2. Regulatory and consumer rating requirements
Driver airbag performance is not evaluated against a single test. It is assessed through a layered matrix of regulatory compliance tests and consumer rating protocols, each with its own impact configuration, dummy type, injury criteria, and pass thresholds. For a DAB (driver airbag) program, the practical consequence is that one hardware set must satisfy multiple, sometimes conflicting, loading requirements. Understanding the matrix structure is therefore a prerequisite for both test planning and CAE correlation, because each configuration produces a different bag deployment state, a different dummy-to-bag interaction, and a different set of correlation targets.
Regulatory Layer: FMVSS 208 and ECE R94/R95
FMVSS 208 (49 CFR 571.208) is the most demanding regulatory driver-side requirement in terms of configuration count. For the driver position, the relevant test set includes:
- Rigid barrier, 0° frontal, 56 km/h — unbelted 50th percentile male (Hybrid III), the primary driver airbag sizing case.
- Rigid barrier, 0° frontal, 40 km/h — unbelted 50th percentile male, lower-severity deployment case.
- Offset deformable barrier (ODB), 40% overlap, 56 km/h — belted 50th percentile male, added under the 1997 sled/offset amendments.
- Angled rigid barrier, ±30°, 56 km/h — unbelted 50th percentile male, introduced to address oblique loading.
- Low-risk deployment (LRD) static and dynamic tests — out-of-position (OOP) occupants, including 1-, 3-, and 6-year-old dummies and the 5th percentile female, at various proximity positions.
The injury criteria for the 50th percentile male are HIC15 ≤ 700, chest deflection ≤ 63 mm, and neck Nij ≤ 1.0, with femur axial force ≤ 10 kN. The LRD tests use different, position-dependent thresholds and are the primary driver of bag venting, tethering, and adaptive deployment strategies.
ECE R94 (frontal impact) and ECE R95 (lateral impact) apply in UNECE markets. R94 uses a 40% offset deformable barrier at 56 km/h with a belted Hybrid III 50th percentile male, and additionally a 0° rigid barrier at 50 km/h. R95 is a lateral impact regulation and, while not a driver airbag sizing case, it constrains the side structure and seat design that interact with the driver's seating position. The key difference from FMVSS 208 is that R94 is fundamentally a belted test, so the airbag is tuned to supplement rather than replace the restraint load path.
Consumer Rating Layer: Euro NCAP and C-NCAP
Consumer protocols are more severe than regulation in several respects and change on a roughly three-year cadence. For driver airbag work, the following configurations matter most:
| Protocol | Key driver-side configurations | Dummy / criteria focus |
|---|---|---|
| Euro NCAP (2023+) | Frontal ODB 50 km/h, 50% overlap; MPDB 50 km/h, 50% overlap; frontal full-width 50 km/h | THOR 50th male; HIC, chest deflection, knee/thigh, and new "far-side" and "oblique" metrics |
| C-NCAP (2021/2024) | Frontal ODB 64 km/h, 40% overlap; MPDB 50 km/h; frontal full-width 50 km/h | Hybrid III 50th male; HIC15, chest deflection, femur force; OOP and child presence detection |
| Euro NCAP OOP | Static deployment at multiple proximity positions; dynamic OOP sled | 5th female, 6-year-old, 3-year-old; bag pressure and injury risk |
Two structural differences deserve emphasis. First, Euro NCAP's move to THOR and the MPDB (Mobile Progressive Deformable Barrier) changes the chest loading profile relative to Hybrid III, which directly affects bag pressure and vent tuning. Second, C-NCAP's 64 km/h ODB is a higher-energy offset case than either FMVSS 208 or ECE R94, making it a frequent worst case for bag cushion volume and tether design.
How the Matrix Drives DAB Design and CAE Correlation
The practical workflow is to map each configuration to a distinct bag state and then build a CAE matrix that mirrors it. A representative driver-side matrix for a mid-size sedan program might look like this:
Config ID | Protocol | Speed | Barrier | Dummy | Belt | Key target
D01 | FMVSS 208 | 56 km/h| Rigid 0° | H3-50M | No | HIC15 ≤ 700
D02 | FMVSS 208 | 40 km/h| Rigid 0° | H3-50M | No | HIC15 ≤ 700
D03 | FMVSS 208 | 56 km/h| ODB 40% | H3-50M | Yes | Chest defl ≤ 63 mm
D04 | FMVSS 208 | 56 km/h| Rigid ±30° | H3-50M | No | HIC15, Nij
D05 | FMVSS 208 LRD | static | — | H3-5F, child| — | OOP injury risk
D06 | ECE R94 | 56 km/h| ODB 40% | H3-50M | Yes | HIC15, chest
D07 | Euro NCAP | 50 km/h| MPDB 50% | THOR-50M | Yes | HIC, chest, far-side
D08 | C-NCAP | 64 km/h| ODB 40% | H3-50M | Yes | HIC15, chest, femur
Each row implies a different CAE setup: barrier model, initial velocity, belt routing, dummy positioning, and bag trigger time. The correlation target is not a single number but a set of time-history curves — bag pressure, dummy head acceleration, chest deflection, and belt force — that must match within a defined tolerance band. A common engineering tolerance is ±15% on peak values and a correlation rating (e.g., CORA or ISO/TS 18571) above 0.7 for the primary channels.
Practical note: The same physical bag is often required to pass D01 (unbelted, high energy) and D05 (OOP, low energy, proximity). These two cases pull the design in opposite directions: D01 wants a large, well-vented cushion that stays inflated long enough to absorb the unbelted occupant's energy, while D05 wants a low-output inflator and rapid venting to avoid injuring a close occupant. The resolution is usually an adaptive inflator with two output levels plus a tether/vent architecture that changes effective vent area with bag internal pressure.
Reading the Matrix for CAE Planning
When building the correlation plan, group configurations by the physical mechanism they exercise rather than by protocol name. D01, D02, and D04 all exercise the unbelted, full-deployment load path and can share a bag model with different initial conditions. D03, D06, and D08 exercise the belted offset load path and share belt routing and seat position. D05 and the Euro NCAP OOP cases exercise the low-energy, proximity load path and require a separate inflator model with time-dependent mass flow. D07 is unique because THOR's chest and abdomen response differs from Hybrid III, so the dummy model itself must be validated before the bag correlation is meaningful.
The following diagram summarizes how the regulatory and consumer configurations map onto the three driver-side load paths and the corresponding CAE model families.
The matrix also determines the order of CAE work. Start with the dummy model validation for the specific dummy type used in each protocol, because a bag correlation built on an unvalidated dummy model will absorb dummy error into the bag tuning and produce a non-physical result. Then run the unbelted full-deployment cases to size the cushion and inflator, followed by the belted offset cases to tune venting and tether geometry, and finally the OOP cases to verify that the adaptive deployment logic does not create a new injury risk. Each stage produces time-history data that must be compared against the corresponding physical test, and the correlation rating should be reported per channel rather than as a single aggregate number.
Correlation tip: When a configuration fails to correlate, check the bag pressure trace first. A mismatch in peak pressure usually points to an inflator mass-flow or vent-area error, while a mismatch in the pressure decay slope points to a bag permeability or fabric leak issue. Head acceleration mismatches that appear only after the pressure trace has matched are more likely to come from dummy positioning or belt routing than from the bag itself.
In summary, the regulatory and consumer matrix defines not just pass/fail thresholds but the entire CAE correlation architecture. FMVSS 208 supplies the unbelted and OOP cases, ECE R94 supplies the belted offset case, Euro NCAP adds THOR-based and MPDB loading, and C-NCAP adds a higher-energy 64 km/h offset. Grouping these by load path rather than by protocol name keeps the CAE model count manageable and makes the correlation targets explicit for each bag state.