Bill Muller has shown it too !

[pinobot]

The balls are pulled away from the magnet by gravity, it's best to put the magnet where gravity has the most influence, halfway down.
The only thing that can spoil your day is the fact that the energy of the magnet-system (magnet+balls) must be greater than the Potential gravitational Energy that would be produced when you let the top ball fall the distance top ball - botom ball.
The Potential Gravitational Energy is easily calculated, the energy of the magnet-system (magnet+balls) is an other matter.
http://hyperphysics.phy-astr.gsu.edu/hbase/gpot.html#mgh

Btw, if it works, run like hell !

[hartiberlin]

Pinobot, I did order today a few steelballs via Ebay.
When they come in, I will try it too.
Tommorow I go out buying a plastic hose for the balls to  go in there.

I still have 6 or 8 neodymn magnets here, which I will first try it with.

Regards, Stefan.

[pinobot]

The test is simple, first you drop one ball in on the left without the magnet and balls in place, and record the hight of the ball on the right, that way you know how much friction there is (if the ball for instance goes up 90% on the right you have 10% friction loss).
Then you put in the magnet and balls in place and drop the ball on the left, if the ball goes higher on the right than the first time you've gained energy. If the ball jumps out of the tube you have OU for sure.

[hartiberlin]

Pinobot,
reminds my of my old SBMOT.
See:
http://www.keelynet.com/bedmot/sbmot01.htm

I unfortunately never built it.
But now it is time to do so...
Also you would need more than 1 steel ball at each end and in the center.
so go with at least 2 balls in each place, then it should work.

Regards. Stefan.

[pinobot]

I've tried that configuration, it's a normal Gauss Rifle but i couldn't get i close looped.