Circular cylinder delayed separation on star-ccm+

lovecraft22 · March 5, 2010, 10:45

Hi everybody, I just registered.
I'm running a simulation on a finite lenght circular cylinder on star ccm+. This simulation will be used to assess some wind tunnel data I collected on the same model. In my experimental tests I obtained a separation line on the lateral surface of the cylinder located at about 80° from the frontal stagnation point. The separation line on star ccm is located at about 100°. This results in a lower drag coefficient. I've already reduced the free stream turbulence to 0.001 but the issue persists. The flow seems to be too turbulent and that's pobably the cause of the delayed separation.
How can I solve the situation?

Thanx!

f-w · March 5, 2010, 11:22

Not sure what your setup is in terms of models and boundary layer mesh density, which also play a role; but have you experimented with the Gamma ReTheta Transition Model?

lovecraft22 · March 5, 2010, 11:25

That's what I'm trying right now. Thank you, I'll get back to you!

lovecraft22 · March 8, 2010, 09:09

Nothing seems to work well, separation is still at 100 deg from stagnation point. Any idea?
Thanx

f-w · March 9, 2010, 11:50

Give us details on your physics models, your mesh, especially in the boundary layer, and your boundary conditions.

lovecraft22 · March 9, 2010, 12:03

Free stream velocity=20m/s
Cylinder diameter=0.06m
Cylinder height=0.18m
10 layers of prism layer mesh
K-omega model + Theta Re Omega

Thanx

f-w · March 9, 2010, 17:24

Reynolds number suggests it should be as you say (sanity check). How are you defining "separation" in Star-CCM? I'm guessing the reason Gamma ReTheta doesn't work for you is because it's not trivial to setup. So, let's go back a step; as a check, in your physics model, try suppressing turbulence over your entire cylinder boundary.

Here's what I would try for the mesh:
cell size: 0.00075m
prism layer stretching mode: wall thickness
stretching/layers: 1.5/15
near wall thickness: 0.0001m
prism layer thickness: 0.05*diameter

March 5, 2010, 10:45	Circular cylinder delayed separation on star-ccm+	#1
lovecraft22 Senior Member lore Join Date: Mar 2010 Location: Italy Posts: 460 Rep Power: 18	Hi everybody, I just registered. I'm running a simulation on a finite lenght circular cylinder on star ccm+. This simulation will be used to assess some wind tunnel data I collected on the same model. In my experimental tests I obtained a separation line on the lateral surface of the cylinder located at about 80° from the frontal stagnation point. The separation line on star ccm is located at about 100°. This results in a lower drag coefficient. I've already reduced the free stream turbulence to 0.001 but the issue persists. The flow seems to be too turbulent and that's pobably the cause of the delayed separation. How can I solve the situation? Thanx!

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
[Commercial meshers] Using starToFoam	clo	OpenFOAM Meshing & Mesh Conversion	33	September 26, 2012 04:04
[Other] StarToFoam error	Kart	OpenFOAM Meshing & Mesh Conversion	1	February 4, 2010 04:38
theta separation over circular cylinder	ND	FLUENT	0	August 3, 2008 04:15
[Commercial meshers] Trimmed cell and embedded refinement mesh conversion issues	michele	OpenFOAM Meshing & Mesh Conversion	2	July 15, 2005 04:15

March 5, 2010, 11:22		#2
f-w Senior Member Join Date: Apr 2009 Posts: 154 Rep Power: 17	Not sure what your setup is in terms of models and boundary layer mesh density, which also play a role; but have you experimented with the Gamma ReTheta Transition Model?

March 5, 2010, 11:25		#3
lovecraft22 Senior Member lore Join Date: Mar 2010 Location: Italy Posts: 460 Rep Power: 18	That's what I'm trying right now. Thank you, I'll get back to you!

March 8, 2010, 09:09		#4
lovecraft22 Senior Member lore Join Date: Mar 2010 Location: Italy Posts: 460 Rep Power: 18	Nothing seems to work well, separation is still at 100 deg from stagnation point. Any idea? Thanx

March 9, 2010, 11:50		#5
f-w Senior Member Join Date: Apr 2009 Posts: 154 Rep Power: 17	Give us details on your physics models, your mesh, especially in the boundary layer, and your boundary conditions.

March 9, 2010, 12:03		#6
lovecraft22 Senior Member lore Join Date: Mar 2010 Location: Italy Posts: 460 Rep Power: 18	Free stream velocity=20m/s Cylinder diameter=0.06m Cylinder height=0.18m 10 layers of prism layer mesh K-omega model + Theta Re Omega Thanx

March 9, 2010, 17:24		#7
f-w Senior Member Join Date: Apr 2009 Posts: 154 Rep Power: 17	Reynolds number suggests it should be as you say (sanity check). How are you defining "separation" in Star-CCM? I'm guessing the reason Gamma ReTheta doesn't work for you is because it's not trivial to setup. So, let's go back a step; as a check, in your physics model, try suppressing turbulence over your entire cylinder boundary. Here's what I would try for the mesh: cell size: 0.00075m prism layer stretching mode: wall thickness stretching/layers: 1.5/15 near wall thickness: 0.0001m prism layer thickness: 0.05*diameter