Flow over a cylinder (1 way FSI)

vmlxb6 · March 4, 2011, 19:04

I know there has been tons of post on flow over a moving cylinder (transient simulation), so pardon me please.
I would like to know the following:

1. what should be my mesh stiffness so that the boundary layer does not get deformed to give me weird results.
2. Right now am using a hexa structured mesh. I need to know what kind of mesh would be appropriate ??? My struct mesh gives me good results for Re = 525 but not for 200 (solution diverges).

Thank you.

ghorrocks · March 6, 2011, 17:13

A good quality structured mesh will always help, it is not the cause of the divergence at Re=200. Providing it is good quality the cause of divergence is something else.

vmlxb6 · March 6, 2011, 17:53

I have a square domain with a circle in the center. Velocity inlet to the left, Pr. outlet to the right, and velocity inlet to the top and bottom. The circle is a wall.
The document you always refer us to " CFD Online ANSYS FAQ" talks about steady state convergence but not about transient convergence. Any help on that.

Thank you.

ghorrocks · March 7, 2011, 18:56

Yes, the FAQ talks about steady state but most of the same principles apply to transient. Obviously you have the additional requirement of choosing an appropriate time step size.

If you are getting good convergence at Re=525 and poor at 200 then a flow feature (eg vortex shedding) is probably causing it.

vmlxb6 · March 7, 2011, 21:08

any suggestions as to what I must be doing to resolve it ???

ghorrocks · March 8, 2011, 16:54

Try smaller timesteps. For more detail you would have to post the relevant bit of the out file to diagonse the problem.

March 4, 2011, 19:04	Flow over a cylinder (1 way FSI)	#1
vmlxb6 Senior Member Ugly Kid Joe Join Date: Aug 2010 Posts: 193 Rep Power: 15	I know there has been tons of post on flow over a moving cylinder (transient simulation), so pardon me please. I would like to know the following: 1. what should be my mesh stiffness so that the boundary layer does not get deformed to give me weird results. 2. Right now am using a hexa structured mesh. I need to know what kind of mesh would be appropriate ??? My struct mesh gives me good results for Re = 525 but not for 200 (solution diverges). Thank you.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
benchmark: flow over a circular cylinder	goodegg	Main CFD Forum	12	January 22, 2013 11:47
1 Was FSI for a flow over a cylinder	vmlxb6	CFX	17	May 16, 2011 02:29
reference calcultion, flow around a cylinder	butch85	Main CFD Forum	3	January 31, 2011 16:10
Flow over a flat plate & Flow over a cylinder	cfdxue	Main CFD Forum	0	November 26, 2007 23:26
Flow over a cylinder	Anna	Main CFD Forum	9	March 24, 2006 14:32

March 6, 2011, 17:13		#2
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,703 Rep Power: 143	A good quality structured mesh will always help, it is not the cause of the divergence at Re=200. Providing it is good quality the cause of divergence is something else.

March 6, 2011, 17:53		#3
vmlxb6 Senior Member Ugly Kid Joe Join Date: Aug 2010 Posts: 193 Rep Power: 15	I have a square domain with a circle in the center. Velocity inlet to the left, Pr. outlet to the right, and velocity inlet to the top and bottom. The circle is a wall. The document you always refer us to " CFD Online ANSYS FAQ" talks about steady state convergence but not about transient convergence. Any help on that. Thank you.

March 7, 2011, 18:56		#4
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,703 Rep Power: 143	Yes, the FAQ talks about steady state but most of the same principles apply to transient. Obviously you have the additional requirement of choosing an appropriate time step size. If you are getting good convergence at Re=525 and poor at 200 then a flow feature (eg vortex shedding) is probably causing it.

March 7, 2011, 21:08		#5
vmlxb6 Senior Member Ugly Kid Joe Join Date: Aug 2010 Posts: 193 Rep Power: 15	any suggestions as to what I must be doing to resolve it ???

March 8, 2011, 16:54		#6
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,703 Rep Power: 143	Try smaller timesteps. For more detail you would have to post the relevant bit of the out file to diagonse the problem.