Mechanical and durability testing of aerospace materials
The selection of materials for aircraft structures and engines is assessed according to a multitude of parameters such as cost, ease of manufacture, weight and a host of other factors. Central to the selection of materials is their mechanical properties such as stiffness, strength, fatigue resistance and creep performance. The durability properties of structural and engine materials in the aviation environment is also critical to their selection. Metals must be resistant to corrosion and oxidation whereas fibre–polymer composites must resist absorbing an excessive amount of moisture from the atmosphere. Aerospace materials should be durable enough to resist degradation and damage over the design life of the structural component, which may range from several hours for rocket engine parts to longer than thirty years for airframes.
Materials are selected by matching their properties to the design specifications and service conditions of the aircraft structure or engine component. The preliminary design of new structures or engines requires the aerospace engineer to analyse the performance requirements for the materials. Key information on the property requirements for the materials is determined early in the design process. For example, in the design of an aircraft wing, the minimum stiffness, strength and toughness properties needed by materials used in the spars, stringers, skins and other load-bearing components must be known. The environmental conditions in which the materials operate must also determined during the design process. For example, materials used in satellites and spacecraft must be built using materials that have high mechanical properties at extremely low temperatures and are not damaged by ionising radiation, micro-meteor impact or the low pressure conditions of the space environment.
The aerospace engineer must understand the mechanical and durability properties of materials to ensure they function over the design life of the aircraft component without the need for excessive maintenance and repair. Unfortunately, many of the mechanical and durability properties of materials cannot be calculated using mathematical models and therefore must be measured. For example, it is not possible to calculate the strength and hardness of metals or the fracture toughness and fatigue life of fibre–polymer composite materials. Likewise, the corrosion resistance of metals or the durability of composites in hot and moist environments cannot be calculated. No theory exists to determine most of the mechanical or durability properties of metals. Calculating the properties of metals is too difficult because they are dependent on too many factors, such as their alloy content, crystal structure, microstructure, heat treatment and processing conditions. Similarly, the complexity of the microstructure, residual stress state and damage modes of composite materials make it difficult to calculate many of their mechanical and durability properties. It is possible to calculate the elastic properties of composites using theoretical models, but many other important properties, strength, fracture toughness, fatigue, creep, and so on, cannot be accurately calculated. Because many of the mechanical and durability properties of metals and composites cannot be calculated, they must be measured under standardised, controlled test conditions.
The testing procedures used to measure the properties of materials are performed under conditions specified by standards organisations, such as the American Society for Testing and Materials (ASTM) or the International Organization for Standardization (ISO). The aerospace industry uses these standards and, in some cases, uses their own specialist test procedures when a standardised method does not exist.
The aim of this chapter is to introduce the mechanical and durability properties of materials and describe the tests used to measure the properties. The main mechanical properties for aerospace materials, including the elastic modulus, yield strength, ultimate strength, ductility, hardness and fracture toughness are described as well as the test methods used to measure these properties. The methods examined include the tension, compression, flexure, fatigue, hardness, fracture toughness and creep tests. Several test methods used to measure the durability properties of materials are introduced. Tests to measure the corrosion resistance of metals and the moisture resistance of fibre–polymer composites are described. Towards the end of the chapter we examine how test data on the mechanical and durability properties of materials is used for certification of new aircraft and the structural modification of existing aircraft. The information given in this chapter provides an understanding of the engineering properties of materials; how the properties are measured; and how the property data is used in aircraft certification.
The tension test is one of the most common and important methods for measuring the mechanical properties of materials. The tension test is popular because a large number of properties can be determined in a single test: elastic modulus, strength, ductility, and other properties. The test is also popular because it is simple, quick and inexpensive. An important aspect of the tension test is that it is fully standardised. The test is performed under a controlled set of conditions using standardised equipment which ensures reliable results that are consistent and repeatable when measured anywhere in the world.
The tension test measures the resistance of a material to a slowly applied pulling force. Figure 5.1 shows the design and equipment for the test. A coupon specimen made of the test material is held at the two ends by the grips of the tensile machine. A tension (pulling) force is applied to the specimen by holding one end firm and forcing the other end away. A strain gauge or extensometer is attached to the specimen to measure the material extension under increasing force. The reaction of the material to the pulling force until it breaks is recorded and analysed. This data is used to quantify how the material responds to tensile forces applied in practical situations such as airframes and engines.
Tension specimens can be flat, round or dog-bone in shape, and are usually in the size range of 5–20 mm wide and 100–200 mm long. The flat specimens are either continuous or contain an open centre hole, as shown in Fig. 5.2. Specimens without the hole are used to measure the tensile properties of materials. Specimens containing an open hole are used by the aerospace industry to determine the tensile properties of materials affected by geometric stress raisers, such as fastener holes, which reduce the tensile strength.
During the tension test, the reaction of the material to an increasing applied load is measured until the sample breaks. The machine records the extension of the specimen with increasing load, and this is plotted as the applied force (or load)–extension curve. Figure 5.3 shows force–extension curves for aluminium measured using specimens with different sizes and shapes. As expected, the larger the specimen the greater the force that must be applied to elongate and deform the material. The force-extension curve has limited use in engineering design because its data is dependent on the dimensions and geometry of the test piece. This means the force–displacement curve measured for a material cannot be used to understand the tension properties of the same material used in a smaller component (e.g. aircraft fastener) or larger component (e.g. wing or fuselage) with a different shape.
To overcome this problem, the tensile load is converted into tensile stress (force per unit cross-sectional area) and the extension is converted into strain (or percent elongation). Expressed mathematically, the tension stress (σ) is calculated using:
where P is the applied force and A is the load-bearing area of the test specimen (see Fig. 5.4).
Strain can be expressed in two ways: engineering strain and true strain. Engineering strain is the easiest and most common expression of strain, and is the ratio of the change in specimen length to the original length:
where L is the specimen length under load and Lo is the original length (before loading). The true strain is calculated using the instantaneous length of the specimen as the tension test progresses to failure, and is calculated using:
The benefit of converting a force–displacement curve to a stress–strain curve is that direct comparisons can be made on test results from specimens of different sizes and shapes. The stress–strain curve is independent of the specimen size and geometry, and its data can be used to determine the tensile reaction of the material used in any situation. For instance, the same stress–strain curve can be used to assess the tensile properties of a tiny fastener or large wing panel made using the same material.
Tensile stress–strain curves that typify brittle materials (e.g. carbon-fibre composites and ceramics) and ductile materials (e.g. most metals and polymers) are presented in Fig. 5.5. Actual graphs for several aerospace structural materials are given in Fig. 5.6. The curve for a brittle material shows a linear (elastic) relationship between stress and strain to the failure point. The curve for a ductile material can be divided into the linear (elastic) and the nonlinear (plastic) regimes. The initial linear portion of the stress–strain curve defines the elastic regime. Within the elastic regime the material stretches when stress is applied, and then relaxes back to its original shape when the stress is removed. The material does not experience any permanent deformation or damage when loaded within the elastic regime for a short period of time. The plastic regime covers the nonlinear section of the graph between the elastic regime and point of final failure. The material in the plastic regime is permanently deformed under the applied stress, which causes a nonrecoverable change in shape when the stress is removed. When the material is deformed too much it breaks.
The stress applied to materials used in aerospace structures must always remain within the elastic regime to ensure they are not permanently deformed. All brittle and ductile solids are linear elastic at low strain (typically less than 0.1%), but at higher strains brittle materials suddenly break whereas ductile materials plastically deform. A few solids, such as rubber, are elastic up to very high strains (of the order 500–1000%), although such materials are not used in aircraft structures for other reasons (such as low strength and creep resistance).
The modulus of elasticity or Young’s modulus is the measure of stiffness, and is one of the most important engineering properties for aircraft structural materials. The elastic modulus defines how much a material stretches under an applied tension stress. The greater the modulus, the less the material elastically deforms under the application of a given stress. For instance, materials with low elastic modulus, such as rubber or plastic, are more flexible than higher modulus materials, such as steel or titanium, under the same applied stress.
There is a requirement for aircraft structures and engine components to have high stiffness to resist excessive deformation under load. Therefore, materials with high elastic modulus are used, such as aluminium, titanium, steel and carbon fibre–epoxy composite. Occasionally, there is a requirement to use low modulus materials in aircraft, such as rubber seals for doors and other openings, but they are not required to carry high loads.
The Young’s modulus (E) is calculated from the gradient of the straight line region of the stress–strain curve. In this rectilinear region, the material obeys the relationship defined as Hooke’s law, where the ratio of the applied stress (σ) to strain (ε) is constant:
The elastic modulus of metals is closely related to the binding energy between their atoms. The binding energy describes the magnitude of the attraction force between atoms. The elastic modulus of metals increases with the binding energy. A steep slope to the linear region of the stress–strain curve indicates that a high force is required to separate the atoms and cause the material to stretch, thereby resulting in high elastic modulus. The binding energy between atoms is constant and cannot be changed, and therefore the elastic modulus of a metal is constant. The elastic modulus of metals is one of very few properties not sensitive to the microstructure. This is because the modulus is determined by the strength of atomic bonds, and not by microstructural features such as dislocations or grain structure. The elastic modulus is not changed significantly by heat treatment, cold working, grain size and other microstructural features that have a large influence on other tensile properties such as strength and ductility. The elastic modulus can be changed by the addition of alloying elements, although their concentration must usually be high to have a noticeable effect. The elastic modulus of structural polymers and fibre–polymer composites is determined by other factors, which are described in chapter 13 and 15, respectively.
The elastic modulus for engineering materials is shown in Fig. 5.7; they range over six orders of magnitude from around 0.001 to 1000 GPa. The main types of aircraft structural materials have elastic modulus of 40–400 GPa: magnesium (45 GPa), aluminium (72 GPa), titanium (110 GPa), steel (210 GPa) and carbon–epoxy composite (anywhere from about 70 to over 300 GPa depending on the type, volume content and orientation of the fibres).
Poisson’s ratio defines the amount of lateral contraction versus the amount of axial elongation experienced by a material under the action of an applied elastic load. The Poisson effect is shown in Fig. 5.8. The Poisson ratio v is calculated using:
where εx and εz are the elastic strains in the lateral and longitudinal directions at the same tensile stress. The Poisson’s ratio for most metals is about 0.3 and for carbon fibre–epoxy composites is typically 0.2–0.3. Poisson’s ratio is an important engineering property for aerospace materials because of the need for close tolerances in aircraft structures and engines. For example, a material having a high Poisson’s ratio (i.e. v → 1) used in an engine turbine blade contracts laterally by an excessive amount when under load, resulting in a large loss in propulsion efficiency.
All materials have an elastic limit beyond which something happens and its original shape can never be recovered. The stress–strain curve is rectilinear over the entire strain range up to the elastic limit, which is also the failure strain limit for brittle materials. Ductile materials behave differently beyond the elastic limit; they permanently change shape owing to plastic deformation. The point when a material changes from elastic to plastic behaviour is called the yield point (also known as the proportional limit). Before the yield point, a material returns to its original shape when the load is removed. After the yield point, the material undergoes permanent plastic deformation, and when the load is removed it cannot relax back into the original shape.
The yield point is the point on the stress–strain curve where it is no longer rectilinear. The yield strength is defined at this point as the amount of stress needed to start plastic deformation. In metallic materials, the yield strength is determined by the amount of stress needed to initiate dislocation slip. Dislocations cannot move when the stress is below the yield point, and begin to move when the applied stress exceeds the yield stress. Metals that contain dislocations that can move easily have low yield strength whereas high-strength metals contain dislocations that are highly resistant to slip. (The process of dislocation slip was described in the previous chapter). In polymeric materials the yield strength is determined by the stress needed to begin permanent disentanglement and sliding of the polymer chains. (The behaviour of polymers under tensile loading is explained in Chapter 13).
Figure 5.5 shows the tensile stress–strain curve for a ductile material with well-defined yield strength owing to an abrupt change from elastic to plastic deformation. The occurrence of a sharp yield point depends on the sudden increase in the number of mobile dislocations. However, not all stress–strain curves show an abrupt change from elastic to plastic behaviour. The curve for many metals and polymers shows a gradual transition from the elastic to plastic regimes, as shown in Fig. 5.9, and the exact point where permanent deformation begins is hard to locate. The yield strength of materials which display this behaviour is determined from the curve using the offset method. The method involves specifying an offset percentage strain, which is usually 0.2% for metals. A line starting from the offset value is drawn parallel to the linear (elastic) portion of the curve. The stress corresponding to the intersection of this line with the stress–strain curve is the offset yield strength or (more commonly called) the proof strength. The proof strength is used to define the stress required to cause plastic deformation to materials which lack a sharp yield point on their stress–strain curve.
Figure 5.10 shows yield and proof strength for many types of materials. Similar to the Young’s modulus, the yield strength ranges over nearly six orders of magnitude (from 0.1 to 10 000 MPa). Unlike the modulus, however, the yield strength of metals is sensitive to the alloy content, type of heat treatment, amount of cold working, grain size and other microstructural features. By appropriate alloying and processing it is possible to greatly increase the yield strength of metals. The strengthening of metals is critical to the design of weight-efficient aircraft structures (i.e. high strength-to- weight ratio). Without the ability to increase the yield strength of metals by alloying, heat-treatment, work hardening and other strengthening processes then aircraft structures would be heavy and bulky because of the need to increase their load-bearing section thickness.
Yield strength is an important property in the design of aerospace structures. It is essential that aerospace materials are not subject to stress levels exceeding their yield point, otherwise the structure permanently deforms. The yield strength is used in aerospace design to define the upper stress limit of the material (called the design limit load). However, aerospace structures are always designed to operate at stresses well below the design limit load to avoid permanent damage caused by unexpected overloading of the airframe. Safety factors are used in the calculation of the maximum operating stress and strain limits on materials to avoid overloading and damage.
The stress–strain curve for ductile materials increases between the yield point and the point of maximum stress. This maximum point is the ultimate tensile strength (UTS, more often called the tensile strength) of the material. The UTS is the maximum stress the specimen can sustain during the tension test. The tensile strength is the parameter most often quoted from the results of a tension test; although it is of little importance in engineering. Aerospace structures are never designed using materials that are loaded to their ultimate tensile strength; otherwise there is a high risk of failure. The yield strength is a more meaningful property than the ultimate strength for determining the maximum stress loading on an aircraft structure.
During tensile testing, the deformation of a ductile specimen is uniform along its length up to the ultimate tensile stress. When loaded beyond this point, the deformation becomes nonuniform. At some point along the specimen, one region plastically deforms (or stretches) more rapidly than the other regions, usually at a local imperfection (e.g. void or second-phase precipitate). There is a local decrease in the cross-sectional area in this region, which develops into a neck (Fig. 5.11). When the specimen is subjected to increasing strain beyond the point of ultimate tensile stress the cross-sectional area of the necked region becomes progressively thinner. A lower force is then required to continue the deformation, and therefore the engineering stress–strain curve decreases until the failure point.
Tensile failure of ductile materials is a complex process, and often involves the formation of submicrometre-sized cavities within the neck region when stretched beyond the point of ultimate tensile strength. The number and size of the cavities increases with the amount of strain, and the cavities eventually link-up into a single crack which causes the specimen to break (Fig. 5.12). The amount of extension experienced by the material up to the point of failure is called the elongation-to-failure, and this property is used to define the ductility of the material. The elongation-to-failure for the specimen is taken from the stress–strain curve at the point of maximum strain.
There are two types of tensile stress–strain curves for ductile materials: the engineering stress curve and the true stress curve. A comparison of the two curves for the same material is shown in Fig. 5.13. The engineering stressstrain graph is based entirely on the original dimensions of the specimen. It is assumed that the load-bearing area of the specimen does not change during testing, and therefore the tensile stress is calculated using equation [5.1] by dividing the applied force P by the original load-bearing area Ao. In reality, the load-bearing area reduces continuously over the course of the test. The area becomes progressively smaller as the material is stretched under increasing load to the ultimate tensile stress. The loaded area then reduces rapidly between the points of ultimate stress and final failure owing to necking. The true stress–strain curve takes into consideration the reduction in the load-bearing area. The true stress is determined (again using equation 5.1) by dividing the force P by the actual (or true) cross-sectional area of the specimen Ai. When the stress is calculated in this way, the true curve rises continuously up to final failure owing to strain hardening of the material. Both engineering and true stress–strain curves are important in determining the tensile properties of aerospace materials.
The compression test determines the mechanical properties of materials under crushing loads. There are many aircraft structures that carry compression loads, such as the undercarriage during take-off and landing or the upper wing surface during flight, and therefore the mechanical behaviour of their materials must be determined by compression testing. It is often assumed that the tension and compression properties of materials are the same, yet this is only true for the elastic modulus. Other important mechanical properties, such as yield strength, may be different under tension and compression. It is essential that the mechanical properties of materials used in compression-loaded aerospace components are measured under compression loading, rather than assuming they are identical to the tensile properties.
The compression test involves squashing a small material specimen under increasing load until the point of mechanical instability. Compression specimens are usually short, stubby rods with a short length L to wide diameter D ratio (usually L/D < 2) to prevent buckling and shearing modes of deformation. The reaction of the specimen against an increasing compression load is measured during the test and, from this, the compression stress–strain curve is determined. The compression curve is simply the reverse of the tension curve at small strains within the elastic regime because the elastic modulus is the same. As the strain increases the difference between the compression and tension curves often becomes significant. As the specimen is squashed, it becomes shorter and fatter, and the load needed to keep it deforming rises. The specimen does not experience necking and so the stress increases until an instability occurs, such as cracking. The main mechanical properties which are determined from the compression test are the compression modulus and compression yield strength.
The flexure test measures the mechanical properties of materials when subjected to bending load. A flat rectangular specimen is loaded at three or four points, as shown in Fig. 5.14. The load causes the specimen to flex, thus inducing a compressive strain on the concave side, tensile strain on the convex side, and shear along the mid-plane. The separation distance between the support points must be sufficiently large to avoid the generation of high shear stress. This is achieved by setting the ratio of the length of the outer span L to the specimen height h to 16:1 or greater. This ensures failure occurs by tension or compression failure at the surface, rather than by shear.
During the flexure test, the reaction of the specimen to increasing force is measured to calculate the bending stress–strain curve. The flexure strain is related to the vertical displacement of the specimen (δ) by:
Flexure testing can determine several important mechanical properties, including the flexural modulus, yield stress and breaking stress. Flexure tests are often done on brittle materials and fibre–polymer composites, but only rarely on ductile materials such as metals.
Simply stated, hardness is the resistance of a material to permanent indentation. Hardness is not a precisely defined engineering property, such as elastic modulus or yield strength, but it is still widely used to describe the resistance of materials to plastic deformation. The hardness of ductile materials is related to their yield strength, and the hardness test (which is simpler, faster and less destructive to perform than the tensile test) is often used to obtain a measure of strength.
Hardness is measured by pressing a hard indenter into the surface of the test material under a specific force (usually about 2 N), as shown in Fig. 5.15. The further the indenter sinks into the material, the softer is the material. The depth or size of the indentation left on the material surface after the indenter is removed is used to determine the hardness. The indentation should be large enough to obtain a bulk measurement of the hardness but small enough that it does not damage the surface finish or act as a stress raiser. The indentations produced by the hardness test are usually 0.1–1 mm wide and less than 0.5 mm deep.
Several methods are used to measure the hardness of metals, and the most common are the Brinell, Vickers and Rockwell tests. These tests are popular because they are quick, easy, inexpensive, nondestructive and can be performed on components of virtually any shape and size. However, hardness is not an intrinsic material property and cannot be used to describe the structural behaviour of a material under load. Instead, hardness numbers are used mainly as a quantitative basis for comparison between materials and to determine the consistency of the same material when produced in batches.
Figure 5.16 shows a Brinell test machine and the hardness indentation. The test involves pressing a hardened steel ball into the test material. The ball is 1, 5 or 10 mm in diameter and is pressed into the surface with an applied load of 30, 300 or 750 kgf. The ball size and load used increase with the hardness of the material. The ball is pressed into the material for a fixed period of time, usually 30 s, and then removed. The diameter of the indentation is measured using a microscope and the Brinell hardness number (HB) is calculated from the expression:
where F is the applied force (in kg), D is the diameter of the indenter (mm), and d is the diameter of the indentation (mm). The Brinell hardness number has the units of stress (Pa). Although hardness testing should only be used for quantitative comparisons between materials, the Brinell hardness number is closely related to the tensile yield strength of many metals by the simple relationship:
The Vickers hardness test works on a similar principle to the Brinell test, but with the key difference being that the indenter is a square-based pyramidal diamond rather than a hard ball (Fig. 5.17). The pyramid indenter is pressed into the material for a fixed time (usually 10–15 s), and the size of the two diagonals (d1 and d2) in the indentation are used to calculate the hardness. The hardness is reported as the Vickers hardness number (VHN).
The Rockwell hardness test uses a small diameter steel ball indenter for soft materials and a diamond cone indenter for hard materials. The indenter is pressed into the material for a fixed time and its penetration depth is measured automatically by the Rockwell test machine and converted into the Rockwell hardness number (HR).
Fracture toughness is an engineering property that defines the resistance of a material against cracking. Tough materials require large amounts of energy to crack whereas low toughness materials have little resistance against cracking. For the materials used in aircraft structures, fracture toughness is just as important as other mechanical properties such as elastic modulus and strength. Aerospace materials need high toughness to resist the growth of cracks initiating at damage sites (e.g. corrosion pits, impacted regions) or sites of high stress concentration (e.g. fastener holes, windows, doors and other access points in the aircraft).
There are several test methods to measure fracture toughness, and the most common is the single-edge notch bend (SENB) test. The SENB specimen is a rectangular block of the test material containing a machined notch with a V-shaped tip (Fig. 5.18). The notch extends about 50% through the specimen. The SENB test involves applying a three-point load to the specimen to generate a tensile stress at the notch tip. The applied load required to grow a crack from the notch tip through the specimen is used to calculate the fracture toughness. Tough materials require a high load to cause complete fracture of the specimen. Another popular fracture test is the compact tension (CT) method, which involves tension loading a block-shaped specimen containing a sharp notch (Fig. 5.19). The load required to break the specimen is used to calculate the fracture toughness.
The toughness values for materials vary over a wide range (five orders of magnitude), from the very tough to extremely brittle. Tough metals have a fracture energy of 100 kJ m−2 or more, whereas those of weak brittle materials are under 0.01 kJ m−2. Most high-strength alloys, including those used in aircraft structures, have moderately high toughness (20–100 kJ m−2). Fibre–polymer composites have anisotropic toughness properties because of their microstructure, and the highest toughness (10–30 kJ m−2) is when the direction of crack growth is perpendicular to the fibre orientation. The fracture toughness properties of aerospace materials are fully explained in chapter 18 and 19.
The toughness properties of materials are also measured using impact tests. These tests involve measuring the energy required to fracture the material specimen when impacted at high velocity by a heavy object. The impact energy absorbed by the specimen is related to its toughness; tough materials require higher impact energies than brittle materials. The most popular methods are the Charpy and Izod impact tests. Both tests involve striking the specimen containing a V-notch with a pendulum travelling at a set speed. Impact loading of the specimen produces much higher strain rates than those generated in the SENB and CT fracture toughness tests. Therefore, the Charpy and Izod tests are useful for determining the dynamic toughness of materials under high loading rates, such as those experienced during an impact event (e.g. bird strike). However, the tests are qualitative; the result can only be used to compare the relative impact toughness properties between materials. Impact test results cannot be used to calculate the intrinsic fracture toughness of the material, although the test is used as a simple, accurate and inexpensive method for screening materials for toughness.
Fibre–polymer composites are susceptible to damage from impact events such as bird strike, dropped tools during aircraft maintenance, tarmac debris kicked-up by the wheels during take-off or landing, and large hail stones. Impact testing is performed at different impact energy levels to screen composite materials for damage resistance and damage tolerance. The most common and simplest test for measuring impact resistance is the falling- weight test, as shown in Fig. 5.20. A flat specimen panel of the composite material is impacted perpendicular to the surface by a hard object, usually a hemispherical steel tub. The weight is dropped from a height of 1–2 m with a mass of 4.5–9 kg to replicate a low-velocity impact event on an aircraft. The amount of damage caused by the impact event is measured (often by a nondestructive inspection method such as ultrasonics or radiography) to determine the damage resistance. The residual mechanical properties, such as compression strength or fatigue life, may also be measured to determine the damage tolerance. Civil aviation authorities specify no reduction to the mechanical properties following an impact at an energy level in the range of 35–50 J whereas the USAF specifies a minimum energy level of 135 J.
Other impact tests are often included at higher energy levels to fully characterise the impact resistance of aircraft composite components. These may include ballistic impact for critical military structures (e.g. main rotor blades), ice-hail simulation and bird strike simulation.
Fatigue tests measure the resistance of materials to damage, strength loss and failure under the repeated application of load. Aerospace materials must withstand repeated loading for long periods of time, which is in the order of 15 000–20 000 flight hours for modern jet engine materials and anywhere from 80 000 to 120 000 h for airframe materials. Materials can be damaged by repeated loading, which causes a loss in strength and eventually leads to complete failure. Fatigue tests are performed to measure the reduction in stiffness and strength of materials under repeated loading and to determine the total number of load cycles to failure.
Fatigue tests are performed by repeated tension–tension, compression– compression, tension-compression or other combinations of cyclic loading. The fatigue stress is applied repeatedly to the specimen using a variety of load waveforms, as shown in Fig. 5.21. The shape of the loading wave is usually sinusoidal, although triangular and block loads are also used. The frequency of the load cycles is generally low, typically 1–20 Hz (i.e. load cycles per second), to avoid heating of the specimen, which can affect the fatigue results.
The basic method of determining the fatigue resistance of materials is the fatigue life (S-N) curve. This curve is a plot of the maximum fatigue stress S against the number of load cycles-to-failure of the material N. The curve is plotted with the fatigue stress as a linear scale and load cycles-to-failure as a log scale. Data for the curve is generated by fatigue testing specimens at different fatigue stress levels to measure the number of load cycles-to- failure. Examples of fatigue life curves for aluminium alloy and carbon-epoxy composite are presented in Fig. 5.22. A fatigue test can be stopped after a specific number of load cycles and then the specimen is loaded to fail to measure the residual fatigue strength. Chapter 20 provides a full description of the fatigue properties of aerospace materials.
Creep is a plastic deformation process that occurs when materials are subjected to elastic loading for a long period of time, often at high temperature. Engineering materials do not plastically deform when loaded within the elastic regime for short times. However, when the elastic load is applied for a sufficient period the material eventually deforms plastically by creep. The amount of creep experienced by structural and engine materials used in aircraft is negligible at room temperature. However, the creep rate increases rapidly with temperature (usually above 40% of the absolute melting temperature for metals) and therefore creep is an important engineering property for materials required to operate at high temperature, such as jet and rocket engine materials.
The creep properties of materials are determined using the creep test. The test involves measuring the elongation of a material specimen while under a constant applied tensile stress at high temperature (Fig. 5.23). The specimen is heated inside a thermostatically controlled furnace attached to a creep testing machine that applies the tensile load. The level of applied stress is below the yield strength of the material. The elongation of the specimen over time is measured using an extensometer or other device capable of measuring the strain. Creep tests are usually performed over long periods, with the loading times typically ranging from 1000 to 10 000 h. The extension of the specimen must be measured using a very sensitive device because the actual amount of deformation before failure may be less than a few percent. Ideally, the test should be performed under conditions that replicate, as close as practical, the temperature, stress and time-scale of the aircraft material when used in service. For example, creep tests performed on materials for engine turbine blades should be performed close to the operating stress and temperature of the blades, which is about 180 MPa and 1200 °C.
The test results are plotted as strain against time to give a creep graph, as shown in Fig. 5.24. The curve for most materials can be divided into three regions: primary creep, secondary creep, and tertiary creep. A series of creep tests performed at different stress levels and temperatures are performed to obtain a complete assessment of the creep properties for a material. The creep properties of aerospace metals and composites are described in chapter 22.
One of the most damaging environmental effects of aerospace metals is corrosion. Corrosion of the metal alloys used in aircraft structures and engines occurs in many forms, including general corrosion, stress corrosion, pitting corrosion, crevice corrosion and exfoliation corrosion. Chapter 21 describes the corrosion properties of metals. There is no single, universal corrosion test; instead there are a large number of tests used to measure the resistance of metals to different types of corrosion. The tests are often performed under extreme corrosion conditions to accelerate the corrosion rate and thereby determine the long-term corrosion performance of metals.
Most tests have been designed to measure (either quantitatively or qualitatively) the resistance of metals to a specific type of corrosion or corrosive environment. For example, tests are available to determine the resistance of metals to pitting corrosion whereas other tests are used to measure the resistance of metals to crevice corrosion. Aerospace companies may sometimes use their own corrosion test methods, such as exposing structural metals to aviation fuels, lubricants or paint strippers that may contain potentially corrosive chemicals. Jet engine manufacturers perform specialist corrosion tests of the materials used in turbine blades and discs that involve hot combustion gases produced by aviation fuel.
The selection of corrosion tests for metals used in aircraft should be based on the types of corrosion and corrosive environments most likely to be experienced in service. An example of a test used to measure the corrosion rate of aircraft metals is the salt spray test. This test involves exposing metal specimens to a dense saline fog within a closed chamber. The fog is produced by producing a fine mist of salted solution (usually water containing sodium chloride to replicate seawater). The appearance of corrosion products and damage on the specimen is evaluated after a period of time. Mechanical tests can be performed on the specimen after testing to determine whether the corrosion affected the strength and fatigue properties.
The greatest environmental problem with fibre–polymer composite materials used in aircraft structures is moisture absorption. Water moisture in the atmosphere (humidity) is absorbed into the polymer matrix causing swelling, damage and softening of the material under extreme conditions. Moisture absorption can also reduce the maximum operating temperature of a composite, sometimes by 20 °C or more. Composites typically absorb anywhere from 0.5 to 2% of their own weight in water, and this resides as water molecules in the polymer matrix and along the fibre–polymer interface. The moisture content is often higher in sandwich composites because the water can condense in the core material. Composites can also absorb other types of fluids, including aviation fuel, although airborne water is the biggest problem.
Material specimens are tested under conditions representative of extreme cases of environment: cold temperature and dry; room temperature and dry; hot and dry; and hot and wet. Of these conditions, the worse case for polymer composites is hot and wet representative of a tropical environment. Specimens are placed inside a sealed chamber which simulates a severe tropical environment with a temperature of 71 °C (160 °F) and high humidity (85–95% water). At regular intervals the specimens are removed from the chamber to be weighed. The change in specimen weight indicates the amount of moisture absorbed by the composite. Figure 5.25 shows the weight gain for several types of carbon-fibre composite materials during exposure to hot–wet conditions. The mechanical properties of the specimens are measured at increasing intervals of exposure time, particularly the matrix-dominated properties such as compression strength. This data is used to assess likely changes to the mechanical behaviour of structural composites over the life of an aircraft. Other environmental tests may also be performed on composites, including exposure to ultraviolet radiation.
The certification of structural and engine materials is one of the most important issues with the testing and evaluation of new aircraft. Certification is also performed when new materials are used in major structural refits of existing aircraft, usually for life extension. Certification is essential to ensure the materials are safe, reliable, durable and functional in their structural application. New and improved materials cannot be introduced into aircraft without thorough analysis, testing and evaluation. A rigorous engineering assessment of the structural materials must be undertaken and passed before they are certified to use on aircraft and helicopters. The certification assessment involves mechanical and environmental (durability) testing of the material together with computation analysis using finite element modelling and other analytical methods. Aircraft certification is a complex, expensive and time-consuming process.
Certification regulations for structural and engine materials are specified by aviation regulatory agencies such as the Federal Aviation Authority (FAA) in the United States. Aerospace companies must ensure the materials pass the certification compliance standards specified by the regulation agency before being introduced to civil aircraft. Military organisations often use their own certification specifications and, although generally not as demanding as the civil specifications they still require extensive physical testing of the materials.
The certification procedure is often represented by the testing pyramid shown in Fig. 5.26. Often called the ‘building-block’ approach, this method is widely used by the aerospace industry to establish mechanical property data, property knock-down factors, and validation of critical design features for structures. Certification begins at the bottom of the pyramid. Mechanical properties of the material are determined by a series of tests at the ‘coupon level’, which means sample sizes about 100–200 mm long and 10–50 mm wide. Coupon tests are performed to determine basic property data, such as Young’s modulus, strength, fracture toughness, fatigue life and so on. The coupon tests are carried out under a standardised set of conditions which specify test parameters such as the sample size and loading rate. The test conditions are specified by standardisation organisations, such as the American Society for Testing and Materials. Large aerospace companies also have their own specifications for certain mechanical or durability tests not covered by the standards organisations.
Coupon tests are divided into quantitative or qualitative. Quantitative tests provide data that can be used for design purposes as well as certification. Examples of quantitative tests are the tension, compression and creep tests. Qualitative tests give results that can only be used for comparison purposes. For example, the hardness test and Charpy impact test provide a simple ‘go/no go’ assessment of materials. Certification testing at the coupon level involves measuring the material properties under different load conditions (e.g. tension, compression, bending) and operating environments (e.g. corrosive fluids, humidity, temperature). Simple drop weight impact tests are also performed, although other impact tests are often included at higher levels of the building block approach. These may include ballistic impact for critical military structures (e.g. main rotor blades), ice-hail simulation, bird strike simulation, and other program-specific impact tests.
The number of coupon tests performed on a new material can range from several hundred to many thousand, depending on the material itself and its intended aircraft structural application. For example, Table 5.1 shows the types of tests and numbers of specimens for a composite material intended for a new aircraft design. A large number of coupon tests are necessary to provide a statistical database on the mechanical properties. This is because many properties, including yield strength, fracture toughness and fatigue life, are sensitive to small variations in the material. Minor differences in the alloy content, microstructure and defects (e.g. gas holes) between metal samples can result in significant differences in the properties. When a large number of coupon tests are performed on the same material then the amount of scatter in the mechanical properties is quantified. For example, Fig. 5.27 shows the variation in the mechanical property of a material which has a Gaussian-type distribution. The mean property value is 100, although the measured properties vary from about 72 to 128.
|Test type||Number of tests|
|Bolted joints (stiffness, strength)||3025|
|Laminate strength (tension, compression, etc.)||2334|
|Durability and long-term exposure||585|
|Interlaminar shear strength||574|
|Defect effects on mechanical properties||494|
|Bonded repairs (stiffness, strength, etc.)||239|
|Ply properties (Young’s modulus, Poisson’s ratio, etc.)||235|
|Stress concentrations (open-hole tension, etc.)||118|
Large databases which quantify the amount of scatter in the mechanical properties are essential for safe aerospace design. The databases are used to determine the so-called material allowables that are used in design analysis. The material allowables are often given under the headings of A or B. The A basis allowable defines the mechanical property values for materials used in safety critical aircraft structures such as the fuselage, wings and undercarriage. The confidence required in the property values for these materials is critical to aircraft safety. The mechanical property value using the A basis allowable is defined as the value above which 99% of the population of values fall within the distribution with a confidence of 95%. The B basis allowable is less stringent, and is applied to components where material failure does not result in the loss or excessive damage to the aircraft. The property value using the B allowable is the value above which 90% of the population of values is expected to fall with a confidence of 95%.
After the mechanical and environmental properties of the materials have been determined by an exhaustive series of coupon tests, the structural and environmental properties of aircraft components built using the materials are then measured by more testing on a larger scale. As shown in the certification pyramid, structural elements, details and subcomponents that represent increasingly complex and more complete sections of the final aircraft are tested. The final component (e.g. wing or fuselage) is constructed of different subcomponents which, in turn, are assembled from many structural details which contain a large number of elements. The elements, details and subcomponents contain structural design features not present in the coupon specimens, such as cut-outs, stringers, rib attachments, changes in section thickness, bolted or bonded connections. Tests that replicate the actual loading on the final component are performed on the elements, details and subcomponents to ensure they comply with the design specifications (Fig. 5.28).
Testing of the entire aircraft is the final stage of the certification process. The full-scale test is one of the most important ways of proving how well the aircraft meets its performance requirements. The test is extremely important because it tests all the components and materials of the aircraft in the most realistic manner by simulating actual flight conditions. A full-scale structural test is usually performed on one of the first four aircraft built. Full-scale tests are also performed on in-service aircraft that have undergone a major structural design change. Full-scale tests are used to ensure the aircraft is structurally sound after all the materials, elements, details and subcomponents have been fully integrated. The full-scale test is also important to determine the effect of secondary loading caused by complex out-of-plane loads, which may not be determined in earlier testing. Such loads arise from eccentricities, stiffness changes and local buckling which may not be fully predicted or eliminated in design nor represented by the structural detail specimen. Another important aspect of the full-scale test is the confidence that the aircraft is safe.
The full-scale component is tested under conditions given in the certification specifications to ensure it is fully compliant. For example, wing testing by Boeing of their B787 aircraft involved bending a complete wing upwards beyond the ultimate load (or 1.5 times the load limit) to the point of destruction (Fig. 5.29). The point was reached when the wing tips deflected by about 7.6 m above their normal position while enduring loads equal to about 500 000 lbf. In addition to this full-scale test, the aircraft was tested for 120 000 simulated flights, or double the expected life cycle, for fatigue performance. The simulated flights lasted about 4 min each and replicated taxiing, climb, cabin pressurisation and depressurisation, descent and landing. To add realism, the artificial flight conditions varied from completely smooth and level to extremely turbulent. It is only after such an exhaustive series of passed tests from the coupon to final component levels that the aircraft is certified by the aviation safety authority.
The mechanical properties of aerospace materials should be measured under the stress conditions experienced in service. As examples, the properties of materials used in tension loaded structures, such as the fuselage, are determined by tensile testing whereas materials used in compression loaded structures are measured by compression testing.
Mechanical property tests are classified as quantitative and qualitative. Examples of quantitative tests include tension and compression tests which give absolute measurements for properties such as elastic modulus and yield strength. Qualitative tests, such as Charpy impact and hardness tests, measure property values which should only be used for ranking materials.
Durability tests are performed on aerospace materials to ensure they do not lose functionality and degrade when used in the aviation environment. The main tests for metals determine the corrosion and oxidation (for engine applications) properties whereas the main tests for fibre–polymer composites determine the moisture absorption behaviour.
The properties of engineering materials are rarely constant, and there is usually scatter in the measured values. The certification testing of aerospace materials requires a large number of mechanical property tests to be performed to quantify the variability in properties, which is then expressed as an A- or B-basis allowable in the design of safety critical structures.
The mechanical and durability testing of materials forms an important part of the certification of aircraft. Coupon tests performed using small samples of the material are used to obtain basic data on the mechanical and durability properties. Further tests are then performed at the element, structural detail, subcomponent and final component (which may be the entire aircraft) levels as part of the complete structural certification process.
Elastic limi: he maximum strain applied to the specimen without causing plastic deformation. The elastic limit is the transition point between elastic and plastic deformation. Also called the proportional limit.
Offset metho: ethod used to calculate the proof strength of a material that does not have a sharp yield point. A straight line is drawn parallel to the linear elastic part of the stress–strain curve at a specific strain offset value. The intercept point of the offset line and the stress–strain curve defines the proof strength of the material.
Poisson’s rati: he ratio of the transverse (lateral) strain to the corresponding axial (longitudinal) strain resulting from uniformly distributed axial stress below the proportional limit of the material.
Proof strength (also called offset yield strength): The stress at which a material undergoes a specified amount of deformation (usually 0.2% strain). The stress used to define the onset of plastic deformation in ductile materials that do not have a clearly defined yield point on the stress–strain curve.
Proportional limi: he highest stress at which stress is directly proportional to strain. It is the highest stress at which the stress–strain curve is rectilinear, and marks the point between elastic and plastic deformation.
Tensile toughnes: roperty that describes the amount of elastic work to the material before the onset of plastic deformation. Determined from the area under the stress–strain curve in the elastic regime.