extra parameter needed

Hardening parameters determine rate of hardening in phenopowerlaw.
This flexibility is also needed in J2.
This commit is contained in:
Martin Diehl 2021-06-21 23:22:51 +02:00
parent 27431bd60c
commit 4b82cb4c14
1 changed files with 3 additions and 1 deletions

View File

@ -15,6 +15,7 @@ submodule(phase:plastic) isotropic
dot_gamma_0, & !< reference strain rate
n, & !< stress exponent
h_0, &
h, & !< hardening pre-factor
h_ln, &
xi_inf, & !< maximum critical stress
a, &
@ -99,6 +100,7 @@ module function plastic_isotropic_init() result(myPlasticity)
prm%dot_gamma_0 = pl%get_asFloat('dot_gamma_0')
prm%n = pl%get_asFloat('n')
prm%h_0 = pl%get_asFloat('h_0')
prm%h = pl%get_asFloat('h', defaultVal=3.0_pReal) ! match for fcc random polycrystal
prm%M = pl%get_asFloat('M')
prm%h_ln = pl%get_asFloat('h_ln', defaultVal=0.0_pReal)
prm%c_1 = pl%get_asFloat('c_1', defaultVal=0.0_pReal)
@ -280,7 +282,7 @@ module subroutine isotropic_dotState(Mp,ph,en)
endif
dot%xi(en) = dot_gamma &
* ( prm%h_0 + prm%h_ln * log(dot_gamma) ) &
* abs( 1.0_pReal - stt%xi(en)/xi_inf_star )**prm%a &
* abs( 1.0_pReal - stt%xi(en)/xi_inf_star )**prm%a *prm%h &
* sign(1.0_pReal, 1.0_pReal - stt%xi(en)/xi_inf_star)
else
dot%xi(en) = 0.0_pReal