Off Roading Forums banner

Checking axle housing for straightness

13K views 29 replies 10 participants last post by  jeepgod 
#1 ·
Anyone have a link or know how to make sure that my D44 housing is straight? Anyone know how much off it can be and still last?
 
#5 ·
Jim, unless you've got some lasers and sights, the only way I know for the average guy is to lay a long perfectly straight piece of metal across the inside of the tubes and check for clearance between the tube and the housing. That's not going to be easy. Then, if it is bent, how do you propose to fix it without a couple of 100 ton presses and jigs?

As for how far out can it go... check the slop in a new set of bearings, cut the slop in half and that's about the amount of out of round/straight tolerance you can have. But the axle will still be running crooked,and that is going to wear the seal.

Unless you have the equipment to check for true and restore the axle tubes if their bent... it may be a good idea to take this one into a shop.

With your shop, and experience, IMHO, you'd be better off removing the tubes and lathing new thicker tube stock for insertion and welding. The you'd just have to make the jig to hold everything true when you're welding. But at least you know you'd be starting with straight tubes.
 
#6 · (Edited)
How about make a couple of tapers to fit into the axle bearings. Put a threaded rod through them and tighten nuts to stretch it. Then clamp a dial indicator to the rod so that it reads against a carrier bearing journal. Then rotate the rod in the axle bearings. If the indicator reading doesn't change, the bearings must be in line.

I don't know what I'd do if it isn't straight, but I have a 50-ton press and access to a 200 ton press. I can move it around for sure.
 
#12 · (Edited)
1. I was just curious as why you think the axle tube/carrier may be bent in the first place?

2. Shouldn't all the bearing journals all the way thru the axle tube assembly have to be in line with each other ? I think you would need to run the threaded rod all the way thru; centering the all thread at each journal. But I don't know if I would trust the reference "straightness" of all thread over that length in that case.
 
#8 ·
Jim, I think you idea would work. I was thinking of using a piece of pipe instead of a threaded rod. I thought about checking straightness of my HD AMC 20 housing, before building it but decided it was a waste of time.

Everything I have read says the bearings need to be in an exact line. The problem is that "exact line" is a meaningless term. If I tested the housing I would most likely have found some run out. I would then have to decide if it was enough to make a problem (0.0005" 0.005" 0.05") Who knows. Rather then get results that I had no idea of the meaning of, I decided to just ignore the issue. My understanding is that the axle splines will wear quicker if the housing is bent as they will need to act like CV joints.

One could even argue that the housing should not be completely straight while on the work bench. The housing will flex: as weight is taken on and off, as the gears are loaded, and on a semi float the axles will flex as well. Even changing the backspacing of the wheels will result in a different loading of the housing. Resulting in the housing bending slightly.

I know none if this is much help. It just seemed to me that no matter how many tests I did, I would still only know it was a straight enough housing after I ran it and it did not eat parts.
 
#9 ·
It also doesn't tell you where it's bent, and how to compensate.
Where it's bent really doesn't matter. One would have to assume that the carrier bearings are where they're supposed to be, and the ends of the tubes must be in line with them.

It would be easier to do if one could mount the arbor to the carrier and then rotate it. Then you would know exactly how far and in which direction each tube end was out of line.

But it really is pointless, as Wilhelm points out, unless there is an allowable tolerance somewhere. There's nothing in the '78 FSM about it.
 
#10 ·
Then you would know exactly how far and in which direction each tube end was out of line.
And that's the rub! You're using very tight tolerance on one end of the tube. By the time you extrapolate that out to the outside of the tube, the measurement has splayed. Also it really doesn't detect if the two tubes are square to one another.

What would it take for you to rig up a jig to mount and rotate the whole assembly?
 
#11 ·
Rotating the whole assembly wouldn't help. It's the inside that has to be straight. An indicator on a rod stretched between the axle bearings would show if the carrier bearings weren't in line with the axle bearings, and which way they were out. That's all you need to know.
 
#13 ·
I had thought mine had some bow in it at one time.

I pulled it and put it on my bench minus shafts and proceeded to check it every where I could with a straight edge (steel ruler).

The only places I found to be untrue were where I had ground off brackets.

I also checked the axle shaft bearing races for any discoloration or unusual wear marks. No bad signs there.

That was about the best I could do at home, but I found this article while researching bent tubes,

Straightening a rear end housing - Crankshaft Coalition Wiki
 
#14 ·
I wonder just how critical it really is.

Remember, the axles only extend 1/2 way through. They are supported in only 2 places each - outer and inner bearings, not 4. If it was 4 (if the axle was one long piece going all the way through) it would be critical. It would have to be line honed.

The bearings are made with a little "slop" anyway.

The splines on the inside are on the side gears - they have huge tolerances and go wherever the axle ends goes.

We've all seen diffs that had the wheels with too much positive camber - from too heavy a load or from jumping - but they never seemed to wear out. Some rigs that have been jumped get way off, but even they don't seem to have trouble.

If it is way out, I'd worry, but just a tad I wouldn't. I'd think that an eyeball sighting would tell enough. Maybe put the side bearings in the carrier and the axle bearings in the tubes - without the axles, and sight through them.

And, you may even want to call Currie or someone who builds diffs from scratch and ask how critical, and how they check it.
 
#15 ·
Come to think of it --
Many larger heavier cars have rear specs for rear axle alignment. Some have a Positive Camber, and some have a TOE OUT spec.
Toe OUT on drive wheels, toe IN on driven wheels.
Positive camber adds stability in the rear, and helps reduce understeer.

So - if a straight rear axle has specs that keep it from being straight, then obviously it's not real critical for it to be absolute straight.

Ever look at a row of the same model of new cars? Notice all the front ends, upper control arms, have the same number of shims. Is Detroit that accurate - knowing sometimes they can't even get a door straight?
No!
They assemble it all the same, one of the last steps is to check the alignment - then big hydraulic rams come out and bend or tweak components or the frame till it's in spec. Same for the rear - they bend the axle to specs if needed.

If they can do it, so can you!
 
#16 ·
i usually lay that puppy down on the ground.. stand back a good 20 feet or so.. and give a good hard look.. then drink a few beers.. and give her another good long look.. then walk out to the axle and look at it one last time.. if it looks straight.. well.. it's straight enough.. slap that puppy in.. and drive it.. watch your tires.. they will tell ya.. and also if you give it a 4 wheel alignment.. and its still out of wak when driving behind it.. that will let ya know also.. but i'm a cheap ass.. and just keep a keen eye on the rubber.. by the time you see any goofy wearing.. it will be time to rotate the tires.. and you won't be out much.. :D
 
#17 ·
1. I was just curious as why you think the axle tube/carrier may be bent in the first place?

2. Shouldn't all the bearing journals all the way thru the axle tube assembly have to be in line with each other ? I think you would need to run the threaded rod all the way thru; centering the all thread at each journal. But I don't know if I would trust the reference "straightness" of all thread over that length in that case.
1) I have no reason to think it's bent. I just want to check it and straighten it if necessary.

2) Running the thread rod all the way through is what I was proposing. And it doesn't have to be straight at all. It just has to be stretched tight enough that it doesn't sag as it's turned.

In view of Rich's comments about big solid axle cars, I may just use the 'beer and stare' technique.

What started me on this quest is that one of the shops I called about setting up the gears said that they check the housing before building it up.
 
#19 ·
The amount of beer used also makes a difference. Several beers insures the straightness, but too many will bend it like a pretzel. When it's a pretzel, don't try to straighten it - eventually it'll fix itself.

Maybe something to consider while it's out and stripped. Like LEVE said, the tubes aren't very stout. I've seen the tubes get ripped open on rocks. On my Rubi I've ripped the mounting brackets off because the tubes were so thin.

How 'bout splitting another tube lengthwise slightly larger in diameter. Weld it to the bottom of the existing axle tube - as a stiffener and skid plate. Making armor on about the bottom 1/3 of the tube.

I haven't done mine yet - but have done 2 others. I've stitch welded the stiffener a little at a time, and let it totally cool between each short stitch without distortion.
Clamp it on tight. Tack one end solid, cool, then tack the other end, cool, then the middle, cool, then stitch between them cooling each time, a little at a time till it's on solid, or close to it.

--- Do not use water or anything to help it cool! Air cool only! ---

Remember, it will try to pull toward the welds. If your piece reaches about 1/2 way round the axle tube, then the welds end up at the front and back of the axle tubes.

The hardest part is splitting the tube to be used as the skid, easy with a plasma cutter.
 
#20 ·
I have a plasma cutter, so that's no problem.

That's a good idea for hard rock wheeling, but that isn't what we do around here. We deal mostly with mud and wood. What rocks we do see are primarily sandstone or limestone, although there is granite in the Ozarks.

I think, but haven't proven yet, that these tubes are thicker than the LD AMC20 that's been under it for thirty years without damage. Do you know the diameter and thickness of the Rubi tubes?
 
#21 ·
Ohhh I see now, I thought the axle was installed on your Jeep but you're speaking of a replacement axle. That's why I asked what made you think it was bent; I was thinking you hit something. :crazy:

Maybe a quick way to satisfy yourself that it is straight would be to measure from the edge of differential carrier on a diagonal out to each end of the tubes, like you would if you were going to truss the axle like RRich suggested. That would at least give you an idea of straightness in the vertical plane. :)
 
#24 · (Edited)
Actually, Jim, I was thinking about using some point on the bottom of the differential housing. Like maybe a pencil mark transfered on to the bottom center of the gasket surface. Then measure out to each top of the tubes from there. Another method could be to measure from the bottom left hand side cover bolt hole to the top end of the left tube and then from the right bottom hole to the top of the right tube. I'm assuming that the cover bolt holes are at equal distances around the cover so if those two measurements are the same then the axle assembly would be straight, at least to the accuracy of your tape measure (about 1/32"). :)
 
#27 ·
If it looks straight with just your eye, it should be close enough. A more important thing to check is the shaft itself. One of my shafts had a bent flange so the tire wobbled and went up and down. Moser can fix that too or make you new stronger ones for about $170 each I think.
 
#28 ·
The seal surfaces on these shafts are worn, and sleeves for them are going to cost $64 per shaft. I checked Moser and they want $160 per for new shafts, so I'll stick with what I have. But I will chuck them in the lathe and check for runout on the flange before I spend that much on them. Good pointer. Thanks!
 
#29 ·
You don't think fixing them the ******* way will work?

Fill weld the grooves on the shafts, then make them round again with a hand grinder?



He He - a few days ago a friend at a 4x4 shop showed me a "******* Repair" - a driveshaft - the U-joint ears had been broken off - so someone brazed the ears back on - with brass rod that barely stuck. When my friend told the guy he needed a new driveshaft, the guy got mad!

He guy said he already fixed the shaft and U-joint. He brought it in for a vibration! Called him a crook trying to sell something he didn't need!

Go figure -- It takes all kinds.
 
#30 ·
ya jim.. axles don't need to be that straight for driving.. i used to think that way also.. until i cut an axle tube in half.. to cut out 3 inches.. then welded it back together and then cut two tubes length wise.. then put them on the top and bottom of both axle tubes.. and welded them on.. fulling welding them.. (the oringinal tubes were pretty rusted where the spring perches were and it was thin tube to begin with)...

i thought to myself.. this will never be straight..

but i ran it in my orange jeep.. when it was my daily driver.. drove some 40 thousand miles on it.. just kept an eye on the rear tires.. and they wore even.. so after about 15 thousand.. i rotated them.. and kept the rotation cycle at around 5k miles.. and i could never tell if it was straight or not.. but it drove straight... tires wore evenly.. that's all i asked for..

the axle was a semi-float dana 60 35 spline with a detroit in it..
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top