Bicycle Mechanics - Repairing pitted cones with valve grinding compound

I bought an old Francesco Moser bike with Gipiemme hubs and all was in great shape except the rear hubs which had significant pitting in the cones. I was going to buy another older Gipiemme hub and cannibalize the cones but the mechanic at my LBS (which is an old school pro in his 60's made an unexpected suggestion. He said "in the old days we used to pack the bearings with valve grinding compound instead of grease and work the bearings back and forth until we ground smooth custom seats inside, then cleaned it really well and reassembled it with grease. Now, I had three thoughts on this: (1) Is this why he never made it into the record books; (2) will it work; and (3) really, what have I got to lose as the hub as it is toast anyway. So I gave it a shot, first with 350 grit and it was going really slow as the seats are really hard. So I used a drill to speed things up and switched to finer and finer grits until I finished with some 80,000 grit that left a nice shinny polish. Then I replaced the balls with new ones and reassembled. I cannit say good as new, but WAY better than what it was. Has anyone heard of doing this?

I've heard about it, but i don't think it'll work. The cones are case hardened, and even if the abrasive magically ground them round again, they'll become "soft" and will wear out really fast. And what happened to the cups?

The cups seemed unmarked. Even the balls had little measurable wear, just the cones seemed to be "ground", and not very much but enough to smooth out the pits and rough spots. You mat be right that if I went through the hardened surface they may wear faster. Regardless, it is done so I'll just have to ride it a while and see. It was basically unusable before though, and now it is pretty smooth. I could not grind out all of the pits though as some were pretty deep but I'd say I ground out 70 - 80% of it. Like I said, I had nothing to lose and worst case I was going to have to buy a replacement hub anyway, so I was curious to try it. All in it was $3 for the coarse compound (I had the finer ones), $1 for the new balls after I ground it, and about an hour of my time. I'll just have to see how it holds up I guess... I am not recommending this for everybody, but maybe a consideration as last resort. I am also curious if anyone else has tried this...

Hope it works, if so please let us know. I've heard also about turning the axle w/cones on a drill press and using emery cloth, but never tried it.

Saw Aaron's bike shop in Seattle demonstrating a How to do It .. But.. it takes 2 drill Motors.

[I think the project, was a 'cold-day in Pergatory' chance of finding any spares]

One ,in a vise, you chuck an axle Up with the cone on it , and that one spins the cone..
not too very fast..

Second one you chuck up a Conic shaped stone bit and use the part of that cone grind stone
that matches the arc of the cone..
so Both are spinning..

as with any polishing project , you need to go down in grit size [up in number],
-> and keep the work cool so you dont take the hardness temper out..
a cup of water to dunk the part in will do.

once the pits are ground out , bringing the surface down to their level, then the further polishing can be with carborundum grit, Cloth..

a shoe shine stroke, back and forth, with the cloth strip will keep the grit in any one spot from all going away.
the fine lapping compounds can be a finishing step>

[ engine Valve lapping is done on the bench, cylander off the motor.. then all the grit is cleaned off, thoroughly
or the engine gets a few thousand miles of wear in, just the start up.]

i would just try it. repack them as suggested, then ride it for a fifty or a hundred miles or so. maybe check it once or twice for smoothness.

like you said, there's nothing to lose. if you've got the time. go for it.

It depends on the depth of the pitting, because as was pointed out earlier the steel below the surface is softer than the skin.

But there's a more serious issue, and that's the radius of curvature of the cup and cone.

As produced the curvatures are both greater than that of the balls, so they roll with point contact. If you use the same size balls to lap and you use to run, you'll form grooves with curvatures exactly like that of the balls. That means that the contact area will be wider increasing the friction, but more importantly there'll be contact at various radii. That means that whenever the balls run, there' be some rolling and some sliding contact. This increases friction and the wear rate.

So the lapping trick may be OK for shallow pitting, or as a quick patch for a bearing you plan to replace soon, or to save a collectible and make it workable with the understanding that it won't see much use.

until I finished with some 80,000 grit that left a nice shinny polish.

Wow. That's awfully fine.

Case hardening on bearing races in the hub is about .010-.015 thick.You would be lapping a long time before you get through that.They will last for awhile.

Some of the older and newer cones are/were not case hardened,they were hard chromed.Once the pits go through the chrome,it's toast.

This is why the cones are sometimes pitted and the races are not.

Just buy a replacement cone for God's sake.

Case hardening on bearing races in the hub is about .010-.015 thick.You would be lapping a long time before you get through that.They will last for awhile.

Some of the older and newer cones are/were not case hardened,they were hard chromed.Once the pits go through the chrome,it's toast.

This is why the cones are sometimes pitted and the races are not.

Not that it matters but some minor corrections and clarifications.

Whether through hardened, case hardened, or hard chromed, the surface metal will be harder than the core material to varying depths and degrees. In unplated parts this relates to the quenching process and the near instantaneous cooling of the surface vs. the slightly slower cooling of the core. Also case hardening isn't to a fixed depth, but to varying depth, anywhere from one to a few thousandths of an inch, as determined by the heat treating process, mostly involving the time spent at temperature in a carbon rich atmosphere.

Ground races will have some of the surface layer ground off in a final operation after heat treating, cutting into the available depth of harder material.

Hard chroming will add a yet harder surface to the ground cone, reducing wear until the very thin layer of chrome is worn off, but the chrome itself has no structural strength and depends on a hard layer below to support it.

The reason cones wear faster than cups is two fold.

1- the wear track on the cone is roughly 1/2 the length of the wear track of the cup so the cone wears faster.

2- cups don't need to be threaded so manufactures have the freedom to choose different grades of steel better suited to bearing use. That combined with the typically thinner sections generally means a better case to core ratio which reduces pitting as the surface layer wears.

To understand the importance of the case to core ratio, imagine skating on a frozen pond. When the ice is strong enough to support the skater the skates wear the surface slowly. On thinner ice, the water below doesn't provide support, and the ice begins to fracture and eventually the skater falls through.

Likewise on the wear track of a bearing. With good support from below the track wears smoothly as the balls roll over it. After a while the track wears enough that the remaining hard layer is too thin, and lacking good support below it begins to fracture and collapse leading to the characteristic pitting and gouging we see in badly worn cones.

So if a cone is worn to the point that pitting is happening, there's no way to restore a long lasting load bearing track, though you can buy some time via the lapping trick.

"Just buy a replacement cone for God's sake." Well, I tried that, but I went through all of the cones available at my LBS and not one was a decent fit. I was going to but a similar old hub on eBay and cannibalize the parts, but it had been months since a similar hub had sold on eBay and I did not want to wait a month and go through the hassle if watching and waiting, and then paying a ridiculous price for a "collectable" hub (last one was a pair for $120) and shipping it, etc.. This took a little longer than an hour to do, cost a couple bucks, and it is done and I am riding it. That is why. If it was a new Shimano, no problem, but a 35 year old Gipiemme, not so easy to just buy a replacement...

Thanks for all of the interesting information and suggestions. I clearly came to the right place for info! I guess this is not anything new then. I was able to save an old hub for a couple of bucks, though it may not last too long. I'll have to just ride it and see how it holds up. I'll post a follow-up this summer after I have a thousand miles on it or so and let you know how it holds up... Thanks everyone!

I would think it would depend on the depth of the surface hardening of the race. As suggested before, just replacing the bearing would be he long lasting solution.

I'd just case-harden the cones myself. Use a torch to get them up to temperature, drop them in motor oil. It certainly won't be as good as the factory hardening, but it will be better than not at all.

I'd just case-harden the cones myself. Use a torch to get them up to temperature, drop them in motor oil. It certainly won't be as good as the factory hardening, but it will be better than not at all.

In all likelihood, this will lower rather than raise the hardness of what had been a properly heat treated part. Heating in air usually results in loss of carbon at the surface making it less wear resistant.

In all likelihood, this will lower rather than raise the hardness of what had been a properly heat treated part. Heating in air usually results in loss of carbon at the surface making it less wear resistant.

The cone could also warp during the hardening process, which would effectively ruin it. I would rather accept a greater rate of wear than risk ruining an irreplacable cone.

Theoretically, it could work though. There are lots of people out there doing home hardening and tempering of steel knives, woodworking tools, and so forth.

On problem is that one might not know the type of steel used to make the cone, so the optimum temperature and quenching medium might not be known. Also, should the cone be tempered (I would guess yes), and at what temperature?

In all likelihood, this will lower rather than raise the hardness of what had been a properly heat treated part. Heating in air usually results in loss of carbon at the surface making it less wear resistant.

Reducing flame on a carbon block. The trick is quenching at the right temperature. But you're right- who knows what type of steel was used.

But, if it's wearing out in a couple hundred miles, and it's truly irreplaceable, then a chance of extending its life is better than no chance of extending its life. Nothing ventured, nothing gained.

Anybody know anything about induction hardening? ED

Anybody know anything about induction hardening? ED

Yes, but it won't help. Induction hardening use induction to heat the part. The part is subjected to a high frequency changing magnetic fields. This induces electrical eddy currents within the part causing it to heat up. It's the same basic principal used in magnetic trainers =, but at higher energy levels.

Induction heating is useful because it doesn't involve gas, so can be done in an inert atmosphere. It's also very energy efficient, and has been used in Italy for gasless brazing of production frames. BTW- induction heating has migrated from the factory to the kitchen in the form induction stoves, which can heat a metal pot, without getting hot themselves.

I have used Casenite to harden bolts and screws with no problem. Ed

By the way, grinding was still going really slow when I stopped so maybe a good chance I did not go through the hardened surface. I was amazed at how slow it went with all of that tough abrasive in there. I guess that goes to show just how durable the bearings must be.

It's hard to grind with rolling friction.

This thread is fascinating, BTW

A machinist told me it's not a good idea to use abrasive paste to grind moving parts against each other, because some abrasive keeps imbedded no matter how you clean it and that accelerates the wear.

A machinist told me it's not a good idea to use abrasive paste to grind moving parts against each other, because some abrasive keeps imbedded no matter how you clean it and that accelerates the wear.

That doesn't make sense. Lapping has been standard practice for eons. It's still standard practice for establishing perfect seals on valves when rebuilding engines.

But lapping a ball bearing will be a slow tedious and ineffective process because instead of the sliding motion that allows the grit to cut, you get a rolling operation where the grit is stationary and crushed under the ball at the point of contact. With superfine grit, you might get some polishing action, but this shortcut is far from ideal.

If you want to salvage a vintage irreplaceable cone (not worth the effort otherwise) you can spin grind it. You'll need a bench grinder with a fine wheel, a single point diamond dresser, and a power drill, preferably reversable. Mount the nut on an axle cone out out the end, and secured by a locknut behind. Use the dresser to create the desired curvature on the outer corner of the grinding wheel. Use the cone as a template.

Now you can spin the cone and grind it transferring the curve from the wheel to the cone. It works best if the cone spins opposite the wheel at high RPM, but if you don't have a reversible drill, low rpm in the same direction is OK too. This is a common method for grinding complex shapes in cylinders and works very well, but takes a steady hand, when done without a fixture. Be warned it's very fast even with a fine wheel, so you might want to practice with some scrap before doing the real thing.

Rubberized-compound grinding wheels can be found at jeweler's supply stores (google "Rio Grande"). They'll cut a lot more slowly than your typical hard abrasive wheel.

Rubberized-compound grinding wheels can be found at jeweler's supply stores (google "Rio Grande"). They'll cut a lot more slowly than your typical hard abrasive wheel.

True, but I don't know if they can be shaped. That's OK if you have good eyes and a steady hand and can produce the curve freehand.

BTW- however the grinding is done, once the shape is produced. You can switch to cloth wheels and various grits of polishing compound, ie, white diamond, tripoli and rouge, to bring up a mirror finish. This removes very little metal, so if you continue with the spin grinding method, you'll get a like new cone (except for the issue of material hardness).

I bought an old Francesco Moser bike with Gipiemme hubs and all was in great shape except the rear hubs which had significant pitting in the cones. I was going to buy another older Gipiemme hub and cannibalize the cones but the mechanic at my LBS (which is an old school pro in his 60's made an unexpected suggestion. He said "in the old days we used to pack the bearings with valve grinding compound instead of grease and work the bearings back and forth until we ground smooth custom seats inside, then cleaned it really well and reassembled it with grease. Now, I had three thoughts on this: (1) Is this why he never made it into the record books; (2) will it work; and (3) really, what have I got to lose as the hub as it is toast anyway. So I gave it a shot, first with 350 grit and it was going really slow as the seats are really hard. So I used a drill to speed things up and switched to finer and finer grits until I finished with some 80,000 grit that left a nice shinny polish. Then I replaced the balls with new ones and reassembled. I cannit say good as new, but WAY better than what it was. Has anyone heard of doing this?

If the temp fix doesn't hold long term, might just put the old wheel up on a hanger to be admired and fetch a usuable modern wheel off the net to keep you on the road. On the side; I thought the Gimpy's were like a Campy clone. If so, could be a source of parts.

On the more engineering side; if have access to the right tools, could machine out the old races on a lathe and press in replacments made of the correct type and hardness of material. Making the cones to match would be the easier half of the job.

This thread is fascinating, BTW I agree, way more information than I expected, and very interestingI I guess I should have posted before I did this instead of after...

I agree, way more information than I expected, and very interestingI I guess I should have posted before I did this instead of after...

No, you did the right thing. You have a working hub now, and you did it in the manner most likely to work without screwing it up more. :)

No, you did the right thing. You have a working hub now, and you did it in the manner most likely to work without screwing it up more. :) Thanks, it was toast as it was... It will be interesting to see how long it holds up though...

Bookmark this page and let us know in a year or so.

Just buy a replacement cone for God's sake.

That's what I was thinking! The OP said that his hub races were ok. Usually it's pitted hub races that are the problem,
and a rather difficult one to fix. Bad cones are much easier to find and replace. Using grease-grit in such a situation risks
damaging perfectly fine hub races...So I'd go with finding some new or good used complementary cones and intalling new
bearings. That makes much better sense to me. Of course if the hub races are shot, you might as well try the grease-grit
technique...or just trash the hub and rebuild with a new hub or just get a new wheel.