OpenRadioss 3DNLG-1_Elastic_large_deflection

disla

Hello and Happy new year 2025 everybody.

I’m using known benchmarks to get a taste of OpenRadioss.

In this case 3DNLG-1_Elastic_large_deflection_response_of_a_Z-shaped_cantilever_under_an_end_load.

I’m not sure if I didn't configure something right but regardless of the results accuracy which I’m not comparing now, I’m not obtaining the complete time Period on the results bar.
Monitor reports normal termination but I requested 1 second overall analysis time and I can only see 0.769 seconds.
By other hand, I requested solutions at each 0.1s starting at 0. When I move the slide bar , the solutions seems to be reported at 0.05s, 0.15s, 0.25s. Seems like offset.

Thanks

Victor

I'm not seeing the 0.05 s offset you described - maybe it's just when the display changes as you slide the slider?

However, after 0.7 s, the time values in the solution deviate from 0.1 s increments. That's because the the solution time steps are really the nearest internal OpenRadioss time step, and internal time step size is usually tiny so there's hardly any difference. However, at about 0.7 s, the internal time step grows huge (Solution -> Mechanical misc. -> time step). Something goes wrong there with the loss of displacements too.

With solution every 0.01 s, you can see the problem happening more clearly, but I don't know what's causing it.


Density is unphysically large. Could that be it?

disla

Thanks Victor.

I solved the problem properly with ccx nonlinear. It solves fine except for a slightly large bending moment at point A.
My surprise was with Openradioss. I changed some parameters (Density ,K, F) to see if there was some automatic mass scaling activation and how stable solver was.
It's a new solver and it surprised me to see the green results window with the message like ended fine but then later found the results have disappeared beyond 0.7s. I thought unconditionally stable meant the analysis ends no matter what. Seems I was wrong.