[gmx-users] The pull code

[Majid Rezaei]

Dear gromacs users,
I am tryingto make a shear flow in an electrolyte by moving two parallel bilayers immersedin the electrolyte in opposite directions. I use the following pull code (wherethe groups low and up are the two moving bilayers):
pull = yespull-coord1-type = umbrellapull-coord1-geometry= direction-periodicpull-pbc-ref-prev-step-com= yespull-coord1-dim = Y N Npull-coord1-vec    = 1 0 0pull-coord1-start  = yespull-ngroups = 2pull-group1-name = lowpull-group1-pbcatom= 2pull-group2-name = uppull-group2-pbcatom= 1202pull-coord1-groups= 1 2pull-coord1-rate   = 0.030 pull-coord1-k = 500   pull-coord1-init   = 1.3pull-nstxout = 250pull-nstfout = 250
I expect thevelocity of the bilayers to be ±15 m/s. But the calculated velocityprofile shows a velocity of around 5 m/s. I also tried some other rates ofchange of the reference position and found that the calculated velocity isalways around 3 times less than what I expect. Could you please let me know whatis the problem with my code?
“It shouldbe noted that I am using Gromacs 2019 and that I didn’t encounter such problemwhen using the same code (minus the pull-pbc-ref-prev-step-comcommand) and the same simulation system in Gromacs 2016”
Best wishes,Majid