[UsamaQ]

Hello,

I am working on steady state simualtion of centrifugal compressor and I want to evaluate pressure ratio in CFD post. When I go to function calculator, I see average and massflowaverage values of pressure. Which one is right in this case? I checked one thread for transient simulation and he used massflowaverage of total pressure at inlet and oulet. However when I do this I get very high pressure ratio. When I use simple average of total pressure at inlet and outlet, I get more reasonable value of slightly greater than 3. which one is right?

One more question is I have a converged solution for one mass flow rate, how can i use this one for other mass flow rates so that simulation converges faster?

Thank you.

[Opaque]

A few questions:
1 - are you using a non-zero Reference Pressure?
2 - is there any recirculation at the boundaries?

Then, for the calculation:
1 - I would use massFlowAveAbs(Total Pressure in Stn Frame)@Boundary,
2 - I would include the Reference Pressure, just in case.

Ratio = (Reference Pressure + massFlowAveAbs(Total Pressure in Stn Frame)@Boundary 1) / (Reference Pressure + massFlowAveAbs(Total Pressure in Stn Frame)@Boundary 2)

If it does not match your expected value, there is something not accounted for:
1 - Well converged solution?
2 - Proper mesh independent solution?
3 - Relevant features included in the model?

The averaging type should not be a knob to adjust to match to data.

Hope the above helps

[UsamaQ]

Originally Posted by Opaque

A few questions:
1 - are you using a non-zero Reference Pressure?
2 - is there any recirculation at the boundaries?

Then, for the calculation:
1 - I would use massFlowAveAbs(Total Pressure in Stn Frame)@Boundary,
2 - I would include the Reference Pressure, just in case.

Ratio = (Reference Pressure + massFlowAveAbs(Total Pressure in Stn Frame)@Boundary 1) / (Reference Pressure + massFlowAveAbs(Total Pressure in Stn Frame)@Boundary 2)

If it does not match your expected value, there is something not accounted for:
1 - Well converged solution?
2 - Proper mesh independent solution?
3 - Relevant features included in the model?

The averaging type should not be a knob to adjust to match to data.

Hope the above helps

Yes, my reference pressure is set to 1 atm. Can you please specify what you mean by re circulation at boundaries? I have specified the inlet pressure but the converged solution indicates pressure rise in inlet domain as well. is this be the problem? I have double checked boundary conditions and everything is specified correctly. Could this be due to small inlet domain?

and for the solution yes it is converged and the results are not changing with further refinement.

I am using stage mixing model.

[UsamaQ]

Originally Posted by Opaque

A few questions:
1 - are you using a non-zero Reference Pressure?
2 - is there any recirculation at the boundaries?

Then, for the calculation:
1 - I would use massFlowAveAbs(Total Pressure in Stn Frame)@Boundary,
2 - I would include the Reference Pressure, just in case.

Ratio = (Reference Pressure + massFlowAveAbs(Total Pressure in Stn Frame)@Boundary 1) / (Reference Pressure + massFlowAveAbs(Total Pressure in Stn Frame)@Boundary 2)

If it does not match your expected value, there is something not accounted for:
1 - Well converged solution?
2 - Proper mesh independent solution?
3 - Relevant features included in the model?

The averaging type should not be a knob to adjust to match to data.

Hope the above helps

I just found out that the pressure is ridicolously high at outlet. I am not sure why is it sky rocketing and it is not changing with mesh refinement. I ran a simulation with coarse mesh and using same setup but the streamlines are very different in both cases. What could be the problem? I am really confused. My inlet pressure is 2.42 bar. But at outlet it is in e7 Pa so something is wrong. Can somebody help?

[Sumanth_094]

The streamlines usually have a starting point which you define and these are arbitrarily chosen in CFX based on number of elements on the starting surface and the number of streamlines you wanna visualize. They start at the cell center and since your meshes are not the same, the streamlines you will be visualizing are not gonna be the same as their starting points are gonna be different. I hope this helps.

Moving on to the pressure values, I am really not sure what is a practical solution in your case. You need to first do a mesh sensitivity study. May be that will give you some insight in this problem. Also, check literature and compare with these results.