Using temp/rescale can be considered static simulation?

Dear all,
I have a simulation fully relaxed, but temperature becomes unrealistically high when the interaction of parts start. No chemical reactions, just a very simple physical interactions according to potential file used. This is probably due to the rate of movement, which cannot go lower in this certain case.

If I use temp/rescale command and set the temperature low, the kinetic energy is very small compared to potential energy so that it can almost be ignored when added as total energy.

Does this mean my simulation can be considered static instead of dynamic? I mean, static result is ok in a way for me. Just wondering temp/rescale is somehow messing up even the static result.

Literatures similar to my simulation do not say they used temp/rescale, but the wording is very vague and never in detail…

Thank you,



That means probably your simulation settings are not suitable for your system. Most common is a timestep that is too large for the masses or the potentials involved.
Please note that, depending on the size of the system, you will have fluctuations. Temperature is only really well defined (and constant) like in macroscopic systems for very, very large systems.

Never EVER use fix temp/rescale for a production calculation.

No. Static would require no kinetic energy at all and that would be the result of a minimization.

Of course not.

Please complain to the authors or the editors of the journal. Before you do that, however, check if the authors provide input files in the supporting information (they should if they take reproducibility seriously). In many cases details of the system setup are glossed over since it is assumed that you have learned how to do that from your adviser or tutor.

Thank you sir for the quick reply.

your first explanation gave me chills because it seemed like you were seeing right through what I am doing.

Masses and potentials involved are indeed different.

I will continue my work on this, try to change the timescale, and see if there is any other methods like quasi-static simulations instead of this.

Thank you,