4. Folded airbag geometry

The folded airbag is not merely a compressed version of the deployed cushion. It is a deliberately engineered three-dimensional structure whose fold pattern, layer stacking order, and retention strategy determine how the cushion unfolds, in what sequence gas fills each region, and how quickly the occupant-facing surface reaches its target position. In a deployment test and CAE correlation workflow, the folded geometry is the single most influential input that is rarely measured directly and almost always approximated. Understanding how fold patterns are constructed, what packaging constraints they must satisfy, and how they map to unfolding kinematics is therefore essential before any correlation between test and simulation can be trusted.

Fold Pattern Taxonomy and Construction

Most driver and passenger airbags use a combination of three fold families. Accordion (Z-fold) folds stack the fabric in alternating pleats along one axis, producing a compact rectangular bundle that unfolds sequentially like a bellows. Roll folds wrap the cushion around a mandrel or the inflator axis, creating a spiral cross-section that releases fabric progressively as the roll unwinds. Dart or tuck folds remove local material to shape the cushion into a non-rectangular envelope, controlling where the cushion bulges first.

A typical passenger-side cushion might be constructed as follows: the main panel is first Z-folded along the vertical axis into a strip approximately 120 mm wide, then roll-folded from the outboard edge toward the inflator, and finally compressed to a package thickness of 35–45 mm inside a chute. The order matters. If the roll is placed outboard of the Z-fold, the cushion deploys outward and upward before expanding laterally; if reversed, the cushion expands laterally first and may contact the instrument panel edge. These are not cosmetic differences — they change the timing of occupant contact by several milliseconds.

Packaging Constraints

The folded geometry must satisfy a set of hard constraints imposed by the module and vehicle environment:

  • Module envelope: the folded cushion plus inflator, retainer, and cover must fit within the housing cavity, typically with less than 2 mm clearance to the cover tear seam.
  • Cover tear force: the folded stack must not pre-load the cover so heavily that tear seams open prematurely, nor so lightly that the cushion shifts during shipping vibration.
  • Retention: the fold must survive 10–200 Hz random vibration and thermal cycling from −35 °C to +85 °C without permanent deformation that would alter deployment.
  • Inflator interface: the gas inlet region must remain unobstructed so that the initial pressure pulse reaches the cushion interior without being throttled by folded fabric.

These constraints frequently conflict. A tighter fold improves packaging but increases fabric-to-fabric friction, which delays unfolding. A looser fold deploys faster but may not fit the module. The fold pattern is therefore a negotiated compromise, and the CAE model must represent that compromise rather than an idealized geometry.

From Fold Pattern to Unfolding Kinematics

When the inflator fires, the cushion does not inflate uniformly. Gas enters the inlet, pressurizes the first fabric layers, and the cushion unfolds in a sequence determined by the fold order. The occupant-facing surface may travel 200–400 mm in 20–40 ms, but the path is not straight. It typically follows a trajectory that first moves outward from the module, then upward or downward depending on the fold orientation, and finally stabilizes as the cushion reaches full volume.

Three kinematic quantities are most sensitive to fold pattern:

  • Initial breakout direction: the direction the cushion takes as it exits the module, governed by the outermost fold layer.
  • Unfolding sequence: the order in which fabric regions become pressurized, which determines whether the cushion contacts the occupant with a flat surface or an edge.
  • Time to full deployment (TFD): the interval from inflator trigger to the cushion reaching its design position, typically 25–50 ms for passenger airbags.
Example — fold order effect on breakout: Consider a passenger airbag with a 120 mm Z-folded strip and a 3-turn roll fold. If the roll is the outermost layer, the cushion exits the module along the roll axis (typically upward and outboard). If the Z-fold is outermost, the cushion exits along the Z-fold axis (typically forward). In a CAE model, swapping the fold order without changing any other parameter can shift the occupant contact time by 4–7 ms — enough to change injury criteria by 10–20%.

Representing Folded Geometry in CAE

Explicit dynamics solvers such as LS-DYNA or PAM-CRASH require the folded geometry as a mesh. Two approaches dominate:

  1. Direct folding simulation: the flat pattern is meshed, then folded using prescribed displacements or a folding tool. This preserves fabric continuity and layer contact but is computationally expensive and sensitive to mesh quality.
  2. Mapped or idealized folding: the folded geometry is created in CAD, meshed, and imported. This is faster but loses the exact layer-to-layer contact history, which must be approximated with contact definitions and initial gaps.

For correlation work, the second approach is common but must be calibrated. Key calibration parameters include:

*MAT_FABRIC
  1, 0.0012, 2.1e9, 0.3, 0.0, 0.0, 0.0, 0.0
*CONTACT_AUTOMATIC_SINGLE_SURFACE
  1, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0
  0.15, 0.15, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0
*INITIAL_VELOCITY_GENERATION
  1, 0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0

Here, the friction coefficient (0.15 in this example) between fabric layers is a first-order parameter. A value that is too low causes the cushion to unfold too quickly; too high and the cushion stalls. Typical calibration ranges from 0.10 to 0.25 depending on fabric coating and fold tightness.

Deployment Timing and Correlation Metrics

Deployment timing is evaluated against three primary metrics: breakout time (when the cushion first exits the module), time to full deployment (when the cushion reaches 95% of its design volume), and occupant contact time (when the cushion first touches the ATD). In a correlation study, these are compared between test and simulation at multiple time points, often using high-speed video at 1000–2000 fps.

A common failure mode in correlation is a simulation that deploys too fast because the folded geometry was modeled with uniform layer spacing and no initial contact. The fix is to introduce an initial gap or a pre-stress condition that mimics the actual fold compression. Another failure mode is a simulation that deploys in the wrong direction because the fold order was reversed during meshing. Both are detectable by comparing the breakout trajectory in the first 10 ms.

Practical check: Before running a full correlation, extract the folded geometry from the CAD model and compare its bounding box and layer count against the physical folded cushion. A mismatch in layer count greater than 10% almost always produces a timing error greater than 3 ms.

Visualizing the Fold-to-Deployment Chain

The following diagram summarizes how fold pattern, packaging constraints, and unfolding kinematics are linked, and where CAE correlation parameters enter the chain.

Fold Pattern → Packaging → Unfolding Kinematics → Deployment Timing Fold Pattern Z-fold (accordion) Roll fold Dart / tuck fold Layer stacking order Fold axis orientation (CAD / physical sample) Packaging Constraints Module envelope Cover tear force Vibration retention Inflator inlet clearance Thermal stability (Module CAD + test) Unfolding Kinematics Breakout direction Unfolding sequence Layer friction Gas fill path Trajectory of face (High-speed video + CAE) Deployment Timing Breakout time Time to full deploy Occupant contact time Injury criteria (Test vs CAE) Correlation feedback: adjust fold pattern or friction to match timing CAE Correlation Parameters • Fabric friction coefficient (0.10–0.25) • Initial layer gap / pre-stress • Contact definition (single surface, thickness) • Inflator mass flow rate curve calibrates

The diagram shows that fold pattern and packaging constraints are not independent — they feed into each other through the module design, and both determine the unfolding kinematics that CAE must reproduce. The correlation loop at the bottom is where engineering effort is concentrated: if breakout time or trajectory does not match, the fold pattern or friction coefficient is adjusted, and the simulation is re-run. This iterative process is the core of folded airbag CAE correlation.

Key Takeaways for Correlation Work

  • Fold pattern is a design variable, not a manufacturing detail. It must be documented and modeled explicitly.
  • Packaging constraints define the feasible fold space; any CAE model that ignores them may produce a physically impossible deployment.
  • Unfolding kinematics are governed by layer order and friction. These are the primary calibration knobs when timing does not match.
  • Deployment timing metrics — breakout, full deployment, and occupant contact — are the quantitative targets for correlation.
  • A mismatch in folded geometry layer count or bounding box is a leading indicator of timing error; check it before running the full simulation.

With the folded geometry properly characterized and its influence on kinematics understood, the next step in the workflow is to set up the deployment simulation with the correct initial conditions and contact definitions. That is where the fold pattern becomes a numerical input rather than a physical object, and where the quality of the folded geometry representation directly determines whether the CAE results can be correlated with test data.

← Previous 📖 Contents Next →