[MGE_CFD]

Dear,

I am finishing an academic paper and I use CFX.

I am simulating a centrifugal pump rotor with the diffuser, the rotor being a rotating domain and the diffuser stationary, the interface is Stage. I am simulating only the blades (rotor with 6 blades and diffuser with 8 blades) using periodicity. With input condition I inform the pressure 0 Pa and the output mass flow, without problems in the setup!

The problem is that I did some tests modifying the pitch of the diffuser blades and the result does not change, alias the difference is around 0.5 mca in the pressure and 100 W in the power. I inserted totally divergent angles of the theoretical, almost blocking the passage of the fluid and the result does not change. Could you give me a tip?

Thanks

[MGE_CFD]

I do not know if I was clear on the question, but someone could point me some way to try to have differences in the simulation. Example: mesh, setup, etc ...

[ghorrocks]

FAQ: https://www.cfd-online.com/Wiki/Ansy..._inaccurate.3F

[MGE_CFD]

Thanks for the feedback.

I have already made the following changes: mesh refinement, k-epsilon simulation and SST, biphasic simulation with Rayleigh-Plesset model activated and values ​​do not change.

I do not know if I could be wrong in the interface with the care of the mesh, but also had no result.

[urosgrivc]

maybe a picture or a bit more info, like are you monitoring Torque on the rotor or what?
it is not posible that the result would stay the same unless something else is set that way.
maybe it is the stage interface.
Always remember that if something fishy is going on with the simulation it is more than 99% probability that it is the users fault, computers just turn numbers around.

[Jiricbeng]

Send the expressions you use for the evaluation.

[MGE_CFD]

Hi

Pe = massFlow(TotalPressure in Stn Frame)@Inlet

Ps = massFlow(TotalPressure in Stn Frame)@Outlet

Head = (Ps-Pe)/(998 [kg m^-3]*g)

Qo = massFlow

[MGE_CFD]

I made this post in the meshing and had no answer, I know that here is not the correct forum, but come on:

The centrifugal pump simulation, I have a rotating (rotor) domain and a stationary (diffuser), but I'm not sure of the mesh of the interface region. I'm using Stage in setup, and the two domains are not in a single body in design modeler, for the reason I'm out of control on the node of each interface. How do I check if the interface region has no influence on my result?

[ghorrocks]

How do I check if the interface region has no influence on my result?

With a sensitivity analysis. Take multiple meshes with significant differences between them and compare the results from the simulation. A typical example is mesh refinement. As the mesh is refined the results should converge to a value. You can then use the mesh refinement which gives you an accuracy you are happy with.

[Christophe]

I think the expression needs to be “massFlowAve” not massFlow