Talk:DL POLY: Difference between revisions

From SklogWiki
Jump to navigation Jump to search
No edit summary
No edit summary
Line 1: Line 1:
what's the unit of force in CONFIG and REVCON file?
what's the unit of force in CONFIG and REVCON file?
:Interesting question. Having just taken a look at a [http://www.cse.scitech.ac.uk/ccg/software/DL_POLY/MANUALS/USRMAN3.09.pdf DL_POLY User Manual] there does not appear to be an explicit expression (CONFIG section 5.1.2, and REVCON, section 5.2.4). Given that force has the SI units of kg·m/s², presumably one would use the corresponding 'DLPOLY' units (section 1.3.8), i.e. [[amu]] for the mass, Angstroms for the length, and picoseconds for the time, and some corresponding pre-factor. All said and done I think it is worth asking why do you want to know the units of force in CONFIG. For example, if you are starting a new simulation (if I remember correctly since it has been quire a while now) just start the simulation with the coordinates (levcfg =0) and let DLPOLY do the rest via a [[Thermostats | thermostat]]. All the best --[[User:Carl McBride | <b><FONT COLOR="#8B3A3A">Carl McBride</FONT></b>]] ([[User_talk:Carl_McBride |talk]]) 17:56, 24 November 2008 (CET)
:Interesting question. Having just taken a look at a [http://www.cse.scitech.ac.uk/ccg/software/DL_POLY/MANUALS/USRMAN3.09.pdf DL_POLY User Manual] there does not appear to be an explicit expression (CONFIG section 5.1.2, and REVCON, section 5.2.4). Given that force has the SI units of kg·m/s², presumably one would use the corresponding 'DLPOLY' units (section 1.3.8), i.e. [[amu]] for the mass, Angstroms for the length, and picoseconds for the time, and some corresponding pre-factor. All said and done I think it is worth asking why do you want to know the units of force in CONFIG. For example, if you are starting a new simulation (if I remember correctly since it has been quire a while now) just start the simulation with the coordinates (levcfg =0) and let DLPOLY do the rest via a [[Thermostats | thermostat]]. All the best --[[User:Carl McBride | <b><FONT COLOR="#8B3A3A">Carl McBride</FONT></b>]] ([[User_talk:Carl_McBride |talk]]) 17:56, 24 November 2008 (CET)
:I am pretty sure the answer above is right. The force is given in the units natural to DLPOLY, which are perhaps easier to obtain as energy/distance. I.e. 1 force unit would be 0.166054 pN, which seems right, pN being the usual force scale at the molecular level. It is true forces are usually not needed, not even velocities, when restarting. However, one may be interested in the forces themselves: at variance with velocities, which must have a Maxwell-Boltmann distribution, forces may contain some interesting information for non-homogeneous systems. Forces are also vital for virial calculations, but for these one needs f_(ij), not just sum_j f_(ij). --[[User:Dduque|Dduque]] 11:39, 25 November 2008 (CET)

Revision as of 11:39, 25 November 2008

what's the unit of force in CONFIG and REVCON file?

Interesting question. Having just taken a look at a DL_POLY User Manual there does not appear to be an explicit expression (CONFIG section 5.1.2, and REVCON, section 5.2.4). Given that force has the SI units of kg·m/s², presumably one would use the corresponding 'DLPOLY' units (section 1.3.8), i.e. amu for the mass, Angstroms for the length, and picoseconds for the time, and some corresponding pre-factor. All said and done I think it is worth asking why do you want to know the units of force in CONFIG. For example, if you are starting a new simulation (if I remember correctly since it has been quire a while now) just start the simulation with the coordinates (levcfg =0) and let DLPOLY do the rest via a thermostat. All the best -- Carl McBride (talk) 17:56, 24 November 2008 (CET)
I am pretty sure the answer above is right. The force is given in the units natural to DLPOLY, which are perhaps easier to obtain as energy/distance. I.e. 1 force unit would be 0.166054 pN, which seems right, pN being the usual force scale at the molecular level. It is true forces are usually not needed, not even velocities, when restarting. However, one may be interested in the forces themselves: at variance with velocities, which must have a Maxwell-Boltmann distribution, forces may contain some interesting information for non-homogeneous systems. Forces are also vital for virial calculations, but for these one needs f_(ij), not just sum_j f_(ij). --Dduque 11:39, 25 November 2008 (CET)