Uhyper and Prony

[Bardi]

Dear all

I have implemented in ABAQUS a user defined Hyperelastic function using Uhyper. The subroutine works well in the rate indipendent case.
Then I coupled it with viscoelasticity, specifying the Prony terms.
The Prony series for the purely linear case have been specified correctly, i proved it matching the abaqus curves with analytical ones (The shear relaxation modulus is a know function which I approximate with prony parameters).
The Problem appears when I use my function wich takes into account non linear elastic property via a parameter n.
For high values of n the analytical solution and the Abaqus results are differs a lot.
My question is: How the time dependency works on the User defined Hyperelastic function?

In the manuals it appears that the strain energy function is multiplied by the variation of the normalized shear relaxation modulus.
I have a relationship for calculate the stresse from the strain energy function. Do this variation remain the same for strain energy function and then for stresses?

I specified my Uhyper with function, properties and the derivatives as in ABAQUS Manual.

Thank you

Carlo

[Bardi]

should the strain energy equation be written so that the strain dependent part and the time dependent part are separated????

Carlo

[Jorgen]

It sounds like you have written a UHYPER subroutine that you would like to use together with a linear viscoelastic model specified by a set of Prony parameters.

That approach should work just fine - as long as your UHYPER subroutine is correctly implemented. There should be no need to consider time-dependent effects from within the UHYPER subroutine.

- Jorgen

[Bardi]

[Bardi]

Dear Jorgen

I found the error. it was a numerical error

thank you again and for this forum


Carlo