[Hans John]

Hi guys

I am currently conducting the numerical simulation about the ducted fan. And the case from the reference "Grunwald, K. J. and K. W. Goodson (1962). Aerodynamic loads on an isolated shrouded-propeller configuration for angles of attack from -10 degrees to 110 degrees. WASHINGTON, NASA." is simulated. I have generated unstructured mesh and structured mesh by ICEM CFD. yplus approximately equals to 1. The solver is Fluent. The turbulent model is k-omega SST. The ambient condition is applied. The boundary condition is pressure inlet and pressure outlet. I have done the cases at axial flow 8000 rpm. But the relative error is 33%! I have tried many times and results are similar.

I don't know what is wrong with me. If you are conducting simulation of ducted fan, have you done the case from the reference and how can you verify the case? Sometimes I wonder the experiment result may be wrong. I have been tortured for two months. Help me please! Thank you in advance.

My E-mail: hhanabc@gmail.com

[FMDenaro]

Originally Posted by Hans John

Hi guys

I am currently conducting the numerical simulation about the ducted fan. And the case from the reference "Grunwald, K. J. and K. W. Goodson (1962). Aerodynamic loads on an isolated shrouded-propeller configuration for angles of attack from -10 degrees to 110 degrees. WASHINGTON, NASA." is simulated. I have generated unstructured mesh and structured mesh by ICEM CFD. yplus approximately equals to 1. The solver is Fluent. The turbulent model is k-omega SST. The ambient condition is applied. The boundary condition is pressure inlet and pressure outlet. I have done the cases at axial flow 8000 rpm. But the relative error is 33%! I have tried many times and results are similar.

I don't know what is wrong with me. If you are conducting simulation of ducted fan, have you done the case from the reference and how can you verify the case? Sometimes I wonder the experiment result may be wrong. I have been tortured for two months. Help me please! Thank you in advance.

My E-mail: hhanabc@gmail.com

many reasons can drive to discrepancy with experimental measures

1) wrong experimental measurements!
2) The numerical geometry and boundary conditions do not reflect the real experiment.
3) different use of the statistical variables to compare.

and some less relevant issues are always possible....

Therefore, as general guidelines:
1) try refining the grid and see the variation in the solutions
2) try different turbulence modelling
3) check carefully into experiment details that can be different from your computational setting

[Hans John]

Originally Posted by FMDenaro

many reasons can drive to discrepancy with experimental measures

1) wrong experimental measurements!
2) The numerical geometry and boundary conditions do not reflect the real experiment.
3) different use of the statistical variables to compare.

and some less relevant issues are always possible....

Therefore, as general guidelines:
1) try refining the grid and see the variation in the solutions
2) try different turbulence modelling
3) check carefully into experiment details that can be different from your computational setting

The reference details the geometry and the flow condition. The boundary condition is pressure inlet/pressure outlet. I have generated unstructured mesh and structured mesh and I have run the cases many times. The statistical variables to compare is just the Cx, CT,Tp/T and so on.

I have a question whether and how you should verify the numerical simulation. Do you have some validations about the ducted fan. The file exceeds the forum's limit. If you want to read it, I can send to your e-mail. Thank you very much.

Hans John

[FMDenaro]

Originally Posted by Hans John

The reference details the geometry and the flow condition. The boundary condition is pressure inlet/pressure outlet. I have generated unstructured mesh and structured mesh and I have run the cases many times. The statistical variables to compare is just the Cx, CT,Tp/T and so on.

I have a question whether and how you should verify the numerical simulation. Do you have some validations about the ducted fan. The file exceeds the forum's limit. If you want to read it, I can send to your e-mail. Thank you very much.

Hans John

I never solved such problems.
Have you compared the velocity fields?