Produces the largest share of the car's downforce at the lowest drag cost, by accelerating air under the vehicle through venturi tunnels.
Lift is generated primarily by the pressure drop from continuity and Bernoulli through the tunnel throat, plus the Coanda effect and momentum change through the diffuser. Outboard tunnels are optimised in 3D sweeps and use gurney flaps. A center tunnel was added on CR26I to make use of the space freed by moving away from a front nose. The shelf and half-tube along each side seal the tunnel against atmospheric pressure so the low-pressure region is not bled off from outboard. Design Philosophies and Processes: Parametric CFD sweeps of expansion ratio, choke height, inlet shape and outlet length. Peak efficiency at expansion ratio 5.14, peak downforce at 5.44, flat inlets greatly reduce drag, greater choke heights help, outlet length has little effect, stall is the limit. Choke position is used to locate the center of pressure.
Dimensionally constrained by suspension, exhaust, jack bar, oil pan and the radiators and sidepods. Inlets and outlets positioned to avoid direct tire effects. Shelf ride height set by worst-case heave and roll to avoid ground contact. On CR26I the wide EV frame cut maximum outboard tunnel width from 11.63 in to 6.94 in, which is why the center tunnel exists. On CR26E the accumulator position constrains the design so the tray does not have to come off for accumulator service.
ANSYS Fluent half-car iteration at 35 mph, full-car performance map for heave, pitch, roll, slip and steering angle. Worst-case corner case (heave -0.2 in, pitch -0.7 deg, roll 0.4 deg, slip 2.97 deg, steering 13 deg) gave undertray CLA 0.75, CDA 0.41, efficiency 1.83. Steady-state Excel calculation generates the ride height surface map. From the design reports and one-pagers: Validation per the one-pager: coast down and constant speed testing, plus a physical check using a cardboard mock of the floor nose with the car placed at maximum pitch angle and visually inspected. That second check exists precisely because of the scraping history.
Molds moved from a 3-axis CNC mill to a 5-axis gantry. Single layup by vacuum infusion, with mounting hardpoints included in the laminate. CR25 required a multi-step honeycomb sandwich panel with film adhesive bonding for the honeycomb; CR26I's single infusion removed that step. From the design reports and one-pagers: CR25 one-pager costs: Nomex $205.80; adhesive either Loctite EA 9460 at $311.86 or 3M AU200 0.06 film adhesive at $202.87; mounting $173.43 assuming no scrap stock and no reused hardware. Estimated total $691.09 with EA 9460 or $582.10 with film adhesive.
CFRP skins with aluminium hardpoints embedded at mounting locations. CR26E moved the core to Innegra and G-10 phenolic. CR26I laminate 6.77 lb over 21.87 ft^2, a laminate density of 0.309 lb/ft^2 against CR25's 0.351.
Three Fault Vault entries, all CR26I, all traced to ride height. 2026-03-30, Design, Complete Failure. The floor rode too low and was sanded through to the point of seeing daylight. CR26I, Design, Complete Failure. The flexural nature of the undertray caused scraping. Fixed with spars and safety wire. CR26I (Aero/Suspension joint entry), Design, Compromised Failure. Improper ride height integration between the front wing and the undertray meant packers had to be added. These are the same root cause the design data shows: CR26I was designed at 0.25 in ride height and built at 0.61 in, and the tray is stiffness-limited rather than strength-limited. They are why the transverse stiffeners (top hat centre spar, elliptical fore and aft spars) exist. Separately, the CR25 undertray mold warped badly enough to be unusable and the CR25 undertray itself was damaged in a summer testing incident, which together forced CR26I onto a cross-compatible EV part.
Transverse stiffeners added: a top hat centre spar running the span of the tray and elliptical fore and aft spars between the inboard mounts, plus safety wire, after the flexural behaviour of the tray caused scraping. Inboard mounting was revised to cut attachment and removal time, dropping mount weight from 0.889 lb to 0.556 lb while raising the load requirement from 400 lb to 486 lb.
CR26I: CLA 1.84 designed / 1.63 as built, CDA 0.095, 9.48 lb, L/D 18.90. CR25: CLA 0.91 designed / 0.92 as built, CDA 0.083, 10 lb, L/D 10.94. Design ride height CR26I 0.25 in, as built 0.61 in. From the cost report drawings and CAD: Weighed mass: 3628.7 g (8.000 lb), 3764.8 g (8.300 lb). Recorded masses spanning design variants, 6 values from 3403.0 to 4535.9 g; first three: 3403.0 g (7.502 lb), 3694.4 g (8.145 lb), 3862.7 g (8.516 lb) (recorded as: Undertray Skin). CAD mass, CR22 master assembly: 8.967 lb (4.067 kg). CG at X 51.64 in, Y -0.00 in, Z 1.63 in in vehicle coordinates. (CAD component: Undertray Skin - Manufacturing - Final) Dimensions called out on the cost report drawings. These are the figures printed on the drawing, not a labelled specification; open the drawing itself for what each one refers to. Undertray Joint (mm): TRUE R2.4 TRUE R7.3, 7.1, 50.1, 12.7 Source: Subteam weight books: CR22/Aerodynamics/Manufacturing/CR22_Aerodynamics_Weight.xlsx | SolidWorks mass model exports: CR22 (5 May 2024), CR25 (4 Jul 2025), CR26I (30 Jul 2025), CR26E (22 Mar 2026) | Cost report drawings: CR26E/Aerodynamics/Cost Report & Drawings/Undertray/Undertray Joint.pdf
Nothing in CrimsonStorage covers these yet.