# Traction control: slip targets, TC maps & cut strategies

> TC keeps driven-wheel slip near the tyre’s peak by cutting torque when measured slip exceeds a target that depends on speed, gear, lateral g and throttle. How it cuts matters as much as when.

Area: Engine & Hybrid · Starting level: L2 Practitioner · Updated 2026-11-16
Source page: https://race-engineer-academy.vercel.app/lesson/engine-traction-control

## L1 · Foundation

A driven tyre produces longitudinal force only by **slipping**: the tyre surface moves slightly faster than the road. The **slip ratio** compares driven-wheel speed to a reference (free-rolling) speed:

$$SR = \frac{V_{driven} - V_{ref}}{V_{ref}}$$

Force rises with slip up to a peak — for racing slicks typically around **6–12 %** (lower for stiff-carcass, high-downforce tyres, higher for wet tyres) — and then falls off. Beyond the peak the tyre also loses lateral grip, which is why wheelspin on exit turns into oversteer.

**Traction control (TC)** measures slip continuously and, when it exceeds a target, reduces engine torque until slip returns below target. In Segers' Bosch example, actual slip briefly reaches 6 % against a 4 % target and the system retards ignition until it is back under.

Basic building blocks:
- **Reference speed** – from the undriven wheels (or vehicle-speed estimate in AWD cars).
- **Slip target** – a table, e.g. vs speed and gear; often a driver-selectable **TC level** shifts the whole table.
- **Intervention** – ignition retard, ignition/fuel cut of some cylinders, or throttle closing.

What TC is **not**: a cure for a car with poor rear grip. If TC is cutting a lot, the tyres are being asked for more than they have; the lap time is in setup, tyre management or the driver's throttle application. Well-set TC is mostly *invisible* in the data — brief interventions at the peak — and lets the driver go to full throttle earlier with confidence.

Typical numbers: GT3 TC dials with 8–12 positions; a step is a few percent of slip target. Drivers often reduce TC (less intervention) on new tyres and add it as the rear degrades.

## L2 · Practitioner

**Reading TC in the data (Segers §4.3).** Essential channels: driven-wheel speeds (both sides), reference speed, slip target (from the ECU), TC torque reduction or ignition-cut %, throttle/pedal, gear, lateral g.

- Compute your own slip channel, $SR = 100\cdot(\tfrac12(V_{RL}+V_{RR}) - V_{FL})/V_{FL}$ (Segers Eq. 4.13). Better: use the mean of both front wheels to remove the cornering-radius bias of a single front wheel.
- Overlay **slip vs target**: brief overshoots followed by cuts = TC working. Slip sitting well under target with cuts anyway = target table too conservative or a noisy reference speed.
- **Inside-rear slip**: with lateral load transfer the inside rear spins first. Separate left/right slip channels show whether the diff is distributing torque or letting the inside wheel go.

**TC maps.** The slip target is rarely a single number. Typical axes:

| Axis | Why |
|---|---|
| Vehicle speed | at low speed small absolute speed errors give large % slip; targets often higher at low speed |
| Gear | more wheel torque in low gears → faster spin-up, needs earlier/firmer intervention |
| Lateral g / steering | combined slip: allow less longitudinal slip when the tyre is carrying cornering force |
| Throttle / torque request | relax target at small throttle to avoid nuisance intervention |

Most GT3/GT4/LMP2 cars expose two driver dials: **TC slip level** (how much slip is allowed) and **TC cut/aggression** (how hard it reduces torque once triggered). Higher slip level = more rotation and drive but more tyre wear; higher cut = smoother but slower recovery of torque.

**Cut strategies:**
- **Ignition retard** – fast (next combustion), smooth, but hot exhaust gases; sustained use cooks exhaust, turbo and catalyst.
- **Ignition/cylinder cut** – very fast, large torque steps, audible "rattle"; harsher on the driveline.
- **Fuel cut** – slower to act (wall-wetting, port injection), risk of lean cylinders; used for coarse reductions.
- **Throttle closing (DBW)** – slow (≈ 50–150 ms + manifold), but efficient; used for the base torque level, with ignition for fast trimming.

**Pitfalls:** TC on a wet track with a *front-wheel* reference speed that itself locks or aquaplanes; incorrect tyre circumference after a tyre change (a 1 % circumference error is a 1 % slip offset!); and reading "TC active" as "car has no traction" when it is only trimming peaks.

## L3 · Advanced

**Control loop structure.** A typical engine-based TC is a PI(D) controller on slip error with a feed-forward term:

$$\Delta T = K_p (SR - SR^*) + K_i \int (SR - SR^*)\,dt \quad (SR > SR^*)$$

plus a torque **ramp-back rate** after intervention. The trade-off: aggressive gains cut slip overshoot but cause limit cycles (torque oscillating at 3–8 Hz, felt as "rattle" and often visible in longitudinal g). The **driveline torsional mode** (half-shafts + tyre longitudinal stiffness, typically 5–15 Hz) limits the usable bandwidth: cutting faster than that excites the mode and makes slip oscillate.

**The slip target should follow the tyre, not the other way round.** The slip at peak $F_x$ depends on load, temperature, wear and combined lateral slip. With high lateral force, the friction-circle reduces longitudinal capacity; the peak shifts and the post-peak drop becomes steeper. A lateral-g axis that **lowers** the slip target at high $a_y$ prevents the TC from allowing the slip that would destroy lateral grip at the rear — the main cause of TC-on oversteer.

**Reference speed challenges.**
- Single front wheel: in a 50 m radius corner with 1.6 m track, inside and outside front speeds differ by ≈ 3 %. A single-wheel reference gives ±1.5 % slip bias left vs right-hand corners.
- Under braking or trail braking the fronts slip negative; TC reference must not be used while braking.
- AWD (LMH above deployment threshold): no free-rolling wheel. Reference is estimated by fusing wheel speeds, IMU longitudinal acceleration and GPS; errors here directly become TC errors.

**Choosing a TC level as an engineer.** Use a matrix: per corner, compute (a) peak slip on exit, (b) TC torque reduction integral (Nm·s), (c) time from apex to full throttle. Plot each against TC level from a test. The fastest setting usually has *small but non-zero* intervention integral. Over a stint, track TC intervention per lap as a **rear degradation proxy**: rising intervention at constant TC level means the rear is going off — that is information for strategy as well as for the driver's TC dial.

## L4 · Expert

**Engine-based TC vs brake-based TC vs diff.** Engine torque reduction acts on the axle *sum*; it cannot separate left and right. If only the inside rear is spinning, an engine cut reduces both wheels’ torque, sacrificing outside-wheel drive force that the tyre could have used. Brake-based TC (common in road cars, rare and often banned in racing) applies the inside brake, mimicking a locked diff. In a racing car the **differential locking** does this job mechanically — which is why diff setup and TC calibration must be done together: a more open diff produces more inside-wheel spin and more TC activity, a more locked one gives less TC but more power-on understeer/oversteer transitions.

**Ignition retard thermal budget.** Retard reduces torque by moving combustion late; the energy goes into exhaust heat. 30° retard for a couple of seconds per corner across a lap can raise EGT by 50–100 °C and turbine inlet temperature beyond limits; ECUs include a **retard budget** that switches to cylinder cut when exhaust temperature is high. When TC "feels different" late in a stint, check whether the strategy switched modes.

**Torque-domain TC (hybrids).** In a hybrid, an MGU on the driven axle can modulate torque within milliseconds and with no thermal penalty — and on an LMH front axle it effectively becomes a per-axle torque controller. Regulation limits (deployment threshold, ramp rates, total power cap) are written precisely to stop this becoming a full torque-vectoring traction system.

**Limits of the slip-target model.** Peak-μ slip is not constant: transient exits pass through the curve faster than the tyre's relaxation length allows equilibrium (relaxation length ~ 0.3–0.6 m longitudinally). At low speed, the tyre's slip response time is longer and the controller can overshoot. Good calibrations therefore use a slip-rate term (the derivative) as an early trigger rather than raising gains.

## Key formulas

**Slip ratio**

$$SR = \frac{V_{driven} - V_{ref}}{V_{ref}}$$


**Segers slip channel (Eq. 4.13)**

$$SR\,[\%] = 100\left(\frac{0.5\,(V_{RL}+V_{RR})}{V_{FL}} - 1\right)$$


**Driven-wheel speed from rpm (Eq. 4.14)**

$$V = \frac{2\pi\, r\, n_{engine}}{60\, i_{tot}}$$


**PI traction controller**

$$\Delta T = K_p (SR - SR^*) + K_i \int (SR - SR^*)\,dt$$


## Trackside takeaways

- Slicks peak at roughly 6–12 % slip; TC holds slip near a target that should drop when the tyre is carrying lateral load.
- A tyre circumference error of 1 % is a 1 % slip offset — update TC/logger rolling circumference after tyre changes.
- TC level = how much slip is allowed; TC cut/aggression = how hard it reduces torque. Tune them separately.
- Rising TC intervention per lap at the same level is a rear-tyre degradation indicator.
- Ignition retard is smooth but heats the exhaust; cylinder cut is fast but harsh.
- Diff locking and TC must be tuned together: inside-rear spin triggers engine-wide cuts.

## By car category

### GT3

Bosch/Magneti/Cosworth-based TC with typically 8–12 driver positions plus a separate cut/aggression dial on many cars. Drivers adjust per stint phase and per tyre condition.

### GT4

Road-car-derived TC with motorsport calibration; fewer positions, generally more intrusive.

### LMP2

TC available to the driver via wheel dials; wheel-speed-based with spec electronics.

### Hypercar (LMH/LMDh)

TC operates on a powertrain-torque level; LMH reference-speed estimation is harder above the AWD threshold. TC cut via MGU is fast and thermally free.

### Formula 4

No TC — the spec ECU does not allow it. Driver throttle modulation is the controller.

### FIA F3

No TC. Wheel-speed sensors are limited/spec; slip is analysed post-session from logged data.

### FIA F2

No TC with ≈ 620 hp turbo power — exit traction management (and tyre preservation) is a key driver skill in F2.

### IndyCar

No TC; slip analysed from data. Diff and driver are the traction controllers.

## References

- Segers, Analysis Techniques for Racecar Data Acquisition, 2nd ed. (SAE 2014): §4.2 Traction and Longitudinal Slip (slip ratio definitions, sensor configurations, Eq. 4.11–4.13)
- Segers, Analysis Techniques for Racecar Data Acquisition, 2nd ed. (SAE 2014): §4.3 TCS and Slip Ratios (Bosch ASR example, nominal slip vs actual slip, ignition angle during ASR, driven-wheel speed from rpm (Eq. 4.14))
- Milliken & Milliken, Race Car Vehicle Dynamics (SAE 1995): Ch. 2 Tire Behavior (longitudinal slip characteristics)
- Milliken & Milliken, Race Car Vehicle Dynamics (SAE 1995): Ch. 20 Driving and Braking
