Centrifugal pump simulation using fluent

Ahmed Abotayel · April 25, 2023, 11:00

Hello everyone,
I am working on centrifugal pump simulation case via ansys software, I have set the boundary conditions of the case then by mistake I put the rotational speed 1500 rad/s instead of 1500 rpm the contours of pressure and velocity in these iteration were normal when I changed the rotational speed to 1500 rpm with the same settings and boundary conditions the contours were not normal and the residuals were not stable. Can someone help me to find out why the simulation goes wrong in the case of 1500 rpm?

NickFL · April 25, 2023, 11:12

There could be all kinds of reasons. Did you continue your simulation from the 1500 rad/s when doing the 1500 rpm case? Or did you give a new initial condition? If the former it was converging towards a solution, it may take a while for the solution to adapt to the new flow conditions.

What do you mean that the contours are not normal? Not normal does not provide much information for us to help you.

Ahmed Abotayel · April 25, 2023, 12:06

They are two separated iterations but with the same settings and boundary conditions
I said not normal because the results of pressure and velocity contours were not acceptable according to the analysis of flow inside centrifugal pump as shown here https://prnt.sc/r8YrZiBfkEwp
That is the link of the two projects if you can check them then give me your recommendations https://drive.google.com/drive/folde...WhxfLSZmWYhOzE
thank you for your time

NickFL · April 27, 2023, 03:03

Quote:

Originally Posted by Ahmed Abotayel

They are two separated iterations but with the same settings and boundary conditions
I said not normal because the results of pressure and velocity contours were not acceptable according to the analysis of flow inside centrifugal pump as shown here https://prnt.sc/r8YrZiBfkEwp
That is the link of the two projects if you can check them then give me your recommendations https://drive.google.com/drive/folde...WhxfLSZmWYhOzE
thank you for your time

Please do NOT send private messages simply to grab attention. These are voluntary forums, and many of us have other responsibilities that pay the bills. Being on here, unfortunately, does not.

Further I do not go to or download from random websites. My office computer has restrictions and I don't want any problems on this. I am sure I am not the only one. The image hosting on this forum works quite well.

For your specific problem, I would assume you have validated that the impeller is running in the correct direction? As crazy as it sounds, that is one of the most common mistakes. Are the residuals converged? Also are you using the coupled solver? If so, try reducing the pseudo-timestep. Generally, what the solver chooses works, but sometimes not, especially at early iterations.

Ahmed Abotayel · April 27, 2023, 13:58

Sorry for all what you said I just taught you have free time enough to help me so I sent the message but there is no problem at all thank you for making me realize something like that

That is velocity vectors that ensures that the impeller rotates in the right direction

NickFL · April 30, 2023, 03:32

Are the residuals converged in the second case? Again, without some pictures or more information we really don't have anything to diagnose.

April 25, 2023, 11:00	Centrifugal pump simulation using fluent	#1
Ahmed Abotayel New Member Ahmed Join Date: Apr 2023 Posts: 8 Rep Power: 3	Hello everyone, I am working on centrifugal pump simulation case via ansys software, I have set the boundary conditions of the case then by mistake I put the rotational speed 1500 rad/s instead of 1500 rpm the contours of pressure and velocity in these iteration were normal when I changed the rotational speed to 1500 rpm with the same settings and boundary conditions the contours were not normal and the residuals were not stable. Can someone help me to find out why the simulation goes wrong in the case of 1500 rpm?

April 25, 2023, 11:12		#2
NickFL Senior Member Join Date: Jun 2009 Location: Technische Universität Chemnitz Posts: 107 Rep Power: 16	There could be all kinds of reasons. Did you continue your simulation from the 1500 rad/s when doing the 1500 rpm case? Or did you give a new initial condition? If the former it was converging towards a solution, it may take a while for the solution to adapt to the new flow conditions. What do you mean that the contours are not normal? Not normal does not provide much information for us to help you. Ahmed Abotayel likes this.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
How to set Outlet Guage Pressure in ANSYS Fluent for Centrifugal Pump?	PratikChalke	FLUENT	0	March 6, 2023 14:34
axial piston pump simulation in Fluent	Silence	FLUENT	4	October 14, 2022 00:02
Centrifugal pump impeller simulation doesn't converge	JuanRincon	CFX	8	June 25, 2019 05:59
problem for centrifugal pump transient simulation	renyun0511	OpenFOAM Running, Solving & CFD	8	January 15, 2014 04:55
centrifugal pump simulation OK but result with 30% error	ARohit	FLUENT	0	January 1, 2014 01:54

April 25, 2023, 12:06		#3
Ahmed Abotayel New Member Ahmed Join Date: Apr 2023 Posts: 8 Rep Power: 3	They are two separated iterations but with the same settings and boundary conditions I said not normal because the results of pressure and velocity contours were not acceptable according to the analysis of flow inside centrifugal pump as shown here https://prnt.sc/r8YrZiBfkEwp That is the link of the two projects if you can check them then give me your recommendations https://drive.google.com/drive/folde...WhxfLSZmWYhOzE thank you for your time

April 27, 2023, 13:58		#5
Ahmed Abotayel New Member Ahmed Join Date: Apr 2023 Posts: 8 Rep Power: 3	Sorry for all what you said I just taught you have free time enough to help me so I sent the message but there is no problem at all thank you for making me realize something like that That is velocity vectors that ensures that the impeller rotates in the right direction

April 30, 2023, 03:32		#6
NickFL Senior Member Join Date: Jun 2009 Location: Technische Universität Chemnitz Posts: 107 Rep Power: 16	Are the residuals converged in the second case? Again, without some pictures or more information we really don't have anything to diagnose.