# Brake bias in practice: balance bar, master cylinders, in-car adjust & migration

> How the hydraulic bias is actually set — balance bar position, master-cylinder sizes, driver adjusters — how to measure it with pressure sensors, and how brake-by-wire migration in hybrid prototypes changes the bias through the braking event.

Area: Brakes · Starting level: L2 Practitioner · Updated 2026-11-17
Source page: https://race-engineer-academy.vercel.app/lesson/brakes-bias-hardware-migration

## L1 · Foundation

**The hydraulic system.** The pedal pushes on a **balance bar** (bias bar) that drives two separate **master cylinders**, one for the front circuit, one for the rear. Each master cylinder converts force into pressure: $p = F/A_{MC}$. Two ways to change the front/rear pressure split:
1. **Move the balance bar pivot** (the driver's bias adjuster): the cylinder closer to the pivot gets a larger share of the pedal force.
2. **Change master-cylinder sizes**: a smaller bore gives *more pressure* for the same force (but needs more pedal travel to displace the same fluid volume).

Common master-cylinder bores (inches): 0.625, 0.700, 0.750, 0.813, 0.875, 0.938, 1.000.

**Example.** Balance bar centred (50/50 force split), front 0.700" (248 mm²), rear 0.750" (285 mm²): $p_F/p_R = 285/248 = 1.15$ → **pressure bias 53.4 %**. Changing the rear to 0.813" (335 mm²) gives $p_F/p_R = 1.35$ → **57.4 %**, a 4 % step — too coarse for fine tuning, which is why the bar adjuster exists.

**See it on the pedal box.** Below, a ram pushes the brake pedal like the driver's foot. The pedal pushes the lit balance bar, whose two ends drive the front and rear master cylinders. The bars show front and rear line pressure on one scale; the gauge shows the pressure bias, with a small marker for the force split.

**Try it:**
- Set **600 N** with the bar centred and the **.875″** rear cylinder: the force splits 50/50, yet the front reads about **47 bar** and the rear **35 bar** – a **57.6 %** pressure bias from the bore difference alone.
- Change the pedal force: both pressures scale together and the bias does not move – bias is a ratio, not a force.
- Turn the **bias adjuster** to +4 mm: the bias rises to about **64 %**. Select the **.750″** rear (same as the front): with the bar centred the bias is exactly **50 %**.

**Takeaway:** the master-cylinder bores set the coarse bias in steps of a few percent; the balance bar fine-tunes it (about **1.7 % per turn** here), which is the adjustment the driver makes from the cockpit.

**Brake bias definition (Segers Eq. 5.3):**

$$BB_F = \frac{p_F}{p_F + p_R}\times100\,\%$$

measured with front and rear line-pressure sensors. Some cars display this computed value; others display the *adjuster position* (turns or clicks). Know which one your dash shows — "55" on one car is not "55" on another.

**Driver in-car adjustment.** Most race cars have a bias knob or lever (cable to the balance bar, or electric motor). Drivers adjust for fuel load, tyre wear, track evolution, rain and even per corner in some categories. A typical click is a fraction of a percent of pressure bias (car-dependent — measure it on your car with the pressure sensors).

**Typical values:** GT3 ~52–58 % front; formula cars ~53–62 %; prototypes depend on the hybrid system. These numbers only mean something with the specific caliper/disc/pad package of that car.

## L2 · Practitioner

**Measuring and checking the bias on track (Segers §5.5).**
- **Bias math channel** $BB_F = p_F/(p_F+p_R)$, *gated*: only valid when front pressure exceeds a threshold (5–15 bar), otherwise the channel divides by near-zero. When computing lap averages, exclude gated-off zeros or use the per-lap maximum.
- **Run chart of bias per lap** shows when the driver moved the adjuster (driver changes, drying track — Segers Fig. 5.8/5.9).
- **XY plot $p_R$ vs $p_F$** should be a straight line through the origin. Scatter or a kink indicates compliance or friction: caliper deflection, pedal-box free play, hose expansion, seal friction.
- **Bias vs front pressure** plot: a well-built system shows < 5 % variation in bias above ~15 bar. Large variation at low pressure = free play or friction; a "bump" at a specific pressure = caliper deformation or a mechanical stop (Segers Fig. 5.11).
- **Bias through a braking zone**: watch for the bias drifting forward as the driver releases (static friction in the pedal box / MC seals). That shifts the effective bias at turn-in, exactly where the driver feels balance — a 5–10 % shift there can completely change entry balance.

**Pedal and master-cylinder choice.**
- Pedal ratio (typically 3:1 to 5:1) × master-cylinder area sets pressure per kg of pedal force: e.g. 4:1 ratio, both MCs ~0.70–0.75" → ~56 kgf pedal force for ~43 bar front and ~38 bar rear with the bar centred (around 1.3 g on a GT3-like car).
- Smaller MCs = lighter pedal, longer travel; bigger MCs = firmer, shorter pedal but more effort. Drivers often care more about **pedal feel and modulation** than the absolute force.
- Change MC sizes for gross bias shifts (new pad compound, carbon vs iron) and keep the bar near its centre so the driver has adjustment range both ways.

**Migration (brake-by-wire, hypercars).** On LMDh cars the rear axle carries the hybrid MGU, and on LMH cars hybrid recovery can act on the front axle. The driver's pedal commands a *total* brake demand; an ECU splits it between friction brakes and electric regeneration. Because regen torque depends on speed, state of charge and strategy, the hydraulic pressure on the regen axle is actively controlled (brake-by-wire) to keep the **target bias**. **Migration** maps let the target bias change during the braking event — usually as a function of pedal pressure or speed — e.g. moving the bias forward or rearward as pressure decreases towards turn-in. Drivers adjust base bias and migration from the steering wheel.

**Conventional cars (F2, F3, GT3, LMP2):** no electronic migration; any variation through the zone comes from hydraulics (compliance, friction, proportioning valves where used) and from the driver's pedal profile.

**Misconception:** "once the bias is set in the garage, it's set." Pad wear, pad μ vs temperature, compliance and driver technique all move the effective bias on track; the data channel is the truth.

## L3 · Advanced

**Balance bar mechanics.** For a balance bar with pivot at distances $x_F$ and $x_R$ from the front and rear clevis centres, static equilibrium gives $F_F\,x_F = F_R\,x_R$, so $F_F/F_{pedal} = x_R/(x_F + x_R)$. Pressure bias:

$$BB_F = \frac{F_F/A_F}{F_F/A_F + F_R/A_R} = \frac{x_R A_R}{x_R A_R + x_F A_F}$$

Sensitivity: with $x_F + x_R = 50$ mm, moving the pivot by 1 mm changes the force split by 2 %; the pressure-bias change depends on the area ratio. The adjuster's thread pitch (e.g. 1 mm/turn) therefore defines "% per turn". Note that a balance bar only works as designed if both master cylinders are in their linear range — if one circuit bottoms out (pad knock-back, air, MC at end of travel) the bar *tilts* to its stop and the bias jumps. Excessive bar tilt (> ~5–7°) also changes the effective lever arms.

**Compliance and dynamic bias.** Each circuit has a volume–pressure characteristic (caliper body flex, piston seal deflection, hose expansion, pad compressibility). The front and rear circuits rarely have the same compliance, so the pressure split varies with pressure: at low pressure the stiffer circuit builds pressure first. Thermal effects (fluid expansion, caliper growth) add drift over a stint. Segers notes that a driver with very consistent pedal application hides some of this, which is why bias channels look different between drivers in the same car.

**Brake-by-wire control.** In an LMDh-type system the pedal feeds a pedal simulator and the front circuit directly (often); the rear circuit is controlled by an electro-hydraulic actuator. Control law (simplified):

$$T_{R,fric} = T_R^{*}\big(\text{demand},\,BB_{target}\big) - T_{regen}$$

where $T_R^{*}$ is the desired total rear torque derived from the driver's demand and the target bias (base + migration). If $T_{regen}$ falls suddenly (battery full, regen limit, failure), the friction torque must rise within tens of milliseconds or the effective bias jumps forward; if regen exceeds the target (e.g. overrun with high regen), the rear can be over-braked even with zero hydraulic pressure. Fault modes and response times are therefore key reliability items for hypercar engineers.

**Migration philosophy.** The physics of the ideal distribution (previous lesson) says the target should follow axle loads; the driver's need at turn-in (lateral capacity at the rear) and rotation preference also matter. Typical tuning: start with a migration that holds the force split near the ideal in the high-pressure phase, then shape the low-pressure end (trail braking) to the driver's entry-stability/rotation preference.

## Key formulas

**Pressure bias (Segers Eq. 5.3)**

$$BB_F = \frac{p_F}{p_F + p_R}\times 100\,\%$$


**Master cylinder pressure**

$$p = \frac{F_{pushrod}}{A_{MC}} = \frac{F_{pushrod}}{\tfrac{\pi}{4} d_{MC}^2}$$


**Balance-bar bias**

$$BB_F = \frac{x_R A_R}{x_R A_R + x_F A_F}$$

- x_F: pivot to front MC clevis (mm)
- x_R: pivot to rear MC clevis (mm)
- A_F: front MC area (mm²)
- A_R: rear MC area (mm²)

## Trackside takeaways

- Bias = p_F/(p_F+p_R), from pressure sensors, gated above 5–15 bar.
- Gross bias changes: master-cylinder sizes. Fine changes: balance-bar adjuster. Keep the bar near centre.
- Know whether your dash shows computed % or adjuster position.
- Check consistency: p_R vs p_F should be a straight line; < 5 % bias variation above ~15 bar.
- Watch bias drift at release — that's the bias the driver feels at turn-in.
- Brake-by-wire cars blend regen and friction to hold a target bias; migration shapes it through the zone.

## By car category

### GT3

Driver bias adjuster plus MC options; ABS is active, so bias shapes entry balance and ABS workload rather than raw lock-up. Typical 52–58 %.

### One-make Cup

Limited hardware freedom (spec MCs/calipers); bias adjuster is the main tool. In a Cup car without ABS (992.1 Porsche 911 GT3 Cup), bias is directly a lock-up and stability tool; the 2026 992.2 911 Cup has Bosch M5 ABS.

### FIA F2

Conventional hydraulic system with driver adjuster and carbon brakes; bias commonly moved during a race as fuel burns and brakes/tyres change.

### FIA F3

Conventional hydraulics with in-car adjuster; drivers use it actively in qualifying and races.

### LMP2

Conventional hydraulics, no hybrid: classic balance bar/MC tuning with in-car adjuster.

### Hypercar (LMH/LMDh)

Brake-by-wire on the hybrid axle with base bias and migration maps adjustable from the wheel; regen strategy and SOC interact with braking feel. Verify response to regen faults.

### IndyCar

Driver-adjustable bias bar; teams log bias position and line pressures. The hybrid (from 2024) adds driveline regen considerations to braking.

## References

- Segers, Analysis Techniques for Racecar Data Acquisition, 2nd ed. (SAE 2014): §5.5 Brake Balance (Eq. 5.3, Figs 5.6–5.13)
- Segers, Analysis Techniques for Racecar Data Acquisition, 2nd ed. (SAE 2014): §5.6 Pedal Travel
- Milliken & Milliken, Race Car Vehicle Dynamics (SAE 1995): Ch. 20 Driving and Braking
