Crystal_Growth with Phase Change

Numerical methods and mathematical models of Elmer
Post Reply
Roland
Posts: 226
Joined: 12 Apr 2018, 11:29
Antispam: Yes

Crystal_Growth with Phase Change

Post by Roland »

Hi,
In a former topic (Transient Phase Change Solver) I worked on the setup of a crystal growth 2D axisymetric transient model with use of the Transient Phase Change Solver(many thanks for your help, Peter, Kevin and Rich!)
After some progresses the model works now over 20s and shows the upper solid crystal growing.
Here attached are an animation over 20s and the zipped GUI Elmer project.
For the moment there is no steady state reached since there is no driving of parameters like the hot boundary temperature(right vertical liquid wall) or the "v_cast" vertical upward pulling velocity (which is imposed on the upper horizontal boundary, with the assumption that the solid phase (the upper solid pulled crystal) is considered as a liquid with a very huge dynamic viscosity (7e10 Pa*s!) so that it behaves like a solid). This is why afer 20s the triple point moves so that the crystal width narrows and the mesh in this area flattens with some nodes overlaps.
In order to reach a stabilized steady state I would like to use the "Pull Rate Control" and "Triple Point Fixed" command lines in the Transient Phase Change Solver. I made a few tries but it does not seem to work up to know.
It would be very wellcome if somebody could take a look at the current attached model and give some help to have it work with the former "Pull Rate Control" and "Triple Point Fixed" in order to solve over a longer time and reach the steady state (for the moment it solves over 20s but with very small time steps of 0.01s which requires very much memory and a long solve time if we want to store over longer physical times like 1000s required in such crystal growth processes !).
Thanks in advance for your help
Roland
Attachments
CG_04.zip
(612.04 KiB) Downloaded 33 times
CG_04_animation.gif
(959.53 KiB) Not downloaded yet
annier
Posts: 1168
Joined: 27 Aug 2013, 13:51
Antispam: Yes

Re: Crystal_Growth with Phase Change

Post by annier »

Dear Roland,
The use of mesh update solver is the limiting factor in my guess.
It might be useful for shape change or mesh deformation , that happens at a small rate.

I do not know whether the eulerian method such as "Level-set solver" is well suited for your case.
There used to be an issue with re-initilialization procedure.

You can give it a try.

Best Regards,
Anil
Anil Kunwar
Faculty of Mechanical Engineering, Silesian University of Technology, Gliwice
Roland
Posts: 226
Joined: 12 Apr 2018, 11:29
Antispam: Yes

Re: Crystal_Growth with Phase Change

Post by Roland »

Hello Anil,
Thank you for your answer. As you suggest it I can try to make a model with the Level-Set solver, but the wellknown drawback of this method is that it requires a very fine mesh around the moving phase change interface.
We have developped some models with Comsol and they begin now to work quite well as you can see in the following attached animation.
Crystal_Growth.gif
Crystal_Growth.gif (400.49 KiB) Viewed 513 times
This is why we would like to get the same kind of result with Elmer.
It would be very wellcome if you (or anybody who could be interested) could bring some ideas about how to improve this kind of crystal growth model.
Thanks in advance!
Roland
Post Reply