Carbon-fibre aircraft fuselage in a composites workshop

Project 08

Case study on Failure of Aircraft Composites and Alternative Manufacturing Approaches

A manufacturing case study tracing delamination, voids and fatigue failures in aircraft and wind-turbine composites back to hand lay-up and vacuum infusion, and matching each failure to prepreg, resin transfer molding or carbon-fibre 3D printing.

Project overview

From failure evidence to process selection

Documented failures in UAV structures, aircraft cabin panels and wind-turbine blade roots were traced to their manufacturing root causes, then compared against advanced manufacturing alternatives to recommend a better-suited process for each component.

Case studyComposite failure analysisPrepregRTMCF 3D printing

01 / Context

Composites and conventional processes

Composites offer a high strength-to-weight ratio, corrosion resistance and design flexibility, and are used in primary and secondary aircraft structures (fuselage, wings, fairings). Wet (hand) lay-up applies liquid resin manually over the reinforcement before curing. Vacuum infusion places dry reinforcement in a mold and draws resin through it under vacuum pressure.

Technician applying resin by hand during wet layup — Wet (hand) lay-up.

Composite part under vacuum bag with resin feed line — Composite part bagged for resin infusion.

02 / Failure cases & root causes

Three documented failures

Case 01

UAV wing skins & empennage panels

Composite UAV wing skins and tail panels

Observed failures

Premature buckling under aerodynamic loads and delamination during in-flight trials. Low compressive strength: 63.66 MPa for hand lay-up (HLU) versus 109.06 MPa for vacuum resin infusion (VRI). Low shear strength with HLU: 13.06 MPa versus 16.88 MPa for VRI.

Source: researchgate.net

Root causes

High porosity and voids from air entrapped during HLU and VRI, lowering mechanical performance. Limited fibre-volume-fraction control: aerospace parts need ~60%, while HLU and VRI reach only ~50%, reducing stiffness and strength. Excess resin content adds weight and reduces structural efficiency.

Source: ecm-academics.plymouth.ac.uk

Crashed composite UAV in rocky terrain — UAV structural failure in the field. Source: researchgate.net

Published bending, shear and impact test tables comparing manufacturing methods — Published bending, interlaminar-shear and impact data by manufacturing method. Source: ecm-academics.plymouth.ac.uk

Micrograph showing resin-rich defects in a composite laminate — Resin-rich defects in a composite laminate. Source: ecm-academics.plymouth.ac.uk

Case 02

Aircraft lavatory wall panels & overhead bins

Lavatory side-wall and ceiling panels; overhead stowage-bin covers

Observed failures

FAA-reported 16% rejection rate due to delamination, disbonding and voids. Structural cracks around fasteners, plus surface waviness and poor finish.

Source: federalregister.gov

Root causes

Manufactured by hand lay-up: manual fibre placement produced non-uniform thickness. Air entrapment from inadequate compaction.

Aircraft lavatory interior composite panels — Composite lavatory wall panels. Source: federalregister.gov

Micrograph labelling matrix, fibre and void — Matrix, fibre and void in a hand-laid laminate.

Hand-laid composite panel under vacuum bag — Hand lay-up panel with visible non-uniformity.

Case 03

Wind-turbine blade roots

Root section of large wind-turbine blades

Observed failures

40% reduction in fatigue life for hand lay-up laminates.

Source: sciencedirect.com

Root causes

Wet lay-up achieves only ~50% fibre volume fraction, reducing stiffness and strength. Entrapped air and poor compaction in hand lay-up. Early resin gelation and vacuum leaks in vacuum infusion.

Source: mdpi.com

Damaged wind-turbine blade sections — Wind-turbine blade damage. Source: sciencedirect.com

Plot of stiffness against impact number for HLU and VI laminates — Stiffness degradation with repeated impacts — HLU versus VI. Source: sciencedirect.com

Diagram of blade damage mechanisms: debonding, buckling and ply-drop delamination — Blade damage mechanisms: debonding, buckling and delamination at ply drops. Source: mdpi.com

03 / Alternative methods

Why change the process

Variability and defects in current methods affect safety and reliability. High porosity thresholds (~2%) drive high component rejection rates.

Thermographic inspection of an aircraft composite panel — Thermographic inspection for defects.

Alternative 01

Prepreg autoclave curing

Heat and pressure cure pre-impregnated fibres, giving optimal resin distribution and minimal defects. Controlled resin content for higher strength and more consistent structures. Higher fibre volume fraction leading to roughly 10–12% better stiffness.

Prepreg component in an autoclave — Prepreg component during autoclave curing.

Bar chart comparing strength of infusion and prepreg laminates — Strength of infusion versus prepreg cured at 90 °C and 120 °C (Hexcel data). Source: biesterfeld.no

Alternative 02

Resin transfer molding (RTM)

Low-viscosity resin is injected into a closed mold containing a fibre preform. High fibre volume fraction and fewer voids, increasing durability.

Robotic preform handling for RTM — Automated preform handling for RTM.

Alternative 03

Carbon-fibre 3D printing

Continuous-fibre reinforcement deposited during 3D printing. Design flexibility with high strength and reduced material wastage. This alternative mainly focuses on continuous-fibre printing; its applications for manufacturing skin composites are very limited.

3D-printed carbon-fibre composite parts — Continuous-fibre 3D-printed composite parts.

04 / Comparison & recommendations

Matching each failure to a process

Setup cost

RTMHigh — complex molds and tooling

Continuous CF printingLow to medium — printer, minimal tooling

PrepregHigh — autoclave, cold storage, tooling

Manufacturing cost

RTMMedium — efficient at moderate volume

Continuous CF printingLow for prototypes, high for batches

PrepregHigh — material and energy intensive

Labour

RTMMedium — partly automatable

Continuous CF printingLow — mostly automated

PrepregHigh — manual lay-up and bagging

Flexibility

RTMMedium — mold-dependent

Continuous CF printingHigh — complex geometry without molds

PrepregLow to medium — lay-up limited

Surface finish

RTMHigh — both mold faces

Continuous CF printingMedium — visible layer lines

PrepregHigh — smooth, especially in autoclave

Strength

RTMHigh if properly infused and cured

Continuous CF printingMedium — fibre volume and orientation dependent

PrepregVery high — aerospace grade

Figure 04-01 — Qualitative comparison. Bar lengths encode ordinal descriptions (low = 1, medium = 2, high = 3, very high = 4; intermediate values are approximate), not measured data. A longer cost or labour bar indicates more, not better.

Component	Observed failure	Recommended method	Why it works
UAV wing skins & empennage	Delamination, resin-rich zones (HLU)	Prepreg	Precise resin content; strong, light load-bearing skins
Lavatory panels & overhead bins	Surface waviness, fastener cracking (HLU)	RTM	Clean finish, mold-controlled geometry, lower rejection
C-17 landing-gear door panel	Dry spots, incomplete wet-out (VI)	Prepreg	Full impregnation, consistent quality at stiffener intersections
Wind-turbine blade roots	Voids, early fatigue failure (HLU/VI)	RTM or prepreg	RTM scales with compaction; prepreg maximises fatigue strength
Prototypes, small empennage & lavatory fittings	Delamination, resin-rich zones (HLU)	CF 3D printing	Fast prototyping, complex geometry, low tooling cost

Table 04-02 — Recommended manufacturing method for each failure case.

Return to all projects