Learning Forum

[datou]

Hello everyone, I am using the dynamic mesh method to simulate the dipping process of a pipe into hot melts. When the pipe is dipped into the melts, it will be melted due to high temperature, and the melted parts will flow unrestrictedly in the melt.

I am facing the following problems:

1. if I set the pipe as a solid zone, the UDF "DEFINE_CG_MOTION" can well describe the movement of the pipe, however, I can not simulate the melting process of the pipe (melting and solidification model needs fluid zone setting);
2. if I set the pipe as a fluid zone, the UDF "DEFINE_CG_MOTION" can not give the velocity of the pipe zone, and the pipe zone will deform, and can not keep its shape.

Could you give me some comments how to do this case? Thank you in advance.

[datou]

Does anyone help me. please, please, please...............

[Rob]

If you set the pipe as solid the solidification/melting model won't work. That's for fluid regions.
If you read the theory on the solidification/melting model you'll see it freezes the flow in those cells. That's not going to react well with the mesh motion.
So, what is the purpose of the model, ie can you assume the pipe motion is fast compared to the melting time? If so, why not start with an immersed pipe? If it's melting instantly can you use a material & energy source term? Can you move the liquid up and leave the pipe fixed?

As an aside, the system catches missed threads after a couple of days. Bumping a thread before that script runs means you may be waiting for another period if no one spots it.

[datou]

Dear Rob, Thank you again for your reply. I used a new thread, because this topic is different with my previous one.
This model, I want to simulated this process: the moving pipe solidification and melting in a hot melt, the melted parts will follow in the melt (momentum equation works), but for the un-melted part, it still keeps its original velocity (DEFINE_CG_MOTION works).
The reality is that: during the pipe motion, the pipe has been partly melted.
Yes, I agree with you to use a material & energy source term. However, I found, the method easily causes the solution unsteady, that can not converge well.
I also test some other methods, such as directly fixing the pipe velocity. But when I fixed the pipe motion, the melted parts flow in the melt can not been simulated. Even I used some "if" statement to control the open and shut of the fixed velocity. I guess, once we use the fixed velocity, it will work in the whole modelling. Please correct me, if I am incorrect.
I wonder, if there are any other robust methods for simulating a similar process? something like controlling the switch from a solid zone to a fluid zone according to a specific "if" statement. If so, we can specify the un-melted part surface as a solid wall, while set the melted part wall as a interior (fluid zone).
Thank you in advance.

[Rob]

You can shrink and covert the moving solid to a liquid but that'll need additional UDFs on top of the motion. It's not as simple as an IF statement as you're going to be working on the cell level. It's doable, but not easy, and not something I can cover in any detail because of the forum rules.

[datou]

Dear Rob Thank you for your confirmation about this point.
Could you provide a little more general information, such as which UDF, or where can I to get more information about this.
I don't need your detailed method or any coding. please just give me a general information that I can refer to.
I have go through many times of UDF helps, but I can not find any UDF macros that can covert the moving solid to a liquid.
My method is to control the momentum equation to work or not according to the melted condition. I believe this method is not the one you mentioned.
Thank you very much.

[Rob]

You'd need to deform the mesh as the solid melts and add fluid into the bit where the solid was. Source terms might work, but you'll need to be careful with how to remove the last little bit of solid. Switching specific cells from solid to fluid won't be easy, and I don't know how you'd do that. Fixing a fluid shape so the solid was a fluid that moves at a set rate might be easier?