Hello, community! I am solving a coupled hygroviscoelastic problem using mixed elements of the Taylor-Hood type. As part of the analysis, I decided to test tetrahedral elements with different iterative methods and preconditioners. I observed that the results obtained showed insignificant differences. However, when repeating the same procedure with hexahedral elements, the results exhibited significant variations.
I also validated this behavior using other published codes based on Dolfinx and obtained the same result. It is worth noting that when using the same interpolation order for hexahedral elements as for tetrahedral elements, an identical response is obtained, which is ideal. However, this configuration does not comply with a recommended stability law.
Does anyone have insights into what might be causing this behavior?