Integration scheme in UMAT: implicit versus explicit

[bvrm22]

Hi,

I was wondering about the following:
When implementing a simple visco-elastic material model in the UMAT subroutinge, the Abaqus subroutine manual uses the centralized integration scheme to discrtetize the ordinary differential equation. This is actually an explicit integration scheme which is applied.

Now I was wondering if it is worthed to implement a IMPLICIT integration scheme to calculate the updated stress in the subroutine. I already use dynamic implicit solving for the general simulation process.

If implicit solving of the ode to find the stress updates is worthed, can someone help me to point a implementation of eg. backward euler with newton iteration in Fortran code?

Thanks in advance.

Kind regards
bvrm

 

[IceBreakerSours]

I have little experience with UMATs, if any. However, UMATs are supposed to be used with the Implicit solver (Standard), so I strongly feel you may have misinterpreted something along the way. VUMATs, on the other hand, are supposed to be used with the Explicit solver.

Are you new to this forum? If so, please read these FAQ:

http://www.eng-tips.com/faqs.cfm?fid=376

http://www.eng-tips.com/faqs.cfm?fid=1083

 

[bvrm22]

Thank you Icebreaker. I guess iterative solving of the material equation is only needed when working with plastic strains. The centralized differential scheme to discretize the ode is probably just chosen because of second order accuracy. (?)

So when using an implicit dynamic solving procedure, the materialmodel will also be solved implicit.

I tried to implement a backward euler differential scheme, but obtained an explicit formula in terms of sigma_n+1 or the stress increment. So I guess it is not needed to worry about this.

Kind regards,
bvrm22