4. Airbag folding techniques in MADYMO

An airbag's folded geometry is not a cosmetic detail — it is the initial condition that governs how the cushion unfolds, where the fabric mass travels, and when the occupant first contacts the inflating bag. In MADYMO, the folded state is the starting point of every deployment simulation, so the folding technique you choose directly shapes the deployment kinematics you will later validate against physical tests. This section covers the three fold families available in MADYMO workflows, how fold lines are defined and mapped onto the reference mesh, and how each fold type biases the cushion's early expansion.

Why the Folded State Matters in MADYMO

MADYMO represents an airbag as a finite-element or facet mesh with an initial reference geometry (the flat pattern) and a folded geometry (the gas-volume state at t = 0). The solver inflates the folded mesh using a gas flow model — either a uniform-pressure (UP) or a gas-flow (GF) approach — and the fabric unfolds as the internal pressure rises. Because the fabric is modeled with membrane or shell elements, the fold pattern determines:

• the initial distance between the module and the occupant, which sets the time-to-contact;
• the direction of the first fabric motion (toward the occupant, laterally, or downward);
• the local strain concentrations that appear as the bag opens;
• the effective venting path, since folded layers can temporarily block vent holes.

A poorly folded model can inflate "too cleanly" and miss the real-world lag, snagging, and asymmetric opening that physical tests show. The fold technique is therefore a calibration parameter, not just a preprocessing step.

Flat Folds

A flat fold (also called a Z-fold or accordion fold) compresses the cushion into a stack of horizontal layers. In MADYMO, you define a series of parallel fold lines across the flat pattern, then map each panel onto the next using the FOLD keyword or the folding utility in the preprocessor. The result is a compact, roughly rectangular package that sits flat against the module housing.

Flat folds are the easiest to define and the most stable numerically. They are common for passenger-side airbags and for side curtains where the bag must deploy along a relatively straight path. The deployment kinematics are dominated by the first fold line: the top layer peels away first, then the stack unfolds sequentially. This produces a characteristic "unrolling" pressure trace — a slow initial rise, then a steep ramp as more layers separate.

Rolled Folds

A rolled fold wraps the cushion around a central axis, similar to rolling a poster. In MADYMO, the roll is defined by a roll axis and a set of angular increments; the mesh nodes are rotated and translated onto a spiral path. Rolled folds are typical for driver-side airbags, where the bag must expand radially from the steering-wheel hub.

The deployment kinematics of a rolled bag are fundamentally different from a flat fold. The outer wrap releases first, and the bag "unwinds" tangentially. This creates a strong tangential velocity component at the bag's outer edge, which is why rolled driver airbags tend to sweep across the occupant's chest before the bag is fully pressurized. If you model a driver airbag with a flat fold instead of a roll, you will typically under-predict the lateral sweep and over-predict the axial force.

Rolled folds are more sensitive to mesh quality. Because the nodes follow a spiral, adjacent elements can invert if the roll radius is too small relative to the element size. A practical rule is to keep the roll radius at least three times the average element edge length.

Complex Folds

Complex folds combine flat, rolled, and tuck patterns within a single cushion. They are used when the deployment path must be shaped — for example, a passenger airbag that must first expand upward along the windshield before rotating toward the occupant, or a knee airbag that must deploy downward and outward simultaneously.

In MADYMO, complex folds are built by chaining fold operations: a flat fold for the upper panel, a roll for the lower panel, and a tuck for the side gussets. Each operation is applied to a node set, and the order matters because later folds operate on the already-displaced mesh. The deployment kinematics are correspondingly complex: different regions of the bag release at different times, producing a staged pressure rise and a non-uniform contact force.

Complex folds demand the most validation effort. Because the fold sequence is user-defined, small changes in fold-line placement can shift the contact timing by several milliseconds. Always compare the simulated bag outline at 20 ms, 40 ms, and 60 ms against high-speed video before trusting the occupant response.

Fold Line Definitions and Mapping

A fold line is a polyline on the flat pattern that separates two panels. In MADYMO, fold lines are defined in the reference configuration and then mapped to the folded configuration through a rigid-body transformation per panel. The key parameters are:

Fold lines must be continuous and must not cross. If two fold lines intersect, the mapping becomes ambiguous and the solver will either error out or produce inverted elements. When building a complex fold, define the fold lines in the order they will be executed and verify that each line lies entirely within the panel it folds.

Influence on Deployment Kinematics

The fold technique sets the initial mass distribution and the initial fabric orientation. During inflation, the gas pressure acts on the inner surface, and the fabric accelerates along the path of least resistance — which is the direction the fold allows it to move. A flat fold releases layer by layer, giving a relatively linear displacement curve. A rolled fold releases tangentially, giving a rotational displacement component. A complex fold releases in stages, giving a stepped displacement curve.

These differences propagate into occupant interaction. A bag that unfolds too quickly may contact the occupant before the pressure has peaked, producing a high force at low displacement. A bag that unfolds too slowly may allow the occupant to "bottom out" against the module. The fold technique is therefore one of the primary tuning levers for matching the simulated occupant response to sled or vehicle test data.

Practical Workflow and Verification

A reproducible folding workflow in MADYMO follows four steps. First, prepare the flat pattern and verify that the mesh is manifold and free of duplicate nodes. Second, define fold lines as node sets, checking that each line is continuous and does not intersect another. Third, apply the fold operations in sequence, using FOLD_TYPE and FOLD_ANGLE to control each layer. Fourth, run a short inflation-only simulation (no occupant) and inspect the bag outline at 10 ms intervals.

Typical failure symptoms and their causes:

• Inverted elements at t = 0: fold radius too small relative to element size, or fold lines crossing.
• Bag inflates asymmetrically: non-uniform fold-line spacing or an unbalanced roll axis.
• Contact force too high at low displacement: bag unfolds too quickly — increase layer count or reduce initial gas mass flow.
• Occupant bottoms out: bag unfolds too slowly — reduce layer count or increase initial gas mass flow.

Once the folded model matches the physical bag outline at 20 ms, 40 ms, and 60 ms, the fold technique is considered validated. From that point, the fold becomes a fixed input, and subsequent tuning focuses on gas flow, venting, and occupant positioning. The fold is the foundation — get it right, and the rest of the deployment model becomes far more predictable.