Hi all !
I have trouble with the potential energy computation. I always get a zero value for potential energy, whatever the conditions (sollicitation type, contact law, number of particles) and the compute used (pe, atom/pe).
I linked a small test case with an input script (two particles in periodic box, compressed at constant strain rate) and the thermo result I get in the log file. The particles indent in one another, but the potential energy remains zero. It seems that there is no pe computation in LIGGGHTS-PUBLIC/examples/LIGGGHTS/Tutorials_public.
I'll be glad if you can spot a mistake in my test case or send me a working example. Any hint is welcome!
Cheers
Robin
| Attachment | Size |
|---|---|
 input_script | 2.13 KB |
| log_results | 7.07 KB |
robin | Fri, 07/24/2015 - 14:44
Hi Omid,
Hi Omid,
Thanks for this idea, it allows us to verify that compute_pe is not completely silent.
From the compute_pe's documentation : "The energy is calculated by the various pair, bond, etc potentials defined for the simulation. [...] Various fixes can contribute to the total potential energy of the system."
From the fix_gravity's documentation : "The fix_modify energy option is supported by this fix to add the gravitational potential energy of the system to the system's potential energy as part of thermodynamic output."
Indeed, when you turn the switch "fix_modify energy" on, the potential energy of the system varies, so this parts seems to work. But I can't find any "hidden switch" for the elastic contribution of potential energy, which should always be included.
Cheers,
Robin
ckloss | Mon, 07/27/2015 - 12:24
Hi Robin,
Hi Robin,
if the pair style does not metion it, it's not implemented! (and pair gran does not)
Best wishes,
Christoph
robin | Mon, 07/27/2015 - 18:05
Hi Christoph,
Hi Christoph,
Thanks for your message, I'm sure now it's not something missing in my codes.
Is it planed at all to implement this feature in a future release?
Anyhow, I suggest mentionning this in the "restictions" sections of compute_pe and pair_gran documentation.
Cheers,
Robin
ckloss | Thu, 12/10/2015 - 23:08
Hi robin,
Hi robin,
yes - I put it on my list of things to mention the restirciton in the next release
Cheers
Christoph