Posted by Rick Denney on June 02, 2002 at 08:12:22:
In Reply to: Re: Re: Re: Re: Re: Re: Swedging Your Rotors posted by Jim Andrada on June 01, 2002 at 14:43:38:
I agree with the suspicion that very few qualified engineers have ever been consulted on tuba design. Tradition and trial-and-error do achieve functional results, but often miss the slap-on-the-forehead breakthrough.
My own short-action valves do not inspire me to believe that the Conn method is necessarily superior to the traditional approach. My Conn's valves are not on the same planet as the Tanabe-rebuilt B&M valves on the York Master, though I agree it isn't a fair comparison.
You are correct that roller bearings are designed for full-rotation applications. Lubrication is essential in these bearings, because the precession of the rollers (or balls) creates friction in the bearing. The smaller the bearing in relation to the roller diameter, the worse the problem.
Ball bearings that don't turn will pump out the lubricant, and wear holes in the bearing races by a microscopic fretting process. I'm not sure that this would happen with rotary valves, which turn a full 90 degrees in routine use. It is a problem with bearings that carry a load but that move over an arc smaller than a degree. Your example of a hard-disk pickup-arm bearing is one such case, but one more folks have likely encountered is involves the steering tube bearings on a bicycle. These are easily worn into notches, but not because of impact as most folks assume, particularly when people ride on very straight bumpy roads, such as fast downhills.
Interestingly, precision clocks and watches are not made with roller bearings of any sort. They are made with steel pins that ride in mineral sleeves (called jewels) that are very hard, and provided the absolute minimum of lubrication. They will last longer than a roller bearing and operate with less friction. Even without jewels, clocks use steel pins in brass or bronze bushings that are often swaged to remove wear. But in most clocks, a worn-out bushing hole can be drilled out and a bronze replacement bushing can be hammering into place, which makes me wonder if replacement bushings for the current design might be possible for someone with a lathe, especially for rare and old horns that no longer have available replacement parts.
For rotary valves, a hardened steel pin in a replaceable porous bronze bushing (that is permanently oiled) would be just the ticket. But it could only be repaired using the correct replacement parts. No swaging here to get you through the night on an old horn for which replacement parts are no longer available. And thrust clearance should be controlled with thin steel washers.
But sealed bearings of any sort would be a poor choice. The seals themselves provide considerable friction, and would make the valves feel mushy. Plus, you need a fairly heavy grease to make sure the seal is not broken; otherwise the rubber seal will harden and shrink, breaking the seal and complicating repair.
By the way, the front bearing of a rotary valve gets almost all the force. That front bearing is the axis for the bending moment caused by the linkage, and thus absorbs almost all the lateral force. The rear bearing gets only perhaps a fifth of the linkage-caused force that the front bearing gets. Of course, the rear bearing usually has a smaller bearing surface.
Rick "an engineer" Denney