composite-manufacturing

name	composite-manufacturing
description	Composite manufacturing processes — autoclave cure (Darcy's law resin flow, cure kinetics dα/dt = A×exp(−E_a/RT)×f(α)), prepreg layup, hand layup, RTM/VARTM (resin transfer molding), filament winding (geodesic/helical/hoop angles), pultrusion, void content measurement (void% from density or acid digestion per ASTM D2584), fiber volume fraction V_f, cure cycle design (ramp/dwell), out-of-autoclave (OOA) prepregs, NDT of composites (UT C-scan), ASTM D5687, ASTM D2734, AS9100 process control.
metadata	{"priority":7,"promptSignals":{"phrases":["composite manufacturing","autoclave cure","RTM","filament winding","prepreg layup"],"minScore":3}}

Composite Manufacturing — Complete Skill

Material Forms

Prepreg

Prepreg: reinforcement fabric or UD tape pre-impregnated with B-staged resin Fiber types: carbon (AS4, IM7, T300, T800); glass (E-glass, S2-glass); aramid (Kevlar 49) Resin systems: epoxy (120°C/177°C cure), BMI (bismaleimide, 190°C), PEEK (thermoplastic, 380°C)

Prepreg storage: Out-time limit: typically 30 days at room temperature before partial cure advances (B→C stage) Frozen storage: −18°C; extends shelf life to 12–24 months; warm to RT before opening (prevent condensation)

Fiber areal weight (FAW): FAW [g/m²] = mass of fiber per unit area; typical UD prepreg: 100–200 g/m² Resin content: Wᵣ = (W_prepreg − W_fiber) / W_prepreg × 100% [typical 32–42% by weight]

Fiber volume fraction V_f: V_f = (W_f / ρ_f) / [(W_f/ρ_f) + (W_r/ρ_r)] [V_f ≈ 0.55–0.65 in autoclave; 0.45–0.55 RTM/VARTM] Rule of mixtures: E₁ = V_f × E_f + V_m × E_m [longitudinal modulus; exact] E₂ = (E_f × E_m) / (V_f × E_m + V_m × E_f) [transverse; Reuss model]

---

Cure Kinetics

Epoxy Cure Model

Degree of cure α (0 = uncured, 1 = fully cured): dα/dt = K(T) × f(α) [cure rate = rate constant × conversion function] K(T) = A × exp(−E_a/(R×T)) [Arrhenius; A = pre-exponential [s⁻¹]; E_a = activation energy [J/mol]]

Kamal autocatalytic model: dα/dt = (K₁ + K₂ × αᵐ) × (1 − α)ⁿ [autocatalytic behavior; m ≈ 0.5–1.0; n ≈ 1.5–2.0] For Hercules 3501-6 epoxy: K₁ = 2.1×10⁹ exp(−80,200/RT); K₂ = 2.0×10⁹ exp(−77,800/RT); m=0.81; n=2.74

Glass transition temperature (Tg) evolution: DiBenedetto equation: Tg(α) = Tg₀ + (Tg∞ − Tg₀) × (λ × α)/(1 − (1 − λ) × α) [Tg₀ = uncured; Tg∞ = fully cured; λ = lattice stiffening parameter ≈ 0.4–0.7] Constraint: cure temperature must be ≥ Tg(α_current) at all times to prevent vitrification before full cure

Exothermic heat generation: q_rxn = H_total × dα/dt [W/m³; H_total ≈ 300–400 J/g for epoxy; can cause exotherm for thick laminates] For thick sections (>10 mm): model heat generation to avoid exceeding T_cure_max → thermal degradation

---

Autoclave Processing

Cure Cycle Design

Typical two-dwell autoclave cure (epoxy 177°C):

1. Ramp at 1–3°C/min to T₁ = 120°C (dwell 1: 30–60 min) → allows resin flow and bag consolidation before gelation
2. Ramp at 1–3°C/min to T₂ = 177°C (dwell 2: 120 min) → full cure (α → 0.97+)
3. Cool at ≤ 3°C/min to 60°C before pressure release (prevent delamination from thermal stress)

Pressure: Applied at: before first ramp (or at gel point); prevents void growth Autoclave pressure: 700 kPa (100 psi) typical; bag differential: 85–90 kPa vacuum + autoclave pressure Compaction: applied pressure squeezes excess resin → consolidates → target V_f

Darcy's Law for resin flow: v_resin = −(K/μ) × ∇P [v = resin velocity; K = permeability [m²]; μ = resin viscosity [Pa·s]; ∇P = pressure gradient] Flow distance: L² ≈ 2 × K × ΔP × t / (μ × e) [e = porosity of prepreg stack; ΔP = pressure driving flow]

Bag assembly (vacuum bag): Peel ply → bleeder cloth → barrier film → breather cloth → vacuum bag Bleeder controls resin flow: more bleeder = lower final V_f; prepreg with controlled RC: no bleeder needed

---

Resin Transfer Molding (RTM) and VARTM

RTM (Closed Mold)

Process: Dry fiber preform placed in matched metal mold → mold closed → resin injected under pressure (100–700 kPa) Injection pressure: p_inj = (v × μ × L²)/(2 × K) [Darcy's Law for mold fill; v = resin velocity] Fill time: t_fill = (φ × L²)/(2 × K_Darcian × p_inj / μ) [φ = fiber preform porosity]

Preform design: Fiber architecture controls permeability K; measured by Darcy experiment K for woven fabric: 10⁻¹⁰–10⁻¹² m² (plain weave); lower for stitched NCF

Gate/vent locations: injection gates at lowest point → resin flows up; vents at farthest/highest points → prevent dry spots Race-tracking: resin flows faster along tool edges → premature vent → dry spot in interior → gate seal critical

VARTM (Vacuum-Assisted RTM)

VARTM: vacuum only drives resin (no matched mold pressure); one-sided tooling; cheaper for large parts Driving ΔP: 85–90 kPa (limited by atmospheric pressure) Distribution medium: high-K flow media (4×10⁻⁸ m²) placed on top to distribute resin across surface; then infuses through thickness V_f: lower than autoclave (0.45–0.55) due to lower compaction pressure; acceptable for marine, wind, infrastructure

SCRIMP (Seemann Composites Resin Infusion Molding Process): trade name; distribution media variant

---

Filament Winding

Winding Angles

Geodesic path on cylinder: Clairaut's law: r × sinα = constant [r = radius from axis; α = winding angle from axis] For cylinder (constant r): α = constant along cylinder → pure helical winding

Netting analysis for pressure vessel: Hoop stress: σ_θ = p × r/t [p = internal pressure; r = radius; t = wall thickness] Axial stress: σ_z = p × r/(2t) Optimal winding angle for biaxial load (cylinder): α_opt = arctan(√(σ_θ/σ_z)) = arctan(√2) = 54.7° (from vessel axis) Pure hoop (90°): only σ_θ carried; axial carried by liner or separate layers Helical (54.7°): balanced biaxial Low angle (15–30°): high axial strength; used for end dome (polar opening region)

Fiber volume fraction in winding: V_f = n_t × tex × cos(α) / (ρ_f × t × w_band) [n_t = tows/band; tex = linear density [g/km]; ρ_f = fiber density] Compaction: controlled by winding tension T [N]; resin squeeze-out → V_f

---

Void Content

Measurement Methods

Theoretical density method (ASTM D2734): ρ_theoretical = 1/[w_f/ρ_f + w_m/ρ_m] [w_f, w_m = weight fractions; ρ_f, ρ_m = densities] V_void = (ρ_theoretical − ρ_measured) / ρ_theoretical × 100% [%] Measure ρ_measured by water displacement (Archimedes)

Acid digestion (ASTM D3171): Dissolve matrix with H₂SO₄ + H₂O₂ or 30% H₂O₂ at 60°C → weigh remaining fiber → calculate V_f, V_m, V_void Most accurate; destructive

Acceptance: Aerospace structures: V_void ≤ 1% (high performance); ≤ 2% (general aircraft) Marine/infrastructure: V_void ≤ 4% (VARTM structures)

---

Out-of-Autoclave (OOA) Prepregs

OOA prepreg design: Partially impregnated (PI) prepreg: dry regions allow vacuum channels for air evacuation during cure Cure in convection oven at 120–135°C under vacuum bag only (no autoclave pressure) V_void: 1–2% achievable; approach autoclave quality Cost: OOA processing 50–70% lower cost than autoclave for large parts

VBO (Vacuum Bag Only) process: same as OOA; emphasizes bag-only pressurization

---

Non-Destructive Testing of Composites

Ultrasonic C-scan (immersion or through-transmission): Void content: void scatters/absorbs US → signal attenuation map → compare to reference panel Delaminations, disbonds: echo from interface Frequency: 1–10 MHz (higher f = better resolution but lower penetration) Accept: attenuation ≤ 6 dB from reference good-area [ASTM C1533; SAC from SAM or immersion UT]

Thermography: Pulse thermography: flash lamp → transient thermal response → delaminations appear as hot/cold spots ASTM E2582; good for large area survey; less precise depth estimation

---

Standards

Standard	Scope
ASTM D2584	Ignition loss / fiber content (glass composites)
ASTM D3171	Constituents of composite material (acid digestion)
ASTM D2734	Void content
ASTM D5687	Standard guide for composite panel fabrication
ASTM D7136	Drop weight impact (composite panels)
AMS 2766	Heat treatment of composites
NASM 1515	Fastener requirements for composites
AS9100 Rev D	Aerospace QMS (process control for composite manufacturing)
SAE ARP5765	Composite repair guidance
ISO 1268	Filament winding process

Output

Provide: material system (fiber: type, FAW [g/m²]; resin: type, T_cure [°C], T_g [°C]; prepreg or dry+RTM), layup plan (ply count; orientations [0/±45/90]; stacking sequence; symmetric/balanced?; total cured thickness [mm]), cure cycle (ramp rates [°C/min]; dwells: T₁ [°C] × t₁ [min]; T₂ [°C] × t₂ [min]; pressure application timing; cool rate [°C/min]), process method (autoclave: P [kPa]; RTM: injection P [kPa], fill time [min]; VARTM: K_media, flow length [m], fill time [min]; filament winding: angle [°], tension [N], winding pattern), quality targets (V_f target [%]; V_void target [%] ≤ 1% for aero; cure state α ≥ 0.97), verification (density test per ASTM D2734 or acid digestion D3171; UT C-scan attenuation limit; Tg measurement by DSC: target Tg [°C]), mechanical properties (E₁ from ROM [GPa]; laminate properties from CLT; strength per CMH-17), and applicable standard (ASTM D5687 panel; AS9100D for process control; AMS spec for prepreg material).