Thanks, broli. Added to my "favorites"!
The green ones are mostly about the Tolchin devices, in several incarnations. Some of the later ones (the domed, beetle-like devices) are the work of Gennady Shipov and his collaborators, carrying on the Tolchin work.
The air track and table demos are impressive. But unfortunately, to my knowledge, these devices still fail the "pendulum test". The center of mass, when carefully tracked, still averages out over time to be motionless. The devices cannot sustain a constant deflection of the center of mass when suspended as pendulum.
I met Shipov in 2000 or 2001, and tested a small mechanical Tolchin device that he brought over. It had three wheels and the paired offset weights, driven by clockwork. It would lurch along nicely, and Shipov claimed that it would do the same on an air table. I didn't have an air table at that time, so I borrowed a machinist's surface plate, leveled it precisely, put a couple dozen marbles on top, and a piece of plate glass on top of that, with the plate glass weighted to the same weight as the Tolchin device. Then I put the Tolchin machine on the glass and started it up.
My reasoning was this: If the device worked, as claimed, by interacting with "torsion fields" and was truly a reactionless inertial propulsion drive, it would go happily along on the plate glass and the glass would simply sit there motionless. On the other hand, if the device worked according to Newton's laws of motion, that is, normally and not as a reactionless and/or inertial propulsion drive, the Tolchin machine should go one way and the glass the other, with the Center of Mass of the system formed by the glass and the device remaining, within experimental error, motionless.
It's amazing how angry people can get over a simple control experiment, performed right before their eyeballs with ten dollars worth of apparatus.