Designing Composite Laminates
Given load case, geometry, and manufacturing constraints, produce a laminate stacking sequence with justified ply angles, verify interlaminar integrity, and confirm the process can realistically produce it.
Example minimal workflow:
Input: Panel under biaxial in-plane load (Nx=800 N/mm, Ny=200 N/mm) + out-of-plane pressure
Material: Carbon/epoxy UD, [0/±45/90] family
1. Classical Lamination Theory (CLT) -> compute [A][B][D] for candidate stack
[45/-45/0/90]2s, symmetric & balanced -> B=0
2. Failure check (Tsai-Wu or LaRC03) per ply at max load, with 1.5 safety factor
3. Interlaminar shear/normal stress via 3D FE or closed-form (e.g., Pagano) at free edges
4. Check manufacturability: AFP min steering radius, RTM permeability for ±45 stacks
Output: Stack [45/-45/0/90]2s, margin of safety 0.18 (strength), 0.22 (delamination onset),
AFP-compatible (no cut course under 50mm), RTM flow time 340s (OK, gel time 600s)
Progress:
- Step 1: Define load cases, environmental conditions (temp, moisture), and design drivers (stiffness vs strength vs weight)
- Step 2: Select fiber/matrix system and ply properties (E1, E2, G12, ν12, thicknesses, strength allowables)
- Step 3: Propose candidate stacking sequences respecting design rules
- Step 4: Compute ABD matrix via CLT; verify balance/symmetry unless intentional asymmetry needed
- Step 5: Run ply-by-ply stress/strain recovery and failure criteria (Tsai-Wu, Hashin, LaRC03, Puck)
- Step 6: Analyze interlaminar stresses (free-edge effects, delamination onset via VCCT/cohesive zone or fracture mechanics G vs Gc)
- Step 7: Iterate stacking sequence to reduce stress concentrations (ply grouping, blending, dispersed angles)
- Step 8: Evaluate manufacturability per chosen process (RTM/autoclave/AFP) — see Best Practices
- Step 9: Document final layup, margins of safety, and process parameters
Step 3 design rules (apply by default unless overridden)
- Balanced & symmetric about midplane (eliminates B matrix, avoids warping)
- No more than 4 consecutive plies of same orientation (reduces matrix microcracking/delamination)
- Minimum 10% rule: each of 0/±45/90 ≥10% of total thickness for damage tolerance
- Surface plies at ±45 to improve impact/damage resistance, avoid 0° on outer surface near bolt holes
- Disperse ply angle changes gradually through thickness near free edges to cut interlaminar stresses
- Free-edge effect: Even under pure in-plane load, mismatched Poisson/shear-coupling between adjacent plies generates interlaminar normal (σz) and shear (τxz, τyz) stresses at edges — dominant delamination initiator in many laminates.
- Methods, in order of fidelity/cost:
- Closed-form (Pipes-Pagano, Pagano's elasticity solution) — fast screening
- 2D plane-strain FE at the edge cross-section — good balance
- Full 3D FE with cohesive zone elements or VCCT — for crack growth/propagation
- Failure assessment: compare computed energy release rate G (mode I/II/III mix) against critical Gc from DCB/ENF/MMB tests using a mixed-mode criterion (e.g., Benzeggagh-Kenane).
- Report delamination onset load and margin separately from in-plane first-ply-failure margin — they often govern different regions of the design envelope.
Autoclave (prepreg)
- Any symmetric/balanced layup is feasible; main constraints are cure cycle, debulk frequency (every 4-6 plies), and tool/part thermal lag.
- Check for resin-rich/starved zones at radii (minimum bend radius ~ 6-8x ply thickness).
RTM (dry fabric + resin infusion)
- Verify fiber volume fraction achievable with chosen preform compaction.
- Check resin flow: permeability drops sharply with high ±45 content and tight weaves — run flow simulation (Darcy's law) to ensure fill time < gel time with margin ≥1.5x.
- Avoid very thick local buildups (>~8mm) without flow-assist media; risk of dry spots/voids.
AFP (Automated Fiber Placement)
- Enforce minimum steering radius (course curvature) per tow width; typical limit ~ steering radius > 635mm for 1/8" tow unless using cut/add-course strategies.
- Avoid gaps/overlaps >0.5mm; plan course boundaries away from high-stress zones.
- Minimum course length to avoid "cut course" defects; check against laminate's curved/tapered regions.
- Tack time and tow tension affect ply wrinkling on double-curvature tools — flag tight Gaussian curvature regions.
Example 1:
Input: Wingskin panel, tension-dominated Nx=1200 N/mm, autoclave prepreg, damage tolerance required
Output: Stack [45/-45/0/0/90/-45/45/0]s (16 plies, t=2.08mm), balanced/symmetric, 0° content 50% (meets load direction), ±45 at surfaces for impact resistance. First-ply failure MS=+0.31 (Hashin fiber tension). Free-edge delamination MS=+0.15 (mode I/II mixed, Benzeggagh-Kenane). Autoclave: standard debulk every 4 plies, no concerns.
Example 2: Input: Thick RTM bracket, 40 plies biaxial woven fabric, complex double-curvature Output: Flag: permeability simulation shows fill time 920s > gel time 600s at outer radius due to compaction-induced fiber volume fraction 62% (vs target 55%). Recommend: add flow-distribution medium or increase injection ports; reduce local ply count via tapering (blend 40→28 plies away from critical region) to cut flow path thickness.
Example 3: Input: AFP-manufactured fuselage frame with tight radius (R=400mm) steered 0° courses Output: Flag: required steering radius violates AFP minimum (635mm) for straight 1/8" tow — recommend switching to 1/4" slit tape with lower min radius, or re-route fiber path with cut-course segments every 150mm, verified against local strength requirement (reduced continuous fiber at cuts drops local compressive strength ~8%, still meets MS+0.05).
- Always separate in-plane (first-ply failure) margins from interlaminar/delamination margins — do not let one mask the other.
- Run manufacturability checks before finalizing the structural optimization; a structurally optimal but unmanufacturable layup wastes iteration cycles.
- Use ply blending/tapering with staggered drop-offs (minimum 1:20 taper ratio, drop no more than 2 plies per step) to avoid resin pockets and stress concentrations at thickness transitions.
- Validate CLT/FE predictions against coupon test data (tension, compression, OHC/OHT, DCB/ENF) before committing to full-scale design.
- For pressure vessels/fatigue-critical parts, always check matrix-dominated properties (90° ply cracking onset) in addition to ultimate strength.
- Ignoring free-edge interlaminar stresses because in-plane CLT margins look comfortable — delamination often governs first.
- Using non-symmetric layups without explicitly needing tailored bend-twist coupling (causes warping during cure).
- Specifying stacking sequences that violate AFP steering limits or RTM flow feasibility, discovered only at the manufacturing stage.
- Treating 10% rule and 4-ply grouping limit as optional — skipping them often causes unexpected microcracking in service.
- Assuming woven fabric and UD tape share permeability/draping behavior — always re-run process simulation per material form.