Supra Forums banner

It is official. T1 Torsen diff's are weak.

12K views 80 replies 23 participants last post by  aaron300  
#1 ·
Took the chance running a JDM Torsen diff for a few months, since they give much better traction than a clutch type LSD, but it is now toast. It went out in a strange way too. It developed a lot of slack over the course of a day, causing lots of driveline bouncing and surging. I had the car up on jackstands checking for loose tranny mounts or a broken subframe. With the engine revved and one quick pop of the clutch with the wheels in the air broke it, confirming my suspision.

With daily RWHP around 350-370, this is pretty weak. Never did any burnouts.

Before I bolt my stock LSD back in, should I beef up the clutch packs? I'm sick of swapping this shit in and out!
 
#4 ·
ummm not for anything but drjonez posted the crappyness of the jdm torsen a while back.

basically shovel + bucket to pick up what was left.
 
#6 ·
Kaaz makes a better clutch based LSD for our cars (the TRD on is a rebadged Kaaz unit)..

if you can get search to work it's on the forums somewhere...
 
#7 ·
My stock diff worked well, and seemed to hold 350-400 rwhp. I'm just not sure how long it will last as it has 150,000 miles on it. I do plan on hitting 500+ hp down the road, so I wish I had the extra $ to slap in a TRD or Kaaz. How streetable are these diffs? Anyone got one?
 
#10 ·
funny, I have the a TT-R and haven't had any trouble from my torsen diff yet and I have been far from easy on it. Burnouts, Drag Launching, and Sliding around and its still fine. The trick is to keep both wheels on the ground while power is being applied to them otherwise they break like no ones business.
 
#11 ·
Derek1224 said:
funny, I have the a TT-R and haven't had any trouble from my torsen diff yet and I have been far from easy on it. Burnouts, Drag Launching, and Sliding around and its still fine. The trick is to keep both wheels on the ground while power is being applied to them otherwise they break like no ones business.
perhaps you're lucky?
 
#14 ·
drjonez said:
perhaps you're lucky?
possibly, the car has 89,xxx kms on it right now and I don't think it was driven hard at all while it was in japan before being shipped over here. At any rate if the torsen goes out I'm going with a Kaaz unit as those are amazing upgrades for stock.
 
#15 · (Edited)
I thought I was lucky too, after it worked great drag racing, autocrossing, and hard launching all the time. It just started failing at random after 6 months. I looked inside it too before I installed it. It was a tight, mint, JDM unit with around 50,000 miles. I'm assuming that I stripped off some teeth on the helical gears since I could still drive on it but it had HUGE slack.
 
#16 ·
drjonez said:
i'm pretty sure it was started by a fracture. that's the problem w/the thhe T1, no tolerance for shock loading.
Wow. That is pretty wild. I guess I have been lucky with mine so far. If it was a fatigure failure you should be able to see a noticable difference in the fracture surface with a fine lined structure in the fatigue region and a coarse 'ripped apart' region in the area when it failed completely through the remaining cross section in an abrupt fashion. If the fracture surface is not too screwed up you might even be able to find an 'initiation site'... possibly a bad spot or material defect. If it was a tensile overload (failed all at once) the surface would look about the same everywhere. Likely spot would be on the OD of the housing just above the ring gear seating surface.

I had never really thought about this before, but it looks like the T1 housing is a casting with features machined into it for the helical diff gears. Castings can have small defects randomly dispersed throughout the part. Somebody mentioned being 'lucky', and that may very well be the case. Higher loads and a poorly placed material defect is probably all it takes to initiate failure. If you get a 'good' casting you may never run into trouble.

I've been running mine at 330 RWHP for quite some time now. I guess the real test will be when I finally get my motor project finished and up it to around 450.

I guess I will be keeping the 4.56 open diff around just in case... That's great... Just what I need... another reason to have more 7M parts laying around my house.
 
#18 ·
aaron300 said:
Wow. That is pretty wild. I guess I have been lucky with mine so far. If it was a fatigure failure you should be able to see a noticable difference in the fracture surface with a fine lined structure in the fatigue region and a coarse 'ripped apart' region in the area when it failed completely through the remaining cross section in an abrupt fashion. If the fracture surface is not too screwed up you might even be able to find an 'initiation site'... possibly a bad spot or material defect. If it was a tensile overload (failed all at once) the surface would look about the same everywhere. Likely spot would be on the OD of the housing just above the ring gear seating surface.
from what I can see it is a completely brittle break (tensile overload). there are the distinctive river marks, indicating the direction the break propagated. It is unlikely you'd get a good idea of the initiation site unless you threw it under an SEM or at least a good stereomicroscope. Either way, the material used was too brittle. If the metal was the least bit ductile, you would have distinct ductile tearing around the initiation site with a chance of some fatigue damage, not a complete brittle rupture.
 
#19 ·
maybe the heat treating is wrong? also, a lot of people don't realize you have to break in a rear-end properly with heat cycles, that could cause issues in the metal as well...

Would be nice to figure out a way to make them better instead of just breaking them and then getting another one...
 
#21 ·
I think they just couldn't make it any bigger and have it fit the A70 diff case without completely redesigning the case, like they did with the MKIV which are WAY stronger. If you look at a TruTrac diff you can see there are 4 pairs of worm gears compared to the T1's 3, plus the overall beefening of the unit. I think a custom TruTrac install is the only real way to get a reliable Torsen into a Mk3. Argh!!! $$$$$
 
#22 ·
Poodles said:
maybe the heat treating is wrong? also, a lot of people don't realize you have to break in a rear-end properly with heat cycles, that could cause issues in the metal as well...

Would be nice to figure out a way to make them better instead of just breaking them and then getting another one...
how hot does a differential get? Unless the rear end is getting close to red hot for an extended period of time (think several hours) nothing will happen. "Break-in" does not constitute in-situ heat treatment.
 
#23 ·
tekdeus said:
I think they just couldn't make it any bigger and have it fit the A70 diff case without completely redesigning the case, like they did with the MKIV which are WAY stronger. If you look at a TruTrac diff you can see there are 4 pairs of worm gears compared to the T1's 3, plus the overall beefening of the unit. I think a custom TruTrac install is the only real way to get a reliable Torsen into a Mk3. Argh!!! $$$$$
yes and no. the MKIV T-2 isn't that much bigger @ all, i have pix somewhere.

and yeah, the only way we're getting a (good) helical diff is a custom app....<sigh>...
 
#24 ·
drjonez said:
yes and no. the MKIV T-2 isn't that much bigger @ all, i have pix somewhere.

and yeah, the only way we're getting a (good) helical diff is a custom app....<sigh>...
I will pay to have the busted pieces of your torsen sent down here if you still have them and have no use for them. I would like to take a look at it and try and figure out how/where it failed. I might even be able to take it into work and have my MR2 buddy that works in the materials lab take a look at it on the SEM if it fits. It looks like the problem might be with the housing configuration. If that is the case there might be a way help hold it together a little better.

What ratios have the guys that are busting these things running? I swapped in my 3.73 gears, so I'm probably running at 10% less torque than the 4.1's.

I guess I have been to busy working on my CT12 upgrade... I totally missed the OBX debacle LOL. Any time gears are involved somebody should hunt me down if I'm not paying attention. I like to try and get some use out of my MS from Ohio State whenever I get a chance... gears don't come up too much on jet engines.
 
#25 ·
OHIO STATE??!!??? nooooooooooooooooooooooooo!

j/k. ;)

yea, np on sending you thhe trashed stuff.......only thing is i'd like it back (i keep a collection of fragged parts...yea, i'm a nerd). how about we each pay shipping 1 way? it shouldn't be too much since you're just in cincy. i'm curious about the results as well.....

the torsen i killed was a factory 3.91.