[gmx-users] Re: heating the system

[Anton Feenstra]

nk yx wrote:
> Hi,Anton
>   Do you mean that it'd better use 'simulated annealing' method to heat 
> the system with the zero_temp_time set to negtive? If i want to heat the 
> system from 0K to 100K during 10ps period, how do i set up the 
> parameters of heat.mdp?
>  dt=0.002ps
>  nsteps=5000
>  annealing=yes
>  zero_temp_time=?
>  ref_t=?(0K or 100K)
> Is it so kind of you to tell me the last two parameters?
> Thank you very much!

This is a bit tricky one, I'm not sure it is actually
possible. Try setting init_t (or t_init, check the mdp
options page) to something other than 0, and then play
with zero_temp_time. In any case, ref_t should be 100K,
and your other parameters are OK, too.

What is your reason for wanting to heat from 0-100K? As I
said before, if it's simply to start your run with a reasonable
set of velocities, you'r probably much better off by setting
gen_vel=yes and gen_temp=100K (or whatever temp you want).
This is used consistenly and extensively, and has no artefacts
that I am aware of.


-- 
Groetjes,

Anton
 _____________ _______________________________________________________
|             |                                                       |
|  _   _  ___,| K. Anton Feenstra                                     |
| / \ / \'| | | Dept. of Pharmacochem. - Vrije Universiteit Amsterdam |
|(   |   )| | | De Boelelaan 1083 - 1081 HV Amsterdam - Netherlands   |
| \_/ \_/ | | | Tel: +31 20 44 47608 - Fax: +31 20 44 47610           |
|             | Feenstra at chem.vu.nl - www.chem.vu.nl/~feenstra/       |
|             | "If You See Me Getting High, Knock Me Down"           |
|             | (Red Hot Chili Peppers)                               |
|_____________|_______________________________________________________|