Solid Fluid Interface

  • Last Post 27 November 2018
Bibek Poudel posted this 13 November 2018

Hi everyone, 

I am doing a simulation in Ansys Fluent where I have cylindrical fluid surface ( a suspension, with mixture of iron, cbn and silicon oil) with very high viscosity. It is in contact with the solid surface ( a steel) in below. I am giving magnetic field from top so that all the mixture should be held together and retain the cylindrical geometry. But the fluid here is flowing in the surface of the solid surface below. How can I define the lower boundary of fluid cylindrical geometry also to be reflecting surface? Is there any way that I can define the fluid geometry to be as reflecting wall. I can define the other walls of cylindrical geometry as a reflecting wall but Fluent didn't give me option for defining the lower surface? What can I do?? 

Thank you for the help in advance!! 

Order By: Standard | Newest | Votes
Kremella posted this 13 November 2018


What models are you using to solve your problem? What is your end goal?

Also, could you please share a screenshot of your geometry? This will certainly help us understand your geometry better.

Thank you.

Best Regards,


kkanade posted this 14 November 2018

Please insert image of geometry, mesh and set up. Please note that as ANSYS employee, we can not download attachments. 

Bibek Poudel posted this 14 November 2018

Hi everyone, 

Thanks for the reply. What I am doing is the simulation where I want to track the iron and cbn particles injecting them in a Si-oil. There is a mixture of Si-oil, iron particle and cbn particle in the cylindrical geometry (but a fluid geometry, not a solid geometry) . We are trying to rotate that cylindrical model and give the magnetic force from the top. Alongside, there is a solid steel plate in contact with the lower face of cylindrical slurry mixture of Si-oil, iron and cbn as shown in figure. 



Now the problem arose when I tried to insert the silicon oil in the cylindrical fluid geometry. The Si-oil began to flow to the steel surface. It's like the lower boundary wall of cylindrical surface is open; so all the Si-oil are flowing out to the steel surface like the picture below: 

What I wanted was to insert the Si-oil until cylindrical geometry was full. Then I wanted to inject iron and cbn particles and then rotate the cylindrical geometry keeping the lower steel surface stationary until the Si-oil, iron and cbn particles mix homogeneously. And then insert magnetic force from the top of cylindrical geometry. And throughout the process, I wanted to track the iron particles. 

So, is there any chance I can block the lower wall of cylindrical surface so that Si  is not spilled. How can I define the lower wall of cylindrical surface which is in contact with steel surface in the bottom? If there was no steel in the bottom I could've changed that surface as I need; like a stationary wall or moving wall; reflecting wall or trapping wall; but now it seems like I can not change anything. So, could  you guys please help me with this? 

Thank you in advance,

Kremella posted this 19 November 2018

Hello Bibek,

Apologies for our late response. 

Could you please share a screenshot of your boundary conditions? I am specifically interested in the boundary condition at the interface. Also, do you have any mesh interfaces between the cylindrical geometry as well as your steel plate? If so, could you please share the screenshot of this as well.

Thank you.

Best Regards,


Bibek Poudel posted this 20 November 2018

Hi Karthik, 

My inlet and outlet are both default conditions with zero inlet velocity and zero gauge pressure at outlet. Because I have already inleted Si-oil and then used DPM to inject iron and cbn particles, right now they are all set with default boundary conditions. In mesh interface, it is a coupled wall as I have fluid cylindrical geometry in the top and solid steel workpiece in the bottom. Also I am intend to use frame motion and mesh motion for all Si-oil, iron particles and cbn particles to mix homogenously.


So, I don't know where I am running wrong. I hope you can help. 

Thank you, 

kkanade posted this 21 November 2018

Apart from boundary condition, I have couple of thoughts. 

You can patch the cylindrical volume with the material you want so that you have cylinder full with that material. 

Then in injection you can use file injection to get homogeneous mixture. 

kkourbat posted this 21 November 2018

display the same contours of VOF, but with Node Values unchecked in the contour panel? This will show cell centered values calculated by the solver. This is a sanity check to very it is not a post-processing artifact due to node interpolation.

Bibek Poudel posted this 26 November 2018

Hi, Thanks for your reply. But whenever I do that too, during the injection time, the fluid already begin to slide inside the workpiece under. It doesn't remain within the cylindrical boundary. 


Bibek Poudel posted this 26 November 2018

Hi, Also, how can we use file injection?? I saw many people saying use this: 

((x y z u v w diameter temperature mass-flow) name )

But how do you use it? I mean which format do you save? Can I just type like this in notepad and save it in txt. or inj. format or what?? It would be great help if you tell me how to use this. 


kkanade posted this 27 November 2018

yes, save it in inj format.