Hi to all,
maybe some developer can me help:
For gran/hooke there is only one friction coefficient necessary. But in reality are 2 friction states possible: Sliding (or dynamic) friction and static friction. Is there any difference in Liggghts, or is this always the same? For some materials is in reality a big difference between static coefficient of friction and sliding cof....
Thx
Orangensaft
ckloss | Wed, 07/15/2015 - 22:35
Hi,
Hi,
be careful not to mix this us: What people usually refer to as friction is the ratio of shear to normal force. This shows the static/dynamic transition behavior. On the microscopic side, the tangential force on a partcle-particle contact is composed of a spring and a dashpot part, limited by the Coulomb friction limit (which is assumed to be a modeled by a constant friction coefficient)
If you model a shear cell using LIGGGHTS(R) (a macroscopic friction experiment), you WILL see a change in macroscopic friction between static and dynamic
Best wishes
Christoph
OrangensaftDE | Thu, 07/16/2015 - 13:48
Hi Christoph,
Hi Christoph,
thank you for your answer!
You're right, I want to model a shear cell using LIGGGHTS (and compare it with my experiment).
The physics of the particle-particle contact I totally understood. Here -as you say- the tangential force are described by a spring and a damper, limited by Coulomb friction, which I can define by the cof. Inside the cell, I simulate the experiment with determined parameters. Therefore liggghts get me a acceptable solution. So you're right, I see a change between static and dynamic friction between particle-particle.
But at the contact between particle-boundary (represented by a rigid wall) I got some trouble with my particles. How calculate LIGGGHTS the tangential force between particle and wall? Is it also with a spring and a damper? Or only F_t=cof*F_n and I have to decide which cof I define? The latter need an new contact model to differentiate between sliding and static friction coefficient, is that right?
Greetings
Thomas
ckloss | Mon, 07/27/2015 - 12:47
Hi Thomas,
Hi Thomas,
just think of particle-wall contact as a particle-particle contact where one particle has infinite radius. So the contact law is the same
Best wishes
Christoph
JoshuaP | Fri, 12/11/2015 - 11:14
hi
Still you could consider dynamic friction in particle scale. I changed the friction model a bit, so that I first reach the Coulomb friction and then it goes down to a lower friction between the particles. This is an effect that happens because particles cannot interconnect the same way in static and dynamic conditions. What I hope to get is more compressibility and irrecoverable settlements, because my macro scale behaviour is to elastic at the moment.
regards
joshua
ckloss | Tue, 12/22/2015 - 19:52
Hi Joshua,
Hi Joshua,
for getting compressibility on the macro scale (e.g. for soil models) there are (in my opinion) better alternatives to achieve this. I do not agree that this comes from the sliding friction model. If you want, we can continue the discussion also via email (since you have a support contract with DCS).
Best wishes
Christoph