in the actual slip load it will take to to break loose. I need to find out what Auburn does to get their numbers.

I would imagine a jig with one side locked down and a force gauge on the other. Apply torque and find the break away point. Dunno for sure...

I just realized I forgot to mention something important!

If your going to do this with the part numbers above, I'd reccomend to cut .1421" deep instead of .1500" like I did. This will put the bearing flush with the surface of the housing and leave slightly more meat for strength. I added the .008" shims to account for this over machining and give slightly more preload than stock. If you machined .1500" and didn't add the shims, you would have less preload than stock. I just wanted to make that clear. I appologize for any confusion. :dunce:

Is this problem inevitable with the TT?

Given the apparent lack of hardened bearing surfaces, coupled with the typical disdain for changing diff lubs, I should imagine thit is rather inevitable. Good to know they can be resurrected for what amounts to pocket change. Even if you are charged $75.00 by a machinist, an out of pock of around $100.00 is chump change to get a non clutch LSD.

My Wife's 96 Bronco has a factory LSD with the dreaded clutches and it is a pain. We had to rebuild it a couple of years ago. All clutch type LSDs will fail. Properly maintained with friction reducing modifiers, I would think a TT would run for near ever...

As a plus, they can be had new for a tad over $500.00.

Hey Boostwerks, any luck on the calculator? I am kind of curious. You said the travel on the Belleville is 0.050", with 0.015" of wear present that represents a solid 30% loss of preload. Given that a Bellevile is not exactly a linear rate spring, I would guess a loss of around 40%...

For those interested in how Belleviles work anf the history thereof, I am recommending this pdf document http://www.schnorr.com/docs/Handbook.pdf Do not be shocked by the Swastika on the patent document... Yup, another German invention.

Yeah, I would also go out on a limb and say it's inevitable. There might be some unknown super lube out there that would reduce the wear, but I'd be willing to bet that if you cracked open any moderately used true trac you'd see the same.

I havn't done the calculator yet. I'll try to get to it this weekend however. Right now I just need some sleep lol. I'm hoping to get in touch with Eaton here soon.

First off I would like to say very nice job on the machine work and thinking outside the box to fix the problem. Everything looks very well done. However I do not reccomend these modifications.

The case in a Trutrac is heat treated steel and has a very hard surface. While the case being worn does show it has been highly used the gear is still riding on a hard surface designed for wear. By machining the case 0.150" you have machined away the entire hard surface of the case. You now have a bearing wearing on the soft core of the material. Bearings are not designed to wear on soft material.

Along with the wear issue the Trutrac's design works on friction between the gears and case. You have removed nearly all friction between the case and the side geasr by adding a bearing in this application. While you have increased the belleville springs force back to the original design you have eliminated a good portion of some of the overriding forces.

I would also like to note that a good majority of the Trutrac models do not have any belleville washers in the assembly. These units work very well without this feature. The bellevilles are only added to produce some amount of pre-load on the unit.

I have seen Trutracs with 100,000 miles that show almost no where at all. I would say the unit in this thread is not the norm.