Questions about a Turbulent Flat Plate Case

tstorm · August 11, 2009, 13:02

I'm trying to validate a user-defined turbulence model. As a baseline, I'm generating results using the Spalart-Allmaras, k-e, and k-w turbulence models for a turbulent flat plate. Freestream velocity is 1.46 m/s, with the molecular viscosity dropped to about 1.5e-7.

The plate is 1m long, with 0.2m in front of the plate and 0.2m behind the plate, both defined by the 'symmetry' boundary condition. At the symmetry boundaries upstream and downstream of the plate, the cases are showing very high turublence kinetic energy. Is that right? Can anyone explain why I'm seeing this?

Here's a contour plot of k: http://picasaweb.google.com/lh/photo...eat=directlink

Thanks!

holg · August 11, 2009, 15:12

I will give you my thoughts: so if you have a symmetry condition in front the plate, you have a stagnation point at the tip of your plate. that, might cause turbulence upstream. I assume you applied the no slip condition at the plate surface. So you will have a laminar sublayer which by definition has no turbulence. That might explain the low turbulence above the plate. Downstream the plate you have free air flow again an turbulence can spread. Did you use any wall functions? greets holger

tstorm · August 11, 2009, 15:16

Thanks for your response - with the k-e model I'm using enhanced wall functions. y+ is below 1 though (99% of the wall cells are at about y+ = 0.5.) I'm seeing similar k contours for the k-w model as well, and I'm not using wall functions for that case. The wall does have a no-slip BC. I expected no turbulence in the laminar sublayer, but there doesn't seem to be any appreciable buildup of turbulence in the log layer, which I find surprising.

August 11, 2009, 13:02	Questions about a Turbulent Flat Plate Case	#1
tstorm Member Join Date: Apr 2009 Posts: 78 Rep Power: 17	I'm trying to validate a user-defined turbulence model. As a baseline, I'm generating results using the Spalart-Allmaras, k-e, and k-w turbulence models for a turbulent flat plate. Freestream velocity is 1.46 m/s, with the molecular viscosity dropped to about 1.5e-7. The plate is 1m long, with 0.2m in front of the plate and 0.2m behind the plate, both defined by the 'symmetry' boundary condition. At the symmetry boundaries upstream and downstream of the plate, the cases are showing very high turublence kinetic energy. Is that right? Can anyone explain why I'm seeing this? Here's a contour plot of k: http://picasaweb.google.com/lh/photo...eat=directlink Thanks!

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Turbulent Flat Plate Validation Case	Jonas Larsson	Main CFD Forum	0	April 2, 2004 11:25
Free convection flow over vertical flat plate	Polly	Main CFD Forum	1	February 11, 2003 14:25
Flat plate trouble	Chetan Kadakia	FLUENT	2	October 19, 2001 04:05
Supersonic flow over flat plate - best model	Chris Rand	FLUENT	4	October 3, 2000 07:58
Turbulent flow over flat plate	Narayanan	Main CFD Forum	1	May 8, 2000 08:03

August 11, 2009, 15:12		#2
holg New Member holger thiele Join Date: Jul 2009 Posts: 2 Rep Power: 0	I will give you my thoughts: so if you have a symmetry condition in front the plate, you have a stagnation point at the tip of your plate. that, might cause turbulence upstream. I assume you applied the no slip condition at the plate surface. So you will have a laminar sublayer which by definition has no turbulence. That might explain the low turbulence above the plate. Downstream the plate you have free air flow again an turbulence can spread. Did you use any wall functions? greets holger

August 11, 2009, 15:16		#3
tstorm Member Join Date: Apr 2009 Posts: 78 Rep Power: 17	Thanks for your response - with the k-e model I'm using enhanced wall functions. y+ is below 1 though (99% of the wall cells are at about y+ = 0.5.) I'm seeing similar k contours for the k-w model as well, and I'm not using wall functions for that case. The wall does have a no-slip BC. I expected no turbulence in the laminar sublayer, but there doesn't seem to be any appreciable buildup of turbulence in the log layer, which I find surprising.