When I assign a velocity to a wall of cube I can see the necking and failure.
But when I use fix deform command even though if I set final strain to 5…!!! there is no necking.
Can I conclude that fix deform is for tiny strain values and estimates that there is no necking in tiny strain values.
Please take a look at the way I calculate stress vs strain.
fix 2 all deform 1 z vel .32 x volume y volume remap x
variable sigmavolume equal c_fz[3]vol10^-5
variable Lzero equal 32.4
variable strain equal (lz-v_Lzero)/v_Lzero
compute perfz all stress/atom pair
compute fz all reduce sum &
c_perfz[1] c_perfz[2] c_perfz[3] &
c_perfz[4] c_perfz[5] c_perfz[6]
The strain rate is an input to fix deform, so that's up
to you. Fix deform applies the strain homogenously
to the entire simulation box. Just applying a velocity
to atoms at the ends of a sample is different and typically
a more harsh change to the sample.
Reza, there have been many simulation studies of this sort of problem - that have looked into the details of what happens when you apply different deformation protocols. I’d suggest a Web of Science (or pick your favorite online paper-database) search, with various keywords, to look and see what others have done. That’ll give you a lot of guidance. It can be time consuming, but people do this sort of thing all the time, and probably more useful than asking the question on this list.