Temperature control

Submitted by rasoul on Tue, 08/20/2013 - 21:01

For doing a project I face some unknowns.

The one is that I want to simulate crack propagation in shale (a type of
sand) and investigate crack pattern in shale if I decrease the temperature by LIGGGHTS. I need to know how I can decrease it (temperature of the simulation box)?

Second is that do you know the material properties of shale?

Thanks a lot for your attention in advance.


ckloss's picture

ckloss | Fri, 08/30/2013 - 14:56

>>I decrease the temperature by LIGGGHTS
you mean macroscopic temperate?
you'll have to implement a bond or contact law that depends on temperate, there is nothing out-of the box that could do this. but any pair style can read fix property/atom values

>>Second is that do you know the material properties of shale?
I guess you will have to calibrate this


rasoul | Fri, 08/30/2013 - 22:25

Thanks a lot for the response,
The main problem is module of elasticity of shale that is very high (50E9).
I need to know what is the appropriate pair_stale for modelling shale?
What do you mean by calibrate?

gukargl | Mon, 09/02/2013 - 17:10

Thermal properties of granular materials depend quite a lot on the particle particle contact points or bridges connecting the particles. That means when you use spheres you need to implement thermal properties which represent the actual grain properties of shale i.e. make sure your simulation spheres behave like the real material which is for sure not spherical. Thus you need to calibrate your model to behave like the real material you need to simulate. This applies of course also to other material properties as well. To decrease the temperature of your model you need some heat sink otherwise energy conservation will not allow the temperature to change.
You can e.g. change the temperature of a wall to change the temperature of you simulation volume but for "realistic" materials this can take ages to achieve a homogeneus temperature distribution. In real life the presence of air or water/liquid is speeding this process up via convection. Which of course would need a CDFDEM model implementation.
I face myself the problem to calibrate a mechanical model of granular soil to match real materials and would be interested to share any progress in this respect.

rasoul | Mon, 09/02/2013 - 22:57

Dear Gunter,

As you said, I have to calibrate a mechanical model of granular soil to match real materials and I am working on it and I got some new things. I will inform you and share my achievements whenever I get useful information about calibration. I am heavily interested in cooperation with you in order to solve our difficulties.


rasoul | Sat, 09/07/2013 - 15:41

Do you know what is the appropriate pair_style for sand or shale or clay simulation?

Thank you for the response in advance.