Kw-SST not converging on vehicle CFD

thehappiest · October 16, 2021, 06:38

Hello,
I'm a student trying to do a simulation of external aerodynamics of a car with Kw-SST in Star CCM 2021.1.1. However, the solution doesn't seem to converge.
Going over a Wall y+ scene over the car, I get that there is a lot of area with an y+ between 5 and 30. I suposse that is an error I need to fix. Also, the drag coeficcient keeps oscilating. I'll attach images of most of the results.
Here is my setup for the mesh (Surface Remesher, automatic repair, trimmed, prism layer mesher):

Base Size: 0.05m
Target: 400%
Minimum: 10%
Surface Curvature: 55 points
Surface growth rate: 1.05
Number of prims layers: 4
Prism stretching: 1.5
Prism thickness: 33.33%
Volume Growth: Slow
Max Cell Size: 800%

Customs controls:

Box around car: size 200%
Ground: Target surface 400%
Vehicle: Target surface 50% Prism layer:6
Wake: size 100%
Walls: Target surface 800%

Models:

constant density
gas
k-omega turulence
SST (Menter) komega
Segregated
Turbulent
Steady
Three dimensional

The inlet velocity is 33m/s and I also put that velocity to a moving ground. Reference Pressure is 0 Pa as also is in the pressure outlet. Any other configuration is as default.

I really don't know how to fix it. I've tried finer mesh, but seems to be pointless. ¿How, and would you try to get a Y+ lower than 5 or higher than 30? ¿Do you think that will be it? Also, have to say that with default K-epsilon, the same case converges getting most of the car with an y+ higher than 30.
Any help will be much appreciated. Thanks!

residuals.jpg

wall_y+.jpg

mesh1.jpg

velocity.jpg

CD.jpg

fluid23 · October 18, 2021, 08:10

I think your biggest issue is that you are expecting a steady result from an unsteady flow. Try switching to implicit unsteady and you should see better residual convergence.

Y+ between 5 and 30 isn’t great but it isn’t the end of the world either. Make that fix too and consider using wake refinement.

LuckyTran · October 18, 2021, 09:41

Something you can do is create a field mean and variance monitor. Collect statistics while the case is running. Then plot the variances of pressure and so on. You'll be able to see where the solution is oscillating this way and see where you need to make improvements.

October 16, 2021, 06:38	Kw-SST not converging on vehicle CFD	#1
thehappiest New Member Javier Join Date: Oct 2021 Posts: 1 Rep Power: 0	Hello, I'm a student trying to do a simulation of external aerodynamics of a car with Kw-SST in Star CCM 2021.1.1. However, the solution doesn't seem to converge. Going over a Wall y+ scene over the car, I get that there is a lot of area with an y+ between 5 and 30. I suposse that is an error I need to fix. Also, the drag coeficcient keeps oscilating. I'll attach images of most of the results. Here is my setup for the mesh (Surface Remesher, automatic repair, trimmed, prism layer mesher): Base Size: 0.05m Target: 400% Minimum: 10% Surface Curvature: 55 points Surface growth rate: 1.05 Number of prims layers: 4 Prism stretching: 1.5 Prism thickness: 33.33% Volume Growth: Slow Max Cell Size: 800% Customs controls: Box around car: size 200% Ground: Target surface 400% Vehicle: Target surface 50% Prism layer:6 Wake: size 100% Walls: Target surface 800% Models: constant density gas k-omega turulence SST (Menter) komega Segregated Turbulent Steady Three dimensional The inlet velocity is 33m/s and I also put that velocity to a moving ground. Reference Pressure is 0 Pa as also is in the pressure outlet. Any other configuration is as default. I really don't know how to fix it. I've tried finer mesh, but seems to be pointless. ¿How, and would you try to get a Y+ lower than 5 or higher than 30? ¿Do you think that will be it? Also, have to say that with default K-epsilon, the same case converges getting most of the car with an y+ higher than 30. Any help will be much appreciated. Thanks! residuals.jpg wall_y+.jpg mesh1.jpg velocity.jpg CD.jpg

October 18, 2021, 08:10	Kw-SST not converging on vehicle CFD	#2
fluid23 Senior Member Matt Join Date: Aug 2014 Posts: 947 Rep Power: 18	I think your biggest issue is that you are expecting a steady result from an unsteady flow. Try switching to implicit unsteady and you should see better residual convergence. Y+ between 5 and 30 isn’t great but it isn’t the end of the world either. Make that fix too and consider using wake refinement.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Free CFD code with implementation of Menter's k-omega SST model	nikola_m	Main CFD Forum	2	May 8, 2020 13:13
[ANSYS Meshing] Divide entire ground into sub sections for moving ground for vehicle aerodynamics CFD	johny_walker	ANSYS Meshing & Geometry	0	February 24, 2020 10:42
The role of CFD in vehicle aerodynamics design and the reasons	Anna Tian	Main CFD Forum	11	February 4, 2014 05:56
ASME CFD Symposium, Atlanta, 22-26 July 2001	Chris R. Kleijn	Main CFD Forum	0	August 1, 2000 11:07
public CFD Code development	Heinz Wilkening	Main CFD Forum	38	March 5, 1999 12:44

October 18, 2021, 09:41		#3
LuckyTran Senior Member Lucky Join Date: Apr 2011 Location: Orlando, FL USA Posts: 5,761 Rep Power: 66	Something you can do is create a field mean and variance monitor. Collect statistics while the case is running. Then plot the variances of pressure and so on. You'll be able to see where the solution is oscillating this way and see where you need to make improvements.