3. Occupant kinematics and injury biomechanics fundamentals
Every restraint architecture decision — where to place an anchor, how much webbing payout to allow, when to fire a pretensioner, how stiff the load limiter should be — is ultimately a bet about how the occupant will move and what loads the body can tolerate. This section builds the physical intuition behind those bets: the crash pulse that drives the event, the coupling between occupant and restraint, the injury criteria used to judge outcomes, and the way all three feed back into architecture choices.
3.1 The Crash Pulse as the Input Signal
The crash pulse is the vehicle body's deceleration history measured at the occupant compartment, usually at the B-pillar base or a seat track accelerometer, expressed in g versus time. It is the single most important input to any occupant simulation or sled test, because the occupant only "feels" the vehicle through this pulse plus whatever the restraint system adds.
Three parameters dominate the pulse's effect on occupant loading:
- Delta-v (Δv) — the velocity change, obtained by integrating the acceleration trace. A 56 km/h frontal barrier test produces roughly 15.6 m/s of Δv regardless of pulse shape.
- Peak acceleration and time to peak — a short, high-amplitude pulse (e.g., 40 g at 40 ms) loads the occupant earlier and harder than a long, low pulse (e.g., 25 g at 80 ms) with the same Δv.
- Pulse duration and shape — square, triangular, and haversine pulses with identical Δv produce measurably different chest deflection and head excursion.
For architecture work, the practical consequence is that a stiffer front structure shortens the pulse and raises its peak, which pushes the occupant forward earlier and demands earlier restraint engagement. A softer, longer pulse gives the restraint system more time to couple but increases ride-down distance requirements. Neither is universally better; the restraint architecture must be tuned to the pulse the vehicle actually produces.
3.2 Restraint Coupling: From Free Flight to Ride-Down
Between the moment of impact and the moment the occupant contacts the restraint, the occupant is in free flight — moving forward at the pre-crash velocity while the vehicle decelerates beneath them. The distance traveled in this phase is the free-flight displacement, and it is governed by the pulse and the initial gap between occupant and restraint (belt slack, seatback compliance, airbag standoff).
Once contact occurs, the restraint begins to apply force. The goal is ride-down: keeping the occupant coupled to the decelerating vehicle so that the occupant's velocity is reduced gradually rather than in a single impact. The quality of ride-down is captured by the occupant load path — the sequence of forces transmitted through the belt, airbag, seat, and knee bolster into the occupant's body.
Coupling is not binary. A belt that is too loose allows excessive free flight and a late, sharp engagement. A belt that is too tight or a load limiter set too high transmits the full pulse into the chest. The architecture must balance:
- Engagement timing — pretensioners and retractor locking determine when coupling begins.
- Force level — load limiters and airbag venting determine how much force is transmitted.
- Load distribution — belt geometry, airbag shape, and seat design determine where on the body the force lands.
These three are not independent. A pretensioner that removes slack earlier allows a lower load-limiter threshold without increasing excursion, because the occupant is already coupled when the pulse peaks. This is the core trade space that architecture decisions navigate.
3.3 Injury Criteria and Their Architectural Implications
Injury criteria translate occupant kinematics into a pass/fail judgment. The most relevant for restraint architecture are:
| Criterion | What it measures | Typical limit (frontal) | Architecture lever |
|---|---|---|---|
| HIC15 / HIC36 | Head resultant acceleration over 15/36 ms | 700 (HIC15) | Airbag timing, venting, head restraint geometry |
| Chest deflection | Rib cage compression (mm) | 50 mm (Hybrid III) | Load limiter force, belt routing, airbag pressure |
| Chest acceleration (3 ms clip) | Peak chest g sustained 3 ms | 60 g | Belt force distribution, seatback stiffness |
| Femur load | Axial force on femur (kN) | 10 kN | Knee bolster stiffness, seat pan angle |
| Neck injury (Nij) | Combined axial force and moment | 1.0 | Airbag tethering, head restraint position |
Each criterion maps to a physical load path, and each load path is controlled by specific architectural elements. For example, chest deflection is dominated by the belt force applied over the rib cage. If the load limiter is set to 4 kN, the chest sees roughly that force minus what the airbag and knee bolster absorb. Raising the limiter to 6 kN increases chest deflection but reduces forward excursion — a trade that may be necessary if the occupant is large or the pulse is severe.
3.4 From Biomechanics to Architecture: A Worked Example
Consider a frontal sled test at 56 km/h with a 50th-percentile male Hybrid III. The pulse is a 30 g haversine with 100 ms duration. The baseline restraint is a 4 kN load limiter, no pretensioner, and a 60 L airbag with two vents.
Baseline results (simulated): HIC15 = 620, chest deflection = 42 mm, femur load = 7.5 kN. All pass, but chest deflection is close to the 50 mm limit. The team wants to reduce chest deflection without increasing head excursion.
Option A: Lower the load limiter to 3 kN. Predicted chest deflection drops to ~35 mm, but head excursion increases by ~40 mm because the occupant moves further forward before the belt force balances the pulse. HIC15 rises to ~680 — still passing, but the margin shrinks.
Option B: Add a pretensioner that removes 40 mm of slack in the first 15 ms. This couples the occupant earlier, so the same 4 kN limiter now produces ~38 mm chest deflection because the belt engages before the pulse peak. Head excursion stays roughly constant. HIC15 drops to ~590.
Option C: Increase airbag venting to reduce internal pressure. This lowers chest deflection slightly but increases head excursion significantly because the bag bottoms out earlier. HIC15 rises to ~700 — marginal.
The architecture decision here is not "which option is best" in isolation. It is: given the pulse, the occupant, and the injury criteria margins, which combination of pretensioner, load limiter, and airbag tuning delivers the most robust performance across the expected occupant population? Option B is attractive because it improves chest deflection without sacrificing head margin, but it adds cost and complexity. Option A is cheaper but eats into the head margin. The biomechanics data makes the trade explicit.
3.5 Reading the Trade Space in Practice
The worked example above is a simplified simulation, but the pattern is real. In practice, the architecture team runs a matrix of sled or CAE simulations across occupant sizes (5th percentile female to 95th percentile male), pulse severities, and restraint configurations. The output is not a single number but a trade surface: chest deflection versus head excursion versus femur load, with each point representing a specific combination of pretensioner timing, load limiter force, and airbag venting.
The architecture decision is then a selection problem: choose the configuration that maximizes the minimum margin across all criteria and all occupants. This is why restraint systems are increasingly adaptive — a fixed load limiter cannot be optimal for both a 50 kg female and a 100 kg male. Dual-stage pretensioners, adaptive load limiters, and occupant classification sensors exist precisely because the biomechanical trade space is occupant-dependent.
When reviewing a restraint architecture proposal, ask three questions: What pulse was assumed? What occupant sizes were evaluated? Which injury criterion is the binding constraint? The answers reveal whether the architecture is tuned to the vehicle's actual crash behavior or to a generic test condition.