Learning Forum

[dan_123]

Hi

My model is a cantilever cylinder with a time-dependent force acting in one direction. This force is within the linear elastic range of the material, so everything should be linear. I am using the transient structural module to solve this model. The problem is that ANSYS takes a lot of time to solve it. With 40 time-step loads the model takes a couple of hours to finish. The problem is that the time-history of the force applied is composed by thousand of points. I don't understand why is taking that long since this model is not complicated at all.

 

Thank you,

[peteroznewman]

Hi Dan,

Transient Structural models take a long time to solve, so you want to do as much work in Static Structural before you switch to Transient.

Please reply with a graph of the Force vs. Time.  If you calculate an FFT on that Force-Time graph, what kind of frequencies are in that data?

Have you done a Modal analysis on your structure?  What kind of frequencies are in the first few modes?

If the frequencies in the Force-Time graph are much, much lower than the first natural frequency of the structure, that's an indication there will not be much difference between a Static solution and a Transient solution.

If you need help calculating an FFT on your Force-Time graph, you can put the data in a zip file and use the Attach button to upload the zip file to your reply, after you post, when the Attach button appears.

Even if you need a Transient, there is a much faster way for linear models, and that is the MSUP Transient. You can learn more about that by reading the Transient Dynamics section of the ANSYS Help

[dan_123]

Hi Peter,

 

Thank you! The MSUP transient is incredible faster than the full transient analysis. I am doing a sensitivity analysis regarding the number of frequencies considered in the model.