[gmx-users] unusual behavior of CV in steered MD (gromacs+plumed)

[tarak karmakar]

Dear All,
I am running a SMD simulation (gromacs+plumed) pulling a ligand out of the
enzyme. The parameters are,
pulling velocity= 0.00005 per 1000 steps (2 ps, as in gromacs)
force constant = 20920.0 kJ/mol/nm-2
CV from 0.3 nm to 1.7 nm

STEER CV 1 FROM 0.3 TO 1.7 KAPPA 20920.0 VEL 0.00005

So, I need to run a trajectory up to a time span of 56 ns [(2 ps /0.00005
nm)*1.4 nm].
part of nvt.mdp
dt                      = 0.002                 ; [ps] time step for
integration
nsteps               = 28000000          ; maximum number of steps


While running this simulation, I observe that after CV=1.0 (~23 ns), the
graph becomes parallel to x-axis; the CV is not increasing (in principle,
it should increase up to 1.7).

What could be the issue in this SMD simulation?
Is it that the velocity is too low, and the stiff-spring approximation is
not getting obeyed?

I appreciate your comments and suggestions.

Thanks,
Tarak