[A7A]

Hi,

I'm trying to simulate air flow around peugeot 206CC to calculate the values of drag and lift coefficients, but after finishing the simulation and doing the calculations i get good results of drag coefficeint(approximatly Cd=0.28), unlike the results of the lift coefficient(Cl=-0.4)
The turbulence model being used is (Standard k-e), and the solution method is (first order upwind).

Could you give me any advice.
Thank you in advance.

[blackmask]

Why is this result incorrect/inappropriate? What should be the reasonable value? It is my suggestion that never use the first-order scheme to produce the final result.

[xiaomi]

may be you can use the other turbulence model and try to use the second-order scheme!

[A7A]

Originally Posted by blackmask

Why is this result incorrect/inappropriate? What should be the reasonable value? It is my suggestion that never use the first-order scheme to produce the final result.

Thank you for your reply,
After searching on the internet i found that the car lift coefficient for this type of cars should be positive (around 0.1). I've tried to use second order scheme, but when using it i get convergence problems.

[A7A]

Originally Posted by xiaomi

may be you can use the other turbulence model and try to use the second-order scheme!

Thank you for your reply,
I've tried to use RNG K-e but it gave me the same results.

[oj.bulmer]

Have you done mesh independence? Are you sure the solution is converged? Have you tried second order scheme along with RNG? Although, my choice of model would be Realizable model.

OJ

[A7A]

Originally Posted by oj.bulmer

Have you done mesh independence? Are you sure the solution is converged? Have you tried second order scheme along with RNG? Although, my choice of model would be Realizable model.

OJ

Thank you for your reply,
I've tried the Relizable model+Second order, but i still get the same results (negative lift coefficient), even though the drag coefficient results are good (which means that the pressure coefficient variation along the car should be true)

[xiaomi]

if possible, you can upload you .cas and .dat files then we can check them to find the problem

[jthiakz]

Why dont you try with one-equation model spallart allmaras, recommended for external flows

[A7A]

Originally Posted by xiaomi

if possible, you can upload you .cas and .dat files then we can check them to find the problem

Unfortunately, i can't upload them because of the big size, but i've uploaded some pictures of the mesh and B.C

[xiaomi]

hi i think the result of lift may be reasonable due to the ground effect, as the ground is moving, the pressure between the car and ground may be less than the surface of the car, then the lift may be negative, you can check the pressure contour to see what happened!

[Umuteng]

i think negative lift coefficient is normal. For a car you dont want to fly. You want to feel the ground and attached to it. Race cars are designed some diffuser mechanisims to get more downforce. They have even higher (but negative) lift coefficient then you found. I think you should search more about car aerodynamics. I belive your results are okey. However, you should see mesh independence. Although i am not sure about spalart almaras it seems kind of enough for your case. And you should use second order methods. Furthermore the negative lift is partially is due to ground effect, therefore you must be very carefull when you are determining ground to car distance. There must be a some treshold value at which lift of the car becomes negative. Think like that; if ground clearance is very low flow velocity at the bottom of the car will be higher then the flow at the upper surface of the car which results in downforce(negative lift). But if your ground clearance is high(er then some specific value) the flow at the bottom of the car is slow and this results in positive lift coefficient.