Convergence issues for modeling of flow over a car

blin · January 10, 2010, 19:31

Hi,

I am an amateur at this and would like to ask some CFD gurus for some help

.

I am trying to model flow over a car body in Fluent (actually, I'm only modeling half of the car split upon its symmetry plane). I have chosen a spalart-allmaras model, with a turbulence intensity of 0.21% and a turbulence length scale of 0.16m.
My discretization values are as follows:

Pressure 2nd Order (0.2)
Momentum QUICK (0.6)
Pressure-Velocity Coupling SIMPLE
Turbulence QUICK (0.7)

The solution converges to residuals of 0.001 when I run it using first order upwind (default) setting. However, when I change it to the above setting the solution does not converge.
I have tried running it using first order upwind, and then change the discretization parameters to the above setting and continue running it without initalization, but that does not converge as well.

I'm really stuck as to what I can do further.

Also, I'm getting a bunch of warnings that says "turbulent viscosity limited to viscosity ratio of 1.000000e+05 in xxxx cells", with the number of cells increasing. I have read on the forums on discussions regarding this topic, but none relating to how to solve it in the spalart-allmaras model, so I'm wondering as to the solution to that as well.

Thanks in advance for any input.

benji2642 · January 11, 2010, 06:01

What kind of data do you want to extract from your simulation?
Do you just want to visualize the flow?

Why did u choose the QUICK parameter?

blin · January 11, 2010, 11:58

I want to visualize the flow, along with gather data on the velocities and pressures around the car.

I chose QUICK because it was according to an similar example that I have seen in literature.

January 10, 2010, 19:31	Convergence issues for modeling of flow over a car	#1
blin New Member Join Date: Dec 2009 Posts: 10 Rep Power: 17	Hi, I am an amateur at this and would like to ask some CFD gurus for some help . I am trying to model flow over a car body in Fluent (actually, I'm only modeling half of the car split upon its symmetry plane). I have chosen a spalart-allmaras model, with a turbulence intensity of 0.21% and a turbulence length scale of 0.16m. My discretization values are as follows: Pressure 2nd Order (0.2) Momentum QUICK (0.6) Pressure-Velocity Coupling SIMPLE Turbulence QUICK (0.7) The solution converges to residuals of 0.001 when I run it using first order upwind (default) setting. However, when I change it to the above setting the solution does not converge. I have tried running it using first order upwind, and then change the discretization parameters to the above setting and continue running it without initalization, but that does not converge as well. I'm really stuck as to what I can do further. Also, I'm getting a bunch of warnings that says "turbulent viscosity limited to viscosity ratio of 1.000000e+05 in xxxx cells", with the number of cells increasing. I have read on the forums on discussions regarding this topic, but none relating to how to solve it in the spalart-allmaras model, so I'm wondering as to the solution to that as well. Thanks in advance for any input.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
mass flow in is not equal to mass flow out	saii	CFX	12	March 19, 2018 06:21
3D Supersonic flow over a car	TeQuiero_Q8	FLUENT	0	August 13, 2009 04:42
Please help with flow around car modelling!	Tudor Miron	CFX	17	March 19, 2004 20:23
Flow field analysis around the car	Cenk	FLUENT	1	December 26, 2003 19:09
How i can to simulate the air flow over the car?	fernando vicente	CFX	3	August 11, 2003 23:35

January 11, 2010, 06:01		#2
benji2642 New Member Benjamin Join Date: Mar 2009 Posts: 5 Rep Power: 17	What kind of data do you want to extract from your simulation? Do you just want to visualize the flow? Why did u choose the QUICK parameter?

January 11, 2010, 11:58		#3
blin New Member Join Date: Dec 2009 Posts: 10 Rep Power: 17	I want to visualize the flow, along with gather data on the velocities and pressures around the car. I chose QUICK because it was according to an similar example that I have seen in literature.