Diff is overheating after rebuild
Discussion
I fitted an alternate ratio ring and pinion into an LSD on my Vit. I set the preload, used the official Suzuki diff rebuild kit, set the backlash etc. But after a few miles the diff was very hot. I took it apart, rechecked the backlash... all looks good.
It does feel like the preload is very high on the bearings but I double checked it was within specifications.
The only thing I can think off is that I have preloaded used bearings using the factory manual specifications, which are for new bearings. Maybe that's the problem. I've bought some new bearings now and I'll try again.
I spoke with a few diff builders and none of them mentioned the need for fitting new bearings, even though the factory manual specifies it. Maybe if using the original bearings I should have set the preload to what it was at when the bearing was removed. I remember one specialist saying similarly about the backlash... don't use the factory specification, if a second hand R+P is used, then set it to what it had bedded itself in at when it was removed.
It does feel like the preload is very high on the bearings but I double checked it was within specifications.
The only thing I can think off is that I have preloaded used bearings using the factory manual specifications, which are for new bearings. Maybe that's the problem. I've bought some new bearings now and I'll try again.
I spoke with a few diff builders and none of them mentioned the need for fitting new bearings, even though the factory manual specifies it. Maybe if using the original bearings I should have set the preload to what it was at when the bearing was removed. I remember one specialist saying similarly about the backlash... don't use the factory specification, if a second hand R+P is used, then set it to what it had bedded itself in at when it was removed.
Equally important as lash etc is pinion depth, although it's much harder to set.
Might be worth taking it apart and getting some gear marker, and actually checking toot contact patterns
If something is wrong it wont take long to do damage.
Bearing pre-loads...you can usually get away with a fair bit there, wouldnt be overly concerned on that front.
Might be worth taking it apart and getting some gear marker, and actually checking toot contact patterns
If something is wrong it wont take long to do damage.
Bearing pre-loads...you can usually get away with a fair bit there, wouldnt be overly concerned on that front.
IMO bearings are unlikely to be the problem.
These will impart so little heat into the oil. Tight gears meshing would probably be worst.
Loose...not really sure.
Stupid/obvious question...
You changed the rear diff ratio.
I assumed you also changed front to match ?
Diff's front and rear would need to match otherwise you'd be winding both them and the centre diff up like mad
These will impart so little heat into the oil. Tight gears meshing would probably be worst.
Loose...not really sure.
Stupid/obvious question...
You changed the rear diff ratio.
I assumed you also changed front to match ?
Diff's front and rear would need to match otherwise you'd be winding both them and the centre diff up like mad
Steve, if you're sure it's not the bearings (though I would have thought overtight bearings would heat up), then the only other explanation is the the Ring and Pinion (or CWP as you say it's called) are at fault. There's lots of reasons to suspect them:
1. They came from a dodgy factory that most Suzuki owners avoided (but the only one that still does low ratio R+Ps)
2. Every reseller that used to import them into the UK has now stopped selling them
3. After speaking with the factory about the fault the only thing they said was to loosen the backlash.
There was also a suspicious amount of metal in the diff oil after the test run which was approx 3 miles.
I'm pretty sure that the backlash, mesh and preload was set correctly because I had 8 original factory diffs lying around, and I measured them all as a test. All of them had backlash according to factory specs. The preload was mostly below, except for the two very low mileage diffs.
1. They came from a dodgy factory that most Suzuki owners avoided (but the only one that still does low ratio R+Ps)
2. Every reseller that used to import them into the UK has now stopped selling them
3. After speaking with the factory about the fault the only thing they said was to loosen the backlash.
There was also a suspicious amount of metal in the diff oil after the test run which was approx 3 miles.
I'm pretty sure that the backlash, mesh and preload was set correctly because I had 8 original factory diffs lying around, and I measured them all as a test. All of them had backlash according to factory specs. The preload was mostly below, except for the two very low mileage diffs.
Tight bearings on the pinion or side bearings will heat things up, as it'll effectively try and burn the bearings out. But it should be easy to identify by hand because the diff would be difficult to turn. How did you determine the pre load was correct? Do you have a rolling torque gauge ?
An incorrectly seated bearing race within the diff housing can easily increase the loading on the bearings, so make sure the races are fully "home" ( if using a hammer and drift to seat them the sound will alter when the race has bottomed out)
On the re-strip were there any visible markings on the bearing rollers and races, and did it run long enough to show a pattern on the crown wheel teeth which would show the depth of mesh? The polished patch on the teeth can move on 2 axis which is a great indicator for the set up and often a truer indication than a splodge of engineer's blue on the teeth.
An incorrectly seated bearing race within the diff housing can easily increase the loading on the bearings, so make sure the races are fully "home" ( if using a hammer and drift to seat them the sound will alter when the race has bottomed out)
On the re-strip were there any visible markings on the bearing rollers and races, and did it run long enough to show a pattern on the crown wheel teeth which would show the depth of mesh? The polished patch on the teeth can move on 2 axis which is a great indicator for the set up and often a truer indication than a splodge of engineer's blue on the teeth.
There is a shiney area visible on the teeth, and the pattern looks about right, with the contact area in the middle of the teeth on the coast and drive side. The original bearings look good as well, but like I said, I only drove a few miles.
I'm just waiting for a surface plate (needed for setting the pinion depth using the Suzuki diff tool) and breaker bar and I'll be rebuilding it with new bearings. If anyone has any other suggestions for what to look for then I'll be glad to hear them.
I'm just waiting for a surface plate (needed for setting the pinion depth using the Suzuki diff tool) and breaker bar and I'll be rebuilding it with new bearings. If anyone has any other suggestions for what to look for then I'll be glad to hear them.
Forgot to answer how I measured the bearing pre-load. I used a spring scale.
The procedure in the factory manual is to set the preload on the pinion with the diff removed. Then add the diff, and set the preload on the diff. At this point you can only measure the total preload of the diff and pinion, but that's fine as you can just deduct the pinion preload to know the diff preload.
The procedure in the factory manual is to set the preload on the pinion with the diff removed. Then add the diff, and set the preload on the diff. At this point you can only measure the total preload of the diff and pinion, but that's fine as you can just deduct the pinion preload to know the diff preload.
I managed to speak with two professional diff builders and got the following info:
One says its normal for a diff to get very hot during break in.
The other says that preloads on used bearings should be lower than on new ones, and also says that diffs can run upto 120 C.
I've got the diff apart now, just waiting for new surface plate to arrive so I can set new lower preloads and put it back together. Surface plate is needed to zero the Suzuki diff rebuild kit.
One says its normal for a diff to get very hot during break in.
The other says that preloads on used bearings should be lower than on new ones, and also says that diffs can run upto 120 C.
I've got the diff apart now, just waiting for new surface plate to arrive so I can set new lower preloads and put it back together. Surface plate is needed to zero the Suzuki diff rebuild kit.
May have stumbled upon another possible reason.
When I measured the preload, the dial would jump up in order to get the pinion turning (due to stiction?) and then would come down to a lower torque to keep the pinion turning thereafter. I was using the second, lower reading as my preload setting.
Which ones should I be using?
When I measured the preload, the dial would jump up in order to get the pinion turning (due to stiction?) and then would come down to a lower torque to keep the pinion turning thereafter. I was using the second, lower reading as my preload setting.
Which ones should I be using?
Yes, new builds will run hotter but not usually drastically hotter. To hit 120 degrees under anything other than high speed continuous or performance driving would be difficult. 70-80 after a spell on a dual carriage way is more likely.
Yes, preload on old bearings is a couple of Nm lower as they've already bedded in. If building a diff I'd always use new bearings anyway. The break away reading is always much higher than the "rolling" or continuous effort, if you can get hold of a rotational torque meter you'll have better results than using a spring scale.
Plain ceramic tiles and even normal mirror glass is usually good enough to use in place of a surface plate too (just don't drop the bits on them )
Yes, preload on old bearings is a couple of Nm lower as they've already bedded in. If building a diff I'd always use new bearings anyway. The break away reading is always much higher than the "rolling" or continuous effort, if you can get hold of a rotational torque meter you'll have better results than using a spring scale.
Plain ceramic tiles and even normal mirror glass is usually good enough to use in place of a surface plate too (just don't drop the bits on them )
A dial gauge (dti) isn't the same as a rolling torque meter.....just incase there's any confusion. The dti would be used for measuring the pinion's height and the crown wheels back lash.
A little tip too, I've always found it beneficial to give the diff casing a whack or two with a hammer in the area of the bearing races after they've been initially set to settle them down while rotating the diff flange. It also seems to slacken off the brgs if they've gone a touch tight during set up.
A little tip too, I've always found it beneficial to give the diff casing a whack or two with a hammer in the area of the bearing races after they've been initially set to settle them down while rotating the diff flange. It also seems to slacken off the brgs if they've gone a touch tight during set up.
Gassing Station | Engines & Drivetrain | Top of Page | What's New | My Stuff