[gmx-users] magnetic field - segmentation fault

[Man Hoang Viet]

Hi Maryam Kowsar,

I have implemented external magnetic field to GROMACS and tested it. The
code ran well and the test result is consistent with experiment for simple
case (say Na+ in vacuum and without t_coupling). I may guest why your code
got the error. But I am sorry that I can not share my code to anyone now
(but I will do share it to this forum in near future).
However, there is an important issue that MD simulation with t_coupling
always does re-scale velocity of all atoms to keep the given temperature
and it strongly effect on the Lorentz Force of the external magnetic
field. Otherwise, If the t_coupling is not applied, the system temperature
will increase forever. Therefore, I wonder whether we really can estimate
effect of the magnetic field on protein (or other target) in simulation?


>On 23 Jul 2015, at 11:52, Maryam Kowsar <maryam.kowsar at gmail.com> wrote:

> Dear users,
>
> Iam trying to add magnetic field code to gromacs. I think I did all the
> modifications necessary in all source codes but when I use mdrun command
> it
> gives me "segmentation fault" error while in md.log the magnetic field is
> present. When I change tpx_version the error is gone but in md.log no
> magnetic field is implemented. How can I overcome segmentation fault
> error?  Should tpx_version be changed after adding a code?
>
> Thanks!

Man Hoang Viet
Institute of Physics,
Polish Academy of Sciences.
Al. Lotnikow 32/46
02-668 Warsaw, Poland.