# Caster, kingpin inclination, scrub radius & mechanical trail

> The steering axis geometry decides steering weight, self-centring, camber change with steer and the diagonal weight jacking that happens every time the driver turns the wheel.

Area: Suspension Geometry & Kinematics · Starting level: L1 Foundation · Updated 2026-11-16
Source page: https://race-engineer-academy.vercel.app/lesson/sg-steering-axis

## L1 · Foundation

**Steering axis.** The front wheel steers about a line through the upper and lower ball joints (or strut top and lower joint) – the **kingpin axis**. Its orientation is described by two angles and two offsets:

- **Caster angle** – side-view tilt; top rearward = positive. Race cars ~5–10°.
- **Kingpin inclination (KPI / SAI)** – front-view tilt; top inward. Race cars ~6–14° (larger on GT cars with big brakes inside the wheel).
- **Mechanical trail** – side-view distance on the ground from where the axis hits the ground to the contact-patch centre (patch behind = positive). Typical 15–40 mm.
- **Scrub radius** – front-view distance on the ground between axis intercept and contact-patch centre (patch outboard = positive). Race cars often +10 to +40 mm, because large brakes push the ball joints inboard.

**Mechanical trail** is the lever arm on which the tyre's lateral force acts around the steering axis. It produces self-centring and part of the steering weight:

$$t_m \approx R\,\tan(\text{caster}) - e$$

where $R$ is tyre loaded radius and $e$ any hub (caster) offset. With R = 330 mm and 7° caster: about 40 mm before offset.

**Caster also changes camber with steering.** When the driver turns, the **outside** front wheel gains **negative camber** (roughly caster × steer angle in radians) and the inside wheel positive. With 8° caster and 15° road-wheel steer at a hairpin: about −2° extra on the outside tyre. That is why high caster helps slow-corner front grip.

**Scrub radius** is the lever arm for longitudinal forces acting at the contact patch: braking (outboard brakes) or bumps on one front wheel produce a steering torque. Large positive scrub = kickback over kerbs and tugging under braking on uneven grip. Torque steer on front-drive cars is a different lever: the drive force from a driveshaft acts at the **wheel centre** (the torque is reacted inboard), so what matters is the kingpin offset at hub height, plus unequal driveshaft angles – not the ground scrub radius.

## L2 · Practitioner

**What changes when you change caster (common GT/prototype adjustment via shims or eccentrics):**
- **Steering weight** up with more caster (more mechanical trail).
- **Camber in steer:** more negative camber on the loaded outside front at large steering angles → better slow-corner grip; negligible effect in fast corners where steer angles are 1–3°.
- **Diagonal weight jacking:** with positive caster, steering lowers the outside-front corner and raises the inside-front corner of the chassis. That loads the **inside-front / outside-rear** diagonal. Effect: less front lateral load transfer and slightly more rear – can help front grip in slow corners, but unloads the inside rear (traction on exit). The effect grows with steer angle, so it matters in hairpins.
- **Self-centring and straight-line stability:** better with more caster; also the steering returns quicker after oversteer corrections.
- **Feel at the limit:** more mechanical trail means the pneumatic-trail drop at the limit is a smaller *fraction* of total steering torque – the driver feels the front saturation less (see the steering-feel lesson). Too much caster can mask understeer.

**KPI** gives camber change with steer (positive on both wheels – helpful on neither), and lifts the front of the car as the wheels steer, which produces a gravity self-centring torque at low speed. Designers use it mainly to reduce scrub radius within the wheel package.

**Scrub radius trackside:** wheel offset/spacer changes alter it directly (a 10 mm spacer adds 10 mm scrub radius and steering kickback). Teams often discover this when running different wheel suppliers or spacers to adjust track.

**Measuring:** caster/KPI with a caster-camber gauge using the 20° swing method on turn plates (read camber at ±20° steer; caster ∝ difference). Mechanical trail and scrub from CAD or from pickup coordinates; on a car with unknown geometry measure ball-joint positions with a FARO arm or careful plumb-line work.

**Typical consequence of a front caster change:** +1° caster on a GT3 typically gives noticeably heavier steering and better hairpin rotation; the driver may report "more front" in slow corners and "nothing changed" in fast ones – which is exactly what the physics predicts.

## L3 · Advanced

**Steer kinematics derivation.** Rotate the wheel about an axis inclined by caster $c$ (rearward at top) and KPI $\lambda$. For small steer $\delta$, the camber change of a wheel is approximately:

$$\Delta\gamma \approx -\delta\,\sin c + (1 - \cos\delta)\,\sin\lambda$$

(sign: negative = top towards car centre, for the outside wheel with $\delta$ positive towards the turn). Caster dominates and is first-order in $\delta$; KPI is second-order and adds positive camber to both wheels.

**Vertical motion of the wheel centre.** Using Rodrigues' rotation, the wheel centre (lateral offset $d$ outboard of the axis at hub height) moves vertically by $\approx -d\,\sin c\,\sin\delta$ (first-order, caster) plus a second-order KPI term $\approx d\,\sin\lambda\,(1-\cos\delta)$. The first-order caster term is opposite on left and right: the inside wheel moves down relative to the chassis (lifting that corner), the outside wheel up (lowering it). This is the **caster jacking** that re-distributes corner weights diagonally. The KPI term raises the front symmetrically → self-centring under gravity.

**Steering torque about the axis (simplified, per wheel):**

$$M_{kp} = F_y\,(t_m + t_p)\cos c + F_x\,r_s + F_z\,(\text{jacking terms})$$

where $t_p$ is pneumatic trail and $r_s$ scrub radius. With equal left/right braking forces the scrub terms cancel; with split-μ or ABS cycling on one wheel they do not → kickback and steering tug. Negative scrub (patch inboard of axis) is used on road cars to make a split-μ brake force steer the car *towards* stability; race cars accept positive scrub for brake packaging.

**Dynamic caster.** Under braking the body pitches nose-down by $\theta$, which reduces caster relative to the ground by $\theta$ (≈0.5–1° on a GT3 at 1.5 g). Anti-dive geometry rotates the upright the other way: with the side-view IC behind and above the front wheel, the upright rotates nose-up about it as the wheel moves into bump, adding caster relative to the body by ≈ $dz/SVSA$ (e.g. +0.4° for 15 mm of dive with a 2 m side-view swing arm). So anti-dive *partly offsets* the pitch-induced loss. A car with little anti-dive and soft front springs loses the most caster on entry, which lightens the steering exactly when the driver is trail-braking – account for it when the driver says "steering goes vague on entry".

## L4 · Expert

**Jacking moments as a balance tool – limits.** The diagonal weight change from caster jacking is proportional to steering angle and to the front suspension's *wheel rate* (it is a displacement imposed at the wheel), then distributed by the roll stiffness distribution. On cars with very stiff front ARBs, a few mm of caster-induced corner height change can move hundreds of newtons diagonally in a hairpin. This partly explains why the same caster step feels larger on a stiff-front setup. A full model needs steering-angle-dependent corner height in the kinematic tables, not a static caster number.

**Pneumatic vs mechanical trail ratio.** At small slip, pneumatic trail on a race slick is roughly 1/6 of contact length – ~20–35 mm. If mechanical trail is 40 mm, roughly 40 % of steering torque comes from the tyre's pneumatic trail at low slip; that fraction collapses near the peak. Designers choose caster/trail to give a target "steering torque drop-off" at the limit: too little mechanical trail gives very light, nervous steering that falls off a cliff; too much gives heavy steering where the drop-off is hidden. Power steering gain shapes the same thing at the driver's hands.

## Key formulas

**Mechanical trail**

$$t_m \approx R_l\,\tan c - e$$

- R_l: loaded tyre radius (mm)
- c: caster angle
- e: hub/caster offset (mm)

**Camber change with steer**

$$\Delta\gamma \approx -\delta\sin c + (1-\cos\delta)\sin\lambda$$

- δ: road-wheel steer angle (rad)
- λ: kingpin inclination

**Steering-axis moment (per wheel)**

$$M_{kp} \approx F_y\,(t_m + t_p) + F_x\,r_s$$

- t_p: pneumatic trail
- r_s: scrub radius
- F_x: longitudinal force at the contact patch

## Trackside takeaways

- Caster → mechanical trail (steering weight) and negative camber on the outside wheel with steer.
- Caster jacking loads the inside-front/outside-rear diagonal in proportion to steer angle.
- KPI lifts the car symmetrically when steering (self-centring) and adds positive camber.
- Positive scrub radius = kickback and brake tug; spacers change it.
- Too much mechanical trail can mask the steering-torque drop that tells the driver the front is saturating.

## By car category

### GT3

Caster adjustment via shims/eccentrics is typical; steps of ~0.5–1° are noticeable. Large brakes force big KPI and positive scrub; wheel/spacer choices move the scrub radius.

### TCR

MacPherson + FWD: scrub radius matters for brake tug and kerb kickback, while torque steer is set by the kingpin offset at hub height and by unequal driveshaft angles; caster is limited by the strut geometry, but caster-induced camber is one of the few ways to get the outside front tyre working in slow corners.

### FIA F3

Spec geometry with limited caster options; steering is unassisted, so caster choice directly sets driver effort and fatigue over a race.

### IndyCar

Steering is unassisted on the IR-18, so caster and mechanical trail set both the feel and the driver effort over a race distance; oval packages use different caster left/right.

### LMP2

Power steering hides steering weight, so caster is chosen more for camber-in-steer and feel modulation than for effort.

## References

- Milliken & Milliken, Race Car Vehicle Dynamics (SAE 1995): Ch. 19 Steering Systems (kingpin geometry, trail, scrub, jacking)
- Milliken & Milliken, Race Car Vehicle Dynamics (SAE 1995): Ch. 17 Suspension Geometry
