Particle damper - dissipated energy

Submitted by Tijmus on Fri, 01/04/2013 - 16:53

I'm trying to simulate a cylindrical particle damper and basically have my simulation up and running now. But... I can't figure out how to extract the quantity I'm interested in: the energy dissipated by the interparticle and wall collisions (either integrated over time or per timestep).

Any ideas? Thanks.

Forums:

richti83's picture

richti83 | Fri, 01/04/2013 - 17:55

you can write Ekin into a variable and dump it with fix print:
add ke to your thermo modify command and then

variale t equal step*dt
variable Ekin equal ke
fix outfile all print 1 "$t,${Ekin}" file energy.csv title "time,Ekin" screen no #dump every integration step

you will see a dramatic dependency between dt and Ekin.
If you are interested in we can exchange results next week, at the moment I have no access to my storage in the office.

I'm not an associate of DCS GmbH and not a core developer of LIGGGHTS®
ResearchGate | Contact

ckloss's picture

ckloss | Sun, 01/06/2013 - 22:24

Hi Tijmus,

alternatively to tracking the global energy as described by Christian, you could also modify compute pair/gran/local to track the dissipated energy per-contact.

Cheers,
Christoph

Tijmus | Mon, 01/07/2013 - 10:15

Thanks guys. Correct me if I'm wrong, but I think tracking the kinetic energy wouldn't be enough. I have a moving mesh feeding the system with energy, so I wouldn't know how well energy is dissipated just by tracking the kinetic energy.

I'll have a look at modifying pair/gran/local.

Tijmus | Tue, 01/08/2013 - 15:10

Christoph, is it only compute pair/gran/local that needs to be modified? Any pointers as to which variables are accessible by the routine (my understanding of the inner workings of the code are still very limited)?

Surprised that outputting damping quantities isn't in there by default btw; hasn't anyone done something similar before (couldn't find much on the forums anyways)?.

ckloss's picture

ckloss | Tue, 01/08/2013 - 16:39

>>Surprised that outputting damping quantities isn't in there by default btw
Assuming defaults in Open Source software can or cannot lead to surprise :-)

You'd need to modify the interface to the cpl->add_pair function that is called from the gran pair styles and pass the damping parts of the force to this function

Cheers, Christoph

Tijmus | Tue, 01/08/2013 - 16:58

That's of course very true :). I'll see what I can manage.

You agree though, that there's no way of getting the info I need without modifying the code, right?

Thanks,
Tijmen

marcosmadrid | Tue, 08/29/2017 - 21:08

Hi Tijmus,
did you solve this?
did you found how to modify the code to extract the dissipated energy?
in that case, could you help me with that! I'm trying to do the same (and I fail with all success!)
Marcos