Unable to achieve a steady flow

lost.identity · April 29, 2009, 07:09

Hi

I'm trying to obtain a steady flow solution for flow in a convergent-divergent duct using a RANS calculation with the SA turbulence model. The Mach numbers in the duct varies from 0.03 to 1.1.

I'm not using wall functions in my simulations and I use a very fine grid spacing near the wall (with the first grid node well within a y+ =1). I know that you should avoid discontinuities in grid spacing and the ratio should be below 1.3. However, I do obtain a steady flow when the ratio between one grid spacing to the next was left as 2. But when I used a grid spacing of 1.3 I fail to get a 3-order-of-magnitude drop in the residuals and never achieve a steady flow.

I do not understand why this might be the case (surely when I have a ratio of 2 it should fail?) and I can't find any other problem with my setup.

johndorian · April 29, 2009, 10:22

"Better" RANS grids (as in more refined) usually lead to the resolving of energy-carrying scales of motion in the flow. This manifests itself as an inherent unsteadiness in the solution and the residuals will not drop to a seemingly converged value.

My guess is that there is some separation in the diverging part of your duct. Try to find out where the maximum change in flow residual is occurring and check the flow field in this area to look for signs of separation or vortex shedding.

April 29, 2009, 07:09	Unable to achieve a steady flow	#1
lost.identity Senior Member Join Date: Apr 2009 Posts: 118 Rep Power: 17	Hi I'm trying to obtain a steady flow solution for flow in a convergent-divergent duct using a RANS calculation with the SA turbulence model. The Mach numbers in the duct varies from 0.03 to 1.1. I'm not using wall functions in my simulations and I use a very fine grid spacing near the wall (with the first grid node well within a y+ =1). I know that you should avoid discontinuities in grid spacing and the ratio should be below 1.3. However, I do obtain a steady flow when the ratio between one grid spacing to the next was left as 2. But when I used a grid spacing of 1.3 I fail to get a 3-order-of-magnitude drop in the residuals and never achieve a steady flow. I do not understand why this might be the case (surely when I have a ratio of 2 it should fail?) and I can't find any other problem with my setup.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Creating streamline animation with steady flow	CFDnewbie	CFX	4	October 3, 2005 15:38
Unsteady AND Steady mode for Fully Developped Flow	Dominique	FLUENT	5	April 16, 2004 17:58
Turbulent steady flow around a circular cylinder	Mirek Kabacinski	FLUENT	0	July 23, 2003 18:40
Inviscid Drag at subsonic, subcritical Mach #	Axel Rohde	Main CFD Forum	1	November 19, 2001 12:19
steady and unsteady flow	shisheng	Main CFD Forum	10	July 23, 1999 06:00

April 29, 2009, 10:22		#2
johndorian New Member Join Date: Mar 2009 Posts: 11 Rep Power: 17	"Better" RANS grids (as in more refined) usually lead to the resolving of energy-carrying scales of motion in the flow. This manifests itself as an inherent unsteadiness in the solution and the residuals will not drop to a seemingly converged value. My guess is that there is some separation in the diverging part of your duct. Try to find out where the maximum change in flow residual is occurring and check the flow field in this area to look for signs of separation or vortex shedding.