Mad Honk 

A couple of weeks ago in the C&V sales area I put up a post in Malacal pays it forward to send in pics and discussion about how I go about reconditioning bearing races in a head set. So Ben this one is for you.
This will take a few posts to get everything in so bear with me.
I have had a headset in the shop now with some minor damage to the races. Here are pics of the races that need work and the process involved in bringing them back to functional use.

Lower outer race and damage at the 11:30 position

Dimpling on the upper outer race

The upper inner race is fairly clean and will just need some polishing. The lower race was missing, so a Brampton will be the replacement and it will need significant help to get to mirror finish.

These are the bits I use for cleaning the outer races. They are rubber infused with diamond dust and create a mirror finish for the races

If you note in the previous picture the bits are squared off so they must be contoured to match the outer race dimension This is done with the bit in the tool, running and a belt sander.

Mad Honk 

I start the first part of the process by giving all of the parts a bath. A small container and some high VOC cleaner. Next about a half hour of deep cleaning with small brushes and white cleaning cloths to make sure all dirt is gone.

bath tub for the parts

The first brush and the cleaning solution.
Now I will get to work and after a bit, I can post pics of the cleaned parts and the first polishing steps. Smiles, MH

WGB

Looking forward to this. I just ripped down a mystery bike and the headset has the same issues. Too good to toss and to rough to install.

A week or so ago I took a dremel with nylon abrasive wheels to that bike's hubs (rusty and slightly brindled) and used WD-40 as a lubricant, now they gleam. I was thinking of trying that on the headset but I suspect this gentleman's methods will be better.

Mad Honk 

After some cleaning, the parts are clean enough to work on. Maybe I am just OCD but I hate working on anything dirty. First pic is of the cloth after the parts got out of the bath. Not a lot of dirt but enough for me to be complaining about. Then some pics of the process for polishing the outer races, and then the finished races.

Argh! Dirt!

The polishing of the outer races

Perhaps a better pic of the bit on the races. There is a bit of heat generated so I use the gloves

After polishing the race is once again mirror smooth. Note that the dimpling is gone.

Lower race is also ride ready at this point.
I am going to take a break and will get back after this project after re-setting the tools needed for the inner races. About an hour into the project so far. Smiles, MH

xiaoman1 

MH,
Thanks for doing this, I know that it takes a lot of effort on your part.
UP!
Best, Ben
Mad Honk

Mad Honk 

Wesley,
Look for the bits I mentioned earlier. The Temo's are available for about a buck apiece at Colden Coulee a hardware store in LaCross, WI . I try to buy multiples to keep shipping costs low. But they are the bomb for this process!
Ben,
You asked for this. It will be a couple of days before I have it all documented for ya. I'm gonna watch some racing and then reset the tools for the other races. Smiles, MH

oneclick 

I'd be worried that I'd grind the races out-of-round in spots, if I was doing it free hand. One ten-thousandth of an inch (roughly a thousandth of the thickness of a caucasian hair) and you're over the limit for grade 25 balls.

Reynolds 

I'm not a fan of refinishing races, but when I did I held the cone on an old axle and turned it on a bench drill. Then I used a Dremel to polish the race on the spinning cone.

oneclick 

The issue is that when you have a place where the distance between the cup race and the cone race is larger than the ball diameter, that ball no longer takes any of the load. That's why a bearing with only one or a few (spacing is a factor with multiples) pits can be perfectly functional. As the number of places increase, fewer and fewer balls take the load at that point, and eventually load exceeds the elastic limit of either the ball or one (or both) of the races at another point and you get a(nother) deformation of some kind. This is a cascading situation.

The time a ball is in such a place is thus more important than how far away it is; a ten-thousandth is enough, deeper makes no difference. Widen the pit by grinding or polishing and it may look better but it'll last worse. The number of simultaneous times it happens is perhaps an even more significant factor; that is one of two reasons that uncaged balls are better - their circulation is somewhat random. If you want a bearing to last (at bicycle rpms) get rid of the ball cages. The only operating benefit of a ball-cage is eliminating ball-ball-contact wear; in the main the degree of such contact is a result of ball inertia not load, and so high speed operation is the only place it really helps. If you've got random pits and enough balls that's better than evenly spaced pits, and *much* better than evenly spaced pits with cages.

With the inevitable innacuracy of hand grinding I doubt the situation would improve. A rigid setup on a mill or lathe or a specialized tool like a Quorn (and skill) could grind and polish a new race accurately enough, but then you may have a problem with the race actually shifting location.

Reynolds 

Even if you could refinish the race on a lathe with a grinding wheel, if the pits are deep you could end up removing the hardened surface I guess.

unworthy1

Never heard of "Temo" brand but have used the similar (I think) "Cratex" brand which come it many sizes and shapes and degrees of coarseness from "coarse" to extra-fine. Cratex uses mostly silicone-carbide, not diamond abrasive but perhaps they have that option now, too. Hard to find any Dremel bits for a buck each so will have to keep a look out for Temo brand. I used to do drafting so have a Koh-i-nor electric eraser from way back in primitive times, these use 6" long 1/4" diameter eraser strips (12 in a box) and the gray one used for "ink" is a very effective rust eater. Cheaper than Cratex but might be too abrasive as a polisher

oneclick 

Maybe.

If it's a cheap part, quite possibly; they are only hard on the surface. Case-hardening depth is on the order of 1mm, but it varies a lot.

Good bearing are ground after hardening - it allows a more controlled and accurate oven-hardening, which goes all the way through. Poor bearings are ground then case-hardened. Normandy bearings are stamped/punched/oozed out of a toothpaste tube, and then hardened in an ashtray full of Gitanes.

Mad Honk 

Ben, and Wes,
Got into the shop today after doing a bit of golf related work (got pay the bills ya know?). So I set up the drill press for the grind on the lower race. I picked a dark steel lower race to show the contrast of the work as it is done. And I used an appropriate stainless steel rod diameter with a a slot cut into it to accept paper for the grinding of the surface for the bearings to ride on. I started with a 220 grit paper for two passes to remove damage and clean the race for final polishing. There are pics here of the race and the amount of material pulled off the steel race and left on the paper. The final two passes were with 600 grit wet paper and then 1500 grit wet paper. The pics show the change and amounts of material removed. The final result was tested with a new set of caged Campy bearings and the lower outer race. Bearing assembly rides smooth and straight in the refurbished races. In a further part of this thread I will cover my process for making sure the caged bearings are functional and perfectly clean before any re-assembly. here are some pics, Smiles, MH

Hardened black steel lower race to be re-conditioned.

Note the damage to the race from dirt and debris and old bearing wear

The set-up for the polishing, using a diameter specific split rod in my drill press with 220 grit paper, and the race secured in a high speed drill to make sure the race is polished uniformly.

Two passes with 220 grit paper to remove damage to the race. Paper show the amount of material removed.

One pass with 600 grit wet paper and paper with material removed. By now the finish is starting show the mirror effects.

Final pass with 1500 grit wet paper. note mirror and the amount of material left on the paper.

xiaoman1 

Mad Honk,
Very impressive work, I was wondering how you would machine the race to the proper dimensions....I assume that you have many different "barrels for the varying bearing sizes etc.
What method did you use to secure the race to the high-speed drill, it looks like it's riding on a rubber mandrel, is it secured on the opposite side of the drill....how?
Thanks very much for this very informative thread., looking forward to the second part.
Best, Ben


Like this one?
Mad Honk

Mad Honk 

Ben,
My mandrel is a bit older than the one you show but they work the same. If you go back and look real close the rubber section is covered by about .030" of masking tape made by 3M. Very consistent thicknesses so I use it. I spoke earlier today with a collector form Chicago who has been a friend for 40 years. We only get to talk infrequently these days, but we covered all sorts of things, being friends with Othon Ochsner, Years of racing and coaching in the Little 500, the bikes from Breaking Away. Bearing tolerances came up and he reminded me of the courses I used to teach for the free university here. I used to tell the students that bearings are like the definition of a circle; a lot of points all equi-distant from the center. And when you think about a bearing that way, it rests on two of those points on either side of the ball. That point is about a few thousandths of an inch. Now think about putting a rider's weight in that little point. The weight when multiplied out to one square inch works out to tons per square inch. So that is what we are trying to correct when the bearing surfaces are being re-conditioned. The damage occurred at the tons per square inch rate. Hope this helps in explaining what I am up to here. Smiles, Honk

Mad Honk 

OK Ben,
Today was supposed to be a golf day but the rain and mid 40's kept me in the bike shop. So another head set is in from sunny CA. I bought it off the bay, and the seller sez it is just fine for a daily rider. Take a look at the pictures and tell me it is ride-able in the shape it came in. The damage is a bit rough, but there is enough to save it. I am going to re-furbish the lower race only on this one as it has the most damage.

Crown race, note the dimpling on the race where the bearings ride.

The lower head tube race and note the dimples. These two races when used together will result in an indexed steering system

Lower race after reconditioning. Dimples gone and nice shinny surface for bearings to ride on.

This one took a little extra work to remove about .010" to take the dimpling out of the race. These two races when used together now will steer smoothly and not be indexed.
I have this one finished and next post will cover the bearing retainers and how I clean them and replace the bearings. Smiles, MH

xiaoman1 

M.H.
Amazing work....and a pleasure to see someone that is able to solve problems with their hands that others would say are not achievable. I still hand lap the valves on my V-8's and a feeler gauge (not a dial) to set my valves!
I hope the weather isn't inclement so you can at least get the front 9 in!
Thanks again for the education.
Best, Ben

Mad Honk 

Ben,
Today was a day in the golf shop. Ping irons getting a new lease on life and perhaps more playable for the owner. Same thing tomorrow, just going to add a 3 wood to the mix that didn't survive "taking a knee".
I am going to find a head set that came in with the damaged bearing cages that is normally what I see. Smiles, MH
BTW, You wanna come to middle earth and build a 7000 rpm 350. One blew up last season, and it will need to be reconstructed. Broke the block at the pan section when a rod failed. I'm glad it is the off season!

WGB

I think the key to enjoying life is to own an insulated, proper machine shop (and wood shop). I have an un insulated garage with winter heading in. Minimum I need a vise to hold small items still instead of a shaking hand.

trailangel

I would've just put it together with valve lapping compound, worked it around a bit... cleaned it, grease it and assembled with fresh bearings and called it good.

xiaoman1 

M.H.
I think you may have discovered another sideline.....I am looking forward to your new post there on how to revive a set of Beres"s, are you a member of "Golf Forums" yet?
Best, Ben

xiaoman1 

DBL post

oneclick 

I wonder what a DTI would say if he mounted a worked-on bearing on something with better runout than a hand-drill and measured it. A straight valve in a guide in good condition might do, they should be 2 thousandths or less.

Mad Honk 

Ben.
I finally got my regular computer back so I can post pics. Here are the pics of cleaning and replacing the bearings in the caged headset bearings. Most that come in have old grease packed into the spaces between the bearings, and the use of small Dremel wire brushes to clean the spaces out. What comes out of the spaces is like a wax that hold dirt and debris so removing it is necessary for a good re-conditioning of the bearing cages. Pics are of the sequence I use to clean and replace the bearings.

old bearing cage with damaged bearings

process of removing bearings from the cage

old bearings on the right side new bearings on the left and empty cage after cleaning

process of cleaning the old grease residue out of cage

Hilander

I know this is an older thread, but anyway...

I wonder how regrinding affects longevity?
Not sure if headset or other bicycle bearing races in bottom brackets or wheel axles are through hardened, but if these bearing races are surface (case) hardened only, then the hardened layer will be lost during regrind, leading to a fast reproduction of new damages to them.