4. Comparative analysis of test configurations

FMVSS and UN regulations share a common engineering goal—protecting occupants in a crash—but they express that goal through different test configurations, different barriers, different dummies, and different acceptance criteria. For anyone building an evidence crosswalk, the test configuration is the anchor point: it defines what physical event is being reproduced, what measurement channels exist, and therefore which evidence can legitimately be mapped from one regulatory regime to the other. This chapter compares the four principal crash modes—frontal impact, side impact, rear impact, and roof crush—and identifies where FMVSS and UN requirements align closely enough to share evidence, and where they diverge so far that evidence must be regenerated.

Frontal Impact: The Closest Alignment

Frontal impact is the mode with the strongest structural correspondence. FMVSS 208 (occupant crash protection) and UN Regulation No. 94 (frontal impact protection) both use a rigid or deformable barrier impact at approximately 56 km/h, both evaluate a Hybrid III 50th percentile male in the driver position, and both measure head injury criterion (HIC), chest deflection, and femur loads. The barrier face differs: FMVSS 208 uses a full-width rigid barrier at 56 km/h for the belted test, while UN R94 uses a 40% offset deformable barrier (ODB) at 56 km/h. This is not a cosmetic difference. The ODB concentrates energy into one side of the vehicle, loading the longitudinal rail asymmetrically and producing occupant kinematics that a full-width rigid barrier does not reproduce.

For crosswalk purposes, the practical consequence is that HIC and chest deflection values from a full-width rigid barrier test cannot be substituted for ODB results without justification. The measurement channels are compatible, but the loading path is not. A defensible crosswalk entry should record the barrier type, overlap percentage, impact speed, and dummy position as mandatory metadata, and should flag any claim of equivalence between full-width and offset results as requiring engineering rationale rather than simple numeric comparison.

Side Impact: Divergent Barriers and Dummies

Side impact exposes the widest divergence. FMVSS 214 specifies a moving deformable barrier (MDB) with a crabbed 27° impact angle at 53.6 km/h, using either the ES-2re or SID-IIs dummy depending on vehicle type, and evaluates rib deflection, abdominal force, and pelvic acceleration. UN Regulation No. 95 uses a different MDB—the AE-MDB in its later amendment series—at 50 km/h with a perpendicular impact, and relies on the ES-2 dummy. The barrier mass, face geometry, honeycomb stiffness, and impact angle all differ.

These differences mean that a side impact test conducted to FMVSS 214 does not produce evidence that can be directly entered against UN R95 criteria. The dummy channels are broadly analogous—rib deflection in millimeters, abdominal force in kilonewtons—but the loading environment is different enough that a pass in one regime says little about the other. In crosswalk practice, side impact is typically treated as a "regenerate" category: the evidence must come from a test performed to the target regulation's own configuration, or from a validated CAE model correlated to that configuration.

Rear Impact: Fuel System Integrity vs. Seat Strength

Rear impact reveals a structural mismatch in regulatory scope rather than in test hardware. FMVSS 301 (fuel system integrity) evaluates rear impact at 80 km/h with a moving barrier, focusing on fuel leakage and post-impact fire risk—not on occupant injury metrics. UN Regulation No. 34 (fuel system integrity) covers a similar rear impact at 80 km/h but with its own barrier specification and leakage limits. Meanwhile, UN Regulation No. 17 (seat strength) and FMVSS 207/210 address seat anchorage and head restraint performance under rearward loading, which FMVSS 301 does not.

The crosswalk implication is that rear impact evidence must be split by regulatory objective. Fuel-system evidence from FMVSS 301 maps to UN R34 only after confirming barrier mass, impact speed, and leakage measurement method match. Seat-strength evidence has no FMVSS 301 counterpart and must be sourced separately from FMVSS 207/210 testing or UN R17 testing. A single "rear impact" crosswalk row is therefore misleading; the row should be decomposed by objective.

Roof Crush: A Quasi-Static Outlier

Roof crush is the clearest divergence in test type. FMVSS 216a applies a quasi-static load to the roof rail through a flat rectangular platen, requiring the roof to withstand a force equal to a multiple of vehicle weight (up to 4× unloaded weight for vehicles under 2,722 kg) without exceeding a specified platen displacement. UN Regulation No. 29 is structurally similar—quasi-static platen loading—but its force requirements, platen dimensions, and displacement limits differ, and it applies to commercial vehicle cabs rather than passenger cars in its primary scope.

Because both are quasi-static, the measurement is force versus displacement rather than dummy injury metrics. This makes roof crush evidence relatively portable in principle: a force-displacement curve can be re-evaluated against a different force threshold. But the platen geometry and loading location must match, and the vehicle weight normalization must be recomputed. In practice, roof crush crosswalk entries are the most amenable to analytical re-evaluation and the least dependent on dummy hardware.

Building the Crosswalk Rows

The comparison above suggests a practical rule for structuring crosswalk entries. Each row should be keyed not by crash mode alone but by the tuple (crash mode, regulatory objective, test configuration). Frontal impact yields two rows—one for full-width rigid barrier evidence and one for ODB evidence—because the barrier type changes the loading path. Side impact yields a single row marked "regenerate" unless a validated CAE correlation exists. Rear impact yields at least two rows: fuel-system integrity and seat strength. Roof crush yields one row that is analytically re-evaluable provided platen geometry and weight normalization are recomputed.

A worked example makes this concrete. Suppose a vehicle has undergone FMVSS 208 full-width rigid barrier testing at 56 km/h with a Hybrid III 50th percentile male, producing HIC = 620 and chest deflection = 42 mm. To crosswalk this against UN R94, the analyst must first confirm the barrier type mismatch. The HIC and chest deflection channels are compatible, but the ODB loading is asymmetric. A defensible entry would state: "Frontal evidence available for full-width rigid barrier configuration; UN R94 ODB evidence not established; CAE correlation to ODB configuration required before mapping." The numeric values are recorded as reference data, not as compliance evidence.

For roof crush, the re-evaluation is more tractable. Given a force-displacement curve from an FMVSS 216a test, the analyst can extract the peak force at the specified platen displacement, normalize by vehicle unloaded weight, and compare against the UN R29 threshold—provided the platen dimensions and loading location are documented and acceptably close. If the platen geometry differs materially, the curve cannot be re-evaluated without correction, and the entry should be flagged accordingly.

Alignment and Divergence Summary

The pattern across the four modes is consistent: alignment is highest where the physical loading and measurement channels coincide (frontal, with caveats on barrier type), lowest where barrier hardware and dummy selection diverge (side), split where regulatory objectives differ (rear), and analytically tractable where the test is quasi-static (roof crush). The crosswalk is therefore not a single equivalence table but a structured evidence map in which each row carries its own confidence level and its own regeneration requirement. Treating all four modes as interchangeable would produce a crosswalk that looks complete but cannot withstand technical review.