# Hybrid systems in LMH / LMDh: deployment, energy & brake-by-wire

> In Hypercar the hybrid does not add peak power — total power is capped. It changes where energy comes from, which axle drives (LMH above a speed threshold) and how braking torque is split, which makes brake-by-wire a key engineering topic.

Area: Engine & Hybrid · Starting level: L2 Practitioner · Updated 2026-11-16
Source page: https://race-engineer-academy.vercel.app/lesson/engine-hybrid-lmh-lmdh

## L1 · Foundation

WEC Hypercar has two technical routes:

- **LMDh** (e.g. Porsche 963): a spec-supplier hybrid — a ≈ 50 kW Bosch MGU in the Xtrac gearbox on the **rear axle**, a spec Williams Advanced Engineering battery, combined with the manufacturer's ICE.
- **LMH** (e.g. Ferrari 499P): manufacturer-designed. A hybrid, if fitted, drives the **front axle** with up to ≈ 200 kW, making the car four-wheel drive when it deploys.

The crucial rule for both: the **total power** at the wheels (ICE + MGU) is capped by BoP at around 500 kW. When the MGU pushes, the ICE must deliver less. So the hybrid is **not a boost button** — it is:

1. an **energy recovery** device: braking energy is stored and re-used, so the car burns less fuel/energy per lap;
2. for LMH, a **traction** device: front-axle drive is only allowed **above a speed threshold** (set per car by BoP: in 2023 190 km/h for Toyota/Ferrari and 150 km/h for Peugeot, linked to front tyre size; check the current BoP table). Below it the car is rear-driven only, so the front motor cannot give AWD traction out of slow corners;
3. a **braking** device: regeneration provides part of the braking torque on its axle — which means the hydraulic brakes on that axle must be reduced to keep the driver's total braking and balance constant. That is the job of **brake-by-wire (BBW)**.

Energy per stint: BoP also fixes how much energy the car may use between refuels (measured by FIA sensors). Recovered energy helps you stay inside that budget at a given pace.

The engineer's world in Hypercar therefore includes **state of charge (SoC)**, deployment and regen strategies per lap, BBW calibration and energy-per-lap targets alongside the usual chassis work.

## L2 · Practitioner

**Deployment strategy.** Within the power cap, the question is *where* MGU power replaces ICE power. Considerations:
- **LMH**: front deployment adds traction only if the speed threshold is passed while the car is still traction-limited — e.g. high-speed exits on worn rears in the wet. Above the threshold in a straight line, AWD does little for acceleration (car is power-capped) but it shifts drive torque forward, which affects **balance**: front drive torque during exit reduces front lateral capacity (combined slip) → more exit understeer. The deployment profile is therefore also a handling tool.
- **LMDh**: rear MGU can deploy anywhere; it is mostly an efficiency and energy-shaping tool (and can fill turbo lag at low rpm).

**Regen and SoC.** The battery is small (energy store of the order of 1 kWh class, power-limited), so per-lap usage is a balance: regen under braking ↔ deploy on exits/straights. Engineers monitor **SoC at a fixed lap point** (start/finish): drifting down lap by lap means the strategy is energy-negative and the car will lose deployment late in the stint.

**Brake-by-wire implications (the big one for chassis engineers):**
- The driver requests a total braking torque via the pedal; the BBW controller subtracts regen torque from the requested torque on the regen axle and applies only the remainder hydraulically.
- Regen torque depends on SoC, battery/motor temperature, speed and rpm. When regen is reduced (full battery, hot motor), BBW must **instantly** add hydraulic torque. Imperfect blending is felt as **brake balance shifting during a stop** or between laps.
- Data checks: compare **effective brake balance** (hydraulic + regen per axle) vs the driver's set balance through each braking zone; look for steps when regen limits are reached. Lock-ups of the regen axle late in braking often coincide with regen fade-out at low speed.
- Brake temperatures on the regen axle are lower; pad/disc bedding and cooling choices differ from a non-hybrid car.

**Failure modes:** BBW faults default to a fallback hydraulic mode with a fixed balance — drivers must know the feel and the dash alarm.

## L3 · Advanced

**Power-cap accounting.** With a cap $P_{cap}$ measured at the driveshafts, the supervisor enforces $P_{ICE,wheels} + P_{MGU} \le P_{cap}$ (with additional limits on MGU power and ramp rates). For an LMH deploying 200 kW at the front, the ICE is limited to ≈ 300 kW at that instant. Consequences:
- **Fuel energy per lap drops** whenever regen energy is redeployed: every MJ recovered and reused (round-trip efficiency ≈ 70–85 %) is a MJ of ICE output not needed. Over a stint this is the main hybrid benefit under an energy-per-stint rule.
- Lap-time benefit comes from traction (LMH), improved drivability (torque fill) and the ability to spend the energy budget where it is most valuable.

**Balance with front torque (LMH).** Driving the front axle with torque $T_f$ consumes part of the front tyres' friction circle: $F_{y,f,max} \approx \sqrt{(\mu F_{z,f})^2 - F_{x,f}^2}$. Front drive above the threshold therefore *adds* understeer on fast exits, while removing some rear drive load *reduces* rear combined slip and power oversteer. Torque split (how much the supervisor sends front vs rear) is effectively a balance parameter in fast corners — BoP and regulations limit how it can be modulated so that it doesn't become torque vectoring.

**Brake blending dynamics.** With regen on one axle, effective balance:

$$BB_{eff} = \frac{T_{hyd,f} + T_{regen,f}}{T_{hyd,f} + T_{regen,f} + T_{hyd,r} + T_{regen,r}}$$

The BBW controller targets $BB_{eff} = BB_{set}$ continuously. Errors arise from (i) **caliper/pad torque uncertainty** — the controller commands pressure, but pad μ varies with temperature (±10–15 %), so the hydraulic part of the blended axle is less accurate than an all-hydraulic axle; (ii) **regen ramp-out** at low speed (motor torque limited, then zero near standstill) which must be replaced by hydraulic pressure in tens of ms; (iii) **downshift engine braking** passing through the same axle.

**Engineering workflow:** log requested vs achieved torque per axle, compute $BB_{eff}$ per braking zone, and histogram the deviation. A good calibration keeps $|BB_{eff} - BB_{set}|$ within ≈ 1 % for > 90 % of braking time. Driver complaints of "inconsistent brake feel" in hybrids are investigated here first, before pads or discs.

## L4 · Expert

**Optimal deployment as a control problem.** Given track model, power cap, MGU limits, SoC bounds and an energy-per-stint constraint, minimise lap time over a stint. The structure of the solution (Pontryagin / dynamic programming) has a **co-state** $\lambda$, the time value of a unit of stored energy. Deploy where the marginal time **saved** per MJ, $-\partial t / \partial E_{deploy}$, exceeds $\lambda$ (in s/MJ); harvest where the marginal time **cost** of harvesting per MJ is below $\lambda$. Two features matter in Hypercar:

1. **Power cap makes deployment substitutive**, not additive. In capped segments, deploying electric energy only saves ICE fuel energy — its "time value" is via the energy-per-stint constraint, so $\lambda$ is tied to the energy budget's shadow price.
2. **LMH threshold** creates a discontinuity: front deployment has a traction value only on exits that cross the threshold while still traction-limited. These corners (few per lap) are where a small change in threshold (BoP) changes lap time most — which is why BoP authorities treat the threshold as a lever.

**BBW as a balance tool.** Beyond equivalence with a hydraulic system, BBW permits a **brake-balance map vs speed or deceleration** (where regulations allow) — mimicking aero-dependent migration. Rules typically constrain the driver-selectable range and dynamic features; the engineer must document what is calibration (allowed) vs active balance control (often not).

**Thermal limits are the real constraint.** Battery cell temperature (typically managed to stay below ≈ 55–60 °C) and MGU winding/inverter temperatures cause derating; derating changes regen torque, hence brake feel, mid-stint. A stint plan that is energy-optimal on lap 3 may be thermally infeasible by lap 15. Model thermal states alongside SoC in the strategy tool; treat battery temp as a vital sign.

**Limits of the theory:** the optimal-control picture assumes the driver can modulate lift points precisely and that tyre state is unaffected. In practice, regen loading of one axle changes tyre temperature balance over a stint, and lift-and-coast changes brake temperatures — both feed back into handling, so the "energy-optimal" plan is a starting point, not the answer.

## Key formulas

**Total power cap**

$$P_{ICE} + P_{MGU} \le P_{cap}$$


**Effective brake balance with regen**

$$BB_{eff} = \frac{T_{hyd,f} + T_{reg,f}}{\sum (T_{hyd} + T_{reg})}$$


**Energy from MGU power**

$$E = \int P_{MGU}\,dt \quad (1\,\text{kWh} = 3.6\,\text{MJ})$$


## Trackside takeaways

- Hypercar total power is capped: hybrid deployment replaces ICE power, it does not add to it.
- LMH front MGU may only drive above a BoP speed threshold — no AWD traction out of slow corners.
- LMDh uses a spec ≈ 50 kW rear MGU; benefit is efficiency, torque fill and energy management.
- Brake-by-wire subtracts regen from hydraulic torque; check effective balance (hydraulic + regen) per braking zone.
- Watch SoC at a fixed lap point, plus battery/MGU temperatures — derating changes brake feel mid-stint.

## By car category

### Hypercar (LMH/LMDh)

Porsche 963 (LMDh): spec rear MGU and battery, BBW on the rear. Ferrari 499P (LMH): front-axle MGU, deployment above the BoP threshold, BBW blending regen with hydraulic torque on the regen axle. Energy per stint and power caps from BoP.

### LMP2

No hybrid: conventional hydraulic brakes and NA V8 — a useful baseline to understand what BBW changes in Hypercar.

### IndyCar

IndyCar introduced a low-voltage hybrid (supercapacitor + MGU in the bellhousing) in 2024: driver-triggered deployment and regen, energy-budget thinking similar to push-to-pass.

### GT3

No hybrid in GT3 regulations; the brake system remains fully hydraulic with mechanical bias adjustment.

## References

- Other: FIA WEC Hypercar technical regulations & BoP (power cap, deployment threshold, energy per stint — check current values)
- Milliken & Milliken, Race Car Vehicle Dynamics (SAE 1995): Ch. 20 Driving and Braking (brake proportioning and drive distribution)
- Segers, Analysis Techniques for Racecar Data Acquisition, 2nd ed. (SAE 2014): §5.5 Brake Balance (balance channels — extend with regen torque for hybrids)
