[gmx-users] SMD - Pulling Error in mdrun 0.49 times box size

[Agnivo Gosai]

Dear Users

I have put my molecule complex in a box having dimensions 7.5 7.5 15 . The
center of the molecule is fixed as 3.75 3.75 and 4.0.

I use pull geometry = distance.

I did the following SMDs keeping pull force constant unchanged :

1. 500 ps SMD with pull rate = 0.01 nm / ps. mdrun successfully completed.
nsteps = 250000 dt = 0.002

2. 1 ns SMD with pull rate = 0.005 nm / ps. mdrun successfully completed
nsteps = 500000 dt = 0.002

3. 5 ns SMD with pull rate = 0.001 nm / ps. nsteps = 2500000  dt = 0.002 .

Error after the 2383000 th step

Distance of pull group 1 (7.306606 nm ) is larger than 0.49 times the box
size ( 7.455605 ). You might want to consider using "pull-geometry =
direction-periodic" instead.

I also noticed that the "distance at start " for pull group 1 = 2.560 in
grompp output. Initially I thought that this distance plus 5 nm (the total
pulling distance) > 0.49 times 15 nm ( length of box in z direction ) and
is contributing to the error. But this has been the same in cases 1 and 2
also. Hence I am getting confused.

Any suggestions , please ??

Thanks & Regards
Agnivo Gosai
Grad Student, Iowa State University.