Race car simulation, urgently need..

jhthoh · February 19, 2011, 01:47

1)does anyone know about simulation of race car or normal car to get its drag and lift coefficient?
2) how to setup correctly for the Ansys fluent simulation? can i just use the default setting?
3) do i need to create a domain for the air flow?

sorry if my question is disturbing. I m totally new in this CFD bt i need to submit it in mid of march..thanks alot for any help..

rkumar310 · February 19, 2011, 01:59

You are looking to do an external aerodynamics simulation. Here's what you need to do
1. Get the CAD model for the car.
2. Just keep the external surfaces
3. Create an external domain (rectangular) preferably about 10-15 lengths away from the model so that the boundaries dont influence the solution.
3b. You might be able to exploit symmetry to reduce cell count depending on geometry and flow conditions.
4. Create some sort of a hybrid mesh (boundary layers + tet cells)
5. You might be able to just use incompressible flow (since the speed of the car is typically much smaller compared to speed of sound)
6. Specify inlet velocity
7. Solve for flow (defualt settings should be fine)
8. Query the surfaces for forces in the x and y direction to get the drag and lifty coeff)
9. A better way is to monitor the forces as the solution proceeds to make sure it is converged.

jhthoh · February 19, 2011, 05:42

do i only need to run in 2d solver or need to run in 3d solver?
below is in 3d solver case:
then how about the boundary condition?
the domain will become 1 body in the solver,do i need to separate them into few faces?so that i can set them as inlet or outlet?
then the cell zone,i only need to keep them stationary?

rkumar310 · February 19, 2011, 08:26

You will need a 3d simulation. There will be one cell zone which will be stationary. You will specify the speed of the car as the inlet velocity (the boundary towards the front of the car). The outlet will be pressure outlet. The side boundaries can be symmetry or velocity inlet (with the same conditions as the front inlet).

jhthoh · February 19, 2011, 22:06

1) Is that the simulation should run in transient?
2) if i make the air cell zone moving,moving mesh or reference frame is more suitable?
3) In the boundary condition there, it have "interior-air" and "interior-solid", and their type is default set to be interior, should i change them to any type or jz remain them?
4) Need to use dynamic mesh?
5) About the reference value, should be compute from where and which reference zone?
6) Initialization compute from where?

Sorry i have too many question, but i not understand what the Help file say. Thanks for ur patience.

jhthoh · February 19, 2011, 22:34

"Divergence detected in AMG solver: pressure correction". What is this mean?

I have copy paste my setup in word file.bt its too large to upload here. So if u dun mind,can u help me check? if can pls drop me an email in txmjunhong@yahoo.com

thanks alot..

February 19, 2011, 01:47	Race car simulation, urgently need..	#1
jhthoh New Member jhthoh Join Date: Feb 2011 Posts: 11 Rep Power: 15	1)does anyone know about simulation of race car or normal car to get its drag and lift coefficient? 2) how to setup correctly for the Ansys fluent simulation? can i just use the default setting? 3) do i need to create a domain for the air flow? sorry if my question is disturbing. I m totally new in this CFD bt i need to submit it in mid of march..thanks alot for any help..

February 19, 2011, 22:34		#6
jhthoh New Member jhthoh Join Date: Feb 2011 Posts: 11 Rep Power: 15	"Divergence detected in AMG solver: pressure correction". What is this mean? I have copy paste my setup in word file.bt its too large to upload here. So if u dun mind,can u help me check? if can pls drop me an email in txmjunhong@yahoo.com thanks alot.. Last edited by jhthoh; February 19, 2011 at 23:21.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
HELP - Moving car simulation in fluent	Brad Wells	FLUENT	7	January 4, 2018 19:55
Simulation inside car cabin	IronMike	FLUENT	1	April 10, 2016 09:06
Car park ventilation and impulse fans	guillaume	FLUENT	3	June 27, 2011 11:17
Replacing mesh while running a simulation	akultane	CFX	1	November 15, 2009 13:46
3-D Contaminant Dispersal Simulation	Apple L S Chan	Main CFD Forum	1	December 23, 1998 10:06

February 19, 2011, 01:59		#2
rkumar310 New Member Rahul Kumar Join Date: Feb 2011 Location: West Lafayette, IN Posts: 9 Rep Power: 15	You are looking to do an external aerodynamics simulation. Here's what you need to do 1. Get the CAD model for the car. 2. Just keep the external surfaces 3. Create an external domain (rectangular) preferably about 10-15 lengths away from the model so that the boundaries dont influence the solution. 3b. You might be able to exploit symmetry to reduce cell count depending on geometry and flow conditions. 4. Create some sort of a hybrid mesh (boundary layers + tet cells) 5. You might be able to just use incompressible flow (since the speed of the car is typically much smaller compared to speed of sound) 6. Specify inlet velocity 7. Solve for flow (defualt settings should be fine) 8. Query the surfaces for forces in the x and y direction to get the drag and lifty coeff) 9. A better way is to monitor the forces as the solution proceeds to make sure it is converged.

February 19, 2011, 05:42		#3
jhthoh New Member jhthoh Join Date: Feb 2011 Posts: 11 Rep Power: 15	do i only need to run in 2d solver or need to run in 3d solver? below is in 3d solver case: then how about the boundary condition? the domain will become 1 body in the solver,do i need to separate them into few faces?so that i can set them as inlet or outlet? then the cell zone,i only need to keep them stationary?

February 19, 2011, 08:26		#4
rkumar310 New Member Rahul Kumar Join Date: Feb 2011 Location: West Lafayette, IN Posts: 9 Rep Power: 15	You will need a 3d simulation. There will be one cell zone which will be stationary. You will specify the speed of the car as the inlet velocity (the boundary towards the front of the car). The outlet will be pressure outlet. The side boundaries can be symmetry or velocity inlet (with the same conditions as the front inlet).

February 19, 2011, 22:06		#5
jhthoh New Member jhthoh Join Date: Feb 2011 Posts: 11 Rep Power: 15	1) Is that the simulation should run in transient? 2) if i make the air cell zone moving,moving mesh or reference frame is more suitable? 3) In the boundary condition there, it have "interior-air" and "interior-solid", and their type is default set to be interior, should i change them to any type or jz remain them? 4) Need to use dynamic mesh? 5) About the reference value, should be compute from where and which reference zone? 6) Initialization compute from where? Sorry i have too many question, but i not understand what the Help file say. Thanks for ur patience.