Getting started: aerofoil sections and car aero data

This is the starting guide for the aero sub-team. Your job is to work out what range of whole-car Cl (downforce), Cd (drag) and aero balance (front/rear split) is actually achievable from real wing choices — the performance sub-team needs that range to know what’s worth chasing.

The tool: XFoil

Under the hood, this project uses XFoil, a widely-used 2D panel-method solver for aerofoil sections. Give it a shape and an angle of attack and it computes the resulting lift and drag coefficients (Cl, Cd) for that 2D section — this is a real viscous solve, not a lookup table, so it can fail to converge for some AoA/section combinations (more on this below). You never call XFoil directly; api.naca4Aero wraps it for you.

From a wing choice to Cl/Cd

A wing shape here is a NACA 4-digit section, e.g. '2412':

• 1st digit: max camber, as %chord (2 = 2%)
• 2nd digit: position of max camber, in tenths of chord (4 = 40%)
• 3rd/4th digits: max thickness, as %chord (12 = 12%)

'0012' is a symmetric section (no camber); a cambered section like '2412' naturally generates more lift for the same angle of attack.

Two functions matter for you:

[Cl, Cd] = api.naca4Aero(section, AoA)

Runs one wing section at one (or several) angles of attack and returns its 2D Cl/Cd. Useful for understanding one section in isolation. AoA is NEGATIVE for downforce — every wing in this project is modelled mounted upside down to generate downforce (a NACA code can’t express negative camber directly, so this is how it’s done instead), and that’s the default behaviour, so you don’t need to set anything extra to get it.

[CL, CD, aeroBalance] = api.genCarAeroData(frontSection, frontAoA, rearSection, rearAoA)

The function you’ll actually use. Give it a front wing and a rear wing choice and it returns the whole-car Cl, Cd and aero balance — already converted for wing area, already including a fixed body/floor contribution and induced drag, and already in the right sign convention to feed straight into the performance sub-team’s simulation. Same convention as above: frontAoA/rearAoA are negative for downforce. See help api.genCarAeroData for the full set of options (wing chord/span, body aero, etc.) if you want to go beyond the defaults.

Two more functions help you look before you leap:

api.plotNACA(section, AoA)   % plots exactly what that section/AoA looks like — instant, no XFoil
api.genAeroPolar(section)    % sweeps AoA and plots Cl, Cd and the drag polar

api.plotNACA draws the section mounted upside down and pitched at AoA, matching exactly what api.naca4Aero is analysing underneath — run it on your chosen section/AoA before spending time on a real polar, just to sanity-check you’re picturing the right thing.

Suggested first steps

1. Run api.plotNACA(section, AoA) on a few candidate section/AoA combinations to see what they actually look like.
2. Run api.genAeroPolar on one or two sections to see how Cl and Cd respond to AoA — this is your first sensitivity study. Look for where the curve goes ragged or stops: that’s stall/non-convergence.
3. Pick a front and rear section and sweep AoA through api.genCarAeroData for each, building up a table of achievable [CL, CD, aeroBalance] combinations. This is the design space you’ll hand to the performance sub-team.
4. Try a small design-of-experiments (DOE) style sweep — vary front AoA, rear AoA, and/or section choice systematically rather than one at a time — to map out the achievable Cl/Cd/aero-balance space more efficiently than a manual search.

A few things to expect

• XFoil sometimes won’t converge, especially at negative AoA on a cambered section, or near stall. api.naca4Aero/api.genCarAeroData return NaN when this happens rather than erroring — treat NaN as “not achievable,” not as a bug. Sweeping many AoA values in one go and filtering out NaN afterward is easier than chasing every failure individually.
• More downforce always costs more drag — partly the section’s own Cd rising with AoA, partly induced drag (a real, finite wing pays a drag cost for the lift it generates, roughly proportional to Cl²). There’s no way to raise Cl for free; the question is how steep that trade-off is for a given section, which is exactly what your sensitivity study should reveal.
• Front and rear AoA together set aero balance, not just total downforce — the performance sub-team needs both the achievable Cl/Cd range and how balance moves within it, not just a single best point.