# LLTD, the "magic number" and mechanical balance

> The front share of total lateral load transfer (LLTD) is the main mechanical balance parameter. A commonly quoted starting point is front LLTD about 5 points above static front weight, but the right number depends on tyres, drivetrain, aero and driver.

Area: Weight Transfer & Balance · Starting level: L2 Practitioner · Updated 2026-11-16
Source page: https://race-engineer-academy.vercel.app/lesson/wt-lltd-magic-number

## L1 · Foundation

**Lateral load transfer distribution (LLTD)**, also called TLLTD (total LLTD), is the fraction of the total lateral load transfer carried by the **front** axle:

$$\text{LLTD}_f = \frac{\Delta F_{z,f}\,t_f}{\Delta F_{z,f}\,t_f + \Delta F_{z,r}\,t_r} \approx \frac{\Delta F_{z,f}}{\Delta F_{z,f} + \Delta F_{z,r}} \;(t_f \approx t_r)$$

Because of tyre load sensitivity, the axle that carries the larger share of load transfer loses relatively more grip:
- **Higher front LLTD** → more understeer (steady state).
- **Lower front LLTD** → more oversteer.

Compare LLTD with the **static front weight fraction**. If the front carries 45% of the weight and 45% of the load transfer, both axles are equally "punished", so balance is set by other things (tyre sizes, aero, geometry). The **"magic number"** associated with Milliken's work is a starting point of roughly **front LLTD ≈ static front weight % + 5%**, which builds in mild understeer for stability, especially in a RWD car that also needs rear capacity for traction.

**Main levers:**
| Lever | LLTD effect |
|---|---|
| Front ARB stiffer / rear ARB softer | LLTD up (more understeer) |
| Front springs stiffer | LLTD up (and changes ride, aero platform) |
| Front roll centre higher | LLTD up (geometric) |
| Rear track wider | At constant rear roll stiffness (springs/bars re-rated): less ΔF per rear tyre, which helps rear grip; moment-based LLTD unchanged. With the same wheel rates $K_{\phi r}$ rises ∝ $t_r^2$, the rear takes more of the roll moment and front LLTD falls slightly |

ARBs are the preferred tool: they change roll stiffness with little effect on heave/pitch stiffness, so the aero platform is largely unchanged.

## L2 · Practitioner

**How big is a change?** With the tyre load-sensitivity model $\mu = 1.8 - 5\times10^{-5}F_z$, an axle pair with transfer $\Delta F$ loses $10^{-4}\,\Delta F^2$ of lateral capacity. Take the GT3 example (1300 kg, 45% front, no aero) at 1.3 g: total transfer ≈ 4.5 kN. Moving front LLTD from 49% to 54% raises front transfer from 2216 to 2442 N (the inside front still keeps ~430 N of its 2869 N static load) and lowers rear transfer from 2306 to 2080 N. The front pair loses $10^{-4}(2442^2 - 2216^2) \approx 105$ N of capacity and the rear gains ~100 N: a ~2% relative balance swing, comparable to a meaningful aero balance change, and a driver will feel it. Do not run this calculation at 1.6 g without downforce: at 54% the inside front would need a negative load (it lifts), and the quadratic model no longer applies (see wheel-lift saturation, level 4).

**Practical procedure:**
1. Compute LLTD from your setup sheet (wheel rates, ARB rates, motion ratios, roll centres, masses). Calculators or a setup spreadsheet linked to the data system (Segers ch2) make this routine.
2. Log **roll gradient** per axle from damper pots (converted to wheel travel with the motion ratios). Use it to check consistency between runs and to spot packer or bar-link problems. Do **not** treat the front/rear roll-angle ratio as $q$: on a torsionally stiff chassis both axles see the same body roll, and the suspension roll angles differ mainly through tyre deflection, so Segers' ratio (Eq. 10.13) is pulled towards 50% and is only a rough index. To get $q$ you need forces: load cells, or pot travel × known wheel and ARB rates.
3. Correlate LLTD with **balance metrics**: understeer angle (Segers §7.5) or yaw rate vs a bicycle-model reference, gated for **mid-corner in mechanical corners**.
4. Track LLTD changes over a weekend in a table with balance outcomes. That becomes your car's sensitivity map.

**Where LLTD dominates and where it doesn't:**
- Dominates in **slow and medium mechanical corners** at steady state (the effect scales with $a_y^2$).
- Weak in **high-speed corners** where aero balance rules.
- **Transient** phases (entry, direction changes) are shaped by damper distribution and geometric (instant) transfer as much as by steady LLTD.

**Common pitfalls:**
- Fixing low-speed **exit** oversteer only with the rear bar when the root cause is diff locking, TC or throttle application. Rear roll stiffness does affect traction (through the inside-rear load, which limits drive with an open or lightly preloaded diff), but if the data show the diff or the driver is the cause, change those first.
- Assuming LLTD = roll stiffness distribution. Unsprung and geometric terms shift it by several points.

## L3 · Advanced

**Why the "magic number" is above static weight.** Several effects bias the optimum front LLTD above the static front fraction:
1. **Traction**: a RWD car on exit asks the rears for longitudinal force. Lower rear load transfer keeps more rear capacity (friction ellipse) for traction.
2. **Stability margin**: a car slightly front-limited at the limit is easier to drive and less likely to spin from lift-off or a gust.
3. **Tyre size asymmetry**: wider rears (most GT and formula cars) have lower contact pressure and different load sensitivity, so the rear pair tolerates more transfer. That pushes the optimum front LLTD *down*, partly cancelling items 1–2. The "+5%" is a heuristic net of these, not a law.
4. **Drivetrain**: FWD cars (TCR) run *much lower* front LLTD (very stiff rear) to protect the driven front axle. AWD/hybrid-front LMH cars sit differently again.

**Pair-analysis view (RCVD Ch. 7).** Treat each axle as a "pair" with an effective $F_y(\alpha)$ that includes load transfer. LLTD changes the shape of each pair curve: more transfer lowers the peak and rounds the top. Balance at the limit is determined by which pair saturates first at the $a_y$ demanded. Below the limit (linear range), LLTD acts through load-dependent **cornering stiffness**, which can have a different sensitivity from peak μ. A bar change can therefore affect "response" and "limit balance" differently.

**Aero and LLTD.** With downforce $D$ and aero balance $\beta_a$, static axle loads become $W_f + \beta_a D$, $W_r + (1-\beta_a)D$. The neutral-steer LLTD shifts with speed. One rule-of-thumb target is to keep mechanical LLTD offset from the **total** (static + aero) front load fraction at the speeds where mechanical balance matters. Aero balance is usually set so the car's high-speed balance is similar to its low-speed balance from LLTD, giving the driver a consistent car across corner speeds.

## L4 · Expert

**Limits of the LLTD concept.**
- **It is a steady-state, mid-corner metric.** On entry, LLTD combines with longitudinal transfer, brake balance and combined slip; on exit with traction and diff locking. Many "LLTD problems" are really entry/exit problems.
- **Inside-wheel lift** saturates an axle's transfer: once the inside wheel is unloaded, more roll stiffness on that axle does nothing more for that axle. Any additional roll moment must go to the other axle. TCR and some GT4 setups operate in this regime deliberately.
- **Chassis torsional stiffness** limits how much LLTD you can actually dial in. With a compliant chassis, front and rear roll stiffnesses are coupled through the chassis spring, and bar changes partly twist the tub.
- **Non-linear elements**: bump rubbers and packers engaged in roll (common in aero cars) add roll stiffness at high load, so LLTD becomes $a_y$ and speed dependent. Calculating LLTD from linear rates for a car running on its packers at 250 km/h is wrong. Use load cells or measured roll gradient vs $a_y$ to see it.
- **Tyre-temperature coupling**: higher transfer on an axle heats its outside tyre more (Segers §8.6). Over a stint, the effective load sensitivity on that axle grows if it leaves the window, so the "best LLTD" for one lap and for a stint differ.
- **Driver preference** (Segers Fig. 7.28: one driver fastest with near-zero average understeer angle, another fastest with more understeer). The number that maximises theoretical grip is not always the number that maximises lap time with a given driver.

## Key formulas

**Front lateral load transfer distribution**

$$\text{LLTD}_f = \dfrac{\Delta F_{z,f}\,t_f}{\Delta F_{z,f}\,t_f + \Delta F_{z,r}\,t_r}$$


**Roll stiffness distribution**

$$q = \dfrac{K_{\phi f}}{K_{\phi f} + K_{\phi r}}$$

- K_\phi: axle roll stiffness at the wheels (Nm/deg), suspension and tyres combined in series

**Axle roll stiffness from wheel rate and ARB**

$$K_{\phi} = \dfrac{\pi}{180}\,\dfrac{t^2}{2}\left(k_{w} + k_{arb,w}\right)$$

- k_w: spring wheel rate per side (N/m)
- k_{arb,w}: equivalent per-side wheel rate of the bar in pure roll (N/m); if the bar rate is quoted per unit of relative left–right wheel travel, use 2× that value
- t: track (m)
- K_\phi: suspension roll stiffness (Nm/deg); combine in series with the tyres, (π/180)·k_t t²/2 in Nm/deg, for the axle value

**"Magic number" starting point**

$$\text{LLTD}_f \approx \Psi_f + 0.05$$

- \Psi_f: static front weight fraction

## Trackside takeaways

- Front LLTD up = understeer; down = oversteer (steady state, mechanical corners).
- Starting point: front LLTD ≈ static front weight + ~5 points. It is a heuristic, adjusted for tyres, drivetrain and driver.
- ARBs change LLTD with minimal heave/pitch change, so they are the platform-friendly balance tool.
- LLTD ≠ roll stiffness distribution: unsprung and geometric terms move it by several points.
- Packers/bump rubbers in roll make LLTD speed- and g-dependent on aero cars.

## By car category

### GT3

Front-engined (e.g. BMW M4, AMG) and mid-engined (296, Huracán) GT3s have quite different static front weight, so their LLTD targets differ. ARB blade/position changes are the standard mid-corner balance tools, with BoP-fixed aero leaving mechanical balance to do much of the work.

### TCR

Front-drive: front LLTD is set far below static front weight (often very stiff rear bar, inside rear lifting) to keep the front pair evenly loaded for steering and traction.

### One-make Cup

With fixed springs, ARB positions are the main LLTD tool. Small adjustments are felt clearly because the cars have relatively high CG and modest aero.

### LMP2

High-speed balance is aero-dominated; LLTD (ARBs, often driver-adjustable) is tuned for slow corners. Packers in roll make the effective LLTD speed dependent.

### FIA F3

Stiff platform and limited ARB range. LLTD changes are small but effective in slow corners. Most balance work at high speed goes through wing and ride height.

### IndyCar

Driver-adjustable front and rear ARBs (and the weight jacker on ovals) let the driver move LLTD during a stint as fuel and tyres change. Engineers brief target positions per stint phase.

## References

- Milliken & Milliken, Race Car Vehicle Dynamics (SAE 1995): Ch. 18 Wheel Loads (TLLTD and the "magic number" starting point)
- Milliken & Milliken, Race Car Vehicle Dynamics (SAE 1995): Ch. 7 Steady-State Pair Analysis
- Milliken & Milliken, Race Car Vehicle Dynamics (SAE 1995): Ch. 16 Ride and Roll Rates
- Segers, Analysis Techniques for Racecar Data Acquisition, 2nd ed. (SAE 2014): §10.1 Lateral Weight Transfer (Roll-ratio index (Eq. 10.13); see its limits on a stiff chassis)
- Segers, Analysis Techniques for Racecar Data Acquisition, 2nd ed. (SAE 2014): §9.4 Front to Rear Roll Angle Ratio
