[gmx-users] temperatrue control again

[B. Nick]

Dear Eric,

I have attached my input parameters. Integration
timestep is 0.5 fs and coupling time already 0.1 ps.

I have also attached the beginning of the output
of g_traj -f traj.trr -ot
that shows the temperature falling down within
the first picoseconds.

What else can I change/check? 

Thanks for any help,
Gitta

;
;	MD Input File 
;
;
title               =  MD run at 371 K
cpp                 =  /lib/cpp


; *** Run control ***

integrator          =  md
tinit               =  0
dt                  =  0.0005	
nsteps              =  10000000
nstcomm             =  1        
comm_grps           =  ABc HOH

; *** Output control ***

nstxout             = 100000
nstvout             = 100000
nstfout             =    0
nstlog              = 10000
nstenergy           = 10000
nstxtcout           = 10000


; *** Neighbor searching ***

nstlist             =  10
ns_type             =  grid
pbc                 =  xyz
rlist               =  1.2


; *** Electrostatics and VdW

coulombtype         =  Cut-off
rcoulomb_switch     =  0.0
rcoulomb            =  1.4
epsilon_r           =  1.0
vdwtype             =  Cut-off
rvdw_switch         =  0.0
rvdw                =  1.4
DispCorr            =  EnerPres


; *** Temperature coupling ***

tcoupl              =  Berendsen
tc_grps		    =  ABc HOH
tau_t               =  0.1 0.1
ref_t               =  371.0 371.0


; *** Pressure coupling ***

Pcoupl              =  Parrinello-Rahman
Pcoupltype          =  isotropic
tau_p               =  1
compressibility     =  4.5e-5 
ref_p               =  1 


; *** Velocity gerneration ***

gen_vel             =  yes
gen_temp            =  371.0
gen_seed            =  173529


; *** Bonds ***

constraints          =  none
unconstrained_start  =  yes




# This file was created by g_traj
# which is part of G R O M A C S:
# Gravel Rubs Often Many Awfully Cauterized Sores
# All this happened at: Thu Mar 20 13:57:30 2003
#
@    title "Temperature"
@    xaxis  label "Time (ps)"
@    yaxis  label "(K)"
@TYPE xy
@ view 0.15, 0.15, 0.75, 0.85
@ legend on
@ legend box on
@ legend loctype view
@ legend 0.78, 0.8
@ legend length 2
@ s0 legend "ABc"
@ s1 legend "SOL"
 0	376.324	370.866
 50	345.259	249.622
 100	374.621	245.835
 150	417.497	247.743
 200	359.358	250.647


>Hi, 

>Are you really using 1 femtosecond for the coupling time, or did you 
>mean 1 picosecond? 

>The values in the mdp file are given in picoseconds; if you don't get 
>the correct temperature 
>I would first try to use a shorter coupling time, say 0.1 picoseconds. 

>Cheers, 

>Erik 

On Wednesday, Mar 19, 2003, at 22:46 US/Pacific, B. Nick wrote: 

> 
> Dear all, 
> 
> I want to run a MD simulation of a 74 atoms 
> solute in 1600 SPC water molecules at 371 K. 
> I coupled each group (solute and solvent) 
> separately to the temperature control. 
> If I analyze the temperature after MD 
> I see that the solute is about 370 K whereas 
> the water cools down to 251 K during the 
> first ps of the run. This happens 
> irrespectively of the thermostat and timestep. 
> I used Berendsen and Nose-Hoover with a timestep 
> of 1fs or 0.5 fs. 
> Why isn't the temperature of both near 371 K? 
> Can anyone explain this to me? 
> 
> Thanks in advance, 
> Gitta 
> 
> -- 
> Dr. Birgitta Nick 
> Deutsches Wollforschungsinstitut 
> Veltmanplatz 8, D-52062 Aachen, Germany 

----- 
Erik Lindahl, MSc, PhD <lindahl at stanford.edu> 
D109, Fairchild Building 
Dept. Structural Biology, Stanford University School of Medicine 
Tel. 650-7250754 Fax. 650-7238464 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20030320/c0170dd5/attachment.html>