Valildation of OF for external aerodinamics

andrea.ciarella · November 28, 2012, 04:25

Hi,

I have one problem with OF. I am trying to validate SimpleFOAM for external aerodynamics.

The problem is

Airfoil: s822
Mach: 0.1
Reynolds: 500000

I am generating the polar from the angle of attack -10 to 14 with steps of 1 degree.

I compared the results with Experimental data and another code.

I used Spalart-Allmaras turbulence model with y+ = 1 and tested several numerical schemes. I only attached the best results

lift-drag.jpg

Polar.jpg

pressure.jpg

my questions are:

The drag is wrong after the linear range, the cause is a wrong stagnation pressure. Is here any way to improve it?

If I want to use a wall fuction which one do you think I have to use?

Do you think will be better with other solvers or turbulences? I test K-Omega sst but the results are similar.

Does a validation of OpenFOAM for external aerodynamics exist?

Thank you in advance
Andrea

sail · November 28, 2012, 13:52

Hi there.

could if be possible to see your mesh? my suspect is that you don't have enough spatial resolution in the chord direction and not enough layers in the boundary layer. I got excellent results with 100-150 points for in the chord and at lest 20 cells across the BL

also, it the error can be due to the turbulence model. SA is mostly effective for high Mach numbers, say 0.6+

at your Re lot of drag is due to friction more than pressure, so i would reccomand a model that can show laminar-turbulent transition like the k-o SST gamma-re-theta.

there is an excelent implementation for openfoam in the forum. try to google it.

calim_cfd · November 29, 2012, 14:09

hi!

just to emphasize.. a pic of your mesh would be helpfull.

try plotting the y+(y*) values for each aoa too along the wing span.

also, make sure ur boundary conditions match the experimental one.

other than that i dont c where the problem could be

gl

andrea.ciarella · November 30, 2012, 03:51

Thank you for your replay.

The mesh is a semistructured mesh

mesh.jpg

I used different mesh with y+ between 0.1 an 3 to see if it is an effect of the mesh, but I didn't get much difference

I will try the transition model next week and I will post the results.

From my point of view the biggest problem is the wrong stagnation pressure that is the cause of he wrong drag.

I already tested other farfield boundary conditions, but doesn't change the solution.

Did somebody have/know some proper validation of the solver for external aero?

Thank you

Regards
Andrea Ciarella

calim_cfd · November 30, 2012, 12:28

hi again!

im not an expert in lift/drag calculations but are you updating the reference area with the aoa? do ppl usually do that when plotting those charts?

November 28, 2012, 04:25	Valildation of OF for external aerodinamics	#1
andrea.ciarella New Member andrea ciarella Join Date: Nov 2012 Posts: 5 Rep Power: 14	Hi, I have one problem with OF. I am trying to validate SimpleFOAM for external aerodynamics. The problem is Airfoil: s822 Mach: 0.1 Reynolds: 500000 I am generating the polar from the angle of attack -10 to 14 with steps of 1 degree. I compared the results with Experimental data and another code. I used Spalart-Allmaras turbulence model with y+ = 1 and tested several numerical schemes. I only attached the best results lift-drag.jpg Polar.jpg pressure.jpg my questions are: The drag is wrong after the linear range, the cause is a wrong stagnation pressure. Is here any way to improve it? If I want to use a wall fuction which one do you think I have to use? Do you think will be better with other solvers or turbulences? I test K-Omega sst but the results are similar. Does a validation of OpenFOAM for external aerodynamics exist? Thank you in advance Andrea

November 28, 2012, 13:52		#2
sail Senior Member Vieri Abolaffio Join Date: Jul 2010 Location: Always on the move. Posts: 308 Rep Power: 17	Hi there. could if be possible to see your mesh? my suspect is that you don't have enough spatial resolution in the chord direction and not enough layers in the boundary layer. I got excellent results with 100-150 points for in the chord and at lest 20 cells across the BL also, it the error can be due to the turbulence model. SA is mostly effective for high Mach numbers, say 0.6+ at your Re lot of drag is due to friction more than pressure, so i would reccomand a model that can show laminar-turbulent transition like the k-o SST gamma-re-theta. there is an excelent implementation for openfoam in the forum. try to google it. __________________ http://www.leadingedge.it/ Naval architecture and CFD consultancy

November 29, 2012, 14:09		#3
calim_cfd Senior Member mauricio Join Date: Jun 2011 Posts: 172 Rep Power: 18	hi! just to emphasize.. a pic of your mesh would be helpfull. try plotting the y+(y*) values for each aoa too along the wing span. also, make sure ur boundary conditions match the experimental one. other than that i dont c where the problem could be gl __________________ Best Regards /calim "Elune will grant us the strength"

November 30, 2012, 03:51	Mesh	#4
andrea.ciarella New Member andrea ciarella Join Date: Nov 2012 Posts: 5 Rep Power: 14	Thank you for your replay. The mesh is a semistructured mesh mesh.jpg I used different mesh with y+ between 0.1 an 3 to see if it is an effect of the mesh, but I didn't get much difference I will try the transition model next week and I will post the results. From my point of view the biggest problem is the wrong stagnation pressure that is the cause of he wrong drag. I already tested other farfield boundary conditions, but doesn't change the solution. Did somebody have/know some proper validation of the solver for external aero? Thank you Regards Andrea Ciarella

November 30, 2012, 12:28		#5
calim_cfd Senior Member mauricio Join Date: Jun 2011 Posts: 172 Rep Power: 18	hi again! im not an expert in lift/drag calculations but are you updating the reference area with the aoa? do ppl usually do that when plotting those charts? __________________ Best Regards /calim "Elune will grant us the strength"