viscoelasticity

[Muzialis]

Hello,

I am working on a polymeric component which has been modellled using linear viscoelasticity, more specifically the Wiechert (multiple Maxwell ?) model.
The object's shape is quite simple so a closed form solution is obtainable for the elastic stresses: passing to the Laplace plane the strain hystory is rather complicated so I had to solve the equations numerically. The problem is that the solution does not converge, after the expected transient, to a stable pattern.
I am pretty sure a system made only of springs and dampeners under a periodic strain hystory must converge so I can only blame the numecial scheme, which is quite accurate though.

Does anybody had similar problems or have any interesting advice?

Thank you so much.

Muzialis

[Jorgen]

A quick comment: certain sets of material parameters can cause your set of ordinary differential equations (ODEs) to become stiff. And as you likely know, stiff ODEs can create numerical problems if not an appropriate solver is used.

- Jorgen

[Muzialis]

Jorgen,

no doubt you are right, I tried again excluding the fastest decaying elements in the Wiechert model and everything is fine: I will implement a proper solver and I do think the problem will be sorted out.

Thank you for the hint.


Muzialis