Classic crank science: an antigravity drive

XKCD
xkcd c226: swingset

A friend of mine just asked me to comment on a piece recently recommended to him by Google Alert on how to build an "antigravity drive"…

Here's an excerpt:

Chapter 4: How to Build a Flying Saucer

If directed upward, centrifugal force can be used to drive an antigravity engine. The problem engineers have been unable to solve is that centrifugal force …

The full article is here. Unfortunately, the physics of the article makes no sense.

Here's a simple paragraph that's key to understanding the borkenness.

A bicycle wheel is a flywheel. If you remove the rim and tire, leaving only the spokes sticking out from the hub, you still have a flywheel. In fact, spokes alone make a more efficient flywheel than the complete wheel; this is because momentum goes up only in proportion to mass but with the square of speed. Spokes are made of drawn steel with extreme tensile strength, so spokes alone can generate the highest levels of centrifugal force long after the rim and tire have disintegrated.

Wrong. The centrifugal force goes up in proportion to the square of the speed, and to the mass times the distance of the mass from the origin. This is why people make big flywheels with light spokes and heavy rims: they want to be able to go slower to avoid blowing up the flywheel. This is Newtonian physics 101 stuff, so you shouldn't believe anything this guy says after this howler.

But spokes alone still generate centrifugal force equally in all directions from the plane of rotation. All you have to do to concentrate centrifugal force in one direction is remove all the spokes but one. That one spoke still functions as a flywheel, even though it is not a wheel any longer.

I have no idea what this means. Consider this guy's perfectly good one-spoke flywheel. Apparently it "concentrates centrifugal force in one direction". But which one? Over one rotation of the flywheel, it spends an equal amount of time pointing in every direction. Seems pretty symmetric.

The physics to show that rotating unevenly doesn't help is only slightly less obvious. The main thing to notice here is that a flywheel really only resists motion that twists it off its plane of rotation. This is a consequence of Newton's Second Law—bodies in motion want to keep moving in straight lines—and of the fact that the coordinates of motion are independent. In the plane of rotation of the flywheel, the central force of the axle cancels the centrifugal force of the atoms perfectly, and you can move the flywheel freely in that plane without feeling any force. Perpendicular to the plance of rotation, the atoms are effectively stationary, and so they behave just like the flywheel wasn't rotating. Try to turn it any other way, and you make the moving atoms bend their flight path: this will require force.

In short, this is classic crank science. A guy who doesn't even understand Physics 101 is going to tell us how flying saucers work. The sad part is that someone apparently found it worth my friend's attention. (B)