Will refining the grid help in better convergence?

knax · February 4, 2018, 09:36

My 3D mesh with K-omega simulation of a fan has a pretty good quality, with just a few cells out of 0.8 million with an orthogonal skewness >0.7. However, the residuals are stagnating at just about 10e-3 for velocities and 10e-2 for continuity.

Any suggestions?

Kirill-MIPT · February 5, 2018, 07:59

Hello! Please show the mesh and tell which boundary conditions do you use. Which y+ value did you try to obtain?

Gert-Jan · February 5, 2018, 11:40

Which k-omega model?
And settings?

Kirill-MIPT · February 5, 2018, 14:51

I mean pressure, temperature and velocity/Mach number in the inlet

knax · April 11, 2018, 00:09

Hi Kirill and Gert,

Sorry for the later reply. I didn't use y+ for first layer thickness. I used AR of 10 for k-omega (standard).

I'm simulation an axial fan with MRF. 1st inflation layer AR=10 on all walls.

Only pressure outlets. With an RPM of 2500. I worked on max orthogonal skewness and brought it down to 0.7. I've get my residuals to 10e-4 and the parameters of total-pressure and heat-flux are not changing. Hence, I've taken them to have converged. But I just wanted insight into how Fluent actually calculates residuals. Thanks!

Kind regards,

Noel

February 4, 2018, 09:36	Will refining the grid help in better convergence?	#1
knax New Member Noel Join Date: Apr 2017 Posts: 7 Rep Power: 9	My 3D mesh with K-omega simulation of a fan has a pretty good quality, with just a few cells out of 0.8 million with an orthogonal skewness >0.7. However, the residuals are stagnating at just about 10e-3 for velocities and 10e-2 for continuity. Any suggestions?

April 11, 2018, 00:09	Sorry for the later reply. Didn't get notified. How does Fuent calculate residuals?	#5
knax New Member Noel Join Date: Apr 2017 Posts: 7 Rep Power: 9	Hi Kirill and Gert, Sorry for the later reply. I didn't use y+ for first layer thickness. I used AR of 10 for k-omega (standard). I'm simulation an axial fan with MRF. 1st inflation layer AR=10 on all walls. Only pressure outlets. With an RPM of 2500. I worked on max orthogonal skewness and brought it down to 0.7. I've get my residuals to 10e-4 and the parameters of total-pressure and heat-flux are not changing. Hence, I've taken them to have converged. But I just wanted insight into how Fluent actually calculates residuals. Thanks! Kind regards, Noel

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Grid Convergence Study	Hybrid	Pointwise & Gridgen	5	April 18, 2016 09:40
Force can not converge	colopolo	CFX	13	October 4, 2011 22:03
Convergence of CFX field in FSI analysis	nasdak	CFX	2	June 29, 2009 01:17
Re: Convergence problem for refined grid	CFD_Novice	Main CFD Forum	13	February 17, 2008 22:56
Convergence on non-uniform grid	dragon	Main CFD Forum	2	January 31, 2005 02:23

February 5, 2018, 07:59		#2
Kirill-MIPT Member Kirill Borodin Join Date: Mar 2016 Posts: 60 Rep Power: 10	Hello! Please show the mesh and tell which boundary conditions do you use. Which y+ value did you try to obtain?

February 5, 2018, 11:40		#3
Gert-Jan Senior Member Gert-Jan Join Date: Oct 2012 Location: Europe Posts: 1,833 Rep Power: 27	Which k-omega model? And settings?

February 5, 2018, 14:51		#4
Kirill-MIPT Member Kirill Borodin Join Date: Mar 2016 Posts: 60 Rep Power: 10	I mean pressure, temperature and velocity/Mach number in the inlet