Control solver calculation time steps? [SOLVED]
Altair EmployeeHello,
For the given boundary conditions and material model, RADIOSS solves the F=KD matrix at certain time intervals and considers the displacements from previous step for current step.
But how do I control these time steps? I want to have custom time steps and not a constant frequency.
I want only 1 step for the linear region, and varying steps for the non linear region.
Ideally I would like to enter the time steps in a list or tabulated form.
I can do this in Creo Simulate, but cannot find in HyperMesh.
Please guide me how to do this in HyperMesh RADIOSS(block120) profile?
Answers
-
Can anyone please help?
0 -
Please refer below video for timestep in Radioss Explicit.
0 -
Can you tell what type of analysis you are doing and please elaborate on your analysis.
0 -
Rahul, the video was quite helpful.
If I understand, we can only define a constant time step by DT/NODA/CST for explicit analysis.
In Creo Simulate, I could input custom time for each step. This is very helpful because the linear portion does not need a fine time step. Only when deformation increases do you need a finer time step. I am surprised to not find this feature in RADIOSS!
George, I am doing drop test of an object. You must have seen the videos in another thread. Using Tetra elements, RWALL and Initial Velocity with Iterface7. I have been letting RADIOSS choose the time step but seem to not get proper convergence so wanted to define the time step on my own. But it appears that we can define only one frequency for the entire stress-strain curve. Quite shocking.
0 -
In RADIOSS we have a feature called as MULTIDOMAIN where this technique aims at optimizing performance of large scale RADIOSS models containing one sub-domain with significant time step discrepancy, often related to mesh refinement differences. In order to get more accurate results, users may use a fine mesh in the area of interest (high deformation location). Using this method it is possible to use a small time step for the fine meshed part and a large time step for the coarse meshed part.
But for a drop test imposing a time step using DT/NODA/CST is sufficient and Advanced Mass Scaling (AMS) is another method to impose a time step. AMS does not modify the global mass so that the global momentum of the related nodes is conserved.
The various methods to control the time step in RADIOSS is explained above. It is not possible to control the time step in a tabulated form in RADIOSS. And Creo Simulate does not support explicit analysis.
0 -
Thank you for your reply George!
Although I did not quite understand about the multidomain part, it sounds interesting!
What is AMS? Is there any video like the one for NODA?
If I understand correctly, breaking the simulation into time steps means explicit analysis, right?
If time is not taken into consideration then it is implicit analysis. By this way Simulate is also explicit. Correct me if I am wrong.
0 -
Advanced Mass Scaling (AMS) allows setting the time step to a higher value than the usual elementary or nodal time step to the whole model or a group of its parts and it does not modify the global mass of the system. No videos are there for AMS now, but very soon we will make it available. Similarly for MULTIDOMAIN also.
As for as i know Creo Simulate is not an explicit solver.
Yes, you are right about explicit analysis. As opposed to Implicit methods, explicit scheme is a function of time. The explicit method uses much smaller time steps since it is conditionally stable, meaning that the time step for the solution has to be less than a certain critical time step, which depends on the smallest element size and the material properties.
0