the modified DPC model is already implemented in abaqus.
as you know, the DPC parameters are density dependent. thus you need to write a usdfld subroutine to calculate the relative density of each element as a field variable in each time increment. also you have to provide the parameters in different relative densities in property module. when you do that, the field variable (relative density) would be updated in each increment and new parameters would be used in DPC constitutive law.
Thanks Majid Moghaddam. At first, it is obvious to put the wanted parameters desired in DPC model... As in metal powder compaction, i am very much concern with the orientation of density distribution which the uniform density in the compacted part is the ultimate goal of the powder metallurgy research. So, is it recommended to use usdfld first before writing the vumat subroutine?
Thank you.
Why would you want to write a vumat? the uniformity of density distribution depends on the loading and boundary conditions. for example the geometry of the die, strain rate or etc. it has nothing to do with the subroutine. the relative density is a function of volumetric inelastic stain. therefore to achieve a uniform relative density, the magnitude of volumetric stain must be relatively similar in different elements. nevertheless, the elements near the corners of the die would have greater relative densities in most cases.
- Badges
- Science topic