TR2/3/3A Steering Worm Gear Runout

I seem to be having difficulty setting the worm gear preload. With the end cover and shims installed, the resistance to turn changes depending upon the angular position of the shaft. For about 230-degrees, it turns smooth with no up-down play. As it rotates through the other 130-degrees, the resistance to turn builds up and then drops back off. With the end cover off, I noticed that the race moves up and down slightly while sitting on the ball bearings when turning the shaft. I placed a dial indicator on the worm gear (as perpendicular as I could get it) where the ball bearings roll, and detected about 4-mils of runout. I could also detect a very small amount of runout with the naked eye. Has anyone seen this before? Do I have any options for repair other than new worm gear and/or shaft?

Thanks,
Joel

If I understand correctly, it sounds like your shaft is bent. Try removing the upper column bushing.
If now the end of the shaft wobbles as you turn it, but the box is OK, you may be able to straighten the shaft in a press. Mine would almost touch the outer column, but I was able to reduce that a lot by judiciously applying pressure in the press.

But if the worm doesn't run true on the ball bearings, then I think a different worm (either new or used) will be the only good solution. AFAIK, the bearing races are machined directly into the worm and then hardened. So there's no practical way to correct them.

If you have removed the back bearing race, then you know how tight and precise they fit in the housing, and they have to be fully in the back of the case to get all the adjustments and settings correct. If that is the case, I would tap it back with a brass drift. However, you should have felt that earlier before the shaft went inside the box.

When the drag/tension is set on the shaft in the box going left to right, the shaft drags in the center and you would feel that for sure, but it should be in the center as the wheels go left to right and pass over the center. Is the shaft and pitman line lined up?

I removed the upper column bushing and re-checked. The shaft is tilted over to one side of the column but it does not wobble when I turn the shaft - it just spins in that location. And the worm gear now spins smooth in the box. I also used a brass drift to make sure the back race was driven home, and verified that the depth was consistent all around with calipers. If it wasn't for the shaft being tilted over, I think I would be good to go. Any idea on the next step?

View attachment 57285

Just checked the straightness of the column itself, and it is NG. What is the best technique for bending the hollow column back into position without damaging it?

Just to through this out there, the box mounting holes on the bottom are over-sized for adjustment and the same inside the cab. Can you push the whole assembly around some with a bar on the bottom by losing up those 2 bolts? Could you have bound the assembly up some on installation? I have a bent column also. It is bent at the shock tower. I am not using that column now, but I did use it many years ago and did not notice it; it is a one piece column, and I think the column could tolerate some un- straight- ness and still not be felt at the wheel. Heck they are like 5 or 6 feet long with not much inside them. Yes, sure a new one would probably be best. Is your column one piece? I did wrap the column in rages and rugs and put it under my work bench and bent it up and back some, but was always afraid it would fold.

My column is a two piece, and I guess it's possible that it was bound upon assembly. I installed it several months ago, and then realized I probably left too much play in the rocker bushing (I read that no play is needed for better steering). Since this is a frame off, I wanted to check while everything was easily accessible. So I removed the box, replaced the bush, reamed to no play, and now discovered the bent column.

Having said that, I just wrapped the column in towels and manually bent it off the lift arm of my 57 John Deere 420T. It did leave a very small kink in the column, but it's not that noticeable and doesn't even come close to contacting the steering shaft, which is now centered. The preload also appears to be mostly even throughout 360 degrees. If I notice any problems once I finally get to drive the car, I will repair/replace. But for now, I'm just glad to save a little money on a project whose cost is not so little.