[anirudster]

We know, Aluminium sample (solute) which is having a lower melting point compared to Titanium should start melting when poured in the titanium melt which might be at 2000 K. I need to do this fluid flow simulation in Ansys CFX 12.1.

I have done the geometry for both and are connected to each other - meshing is done too - Imported in CFX and started to finalize the boundary conditions, but got stuck.

Heat is directed from the top (ex - Electron beam gun)

In a molten titanium puck, we have inserted the thin aluminium bar from top, dipped from 3 sides in the melt.

Guess, its somehow manageable understand.

So, what I have done for the boundary conditions:

Aluminium solid:
Front and Back face - Symmetry
Top - Wall

Titanium melt:
Front and Back face - Symmetry
Top left - opening
Top right - opening
Bottom - wall
Left and right - wall
Cut right, Cut left, Cut Bottom - Interface


I have defined the material properties as constant for now, to keep things simple. I just want to simulate the melting process for now and for which I am concerned about my boundary conditions being set right.

Can someone please help me with setting me the right boundary conditions for my problem?

Thank you so much!

[ghorrocks]

If your question is urgent and important you should be talking to CFX Support. You get what you pay for.

So what are you trying to model and what do you hope to learn from the simulation?

[anirudster]

Hi Glenn

The intension is to simulate the melting process of Aluminium solid at 25 deg. C into Titanium liquid at 1800 deg. C. I have basic doubts in boundary condition and the material properties I should set.

Initially, to keep things simple and to have least amount of complexities I was wondering if I can keep the material properties for both the material as constant. But I am in doubt, doing this what would be the driving force for the Aluminium to diffuse towards the Titanium melt? If I add buoyancy, surely Al being lighter would move up and Ti being denser would sink down. I will add buoyancy factor to it but I am in doubt in simulating this melting process in CFX which should give me results, which can be comparable to my experimental results, I would do the experimentation later - once I get the model work to a certain extent.

Thanks!
Ani

[ghorrocks]

The intension is to simulate the melting process of Aluminium solid at 25 deg. C into Titanium liquid at 1800 deg. C.

But why? Are you interested in how quickly it melts? Or how the Al diffuses into the Ti? Or the motion of the Al block? Or do you just want some pretty pictures?

[anirudster]

I am interested to learn how Al diffuses into Ti, I want to quantify the dissolution process, the kinetics of it. Like, if I put a piece of aluminium into the melt, how the flow pattern would change, how Al going diffuse into the melt etc. etc.

Thank you!

[ghorrocks]

Can you do this ignoring the solid Al? That is, have the outer skin of the Al as a boundary in the Ti? Or do you need to model what is going on inside the solid Al, or does the solid Al move around?

[anirudster]

Ya, we want to know how Al breaks down, how to flows through the material; also we want this Al specimen to be there for some time and then removed, say for 10-15 sec and then removed, just to see how quickly it melts/evaporates etc etc. and also want to measure the amount of Al which melted over that time when exposed to this 2000 K hot Ti melt.

Yesterday after posting this query, I was trying to simulate based on whatever I know, the results I obtained are weird. I have doubts in the boundary conditions I am setting and also a bit concerned about the material properties. I am doing by keeping all the material properties constant but I guess, I need to input temperature dependent variables.

What my simulation results telling: Ti temperature drops from 2000K to 300K in 1 second; Al specimen is at a lower temperature as well. There is no flow of Al into Ti melt. Ah! upset and worried. Lets see what can be done.

I am not finding any tutorials or good literature where they explain about the modeling aspect of this. If you have some idea or have come across any paper, please introduce them to me. They would really be of great help.

For first week, I just want Al to melt in Ti and then I can go on to increase the complexities in it, as week passes.

Anyways, Thank you so much Glenn for listening to me.

[ghorrocks]

Well you have a very difficult and complicated CFD adventure ahead of you as the physics you describe is very complicated. This is a research topic which will probably take a few years to get working well. I hope you have lots of time to do this project.

[houwy]

Originally Posted by anirudster

Ya, we want to know how Al breaks down, how to flows through the material; also we want this Al specimen to be there for some time and then removed, say for 10-15 sec and then removed, just to see how quickly it melts/evaporates etc etc. and also want to measure the amount of Al which melted over that time when exposed to this 2000 K hot Ti melt.

Yesterday after posting this query, I was trying to simulate based on whatever I know, the results I obtained are weird. I have doubts in the boundary conditions I am setting and also a bit concerned about the material properties. I am doing by keeping all the material properties constant but I guess, I need to input temperature dependent variables.

What my simulation results telling: Ti temperature drops from 2000K to 300K in 1 second; Al specimen is at a lower temperature as well. There is no flow of Al into Ti melt. Ah! upset and worried. Lets see what can be done.

I am not finding any tutorials or good literature where they explain about the modeling aspect of this. If you have some idea or have come across any paper, please introduce them to me. They would really be of great help.

For first week, I just want Al to melt in Ti and then I can go on to increase the complexities in it, as week passes.

Anyways, Thank you so much Glenn for listening to me.

hello! I am working on your problem now. I want to know that have you sovle your problem?