2. Crash test types and regulatory landscapes

Crash test repeatability cannot be discussed in the abstract. Every repeatability target, tolerance band, and uncertainty budget is anchored to a specific test configuration defined by a regulation or consumer rating program. A frontal offset test and a side pole test load the vehicle structure, restraint systems, and instrumentation in fundamentally different ways, and the dominant sources of scatter differ accordingly. This chapter maps the major crash test families and the regulatory frameworks that govern them, with particular attention to the repeatability requirements embedded in each.

2.1 The Four Principal Impact Families

Frontal impacts dominate occupant protection statistics because they represent the most frequent severe collision mode. The two most influential configurations are the full-width rigid barrier test and the offset deformable barrier (ODB) test. In a full-width rigid barrier test, the vehicle strikes a non-deforming wall perpendicular to its longitudinal axis, engaging both longitudinal rails simultaneously. In an offset test, only part of the vehicle front contacts a deformable barrier, typically at 40% overlap, which induces asymmetric loading, yaw rotation, and structural bending. The 56 km/h ODB at 40% overlap is the reference configuration in Euro NCAP and C-NCAP, while FMVSS 208 uses a 56 km/h full frontal rigid barrier for the driver and a 40 km/h full frontal for unbelted occupants.

Side impacts present a different repeatability challenge because the struck vehicle has minimal crush space between the door outer panel and the occupant. The moving deformable barrier (MDB) test, standardized as FMVSS 214 and ECE R95, uses a 1,500 kg barrier with a deformable honeycomb face striking the vehicle at 50–62 km/h perpendicular to its side. The side pole test, required by FMVSS 214 and Euro NCAP, drives the vehicle laterally into a rigid 254 mm diameter pole at 32 km/h. Pole tests are notoriously sensitive to impact point location: a 25 mm shift in the pole's longitudinal position relative to the vehicle can change occupant loading by 10–15%.

Rear impacts focus on seat and head restraint performance. FMVSS 202a specifies a 26 km/h rear sled pulse, while ECE R17 uses a similar sled-based approach. The key repeatability parameter is the sled pulse shape, which must be controlled within tight tolerances because the head-to-restraint geometry and the resulting neck injury criteria (NIC) are highly sensitive to pulse rise time.

Pedestrian impacts are unique because the "struck object" is a pedestrian surrogate rather than a vehicle. The legform, upper legform, and headform impactors are propelled into a stationary vehicle at defined velocities and angles. Euro NCAP and C-NCAP both require pedestrian protection assessment, with headform impacts at 35 km/h and legform impacts at 40 km/h. Repeatability here depends heavily on impactor calibration and the precise positioning of the vehicle relative to the launch system.

2.2 Regulatory Frameworks and Their Repeatability Philosophies

The regulatory landscape splits into two broad categories: mandatory type-approval regulations and voluntary consumer rating programs. Each imposes different repeatability expectations.

FMVSS (Federal Motor Vehicle Safety Standards) are mandatory in the United States. FMVSS 208 governs frontal occupant protection, FMVSS 214 side impact, and FMVSS 202a rear head restraints. These standards specify pass/fail injury thresholds — for example, a HIC₁₅ limit of 700 and a chest deflection limit of 63 mm for the 50th percentile male dummy in FMVSS 208. Critically, FMVSS does not mandate a minimum number of repeat tests or a statistical confidence level. The burden falls on the manufacturer to demonstrate that production vehicles consistently meet the threshold. In practice, this means OEMs conduct internal repeatability studies, often running the same test configuration 3–5 times to establish a coefficient of variation (CV) for each injury metric.

ECE regulations (now UN Regulations under the 1958 Agreement) are mandatory in Europe and many other markets. ECE R94 covers frontal impact, R95 side impact, and R17 rear impact. The type-approval process requires testing by a designated technical service, and the regulation includes provisions for reproducibility — if a second test yields a different result, the approval authority may require additional testing. The implicit repeatability expectation is that the same vehicle configuration tested at different laboratories should produce comparable results.

Euro NCAP is a voluntary consumer rating program that has become the de facto global benchmark. Its protocols are more demanding than ECE regulations and explicitly address repeatability. For example, Euro NCAP's frontal ODB test requires that the vehicle be positioned within ±10 mm of the nominal overlap and that the barrier face be replaced or verified after a specified number of tests. The rating scheme uses a points-based system where a single outlier test can affect the star rating, making repeatability a commercial necessity.

C-NCAP (China New Car Assessment Program) mirrors Euro NCAP's structure but with adaptations for the Chinese market. Its frontal test uses a 64 km/h ODB at 40% overlap, and its side test uses a 1,400 kg MDB at 50 km/h. C-NCAP's 2021 revision introduced stricter repeatability requirements, including mandatory verification of impactor calibration before each test and a requirement that the test laboratory maintain a documented uncertainty budget for velocity, position, and timing measurements.

2.3 Repeatability Requirements in Practice

Across all frameworks, repeatability is quantified through a small set of parameters. The most common is the coefficient of variation (CV) for key injury metrics. A CV below 5% is generally considered excellent for dummy injury criteria; 5–10% is acceptable; above 10% indicates a process that needs investigation. For velocity, the typical tolerance is ±0.5 km/h at impact. For position, the tolerance is ±10 mm for longitudinal placement and ±5 mm for lateral placement. For timing, the tolerance is ±1 ms for trigger signals.

These tolerances are not arbitrary. They are derived from the sensitivity of the injury metrics to each parameter. For example, in a side pole test, a 10 mm lateral shift in pole position can change the pelvic force by approximately 8%, which translates directly into a change in the injury assessment. Understanding these sensitivities is the foundation of any uncertainty budget.

2.4 A Worked Example: Repeatability Assessment for a Frontal ODB Test

Consider a hypothetical 56 km/h ODB test at 40% overlap. The laboratory runs the same vehicle configuration five times. The measured HIC₁₅ values are: 620, 645, 610, 660, 635. The mean is 634, and the standard deviation is 19.4. The CV is 3.1%, which is well within the excellent range. However, the chest deflection values are: 42 mm, 48 mm, 39 mm, 51 mm, 45 mm. The mean is 45 mm, standard deviation is 4.7 mm, and CV is 10.4%. This indicates that chest deflection is more sensitive to test variation than HIC. The laboratory should investigate whether the seat belt pretensioner timing or the dummy seating procedure is introducing scatter.

This example illustrates a key principle: repeatability is not a single number. Different injury metrics within the same test can have vastly different repeatability characteristics. A comprehensive uncertainty budget must account for each metric separately.

2.5 Summary

The four crash test families — frontal, side, rear, and pedestrian — each have distinct loading mechanisms and repeatability challenges. FMVSS, ECE, Euro NCAP, and C-NCAP impose different levels of repeatability rigor, with consumer rating programs generally being more demanding than mandatory regulations. The practical quantification of repeatability relies on CV for injury metrics and tight tolerances for velocity, position, and timing. The next chapter will build on this foundation by examining how these tolerances translate into a formal measurement uncertainty budget.