Soils, Transient consolidation, Time period

[Elliot19]

Good Afternoon,
I'm doing simulations on a knee joint model (tibial cartilages, femoral cartilage, menisci and ligaments - first-order hexahedral elements) which includes an initial contact phase guided by an imposed displacement (Soils, Steady-state), in order to make the various structures come into contact, followed by a step that involves the application of a ramp force (Soils, Steady-state), followed by a "Soils, Transient consolidation" step, in which a ramp force is applied. Could someone explain to me how to set the value of the "Time period" in the correct way and by which parameters it is influenced? Setting a value = 2, the simulation converges, while setting it = 1.9, the simulation stops during the initial part of the last step. It is not clear to me what this depends on and based on what I have to choose one value rather than another.

 

[FEA way]

In steady-state analyses time is just an artificial measure used for load incrementation. However, in case of transient analyses it’s a physical time and thus you should specify the value that corresponds to the actual time duration of the physical process being simulated.

To handle convergence issues try automatic stabilization available in step settings.

 

[Elliot19]

Thanks for your answer. The Time period that I should set, so that it reflects the real physical time, is much lower than the value that brings the simulation to convergence: I should set about 0.125, while the simulation already with 1.9 is not completed.
Adding an automatic stabilization, with a damping factor = 0.02, the simulation goes to convergence also with Time period = 1.5, but we are still far from the desired value.
- To reach convergence with the desired time period, what can I do? Is it correct to increase the damping factor a lot, or do I risk having a wrong simulation?
- What other factors can bring the simulation to convergence?
- Could the mesh be too rough? (I obtained it with a Matlab code, which creates elements with a side of about 1.5 mm)

 

[FEA way]

In some cases you can use larger time period and simply save results for earlier frame.

Increasing the stabilization damping factor too much may influence the results. You should always compare ALLSD and ALLIE energies to make sure that this artificial damping is not too high.

Indeed, the mesh may have impact on convergence. Try refining it.