Using fix servo wall to reach a target confining pressure, but failed.

Submitted by Rachel on Mon, 10/29/2018 - 15:16

Hi all,

I am using LIGGGHTS for a confining process of a cube area. The maximum and minimum of particles are 5mm and 1mm, respectively. Here I use six servo wall to confine the particles, e.g.

fix topwall all mesh/surface/stress/servo file meshes/topwall.stl type 1 com 0. 0. 0.1 ctrlPV force axis 0. 0. -1. target_val ${targetForce} vel_max 20 kp 0.005

I keep updating the value of "targetForce" so as to reach a confining pressure of 50kPa, but I found it very hard even after 5,000,000 step with timestep=2e-7.

I have also tried several other "vel_max" and "kp" values but cannont get a good results. Could anyone please tell me how to select a reasonable "vel_max" value and "kp" value or do I make any mistake during my simulations? Thanks very much.

Best regards,
Rachel

Forums:

arnom's picture

arnom | Thu, 11/15/2018 - 15:39

What Youngs Modulus are you using? vel_max is rather high I have to say, we generally use values around 1. I also suggest you to use the mode auto where you won't need the kp.

DCS team member & LIGGGHTS(R) core developer

Rachel | Thu, 11/22/2018 - 03:44

Hi Arnom,

Thanks for your reply. I use the Young's modulus E=7e10Pa for particles and E=7e11Pa for servo walls. I found I can get the target pressure using vel_max=0.05 and kp=1. I am still a little confused how to determine a suitable vel_max and kp? How to reduce the fluctuation during the loading process using servo walls? The kp value is not important to me, I only need to obtained a stable confining pressure state. Could you please help to explain these? Many thanks.

Best regards,
Rachel

arnom's picture

arnom | Mon, 12/17/2018 - 10:19

You need to make sure that you are within the ranges of the time stepping constraints (see fix check/timestep/gran and associated documentation). Your Youngs moduli are really high so that's why vel_max needs to be low.

DCS team member & LIGGGHTS(R) core developer

iluvatar | Wed, 12/19/2018 - 05:01

Hi Rachel,
Did you find any solution to this problem? If yes, could you please share it? Or, if you prefer, we could talk by email. I have been unable to use the servo module successfully to impose a given pressure on the packing, but I feel I am really close. Thanks in advance.

iluvatar | Fri, 12/14/2018 - 05:08

Hi, I am also compressing a cubic sample with servo wall by it is very difficult to calibrate either kp/vmax or ratio/vmax (the later for mode auto) to reach the target force value. I am measuring the mesh force value using f_SERVONAME[3] (in vertical direction , in this case). The value are always way off the target one. Is there any trick/advise to get this right?

zumack | Thu, 01/03/2019 - 06:55

I recommend to you to read the O'Sullivan Book A discrete element modelling in the Chapter 2 is the considerations of the numeric stability of DEM. And you could read the Book Contacts in Mechanics by K. Johnson. In that you could search the principles of contact models and the limitations of that.