Discussion Forum

Hi

if its the gradient of the stress you are looking for (?) you could try with a d(,x) ... derivative operator, but I' havent tried it out myself, wih respect to the stress tensor that makes quite some derivatives, no ?
--
Good luck
Ivar

Thanks for the reply, Ivar

I don't look for gradient of stress, i look for time derivative - how fast does it change with time. Just like the strain rate, but for stress. I need it to calculate some complex values to evaluate attenuation. But this is not the point. I know the value of function in each element at each time step, I am looking for rate, should I try to define it like analogy for space steps d(,t)? And I am very far from being expert in COMSOL, so I want to specify, where should I try to define this in COMSOL's menu (3.5a).

Thanks a lot.

Hi

via matalb you have full control of what you do by the fact you access the time step data and you treat it at will.
But within COMSOL there is quite alot too, specially in V4.

I would propose that you try it out on a simple 2D canteliever, while aplying a bending force in a transient analysis, and then test out a few ways to display the data, and to postprocess it.

There have been a few threads on time derivatives too (search the Forum) these might help, I have a doubt if the d() operator is not limited to spatial variables (?) cannot tll by heart

--
Good luck
Ivar

I will consider both possibilities.

Thank you very much for help, Ivar.