making "loose ball" steering head bearing races

Like the ones that are on bicycles or older motorcycles with loose bearing balls and an upper and lower race. You know, the ones that dump 40+ balls on the floor when you take them apart or try to reassemble them



How would those be made for small production runs?

There are some aftermarket taper roller bearing kits for older motorcycles where they have ground the ID and/or OD of the TRB to fit the existing components. But it may not be possible to always find a TRB that can be modified to fit.

I turned the OD of some Thomson shafting recently and once I got the right speed/feed the RC60-62 case hardening seemed to machine pretty nicely with one of Curtis' "universal steel" WNMG inserts.

If through-hardened steel could be purchased (1.5-1.75" OD) could a CNC lathe make these kind of bearing races and get a smooth enough surface for the balls to run on? Or would there have to be a grinding operation on the actual race area?

A Bultaco trials bike I just bought needs new races and it looks like replacement parts are available, though I may need to source them from the UK. But it made me wonder how involved making hardened parts like these would be so I thought I'd ask and see if I could learn something.

cheers,
Michael

You might look into Torrington needle thrust bearings. You get two hardened washers, in a choice of thicknesses, and a needle and cage assembly. They come in many sizes, so they can be adapted to many applications.

Larry

Larry, they need to take both axial and radial loads, which is probably one reason that TRB are a popular item for this use.

I've just been told by another Sherpa T owner that there are TRB retrofit kits available and I'll certainly go that way versus chasing the balls over the floor. But I'm still curious about how people would make these parts. I'd only faced the end of Thomson shafting and I'd been expecting that turning down the case-hardened OD would give a pretty crappy finish, but that isn't at all what I got.

http://www.eurospares.com/graphics/Bultaco/SherpaTswingarm05.jpg

That shows the OEM swing arm pivot inner bushings which are hard-chromed aluminum. As you can see, big chunks of chrome are missing. The stepped area on the steel bushing is where a seal rides to keep dirt out of the grease, and I was pleased at getting a finish that doesn't look like something that will chew up the seal.

I'm trapping that needle bearing race between the big bushing and the "top hat" end part (4140HT) for the end thrust. I put a very slight register on the top hat to locate the ID of the race washer.

I've seen reference to "hard turning" in re CAT toolholder's tapers, so maybe this really hard stuff can be turned smooth enough to not need grinding. Since I'm unlikely to be putting precision grinding equipment in the garage that would be good to know about.

cheers,
Michael

Michael - As slow as those bearings turn, I'd think a nice turned finish would work fine. CBN and ceramics are great for turning hardened steel, I've only used CBN in any quantity, some people prefer the ceramic inserts. I believe CBN will cut harder parts than ceramics will from what I've heard here. Here's a CBN insert that will fit your WNMG holder - http://cgi.ebay.com/4-NEW-CNMA-432-...ryZ58211QQssPageNameZWDVWQQrdZ1QQcmdZViewItem

As for what material to start with, I wonder if you can buy a very thick thrust washer or thrust bearing race of the right size? Perhaps you could just resurface the bearing surface of the existing races? Otherwise I imagine a piece of 52100 bearing steel would be the ticket, through hardened. I personally don't know of anywhere to purchase steel barstock like that already hardened, other than 4140HT which is only around 28-32Rc.

Micheal, I can see what your'e thinking of, but is there anything here that might go with a grind on a dim or 2? I was thinking ''single direction '' type
http://www.skf.com/portal/skf/home/products?maincatalogue=1&lang=en&newlink=1_8_3

Slightly at a tangent, but have been known to sleeve the spindle to mix and match, and make up top hat bushes to achieve the same too.
If all else fails, I have been known to select trb bearings for a headstock, then machine up a new yoke (triple tree) spindle from a solid piece of en16t to suit. That caters for most of the remaining eventualities (plus we're usually trying to fit forks to something they weren't fitted to originally.)
There a few specialist suppliers who will supply trb to suit whatever you want, but this usually costs $$$ and means you have to restock from them in future...

Glad you found the timkin conversion kit, much better way to go!

per your "just 'cause I want to learn something new" question, I'm not sure the races used origionally were all that hard to begin with...maybe 4140 (or similar), or case hardened CRS maybe? One of the big issues dealing with the ball-bearing necks was the way the ball bearing would dimple the race over time (most notably in the 'straight-ahead' position where the forks spend most of their time), and you could feel the dimple as you rotated the fork side-to-side. The timkin races will do it too, but at a much slower rate (unless you really beat the front end hard, like doing wheelies on a sport bike). I've also noticed in the occasions where 'destructive removal methods' were required, the BB races were much easier to cut than timkin races

I really don't want to get into making a new stem and the various sleeve and top nuts. The idea in buying this bike was that I'd not have to turn it into a machine shop project. I've got more than enough of those, I want to go riding on this one. Here's a link to a photo of it (it is too large to embed in the thread)

http://www.eurospares.com/graphics/Bultaco/Model1590012a.jpg

I got some feedback on the tapered roller bearing kit and it appears that may be one of those "close enough to work but not really fitting into the original design" deals. I've got the last two pairs of OEM bearings on hand at one of the USA Bultaco parts people coming my way. That will let me take the old bearings (which do have a few "detent" spots on them) and sacrifice one to see how hard it is.

There is one of the SKF single direction bearings (51105) that might fit though I don't have a OA thickness measurement for the stock bearing on hand. It looks like it is right on the ID and either right or a mm off on the OD (I didn't measure things when I had them apart because they looked OK and I hoped they wouldn't need any further attention). SKF seems pretty firm that they aren't to experience any radial loads. Here's what the SKF bearing looks like:



The stock races are more like an angular contact bearing, where the race on the stem has the bearing face skewed on the ID, and the race in the frame has it skewed towards the OD. I don't know if the thrust bearing with the centralized raceway would be a problem or not. The OEM races may also be made that way to give more contact on the stem/in the frame. But if the dimensions are right the other bearings are very inexpensive and worth ordering a couple to try and fit them. Maybe they'd work OK for a year or two and then a fresh set could be tossed in.

A nice thing about those single direction bearings is the balls are caged. I think that will mean fewer of them and possibly higher loads per ball, making them more likely to dent the races. SKF shows a static load of 3570 lbf for the 51105 bearing.

cheers,
Michael

Nice Sherpa Macheal, brought back a few memories, IIRC your idea of using off the shelf bearings and changing them out in a year or two won't be too far off from what Bultacos were like anyway

IMO a good machine at what it did, (helped by the great Sammy Miller, that guy wasn't human! but that's another subject) but some the engineering could be?........ (careful Sami) questionable.

That TRB kit that some think doesn't fit quite as nicely as it is supposed to is said to be a Miller item.

My bike has good hard-chrome in the front brake drum but patches missing in the rear. I've talked to Langcourt about having them do electroless nikasil plating on the rear hub after I strip out the remaining chrome. I wasn't aware that they could do "additives" like the silicon carbide in the plating with the electroless process. I'd like to avoid going to an inserted steel or CI liner. These wheels are really light, and if I can keep them all aluminum with a hard-surfacing I intend to do so.

cheers,
Michael

''That TRB kit that some think doesn't fit quite as nicely as it is supposed to is said to be a Miller item.''

Now I see why, it's a Miller items.....strange how some things never change isn't it?

I would like to go a bit further in regards this thread. I will possibly need to make a similar bearing but 6 5/8" od and a narrow race area taking 5/16" balls. It is the same as the head bearings in the first post.It is a big constant drive bearing in a 100 year old motorcycle. How do I turn the race at the right radius, then get it smooth, then get it hard without warping???

When possible, I have re-ground the old races and added external spacer (washer) to take up the clearance. I have done this for motocycles, bicycles including headrace and cranksets; and once a conical seat thrust bearing for my largest rotab. Have also made races including for woodworking equipment. I use S7 as being the most practical for the application, for small scale one at a time use. It does not get quite as hard as others, but the toughness makes me feel more secure.

I have re-ground races with spindex's & a grinder, on the lathe with a die grinder in the toolpost, and now my preferred option, on a toolpost grinder.

The following photos show making a bearing race from scratch for a missing race from a Euro woodworking machine bought at auction. No replacements are available from the OEM. Fortunately the outer race had been tied to the machine with twine so it was not necessary to remake that in this instance. The S7 blank has been rough turned on a lathe, and then hardened in a furnace, in foil, and tempered only the minimum.



Like lathe work, it's best if all the concentric features can be completed in the same set up. Such as center bore, perpendicular faces, and race.
















smt

Ok, thanks for that. The pictures tell the story. The one I need is pretty big, so will move a lot if case hardened, and I don't have a grinder attachment.