msst

LAMMPS LAMMPS Mailing List Mirror
1

Dear All,
I am using LAMMPS (14 Feb 2014) and have some questions about fix msst :

1.I set Us=6km/s,mu=0, p0 , v0 and e0 are calculated on the first step.timestep=1fs,I run 200000 steps.
By default,q=10,tscale=0.01.In order to get the appropriate ‘q’ and ‘tscale’ value,I have tried different sets of ‘q’ and ‘tscale’.I tried ‘q’ with a range: 5,10,15,20,…, 40 (keeping tscale=0.01 ) and I also tried ‘tscale’ with a range: 0.001,0.005,0.01,0.015,0.02(keeping q=15 and 10).All simulations I tried can see a compression and all outputs(including pressure,vol,lagrangian_speed,etc) except dhugoniot and drayleigh, can come to around stable values after several thousands steps.However,after running 200000 steps ,dhugoniot and drayleigh still cann’t come to stable values.dhugoniot varies from -20 to 20,dray varies from 2000 to -2000.I just wonder how to deal with the problems and get the appropriate ‘q’ and ‘tscale’ value?

2.How to judge whether dhugoniot and drayleigh are small enough?

Thank you very much for your answer

2

Have you searched the mailing list archive?

Ray

3

Yes，I have gone through almost all the similiar questions and get the point that dhugoniot and drayleigh should be around 0.However, dray seems a litter large.Therefore,I just wonder whether dhugoniot and drayleigh are small enough ?

4

Similar to T and P fluctuations in NPT ensemble simulations. How do
you determine if you T/P fluctuations are small enough? By the way,
without providing units, nobody knows how large "dhugoniot varies from
-20 to 20,dray varies from 2000 to -2000" are.

Ray

5

I am sorry,dhugoniot is temperature unit ‘K’, dray is pressure unit ‘bar’,so dray seems a litter large.