Perpetual motion: you work it out ...

[Tim+]

So you believe in perpetual motion??

Tim

[Chris Hogg]

I have one. Needs sunlight to get it going. IR might also work, but never tried it. Focus the sunshine onto the paddles with a hand-lens and it goes like the clappers!

[Martin Brown]

Most consumer grade ones spin around the wrong way!

Radiation pressure of photons is much weaker than the air pressure imbalance created between the black and silver sides of the vanes that occurs in a poor vacuum (the vanes are thicker and insulated).

You have to really work to get a proper hard vacuum where the recoil of photons off the mirror side causes it to spin in the right direction. A true Crooke's radiometer will work with thin foil and good hard vacuum.

[Fredxx]

says it only works in a partial vacuum and cites that vanes are motionless in a hard vacuum.

Are you getting confused with the Nichols Radiometer?

[Paul]

There's no sign of torque ripple.

it would need sinusoidal drive waveforms, to make it that smooth.

A camera with a global shutter, would be able to make frames shot of the wheel, stand still. You would need better quality video than a Youtube one, to make measurements.

With a sensor part number in hand, you can then see what a camera with one of those would cost.

More video cameras have rolling shutters. But a few have global shutters and are better for motion analysis. The fastest I could find is 1000 FPS global, but the resolution on that is pretty low.

If you look at the rotor on a brushless DC motor or on a stepper motor, you get some idea what torque ripple looks/sounds like. I had a hard drive in the lab, way back when, it was an 8" HDD (single platter) and some fool designed it with a stepper for the spindle motor. It made quite a distinctive noise, and you could hear that noise outside the lab doorway as it accelerated. The company making that, went out of business a year later.

If you wanted to pull tricks, it would be easier to do in the hub of the motor, than way out in space like that.

If the bearing losses are low, you would not need much of a kick to keep it running. But with careful measurement, you should be able to spot that happening. Even a sensitive microphone or vibration sensor attached, could give you info.

Paul

[wrights...]

That isn't implied.

Bill

[wrights...]

What's that weird tattoo on the bloke's arm? Bill

[Ridalo]

A metric and imperial ruler, I think. Ridalo

[Rod Speed]

What weird tattoo ?

You should do a

the actual time for the t data. You right click on the video and use the copy url at the current time entry.