AMBER Archive (2005)

Subject: Re: AMBER: The pressure

From: bybaker_at_itsa.ucsf.edu
Date: Wed Mar 09 2005 - 00:38:14 CST


('binary' encoding is not supported, stored as-is)

Dear Ross:

Thank you very much for your advices.

>
> Can I see your input files. Are you running constant pressure periodic
boundary conditions? What are the values of NTB, NTP and PRES0 and TAUP?

I have included my input file as an attachment. I use constant pressure
periodic boundary condition. ntb=2, ntp=1, tautp=2.0.

> it is possible that your system is either: a) not sufficiently
> equilibrated

I equilibrated my system with two steps: 0K to 300K, 50 ps; and constant
pressure 300K, 50 ps.

> b) the run is too short for accurate statistics,

It is a 3-ns MD run.

> or you dump to the output file too infrequently,

dt=0.002, ntwr=5000, ntwx=5000,

> c) you have some kind of restraints defined whichstop the pressure
regulation working correctly.

No restraints on the system.

> Try plotting the average pressure as a function of time (you can use a
moving box car average for this) and see if it is converging over time,

The pressure jumped up and down, but the positive and negative values
remained in the certain ranges. Only when I try to calculate the average
value using Excel, it shows the vaule of ~ 2.--- or ~5.----. The average
value of density is about 1.034. I compared the results from the MD runs
using 6A water box or 10A water box. Both used the same input file and
the average pressure values are above ~2.--- atomsphere. Maybe as you
suggested that I dump to the output file too infrequently

All the best

Bo



  • application/octet-stream attachment: md1a.in
-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber_at_scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu