Drivetrain Clunking...I'm going to need some help.

[thrustnut]

I am not an expert on this, but I did recently get rid of my rear end clunk which sounds to me like the same clunk you have. I had previously replaced the transmission mount, the tranny cross member mounts (with poly) the rear diff mount, and the mustache bar mounts (also with poly)with no success. I recently replaced my 3.54 with a 3.90 from a member here as a last resort. I didn't notice anything abnormal when removing my diff (nothing loose, missing, ect.) that would have caused the clunk. When I had the two diffs sitting on the bench next to each other I noticed that my old diff had just a tiny bit of play if I held the output shafts and spun the input shaft. The 3.90 I replaced it with had NO play at all when doing the same thing. I pulled the covers and again, nothing obviously wrong with the old diff (broken gears, metal shavings). The fluid level was low, but clean... can't say if that had anything to do with the clunk when installed in the car, but with both diffs empty only my old one had the play. The only other things I did when the rear end was out was clean up my half shafts, check play on the half shaft U-joints (they where good) and lube them. The last thing I did which was unrelated was replace my LCA bushings and pins. I personally believe now after this that ANY play between the input and output flanges is too much, but again I'm not an expert, just my opinion after some observations and fixing the problem.

[jacob80]

Yeah, I'm thinking the same thing. I don't like that I have any play in my drivetrain, especially considering the fact that if I'm on the slack side and drop the clutch harshly, I may damage my rear end pinion or ring gear. Hopefully someone has some advice for me as far as this play. Thanks again!

[JMortensen]

I am not an expert on this, but I did recently get rid of my rear end clunk which sounds to me like the same clunk you have. I had previously replaced the transmission mount, the tranny cross member mounts (with poly) the rear diff mount, and the mustache bar mounts (also with poly)with no success. I recently replaced my 3.54 with a 3.90 from a member here as a last resort. I didn't notice anything abnormal when removing my diff (nothing loose, missing, ect.) that would have caused the clunk. When I had the two diffs sitting on the bench next to each other I noticed that my old diff had just a tiny bit of play if I held the output shafts and spun the input shaft. The 3.90 I replaced it with had NO play at all when doing the same thing. I pulled the covers and again, nothing obviously wrong with the old diff (broken gears, metal shavings). The fluid level was low, but clean... can't say if that had anything to do with the clunk when installed in the car, but with both diffs empty only my old one had the play. The only other things I did when the rear end was out was clean up my half shafts, check play on the half shaft U-joints (they where good) and lube them. The last thing I did which was unrelated was replace my LCA bushings and pins. I personally believe now after this that ANY play between the input and output flanges is too much, but again I'm not an expert, just my opinion after some observations and fixing the problem.

This is bad procedure. If you want to check backlash, you shouldn't hold the output shafts and try to spin the input shaft. When you check backlash you should use a dial indicator on the ring gear, set as perpendicular to the tooth face as you can get, and then rock the ring back and forth. It is VERY easy to measure incorrectly by moving the pinion gear, so do it several times to make sure you have an accurate, repeatable measurement.

 

When you grab the output shafts you are introducing the slop in the spider gears into your measurement. Spider gears are designed to have slop in them, so saying you found slop there is saying you found that everything is normal. I can't tell what you're measuring in the video, it looks like you've jacked up the rear end (or one side?) and are moving the wheels back and forth. Either way, this is feeling the slop in the spider gears, which is intended to be there. While it is possible that there may be excessively worn spider gears, I wouldn't try to measure it in this way, I would disassemble the diff and look at the gears and the crosspin to see if you have wear.

Edited May 18, 2010 by JMortensen

[thrustnut]

Like I said, I'm not an expert. I was giving him the information I obtained from having the same problem and finding a solution. I would think that since I only changed the rear end, DID disassemble the old one and couldn't find any hard evidence of broken parts, that the play that was evident in the old rear end and non-existent in the new was the culprit. There's not that much going on back there, and most rear ends are pretty cheep, in most cases cheep enough that it's not worth rebuilding. BTW I have a junk 3.54 if any one wants it.

[JMortensen]

Honestly I would throw your results out the window because your testing procedure was so poor. If you want to ship me your diff I'll look at it and tell you if it has excessive wear, but seriously what you did had zero merit as a test for differential wear. Not trying to be a jerk, just trying to inform you and others not to do what you did, because it really isn't a good test of anything at all. It's like checking a sore throat with a colonoscopy and concluding that there is a problem because it smells like @&%$ in there. By messing with the output shafts you were literally looking at the wrong end.

[thrustnut]

I was messing with the input shaft trying to see if it would turn back and forth with the output shafts fixed (like it was sitting on pavement) which it did and the diff that's in the car now did not. What do you think fixed MY problem if all I did was replace the rear end and the clunk went away? Like I said nothing was loose, everything was the same going in as it was coming out. Not trying to be a jerk, but it seems pretty obvious that if I replaced one part and the problems fixed that the part replaced was the problem.

 

I'm a mechanic also, I was looking for the problem, not to replace the diff. I was hopeing to find something else under there that could have attributed to the problem, but there was nothing there. Sorry my test was lame, but not a whole lot of diffs on aircraft and with no problems with anything on the car and input shaft movement on one diff not evident on another, I thought it was a pretty safe bet (which it was). Thanks for the offer on checking my rear end, but I'm sure I can do the same test. You should offer this service to Jacob80 to help find his problem.

 

Anyway Jacob80, good luck with your diff problems, obviously my results where not valid... I will wait for my clunk to return...

Edited May 19, 2010 by thrustnut

[jacob80]

Well I really don't know what to do at this point. I plan on draining the oil out of my rear end and checking for metal shaving but if there aren't any shavings, what is my next step? I was planning on driving from Nebraska to Branson to the Z fest with this setup. I have yet to replace my trannyy insulator, so that's another item on my list to be replaced.

[JMortensen]

I was messing with the input shaft trying to see if it would turn back and forth with the output shafts fixed (like it was sitting on pavement) which it did and the diff that's in the car now did not. What do you think fixed MY problem if all I did was replace the rear end and the clunk went away? Like I said nothing was loose, everything was the same going in as it was coming out. Not trying to be a jerk, but it seems pretty obvious that if I replaced one part and the problems fixed that the part replaced was the problem.

Could have been any number of bolts that got tightened with the installation of the new diff, from the driveshaft to the rear cover. Could be that the spider gears did have a problem. Could have been backlash. I can't tell you what it was. I can only tell you that your test didn't tell you what it was either.

 

I'm not saying you didn't fix the problem, it certainly sounds like you did. I'm saying you didn't diagnose the problem with the test that you did and that this test shouldn't be used by other people because it is meaningless.

Edited May 19, 2010 by JMortensen

[Lazeum]

I had some clunk here and there. I found out the bolts on mustache bar mounts came loose. It happens twice whereas I always torque them to the spec.

It happened once also the 2 diff nuts holding the diff to the mustache bar came out loose (I've changed them them with new ones since then).

Regarding my rear end setup: R200, new front mount, poly bushing on stock mustache bar, rebuild half-shaft with new u-joints.

 

this weekend, it happens again but was more severe and nothing was loose this time.

So I've lifted up the rear end of the car to try to reproduce the clunk manually. It appears left wheel has more lash than right one. I have yet to check half-shaft (it should occur this weekend)

So I'm going to check untightened half-shaft bolts, u-joints, stub axles. I don't know if this message could be of any help but this is how I'm trying to resolve this issue.

Edited May 20, 2010 by Lazeum

[JMortensen]

So I've lifted up the rear end of the car to try to reproduce the clunk manually. It appears left wheel has more lash than right one. I have yet to check half-shaft (it should occur this weekend)

How is is possible that one tire could have more lash than the other? You're not actually measuring "lash" at the tire, you're measuring all of the lash between the left tire and the right tire. To restate: when one tire is locked and you turn the opposite, you're turning the wheel, which turns the stub axle, which turns the companion flange, which turns the halfshaft, which turns the side gear, which turns the pinion gears, which turns the other side gear, which turns the other halfshaft, which turns the companion flange, which turns the stub axle, which turns the wheel.

 

Whatever results you get on one side should be exactly the same as the other because you're moving exactly the same parts in both cases. And what you're testing (again) is the cumulative slop in ALL of the parts, and doesn't tell you if the slop is in the stub axle, the halfshaft U joints, the splines on the side gears, or the gears themselves.

 

Maybe you have a badly adjusted drum brake or a stuck caliper if you have rear disk, but again, this test doesn't tell you anything useful, so I wouldn't even do it. You really need to test each part individually to find out where the slop is, and if it is in the spider gears, well, there is supposed to be slop in the spider gears. And if you have a limited slip like the OBX or Quaife, there might be a lot more slop in the gears and it's still totally normal.

 

If you set the car down and roll it back and forth in gear by hand and hear a clunk, THEN you could say that you've found a clunk and isolated the normal slop in the spider gears (provided you roll the car straight back and forth). You might even be able to pin it down to one component with a stethoscope. But feeling for slop in the spider gears isn't proving anything useful about the condition of the diff because the spider gears are supposed to have slop in them.

[JMortensen]

Thinking a bit more, even if you roll the car back and forth, that doesn't necessarily mean that the spiders are isolated. If you have a situation where the cross pin shaft is ovaling out the hole in the carrier (as seems to happen with drag raced cars with open diffs a lot), then you could have a clunk from this test and it could be the shaft rocking back and forth in the carrier.

 

So I'll have to revert to my original statement. If you want to know which component is causing the clunk, you really need to test them individually.

[Lazeum]

Jon, that's exactly what I believe it is something on one side of the rear end. My comment was to say the clunk could be something not related with diff. I'm trying to open my mind and think "out of the box".

 

I do hear the clunk when pushing my car back and forth in my garage (with the gearbox engaged).

 

the force I apply on the wheels when up in the iar is quite small, not even enough to make the opposite wheel moving. I'm not short (yet!!!) on ideas. I just hopes I did not destroy the stub axle with my short reduced modified half-shaft. Remember my mod? the shorter half shaft could have popped the axle out from the diff and get the splines not fully engaged during operation.... very unlikely though since it should still be close to bottom out anyway at jounce.

 

 

[JMortensen]

The R200 stub shafts that plug into the diff are retained with a circlip. You could just pry the driver's side out with a prybar, if it snaps out, then it wasn't pulled out. That would be a test to see if it had pulled out. Jiggling the wheel around trying to feel play isn't finding the problem though. It's wasted motion.

[Lazeum]

It was a 20 sec test anyway, worth a try IMO 

 

the circlip on the stub axle were there, I'm 100% sure (hopefully it is not that, I'm on the wrong side of the ocean to find a replacement if needed   )

 

 

[thrustnut]

JM, You think with your wealth of information you could help solve the problem instead of shooting down everything everyone else try's. Why bother replying if you don't even try to help.

[JMortensen]

I have tried to help. Diagnosing a noise over the internet is pretty tough, but I've suggested that they examine each part individually, given ideas of where the clunks may be coming from, advised them to disassemble the diff to check for spider gear wear and described how to measure backlash correctly, told Matt how to check if the halfshafts have popped out of the diff, and told them when what they were doing didn't make any sense. I think that's all very helpful. If someone is trying to diagnose a clunk with a divining rod or a Ouija board, the most helpful thing I think you can do is to tell them that their test isn't going to help. Standing there watching and patting them on the back and saying "Good job" when you know better is not helpful.

[woldson]

If someone is trying to diagnose a clunk with a divining rod or a Ouija board, the most helpful thing I think you can do is to tell them that their test isn't going to help.

 

 

Ohh so going to have to put that in a sig, funny stuff J.

[HICKL]

JM, I have a spare dif on my work bench and want to check it as you recommend. Doing so, what is an acceptable amount of play or backlash?

 

Thanks

Hickl

[JMortensen]

I couldn't find the appropriate section of the FSM. Used to have it, but I had a hard drive failure a while back and lost a lot of stuff. I did download a 240SX FSM from www.carfiche.com and it shows a shortnose, so I wouldn't use that spec. I want to say it was from .005" to .0075" or something like that, but I'm sure someone has the manual and can get you the correct number.

 

Carfiche.com rotates which FSM's are available at any given point, so you can always check back later and get the 280Z or 280ZX manual and get the correct spec that way.

[cygnusx1]

For an R200, backlash should be between 0.0051 and 0.0071 inches.  Check with a dial indicator tangent to the ring gear.