Community
CFD Forum
Welcome to Autodesk’s CFD Forums. Share your knowledge, ask questions, and explore popular CFD topics.
cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 

Pump Ripple Analysis

2 REPLIES 2
SOLVED
Reply
Message 1 of 3
max.morse
312 Views, 2 Replies

Pump Ripple Analysis

Hello,

 

I am looking at modeling a pump ripple analysis, where the pressure in the inlet oscillates harmonically due to a pump upstream. A pressure sensor is located downstream, and the geometry leading to the pressure sensor downstream is causing the pressure to resonate.

I have been attempting solutions in a simple venturi nozzle, as the actual geometry is more complex. I have been concentrating on finding a solution a using a harmic pressure wave of +/- 1psi oscillating at 1 Hz. Most of my attemps result in near immediate divergance, so I am looking for help in how to best set the boundary conditions at the inlet/outlet, and what my solve process should be, in terms of what BC's to use during initial steady state solution (if nessesary, and then what to change for the transient portion), and how small should the time step be.

Real-Life Application Information:
-Pressure: 150-250 psi range
-Temperature: 50 C
-Flow Rate: 18 gal/min

-Fluid: Jet Engine Oil

 Any help that could be provided would be great.

2 REPLIES 2
Message 2 of 3
srhusain
in reply to: max.morse

One way to do this is to

1) Apply a constant zero gauge pressure at the exit

2) Apply a constant 100 psi gauge pressure at the inlet

3) Run steady state for a few hundred iterations

4) Change the inlet pressure to time varying harmonic ranging from 100 to 250 psi gauge

5) Run transient- keep the intelligent solution control option (ISC) checked as this will let the solver determine the time step size for optimal accuracy- the caveat is that it can be smaller than you needs. So experiment with your choice of time step size with ISC turned off.

Message 3 of 3
waited
in reply to: srhusain

I would also add that for time accuracy, I have found that you should have at least 100 time steps per cycle.  For 1 Hz that means time step size = 0.01 seconds.  You might experiment a bit and try 0.005 and 0.001 seconds and compare results to ensure a converged temporal response.

 

Dave

Can't find what you're looking for? Ask the community or share your knowledge.

Post to forums  

Autodesk Design & Make Report