Re: NAMD Anisotropic pressure control

From: Mahdi Tavakol (mahditavakol_at_ymail.com)
Date: Fri Sep 18 2015 - 11:51:29 CDT

Dear Jerome
Thanks very much for the reply. I calculated the surface tension through the formula and converted the value to dyn/cm (as in the manual). However, Pxx and Pyy were converged to another pressure. I used these parameters:
useGroupPressure      no #  no hydrogens in CG hence set this to no inspite of 1 fs stepuseFlexibleCell       yes useConstantArea       no useConstantRatio      yes 
langevinPiston        yes
langevinPistonTarget  -150   #  in bar -> 1 atmlangevinPistonPeriod  200.  #usually 2000 langevinPistonDecay   100.  #usually 1000langevinPistonTemp    $temperatureSurfaceTensionTarget  119.2  #using the formula
Pzz was converged to -150, but Pxx and Pyy were converged to -300.
 Could you please explain me more about the formula?
Yours sincerelyMahdi

     On Friday, September 18, 2015 7:15 PM, Jérôme Hénin <jerome.henin_at_ibpc.fr> wrote:
   

 Hi,

Your problem is equivalent to applying a given pressure along z and a nonzero surface tension along x/y (assuming nearly constant cell height Lz).

You should be able to achieve that using SurfaceTensionTarget.
Pz is then specified by LangevinPistonTarget,and I think the surface tension should be Lz * (Pz - 1 bar).

Jerome

On 18 September 2015 at 15:50, Mahdi Tavakol <mahditavakol_at_ymail.com> wrote:

Dear NAMD users,
I would like to stretch my system using anisotropic pressure control. As I know, using useConstantArea the pressure just be controlled in the z direction. However, the values of Pxx and Pyy remain uncontrolled and increase during the simulation. Is there any way to control Pxx and Pyy in 1 bar, while increasing the value of  Pzz.
Any help is appreciated in advance,
Mahdi Tavakol

  

This archive was generated by hypermail 2.1.6 : Thu Dec 31 2015 - 23:22:04 CST