[zese]

Dear all

I want to simulate a compressible flow passing a blade of turbine in 2D. I chose Ideal gas air as the fluid. the inlet velocity is 1421 m/s with 17.23 degree and total pressure 1.3 Mpa. the geometry and BC are shown in the pic. when I put supersonic as the flow regime of inlet, after 10 iterations it failed because of fail in P-MASS equation.
when the I choose lower velocity with subsonic or mix flow regime the equations are solved perfectly.
because it is the first time I am simulating the compressible flow, I dont know what I can do to solve it.


Best regards

[Opaque]

From the snapshot, it seems you are using symmetry planes (2D representation), translational periodicity top and bottom, and "an opening" at the exit from the arrows.

Is that correct? Openings are subsonic, are they not? What about using a supersonic outlet as well?

[zese]

Originally Posted by Opaque

From the snapshot, it seems you are using symmetry planes (2D representation), translational periodicity top and bottom, and "an opening" at the exit from the arrows.

Is that correct? Openings are subsonic, are they not? What about using a supersonic outlet as well?

Dear Opaque,

Thanks for your response. yes it is correct. I did what you said and the same happened. the pic I attached shows the massage of CFX. I dont know what else I can do

[ghorrocks]

There is an FAQ on this: https://www.cfd-online.com/Wiki/Ansy...do_about_it.3F

I recommend you read it as it has the general advice for this situation.

[Opaque]

Originally Posted by zese

Dear Opaque,

Thanks for your response. yes it is correct. I did what you said and the same happened. the pic I attached shows the massage of CFX. I dont know what else I can do

Have you tried running this particular configuration in the subsonic regime first to get a better understanding of how to solve a simpler problem?

0 - Read Glenn's suggestion, then
1 - Subsonic flow
2 - Initial guess? is it reasonable? or just a wild guess? Suggestion: assume constant total pressure/total temperature throughout, then you are only guessing the velocity field.
3 - You have not specified what timestep was selected.
4 - Have you read the documentation to understand the feedback in the output file? That is, what do all those numbers for the different equations mean?