Jhula principle - increase speed by applying brakes.

[xnonix]

Hartiberlin the air resistance is set  at 0.7 kg/ms in the wagon version.
Cheers.

[prajna]

Well, that blows my spring theory (ha, ha)... I just deleted the spring from the mini model and it still works!  This is back to the simple Jhula model I began with!

I really don't know why it works.  I have attached an image with the pendulum centre of mass tracked, which may help to shed some light on it.  I wouldn't put it past being an artifact of WM2D (despite its $3000 price tag it does have errors, particularly in handling polygons).  Accelerating to destruction is something that I really don't trust since the pin joint on the pendulum 'drifts' and I am pretty sure that indicates a WM2D error.  I guess building a physical model is the only way to test it but it takes a bit of tuning to get it to run well.

[xnonix]

hi all,
I have just discovered what is going on with this model. We all know this doesn't work in reality but we need to know why. Well the thing is called geometry. The problem is the L shaped piece. When you change it for another made by 2 rectangles joined then you get the correct result.
It STOPS faster than the original, coz you have more contact surface with air. Again bad news.

Cheers,
xnonix

[xnonix]

We have to study this model prajna made inside a big circle.

[prajna]

I didn't get positive results with that, xnonix (except when there was no air resistance load).  I designed it that way because I suspected the (polygon) angle bracket was causing the effect. Damn, I wish WM2D would sort their polygons.

I am still not convinced that the design will not work in reality; there is a possibility that a spring could bias the torque to one side.  Worth pursuing, I think.