2. Crash pulse taxonomy for restraint design

A restraint system is not tuned to a single crash. It is tuned to a family of crash pulses, each defined by the barrier or impactor, the impact velocity, the overlap, and the direction of the load path. The airbag firing threshold, the seat belt load limiter, the retractor locking strategy, and the pretensioner timing must all survive the worst case within each family while remaining benign in the others. This chapter builds the taxonomy that the rest of the course depends on: what each pulse looks like, why it looks that way, and which restraint design parameters it actually drives.

2.1 What a Crash Pulse Actually Is

A crash pulse is the acceleration–time history measured on a vehicle structure reference point (typically the B-pillar base or the tunnel) during a barrier or impactor test. It is usually reported as a filtered acceleration signal in g, sampled at 10 kHz or higher and filtered to CFC 60 per SAE J211. Three quantities dominate restraint design:

• Peak acceleration — sets the upper bound of occupant deceleration and the load the belt and airbag must manage.
• Pulse duration — the time window in which the restraint must deploy and then absorb energy; a short pulse leaves no time for a slow airbag.
• Velocity change (Δv) and its shape — the integral of the pulse; the shape determines whether the occupant loads the belt early or late.

Two pulses can share the same Δv and still demand completely different restraint behavior. A rigid barrier pulse rises almost instantly and decays over roughly 80–120 ms; a pole pulse is narrow and violent; a side impact pulse is extremely short and lateral. The taxonomy below is organized by that physical reality, not by regulation number.

2.2 Full Frontal Rigid Barrier (FRB)

The vehicle strikes a fixed, non-deformable concrete or steel barrier at 100% overlap. Because the barrier does not crush, the vehicle must absorb essentially all of the kinetic energy through its own front structure. The resulting pulse is characterized by a very steep initial rise (often exceeding 30–40 g within 10–15 ms), a broad plateau, and a long decay tail. Typical test conditions are 56 km/h (FMVSS 208) or 64 km/h (Euro NCAP frontal, though that test uses an offset deformable barrier).

For restraint design, FRB is the high-Δv, long-duration case. The occupant has time to move forward, so the belt must manage the full ride-down. Key implications:

• Airbag must deploy early enough to be in position before occupant contact, but not so early that it is already deflating.
• Load limiter force level is set by the chest deflection target at this pulse, because the belt does most of the work.
• Pretensioner firing is relatively forgiving; the pulse gives a clear, early trigger signal.

2.3 Offset Deformable Barrier (ODB)

The vehicle impacts a deformable honeycomb barrier at 40% overlap, typically at 64 km/h. The barrier crushes, so the vehicle sees a softer, longer pulse than FRB. The critical difference is asymmetry: only one side of the front structure is loaded, so the vehicle yaws and the occupant experiences both longitudinal and lateral acceleration. The pulse peak is lower (often 20–30 g) but the duration is longer and the vehicle rotation introduces a lateral component that the restraint must tolerate.

ODB is the pulse that exposes asymmetric loading problems. A seat belt that works perfectly in FRB may allow the occupant to slide out of position when the vehicle yaws. Design consequences include:

• Seat belt geometry must resist lateral migration; the buckle pretensioner and anchor positions matter more here than in FRB.
• Side airbags and curtain airbags may be triggered by the yaw-induced lateral acceleration, even though the primary impact is frontal.
• Occupant classification and out-of-position (OOP) logic must account for the rotated posture.

2.4 Oblique Moving Deformable Barrier (OMDB)

The vehicle is struck by a moving deformable barrier at an angle, typically 15° and 50% overlap, at 90 km/h in the NHTSA oblique test. This is the most demanding frontal-family pulse for restraint robustness because it combines high Δv, a short rise time, and a strong lateral component. The occupant loads the belt diagonally, and the airbag must cover a wider area to prevent the head from rotating past the bag.

OMDB is where airbag coverage and belt routing are stress-tested. A bag sized only for FRB may be too narrow; a belt without a lap pretensioner may allow submarining under the oblique load. The pulse shape is often described as a sharp double peak: the first from the barrier contact, the second from the vehicle's structural rebound.

2.5 Pole Impact

The vehicle strikes a rigid pole, typically 254 mm diameter, at 32 km/h at 0° or 15° offset. The pole concentrates load into a small area, producing a very narrow, high-amplitude pulse. Peak acceleration can exceed 50 g, but the duration is short — often under 50 ms. The structural deformation is localized, and the occupant compartment may remain largely intact while the pulse is still severe.

Pole impact is the short-duration, high-peak case. The restraint challenge is timing: the airbag must deploy within a few milliseconds of the trigger, and the belt must lock almost immediately. Because the pulse is short, the occupant's forward excursion is limited, but the acceleration spike can cause high chest loading if the belt is already tight. Load limiter calibration must avoid over-constraining the occupant during the spike.

2.6 Side Impact

Side impact pulses are fundamentally different: the load path is lateral, the crush zone is short (often only 200–300 mm), and the time available is minimal. A typical side impact pulse rises to peak in 5–10 ms and is largely over within 40–60 ms. The occupant is struck directly by the door or the barrier, and there is no frontal crush zone to absorb energy.

Restraint design for side impact is dominated by proximity. The airbag must be between the occupant and the door before the door intrudes. This drives:

• Very fast side airbag inflators, often with a smaller bag volume than frontal bags.
• Pressure sensors or acceleration sensors in the door or B-pillar for early triggering.
• Curtain airbags for head protection, which must deploy downward and stay inflated longer than a frontal bag.

The pulse is also highly dependent on the barrier type: the EEVC side impact barrier, the NHTSA moving deformable barrier, and the pole side impact all produce different pulse shapes. The restraint must be robust across all of them.

2.7 Rear Impact

Rear impact pulses are lower in peak acceleration (often 10–20 g) but longer in duration. The primary restraint concern is whiplash: the seat back and head restraint must manage the relative motion between the torso and head. The pulse itself is less severe for the frontal airbag, but it drives seat design, head restraint geometry, and the activation of seat belt pretensioners in some systems.

Rear impact is also the pulse where the occupant is pushed into the seat rather than forward. The belt may not be loaded at all, so the restraint strategy relies on the seat structure and the head restraint. This is a different design domain from the frontal family, and it is included here because a complete restraint taxonomy must account for it.

2.8 Comparative Summary

The table below summarizes the key pulse characteristics and their primary restraint design implications. Values are typical ranges from published test data and are intended as design guidance, not as regulatory limits.

Pulse Family	Typical Peak (g)	Duration (ms)	Primary Restraint Driver
Full Frontal Rigid Barrier	30–45	80–120	Load limiter force, airbag volume
Offset Deformable Barrier	20–30	100–140	Belt geometry, yaw tolerance
Oblique Moving Deformable Barrier	35–50	60–100	Airbag coverage, belt routing
Pole Impact	40–60	30–60	Deployment timing, belt lock speed
Side Impact	30–60	20–50	Proximity, side bag inflator speed
Rear Impact	10–20	80–150	Seat and head restraint geometry

2.9 Reading a Pulse for Restraint Calibration

When you receive a pulse trace, the first step is to identify the family and then extract the parameters that matter. A practical workflow:

1. Identify the test configuration — barrier type, overlap, velocity, and impact angle. This tells you which family you are in.
2. Compute Δv by integrating the acceleration trace. Compare it to the expected value for that test; a mismatch indicates a sensor or filtering issue.
3. Locate the trigger point — the time at which the acceleration exceeds the airbag ECU's threshold. This is typically a calibrated value, often around 3–8 g for frontal and higher for side.
4. Measure time-to-peak and the pulse width at 50% of peak. These determine the airbag deployment window.
5. Check for secondary peaks — these indicate structural rebound or a second impact event, and they can cause the occupant to load the belt twice.

For example, a frontal pulse with a 5 g trigger threshold, a 12 ms time-to-peak, and a 90 ms duration tells you the airbag must be fully inflated by roughly 30–40 ms after trigger. If the airbag inflator takes 35 ms to fill, the margin is thin and the calibration may need a faster inflator or a lower trigger threshold. This is the kind of quantitative reasoning that the taxonomy enables.

The schematic above is not a substitute for measured data, but it captures the essential differences: FRB and ODB are broad and relatively gentle; OMDB and pole are sharp and early; side impact is the shortest and most lateral; rear impact is the lowest and longest. Each shape maps to a different restraint calibration strategy, and the rest of this course will build the methods for finding calibrations that are robust across all of them.