Stress calculation in granular bonds

Submitted by SHUBHAM AGARWAL on Sun, 03/21/2021 - 09:15

Hello
I am running a fiber network simulation using LIGGGHTS-flexible fibers version.
I want to calculate the average stress in x direction for a specified region "stress_region". I am trying following commands to execute the same.

compute sigma stress_region stress/atom bond
compute sigma_x reduce ave c_sigma[1]
variable stress_x equal c_sigma_x[1]

I believe variable stress_x should give me the average stress in X direction for a region "stress_region "

However I am getting the following error: "Invalid compute style"

Kindly help me in this regard
Shubham

Forums:

mschramm | Wed, 03/24/2021 - 16:00

Hello,
The ev_tally (what assigns vatom for use inside of compute_stress_atom.cpp) is not carried out in the granular bond so you would not get any bond contribution to your compute call.
However, this does not explain the error you are receiving. Does it work if you drop bond? The function may see that bond_gran does not attribute to it and simply errors out but I am not sure.

mschramm | Wed, 03/24/2021 - 16:03

You can calculate this value in post processing by printing the bond properties to a dump file via
compute bc all property/local batom1x batom1y batom1z batom2x batom2y batom2z batom1 batom2 btype bforceX bforceY bforceZ btorqueX btorqueY btorqueZ beqdist

dump bondforcechain all local ${fileprint_step} post/bfc*.bond &
c_bc[1] c_bc[2] c_bc[3] c_bc[4] c_bc[5] c_bc[6] c_bc[7] c_bc[8] c_bc[9] c_bc[10] c_bc[11] c_bc[12] c_bc[13] c_bc[14] c_bc[15] c_bc[16]

Mateus Kostka | Mon, 03/29/2021 - 11:27

Dear Ladies and Gentlemen,

is it possibile to calculate the stress in the particle contact area?
My Idea is, to calculate it with the overlap area and the contact force.
Is it possible to print the force and conact area for each particle contact?

When I use this command

compute sigma stress_region stress/atom

I get only very small stresses, although I use Si conform units.

Best regards and thank you

Mateus Kostka