F32 Rear Wing | Kip Gaddis Portfolio

Formula SAE Tri-Element Rear Wing

I was an aerodynamics design engineer for RIT's FSAE team this year and part of my job was designing a rear wing for our electric race car. The hardest part of this design was the tradeoffs between down force, drag, mechanical constraints, and manufacturability. The aero team this year learned a lot from our legacy designs and we were able to make some major improvements this year. We were able to make more downforce with less drag compared to the previous year car.

ISO view of the wing system

This year I designed, and we developed, a tri-element rear wing, it is supported by swan necks for reduced disturbance to the low pressure side of the wing and struts to help with potential side loads in a slide.

2024 F32 full car Solidworks model

Wing structure (left), swan neck to rear wing mounting brackets (top right) and strut clevis (bottom right)

The internal structure of the wing is designed to keep weight down while maintaining a rigid and safe system. Each airfoil element has an internal structure of a-fish to help the skin maintain its shape. The primary element is the main structure of the wing. It is where the wing is connected to the swan necks and how the end plates are attached. It has carbon tubes running length wise to facilitate this.

Wing Elements: Primary (top), Secondary (bottom left), Tertiary (bottom right)

The wing elements were chosen based on initial down force and drag requirements for the car set by the vehicle dynamics team.

The primary is a GOE 227, 15” Chord Length, AOA of 2.5°, 37” wingspan

The secondary is a Be 122-155 (Same as previous generation car), 9” chord length w/ 1.125” cut off, Slot Gap 0.633” (2.57%c), Overlap 0.766” (3.11%c), AOA of 30°

The tertiary is a GOE 448 (Same as FW secondary for ease of manufacturing), 4” cord length, Slot Gap 0.299” (1.22%c), Overlap 0.533” (2.17%c), AOA of 52.5°

The tertiary has a gurney flap located at the traiing edge of the element and is at a 90 degree angle to the chord of the wing

Slot gap and overlap was initially chosen based off recommendations from literature and further refined with iterative ANSYS fluent CFD simulations

For our aerodynamic analysis for the aero package we did numerical simulation using Ansys 23R1and 22R2 with the turbulent model k-ω SST. Free-stream flow velocity was 15 m/s and 30 m/s (Re=9.08x10^5) for most models unless there was a specific condition we wanted to analyze differently. These relatively high velocities are not great for our situation, we tend to be driving, on average, far slower than these simulations (average corning speed is 11.1m/s). There are a few issues with the current way we run CFD simulations that I am working to help change. Like I previously mentioned we run too high of velocities but we also currently don't run mesh convergence. I am currently working on this by running an airfoil with pitot tubes along its chord in the wind tunnel at RIT and comparing to an identical CFD sim as well as bringing the car to a wind tunnel for validation.

Tertiary

Tertiary Mold

Tertiary Mold Close Up of Patches

We encountered issues manufacturing the secondary and tertiary element due the mold for the carbon layup being fairly thin and would warp in the oven if we did our normal as4 pre-preg carbon layup for these elements. We did a wet layup for these elements but this brought its own issues post curing. Most notably the surface finish of the elements was quite warped and bumpy, in areas there were wrinkles thicker than the carbon skin. Sanding helped quite a bit to smooth everything out but the deeper warping or larger wrinkles we sanded as far as we could and then smoothed out further with multiple layers of clear coat

Tertiary Skin Wrinkles

Secondary Skin Bumps

Secondary Skin Bumps Close Up

Tertiary mold broken after one use.

The wet layup also caused issues with the skin bonding to the mod in areas it wasn't anticipated it would causing the de-molding process especially difficult. The mold ended up cracking and later breaking because of this.

New tertiary split mold

Due to the mold breaking and the warping issues and poor layup quality of the tertiary we decided to make a new female split mold for the tertiary that would allow us to do a prepreg layup for the element.

Wing Endplate

A notch to reduce vortices through vortex cancellation of wing tip vortex and high pressure spill over vortex, swept at front and bottom edge to reduce vortices by eliminating sharp corners and allow smoother mix of pressure differentials, size maximized to hold low pressure and allow for ease of serviceability of car critical systems such as drive train and accumulator.

Wing Endplate Flow vis CFD Showing Wing-tip Vortex

CFD Simulation Showing Pressure of the RW

Wing endplate manufacturing, water jet TeXtreme spread tow weave carbon sandwich panels with aluminum and 3D printed inserts

Swan Necks

Swan necks were chosen for their reduced disturbance to the low pressure side of the wing given their small frontal area, increased rigidity and easy maintenance compared to legacy system, angled inwards for increased lateral rigidity.

Optimized vs. Unoptimized Swan Necks Static Structural Deformation Simulation

Optimized vs. Unoptimized Swan Necks Static Structural Strain Simulation

Metal Swan Neck Design Consideration Topology Optimization Simulation

Metal Swan Neck Design Consideration

We thought of doing water jet aluminum swan necks instead of carbon but quickly ruled them out for performance and cost reasons.

To test the structure of the system we did ANSYS static structural simulations where we imported the aero load from a CFD sim and used it for the load on the wing to be as accurate as possible.

Simulation

Aero Load Top Side

Aero Load Bottom Side

Aero Load Wing Deformation at 30 m/s

Aero Load Swan Neck Strain at 30 m/s

We also simulated the car in a full sideways slide. Below is the deformation caused by the slide.

Side Load Swan Neck Strain

After seeing the simulation results we decided that adding struts from the end plates to the chassis in combination with the swan necks would be a good decision for safety and reliability. While the swan necks in theory would have been plenty the safety of the diver and car is paramount.

Aero Load Deformation With Struts

~72% Reduction in Deformation

Aero Load Swan Neck Strain 45m/s With Struts

Aero Load Strut Clevis Stress 45m/s

Side Load Deformation With Struts

Side Load Swan Neck Strain With Struts

Building of the Rear Wing

Swan Necks and A-Fish Fresh Off The WaterJet.

Primary element manufacturing went far better then secondary or tertiary but we did have some issues with bonding where the bottom side would sag a bit due to gravity but we fixed this by bonding the skin while the wing was hanging vertically instead of horizontally.