Head slap moment

[Bigwolf]

You know those times when you slap yourself on the forehead and think "what a dumb a$$ I am!".  "I knew better than that"!  Well maybe you don't but I do.  Had one today actually.  I pulled the rear wheel for a new tire and routine maint.  There they were! Bad bearings both sides.  I did the bearing mod just before Inzane XIV.  Those bearings have less than 10K on them!  Maybe even less than 6K.  I am not sure but the last 2 years have not given me much riding time.  Anyway, the head slap, aha, moment was when I thought about it.  If I had thought about it before doing the brg mod, I would have realized the potential it had to destroy those bearings.  But I did not think about it before, I just followed what others had done.
The problem is this: The double row bearing for the right side is an angular contact bearing designed to lock axial movement of the axel to the axial movement of the wheel.  That 5204 brg will allow very little axial movement.  When we put the 5204 in both the left side of the wheel and the right side, both bearings are trying to set the axial position of the wheel in the bike.  The problem is then we have a steel spacer separating the inside races of the bearings and an aluminum wheel separating the outside races of the bearings.  Those spacings may be equal at room temperature but quickly become different lengths when the temperature changes.  That is because aluminum "grows" faster than steel when heat is applied.  That mismatch in length was enough to axially overload and destroy the new bearings that I put in so few miles ago.  The single row 6204 will allow a little bit more axial movement, maybe just barely enough more.
Chances are this would not have been a problem if I used the rear brake less in hot weather and never accidentally got my big foot on the brake pedal for a mile or 2 without realizing it.  But then again, maybe it would have.  Bottom line is, from an engineering standpoint, it is a poor combination.  I intend to see if I can find a better solution to beef up the left bearing.

[Chrisj CMA]

Seems "fishy" that others have done this mod and put lots of miles without issue.

[pancho]

Hmmm,,  you might have something there Bigwolf,,,,,   what you are describing is the very reason you should never install a wheel bearing with a socket or any tool that just contacts the outer race....  unequal loading of the bearing races when it contacts the spacer. 

[pancho]

After doing a bit of research into angular contact 5000 series, and radial load 6000 series bearings, I can't make any conclusions as to weather this is an inappropriate application.... Someone needs to look into this further.. you out there CA?

[Bigwolf]

Hmmm,,  you might have something there Bigwolf,,,,,   what you are describing is the very reason you should never install a wheel bearing with a socket or any tool that just contacts the outer race....  unequal loading of the bearing races when it contacts the spacer. 

Yes, I installed these with a 3/4 inch bolt and 2 custom 3/4 thick flat washers.  It might have been ok but I burned the rear brake up last spring.  Had not ridden in several months and had forgotten to constantly check that my big and nerve dead foot was not on the rear brake.  Went about 6 miles wondering if I was running out of fuel or needed to clean the carbs.  I believe that heat entering the back wheel triggered the overload and early demise of the bearings.  Even so, had I not had the mod,  the bearings would have had a better chance of surviving.

[Tfrank59]

I don't think it's the coefficient of expansion aluminum versus steel going on, but probably something like the length of the inner spacer was allowing an excessive axial load which caused the failure.  Just my 1.5 cents

[Jess from VA]

Seems "fishy" that others have done this mod and put lots of miles without issue.

Yes, it does.  A lot of us have done it.  So far as I know, this is the only reported failure.

[pancho]

This may be the first time that much additional heat was put into the assembly with the bearing mod...  

still, an interesting case

[Ricky-D]

There have been a few reported failures of the double row bearing modification over the past years. There's more to it than simply

slapping a bearing in there and expecting the result will be a long life cycle, although I think most bearing failures are from improper

installation when the discussion is limited to replacement bearings. Failure of OE bearings I think is mainly from power washing.

***

[DK]

Substitute an aluminum spacer?

[Skinhead]

The amount of thermal expansion you are getting is MINIMAL.  The length of the spacer is what 8" max?  The thermal expansion can be calculated using the thermal expansion coefficients for Al and Fe (used to know them, but would have to look them up after all those years of not using this stuff)  but we're talking 0.00000xin/in/degreeF.

Ok, I looked it up, here is an example for a 100 foot long pipe heated from 32 to 212F:
Example - Thermal Expansion of an Alloy Steel Pipe**

An alloy steel pipe with length 100 feet is heated from 32 to 212oF. The expansion coefficient is 8 10-6 (in/inoF).

The expansion of the pipe can be calculated as:

    dl = (8 10-6 in/inoF) (100 ft) (12 in/ft) ((212 oF) - (32 oF))

        = 1.728 inches

**http://www.engineeringtoolbox.com/thermal-expansion-pipes-d_283.html

If our spacer is 8 in long this would be (8 in*0.000008 in/inoF)*(212oF-32 oF)=0.01152 in

for an 8 in piece of Al this would be (8in*0.0000123 in/inoF)*(212oF-320f)=0.017712 in

This means a difference in expansion of 0.01771 - 0.01152=0.00619 in going from 32 to 212 oF.  I'd be more inclined to think you didn't get the bearing fully seated or some how damaged it during installation.

The only other factor would be the parallelism between the surfaces of your cutdown spacer and the correct dimensions of the spacer.

I did the double row mod over 40,000 miles ago on my interstate and it was smooth as silk at last check. 

[pancho]

Nice work Skinhead, if your calculations roughly translate into what happens in the physical layout of the rear wheel, .006 movement applied into angular contact bearings which are designed to have as little axial movement as possible, would create tremendous pressure in the bearings.

[pancho]

Gotta consider that most would never generate the amount of heat that Bigwolf did.

[Bighead]

I think some folks waaaaaaaaay over think stuff put new bearings in and ride it!