Explicit dynamics FEA of a 6-ply laminated glazing assembly under pedestrian headform impact, high-velocity stone strike, and post-fracture fragment retention. HIC(d) = 834 (limit 1 000), zero fragments detached, 22% mass-efficient PVB optimum identified — with one design recommendation issued for the 120 km/h stone-strike envelope.
The client — a Tier-1 automotive glazing supplier — needed a complete impact-reliability assessment of a new 6-ply laminated windshield assembly before committing tooling. Three regulatory scenarios had to be cleared simultaneously: pedestrian headform impact (ECE R43), high-velocity stone strike (FMVSS 205, 30–120 km/h envelope), and post-fracture fragment retention across a −40 °C to +80 °C thermal range.
The laminate stack — outer glass, two PVB interlayers, inner glass, an EAS/IR functional film, and a chemically strengthened thin inner ply — totals 7.06 mm and combines four distinct constitutive behaviours: brittle fracture, viscoelasticity, cohesive delamination, and pre-stressed chemical tempering. Coupled physics, sub-millisecond timescales, and compliance-grade outputs in a single deliverable.
SIMFORGE built an explicit dynamics model in ANSYS Explicit Dynamics with a hexahedral-dominant mesh refined to 0.5 mm in the impact zone and graded to 3 mm in the far field. Glass plies use the JH-2 model with XFEM crack enrichment; PVB interlayers use a Prony series calibrated against SHPB data; the chemically tempered TIP is initialised with the residual compressive stress field as an initial condition.
Glass / interlayer interfaces are modelled with a cohesive zone (BK criterion, Gc ≈ 200 J/m²) calibrated against peel and DCB tests. LC3 is a sequential analysis — a static thermal pre-load step feeds the explicit fracture / retention step, with self-contact enabled for post-fracture fragment interaction.
Post-processing is automated through Python (ANSYS ODB API): HIC computation, delamination area tracking, fragment count, and parametric sweep aggregation are all scripted and re-runnable.
Explicit dynamics simulation campaign on a 6-ply laminated windshield with material-model calibration, parametric optimisation of the outer PVB interlayer, and full compliance reporting against two regulatory standards.
| Load cases | 3 — headform impact (LC1), stone strike (LC2), fragment retention (LC3) |
| Composite stack | 6 plies — OG · PVB outer · IG · PVB inner · EAS film · TIP |
| Total thickness | 7.06 mm nominal |
| Simulation window | 0 – 15 ms · Δt ≈ 0.05 µs (CFL-controlled) |
| Parametric study | PVB outer thickness sweep · 0.38 → 1.00 mm · 6 variants |
| Standards verified | ECE R43 · FMVSS 205 · EN ISO 3537 · EN 1796 |
| Solver | ANSYS Explicit Dynamics (Autodyn kernel) |
| Post-processing | Python (ANSYS ODB API) + custom HIC extraction script |
The six-ply stack is a progressive energy-dissipation system. Each layer performs a distinct structural or functional role — from the sacrificial outer glass to the chemically strengthened inner ply that forms the cabin-side fragment barrier.
Hexahedral-dominant mesh with XFEM crack enrichment, rate-calibrated PVB viscoelasticity, cohesive zone bonding at glass / interlayer interfaces, and pre-stress initial conditions for chemically tempered glass plies.
| Mesh parameter | Value |
|---|---|
| Glass element type | C3D8R · 3–5 layers / ply |
| Impact zone in-plane | 0.5 mm |
| Far-field in-plane | 3 mm · ratio ≤ 1.5 |
| PVB elements | 2 × C3D8R · AR ≤ 5:1 |
| XFEM enrichment radius | 30 mm |
| Total DOF | ~2.8 M |
| Δt explicit | ~0.05 µs (CFL) |
| Mass scaling zone | Far-field only · factor ≤ 10 |
| Model decision | Justification |
|---|---|
| JH-2 for glass plies | Progressive damage + residual strength post-fracture |
| Prony series PVB | Rate-dependent stiffening 10 → 700 MPa at 500 s⁻¹ |
| CZM interface | Fragment retention energy governed by Gc mode I+II |
| TIP pre-stress IC | σ_surf = −300 MPa — omitting gives 20–30 % threshold error |
| Thermal pre-load LC3 | Sequential static step before explicit analysis |
| Self-contact | Post-fracture fragment surface interaction enabled |
Three scenarios cover the regulatory envelope: pedestrian protection (LC1), durability (LC2), and post-fracture safety (LC3). Each load case has its own constitutive emphasis and output set.
LC1 headform impact at 6.5 m/s achieves full compliance with 22 % margin. LC2 stone strike passes up to 80 km/h; 120 km/h penetrates the TIP and triggers a design recommendation. LC3 fragment retention passes across the full −40 °C to +80 °C thermal range.
| Output | Value | Limit | Status |
|---|---|---|---|
| HIC(d) — LC1 · 6.5 m/s | 834 | 1 000 | PASS ✓ |
| Peak principal stress — OG inner face | 142 MPa | — | Monitored |
| Crack initiation time — OG | 0.31 ms | — | — |
| PVB peak strain — LC1 | 187 % | 250–350 % | PASS ✓ |
| Max CZM delamination area — LC1 | 410 mm² | — | — |
| Peak panel deflection — LC1 | 22.4 mm | — | — |
| Stone strike — 30 km/h | No fracture | No penetration | PASS ✓ |
| Stone strike — 80 km/h | Contained · TIP intact | No penetration | PASS ✓ |
| Stone strike — 120 km/h | TIP fracture · penetration | No penetration | FAIL ✗ |
| Fragment detachment — −40 °C | None | None | PASS ✓ |
| Fragment detachment — +80 °C | None | None | PASS ✓ |
| ECE R43 fragment count | Compliant | Compliant | PASS ✓ |
Six explicit variants sweeping outer PVB thickness from 0.38 mm (baseline) to 1.00 mm were run to quantify HIC(d) reduction versus added mass. The optimum sits at 0.76 mm: HIC(d) = 780 with only +18 % mass, giving 22 % margin to the ECE R43 limit. Beyond 0.89 mm the marginal HIC gain per gram inverts the mass-efficiency ROI.
| PVB thickness | HIC(d) | Mass penalty | Status |
|---|---|---|---|
| 0.38 mm — baseline | 1 010 | — | FAIL ✗ |
| 0.50 mm | 965 | +8 % | FAIL ✗ |
| 0.64 mm | 920 | +14 % | FAIL ✗ |
| 0.76 mm ★ optimal | 780 | +18 % | PASS ✓ |
| 0.89 mm | 730 | +25 % | PASS ✓ |
| 1.00 mm | 710 | +32 % | PASS ✓ |
Four conclusions transferred directly into the client's material model card library and laminate specification template — re-usable across the next two glazing programs.
Quasi-static PVB stiffness underestimates peak inner-glass stress by ~35 %. A Prony series calibrated at 100–500 s⁻¹ from SHPB data is mandatory for sub-1 ms impact windows. Anything else gives false-positive compliance verdicts.
CZM delamination at the OG / PVB interface is the dominant energy sink in LC1. Increasing interface Gc via adhesion-promoter treatment is more mass-efficient than adding glass thickness — a finding that re-frames future design DOEs.
Omitting the chemical-tempering residual stress (−150 to −300 MPa) underestimates the fracture threshold by 20–30 %. The result is false-conservative crack maps and incorrect compliance verdicts in opposite directions on different load cases.
The film's compliance buffers the CTE mismatch between TIP (aluminosilicate) and IG (soda-lime). Removing it from the LC3 model triples predicted delamination area at the minimum service temperature — a result with direct functional-stack design implications.
Full traceable documentation package — re-runnable, re-usable, and structured for direct integration into the client's homologation file.
| Deliverable | Format |
|---|---|
| FEA simulation report — methodology, results, compliance matrix | PDF / DOCX |
| Crack-propagation maps — all load cases | PNG · ANSYS result files |
| HIC time-history plots — LC1 | |
| Parametric sweep summary — PVB optimisation | Excel + PDF |
| Python post-processing scripts (HIC extractor, delam tracker) | .py (documented) |
| Mesh files + material library (Prony, JH-2 constants) | .inp / .dat |
Production-grade explicit dynamics workflow built around ANSYS, with custom Python automation for compliance metrics and material-card calibration.
| Tool | Usage |
|---|---|
| ANSYS Explicit Dynamics | Primary solver — Autodyn kernel |
| Python + ANSYS ODB API | HIC computation · delamination tracking · sweep aggregation |
| HyperMesh | Pre-processing — hex-dominant meshing, XFEM zone definition |
| CATIA / SpaceClaim | CAD geometry preparation, curvature mapping |
| MATLAB | Prony series calibration from DMA / SHPB data |
| DIC (Vic-3D) | PVB biaxial strain field validation data |
Material-model calibration, explicit dynamics, and compliance reporting in a single engagement. Physics-accurate Prony / JH-2 / CZM card libraries — re-usable across your next glazing programs. Fixed fee, single delivery cycle.
From laminated glazing to CFRP composites and bonded assemblies — if it needs explicit dynamics, fracture mechanics, or compliance verification against ECE, FMVSS, ISO, or internal OEM specs, we scope it in a 30-minute call and deliver a fixed-fee proposal within 48 hours.
Initial response · 24 h · NDA available · Fixed-fee proposal within 48 h