OK this is of no real importance, but it's been a minor mystery to me for years. And then in thinking about it as I was formulating the question to post here, I think I figured it out. So I thought I'd go ahead and post about it, just in case anybody else ever wondered the same thing (or maybe this was obvious to everybody else in the world):
I have your basic vibrating brass tumbler (a hand-me down from Strad, BTW, thank you sir), the kind with the threaded rod sticking up the middle to hold the lid on. When the machine is running, and I loosen the nut, rather than staying in same place or wandering randomly up and down with the vibration, the nut always climbs upward, and at a pretty good rate, maybe a half inch per second. Why is this?
It reminded me of the old electric football games. The player pieces had little spines on their bases, angled backward to provide the 'bias' to move forward as the playing field vibrated (no need to take my word for this, young people):
But what I didn't get was, what provides the upward bias for my nut? Shouldn't gravity provide a downward bias if anything? Is it the way the threads are cut? Does vibration at that frequency reverse gravity?
Then it hit me: the rod isn't just shaking. More precisely, it's swinging, since it's fixed at the bottom. The nut's being thrown, not perpendicular to the rod, but at a tangent to the arc of its path as the rod swings back and forth; a vector, often referred to as "centrifugal force", with a component away from the center of the rotation, just like when you windmill your arm to swing the ketchup to the mouth of the bottle and the cap comes off and you have to clean ketchup off the kitchen ceiling, except in just a couple degrees of arc rather than 360.
OK. I can sleep tonight. After I copy this into an email to the Nobel committee.