Possible error in Cd value

Good morning everyone,

I am working on a research project to improve the aerodynamics of a land rover defender 110. In doing so, I am simulating with different concepts. Now simulating the baseline has gone well. Now I have only created my new concepts from which I have simulated two models in SimScale. One model here has a flat floor and the other has optimised wheel arches. The expectation of this was, of course, that the Cd values would be lower, but they turned out to be higher. Now I just have my doubts about the reliability because new low airspeed areas have appeared in the four corners of the wind tunnel (see appendix). In my opinion, this is not quite right, but I don’t know where the error lies because the settings are exactly the same as those of the simulation of the baseline model, which had gone well. In addition, the mesh at these four corners also looks good.

![image|690x432 (upload://7EXgOUG23YAy9ZT3hj6cFX7rW5a.jpeg)

Here my Project link of the new concept: SimScale
Here my Project link of the Baseline model:

Hi @0987283 ,

Thanks for using the forum! This is Ajit from the team.

I went through your project and it looks interesting. Seems like you have simulated quite a few runs and now trying the Ahmed body validation?
The artefacts you see at the corners of the wind tunnel (shown in the image above), might be caused by one of the following reasons:

  • Insufficient upstream, downstream and side width of the wind tunnel to accommodate the local flow characteristics. - can be solved mostly by extending the wind tunnel dimensions. Please refer the Ahmed body validation or the Tutorial: Aerodynamics Simulation of Flow Around a Vehicle for general best practices on Wind tunnel dimensions

  • Insufficient iterations can also cause such velocity gradients in a simulation. Please try to run the simulations longer (say >1000 iterations) to see if they disappear

  • Convergence might also affect the cd value on the new case. Please try to monitor the pressure on the vehicle body to see if the solution is converged.

  • I am sure you followed a good mesh practice here, but please make sure you resolve the surfaces with required y+ values to achieve good drag results.

I would be happy to see if the above comments helped your case and discuss further.

With Kind Regards,

1 Like