[BK90]

Dear Users,

I have to do little project on my studies, which is solidification model (2d with macrosegregation) of binary alloy (Pb 48-Sn) in Ansys 14.5 but i am new with ansys and i am bit stuck with it.

I read user`s and theory guide in ansys help about solidification modeling. All data are from articles (Prescott, Ahmad).
I enabled models: species transport, solidification/melting with thermal/solutal bouyancy and Lever rule. Mushy zone parameter was calculated with formula: A=(180*u)/k^2, where u=dynamic viscosity and k=secondary dendrite arm spacing.
In material properties i added pb,sn and mixture template properties:
for pb-molecular weight,slope of liquidus,partition coefficient(0.48), eutectic mass fraction (0.619), solutal expansion coefficient and 0for diffusion in solid.
for sn-molecular weight, 0 for slope of liquidius solutal expansion coefficient and diffusion in solid,partition coefficient, 1 for eutectioc mass fraction(default).
For mixture: density(9000 and default method is boussinesq),specific heat, thermal conductivity,viscosity, mass diffusivity,thermal expansion coefficient,puere solvent melting heat, solidus and liquidus temperature(mixing law),pure solvent melting temperature, eutectic temperature.
In operating conditions i enabled gravity and operating temperature (600,65K- melting temperature).
In cell conditions for surface-body is enabled porous zone for mixture and i didn`t change any of its parameters. It was default.

Boundary conditions:
Left wall is for cooling alloy and other 3 walls are adiabatic. For left wall i enabled convection and added heat trandfer coefficient and free stream temperature.
I chose SIMPLE scheme for computations and created volume monitor to check change of liquid fraction during computations.
I started calculations but they are wrong, because monitor shows that liquid fraction equals 0 during calculations.
I enclose a few screenshots.
I have no idea what to do to make this model work.

Thank you for all help.

[s8kthi]

Fluent expects all the material properties of the solute, It is not mandatory to fill the solvent properties.
In your case 'Sn' is the solute 'Pb' is the solvent isnt it.

Have you Patched the Initial temperature and Solute concentration fraction into the interior nodes. using Solution initialization tab.

Keep the operating temperature as T_liquidus.

[bobmarley@]

Dear user BK90

I am solving a similar problem, and not getting expected results.

Please let me know if you were able to short out the problem. Any help will be highly appreciated.

Thanks in advance.

[djamila77]

Originally Posted by BK90

Dear Users,

I have to do little project on my studies, which is solidification model (2d with macrosegregation) of binary alloy (Pb 48-Sn) in Ansys 14.5 but i am new with ansys and i am bit stuck with it.

I read user`s and theory guide in ansys help about solidification modeling. All data are from articles (Prescott, Ahmad).
I enabled models: species transport, solidification/melting with thermal/solutal bouyancy and Lever rule. Mushy zone parameter was calculated with formula: A=(180*u)/k^2, where u=dynamic viscosity and k=secondary dendrite arm spacing.
In material properties i added pb,sn and mixture template properties:
for pb-molecular weight,slope of liquidus,partition coefficient(0.48), eutectic mass fraction (0.619), solutal expansion coefficient and 0for diffusion in solid.
for sn-molecular weight, 0 for slope of liquidius solutal expansion coefficient and diffusion in solid,partition coefficient, 1 for eutectioc mass fraction(default).
For mixture: density(9000 and default method is boussinesq),specific heat, thermal conductivity,viscosity, mass diffusivity,thermal expansion coefficient,puere solvent melting heat, solidus and liquidus temperature(mixing law),pure solvent melting temperature, eutectic temperature.
In operating conditions i enabled gravity and operating temperature (600,65K- melting temperature).
In cell conditions for surface-body is enabled porous zone for mixture and i didn`t change any of its parameters. It was default.

Boundary conditions:
Left wall is for cooling alloy and other 3 walls are adiabatic. For left wall i enabled convection and added heat trandfer coefficient and free stream temperature.
I chose SIMPLE scheme for computations and created volume monitor to check change of liquid fraction during computations.
I started calculations but they are wrong, because monitor shows that liquid fraction equals 0 during calculations.
I enclose a few screenshots.
I have no idea what to do to make this model work.

Thank you for all help.

Dear user BK90

I am solving a same problem, and not getting expected results.

Please help me.

Thanks in advance.

[nguyentbtuyen]

Hello experts

I hope you are now an expert in this field already.

The solidification model in Fluent already treats the mushy zone as porous medium - by an additional sink term. But I am wondering what should happen if the "porous zone" is enabled? For me this option should be disable by default when Solidification model is turned on. Please help me to be clear at this point.

Thank you very much

Tuyen

~~~~~~~~~~~~~~~~~~~~~~~~

Originally Posted by BK90

Dear Users,

I have to do little project on my studies, which is solidification model (2d with macrosegregation) of binary alloy (Pb 48-Sn) in Ansys 14.5 but i am new with ansys and i am bit stuck with it.

I read user`s and theory guide in ansys help about solidification modeling. All data are from articles (Prescott, Ahmad).
I enabled models: species transport, solidification/melting with thermal/solutal bouyancy and Lever rule. Mushy zone parameter was calculated with formula: A=(180*u)/k^2, where u=dynamic viscosity and k=secondary dendrite arm spacing.
In material properties i added pb,sn and mixture template properties:
for pb-molecular weight,slope of liquidus,partition coefficient(0.48), eutectic mass fraction (0.619), solutal expansion coefficient and 0for diffusion in solid.
for sn-molecular weight, 0 for slope of liquidius solutal expansion coefficient and diffusion in solid,partition coefficient, 1 for eutectioc mass fraction(default).
For mixture: density(9000 and default method is boussinesq),specific heat, thermal conductivity,viscosity, mass diffusivity,thermal expansion coefficient,puere solvent melting heat, solidus and liquidus temperature(mixing law),pure solvent melting temperature, eutectic temperature.
In operating conditions i enabled gravity and operating temperature (600,65K- melting temperature).
In cell conditions for surface-body is enabled porous zone for mixture and i didn`t change any of its parameters. It was default.

Boundary conditions:
Left wall is for cooling alloy and other 3 walls are adiabatic. For left wall i enabled convection and added heat trandfer coefficient and free stream temperature.
I chose SIMPLE scheme for computations and created volume monitor to check change of liquid fraction during computations.
I started calculations but they are wrong, because monitor shows that liquid fraction equals 0 during calculations.
I enclose a few screenshots.
I have no idea what to do to make this model work.

Thank you for all help.

[nguyentbtuyen]

Hello experts,

should we enable "porous zone" when solidification is already turned on? as the mushy zone is already treated as a porous media by adding a sink term in the energy equation. If not, why the "porous zone" is still not disable in the "Cell Zone Conditions"?

Any suggestions would be much appreciated!

-----sorry if this question is posted twice, as I could not see the first one----

Originally Posted by BK90

Dear Users,

I have to do little project on my studies, which is solidification model (2d with macrosegregation) of binary alloy (Pb 48-Sn) in Ansys 14.5 but i am new with ansys and i am bit stuck with it.

I read user`s and theory guide in ansys help about solidification modeling. All data are from articles (Prescott, Ahmad).
I enabled models: species transport, solidification/melting with thermal/solutal bouyancy and Lever rule. Mushy zone parameter was calculated with formula: A=(180*u)/k^2, where u=dynamic viscosity and k=secondary dendrite arm spacing.
In material properties i added pb,sn and mixture template properties:
for pb-molecular weight,slope of liquidus,partition coefficient(0.48), eutectic mass fraction (0.619), solutal expansion coefficient and 0for diffusion in solid.
for sn-molecular weight, 0 for slope of liquidius solutal expansion coefficient and diffusion in solid,partition coefficient, 1 for eutectioc mass fraction(default).
For mixture: density(9000 and default method is boussinesq),specific heat, thermal conductivity,viscosity, mass diffusivity,thermal expansion coefficient,puere solvent melting heat, solidus and liquidus temperature(mixing law),pure solvent melting temperature, eutectic temperature.
In operating conditions i enabled gravity and operating temperature (600,65K- melting temperature).
In cell conditions for surface-body is enabled porous zone for mixture and i didn`t change any of its parameters. It was default.

Boundary conditions:
Left wall is for cooling alloy and other 3 walls are adiabatic. For left wall i enabled convection and added heat trandfer coefficient and free stream temperature.
I chose SIMPLE scheme for computations and created volume monitor to check change of liquid fraction during computations.
I started calculations but they are wrong, because monitor shows that liquid fraction equals 0 during calculations.
I enclose a few screenshots.
I have no idea what to do to make this model work.

Thank you for all help.

[vinerm]

Though both are treated in a similar manner, the difference comes from the domain extent. When porous zone option is enabled, this is applicable throughout the cell zone. Of course, the resistance coefficients could be function of space. And this is what happens in solidification model. In the beginning, whole zone may be liquid or solid but slowly the size will change. If it is melting, then the porous zone size will keep on reducing. Therefore, solidification model is enabled separately and not via standard porous zone approach. However, if user wants, he can use it using porous zone and define own profiles for resistance coefficients.

[nguyentbtuyen]

Thanks Vinerm. I can see the settings required as porous zone is selected.
That also means turning on both solidification and porous zone at the same time - like the top post described - is not appropriate!
Thanks again
Hope

[vinerm]

Both may be enabled in certain situations. E.g., in an industrial iron-carbon system, there are always inclusions that create resistance to the flow even when it is not solidified. In that case, both could be used as well. It all depends on morphology.

[nguyentbtuyen]

Hi Vinerm,

I get the point now. Thanks a lot for you comments.

Cheers, Tuyen

[gikamc]

Hello! I am currently involved in a similar project but in my case there is a binary Al-Mg alloy. I have already implemented a casting procedure which gives reliable results and converges well, using the solidification and melting model. I want to compute the macrosegregation of Mg along the cast. Could you please help me to set the model for my case ? Is Spiecies Transport Model suitable for such simulation ?

[vinerm]

Fluent has inbuilt models for microsegregation, however, as far as macrosegregation is concerned, you will have to involve extra physical phenomena to model or simulate it. Simulation will involve prediction of dendritic growth and could be doable using VOF or Dynamic mesh. Modeling of macrosegregation instead of simulating it directly would require some developed model. You can do some literature survey if someone has implemented it in their own code or in some other CFD tool. Then, you can use UDF to implement it.

[gikamc]

Originally Posted by vinerm

Fluent has inbuilt models for microsegregation, however, as far as macrosegregation is concerned, you will have to involve extra physical phenomena to model or simulate it. Simulation will involve prediction of dendritic growth and could be doable using VOF or Dynamic mesh. Modeling of macrosegregation instead of simulating it directly would require some developed model. You can do some literature survey if someone has implemented it in their own code or in some other CFD tool. Then, you can use UDF to implement it.

Thank you for your prompt response!



Could you please guide me a little bit because i am confused about parameters required? I have performed a simulation of DC Casting, could i simulate microsegregation in the mushy region ? I can compute the velocity, temperature field and mass fraction from my current simulation.

[vinerm]

You can model microsegregation using either Scheil law, or Lever rule, or finite back diffusion. All these are available within Fluent.

[gikamc]

Originally Posted by vinerm

You can model microsegregation using either Scheil law, or Lever rule, or finite back diffusion. All these are available within Fluent.

It is very difficult for me to understand how to set up it via the Fluent's Manual. Could you please share with me any example related to microsegregation ?

[vinerm]

I am afraid I do not have any examples but you would many online. Just search for Scheil or Lever rule with Fluent. You may also have a look at the

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4986388/

[gikamc]

Originally Posted by vinerm

I am afraid I do not have any examples but you would many online. Just search for Scheil or Lever rule with Fluent. You may also have a look at the

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4986388/

Thank you very much for your help! If you are interested in teaching me how i can implemented via a paid lesson please contact me.

[vinerm]

I do not offer any paid lessons and would be very expensive as a teacher. So, better learn from me via Forum.

[nguyentbtuyen]

Hi Vinerm again,

I am changing from 2D to 3D mesh of the same problem, i.e. binary alloy solidification.
The temperature was shown as expected, flow field is okay! properties checked! but the liquid fraction and species concentration does not change at all.

Could you advise what should I pay more attention to deal with this problem please?
Thank you
Tuyen

[nguyentbtuyen]

Is the residual problem? I have reduced the scaled residual to E-07 but see they are still going down sharply (not flatten out with iterations). Time step is 0.005, mesh size is ~2mm. OR mesh/under-relaxation problem?

Your advice would be much appreciated

Kind regards
Tuyen