peugeot mongrel

It felt like early indexing. The races are perfect except the top of this race crumbled away. It had 21 3/16 loose balls in it which I think is right. Any idea why it did it? Any idea where to get a replacement?

Michael Angelo 

Ebay.

FBinNY 

I'd just be speculating about what caused it, but I can't believe you couldn't feel something wrong as it was happening. There's no way this kind of damage could happen while the headset still felt OK.

Normally only the lower stack suffers any wear, so you might do OK with a used headset, that maybe has a shot lower stack, or find a NOS headset and use the upper stack only until your lower one dies, then complete the changeover.

BTW- it can't matter now, but one possibility is that you had 1 ball too many, though you would have had to miscount as 1039 hold 22, or trapped one of more above the rim when assembling the headset, and ran with the balls trapped up there and riding between the stepped recess in the top cup and the rim. But as I said, I can't believe it turned at all smoothly.

BTW- I'm curious as to what the top cup looks like, could you post a photo?

onespeedbiker

Too few bearings. Without a cage you need 25 3/16" ball bearings top and bottom.

peugeot mongrel

I did feel it and took it apart to check it out. It seemed a little bit loose and a little bit rough. I rode it very few miles once I felt it. 21 balls top and 21 bottom, I caught them all with a pan when I took it apart. Grease was pretty good.

It only holds 22 3/16 with half a ball space

FBinNY 

Being short a few balls shouldn't cause this problem. As long an there can't ever be an arc of about 90-120° without balls the headset would "tripod" off the balls to either side of the gap. What has me wondering, and why I wanted to see the upper cup was something had to be stressing the lip.

I was hoping to see corresponding damage in the cup, since something had to bar down pretty hard on the lip. Unfortunaely I don't see any obvious signs in the cup, but maybe the OP can. To elaborate on my earlier shot in the dark (not enough info to call it a theory). Look at the step to the inside the ball race. Imaging if a few balls were trapped up there instead of outside in their proper track. Once the headset was assembled they would be trapped, and couldn't hop over the step to the ball track.

That's just a total crap shoot, but the OP should look or any sign of wear t the edge of that step to confirm or eliminate that theory.

onespeedbiker

Perhaps I was a little fast on the 25 bb. I have a NOS 1039 Gran Sport headset that I believe is the same with the NR having better finish, cups and races, and bb..The GS headset comes with 50 loose ball bearings but they are 5/32 not 3/16".

I did load up some 3/16 bb just for kicks and it took 20 with a space just too small for 21.

peugeot mongrel

There is no scarring inside that cup. I can't imagine if balls were that out of place how you could even tighten it up to adjust it. I was certainly surprised when I saw it that way. I installed it used but it was smooth as butter maybe 800 miles ago. I took it apart because it seemed a bit "rattly" on rough roads and I could feel some roughness when I started focusing on it. It wasn't loose. When I opened it up nothing was out of place. Something sure as h... ground it up. I'll look again in the daylight tomorrow.

peugeot mongrel

In the '74 - #17 catalog record track & GS are 1040. They are a little different with 5/32 balls

If it wasn't for that ball size difference I'd be tempted to use a tange top race I have that is 30.2 till I find the right one. I think though I'll use tange top cup & race temporarily & leave the campi bottom so I don't have to pop the crown race off my fork.

FBinNY 

There's always the possibility that something broke off, for reasons unknown, and ended up up in that area, moving and jamming as you turned the fork, spreading the damage like a cancer. It doesn't explain what started it, but might explain the extent.

onespeedbiker

When I took the time to compare the 1039 and 1040, mine is obviously the 1040. But I do have to thank you for getting me to open the packet of bearings. This was before the pre-greased bearings in a plastic bag and it turns out 3/4 of the bearings were rusted so I ordered up some 5/32 (only size not in my parts box)..I'm still curious about what happened to your upper cup/race. If you had been using a retainer cage it looks like something was caught between the cage and the top of the cone..As you can see in this Velobase.com photo, the retainer rides right on the spot that crumbled..

peugeot mongrel

I don't think it is a common sort damage. If I figure it out I'll post it. There has to be a clue still there somewhere. I don't think I had it too tight. When I set it up it was snug & smooth, and not as rattly on the road. I agree something had to get in there. None of the balls are disfigured that I can see. Thanks for input. marvin

FBinNY 

I've gone through a few 1039s on my old Frejus. I'm a bit of a pack rat (maybe more than a bit) so there's a good chance I saved an unused upper stack. Don't know exactly where to start, but I'll poke around and if I find a top cone, I'll give it to you.

peugeot mongrel

That would be too fantastic if you do. I might just have something you can use. thanks Marvin

FBinNY 

There's still a chance the cone is still perfectly OK functionally. Test as follows.

Clean the cup and cone.
Use a dry marker to color the polished races themselves, on both parts.
Use pack with balls, For this it would be better to do it dry, but that's would be difficult without a retainer, so use a grease that doesn't wash away the dry marker. (if you can find some die blue, it's grease/oil proof).

Press both parts together and rotate between your palms for as long as you have the patience.

Open it up. If the ball track worn through the die comes nowhere near the damaged lip, as I suspect it might, you're good to go.

onespeedbiker

+1 I've been meaning to bring that up and kept forgetting. It certainly be worth a try if for nothing else to see if it's still doing what ever it was doing that caused the damage.


Further, in looking at the photo of the upper threaded cup in post #5, it looks like on the inside of the cup, just around the opening (under the writing) there may be a polished circular area, as if the top of the race that disintegrated was rubbing against it; or maybe it's just the photo..

ultraman6970

1st time I see this failure. You can test with like 10 balls w/o the cage and check if it works ok, if it works ok just continue using it w/o the cage. Campagnolo has good headsets but they dont last forever as other brands.

FBinNY 

You need more than 10 balls, not a full complement, but 80% of one at least, Otherwise all the balls will gather on one side, queering the test.

peugeot mongrel

I was thinking sort of the same. I have some new 3/16 retainers that are not campi so they have fewer balls & more space but they stay together. When I put them together dry you can feel the roughness but I suspect there is some burring on the edges at the damage I can probably file smooth. It looks like the chewed part should be above where the balls travel. I'll try the marker to see where they seat. Good Idea! thanks marvin

ultraman6970

You never test with the whole thing (all the balls)... if with 10 balls that clearly arent enough works fine then add the other ones, besides the OP has the cup out already... just stick a few balls inside, see how it feels, add more feel again. If it feels ok put the bike together and ride it until the bike dies.

FBinNY 

In use the damaged area can't touch anything. If it were tall and didn't touch (which it doesn't if all is right), increasing the airspace above can't make a difference.

BTW- it isn't at all possible that you used 5/32" instead of 3/16" balls is it?

peugeot mongrel

That pix had flash. There are no marks in the cup. Here is a better one.

peugeot mongrel

No the balls are 3/16

FBinNY 

If you could keep the balls spaced around the cups then all that's needed is 3 to form a tripod. But in practice you can't maintain location, except with a retainer, or with enough balls that the gap is well less than 1/2 the width.

You can try this and see the problem for yourself, or imaging what would would happen if you cut 2 of the four legs out of a table.

peugeot mongrel

I'll try smoothing the rough edges tomorrow