Dear experts,
I was using “pair_style lj/cut/coul/long 11.0” and I had pair_coeff in my data file and I was able to output com of the system correctly. Then I tried to use lj/smooth but my results are very strange/wrong and also com gives me the same value for all the times (the movie shows movement of the atoms though)…I put pair_coeff in the scrip this time (see below)
I do not know what is wrong, I appreciate your advice and help.
here is part of the script:
I don't know what you're asking. Does the dynamics
of the system look correct and is the thermo output normal?
Steve
Dear Steve,
Actually when I switch from “pair_style lj/cut/coul/long 11.0” to “pair_style hybrid/overlay lj/smooth 10.0 12.0 coul/long 11.0” I get very different results.
pair_style hybrid/overlay lj/smooth 10.0 12.0 coul/long 11.0
pair_coeff 1 1 lj/smooth 0.1552 3.1655 10.0 12.0
pair_coeff 2 2 lj/smooth 0.0000 0.0000 10.0 12.0
pair_coeff 3 3 lj/smooth 0.1444 3.2144 10.0 12.0
pair_coeff * * coul/long
Using hybrid/overlay, If I change the lj parameter (epsilon), there is no difference in output of energy values from thermo_style, like “TotEng, E_vdwl, E_coul, E_long, E_tail, PotEng, KinEng, Temp, Press” and center of mass gives me the same value as initial value for all steps.
To test if hybrid/overlay works properly I tried to separate “pair_style lj/cut/coul/long 11.0”, so I used
“pair_style hybrid/overlay lj/cut 11.0 coul/long 11.0”
and again by changing LJ epsilon value no difference in all thermo_style values mentioned above, was observed.
Would you please explain what is wrong with hybrid/overlay?
Many Thanks,
Jamie
Dear Steve,
Actually when I switch from "pair_style lj/cut/coul/long 11.0" to
"pair_style hybrid/overlay lj/smooth 10.0 12.0 coul/long 11.0" I get very
different results.
pair_style hybrid/overlay lj/smooth 10.0 12.0 coul/long 11.0
pair_coeff 1 1 lj/smooth 0.1552 3.1655 10.0 12.0
pair_coeff 2 2 lj/smooth 0.0000 0.0000 10.0 12.0
pair_coeff 3 3 lj/smooth 0.1444 3.2144 10.0 12.0
pair_coeff * * coul/longUsing hybrid/overlay, If I change the lj parameter (epsilon), there is no
difference in output of energy values from thermo_style, like "TotEng,
E_vdwl, E_coul, E_long, E_tail, PotEng, KinEng, Temp, Press" and center of
mass gives me the same value as initial value for all steps.To test if hybrid/overlay works properly I tried to separate "pair_style
lj/cut/coul/long 11.0", so I used
"pair_style hybrid/overlay lj/cut 11.0 coul/long 11.0"
and again by changing LJ epsilon value no difference in all thermo_style
values mentioned above, was observed.Would you please explain what is wrong with hybrid/overlay?
please provide real proof that something is wrong with hybrid/overlay.
so far everything you mention is circumstantial and not reproducible.
without seeing a complete set of (easy to run) inputs and a clear
description what you expect see and what you get instead.
many people (including me) have been using hybrid/overlay a lot
and it has been working for them very well.
based on previous experience, chances are high, that something
is messed up in your input, that you have not paid any attention to
and that you have not shown us.
cheers,
axel.
Can you use lj/smooth by itself (w/out pair hybrid) and get the same
energy/pressure (for 0 steps)
you get doing pair hybrid with just lj/smooth?
Then can you add coul/long to pair hybrid and get the Vdwl energy unchanged,
which still comes from lj/smooth? Note that the pressure will change
b/c there is part of that which will come from coul/long.
You should be able to do the same thing, substituing lj/cut for lj/smooth.
However lj/cut and lj/smooth may be different I believe (even for the
same epsilon,
b/c some forms of smoothing alter the entire range of the potential.
Steve