LSR (Land speed record) car simulation on FLUENT

Maxime31850 · April 27, 2013, 17:17

Hi,

I'm a student who is doing analysis concerning a wheel driven Land Speed Record car. I've received the datas concerning the maximum speed that the car can reach, which is 460 mph.

I am using the software ANSYS v14.0 to draw, mesh the geometry and I'm using an old version of FLUENT to run the 3D simulations.
The geometry is quite simple, it is without the wheel.

Concerning the FLUENT simulation, I've encountered some problems .

As I said on the top of this post, the maximum speed of the car will be 460mph, which means that the type of flow is compressible ( Mach Number = 0,6 >0,3).

As a consequence, I cannot use the same boundary entities like for a incompressible flow. I used then Pressure Far Field for the right and left side (as pressure far field inlet and pressure far field outlet) and on the top (pressure far field top).

The problem is that when I tried to run the simulation on FLUENT, the residuals never converged or when I change some parameters some errors appeared.

Here are the parameters I've set in FLUENT for my last simulation :

Define : Models : energy equation : Ok
viscous model : k-epsilon model (standard)
Materials : air : ideal gas from the density
Operating conditions : 0 for the gauge pressure
Boundary conditions : gauge pressure : 101325 Pa
mach number : 0,6
x component flow direction : -1
Turbulence : turbulent viscosity ratio
5% and 5 for the ratio

After that in the tab Solve : I initialize from Pressure far field inlet and run the simulation.

I think the residuals don't converged because of the parameters but I don't really exactly know what do I have to put.
Could you give me some advices concerning them ?

Thank you very much

DonQuijote · April 30, 2013, 07:13

Hi Maxime.

I haven't done any compressible analysis in Fluent, but did you check your mesh? Did you make sure that it is fine enough in the important zones? Mesh is usually one of the main reasons of non-convergence.

I can't help you about BC, but have you tried to change them (i.e. making the top wall as 'wall')?

Regards

Maxime31850 · May 1, 2013, 12:15

Hi,
firstly thank you for answering.
Regarding the mesh, I have done a face sizing by selecting all the faces from the car and I have done an inflation around it.

About the inflation, in ANSYS, I have chosen in the tab inflation option total thickness and I put 30 boundary layers with a growth rate of 1.2.

I put one inflation too on the road, in order to have accuracy results concerning the ground effect with less boundary layers than around the car.

Do you think these parameters sound good? When I check the Y+ on FLUENT, I have got one located between 50 and 500.

April 27, 2013, 17:17	LSR (Land speed record) car simulation on FLUENT	#1
Maxime31850 New Member Maxime Join Date: Apr 2013 Posts: 5 Rep Power: 13	Hi, I'm a student who is doing analysis concerning a wheel driven Land Speed Record car. I've received the datas concerning the maximum speed that the car can reach, which is 460 mph. I am using the software ANSYS v14.0 to draw, mesh the geometry and I'm using an old version of FLUENT to run the 3D simulations. The geometry is quite simple, it is without the wheel. Concerning the FLUENT simulation, I've encountered some problems . As I said on the top of this post, the maximum speed of the car will be 460mph, which means that the type of flow is compressible ( Mach Number = 0,6 >0,3). As a consequence, I cannot use the same boundary entities like for a incompressible flow. I used then Pressure Far Field for the right and left side (as pressure far field inlet and pressure far field outlet) and on the top (pressure far field top). The problem is that when I tried to run the simulation on FLUENT, the residuals never converged or when I change some parameters some errors appeared. Here are the parameters I've set in FLUENT for my last simulation : Define : Models : energy equation : Ok viscous model : k-epsilon model (standard) Materials : air : ideal gas from the density Operating conditions : 0 for the gauge pressure Boundary conditions : gauge pressure : 101325 Pa mach number : 0,6 x component flow direction : -1 Turbulence : turbulent viscosity ratio 5% and 5 for the ratio After that in the tab Solve : I initialize from Pressure far field inlet and run the simulation. I think the residuals don't converged because of the parameters but I don't really exactly know what do I have to put. Could you give me some advices concerning them ? Thank you very much

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
HELP - Moving car simulation in fluent	Brad Wells	FLUENT	7	January 4, 2018 20:55
Racing car simulation	Edu	FLUENT	5	October 16, 2009 12:36
Fluent simulation over airships	Aragon	FLUENT	0	March 27, 2009 13:38
Can Fluent start a simulation from a negative time	Ted Chu	FLUENT	10	November 26, 2008 22:06
Car aerodynamics simulation (Pls help, urgent)!	Wang	FLUENT	1	February 23, 2007 17:06

April 30, 2013, 07:13		#2
DonQuijote Member Join Date: Jun 2012 Posts: 32 Rep Power: 14	Hi Maxime. I haven't done any compressible analysis in Fluent, but did you check your mesh? Did you make sure that it is fine enough in the important zones? Mesh is usually one of the main reasons of non-convergence. I can't help you about BC, but have you tried to change them (i.e. making the top wall as 'wall')? Regards

May 1, 2013, 12:15		#3
Maxime31850 New Member Maxime Join Date: Apr 2013 Posts: 5 Rep Power: 13	Hi, firstly thank you for answering. Regarding the mesh, I have done a face sizing by selecting all the faces from the car and I have done an inflation around it. About the inflation, in ANSYS, I have chosen in the tab inflation option total thickness and I put 30 boundary layers with a growth rate of 1.2. I put one inflation too on the road, in order to have accuracy results concerning the ground effect with less boundary layers than around the car. Do you think these parameters sound good? When I check the Y+ on FLUENT, I have got one located between 50 and 500.