# The g-g diagram and using the friction envelope

> Plotting lateral against longitudinal acceleration shows the car’s performance envelope and how much of it the driver uses. It is speed dependent (aero), asymmetric (power, drag) and best analysed in speed bands and with gated grip statistics.

Area: Weight Transfer & Balance · Starting level: L1 Foundation · Updated 2026-11-16
Source page: https://race-engineer-academy.vercel.app/lesson/wt-gg-diagram-friction-envelope

## L1 · Foundation

The **g-g diagram** (Milliken; "traction circle" in Segers §7.2) is an X–Y plot of lateral vs longitudinal acceleration for every sample of a lap. It is the vehicle-level counterpart of the tyre friction ellipse.

Reading it:
- **Bottom** (negative $a_x$): braking. Highest values at high speed because downforce and drag help.
- **Sides**: cornering. Left and right lobes differ if the track has more fast corners one way.
- **Top**: acceleration. It is **flattened** because engine power, not grip, limits it except in low gears.
- **Diagonals**: combined braking-turning (entry) and turning-accelerating (exit).

The envelope is an **ellipse-like shape**, not a circle: braking capability exceeds traction capability, and drag adds deceleration at high speed (giving a "tail" at the bottom of the plot after long straights).

**Typical peak values:**
| Car | Peak lateral | Peak braking |
|---|---|---|
| GT4 | ~1.3–1.5 g | ~1.3–1.5 g |
| GT3 | ~1.6–2.0 g | ~1.6–2.0 g |
| F3 | ~2.5–3 g | ~2.5–3 g |
| LMP2 / Hypercar | ~2.5–3 g | ~2.5–3 g |
| F2 / Super Formula | ~3–4 g | ~3.5–4+ g |

(Ranges overlap; they depend on the track, aero configuration and BoP.)

A driver who fills the envelope edge through the transitions (trail braking, early throttle while unwinding) is using the car. A "cross" shaped plot, with braking and cornering done separately, leaves performance unused.

## L2 · Practitioner

**Speed dependence.** Downforce scales with $V^2$, so the envelope grows with speed (Segers §7.3: separate g-g plots in speed bands such as <130, 130–180, 180–230, 230–280 km/h). Lateral G vs speed and longitudinal G vs speed plots show the same thing as clouds whose upper edge rises with speed. Always compare drivers or setups **within a speed band**.

**Corner speed from the envelope** (constant-radius, steady state). Without aero, $V = \sqrt{\mu g R}$. With aero,

$$V^2 = \frac{\mu g R}{1 - \dfrac{\mu\,\rho\,C_LA\,R}{2m}}$$

**Example:** GT3, $m = 1350$ kg, $\mu = 1.6$, $C_LA = 1.5$ m², $R = 100$ m: denominator $= 1 - 1.6\times1.225\times1.5\times100/2700 = 0.891$, so $V = 42.0$ m/s (151 km/h) vs 39.6 m/s (143 km/h) without aero. Larger radii gain more.

**Grip-factor statistics (Segers §8.1).** Use the combined acceleration $\sqrt{a_x^2 + a_y^2}$ gated by situation to build lap statistics:
- overall: combined > 1 G (grip-limited points only);
- cornering: $|a_y| > 0.5$ G;
- braking: $a_x < -1$ G;
- traction: $a_x > 0$ and $|a_y| > 0.5$ G;
- aero: $|a_y| > 1$ G and $V > 120$ km/h.

(Thresholds depend on the car.) Radar charts of these five factors compare days, setups or tyre sets at a glance. Segers' wind example showed aero and braking factors falling the day the wind direction flipped.

**Quadrant density** tells you where the car spends its time. A track dominated by right-hand traction zones rewards setup changes in that quadrant (Segers' Adria example).

## L3 · Advanced

**From g-g to g-g-V (RCVD Ch. 9).** Milliken's full performance envelope is a 3-D surface: for each speed, the boundary of achievable $(a_x, a_y)$. Its shape is set by:
1. **Tyre capacity**, scaled by total vertical load ($mg$ + downforce) and reduced by load sensitivity and load-transfer losses.
2. **Power**: $a_{x,max} = P_w/(mV) - D/m$ above traction-limited speeds. GT3 at 200 km/h: ~360 kW at the wheels, $m = 1350$ kg, drag ~1.5 kN, giving ≈ 0.37 g.
3. **Balance/stability**: the car may have tyre capacity left but be unable to use it because of yaw-moment limits (entry oversteer, exit understeer). The envelope boundary is then a **control** limit, not a grip limit. The MRA moment method (RCVD Ch. 8) separates the two.
4. **Drag**: adds to braking; at 300 km/h, drag alone is worth roughly 0.4 g of deceleration on an LMP2-class car and approaching 1 g on the highest-downforce single-seaters, with no brakes applied.

**Driver-utilisation metrics.** For each sample, compute utilisation $u = a_{comb} / a_{env}(V, \theta)$, where $\theta = \arctan(a_x/a_y)$ and $a_{env}$ is the envelope radius at that speed and direction (from the best laps, or a QSS lap sim). Mean $u$ by corner phase shows where a driver leaves margin. Engineers often find drivers at 97–99% in mid-corner but 85–90% in transitions.

**Pitfalls in building the plot:**
- Use **CG-referenced** and attitude-corrected accelerations. Pitch/roll tilt the accelerometer, mixing in a $g\sin\theta$ component (1° of pitch ≈ 0.017 g). Banking and grade add further terms (see the banking lesson).
- Filter consistently: kerb strikes create spikes that inflate peak values. Use percentile envelopes (e.g. the 98th percentile per angular bin) rather than maxima.

## L4 · Expert

**Advanced use.**
- **Envelope correlation with lap simulation.** A QSS or optimal-control lap sim produces a g-g-V envelope. Overlaying measured data per speed band shows whether the gap to the sim is in grip (envelope too small everywhere), in power (top flattened lower than predicted), or in transitions (data inside the predicted shape). Each points to a different discipline: tyre model, engine/drag, or vehicle balance and driver.
- **Asymmetry analysis.** Left/right lobe differences beyond what track layout explains point to asymmetric setup (cross-weight, ride heights, camber), damaged aero, or a driver who is less committed one way. Compare lobes in matched speed bands and corner types.
- **Time-varying envelope.** Fuel load (mass) and tyre degradation shrink the envelope through a stint. Grip-factor run charts are the 1-D projection of this. Rain shrinks it dramatically and changes its shape (the braking lobe shrinks more than the lateral lobe on standing water).
- **Limits of the concept.** The g-g plot hides *where* on track and *why*. Two laps with identical envelopes can differ by seconds through line choice (radius, apex speed) and time spent near the edge. Always combine with distance-based traces (speed, combined G, throttle/brake overlap) and time-delta channels.

## Key formulas

**Combined acceleration**

$$a_{comb} = \sqrt{a_x^2 + a_y^2}$$


**Corner speed without aero**

$$V = \sqrt{\mu\,g\,R}$$


**Corner speed with downforce**

$$V^2 = \dfrac{\mu g R}{1 - \dfrac{\mu\,\rho\,C_L A\,R}{2m}}$$

- \rho: air density (kg/m³)
- C_L A: downforce coefficient × area (m²)
- R: corner radius (m)
- m: mass (kg)

**Power-limited acceleration**

$$a_{x,max} = \dfrac{P_w}{m V} - \dfrac{\tfrac{1}{2}\rho\,C_D A\,V^2}{m}$$

- P_w: power at the wheels (W)
- C_D A: drag area (m²)

## Trackside takeaways

- The g-g plot is the car’s envelope and the driver’s utilisation in one picture.
- It is speed dependent: compare in speed bands (or as g-g-V).
- Acceleration is power limited above low gears; braking is helped by drag.
- Gated grip factors (overall, cornering, braking, traction, aero) make great run-chart and radar-chart statistics.
- The boundary can be a control (stability) limit, not a grip limit. Check balance before blaming the driver.

## By car category

### GT3

Envelope roughly 1.6–2.0 g with modest aero growth with speed. Traction quadrants matter at low speed. BoP changes (power, weight, aero) show up directly as changes in the power-limited top and the overall radius.

### GT4

Low downforce: the envelope is nearly speed independent, so it is easier to compare corners directly. Peak values of ~1.3–1.5 g make driver transitions (filling the diagonals) the main differentiator.

### FIA F2

Large aero growth with speed: speed-banded g-g plots are mandatory. Braking peaks above 3.5 g at high speed but well under that in slow hairpins.

### LMP2

Strong speed dependence and drag-assisted braking at the end of long straights (Le Mans). Aero grip factor is the key statistic for comparing low-drag and high-downforce kits.

### Super Formula

Very high downforce for the mass gives some of the largest envelopes outside F1. The difference between new and used tyres is clearly visible in the envelope radius.

### One-make Cup

Identical cars and tyres make g-g overlays between drivers very informative. The gaps are almost always in the transitions (entry diagonal, exit diagonal), not in peaks.

## References

- Milliken & Milliken, Race Car Vehicle Dynamics (SAE 1995): Ch. 9 "g-g" Diagram
- Milliken & Milliken, Race Car Vehicle Dynamics (SAE 1995): Ch. 8 Force-Moment Analysis (Control vs grip limits)
- Segers, Analysis Techniques for Racecar Data Acquisition, 2nd ed. (SAE 2014): §7.2 The Traction Circle
- Segers, Analysis Techniques for Racecar Data Acquisition, 2nd ed. (SAE 2014): §7.3 Effects of Speed
- Segers, Analysis Techniques for Racecar Data Acquisition, 2nd ed. (SAE 2014): §8.1 Estimating Grip Levels (Gated grip factors)
