3500lb axles bad design

I checked the Al-Ko axle and found that one end measured 1.3764" and the other end was 1.3749". The radius where the bearing contacts was large enough to completely eliminate a square shoulder contact surface. I suspect that all dimensions that I checked on two brands of bearings and two brands of axles were within the design tolerances, but without the axle blueprints it can only be a guess. Regardless, it is a bad design any way you look at it.

Lynnmor wrote:

JBarca wrote:
I found this, this tells us having a resultant clearance of 0.0019" between the bearing bore and the spindle OD is out of tolerance by 0.0007" too big.

Your bearing would fall in the 0" to 3" bore bearing chart on page 2

https://www.timken.com/pdf/10829_MDV5-Correct-Fits-For-Your-Wheel-Bearing.pdf

As I thought, the ideal middle of the clearance resultant fit spec is 0.0007" clear between spindle and bearing. They declare a min of 0.0002" clear to 0.0012" max between the shaft and the bearing. See page 2, under bearing cone bore diameter.

It looks like you to much running clearance between the shaft and the bearing bore.

Hope this helps.

John

Thanks for that. I really wanted to find the manufacturing tolerances for the bearing and spindle. We both know that a perfect fit would be best. I did find that the bearing is to be 1.3775" to 1.3780" The spindle size is to be 1.3760" and that is what I have so with unknown tolerances, the clearance could be greater.

So anyway, at .0019" my parts are within the specification of .0015" to .0020", so again that clearance could be even greater if we had the spindle tolerance.

I have another bare axle of a different brand that I will measure tomorrow.

Hi Lynnmor,

This has me bugged now and is a good post to figure out why that bearing spun.

We both are coming up close on the bearing. I tried to find the bearing on the Timken site but that site is so big that trying to find what I want, is always a challenge. I'm an older handbook guy...

So I went to the SKF site where I have more luck most times. I have a set of new SKF L68149/L68111 bearings in my shop now left over from years ago when I had the smaller camper. But my mic's stop at 1". Work always had the bigger stuff.

Here is the bearing, SKF L 68149/111 Tapered roller dim's

And here is the tolerances. You start at this page SKF Tolerance list page

And this puts you into Taper Roller tolerances. SKF Tapered Roller Tolerance You are looking for the t delt tmv

This comes out to a bore of 1.3774" + 0.0005" /- 0.0000"

So what you found is within a tenth, so OK we are good on that.

And if your bearing that you measured is 1.3779" it is sitting on the top edge of the bearing tolerance. By any chance, was that the bearing that scored the shoulder or a new one in a box?

They declare the shaft shoulder to be 1.8504" min. See here, scroll down a little and look for d sub b SKF shaft shoulder recommendations

That 1.8504" min SKF recommendation shoulder seems to conflict "I think" with what spindle is. While the spindle is smaller, I'm still not seeing that the reason to cause the bearing to spin.

So far I cannot find the spindle tolerance. Have you tried calling Dexter what is it they use?

Here are my current thoughts on this. On the bearing and spindle you measured (don't know if that is the scored spindle/bearing) there is 0.0019" clearance.

That's a lot of clearance in my view. If I was spec'ing out a running fit for a shaft and a bronze bushing in a medium duty application, using a rule of thumb of 0.001"/ 1 inch of diameter, on a 1.3750/1.3745 shaft, I would set the bushing bore at 1.3763/1.3768. This will give me a running clearance of 0.0013/0.0023" and the shaft will spin all day/week/year long as long as it has lube and run at normal temps.

If you have a 0.0019" clearance on your bearing to spindle, how is that ever going to lock up enough to hold the bearing race from spinning?

A fundamental unknown in this, what is the max clearance between shaft and bearing to allow the bearing to tip and lock up on the shaft under load so it is easier to spin the bearing on the rollers then the bore?

If you go by the Timken specs I listed above, this appears they do not want any more clearance then 0.0012" max.

We really need a spec on the spindle to help solve this.

I'm curious as all get out right now on how this is supposed to work by the numbers.

Hope this helps

John

I’m on the cell phone right now, I’ll review your resources better when I get to the computer. To answers some questions, I emailed Dexter in hopes of getting answers in writing, but no response so far. The bearings were new Timken with about 6,000 miles on them. I did notice that the Timken is cut away even farther than other bearings I checked. I should have taken a photo of the Al-Ko axle showing that there is only contact with the sharp corner of the bearing with the radius on the spindle. Unbelievable!

I still think that the loss of bearing adjustment caused by the shoulder issue, causes bearing races to lose the proper relationship with each other. That small shoulder will wear or deform at an angle due to the way weight and side loads act on it. In addition, the unground radius and relief is not necessarily running true and it becomes the support surface. With races and rollers skewed with the spindle centerline, some torque is applied and causes the spin. Now that is my opinion, I do like to use facts.

One curious thing is that both bearings spun on the left side and no sing of spinning on the right.

I didn't read all four pages of this but if you are have a problem with the race spinning due a gap of just a couple thou, wouldn't this be a good application for loctite bearing mount ?

Here is a link to the specs for it:

http://na.henkel-adhesives.com/us/content_data/316041_FINAL_LT4680_AAM_Do_It_Guide_v7.pdf

I looked at the information you found, that .130" less shoulder contact is huge when you consider that the #84 spindle has next to nothing. My thought is that nearly all trailer axles have about the same bearing to spindle clearance as I have found.

The idea just mentioned about using an adhesive would work, I even have plenty of the right type on hand. The problem is that the spindle would become very difficult to service and might require heat to disassemble. With PA state inspections pulling the hubs yearly, that is not an option.