It's not only possible, it actually happens - at least something very similar. Motorcycles attempting to set speed records on the Utah salt flats have experienced failure of the inflation valve of a tire due to the high centrifugal force. But as this occurs at high speed the centrifugal force keeps the tire "inflated"; so there's no problem until the rider slows down at the other end of the traps. Brent On 3/22/2015 12:56 PM, Henry Baker wrote:
(This thought experiment was inspired by space shuttles landing at 200+ knots, and having electric motors in the wheels that brought the landing gear wheels up to landing speed prior to actual touch-down, so that the landing would not generate any flat spots on the landing gear tires as a result of the landing.)
The following thought experiment could have been performed in Newton's time, but I doubt anyone would have considered it until after the steam locomotive was invented.
Consider a spoked wheel in which the vast majority of the mass is located in the rim; i.e., the mass of the spokes are insignificant relative to the mass of the rim. Furthermore, although the rim is heavy, it is also elastic and can stretch.
Assume the wheel is moving on a perfectly flat surface with a uniform gravitational field perpendicular to the surface.
Assume the wheel is moving in a straight line; thus the 3rd dimension in the problem can be ignored.
Assume that there is a load on the axle in the center of the wheel. If the wheel is stationary, this load is distributed by the spokes in such a way that the spokes on the bottom are in compression and the spokes at the top are in tension. Since the wheel rim is elastic, the rim itself will be deformed slightly, but we will ignore this deformation for a moment.
Now turn the wheel and have it roll along the surface without slipping at a pretty decent velocity.
At some speed, the mass of the wheel rim will provide enough centrifugal force to counteract the compression force on the lower spokes, so that all of the spokes will be in tension, although the spokes at the top of the wheel will be in greater tension than the spokes at the bottom.
Now that we are going fast enough that all of the spokes are under tension, we now allow the rim to disintegrate into N equal pieces, where N is the number of spokes. Each of the rim pieces becomes a point mass having 1/N of the total rim mass and still attached to its spoke. Let us further assume that there are small wires holding the pieces in a roughly circular shape, but these small wires carry a negligible load, and all of the load in the axle is now being carried by the impulses of the pieces of the rim hitting the "road". Another way to model this would be to have an N-spoked wheel without any rim at all, but the mass of each spoke is concentrated at the periphery; we also assume that each such spoke is constrained to stay within its "ray" from the center of the wheel--i.e., the weights of each spoke slide along its spoke ray, but are constrained to not go beyond a certain radius R from the center of the "wheel".
Of course, there are all sorts of 2nd & 3rd order effects, but is there a scenario under which such a "dis-integrated wheel" would work ? I.e., actually carry a load so long as the speed was higher than some critical speed.
_______________________________________________ math-fun mailing list math-fun@mailman.xmission.com https://mailman.xmission.com/cgi-bin/mailman/listinfo/math-fun