Hello Jesse,
The external potential is normally a function of the positions of the
atoms. That means forces are applied to the atoms automatically without
having to do anything in user_bcon.
For example, the potential P for a cylindrical indenter of radius R is
given in section 2.11.4 of the QC Rerence Manual version 1.4 (top of p.
41). When a repatom moves within a distance R of the indenter center a
repulsive force will act on it pushing outward. The only thing that needs
to be defined is the initial location of the indenter in indent_orig(:).
This should be read in and set by the user_mesh routine. The
user_potential routine expects to find this as a global variable defined
in module mod_uservars.
The user_potential code lowers the indenter relative to the initial
condition by the current load property factor (GetLoadPropFact) which
depends on the value of "time" (incremented by the time macro in your
input file) and how you define your load property table with the "prop"
macro in your input file. This allows you to change the rate that you
lower (or raise) the indenter. You can of course also modify the
user_potential code to make the indenter move in other ways rather than
just up down.
Good luck!
Ellad Tadmor
On Wed, 27 Mar 2013, Jesse wrote:
> Dear professorᅵ ᅵ ᅵI'd like to ask how should I apply boundary conditions if
> I define an external potential energy by the routine user_potential?
> For example,in the QC reference manual,you have provided a user_potential
> code to model cylindrical indenters,but I have not
> make it clear that how the nodes X(j,i) under the indenter be constrained
> and move down. should I realize it in the user_bconᅵ
> routine? On the other hand, how should I assign to the variables like
> indent_stiffness, indent_radius defined in mod_uservars?
> should I assign it in input file or directly in the mod_uservars
> routine.Feel sorry I'm not good at it.
> ï¿œlooking forward to you reply!
> ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ
> ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵyours
> sincerelyᅵ
> ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ
> ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ ᅵ Jesse Fang
>
> --
> You received this message because you are subscribed to the Google Groups
> "qcmethod" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to
qcmethod+u...@googlegroups.com.
> For more options, visit
https://groups.google.com/groups/opt_out.
> ᅵ
> ᅵ
>
>