Choosing ratios: top gear, gaps & corner gears
Pick top gear for the longest straight in race conditions (tow!), space the intermediates for the speed bands the track actually uses, and choose corner gears so the exit sits in the rpm of best acceleration without a forced mid-exit shift.
Choosing ratios: top gear, gaps & corner gears: L1 Foundation explanation
Selecting ratios follows a standard order (Segers §6.5; Van Valkenburgh):
- Top gear for the longest straight. The aim: at the braking point, the engine is just at (or slightly below) the shift rpm — not bouncing on the limiter for the last 100 m (lost speed), not 800 rpm short (wasted acceleration in the gear).
- Intermediate gears to cover the corner speeds of the track with the best acceleration and as few awkward shifts as possible.
- Corner gears: for each important corner (those before long straights first), check where in the rpm band the exit happens.
Important practical points:
- Tow (slipstream): in a race, a car in the tow can be 5–10 km/h faster at the end of a long straight. Leave rpm margin, or the following car hits the limiter exactly when it wants to overtake. Qualifying-only gearing is a different choice.
- Wind: a head/tailwind on the main straight changes end speed by several km/h — check the forecast and the data from previous sessions.
- Fixed ratios: in most homologated categories (GT3, F4–F2, LMP2, one-makes) you cannot change ratios. You then apply the same thinking to shift points, corner-gear choice and wing level (wing changes top speed — you can sometimes "gear" the car with drag).
Corner gear rule of thumb (Segers): find the rpm at which the car achieves its maximum longitudinal acceleration in each gear; the rpm where the car accepts full throttle on the exit should be slightly below that point. If it is far below, the gear is too long for that corner; if the driver must shift almost immediately after reaching full throttle, it is too short.
And a warning from Segers worth memorising: don't choose ratios for a corner exit while the chassis is unbalanced — handling problems move the full-throttle point down the road and you'd gear for a problem you should fix.
Choosing ratios: top gear, gaps & corner gears: L2 Practitioner explanation
Top gear procedure:
- From previous data/sim, predict end-of-straight speed in race conditions (with tow margin). Example: 268 km/h alone, +6 km/h in tow → 274 km/h.
- Required total ratio so that the engine reaches shift rpm (7500) at that speed: .
- Check that the drag-limited top speed isn't much lower than this (otherwise the car would be at low rpm in a gear it can never pull) — then it may be faster to put top gear near the power peak at the real terminal speed.
Gaps:
- Build a speed histogram of the lap: time spent in each speed band. Make ratio steps smaller in bands where the car spends a lot of time accelerating at full throttle.
- Check every upshift's post-shift rpm stays above the point where power drops sharply.
- Avoid shifts in braking zones or mid-corner: a gear that runs out 20 m before a braking point, or in the middle of a fast kink, forces either a limiter hit or an upshift then downshift.
Corner gears — trade-offs beyond pure acceleration:
| Lower gear | Higher gear |
|---|---|
| More wheel torque per pedal %: better acceleration if not traction-limited | Smoother torque delivery: easier traction, less TC |
| More engine braking on entry (more rear coast locking) | Less engine braking: freer entry rotation, less stability |
| Possibly an upshift during the exit | Engine might be below the torque band on pickup (turbo lag) |
| Higher rpm mid-corner: more throttle sensitivity | Fewer shifts per lap |
Drivers often prefer the higher gear in slow corners on worn tyres or in the wet — and they are often right: if the exit is traction-limited, extra torque multiplication adds nothing but wheelspin.
Validate with data: after a change, compare speed at fixed distance points after the exit (e.g. +50 m, +150 m, +300 m), not the minimum speed — exit gain shows up down the straight.
Choosing ratios: top gear, gaps & corner gears: L3 Advanced explanation
Formalising the corner-gear decision. Over the exit and following straight, time is . The gear affects through wheel force and through the shift penalty (≈ 30–100 ms of zero drive plus a torque disturbance). A practical analysis:
- From the theoretical-rpm channels (Segers Eq. 6.4), compute for each candidate gear the wheel force available along the exit (engine map at that rpm × ratio).
- Cap it with the traction limit estimated from the slip/TC data of the current lap (where TC cuts or slip exceeds target, the car is traction-limited).
- Integrate forward, insert shift events, compare at a fixed distance.
This shows that the decision is often dominated by the traction-limited fraction of the exit. If 70 % of the exit distance is traction-limited in both gears, only the remaining 30 % and the shift penalty differ — and the higher gear frequently wins once the shift penalty is included.
Top gear in an energy- or drag-limited regime. With power capped (BoP/hypercar) or restricted (flat power above choke), the car's terminal speed is set by . Gearing top gear "long" so peak power is reached exactly at terminal speed maximises top speed; gearing it shorter improves mid-straight acceleration but hits the limiter in the tow. The best choice depends on whether the track's long straights end in overtaking zones (race) — the tow case usually dominates for race gearing.
Ratio sets and constraints. Real gearboxes have constraints: available pairs per gear position (centre distance fixed, sum of teeth constrained), final-drive options and sometimes a homologated list with a per-season ratio limit. Optimisation is therefore discrete: enumerate feasible sets, pre-screen with the power-gap metric weighted by the track's speed histogram, then run a lap sim on the top few. Include driver workload (shifts in fast corners) and reliability (gear loads in 1st/2nd, dog wear) as constraints, not afterthoughts.
Choosing ratios: top gear, gaps & corner gears: L4 Expert explanation
Ratio selection as a constrained discrete optimisation. The real problem is: choose one pair per gear position plus a final drive from finite lists, subject to centre-distance/tooth-sum constraints, so as to minimise lap time (or race time) with gear-life, driver-workload and fuel constraints. The objective is noisy and non-convex, because a ratio change moves shift points and shift locations in braking zones and kinks on and off. Gradient methods on a continuous ratio are only good for exploring sensitivity. A workable pipeline:
- Enumerate feasible sets (often thousands). Reject any set that puts a shift in a listed corner or a braking zone, or leaves less than the tow margin in top gear.
- Pre-screen with the distance-weighted force-gap metric (Level 3 of the fundamentals lesson) on the current lap's speed trace.
- Lap-sim the best 10–30 with a quasi-steady-state or transient sim that includes shift time, cut/ramp-in and traction limits. Run it in race conditions too (tow on the main straight, worn tyres, heavier fuel).
- Robustness: rank by expected lap time over a spread of wind, tow and grip. A set that is 0.02 s faster in calm conditions but hits the limiter with a 10 km/h tailwind is worse.
Gear life as a hard constraint (Miner's rule). Each gear's teeth (bending and contact stress) and dogs have an S–N curve. From logged data build a torque–cycles histogram per gear (rainflow or simple torque bins × tooth-mesh cycles = rpm × time / gear ratio). Then sum damage across bins. Fatigue life is very sensitive to torque (S–N exponents of ≈ 6–10 for tooth bending), so a ratio change that makes a gear work 10 % harder can halve its life. Changing a pair changes both the tooth load (pinion torque = engine torque × any upstream drop-gear ratio; output-gear torque = that × the pair ratio) and the number of mesh cycles (time and rpm spent in that gear, and where on the track the gear is used). The net damage change is therefore not obvious. Calculate it; intuition misleads here. In endurance racing, a set that is 0.05 s faster but doubles 2nd-gear damage may not survive 24 h.
When the quasi-static assumptions fail.
- Traction-limited corners with TC: the sim's tyre model and TC model decide the result. Validate them against slip data before trusting a "shorter 2nd is faster" result.
- Turbo engines: boost build-up after a shift or on pickup is dynamic. A static torque curve overstates low-rpm torque after a long cut. Use a transient engine model or measured post-shift torque.
- Hybrid/energy-limited cars: the optimum depends on the deployment strategy, because ratio choice changes where the energy-limited power is deployed. Ratio choice and deployment must be optimised together.
Final drive as the cheap lever. Where only the final drive can be changed, all total ratios scale together. Its effect is a nearly pure trade-off between top-gear margin and acceleration in all gears. The sensitivity is then a single number per track that the sim can give you before the event.
Key formulas
Trackside takeaways
Top gear for the longest straight in race conditions — leave margin for the tow.
Space intermediates for the speed bands where the car spends time at full throttle; avoid forced shifts in braking zones and kinks.
Full-throttle point on exit should be slightly below the rpm of max longitudinal acceleration.
Never gear around a chassis balance problem — fix the balance first.
With fixed ratios, "gearing" becomes shift points, corner-gear choice and drag level.
Judge exit-gear changes by speed at fixed points down the straight, not by minimum speed.
On your car
Ratios are chosen per event from the approved set; oval vs road/street packages differ completely. Top gear on road courses must cover push-to-pass and tow.
Check yourself
In qualifying the car reaches 7450/7500 rpm in top gear at the end of the main straight. For the race, what is the main risk with this gearing?
- Segers – Analysis Techniques for Racecar Data Acquisition — §6.5 Determining Correct Gear Ratios · gear selection order, rpm at max longitudinal acceleration, chassis balance first
- Segers – Analysis Techniques for Racecar Data Acquisition — §6.6 Determining in Which Gear to Take a Corner
- Segers – Analysis Techniques for Racecar Data Acquisition — §6.1.1 Shifting Point
- Other — P. Van Valkenburgh, Race Car Engineering and Mechanics (HP Books, 1993; orig. 1976) · ratio selection, referenced by Segers [6-1]