Learning Forum

[Josiah Kleven]

I am trying to do a CFD simple analysis on the shell for a solar racecar. To do so I slightly modified the shell to include wheels that represent the tires that will protrude from the shell. Image for reference.

 

Everything went smoothly through spaceclaim and the fluent mesher. But when I run the solution the velocity and pressure around the wheel spike dramatically. The solver is supposed to simulate motion at 45 mph or 20.1168 m/s. The tires have a radius of 10.65 inches or 27.051 cm. This means that the tires are supposed to rotate at 74.36697 rad/s. I have set the centroid correctly as there clearly is rotational motion in the correct places in the solver. The problem is that the moment air touches or even comes near the tires the pressure and velocity increase astronomically. This leads to incorrect lift values and drag values that are negative. Pathline, vector, and contour plot for reference.

It's worth noting that I've created "contact patches" where the tire meets the ground and where it "dissapears" into the body of the car. the flat portion in red is an idealization that assumes at high enough speeds air will virtually not flow in and out of the body. This idealization as well as both contact patches are set to be entirely frictionless.

I simply cannot understand why the wheels are exhibiting this behavior. The center of rotation is correct, the named selections include only the part of the tire that should be rotating, I cannot understand what could be causing the issues I'm seeing. I should mention that these images are taken on a mesh that is only about 600,000 cells because on my personal computer I'm limited by the student licence. Tomorrow I'm going to use the full licensed computers at my university to see if a more refined mesh solves the issue but I seriously doubt it.

If anybody has any suggestions or experience with this any advice is welcome. I'm new to ansys fluent so I have little to no prior experience to rely on. For reference I'm using the K-omega sst turbulent model, accounting for curvature. Air is set to have a constant density, The inlet is a velocity inlet with speed 20.1168 m/s, the outlet is a pressure outlet set to 0 gauge pressure. The ground is moving at 20.1168 m/s while the shell and contact patches are stationary walls. The wheels are set to have only rotational motion of the previously mentioned angular velocity.

[CFD_Friend]

Hi,

Can you please go through these videos and see if proper boundary conditions were applied. Or else as you mentioned maybe a proper refined mesh can help with this case.

Ansys Fluent Multi-GPU Solver Features: DrivAer Car| Ansys Innovation Courses

Solver Setup in Ansys Fluent — Lesson 3 - ANSYS Innovation Courses

[ergispap2]

For future reference, in case anyone encounters the same issue:

I had a similar problem and I solved it by inspecting the rotation axis origin of the front wheel of my model. For some reason, the dot in the number literals had been omitted in all three cartesian coordinates.

This meant that x = -89.654 became x = -89654 and resulted in a completely misplaced axis of rotation.

So indeed it was the setup conditions after all...