# Slip ratio and longitudinal force

> Braking and traction force need the tyre to rotate slightly slower or faster than free rolling. Peak force on dry slicks comes at a slip ratio of roughly 6–12%, and wheel-speed data shows you where the driver and the electronics operate.

Area: Tyres · Starting level: L1 Foundation · Updated 2026-11-16
Source page: https://race-engineer-academy.vercel.app/lesson/tyres-slip-ratio-longitudinal-force

## L1 · Foundation

To brake or drive, the tread in the contact patch must be strained longitudinally, so the wheel turns slightly slower (braking) or faster (driving) than a free-rolling wheel. The **slip ratio** is

$$\kappa = \frac{\Omega R_e - V}{V}$$

where $\Omega$ is wheel angular speed, $R_e$ the effective rolling radius and $V$ the wheel-centre speed. $\kappa = -1$ is a locked wheel; positive values are wheelspin.

The $F_x$–$\kappa$ curve looks like the lateral curve: steep and linear for small slip, bending over to a peak, then falling. For dry slicks the peak is typically at **6–12% slip** (wets and road tyres higher, around 10–20%). The initial slope is very steep: at half the peak force the tyre is only at ~2–3% slip.

Practical meaning:
- A **locked** wheel ($\kappa = -1$) delivers sliding friction, maybe 70–85% of peak, and **no** lateral force. That is why a front lock-up means the car goes straight on, and why flat-spotting follows from the concentrated heat.
- **Wheelspin** past the peak heats the rear surface fast and gives up both traction and lateral grip on exit.
- ABS and traction control are slip-ratio controllers. They try to hold the tyre near its peak, just short of the drop-off.

Numbers to remember: dry slick peak $\mu_x$ is similar to or a little above $\mu_y$ (around 1.5–1.9). Peak braking decelerations are about 1.5 g (GT4, no big aero) to 2+ g (GT3 at speed), 3–4 g (F2/Super Formula at high speed), and above 4 g in top single-seaters, because downforce adds load.

## L2 · Practitioner

**Measuring slip in data.** You need a reference ground speed and each wheel speed.
- RWD car under traction: use the undriven fronts as reference. $\kappa_{rear} = v_{RL}/v_{ref} - 1$.
- Under braking all four wheels slip, so use GPS/optical speed or the fastest wheel and accept the error. ABS ECUs estimate a reference internally.
- In corners, correct for yaw: each wheel's ground speed differs by $\pm r\,t/2$ (yaw rate times half-track). Otherwise the inside wheel shows false "lock" and the outside false "spin".
- Calibrate **rolling radius** on a straight at constant throttle. Slicks grow with speed (centrifugal growth of a few mm at 250 km/h) and radius drops with load. Without calibration a 1% radius error looks like 1% slip.

**What to look for:**
- Front lock-ups: a wheel-speed dip below reference during braking. It often appears on the *inside front* during trail braking, because that wheel has lost load to both lateral and longitudinal transfer. Brake-bias rearward, softer front roll stiffness or a driver release-rate change are the usual tools.
- Exit wheelspin: rear wheel-speed excursions above reference with TC cutting. Repeated spikes of 10–15% mean the TC map or diff setting is wrong for the corner, or the rear is overheating.
- **TC slip targets** on GT cars are usually set by map in the region of the tyre's peak (a few % to ~10%), decreasing with lower grip. In the wet the lower surface μ moves the peak to lower slip (in the brush model $\kappa_{peak} \propto \mu F_z / C_\kappa$) and the post-peak drop is steeper, so the TC slip target comes down. The wet tyre's softer, siped tread lowers $C_\kappa$ and pushes the other way (that is why L1 quotes higher peak slip for wet tyres), so wet maps must be validated on wheel-speed data rather than scaled from the dry map.

**Brake balance and slip.** The ideal brake balance gives equal *normalised* slip front and rear at the limit. With high longitudinal load transfer (high CG, short wheelbase) the front locks easily on high-grip days. A car that locks rears on downshift has a different problem: engine braking plus diff coast lock adds negative torque to the rears.

## L3 · Advanced

**Physics of longitudinal slip.** In the brush model, longitudinal bristle deflection grows linearly through the patch with slope $\kappa$, giving longitudinal stiffness $C_\kappa = 2c_x a^2$, which is the same structure as cornering stiffness. Tread longitudinal stiffness is usually higher than lateral, and the carcass is much stiffer longitudinally (no sidewall bending mode). That is why peak slip ratios (~0.06–0.12) are small compared with the tangent of peak slip angle (~0.08–0.14).

**Effective rolling radius** $R_e$ lies between the loaded radius and the unloaded radius. For radials it is close to the unloaded radius because the steel belt is nearly inextensible. It varies with load, pressure and speed, so a slip estimate needs either a model of $R_e$ or calibration in each run.

**Heat.** Frictional power in the patch is approximately $F_x \cdot \kappa \cdot V$ (Segers Eq. 8.3 uses this term together with $F_y\,\alpha\,V$). Two consequences:
- Heavy braking from high speed puts big heat into the *front surface* for a short time. Brake-zone surface spikes of 20–40 °C on IR sensors are normal.
- Wheelspin at high speed (third/fourth gear exits) costs far more surface temperature than the same slip at low speed. This is the mechanism behind rear thermal degradation in long stints.

**Drivetrain influence.** The driven axle's slip is shared through the differential. An open diff limits axle force to twice the force of the less-loaded wheel. A locked diff forces equal wheel speeds. The inside wheel then over-runs its shorter path (positive slip, pushes forward) and the outside wheel under-runs its longer one (negative slip, drags). The resulting yaw moment resists rotation (entry/mid understeer). Under high throttle the loaded outside wheel takes most of the drive and the rear can saturate into exit oversteer. Preload, ramp angles and clutch plates tune how much slip difference is allowed. This is why diff settings change "slip-ratio balance" even though they do not change tyre curves.

**Brake balance derivation (no aero, rigid):** front load fraction under decel $D$ (in g) is $\Psi + D\,h/L$. Ideal front brake fraction = $\Psi + \mu\,h/L$. For $\Psi = 0.45$, $h/L = 0.17$, $\mu = 1.6$: about 0.72. Aero shifts this through the speed range, which is why GT drivers adjust bias between Le Mans' Mulsanne chicanes and the Porsche Curves.

## L4 · Expert

**Wheel-spin dynamics.** Each wheel obeys

$$I_w\,\dot{\Omega} = T_{drive} - T_{brake} - F_x R_e$$

Below the peak, extra slip raises $F_x$, which opposes the torque that created the slip, so the wheel is self-stabilising. Past the peak $\partial F_x/\partial\kappa < 0$: extra slip *reduces* the restoring force, the torque surplus grows and the wheel runs away to lock or to spin. No constant torque can hold a tyre just past its peak. ABS and TC therefore work as limit-cycle controllers that oscillate around the peak, typically a few to roughly 15 cycles per second, and their quality is how tightly that cycle hugs the peak without dropping into the post-peak region.

**Where the peak sits.** The brush model gives full sliding at $\kappa_{sl} \approx 3\mu F_z / C_\kappa$, so peak slip scales with $\mu F_z / C_\kappa$. Lower surface μ (wet, dust, cold tyres) moves the peak to lower slip. A softer, siped wet tread lowers $C_\kappa$ and moves it higher. More load raises it only weakly, because $C_\kappa$ also rises with load. Slip targets therefore cannot be copied between compounds or conditions.

**Longitudinal relaxation.** $F_x$ also builds over a relaxation length, usually shorter than the lateral one. The lag $\sigma_\kappa / V$ is negligible at 200 km/h, but at hairpin speeds it becomes a meaningful fraction of an ABS cycle. Together with wheel-speed sensor resolution at low speed, this is one reason ABS control gets coarser in slow corners.

**Rolling-radius error budget.** Centrifugal growth (a few mm at 250 km/h, around 1% of radius), load and pressure (the effective radius moves with deflection), tread wear over a stint and temperature each shift $R_e$ by up to about 1%. That is the same size as the linear-region slip you are trying to read. Production-grade slip estimation uses a calibrated $R_e$ model (load, speed) or adapts it online on straights at constant throttle.

**Locked axle, quantitatively.** In a corner of radius $R$ the inside and outside path speeds differ by $t/R$. A locked axle runs at the mean speed, so each tyre sees roughly $\pm t/(2R)$ of slip: a 1.6 m track in a 40 m corner gives about ±2%. With $C_\kappa$ around 120 kN per unit slip, the linear estimate is ±2.4 kN per tyre and about 2.4 × 1.6 ≈ 3.8 kN·m of yaw moment *out* of the turn. That is the same order as a substantial balance change. In reality the lightly loaded inside tyre saturates first and the friction ellipse caps the outside tyre's longitudinal force at high lateral utilisation, so the true moment is smaller and strongly non-linear. That non-linearity is what preload, ramp angles and plate count are tuned against (RCVD Ch. 20).

## Key formulas

**Slip ratio (SAE-style)**

$$\kappa = \dfrac{\Omega R_e - V}{V}$$

- \Omega: wheel angular speed (rad/s)
- R_e: effective rolling radius (m)
- V: wheel-centre ground speed (m/s)

**Wheel slip from data (RWD traction)**

$$\kappa_{RL} \approx \dfrac{v_{RL}}{v_{ref} \mp r\,t_r/2} - 1$$

- v_RL: rear-left wheel speed
- v_ref: reference (front/GPS) speed
- r: yaw rate (rad/s)
- t_r: rear track (m)

**Locked-axle slip in a corner**

$$\kappa_{in,out} \approx \pm\dfrac{t}{2R}$$

- t: track width (m)
- R: corner radius at the axle (m)

**Ideal front brake fraction (rigid, no aero)**

$$\phi_f = \Psi + \mu\,\dfrac{h}{L}$$

- \Psi: static front weight fraction
- \mu: achievable deceleration (g)
- h: CG height (m)
- L: wheelbase (m)

## Trackside takeaways

- Dry slicks peak at roughly 6–12% slip; a locked wheel gives perhaps 70–85% of peak μ and no lateral force.
- Correct wheel speeds for yaw (±r·t/2) and rolling radius before calling anything lock-up or wheelspin.
- Inside-front lock during trail braking is a load-transfer symptom as much as a bias problem.
- Heat from longitudinal slip scales with F·κ·V, so high-speed wheelspin is the expensive kind.
- ABS and TC are slip-ratio controllers. Their targets should follow the tyre’s peak slip, which moves with compound, temperature and wet/dry.
- Past the peak, ∂Fx/∂κ < 0 makes the wheel unstable, so ABS/TC must cycle around the peak rather than hold it.

## By car category

### One-make Cup

The Porsche 992 Cup runs without ABS or TC, so wheel-speed traces show the driver’s own slip control. Inside-front lock-ups in trail braking and rear lock on aggressive downshifts are common coaching items. Ferrari Challenge and Super Trofeo cars run ABS, so check whether the ABS is actually cycling in the corner entry.

### GT3

ABS and TC maps are the driver’s slip tools. Track the TC intervention count per corner. Persistent intervention on one exit is a setup or diff question rather than a driver one.

### Formula 4

No driver aids. Exit wheelspin on the small-rubbered F4 shows clearly in rear vs front wheel speeds. It is a key metric when comparing drivers.

### FIA F2

Heavy braking from 300 km/h with high downforce puts big, short heat spikes into the front surface. Lock-ups flat-spot the Pirelli quickly, and a flat spot causes vibration that can end the stint.

### TCR

Front-drive: all traction slip happens on the steered axle, and the inside front spins first because it is unloaded. LSD settings and the TC map trade exit drive against front-tyre overheating.

## References

- Milliken & Milliken, Race Car Vehicle Dynamics (SAE 1995): Ch. 2 Tire Behavior (Longitudinal slip and force)
- Milliken & Milliken, Race Car Vehicle Dynamics (SAE 1995): Ch. 20 Driving and Braking (Locked/limited-slip differential yaw moment, brake proportioning)
- Segers, Analysis Techniques for Racecar Data Acquisition, 2nd ed. (SAE 2014): §8.4 Where Does Tire Temperature Come From? (Fx·SR·V heat term)
