Dump STL forces

Submitted by andrea.pasquali on Thu, 02/10/2011 - 14:31

Hi,
Is it possible dump forces on stl?
I'd like to save a dump file to then create with pizza a surface vtk file with force field (like a patch).

Thanks for any reply

Andrea

Forums:

ckloss's picture

ckloss | Thu, 02/10/2011 - 18:00

an empirical constant for wear prediction - see the doc "finnie_wear.pdf"
but it has nothing to do with dumping the wall forces

andrea.pasquali | Fri, 04/01/2011 - 09:14

Hi,
I have an other question relating forces on stl.
The fix mesh/gran/stressanalysis command exports pressure and shear stress, it can also integrate the pressure on stl surface to obtain Force? Or is there a new command?
I'd like to monitor the Force on stl during the simulation, maybe making a plot Force-Time.

An other thing, during the simulation I can see the "KinEng" (total translational kinetic energy) and the "1" (total rotational kinetic energy). Is is also possible to check the Total Energy and/or the Internal Energy?

Thanks in advance for any help

Andrea

raguelmoon's picture

raguelmoon | Fri, 04/01/2011 - 14:30

Hi Andrea,
Yes you can dump forces on STL. Fix mesh/gran/stressanalysis stores a global vector with 6 components for access by various output commands. The first 3 components are equal to the total force on the mesh. See docs page for further info about fix mesh/gran/stressanalysis and output commands.
Best,
Ram

Ram

andrea.pasquali | Fri, 04/01/2011 - 15:25

Hi Ram,
thank you for your reply.
I read the doc but where can I find this output file? In my case directory there is not.
Do I have to write an appropriate dump command? I tried with:

dump dmp all stl 100 post/dump.model

but nothing.

Thanks

Andrea

raguelmoon's picture

raguelmoon | Fri, 04/01/2011 - 17:02

Hi Andrea,
For example your fix command is like this:

fix cad all mesh/gran/stressanalysis example.stl 1 0.01 0 0 0 0. 0. 0.

then use cad as output name like:

fix forceOnMesh all ave/time 50 2 100 f_cad[1] f_cad[2] f_cad[3] file filename.txt

Here [1]. [2], and [3] are force vectors along x-, y- and z-axis respectively. You will find the output in "filename.txt" file.

Best,
Ram

Ram

andrea.pasquali | Fri, 04/01/2011 - 17:16

Thank you very much Ram,
I needed the fix ave/time command!

Do you know also if is it possible to check the total and internal energy?

Thanks

Andrea

ckloss's picture

ckloss | Mon, 04/04/2011 - 09:05

energies are typically calculated via thee thermo output. what do you mean with internal energy?
Christoph

andrea.pasquali | Mon, 04/04/2011 - 11:22

Hi Christoph,
for internal energy I mean the contact energy between particles. I don't know the details of LIGGGHTS, but I think between particles there is like a dumper + spring model. I'd like to check the energy / force of this system in order to "stress" my runtime because I'd like to run a "quasi static" simulation.
Maybe with f_ID I can check the force of particles by fix ave/time command? It could be what I want?

Thanks

Andrea

ckloss's picture

ckloss | Mon, 04/04/2011 - 12:42

You can use the global kinetic energy energy as criterion for a loop, see doc for variable
Christoph

andrea.pasquali | Mon, 04/04/2011 - 22:31

Hi Christoph,
thank you for your time.
I have a stl that penetrates inside my particles until a fixed deep (very slow).
I'd like to check the energy between particles (Ei) in order to compare it with the total energy (Et).
I want increase the velocity of my stl up to Ei = 5% Et.
I this way I'd like to save iteration and speed up my simulation.

Regards

Andrea

ckloss's picture

ckloss | Tue, 04/05/2011 - 08:45

ok, I see. but sorry, the internal energy is currently not calculated. this may be included in the future, but I cannot say a date yet. What you can do is use the fix check/timestep/gran and use the hertz timestep fraction it calculates to find the maximum velocity that is possible

Christoph

andrea.pasquali | Tue, 04/05/2011 - 15:07

Hi Christoph,
I have two questions
1) I'm trying the command you suggest:
fix ts all check/timestep/gran 1 0.1 0.1
I can see this:
stl velocity increased 10x = hertz timestep fraction around 0.012
stl velocity increased 20x = hertz timestep fraction 0.013
stl velocity increased 50x = hertz timestep fraction 0.014-0.015
stl velocity increased 100x = hertz timestep fraction 0.016-0.017

I can increase my stl velocity or my hertz fraction bound is too high?
Is it correct to use Nevery = 1?

2) I'm extracting forces on stl with the fix ave/time command:
fix forceOnMesh all ave/time 1 1 1 f_cad1[1] f_cad1[2] f_cad1[3] file forces.dat
But calculating IntegrateVariables on .vtk with ParaView I obtain a different results. At last iteration I have:
with fix ave/time , fx = 1.53 fy = 4.4 fz = 10.93 , f tot = 11.88
with ParaView IntegrateVariables pressure 58.81
Which is the correct force? Am I missing something?
Also here is it correct use Nevery Nrepeat and Nfreq = 1?

Thanks in advance

Andrea