[ODE] I'm confused....

Gildas BAYARD gildas.bayard at lip6.fr
Wed Mar 5 06:54:02 2003


Le Mercredi 05 Mars 2003 06:37, whitt013 a écrit :
> I've been playing with dJointFeedback, trying to get breaking joints to
> working somewhat realistically, and I noticed something odd.  The amount of
> force a hinge exerts on body 1 is roughly negative (usually exactly) of
> what it exerts on body 2.  However, the amount of torque it exerts on each
> body is nowhere near.... Shouldn't a joint exert equal torques in opposite
> direction (especially in freefall)?  Isn't that one of Newton's laws?  Or
> is this one of the inaccuracies of ODE?  Or am I dumber than I thought in
> the world of physics?

Maybe I'm wrong but here is my feeling about that:
- the forces and torques are equal and opposite only when the two connected 
body don't move (neither rotate nor translate)
- In your hinge joint the sum of the torques from body1 on body2 and from 
body2 on body1 are not zero because the two bodies are rotating one respect 
to another and the torque difference gives the rotational acceleration
- I would use the magnitude of the smallest torque as a measure of the 
"stress". the extra torque between the small est and the largest torque is 
used to accelerate rotation and is not absorbed by the joint

Gildas

>
> So what would make the best measure of whether this joint has taken "too
> much stress"?  I was thinking something along the lines of the magnitude of
> the torque vector, but now kind of wonder about that... and which one to
> use.  Any input would be helpful.
>
> Thanks,
> David Whittaker
>
>
> _______________________________________________
> ODE mailing list
> ODE@q12.org
> http://q12.org/mailman/listinfo/ode