Magical Mystery Globe

Here's how I think it works:
An inner globe floats on a liquid to reduce friction. Fixed inside the inner globe is a permanent magnet, which keeps the inner globe aligned in a fixed direction (north). Inside the outer globe (or outside the inner globe) are one or more solar cells connected to one or more small, geared electric motors which turn the outer globe relative to the inner globe - possibly via a friction drive against the outer globe.
I think a big problem would be overcoming the friction of the three bearings (the stand) which support the whole setup, and must allow it to rotate.
Having played with Crookes radiometers, I doubt that you could get one to do any useful work, since they must operate in a partial vacuum and generate very litte torque.

Upon further consideration, I think the method of propelling the outer globe is through some sort of electromagnetism, with solar cell powered electromagnets positioned on the inner and outer globes attracting each other "maglev syle" and imparting a rotary motion to the globes. The problem I see with this is interference with the permanent "north" magnet.

The clear outside shell sits on the three-pronged base. There is no apparent movement between the outside shell and the base. There are no visible devices between the outside shell and the inside shell. From this I conclude that whatever makes the inner globe spin is located completely within the inner globe.

There is no need for a magnet to maintain “north” orientation. The north pole on the inner sphere is on top – the axis is vertical. This can be achieved by simply weighting the inner globe so that it’s heaviest on the bottom, but has overall neutral buoyancy. As neutral buoyancy is not easily achieved, the inner globe, as it spins, will rub on the outer globe at either the top or the bottom. A small raised bump at top or bottom will provide a small contact area, reducing friction. I have not seen this. Maybe it’s there, maybe not.

If the paint on the inner shell is somewhat transparent, light can get inside. Given this, what makes it go? My guess is that some sort of solar panel drives a rotating device, and as the device rotates, the inner shell rotates in the opposite direction as a reaction force.

As I know nothing about the power capabilities of small solar panels, I can’t go any farther.

I just looked at the globe web site again, and saw some things that tell me I haven't got it quite right.

Read US Patent US6937125 which will tell you more than you probably want to know about the options for implementing this.

Another sneaky option occurs to me which would not depend on the internal solar cells: have a fixed bar magnet inside the hydrostatically floating sphere and then under the table on which it stands generate a suitable rotating magnetic field from a hidden electrical outlet.

Many thanks, Mike!

The thing is certainly ingenious, and a more involved design that I would have guessed. That probably explains the selling price.

quote:

[u]Originally posted by Allen Jones:
Here's how I think it works:
An inner globe floats on a liquid to reduce friction. Fixed inside the inner globe is a permanent magnet, which keeps the inner globe aligned in a fixed direction (north). Inside the outer globe (or outside the inner globe) are one or more solar panels connected to one or more small, geared electric motors which turn the outer globe relative to the inner globe - possibly via a friction drive against the outer globe.
I think a big problem would be overcoming the friction of the three bearings (the stand) which support the whole setup, and must allow it to rotate.
Having played with Crookes radiometers, I doubt that you could get one to do any useful work, since they must operate in a partial vacuum and generate very litte torque.[/u]

I think you are right..I will try out the experiment and will share out the results soon..