[Seroga]

Hi, everybody!

I'm a newbie in CFD-simulations and faced a problem of imposing boundary conditions to a rotating axial rotor.
Where are two next questions:
1. What boundary conditions I have to impose on inflow and outflow boundaries? Inlet total pressure and static outlet? Or inlet mass flow and "I don't know what" outlet?
2. Must I correct the speed of rotor rotation depending on inlet mass flow? Or speed is constant and I just vary mass flow independently?

Thanks beforehand for your answers!

[mabelsfatalfable]

First assumption seems to be fine I guess but btw you should also try to take the rpm into account if you change the inlet mass flow. (since the compressor can choke if you vary the mass flow inside too much)

[rvl565]

Pressure Inlet( total pressure defined ) with a pressure outlet ( static pressure defined) works well for me.
The way you generate the PV curve ( constant RPM ) for the fan is by increasing the outlet static pressure. higher the outlet static pressure less is the massflow rate.

[Seroga]

Thanks Armağan and RacMat for your answers.

But what if I want to impose mass flow on he inlet boundary (some solvers allow to do so). What should I impose on the outlet boundary in this case? Static pressure or smth else?

[rvl565]

You can use a mass flow inlet with a pressure outlet to impose mass flow restrictions, i have also seen people use the "target mass flow rate" option in pressure outlet, I am not sure if that is correct.
You could just vary the outlet static pressure (pressure outlet + pressure inlet B.C.s) and see which outlet pressure gives you your correct mass flow rate.

[rvl565]

Originally Posted by Seroga

Thanks Armağan and RacMat for your answers.

But what if I want to impose mass flow on he inlet boundary (some solvers allow to do so). What should I impose on the outlet boundary in this case? Static pressure or smth else?

Hi,

Are you using rotor 67/37 by any chance?

[Seroga]

Originally Posted by rvl565

Hi,

Are you using rotor 67/37 by any chance?

I doubt about I've understood your question correctly. English is not my native language.

I wanted to carry out a flow simulation in Rotor 67 or 37. Now I've built a CAD model of Rotor 67 on my own and I'm going to simulate flow in it using OpenFOAM.

[rvl565]

I just wanted to know if you have the mesh for rotor 67.

[Seroga]

Originally Posted by rvl565

I just wanted to know if you have the mesh for rotor 67.

Actually I don't have the mesh yet. I'm working on it now.
Why don't you use *.tin mesh from https://grabcad.com/library/nasa-rotor-67-1 or this one: http://aero-comlab.stanford.edu/vdwe...ModTipGap.cgns

[rvl565]

Actually i tried using the stanford mesh... i modified it by removing the tip gap... but it isnt working for me ... heres the link to my post :
http://www.cfd-online.com/Forums/flu...on-fluent.html

I have included the cgns mesh file in a later post in the same thread....

I dont use ICEM so cant really use the first link...

Could you please let me know if you succeed?

[Seroga]

Sure, I will tell you if I succeed.
Tell me please about the mesh, you've used. Is it whole rotating? But as I can see it is larger than a rotating part of Rotor 67.

[rvl565]

I extended the inlet and outlet of the mesh by stanford.
It has 3 zones
inlet zone and outlet zone are stationary
rotor zone is rotating.
Have you started your simulations or are you building your mesh?

[rvl565]

hi,
were you able to make the mesh i shared with you work?

[Seroga]

Hi!
I can't use that mesh because I can't understand how to apply it to this problem. It's larger than rotating part of Rotor and I don't know how to define rotating and stating parts in it. Maybe this is because my little experience in this field.

I have already built my own geometry and built my own mesh. It is very coarse because I just want to find out would it work at all. But I faces a problem of cyclic boundary conditions in OpenFOAM and stuck a bit.

I will write to this post as soon as I will obtain some positive results.

[vijaymali]

Hi Seroga,

I hope you have received the answer to your original question about what boundary conditions you should you. I still want to give my opinion on that question.

The type of boundary conditions you should use, depends on the objective of the simulation (what data you want as a output) and what data is available with you.

In most of the turbo simulation problems that we do, we use mass flow rate at inlet and static pressure at outlet. We provide fix RPM to the rotor and extract what is the pressure developed at inlet. That way we can extract pressure rise, efficiency etc. for one point on the characteristic curve. For other points, we vary mass flow rate maintaining same RPM.

I have just gone through the later part of this discussion. I am curious to know what software you are using for meshing and what type (hexa or tetra mesh) of mesh you are generating ?

Thanks
Vijay
www.learncax.com

[rvl565]

Originally Posted by Seroga

Hi!
I can't use that mesh because I can't understand how to apply it to this problem. It's larger than rotating part of Rotor and I don't know how to define rotating and stating parts in it. Maybe this is because my little experience in this field.

I have already built my own geometry and built my own mesh. It is very coarse because I just want to find out would it work at all. But I faces a problem of cyclic boundary conditions in OpenFOAM and stuck a bit.

I will write to this post as soon as I will obtain some positive results.

hi,

I dont know much about OpenFOAM but in fluent i can set the RPM of the three zones (inlet, rotor and outlet) independently. Thus i set the rotor to have -16043RPM. I leave the other two zones as they are.

[rvl565]

Originally Posted by vijaymali

Hi Seroga,

I hope you have received the answer to your original question about what boundary conditions you should you. I still want to give my opinion on that question.

The type of boundary conditions you should use, depends on the objective of the simulation (what data you want as a output) and what data is available with you.

In most of the turbo simulation problems that we do, we use mass flow rate at inlet and static pressure at outlet. We provide fix RPM to the rotor and extract what is the pressure developed at inlet. That way we can extract pressure rise, efficiency etc. for one point on the characteristic curve. For other points, we vary mass flow rate maintaining same RPM.

I have just gone through the later part of this discussion. I am curious to know what software you are using for meshing and what type (hexa or tetra mesh) of mesh you are generating ?

Thanks
Vijay
www.learncax.com

Hi,

I use pointwise for meshing and generate a hex mesh in the whole domain.

As per your comment on boundary conditions, I have read at many places that for fan simulations people generally use pressure inlet and vary the outlet static pressure to generate a PV curve. I thought this method is correct? If not, what is the advantage of using a mass flow inlet over using a total pressure inlet?

[Seroga]

Originally Posted by rvl565

hi,

I dont know much about OpenFOAM but in fluent i can set the RPM of the three zones (inlet, rotor and outlet) independently. Thus i set the rotor to have -16043RPM. I leave the other two zones as they are.

Sure, I can do so for my mesh. Because I define that zones during the mesh generation. But I don't see that they were defined in this mesh and do I don't know how to define that zones for existing mesh.

[Seroga]

Thanks, Vijaymali, for your answer. It was helpful.

Originally Posted by vijaymali

Hi Seroga,
I am curious to know what software you are using for meshing and what type (hexa or tetra mesh) of mesh you are generating ?

I use Salome to build geometry and mesh generation. I use structured hexameshes.

[rvl565]

Originally Posted by Seroga

Sure, I can do so for my mesh. Because I define that zones during the mesh generation. But I don't see that they were defined in this mesh and do I don't know how to define that zones for existing mesh.

Hi,

I can generate meshes for OpenFOAM using pointwise. I generated the following mesh, let me know if this is working for you,

https://drive.google.com/folderview?...28&usp=sharing