Discussion Closed This discussion was created more than 6 months ago and has been closed. To start a new discussion with a link back to this one, click here.
Simulation of impact experiment using Johnson-Cook plasticity model
Posted May 17, 2021, 11:04 p.m. EDT Structural & Acoustics, Materials, Structural Mechanics 3 Replies
Please login with a confirmed email address before reporting spam
Hi everyone!
I'm trying to simulate a ball falling in a plate using COMSOL solid mechanics. I tried to simulate it by prescribed displacement of the ball with a wave function (so the ball impacts and rebounds without fall infinitely). However, I would like to precisely measure the plate's deformation profile as material energy absorbed response. (I am using the johnson-cook plasticity model for the plate). But the problem is that this boundary imposes the penetration value instead of being a response from the plate's material. I thought to use a contact model that measures the overlap of materials (like LSD, constantly used in DEM simulations). However, COMSOL does not allow the use of elaborated contact models in the mechanical module.
So, what is the more realistic boundary condition to simulate the ball rebounds as a response of the plate material?
Please login with a confirmed email address before reporting spam
The Linear Spring Dashpot (LSD) in DEM is a way to add flexibility and loss to the otherwise rigid interacting particles.
If you are actually modeling the deformations in the material coming into contact, for example with plastic dissipation (as with the Johnson-Cook material model) or with viscoelasticity (as in this example: https://www.comsol.com/model/impact-analysis-of-a-golf-ball-89461 ) , then the dissipative process is already accounted for, so it should not also be part of the contact condition as such.
This said, if you are going to model impact contact, you should use either the Penalty, dynamic or Augmented Lagrangian, dynamic method in the Contact node. Those methods include both a stiffness and a viscosity term, but they should be considered more as numerical tools. You could however also choose to view it as a type of distributed LSD.
-------------------Henrik Sönnerlind
COMSOL
Please login with a confirmed email address before reporting spam
Hi Henrik!
Thank you very much for the comments, I will review my contact condition. Although, I still have doubts about the formulation of the dynamics of the ball itself. I'm trying to simulate a free-fall of a rigid ball in a steel plate (which I set the plasticity material model). And I want to validate this model by measuring the indentation (plastic deformation) in the target plate. However, by assigning a function that describes the ball movement, I am imposing the penetration depth (I am using a wave function for the ball). There is some other boundary condition that you recommend to simulate this case? It is a similar simulation to Xuemei Wang's 2013 work ("Validation of Johnson-Cook plasticity and damage model using impact experiment").
Please login with a confirmed email address before reporting spam
I would suggest that you place the free ball just touching the plate, and assign an initial velocity. Then you let the contact condition and material model do the rest. You will probably have to do some experiments with tuning the contact parameters.
Remember to make use of any symmetries. If the impact is perpendicular to the plate, using 1/4 of the geometry should be sufficient. That will speed up the analysis by one order of magnitude.
-------------------Henrik Sönnerlind
COMSOL