[ghorrocks]

You cannot assess mesh quality by a few images, or simple mesh quality statistics. You need to do a sensitivity study to find out in your case whether it is OK.

But I see there is large mesh size jumps from the last inflation layer to the bulk mesh. That is an area to look at if your current mesh is inadequate.

I did not mention velocity distribution.

You should check:
* Mesh size
* Mesh quality if you think that is an issue.
* Time step size (if transient)
* convergence tolerance

[turbo]

Your mesh looks to me much more than enough. For the other questions, I do not think this forum is for free engineering consulting of turbomachinery.

[Gape]

The mesh looks Ok?

And how works 'ramp up'... I change only the rotating speed? Or the inlet and outlet condition too?

[Gape]

is it better to build up the mesh with ansys icem cfd instead ansys cfx? e.g. with hexa mesh ?

I do not understand why it do not come to convergence.

[jb pouillard]

Dear Gape,

Do you have the expression for the NPSR(r) in CFX ?

If I use the CFX tutorial expression I can't compare it with the pump provider data.

[jb pouillard]

Dear Gape,

I currently working on a centrifugal pump optimization too.
I understood that contours of pressure and velocity have to do in Stn Frame .
So you have to plot a contours of velocity in stn frame, you will find it on the variables, you will see that the profile will look like so much more physical

Maybe you will need to compute a NPSH required , for the moment I didn't find the solution. If you have information plese let me know.

Best regards

[Gape]

I think the Problem is the mesh.

Tell me what ist the...

Maximum of aspect ratio and how can I degree it. I did not unterstand the expression <100^a

And mesh Orthogonality

The solver Tell me there is an "!", I think that is not good

[Gape]

Hi,

another problem, why Ansys do not mesh the inflation about the corner? Ansys makes always stair steps! I can not solve this problem