1953 M38A1 differential setup

[jnorris]

I'm replacing the bearings in both differentials on my M38A1 (Front: Dana 25/Rear: Dana 44). I've read many discussions and guides on this process and watched several youtube videos and am comfortable with the process but have a few practical questions. Some of the original shims are somewhat corroded and slightly mangled from the process of removing the bearing cups. When I measure the pinion depth shims with a micrometer I get .0117, .0113, .0035, .0035, .0035 (5 shims).
1. Are shims typically standard thicknesses like 3 thou, 7 thou, 10 thou, 15 thou etc. ?
2. How accurate are thicknesses of new shims - is something that measures 11 thou most likely a nominal 10 thou?
3. When replacing shims should I try to select them to get the total thickness exactly the same or should I assume the old shims have been thinned a bit by corrosion and abuse and go a bit bigger?
4. Am I too obsessive about the accuracy of shim thickness - should I replace a shim that is around 10 thou with new one that is also around 10 thou?

I've iterated through the assembly process of my rear diff several times now, I get the pinion preload and backlash within spec and the pattern on the coast side started out looking pretty good with shims close to the originals but the drive side is on the toe and at the crown of the tooth. I've progressively removed shims to move the pinion further from the carrier and adjusted the carrier shims to keep the backlash in spec but my pattern is still not right and my pinion depth shim pack is way less that the original (now closer to 10 thou than the original ~33 thou).

I've read that with worn gears (mine are not new but don't appear to be heavily worn) the drive pattern doesn't give a good indication and I should focus more on the coast pattern. Should I go back to a shim pack closer to what I removed and verify the coast pattern and call it a day?

I don't see any signs that the axel has been assembled incorrectly in the past so I am fairly confident that it was shimmed correctly last time but the bearings were in pretty bad shape.

[5JeepsAz]

This requires popcorn! Looking forward to the discussion.

[bmorgil]

1. Are shims typically standard thicknesses like 3 thou, 7 thou, 10 thou, 15 thou etc. ? Yes typicly you get them in .005 increments. They are available in .001 increments however, if you are able to find them. They are used in O.E. assembly and could be in the original shim pack in ,001 increments.

2. How accurate are thicknesses of new shims - is something that measures 11 thou most likely a nominal 10 thou? It will be closer than you can measure without an air gauge. It will be precise to tenths at a minimum.

3. When replacing shims should I try to select them to get the total thickness exactly the same or should I assume the old shims have been thinned a bit by corrosion and abuse and go a bit bigger? Measure each shim individually at the best spot. If they look worn try to find the thickest spot. Do not measure them all together. Use the added individual measurement. Use as few shims as possible on reassembly, it is stronger.

4. Am I too obsessive about the accuracy of shim thickness - should I replace a shim that is around 10 thou with new one that is also around 10 thou? Measure them both and trust your measurements. It isn't the exact number you are looking for. You are looking for the same number on the same micrometer. In other words just go by the micrometer. If the old stack adds up to .110, go with as close to .110 as you can get. I use a stack of shims inches high and find the combination I need.

I've iterated through the assembly process of my rear diff several times now, I get the pinion preload and backlash within spec and the pattern on the coast side started out looking pretty good with shims close to the originals but the drive side is on the toe and at the crown of the tooth. I've progressively removed shims to move the pinion further from the carrier and adjusted the carrier shims to keep the backlash in spec but my pattern is still not right and my pinion depth shim pack is way less that the original (now closer to 10 thou than the original ~33 thou). You cannot get a pattern on the drive side without load. You need to load the carrier HARD. I use a nylon block wedged between the ring gear and the case. You can also use the brakes. Concentrate on the drive side, the coast side is what it is. Remember the load spreads from the toe to the rear and from the root to the top when load increases. On a used gear set it can be difficult to get a good pattern as it is usually quite a large contact pattern. If the shims are original stick to them or their size and it will be darn close.

[51 CJ3]

I am getting ready to do this for the first time myself. The last person in there did not have the bearing caps installed correctly so everything is suspect. Good tips on locking the carrier to put a load on it. What do you think is the best way to put the load on it?

[bmorgil]

Jeff I have used both ways. A nylon wedge or the brakes if they are there. The brakes will apply a sufficient load easily. Typically I have used a Nylon wedge. I will say if you can use them, the brakes work best. You want a load sufficient to make it difficult to turn the pinion.

[jnorris]

Thank you for the very complete answer. I haven't measured many shims (new or old) that are exact but they seem to be close to 3, 5, 7, 10, 15, and 30 thou. I'm using an inexpensive micrometer but it checks out when I verify it on my feeler gauge - I guess my old shims are deteriorated and maybe my new ones aren't the best quality.

I've been using a crowbar jammed between the case and the ring gear to apply load but I have to admit I'm not confident that it's effective - I'll see if I can make up a nylon jig. My vehicle is stripped to it's smallest parts right now - it'll be a long time before I can use the brakes.

Everything else you suggest lines up with what I've been thinking but just don't have the confidence and experience to be sure. Thanks again for the input, I'm ready to take this thing on again.

[51 CJ3]

Jeff I have used both ways. A nylon wedge or the brakes if they are there. The brakes will apply a sufficient load easily. Typically I have used a Nylon wedge. I will say if you can use them, the breaks work best. You want a load sufficient to make it difficult to turn the pinion.

I am missing something. If I apply the brakes, the only thing with a load on it is the brakes. If I turn the pinion with the brakes applied then I have a load on the carrier as well. It reads to me all that needs to be done is keep it from turning.

[bmorgil]

Jeff you are missing something. He is looking for a pattern so you need the gear set to turn but under a solid load so the contact pattern is visible and starts to give an indication of contact. We are not trying to hold it still here.

[51 CJ3]

Then my question remains unanswered. If I have this correct, you are creating drag by using a wedge or light application of the brakes to check how the teeth interact. How are you putting a load to the teeth? Using a wrench on the pinion nut? Connecting the drive shaft and turning it with the engine? Using a drill or impact? Screwdriver through the u-joint u-bolts? Pipe wrench on the output shaft? Like I said, I have not worked a differential before. I have lots of other maintenance experience but most of it is in aviation, some boat and tractor. Overhauled a transmission on a ‘66 CJ5 and did a transfer case conversion on a ‘74 Cherokee about 30 years ago. I hauled a couple of axle assemblies to a shop to have work done but this is the first I have attempted myself.

[bmorgil]

We cannot have a question remain unanswered! Now I get it... you have to paint the picture for me sometimes.

You are going to create drag and apply torque via the pinion nut.

[51 CJ3]

Some things would be easier to watch on YouTube than ask about but internet in this area is really bad. I am stuck using cellular with a data cap in an area with erratic service. One of the power companies has been talking about installing fiber optic cable on their poles for a couple of years but so far it has been more talk than work and it isn’t the company that supplies my house.

[bmorgil]

In the 1980's and 90's I participated in many videos and manuals full of pictures and process. I would completely rebuild products while the video team taped. Then they would edit this (without my mistakes) into technical videos and service manuals. Then as we moved into the digital age, most of this was archived and lost. Everything new was about "New". Digital storage was huge and costly. Since then I have not seen many of those early videos again. It would be nice to find old videos people have done and have them on a disc so we could watch them without the internet. Sadly it seems a lot of that old technical training stuff is gone.

[51 CJ3]

It’s hard to believe memory was $1.00 per megabyte not that long ago. My first “BIG” hard drive was 250MB and cost me $250. Now you can buy a 250GB drive for under $30. Digital video quality was horrible compared to today’s standards because of the storage issues. I agree, we have lost a lot during the switch to digital. Maybe the videos will surface again and someone will convert them.

[gmwillys]

Vintage government transmission video.....

https://www.youtube.com/watch?v=fw_VBiTntq8

[bmorgil]

Hahaha! A classic, what did he say?

[gmwillys]

Side fumbling, axial wain shaft, or dingle arm? Lotus gram meter….

[5JeepsAz]

I don't know what any of those words mean