[mahmoudtarek0]

We are a group of mechanical engineering students making a cfd study on a small centrifugal pump whose impeller diameter is 50mm. We treated the blades as walls and rotated the fluid domain in a clockwise direction in this view. As you can see the velocity in the direction of rotation is less than the velocity in the other direction which is not a logical solution.

The turbulent model is SST-kw and in a steady state condition.

Steps used in our study were conducted from cfx tutorials. We want to know what can cause this problem, any help is appreciated.

[AtoHM]

I assume this will be helpful: what's the mean of the "velocity in Stn Frame" ?

[Gert-Jan]

You are looking to the variable "Velocity. "This is relative to the local coordinate. So for rotating domains, this Velocity is the velocity relative to the rotating speed.
To obtain more logical results, from a view point for you as stationary observer, you should look to the global "Velocity in Stn Frame"



This should be added to the FAQ..........

[mahmoudtarek0]

But is it logical that the relative velocity in the direction of rotation is less than the opposite direction or is this a problem in our procedure?

I attached an image for stationary frame velocity and it's more logical. Thank you.

[ghorrocks]

This is in the FAQ (https://www.cfd-online.com/Wiki/Ansy...f_reference.3F), but it focusses more on the fact it results on discontinuous streamlines at the interface.

So feel free to add some comments on this to the existing FAQ, or add a new FAQ question specifically on this. This question gets asked on the forum enough that it should be in the FAQ and explained clearly.