当前时间步长为零！！

liuzhilin

自己编的umat,通过写出子程序中个参数的值，发现程序第一步传入子程序的dt1=0，各应变值为零。子程序只被调用一次，求解释！

xth07

碰到楼主的问题了，请问楼主解决了吗，谢谢

xth07

我的问题貌似解决了。出现dt1=0很大可能是因为UMAT本身的算法出错，导致计算结果不合理。应检查算法。

xth07

上面的回答貌似不对。
我新写一个UMAT时还是遇到这个问题，神奇的是，我把UMAT开头
      SUBROUTINE UMAT42 (cm,eps,sig,epsp,hsv,dt1,capa,etype,tt,temper,
    . faile1,crv)
改为
      SUBROUTINE UMAT42 (cm,eps,sig,hsv,dt1,capa,etype,tt,crv)
之后，dt1和tt输出就不再为0.
具体原因我也不清楚，共享出来让遇到同样问题的童鞋们参考，也希望有大神出来解释解释。

gdyu_yu

umat42最好别用。

xth07

gdyu_yu 发表于 2013-10-2 00:27
umat42最好别用。

请问为什么呢

gdyu_yu

xth07 发表于 2013-10-2 00:53
请问为什么呢

Question:

Is there a difference in UMAT41-UMAT50? Is the routine name of UMAT only connected with MT in MAT_USER_DEFINES_MATERIAL_MODEL？


Answer:

Avoid using umat42 and the vectorized routines umat48v and umat49v -- these are special,  undocumented subroutines that do not reside in dyn21 and thus the user cannot modify them.
Aside from those exceptions, it doesn't matter which MT (from 41 to 50) you choose.  

The umat subroutine used is controlled directly by MT and by IVECT.  If IVECT is set to 1, the "v" (vectorized) version of the umat subroutine will be called, e.g., subroutine umat41v is called if MT=41 and IVECT=1.