[Pw_forum] problem with control of temperature in pw

Eduardo Ariel Menendez P emenendez at macul.ciencias.uchile.cl
Wed Dec 20 17:54:16 CET 2006

I have some problems to do molecular dynamics using pw.x at a defined
temperature (version 3.2 and 3.1.1).

Please, take a look at the example04. To introduce a starting temperature
I added tempw=300.D0 in si.md8.in.
Following the documentation, this should set 300 K as starting T.
However, the starting temperature remains 0 K.

Moreover, using
    nraise=1,  (to rescale at each step)
it starts at 300 K. However, it is not evident that it rescales, unless
it reports all the magnitudes just before rescaling.
cat si.md8.out |grep 'kinetic energy (Ekin) ='
     kinetic energy (Ekin) =     0.01971053 Ry
     kinetic energy (Ekin) =     0.01951557 Ry
     kinetic energy (Ekin) =     0.01958081 Ry
     kinetic energy (Ekin) =     0.01967049 Ry
cat si.md8.out |grep 'Ekin + Etot (const)   = '
     Ekin + Etot (const)   =   -62.15607665 Ry
     Ekin + Etot (const)   =   -62.15583900 Ry
     Ekin + Etot (const)   =   -62.15540873 Ry
     Ekin + Etot (const)   =   -62.15504154 Ry
 cat si.md8.out |grep temperat      Starting temperature  =   300.00 K
     temperature           =   296.38506727 K
     temperature           =   293.45347455 K
     temperature           =   294.43450387 K
     temperature           =   295.78296164 K

Maybe this is because the magnitudes are reported just before rescaling. I
see that reducing dt the fluctuations are reduced. Is is the case?


More information about the users mailing list