4. OOP test matrix design

An out-of-position (OOP) test matrix is the bridge between the occupant-kinematics simulation model and the physical validation program. Its purpose is to define which occupant postures, which deployment conditions, and which hardware configurations must be evaluated so that the airbag system is proven safe across the realistic envelope of misuse and pre-crash positions. A poorly designed matrix either wastes sled and static-deployment budget on redundant cases or, worse, leaves a critical worst-case posture untested. This chapter explains how to build a defensible matrix from regulatory requirements, CAE sensitivity results, and physical-fixture constraints.

Static vs. Dynamic Deployment Tests

OOP testing splits into two complementary families. Static deployment tests (often called "static OOP" or "bag-out-of-position" tests) hold the occupant surrogate motionless at a fixed position relative to the module and fire the inflator. They isolate the deployment energy and the interaction between the expanding cushion and the occupant surface. Dynamic deployment tests add a relative velocity between occupant and module — typically a low-speed sled pulse — so the deploying cushion interacts with a moving body, capturing the combined effect of deployment force and occupant inertia.

The regulatory baseline is defined by FMVSS 208 (S13, S15, S17, S19, S21, S23, S25) and the equivalent UN ECE R94/R95 and consumer protocols. These specify a set of surrogate positions — for example the 1-year-old, 3-year-old, 6-year-old, and 5th-percentile female surrogates — placed at defined distances and orientations from the airbag module. The matrix must cover every regulated position, but a production program typically adds engineering cases beyond the regulation to cover the real-world envelope.

Building the Matrix from Sensitivity Data

Before committing to hardware, run a CAE sensitivity sweep. Vary the key OOP parameters one at a time and record the peak injury metrics — typically head acceleration (HIC), neck injury criterion (Nij), and chest deflection. The parameters that dominate the response become the matrix axes. A typical parameter set for a driver-side frontal airbag includes:

• Longitudinal gap between the occupant reference surface and the module cover (e.g., 0 mm to 150 mm).
• Vertical offset of the head or chin relative to the module centerline.
• Torso recline angle (e.g., 0° to 30° from vertical).
• Lateral offset for side-impact or curtain modules.
• Deployment timing relative to the sled pulse (for dynamic cases).

Once the sensitive parameters are known, the matrix is populated by combining the regulated anchor positions with the worst-case values identified by CAE. The goal is not to test every combination — that is combinatorially explosive — but to test the corners of the response surface where injury metrics peak.

Positioning Fixtures

A positioning fixture is the mechanical interface that holds the surrogate at the exact posture defined by the matrix. Its accuracy directly determines whether the test is valid. Key requirements:

• Repeatability: the fixture must return the surrogate to the same position within ±2 mm and ±1° across repeated setups. This is verified with a coordinate measuring machine (CMM) or a laser tracker before the test campaign begins.
• Rigidity: the fixture must not deflect or vibrate during deployment. A flexible fixture absorbs energy and produces artificially low injury readings.
• Adjustability: the fixture should allow the gap, vertical offset, and recline angle to be set independently, so the matrix can be executed without fabricating a new fixture for every case.
• Surrogate compatibility: the fixture must accept the specific ATD (e.g., Hybrid III 5th female, Q1.5, Q3) and its instrumentation cabling without interfering with the cushion path.

In practice, most labs use a modular fixture: a base plate bolted to the sled or static rig, a vertical column with a calibrated scale, and a surrogate cradle that can be locked at discrete angles. The setup is documented with photographs and a position record sheet for every test.

Worst-Case Scenario Selection

The worst case is not always the closest position. A surrogate pressed directly against the module may be pushed aside by the cushion, while a surrogate at an intermediate gap can be struck by the cushion at its maximum velocity. Similarly, a reclined torso can direct the deploying cushion into the neck rather than the chest. The matrix must therefore include:

• Closest regulated position — the minimum gap allowed by the standard.
• Intermediate gap — where CAE predicts peak head or neck loading.
• Maximum recline — where the cushion path intersects the neck.
• Lateral offset — where the cushion edge contacts the head asymmetrically.
• Dynamic timing — firing the inflator at the moment the occupant's head is closest to the module.

Each worst-case candidate should be justified by a CAE result or a field-data observation. A case that cannot be justified is a candidate for removal, which keeps the matrix lean.

Executing the Matrix: A Worked Example

Consider a driver-side frontal airbag for a mid-size sedan. The CAE sensitivity study shows that head HIC peaks at a gap of 60 mm with a recline of 20°, and neck Nij peaks at a gap of 30 mm with a recline of 25°. The regulated positions are at 0 mm, 50 mm, and 100 mm gap with 0° recline. The engineering matrix adds the two CAE-identified worst cases plus one dynamic case at the neck-peak posture.

The test sequence is:

1. Set the fixture to the regulated closest position (0 mm gap, 0° recline). Fire the inflator. Record HIC, Nij, chest deflection.
2. Move to the CAE head-peak position (60 mm gap, 20° recline). Fire. Record.
3. Move to the CAE neck-peak position (30 mm gap, 25° recline). Fire. Record.
4. Set the dynamic sled pulse to 15 km/h and repeat the neck-peak posture with the inflator fired at the time the head reaches minimum gap. Record.
5. Repeat the regulated mid-gap and maximum-recline positions for completeness.

Acceptance criteria are defined before testing: HIC ≤ 700, Nij ≤ 1.0, chest deflection ≤ 50 mm (values are illustrative and must be replaced by the program's actual targets). If any case exceeds a criterion, the matrix is expanded around that case — for example, by adding ±10 mm gap and ±5° recline — to map the failure boundary.

Documentation and Traceability

Every matrix case must be traceable to its justification: a regulation clause, a CAE result, or a field observation. A simple matrix table with columns for case ID, posture parameters, fixture setting, justification, and acceptance criteria is sufficient. The table below shows a representative excerpt.

The matrix is a living document. As CAE models are correlated with test results, the worst-case predictions shift, and the matrix should be updated to reflect the new understanding. The goal is not to test everything, but to test the right things with enough margin to demonstrate safety across the OOP envelope.