Question 5 of 8: Advanced composite materials in primary aircraft structure

Question 5: Advanced composite materials in primary aircraft structure (20 marks)

Question text not reproduced: the examination questions are © Engineers and Geoscientists BC. Open the official past paper (linked at the top of this page) to read the question, then follow the worked solution below.

(a) Three advanced composite materials used in primary structure. “Primary” means structure whose failure would be catastrophic — wing skins and spars, fuselage panels, empennage boxes, floor beams — so the three systems below are chosen because each is certified in that role on a transport aircraft in service, not merely because each is a composite.

(i) Carbon fibre in a toughened thermosetting epoxy. Intermediate-modulus carbon such as IM7 or T800 in a rubber-toughened 180 ℃-curing epoxy, laid up as unidirectional tape at 60–65 % fibre volume fraction, is the reference material for primary structure. It builds the wing box and the fuselage barrels of the current generation of widebodies. Its attraction is a specific stiffness roughly three times and a specific strength roughly four times that of 7000-series aluminium, together with the ability to place the fibre where the load runs, so a wing skin can be stiffened for bending, tailored for aeroelastic behaviour and thinned locally without a separate part.

(ii) Carbon fibre in a thermoplastic matrix. Carbon in poly-ether-ether-ketone, poly-ether-ketone-ketone or poly-phenylene sulphide is now used for ribs, clips, brackets, floor beams and control surfaces. The matrix is already fully polymerised, so there is no cure reaction, no frozen storage and no shelf life; parts can be press-formed in minutes rather than cured for hours, they can be welded to one another instead of fastened, and the tough semi-crystalline matrix gives far better resistance to impact damage and to hydraulic fluid than a thermoset. The penalty is process temperature, typically 380 ℃ for PEEK, and the tooling that goes with it.

(iii) Fibre–metal laminates. The best-known is a laminate of thin 2024-T3 aluminium sheets interleaved with unidirectional glass/epoxy plies, used for the upper fuselage skin panels of the largest current widebody. The metal layers arrest the fatigue cracks that would run freely in a monolithic aluminium skin, because a crack in one aluminium layer is bridged by intact glass fibres behind it, and the crack growth rate falls by an order of magnitude. The laminate also retains the metal’s bearing strength at fastener holes, its lightning conduction and its formability, while adding excellent burn-through resistance. It is the hybrid answer for a fatigue-critical, fastener-rich, pressurised skin.

Why these have displaced conventional materials. Five reasons carry most of the weight. First, specific properties: at equal stiffness a carbon/epoxy skin is 20–30 % lighter than aluminium, and weight compounds through the whole aircraft because a lighter structure needs less wing, less fuel and smaller engines. Second, tailorability: the laminate is designed as well as the part, so fibre can be steered and plies dropped to follow the load path, something an isotropic metal cannot offer. Third, fatigue and corrosion: carbon/epoxy has essentially no fatigue crack-growth problem at airframe stress levels and cannot corrode, which removed the cabin-humidity and cabin-altitude limits that aluminium fuselages imposed on passengers. Fourth, part-count reduction: a co-cured, integrally stiffened barrel replaces thousands of sheets, stringers, splices and fasteners, and it is the assembly labour rather than the material that dominates airframe cost. Fifth, damage tolerance in the hybrid case, where the fibre–metal laminate directly attacks the fatigue mechanism that limits an aluminium fuselage. The counterweights — acquisition cost, sensitivity to out-of-plane and impact loading, poor electrical conductivity, and repair that needs a bonded scarf rather than a riveted patch — are real, which is why aluminium, titanium and steel remain in the landing gear, the engine pylons and the highly loaded fittings.

(b) Manufacturing methods and their trade-offs. Turning fibre and resin into a part means consolidating them, and consolidation is the step that decides whether the part is fit for primary structure.

(i) Autoclave curing of pre-impregnated tape or fabric. Layers of prepreg — fibre already impregnated with a partially advanced (B-staged) epoxy at a controlled resin content — are laid onto a mould, vacuum-bagged, and cured under simultaneous vacuum, external pressure of 0.6–0.7 MPa and a programmed temperature cycle to 180 ℃. The pressure collapses entrapped air and dissolves residual volatiles back into the resin.

Its advantage is decisive and is why it remains the reference process for wing skins, spars and fuselage barrels: fibre volume fractions of 60–65 %, void contents below 1 % and tightly controlled ply orientation give the highest and, more importantly, the least scattered mechanical properties of any route, so the design allowables can be set close to the mean. The disadvantages are equally decisive. Cycle times run to six or eight hours, the autoclave is a very large pressure vessel whose capital cost and throughput limit the whole factory, the prepreg itself must be shipped and stored frozen with a finite out-time, and part size is capped by the vessel. Cost per kilogram is the highest of the four.

(ii) Liquid composite moulding: resin transfer moulding (RTM) and its vacuum-assisted variant (VARTM). A dry fibre preform, often stitched or braided to near-net shape, is placed in a closed matched-metal tool (RTM) or under a vacuum bag on a single-sided tool (VARTM), and low-viscosity resin is injected or infused through it, then cured.

The attractions are cost and geometry. Dry fabric is far cheaper than prepreg and needs no frozen storage; complex three-dimensional shapes, thick sections and integrally stiffened parts can be made in one shot, eliminating fasteners and the assembly labour that goes with them; and out-of-autoclave curing removes the bottleneck vessel entirely. VARTM in particular scales to very large parts. Against that, fibre volume fractions typically reach only 50–60 %, void content is harder to control than under autoclave pressure, and the process is sensitive to preform permeability — a dry spot or a race-tracked flow front is a rejectable defect that is not always visible. RTM tooling for a closed mould is expensive and slow to develop, so the route pays for itself only over a substantial production run.

(iii) Filament winding. Continuous tow is drawn through a resin bath and wound onto a rotating mandrel along computer-controlled helical, hoop or polar paths, then cured, and the mandrel is extracted or dissolved.

Where the geometry suits it, nothing competes: it is fast, highly automated, uses the cheapest possible form of fibre, and places each tow under controlled tension along precisely the trajectory the load demands, so pressure vessels, drive shafts, engine-casing rings and rocket motor cases achieve outstanding specific strength. Fibre volume fractions of 60 % and above are routine. The limitation is geometric and absolute: the process makes surfaces of revolution and convex shapes, cannot produce a concave surface or a flat panel with a good finish, and cannot easily place fibre in the axial direction of a small-diameter part. Mandrel extraction constrains the design, the outer surface is rough unless the part is subsequently machined or shrink-taped, and interlaminar properties depend heavily on winding tension and bandwidth control.

(iv) Automated tape laying and automated fibre placement (ATL/AFP). A numerically controlled head lays prepreg tape (ATL, wide tape on gently contoured surfaces) or several narrow slit tows independently (AFP, capable of steering and of dropping individual tows) onto the tool, compacting with a heated roller. The result is either taken to an autoclave or, with thermoplastic matrices and laser heating, consolidated in situ.

This is the route that made large composite airframes economic. It removes hand labour and its variability from the laying operation, achieves very high and repeatable deposition accuracy, supports ply drop-offs and steered fibre paths that let the laminate be tailored locally to the load field, and scales to fuselage-barrel size. Its costs are capital and time: the machines and their programming are extremely expensive and are only justified at rate, deposition slows sharply on tightly curved or highly contoured surfaces where the head must decelerate, and small or intricate parts are still made by hand. Unless in-situ consolidation is used, the autoclave bottleneck remains.