Re: "GroupPressure no" resulting in negative pressure outputs

From: Jérôme Hénin (jhenin_at_ifr88.cnrs-mrs.fr)
Date: Thu Nov 25 2010 - 14:56:08 CST

Hi Sebastian,

On 25 November 2010 20:51, Sebastian Stolzenberg
<s.stolzenberg_at_gmail.com> wrote:
> Happy Thanksgiving, All,

Not everyone on this list celebrates Thanksgiving, but thank you all
the same :-)

> "useGroupPressure no" is a default option and not required, when not using
> "rigidBonds".
>
> For the following, I measure "pressure" from the log files using the NAMD
> Plot plugin by observing "PRESSAVG" or "GPRESSAVG".
>
> For a system with box size >100x100x90A^3 and no rigidBonds (FlexibleCell
> yes):
> Using "useGroupPressure yes",
> I get pressures of about 1+-2bar
> and with "useGroupPressure no",
> I get pressures of about -13+-3bar
>
> Fluctuations are larger for "useGroupPressure no" vs. ""useGroupPressure
> yes", but within these fluctuations, the "useGroupPressure no" doesn't seem
> to give me the correct pressure that I set via:
>
> ..
> LangevinPiston       on
> LangevinPistonTarget 1.01325
> LangevinPistonPeriod 200
> LangevinPistonDecay  50
> LangevinPistonTemp   310
> ..
> PME                 yes
> PMEGridSpacing  1.0
> ..
>
> Talking to colleagues of mine, they appear to have exactly the same issue
> for different systems.

It does sound like a problem. To distinguish whether it comes from
pressure measurement itself or from the barostat, it would be very
useful to repeat this comparative test in constant volume simulations.

Best,
Jerome

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:54:47 CST