Forces varying with time?

[bongirs]

Hello,

I am trying to learn Creo Simulate 2.0 and having a hard time applying time varying forces.
I want to have forces something like this:
Force1: 50N from 0 to 0.5sec
Force1: 100N from 0.5 to 1.2sec
Force2: linearly increasing from 200 to 250N from 1 to 2 sec

Simulate does not show any information on how the forces are applied over the timeline of the analysis for linear as well as non-linear analysis.
Am I missing something or Simulate does not have this functionality?

 

[brep]

Simulate can certainly handle time varying loads, but you have to ask it to do so in the correct context. To start with, a linear static analysis is not the right context. But maybe you want a linear dynamic analysis, which Simulate can handle. But given the slow application of loads, I don't believe you want a dynamics simulation. This leaves nonlinear: you can certainly do a large displacement analysis (with or without contact) with time varying loads. You can specify the time varying load function as a lookup table, an algebraic function etc. Make sure you request the output at multiple points along the way (this is done in the analysis setup dialog box) For this instance, you dont just want to accept the defaults

 

[bongirs]

Yeah I found the function technique for time varying loads for non-linear analysis. But for linear static structural analysis I cannot find the ability to vary the loads wrt time. ANSYS has this functionality, but I wonder why Simulate does not?

Also the non-linear solver hardly converges for me.
Initially I tried making the mesh finer, this resulted in getting lower values for the residual. However at the last step it used to fail to converge.

I also tried taking smaller time steps like 4.0, 4.1, 4.2, etc. but it used to take a long time and then fail. So I started taking much smaller time steps like 4.12, 4.14, 4.16, etc. This seemed to improve the residual but the solution could not converge.

Can you tell me which parameters I should fiddle with so that the solution converges?

 

[brep]

There's a check box near the top of the Analysis creation dialog. It is labeled "Nonlinear / use load histories"...
You can then choose whether you want to use the large deformations (nonlinear geometry) solver or not. You'll notice now that the loads have a time dependence button next to them.

For a true linear *static* analysis (or linear perturbation analysis - I used to work at HKS/Abaqus so forgive me for not knowing the ANSYS terminology) there is no "t" term, so functions of time to vary loads make no sense.

 

[brep]

And I meant to add that when selecting "nonlinear / use load histories" you'll need to choose nonlinear deformations for the dialog to accept the analysis. As I stated previously, using load histories for a linear static analysis is meaningless and Simulate will (correctly) prevent you from doing so.