Low Flow Convergence Help

rypersaud · January 6, 2013, 14:10

Hey guys, need some help with convergence. The application involves a 3" inlet pipe with 52 m/s velocity discharged into a 26"X50" duct. The duct has a U-bend with a 20" outlet.

This is a very slow flow problem due to the large size of the duct. We are expecting the velocity at the outlet to be approximately 1 m/s.

Used a surface monitor plot at the outlet to look at the velocity, and scaled residuals criteria of 1e-5. The velocity of the outlet appears to have approached a value, however the residuals are quite high. I ran 5000 iterations to exclude the effect of under relaxation factors. I have posted the plots below.

Any suggestions, or direction on how to improve the model will be great. Thanks!

residuals 5000.jpg

velocity 5000.jpg

LuckyTran · January 7, 2013, 00:13

Seems like everything is okay except for residuals.

Have you tried a finer grid? What about discretization schemes, 2nd order upwind on all?

rypersaud · January 7, 2013, 11:23

I am using a student version of Fluent so I'm limited to 500000 cells and I'm using 2nd order for everything. I'm also using the Simple solver....would using another one help? any other ideas?

msaeedsadeghi · January 7, 2013, 23:59

It is better to use the simplest form (First Order).
It is obvious that the problem is with your mesh. Refine it and try to make a structured mesh.

January 6, 2013, 14:10	Low Flow Convergence Help	#1
rypersaud New Member Ryan Persaud Join Date: Nov 2012 Posts: 4 Rep Power: 13	Hey guys, need some help with convergence. The application involves a 3" inlet pipe with 52 m/s velocity discharged into a 26"X50" duct. The duct has a U-bend with a 20" outlet. This is a very slow flow problem due to the large size of the duct. We are expecting the velocity at the outlet to be approximately 1 m/s. Used a surface monitor plot at the outlet to look at the velocity, and scaled residuals criteria of 1e-5. The velocity of the outlet appears to have approached a value, however the residuals are quite high. I ran 5000 iterations to exclude the effect of under relaxation factors. I have posted the plots below. Any suggestions, or direction on how to improve the model will be great. Thanks! residuals 5000.jpg velocity 5000.jpg

Thread Tools	Search this Thread
Show Printable Version Email this Page	Search this Thread: Advanced Search
Display Modes
Linear Mode Switch to Hybrid Mode Switch to Threaded Mode

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Convergence	Centurion2011	FLUENT	48	June 14, 2022 23:29
Flow through porous medium using CFX: lack of convergence	Chander	CFX	3	July 23, 2014 06:32
Convergence for axisymmetric flow	bl201	Main CFD Forum	5	August 16, 2005 17:01
Multicomponent fluid	Andrea	CFX	2	October 11, 2004 05:12
CFD code for very low Re Flow	Raj	Main CFD Forum	2	May 31, 1999 22:06

January 7, 2013, 00:13		#2
LuckyTran Senior Member Lucky Join Date: Apr 2011 Location: Orlando, FL USA Posts: 5,665 Rep Power: 65	Seems like everything is okay except for residuals. Have you tried a finer grid? What about discretization schemes, 2nd order upwind on all?

January 7, 2013, 11:23		#3
rypersaud New Member Ryan Persaud Join Date: Nov 2012 Posts: 4 Rep Power: 13	I am using a student version of Fluent so I'm limited to 500000 cells and I'm using 2nd order for everything. I'm also using the Simple solver....would using another one help? any other ideas?

January 7, 2013, 23:59		#4
msaeedsadeghi Senior Member SSL Join Date: Oct 2012 Posts: 226 Rep Power: 14	It is better to use the simplest form (First Order). It is obvious that the problem is with your mesh. Refine it and try to make a structured mesh.