3. Adult occupant protection frontal impact

Frontal impact remains the highest-volume injury scenario in Euro NCAP's adult occupant protection (AOP) assessment. For a vehicle programme, the practical task is not simply to "pass the crash test" but to read the protocol as a set of coupled requirements: two barrier configurations, a defined dummy fleet, a specific injury-criteria set, and performance thresholds that translate raw sensor channels into a score. This section explains how to interpret those requirements so that CAE and test engineers can align a programme's restraint and body-in-white (BIW) targets with the protocol before hardware exists.

Two barrier configurations, two different load cases

Euro NCAP frontal assessment uses two complementary tests. The full-width rigid barrier (FWRB) test at 50 km/h with 100% overlap loads both longitudinal members almost symmetrically. It is the primary discriminator for restraint-system performance — airbag timing, belt force limiting, and seat integrity — because the deceleration pulse is severe and the occupant loading is repeatable. The offset deformable barrier (ODB) test at 64 km/h with 40% overlap against a honeycomb-faced barrier introduces asymmetric loading: one longitudinal member absorbs most of the energy, the barrier deforms, and the vehicle yaws. The ODB test is the primary discriminator for structural intrusion and for the interaction between occupant kinematics and a rotating, intruding cabin.

Reading the protocol correctly means recognising that the two tests answer different questions. FWRB asks "does the restraint system control the occupant within a stable, non-intruding cabin?" ODB asks "does the structure keep the occupant survival space intact while the restraint system still manages the occupant?" A programme that optimises only one will typically fail the other. The figure below maps the two load paths and the resulting occupant-relevant outputs.

Dummy instrumentation: what each channel actually measures

Euro NCAP frontal tests use the Hybrid III 50th percentile male in the driver seat, with a 5th percentile female and/or a 95th percentile male depending on the protocol revision and the vehicle's seating configuration. The dummy is instrumented to produce the channels that feed the injury criteria. Reading the protocol means knowing which channel drives which criterion, because a programme that chases the wrong channel will over-engineer the wrong subsystem.

• Head: three orthogonal accelerometers at the head CG. The resultant acceleration time history is the input to HIC15 and to the 3 ms clip criterion (a₃ms). Head rotation is not directly scored in the classic frontal criteria but matters for interaction with the airbag and for the neck.
• Neck: upper-neck load cell and tilt sensors. The protocol scores neck injury via Nij (combined tension/compression and bending moment) and via axial tension/compression limits. Nij is normalised against the dummy's own intercept values, so it is dimensionless.
• Chest: chest deflection (potentiometer or IR-TRACC depending on dummy build) and chest acceleration. Chest compression is scored as an absolute deflection and as a rate; chest acceleration feeds a₃ms.
• Femur and knee: axial load cells in each femur, plus knee slider displacement. These channels capture load transfer through the lower limbs into the footwell and are sensitive to intrusion and pedal-panel behaviour.

Injury criteria and how thresholds are applied

Each criterion converts a physical channel into a scalar that is compared against a threshold. The protocol uses a two-level structure: a lower "performance" limit and an upper "capping" limit. Scores are interpolated between the two, and results above the capping limit are capped (not zeroed) for that body region. This matters for programme target-setting: a design that sits just below the capping limit still scores, but a design that sits just above the performance limit loses points progressively. The practical target is therefore to keep every criterion comfortably below the performance limit, not merely below the cap.

The values above are illustrative and are included to show the structure of the threshold logic; the binding numbers must always be read from the current protocol revision applicable to the programme's assessment year. Protocol revisions change limits, dummy builds, and even the set of criteria, so a programme that carries forward last year's threshold table is exposed to a silent compliance gap.

Reading a test result: from channels to a judgement

A useful way to read a frontal result is to work backwards from the score to the channel. Suppose the chest compression channel shows a peak of 30 mm at t ≈ 55 ms, with the performance limit at 22 mm and the cap at 50 mm. The criterion is above the performance limit but well below the cap, so the region loses a fraction of its points. The engineering question is then: what produced the 30 mm? In FWRB, a chest compression peak at 55 ms usually points to belt force limiting being too aggressive relative to the airbag's support, or to the seat belt geometry allowing too much forward excursion before the airbag is fully inflated. In ODB, the same peak may instead reflect intrusion pushing the steering column or the belt anchor rearward. The same channel value therefore implies different corrective actions depending on which test produced it — which is exactly why the protocol runs both.

For a programme, the practical workflow is to build a channel-to-criterion-to-subsystem map before the first hardware test. Each criterion is traced to the physical channel, the channel to the dummy location, and the location to the restraint or structural subsystem that dominates it. That map becomes the basis for CAE target-setting and for interpreting sled and full-vehicle results consistently. The following checklist is a compact, fillable version of that map for a frontal programme.

Finally, keep the protocol's scoring logic in view when setting targets. Because scores interpolate between performance and capping limits, a small margin below the performance limit is worth far more than a large margin below the cap. The most efficient programme strategy is to identify, for each body region, which of the two tests is the binding case, and to allocate restraint and structural effort to that case first — then confirm the other test still clears its own limits. That is the discipline the protocol is designed to reward.