Thanks guys. I too have mixed thoughts on replacing the axle, but have ordered (what I hope is correct) the axle from an e-bay dealer; an Omix-ada stock replacement. There was a great deal of confusion on my part due to the way they are listed. Most postings show the axle with wheel studs, retainer plates, seals and bearings, however, there is a large cost difference between the assembly and the bare axle shaft. There also is a big difference in listings for Rubicons and non-Rubicons. Some suppliers list separately for TJ’s and Wranglers, which is just dumb because all 2003 (my year Jeep) and 2004 (the reputed year for the axle assembly) of this type are both Wranglers and TJ’s; the model differences are Sport, X, Unlimited (long wheel base), and Rubicon (or some variation thereof). Essentially what we think of as a traditional military bred style of jeep is a Wrangler, and all models from 1997 to 2006 are TJ’s (coil sprung pre-JK). At least that is how I understand it.
So what I found was that most 2004 Rubicon axles were listed as having ABS and disc brakes. Disc brakes are correct but ABS, and the accompanying tone ring, with apparently longer spacing outboard of the wheel bearing, simply did not jive. As best that I could tell the ABS didn’t happen until after 2006 with the change coming with the 2007 JK. Mostly I found Rubicon listings approaching $340, whereas basic TJ D44 axles where listed down around $130-140; same 30 spline count, same 29.2 inch left rear length, same 5x4.5 bolt circle.
Anyway, after calling a few major suppliers, all of my local recyclers, looking on CL and e-bay for used, and cross referencing the Spicer assembly number and the Omix-Ada part numbers, I think I got what I need… except it won’t be here until a week from Monday (ack!).
So I can see how the rest of the build goes and make a decision as to whether to assemble the “bad” axle for temporary use, spend more money for another set 10 axle bearing kit for the new axle, and have to pull the wheel and drum again to swap the axle; or spend more money to extend the rental car and do it right the first time.
Today I tried to stick the outer pinion bearing on with the old preload shims and found that the bearing is also a slight press fit on the pinion shaft (Karl and I had to tap the 5.13 pinion out of it so I don’t know why I thought this would be a slip fit and not require a separate setup bearing. I dug out the old Toyo brand bearing, did the lathe/grinder mod to it and got it to slip on cleanly; washed it out, wiped down the pinion again (after test fitting and handling under less than sterile conditions) and assembled with the old 1310 yoke.
The yoke wrench I made worked great, except it was a pain to go between bolting that up and checking the preload with the torque wrench. I ended up just clamping a 2x4 to the side of the yoke and letting it block against the saw horse that the rearend was setting on. This was quicker to install and remove, and did the job fine for the low level of torque needed to take up slack clearance. The heavy bolt on wrench will still be needed when it comes time to fully torque the pinion nut.
Using the previous pinion preload shim stack, but not really torquing the nut too much, but tight, I got to the point where there was no wink or slop, but also no drag (i.e. no running torque resistance); so I figured that I would have to remove some preload shims. (The manuals say to do this step w/o any preload shims, but I kind of wanted to get a feel for it with the original baseline.)
I decided to leave that be for now and do an initial check of the carrier; mostly to see if I was on track w/o the ring gear spacer. So in went the carrier with no shims. I got a consistent 0.152 inch thrust into the pinion. I could make it go more, but sensed that the carrier was starting to ride up on the pinion, lifting the opposite side race out of the saddle. Since I could get this to repeat, and could pretty much tell where the carrier started to ride up, I figured that was good for an initial check w/o installing the bearing caps. This was good news, in that it confirmed that the gear spacer would not be needed.
I measured the original shims again, this time using a 0-1 inch micrometer, and found that they were each .001 inch more than I had measured with the calipers. So .131 inch on the gear side and .141 on the opposite side. So if I swapped the .141 shim to the gear side and add a .011 shim from the kit I would get the recommended gear side preload (i.e. zero initial clearance; .141 + .011 = .152).
From before, the total thrust (less preload) was .258, so that means that the prior preload was actually about 0.014 inch (closer to the spec .015; .131 + .141 = .272; .272 - .258 = .014).
So if we subtract the .152 from the measured .258 total we get a .106 remainder on the opposite side (for zero preload). Add in the .015 desired preload and we are looking for a shim pack of .121.
In theory this preload spreads the housing equally, so we should get about .0075 clearance with the pinion after the carrier is forced down between the shims.
Not having a case spreader, not wanting to have to make one, and having read that you can force it by using a dead blow mallet, I decided to try it even though they recommend using the zero preload settings for setup.
Well, unlike the single individual shims that came out of the 5.13 gear setup, the shims included in the setup kit packs turned out to not be hardened; some were even galvanneal plated. We managed to put a pretty good gouge in one of them trying to get the carrier to tap in. Not only that, but they were also not the full dimension. Compared to the good hard steel shims they were more open on their ID, so would have less bearing surface under the cup race. Not sure how much, but I know that Timken specifies a minimum bearing support diameter; in other words, you have to adequately support the back of the cup and can’t just cheap out on these shims. I guess they weren’t any less than the area backed by the casting on the outer side of the saddles (which have a narrower land than adjacent to the saddles), but they just don’t seem to have the same quality as the larger hardened steel shims.
We figured we could get it to go, eventually, but decided to stop and work on the setup with zero preload.
The thinnest shim in the kit was about .005-.006, then .010-.011, then .015-.016, a couple at .025-.026, and two at .099-.100.
Attempting to set zero preload while splitting the difference in lash clearance, I backed off on the gear side shim to .141 + .005 for .146, and backed off on the opposite side to .099 + .011 for .110. This would still be about .004 preload, and was as close as I could get to balanced with the available shims w/o resorting to stacking a bunch of thinner ones (wanting to keep the thickest shim possible up against the bearing cups to avoid damaging them when “dropping” the carrier in).
The carrier went in much easier, but with some tapping. However, it was pretty clear that there was no lash or wink between the ring and pinion. The ring gear was tight and could not be easily turned, while the pinion could be turned but felt “notchy” in doing so. We decided that it couldn’t hurt to do an initial read with the marking compound to confirm what we suspected was a high pinion; and it was. The pattern on the drive side was running off the toe (inside diameter of ring gear) with the tell-tale wedge shape biased toward the flank (valley), while the coast side looked to be well centered between flank and face (tip of gear), but was also biased toward the toe.
(Required reading:
Spicer Dana 44 Service Manual and
Yukon Gear Installation Guide)
So that left me with the choice of not getting any further today, or knocking the inner pinion bearing cup race out of the housing and using the less desirable shims. I decided to call it and look and see if any of the shims in the D30 kit I had at home were the same (they weren’t). I could try to lap the good hard shim down, but keeping the faces parallel would be critical, and knowing how much to take would be trial and error.
Also need to reduce the pinion preload shims some, plus whatever I take away from the depth shim. We have a surface grinder at work with a magnetic table, so I can get Don to show me how to run that, but that will take time, as will sourcing more of the correct shims. Ultimately, once I figure out the final opposite side shim thickness I will grind the good .131 shim down to use instead of the unhardened stacked shims.
Arrgh. I suspect this is only the beginning, but hope that after just a couple of iterations things will fall into place… hopefully sooner rather than later.