|From:||"Thompson, Aidan" <athomps@...3...>|
|Date:||Sun, 18 Jun 2017 23:21:29 +0000|
I agree with Matthias' analysis. I was able to modify fix efield to recover P=0 by adding code to optionally add in the virial contribution from fix efield which exactly cancels the contribution from fix shake. This should be released shortly. The reason this will not be the default is that the decision to add or not add the virial contribution from a fix must be handled on a case by case basis, and so is best left up to the user.
Aidan P. Thompson
01444 Multiscale Science
Sandia National Laboratories
PO Box 5800, MS 1322 Phone: 505-844-9702
Albuquerque, NM 87185 Fax : 505-845-7442
E-mail:athomps@...3... Cell : 505-218-1011
A couple things you can try.
a) use flexible molecules instead of SHAKE, see if the results are different
b) define a compute pressure command that includes keywords that
do not include the contribution from fixes (fix shake in this case),
to see if a pressure w/out that term is what you are looking for
note that you can make the pressure from that compute you define
be output with thermo, or even replace the default value that "press"
reports (see the thermo_modify press command)
I'm CCing Aidan, as he may have better comments. He recently worked
on options to enable the forces from walls to be included in the virial/pressure,
so that is another option.
On Fri, Jun 16, 2017 at 1:31 AM, Matthias Kahk <matthiaskahk@...24...> wrote: