Use of differentials in antweights

Simon Windisch · Post by **Simon Windisch** » Wed Jun 01, 2005 11:24 pm

I agree Gary, as I said at the beginning, a little tricky to build.

The way Firestorm did it was to add extra motors to one of the drive shafts. The ring gear was held steady by the main drive motor. A much more elegant solution than the one I came up with!

Anyway, I've made a page with an explanation http://windisch.co.uk/electra/howto/und ... ntials.htm

It might remain vapourware, but it's worth thinking about.

Post by **BeligerAnt** » Fri Jun 10, 2005 9:25 pm

Having discussed this with my mech eng colleagues, it seems I understood better than I thought!
The concensus of opinion is that it is possible (Firestorm demonstrated that) but the BIG problem is powering the motor that is fixed to the diff cage (which rotates). It seems you need some slip-rings or similar and at that point it all gets a bit complicated.
The final nail in the coffin appears to be that the idea doesn't really scale to antweights. To be competitive you would need to achieve all this in about 20g. You would need two motors, plus gearboxes, and the diff and axles. One motor (plus gearbox) would have to be mounted on the diff cage and rotate!
Also, the steering motor is essentially fighting the drive motor. This puts a big load on each motor, so it would be easy to burn either motor out.

As an aside, I just wonder if it is possible to avoid the rotating motor by just driving the right stub-axle (in the photos) with a fixed motor, possibly via a clutch, and almost certainly with some clever differential electronics?

I'm just not sure it actually gains you anything over a standard tank drive system...

Remote-Controlled Dave · Fri Jun 10, 2005 9:48 pm

Sounds like too much hard work to me, specially for something that would just give you mobility you could get from tank steering...

Simon Windisch · Post by **Simon Windisch** » Sat Jun 11, 2005 7:57 pm

OK, very tricky then. Thanks for the info Gary,

The advantage is that whereas a tank style robot would need several rotations of the wheel motors to rotate itself, a diff robot would perhaps only need one rotation of the diff motor to rotate the robot once, at least that's how it seems to me.

Simon