CFD - Centrifugal Pump help
 

Hi I'm new to CFD and I've been struggling with performing CFD for this centrifugal pump model for the past 3 months. I've done my absolute best to simplify the model (I received the whole pump model with the volute casing, motor housing, etc). I have viewed almost every youtube video related to CFD of a pump and have learnt a bit of Ansys fluent.

So my main problems are
- It is not converging. The omega residual tends to oscillate heavily and the continuity residual never goes below 10e-3.
- The pressure values are abnormally high. My inlet and outlet pressures are of the order of 10e9 and 10e10 respectively. If I calculate the head I am a value of 2500000 m (2.5 million).
And these ranges of order are consistent in each of my attempts even after changing geometries, boundary conditions, etc.

I have attached all the photos below so you can see my mesh, boundary conditions, solution methods, and my results.

My procedure below :-

I reduced my model which had the impeller, impeller casing, etc to only two fluid domains (That was the only way I was able to reduce the skewness below 0.9). A rotating impeller domain and a stationary volute domain.

I believe my mesh is good since I've managed to reduce the max skewness to 0.85.

For boundary conditions, I gave a pressure inlet (0 pressure gradient) and a mass flow outlet (2.03 kg/s). The pump I'm working on is a 110 LPM pump with the working fluid being ethylene glycol. So to get the mass flow rate I converted 110 LPM into Kg/s to get 2.03 kg/s.

I made the impeller domain rotate at a speed of 5000 rpm (in negative since my axis is inverted). I set the impeller blades, hub and walls as moving walls with a relative speed of 0 wrt adjacent cell zone.

I set the solution method as shown in the picture. I used various youtube videos as reference. I have actually tried various methods and have tried solving innumerable times before this. I am just showing what I obtained in my final attempt as I don't know what to change or improve anymore.


Please guide me on what I have to change or incorporate. Thank you.

Hi,
Did you set the inlet pressure to 0 Pa static pressure? If this is the case then something is strange with your pressure results. Did you change the reference pressure for some reason or maybe set a pressure for the outlet? The inlet static pressure should be 0 Pa, if you defined it so..
Consider extending the inlet and outlet to avoid backflow.
Consider refining your mesh to improve convergence and let it run for more iterations. It seems the trend is good.

Reconsider the choice of turbulence model according to your Y+ on the walls.
Last option is to run the case in transient to improve convergence, but you should see more reasonable results for the pressure before this.

Good luck.