Basic Units Question

438 views
Skip to first unread message

carolynl...@gmail.com

unread,
Nov 30, 2015, 12:38:33 PM11/30/15
to PLUMED users
Hi, 

I have a question about using Units in PLUMED2.  

I see that PLUMED has standard units of kJ/mol, nm, and picoseconds.  I notice that these are the same units as GROMACS.
NAMD, in comparison, has units of kcal/mol, angstrom, and I can set how many fs correspond to each timestep.

So the question is, do I need to set the units of PLUMED2 to match that of my MDEngine (using http://plumed.github.io/doc-v2.0/user-doc/html/_u_n_i_t_s.html), or does the PLUMED2 determine the conversion itself?  Alternatively, should I assume that, effectively, PLUMED is going to be operating in the same units system as the MDEngine, because it does not apply any conversion factors before generating bias forces.

So if, for example, in a metadynamics simulation, I am using the default units (i.e. I don't specify anything), I am using NAMD as the MDEngine, and I specify HEIGHT 0.5
then will this HEIGHT correspond to kcal/mol (because that is the units of NAMD and no conversion is applied) or kJ/mol (because that is the units of PLUMED and it is converted to NAMD units when applying bias forces to the NAMD simulation) 

Thanks,
Carolyn





carolynl...@gmail.com

unread,
Nov 30, 2015, 12:47:20 PM11/30/15
to PLUMED users, carolynl...@gmail.com
With a little more reading, I may have answered my own question

If I understand the documentation correctly, then, when I specify HEIGHT 0.5 while using NAMD, this will mean 0.5 kJ/mol unless I also specify that the units are kcal/mol.  If I used the same script with GROMACS, or LAMMPS, it would still mean 0.5 kJ/mol.  

Question: How do you handle interfacing with a LAMMPS simulation using LJ units?

I am referencing this in the manual
>Learning from this, we designed PLUMED 2 in a way that units in the PLUMED input are independent of the MD code. >Technically, this is achieved by doing the conversion when coordinates and forces are passed back and forth. In this way, >the same PLUMED input could be used with GROMACS and NAMD.

carolynl...@gmail.com

unread,
Nov 30, 2015, 12:50:35 PM11/30/15
to PLUMED users, carolynl...@gmail.com
Also, could you point me to where in the source code the NAMD conversion actually takes place?
Reply all
Reply to author
Forward
0 new messages