# Diff effect on yaw: entry stability, mid-corner understeer, exit traction

> A left/right drive or braking force difference produces a yaw moment ΔF·t/2. Its sign depends on which wheel is faster, so the same diff stabilises entry, resists mid-corner rotation and can either push or snap on exit.

Area: Differential · Starting level: L2 Practitioner · Updated 2026-11-16
Source page: https://race-engineer-academy.vercel.app/lesson/diff-yaw-moment-corner-phases

## L1 · Foundation

If the two rear tyres produce different longitudinal forces, the car feels a **yaw moment** about its vertical axis:

$$M_z = (F_{x,out} - F_{x,in})\, \frac{t}{2}$$

Example (GT3, $r$ = 0.34 m, rear track 1.6 m, locking $T_{lock} = 120 + 0.38\,T_{in}$ Nm), with matched conditions:
- **Exit at 1.0 g, 2000 Nm axle torque:** $T_{lock}$ ≈ 120 + 0.38 × 2000 ≈ 880 Nm → ΔF ≈ 2.6 kN → $M_z$ ≈ 2.1 kNm. The rear axle lateral force (≈ 7 kN acting ≈ 1.2 m behind the CG) gives ≈ 8 kNm. The diff moment is **≈ 25 % of that** — a major balance lever.
- **Mid-corner at 1.5 g on preload alone** (input ≈ 200 Nm): $T_{lock}$ ≈ 200 Nm → $M_z$ ≈ 0.45 kNm against ≈ 12 kNm, only **≈ 3–4 %**, yet still enough for the driver to feel a lazy apex.

Using the rule "locking torque goes to the slower wheel", walk through a corner:

**Entry (braking, coast, engine braking):** outside wheel is faster → coast locking puts **more braking** on the outside wheel → moment **resisting rotation** → stability, less turn-in. More coast locking/preload = more stable, lazier entry. Too little = rear nervous on lift-off and downshifts.

**Mid-corner (neutral throttle):** outside still faster → preload adds drive to the inside / braking to the outside → **resists rotation** → mid-corner understeer. Slow corners feel it most: the fixed diff moment is a larger share of the low tyre forces (little downforce), and a large $t/R$ makes an axle that stays stuck scrub heavily. (A slipping clutch transfers the same torque whatever the size of the speed difference.) Less preload = better rotation.

**Exit (power):**
1. Early, while no wheel spins: outside faster → more torque to inside → **resists rotation** — power understeer.
2. As throttle rises the inside wheel (unloaded) starts to spin and becomes faster → torque flips to the **outside** → moment **promotes rotation** → power oversteer, potentially sudden.
More drive locking = more traction and more phase-1 understeer, with a sharper transition to phase 2.

This is why diffs are explained badly by "more lock = more understeer". It is true on entry and mid-corner, and in early exit — but at high throttle with an unloaded inside wheel, more lock means more **oversteer** (and more traction).

## L2 · Practitioner

**Driver language → diff phase mapping:**

| Driver says | Phase / mechanism | Diff direction |
|---|---|---|
| "Rear is nervous when I lift / on downshifts" | coast, too little locking | more coast lock (lower coast ramp angle) or more preload; check blip |
| "Car won't turn in on the brakes" | coast, too much locking | less coast lock / preload |
| "Lazy at the apex in hairpins" | mid, preload resisting yaw | less preload |
| "Inside rear lights up, poor drive" | exit, too little drive lock | more drive lock (lower drive ramp angle, more faces) |
| "Pushes on power, then snaps" | exit, high drive lock with state flip | first reduce drive lock; then test preload in BOTH directions (higher preload widens the stuck band, which can delay and sharpen the flip; lower preload + ramp can make it earlier and progressive); check rear roll stiffness and TC slip target |
| "Steering heavy, car fights me on exit" (FWD) | front drive lock, torque steer | less preload/drive lock |

**Interactions you must account for:**
- **Rear anti-roll bar / roll stiffness**: a stiffer rear increases inside-rear unloading → more inside spin → more reliance on diff locking. Diff and rear roll stiffness are solved together.
- **TC**: an open-ish diff with inside spin makes TC cut engine torque for both wheels; more locking reduces TC activity (see TC lesson).
- **Brake balance & engine braking map**: entry stability is the sum of brake balance, engine braking and coast lock — change one at a time and note which.
- **Gear**: wheel torque, hence locking through ramps, is larger in low gears. A diff that is fine in 4th-gear corners can be too locked in 2nd-gear hairpins.
- **Corner speed**: slow corners have low aero load (the diff moment is a larger share of the yaw balance), high wheel torque in low gears (ramp locking) and a large $t/R$ (heavy scrub if the axle stays stuck), so diff effects dominate there; in fast corners aero and springs dominate.

**Misconception check:** "more locking always means more stability". On coast, yes. On power with an unloaded inside wheel, more locking puts more force on the outside wheel — a pro-yaw moment and earlier rear saturation. Highly locked diffs make exits *faster* but *more abrupt* and demand a more precise throttle foot.

## L3 · Advanced

**Yaw moment with tyre-limited forces.** Let the diff lock with capacity $T_{lock}$ and the wheels rotate at different speeds. The axle drive torque splits as $T_{slow} = (T_{in} + T_{lock})/2$, $T_{fast} = (T_{in} - T_{lock})/2$, giving

$$M_z = \pm \frac{T_{lock}}{r}\, \frac{t}{2}$$

with the sign **negative (anti-yaw) when the outside wheel is faster** and **positive (pro-yaw) when the inside wheel is faster**. When the diff does *not* slip (locked state), $\Delta\omega = 0$ and the split is set by the tyres' longitudinal stiffness: the outside tyre is forced to negative relative slip, the inside to positive, generating an anti-yaw moment ≈ $C_x\,(t\dot\psi/V)\,t/2$ per tyre pair (with $C_x$ the longitudinal slip stiffness), limited by $T_{lock}$. At small radius and low speed ($t/R$ large) this moment saturates at the lock capacity.

**State transitions are the real handling problem.** The diff state can switch between: slipping with outside faster → stuck → slipping with inside faster. Over the whole transition the moment travels from $-T_{lock}\,t/(2r)$ to $+T_{lock}\,t/(2r)$, a swing of $T_{lock}\,t/r$. Inside the stuck band it changes continuously as the tyre forces evolve. The final breakaway, sharpened by the static-to-dynamic friction drop and by the inside tyre passing its force peak, can happen within tens of ms. The driver experiences a change of yaw moment of several kNm without having changed steering — the "snap". A high preload makes the locked band wider (the diff stays locked longer), which can make the transition *later and harder*; a lower preload with stronger ramps can make it earlier and progressive. There is no universal rule — test both directions and read the data.

**Combined-slip view.** The outside rear has more vertical load, so giving it more drive torque (inside spinning) uses capacity it has — traction rises — but the inside rear, already spinning, has lost lateral capacity. Net rear lateral force can drop even as drive rises: the classic exit oversteer of a highly locked diff on a car with stiff rear roll stiffness. Softer rear roll stiffness keeps the inside loaded, delays the spin, and lets you run more locking — again, diff and ARB are one system.

**Stability analysis.** Linearising, a stuck axle gives each tyre a slip of $\pm t\dot\psi/(2V)$, so $M_z = -C_x\,t^2\,\dot\psi/(2V)$ and a yaw-damping derivative $N_r \approx -C_x\,t^2/(2V)$ (with $C_x$ the longitudinal slip stiffness **per tyre**, N per unit slip; equivalently $-C_{x,axle}\,t^2/(4V)$ with the axle total). A viscous (speed-sensing) diff adds a similar, smaller term. It increases yaw damping (more stable, less responsive) — the transient analogue of the steady-state understeer contribution (RCVD Ch. 5–6 framework). That's why drivers describe high coast/preload as "stable but lazy".

## Key formulas

**Yaw moment from left/right force difference**

$$M_z = (F_{x,out} - F_{x,in})\, \frac{t}{2}$$


**Yaw moment from slipping LSD**

$$M_z = \pm \frac{T_{lock}}{r}\,\frac{t}{2}$$


**Torque split when slipping**

$$T_{slow} = \tfrac12 (T_{in} + T_{lock}), \quad T_{fast} = \tfrac12 (T_{in} - T_{lock})$$


## Trackside takeaways

- Diff yaw moment = ΔF·t/2 — up to ~25 % of the rear axle’s lateral moment on low-gear exits, only a few % mid-corner on preload, yet still felt by the driver.
- Outside wheel faster (entry, mid, early exit) → locking resists rotation.
- Inside wheel faster (inside spin on exit) → locking promotes rotation: more lock = more oversteer there.
- "Pushes then snaps" = diff state flip; tune the transition, not just the amount.
- Diff, rear roll stiffness, TC, engine braking and brake balance interact — change one at a time.
- Diff effects dominate in slow corners (large ΔV/V, low aero).

## By car category

### GT3

Slow-corner entry stability vs rotation is often tuned with preload; drivers sometimes compensate on the wheel with engine-braking maps where available.

### TCR

Front LSD: locking on power pulls the car straight (power understeer) and adds torque steer; on lift the front diff affects turn-in. Rear yaw tools are limited, so the front diff is central.

### FIA F3

Low-downforce slow corners make diff preload very visible in mid-corner rotation; limited adjustments make each step significant.

### Hypercar (LMH/LMDh)

Rear diff behaviour interacts with hybrid torque distribution and BBW regen on the driven axle; entry stability must be checked with regen active.

### IndyCar

On street circuits the diff dominates slow-corner rotation and exit traction; road/street settings differ substantially from ovals.

## References

- Milliken & Milliken, Race Car Vehicle Dynamics (SAE 1995): Ch. 20 Driving and Braking (differential yaw moments and their effect on steady-state balance)
- Milliken & Milliken, Race Car Vehicle Dynamics (SAE 1995): Ch. 5 Simplified Steady-State Stability and Control
- Milliken & Milliken, Race Car Vehicle Dynamics (SAE 1995): Ch. 6 Simplified Transient Stability and Control (yaw damping)
