Dynamic Implicit Application

[alishst]

Hi

I'm trying to model a 3d liquid tank with acoustic elements and steel shell. I've also used springA elements and a solid mass to serve as a damper. The whole system is subject to an earthquake record. I would like to know which one of these suits this type of analysis better? Transient Fidelity, Moderate Dissipation, or Quasi-static? Or I should I stop using dynamic implicit and go with explicit?

 

[FEA way]

Do you have contact in your analysis ? If yes then moderate dissipation will be used. If not then transient fidelity application is appropriate. Quasi-static is a special type of analysis. It's more like static analysis but with inertia included. I don't think you would like to use it in this case.
Try with implicit first, explicit solver probably won't be necessary here.

 

[alishst]

Thanks for the reply FEA way

At the moment, there's no contact but might include it later. I am using transient fidelity currently but the issue is there seems to be some high frequency noises added to the outputs I receive. For example I feel there are a lot of (unnecessary) sharp peaks and valleys in the base shear response time history (same problem with many other outputs as well)

 

[FEA way]

Increased damping may help you eliminate noises. Also make sure the time increment size is correct.

 

[alishst]

Dear FEA way,

I tried damping with transient fidelity (brown daigram) and as you can see, there are still plenty of high-frequency, resonant noises in the model. The response for quasi-static (blue diagram) looks more sensible but I'm pretty sure it is flawed as well. What do you think is causing the problem? Would you recommend using explicit solvers here?

 

[FEA way]

Moderate dissipation application may be better then, it should reduce noise. Enhanced hourglass control can help too (it can be enabled with *SECTION CONTROLS keyword). If you can't eliminate noise from the analysis you can still reduce it in postprocessing using available filters.

When it comes to the use of Explicit instead of Implicit, I don't think it can help. Noise is usually even worse in Explicit.