3. CRS types and structural architecture

Child restraint system (CRS) crash simulation is only as credible as the structural model behind it. Before meshing a single shell element, the analyst must decide what kind of restraint is being represented, how it transfers load into the vehicle, and which belt or anchorage path carries the occupant's kinetic energy. This chapter maps the CRS landscape onto simulation-relevant architecture: infant carriers, convertible seats, boosters, ISOFIX/LATCH attachments, and the belt-routing configurations that determine how a seatbelt engages the shell.

Four Product Families, Four Load Paths

Regulatory categories (ECE R44/R129, FMVSS 213) group CRS by the child's mass and stature, but from a simulation standpoint the more useful classification is by primary load path. Each family behaves differently under frontal, side, and rear impact because the restraint couples to the vehicle through different hardware.

The critical distinction for the analyst: infant carriers and convertibles are load-bearing structures that absorb and redirect crash energy, whereas boosters are positioning devices that only route the vehicle belt across the child's skeletal load paths. A booster model therefore needs an accurate belt geometry and a validated dummy pelvis/shoulder interaction, but almost no shell stiffness calibration.

ISOFIX / LATCH Architecture

ISOFIX (ECE R14) and LATCH (FMVSS 225) define a standardized rigid or semi-rigid connection between the CRS and the vehicle body. The system comprises two lower anchorages at the seat bight, spaced 280 mm apart, plus an upper tether anchorage or a support leg. In simulation, this is modeled as a set of kinematic constraints rather than a belt.

• Connectors: latch hooks or rigid bars, represented by rigid or stiff spring elements with defined engagement clearance (typically 0–3 mm).
• Top tether: a tension-only element that limits forward head excursion and shell rotation; model as a nonlinear spring with a slack region.
• Support leg: a compression member reacting against the floor pan; model as a stiff axial spring with a defined floor contact.

A common modeling error is treating the ISOFIX connection as perfectly rigid. Real connectors exhibit 5–15 mm of compliance under 10 kN, and this compliance measurably changes head excursion. A calibrated connector spring with a force–deflection curve derived from a component pull test is the recommended approach.

Belt-Routing Configurations

For belt-installed CRS, the routing path determines how belt tension resolves into shell compression and rotation. Three configurations dominate:

1. Rear-facing belt path: the lap belt passes over the shell and the shoulder belt routes behind the shell or through a lock-off. The belt must be locked (ALR or locking clip) to prevent spool-out.
2. Forward-facing belt path: the belt passes through routing guides on the shell; the top tether is mandatory in most markets to control rotation.
3. Booster routing: the lap belt sits low across the pelvis and the shoulder belt crosses the sternum via a belt guide; no shell load path exists.

In a finite element model, belt routing is captured by defining the belt as a 1D seatbelt element with slip rings at each guide. The friction coefficient at guides (typically 0.1–0.3) and the retractor locking behavior are the two parameters that most influence belt tension distribution. A useful sanity check: after a 50 km/h frontal pulse, peak shoulder-belt force in a forward-facing CRS with tether typically falls in the 4–7 kN range; if the model predicts below 3 kN, the retractor lock or guide friction is likely misconfigured.

From Architecture to Model Setup

Translating the architecture into a simulation model follows a consistent sequence. First, select the CRS family and confirm the installation mode from the product label and vehicle owner's manual. Second, define the vehicle interface: for ISOFIX, extract anchor coordinates and connector stiffness; for belt-installed, extract belt anchor points, retractor type, and guide locations. Third, build the shell as a deformable structure with material properties from coupon tests or published data. Fourth, define the occupant model (Q-series or Hybrid III) and its interaction with the shell and belt.

A practical example clarifies the parameter sensitivity. Consider a forward-facing convertible installed with ISOFIX and top tether, subjected to a 50 km/h frontal pulse. If the connector spring stiffness is set to 5000 N/mm (effectively rigid), the model may predict head excursion of 480 mm. Reducing connector stiffness to a calibrated 800 N/mm with 10 mm clearance typically increases head excursion by 30–60 mm and reduces peak tether force by 15–25%. This is not a numerical artifact; it reflects the real compliance of latch hooks and anchor bars. The lesson: connector compliance is a first-order parameter, not a detail to be idealized away.

Belt Routing Details That Change Results

Belt routing is often simplified in early-stage models, but three details consistently matter:

• Lock-off devices: many rear-facing seats use a lock-off clamp that fixes the belt at the shell. Without it, the belt can spool out and the shell rotates excessively. Model the lock-off as a fixed constraint at the clamp location.
• Belt guide friction: each routing guide introduces friction. A shoulder belt passing through two guides with μ = 0.2 can lose 15–20% of its tension between the retractor and the shell. Omitting friction overestimates the restraining force.
• Belt geometry at the pelvis: for boosters, the lap belt must sit below the anterior superior iliac spine. If the model's belt path rides above the pelvis, submarining may be artificially suppressed or exaggerated. Verify belt position against the dummy's pelvis landmark before running the full simulation.

Summary

CRS architecture determines the load path, and the load path determines which simulation parameters dominate. Infant carriers and convertibles are structural systems whose shell stiffness, connector compliance, and belt routing must be calibrated. Boosters are positioning systems whose fidelity depends on belt geometry and dummy interaction. ISOFIX/LATCH connections should be modeled with realistic compliance, not as rigid constraints. Belt routing—including lock-offs, guide friction, and pelvis placement—must be represented explicitly because these details shift head excursion and belt tension by margins that matter for injury criteria. With the architecture correctly mapped, the next step is to build the finite element model and select the appropriate dummy and pulse.