bamboomerang

iTrader: (14)

The short: Anyone know how to measure the mainshaft bearing preload without the fancy tool in the FSM? (page 35 in MT) The input shaft and differential are easy with a dial indicator, but I have no idea how to do the mainshaft ??? Thanks in advance.

The long: So I recently did a bearing/seal rebuild on my MT (pics: http://s647.photobucket.com/albums/u...ion%20rebuild/) , and idiotically put the mainshaft bearings on the wrong ends, which was obvious during reassembly. No sweat, a 5 minute fix, but due to the fact that one mainshaft bearing comes off easily, but the other one has nothing to grip onto, I had to break it off (http://i647.photobucket.com/albums/u...g?t=1266025042) , and for the sake of time (HA!) ignorantly reused the original bearing. Now the big mistake happened, I assumed the old bearings (NTN brand) were 'identical' to the new bearings (NSK brand) since they are both 32206 cone bearings, and I installed the old bearing (which had very little wear) into the new race. (NTN bearing in an NSK race) Drove around for 2 weeks, started hearing the infamous bearing whine which initiated the rebuild project so I got suspicious, and then my shifter got REALLY sloppy to the point where I couldnt shift without double clutching. So I dropped the transmission again, and it was showered with metal shavings and the culprit:


That missing part of the bearing 'rollers' is VERY bad. Remember this is after only a few hundred miles of driving and the bearing was just about ready to explode. The race (which you can see is burnt along the edge) got so hot it virtually popped out of its seat in the trans. case to the point where the whole shaft was wobbling around in there. Caught it just in time. I believe it was caused by the mismatch bearing/race, and not preload (since the bearing on the other end of the mainshaft is still perfect), but maybe I'm wrong and it was shimmed wrong. When I put in the new bearing next week, I want the peace of mind of knowing its shimmed properly). As bad as this all seems, I'm really only out 40 bucks + contaminated amsoil (which I already filtered through a coffee strainer) and a few hours work, but it's definitely a lesson learned the hard way: always use the race that came with the bearing. Anyways, any insight on measuring the mainshaft to properly shim it would be helpful.

bamboomerang

iTrader: (14)

So no ones ever changed the shims in their mainshaft during a tranny rebuild?

JtzMax

iTrader: (5)

There are no shims on the MS, only the IPS and diff.

And for what it's worth, and now you already know! - don't mix and match bearings / races.

The IPS shims are not usually a problem, from the factory. I know and understand that new bearings = new spaces / gaps and should be re-shimmed. However, I don't know any possible way to measure the IPS shim need (other than the solder crush meathod) or by popping out the plug (bronze looking thing) on the case cover and using a dial meter to read it.

When I did my rebuild, I didn't adjust the IPS shims at all. Only the diff shims.

bamboomerang

iTrader: (14)

Thanks for the reply. Mine had 2 shims on each of the the input shaft, mainshaft, and differential. The differential was exactly twice as thick as it needed to be, so I only put 1 shim back in (this is factoring in the fact that the preload needs to be about double the amount of freeplay, meaning the shimming will be twice as much as the differential moves/deflects a dial indicator, according to the chart.) and with that 1 shim, the numbers added up perfectly (I had .013 of freeplay and the shim was .026 IIRC) and there was no play in the dif, and the axle had virtually 0 lateral movement when installed.

Page 35 in the fsm has the procedure for checking mainshaft preload, and page 43 has the mainshaft shim chart, weird that yours didnt have any. Maybe because mine is locking diff? I don't know.

I'll probably end up leaving the mainshaft shims the same, but I figured since the preload on the dif was off from the factory, and considering the fact that I'm using new bearings, figured I would follow the FSM recomendations to change it if changing the bearings. I'm bassically paranoid because of my bearing brand/race mismatch failure, I was thinking it might have been caused in part due to bad shimming preload, but now I'm reassured the bad preload was 100% a race/bearing mismatch issue. But if other people are swapping all the bearings and only adjusting the differential preload, I guess I'll follow suit and cross my fingers.

For the input shaft deflection measurements, I popped out the little brass piece like you mentioned. Needs sealant upon reinstallation, though. In all honesty, because the case fits so tight on the input shaft, I couldn't get it to move, atleast not to get an accurate reading, so the shims stayed the same.

JtzMax

iTrader: (5)

Exactly what I was thinking as I read your post.

Nealoc187

iTrader: (23)

There most certainly are shims on the main shaft bearing... every set of bearings in a transmission has to be shimmed. If yours are not, someone built it incorrectly and you're going to have premature failure...

Use the solder crush method. Search my name and those terms if you don't know what that means. I've explained it a couple times. Conical bearings (tapered roller, like the main shaft and differential) get preload, ball bearings (input shaft) get end play... Use the solder crush method and then the chart in the FSM if you need to order new shims.

JtzMax

iTrader: (5)

Neal -

When I did mine, there were only shims on the IPS, none under the cup (outter race) for the MS. And of course, the diff had shims. Don't think it was rebuilt prior to my doing it. Now I'm perplexed.

bamboomerang

iTrader: (14)

Ill try the solder crush method. So if I have this right, the solder will be between the race and the case, I dont press in the race entirely, and then with the mainshaft installed and the case torqued to spec, the shaft will seat the race into where it needs to be? (then of course, measure the crushed part, and visit the FSM table to shim) Seems simple enough.

questions: Q1) Does the natural play in the cone bearings rollers allow for the race to be pressed down sufficiently to get an accurate reading? I know the shaft isn't going anywhere, but it seems like tightening the case to force the race down would put alot of temporary preload stress on the cone bearings. Not a problem? Q2) If I remeasure all the shafts with this method, should I do 1 at a time, or all shafts once, does it make a difference? Q3) And Neal or anyone else who might know, to help put my ignorant-mind to ease, does it make perfect sense that the mismatched brand race/bearing would be 100% the culprit to cause a failure like this if the rollers were already worn slightly and the race was new? Or would you think that was only part of the problem, along with it being mis-shimmed. This occured after about a month, and only a few hundred miles.

Thanks for your help guys, I'm hitting this up on friday after work. It's a *itch getting all the metal shards out of the gears.

Nealoc187

iTrader: (23)

You either lost them or someone f'd up building or rebuilding it prior to you. Every bearing has to be shimmed without question.

Nealoc187

iTrader: (23)

You're correct on the method of doing the solder crush.

Q1) The solder is soft enough that no undue stress is put on the bearings. It just smashes the solder down because the solder is so soft (lead solder)

Q2) Do all 3 sets of bearings at once. Torque the case down to spec and let it sit for like 20 minutes before taking it back apart. Make sure that you've placed the solder such that it does not imprint the etched numbers and letters on the back of the bearing race, the imprints will mess up your readings obviously, and these are pretty precise measurements you're taking here. Use wheel bearing grease or something similar to hold the little strips of solder in place in the bearing bore. Also grease the outside of the bearing race so it slides easier into the bore (both for the solder crush assembly and your actual final assembly).

Q3) Makes perfect sense that mismatching would cause such a quick failure. Bearings wear in to the races and are thus matched for life and either need to be totally replaced with brand new bearing and race, or kept as a pair if you're going to reuse them. Different brand bearings might also have a slightly different taper angle on them, so that's yet another reason you can't match one race with a different bearing.

bamboomerang

iTrader: (14)

Update: FWIW, So I did the solder crush method (using a digital micrometer) and this is how it turned out:
input shaft deflection (inches)
- 0.058/0.062/0.062/0.060
=average deflection of 0.061
Factory provided input shims were 0.028x0.028= 0.056 which just barely makes the 0 - 0.05 endplay tolerance. I reused the shims.

Mainshaft:
0.037/0.037/0.040/0.043/
= average of 0.039 of deflection.
Factory provided shimes were 0.018 and 0.019 = 0.037 of shim. I reused the shims.

So all the shafts in the tranny is shimmed to factory spec, and this is how things went. I installed the new mainshaft bearing, installed the tranny, started it (which produced a weird super quick squeek that was not a belt or a rotor shield, I assume it was the new bearing before getting throoughly lubed) warmed it up, let it run through the gears with the wheels off the ground, drove it for 40 km and realized I installed the spring on the reverse check assembly wrong because I rushed the job during a weeknight (I couldn't feel the tension when moving the shifter over to the 5-R fork) and couldn't find 3/4 gears on the mainshaft. DUH. So I drop the tranny again to fix it, and I check out my new bearing after 40km's of hwy/city driving, and there is already a bunch of minor spalling on it, and more than a comfortable amount of visible metal flakes in the oil/case. WTF. I was running synthetic amsoil 75w-90. I examined all the other cone bearings and they have a similar spalling despite being virtually new (only 450 KM on them, the outer 5 MM of the cone bearings are shiny and there are minor scratches on the middle portion of each cone bearing, which gives me the impression they have a balanced rotation without excessive preload). When I reinstalled the trans. again, I added a small amount of bearing grease to the race and the cone bearings from advice I got from a heavy equipment mechanic, (he said this is good for the original startup before the oil thoroughly penetrates the bearings, which may have been the reason for the high pitched squeek on startup when I had a new bearing installed with synthetic oil) who also suggested to never use synthetic fluid after a rebuild (which I had originally done). I reinstalled the tranny with a cheap gear lube to break it in. It shifts fine for running crappy oil, but I can hear what I believe to be the mainshaft bearing spinning.

question: 1) During the bearing break in period, is it normal that the oil will become contaminated with small metal flakes while the bearing/race 'polish' one another.
2) When I open the window, I can hear a very faint whirring sound, almost like a supercharger (which is very different from the grind of carrier bearing failure), that is most noticeable when the car slows down while coming to a stop, while the frequency of each rotation spread out and becomes individually perceivable. It doesn't sound like it's grinding, just spinning really fast, and it doesn't get audibly louder at higher speeds, but the frequency of the spinning definatly increases into a high pitch. Is this normal bearing break-in symptoms? It makes this sound regardless of the clutch being engaged or not. I am highly paranoid of this tranny side/mainshaft bearing getting chewed up again, even though the races/bearing are matched. I bought the car with a bad dif. bearing and have no idea what a properly functioning 4th gen MT trans. should sound like. I suspect the bearing grease I added may be the reason I can hear the spinning, but I figure it will get washed away with gear oil.
3) If the shafts are all shimmed properly and the bearings/races are new and installed properly while running a suitable gl-4 fluid, are there any other factors that will result in bearing grinding?