Search

A comprehensive analysis of the defect detection performance of long pulse excitation thermographic NDE is presented. An analytical procedure for predicting the thermal image contrasts of defects of specified size and depth is developed and validated by extensive experimental studies of test pieces having a wide range of thermal properties. Results obtained using long pulse (~5 s) excitation are compared with those obtained using traditional flash excitation. The conditions necessary for the success of the long pulse method are explained and illustrated by both modelling and experimental results. Practical advantages of long pulse excitation are discussed.

A portable thermosonic inspection system has been developed incorporating a miniature microbolometer array camera. The vibrations excited during a test are monitored using a high frequency microphone and assessed automatically to validate the test. An investigation of the correlation between thermosonic heating and vibration excitation energy in composites is presented. The system has been trialled on aerospace components containing impact damage.

Test piece surfaces are sometimes coated with a black, energy absorbing paint before transient thermography is applied. This practice is not acceptable to some thermal barrier coating (TBC) manufacturers and servicers of these systems since thermal barrier coatings are porous so the paint contaminates the coating and it is very difficult and costly to remove. Unfortunately, unpainted TBC surfaces have low emissivity, and after service their colour is usually uneven. The low emissivity gives low signal levels and also problems with reflections of the incident heat pulse, while the variation in emissivity over the surface gives strong variation in the contrast obtained even in the absence of defects. Additionally, the TBC is translucent to mid-wavelength IR radiation which negatively affects the location of disbonds based on the thermal responses. This paper investigates the effects of uneven discolouration of the surface and of IR translucency on the thermal responses. It has been shown that unpainted TBC systems can be inspected reliably by using higher power flash heating equipment assembled with an IR glass filter and a long wavelength IR camera. The paper also shows that the problem with uneven surface emissivity can be overcome by applying 2nd time derivative processing of the log–log surface cooling curves.

High power light emitting diode (LED) arrays have been investigated as excitation sources for long pulse and lock-in thermography. Images of artificial defects in a carbon fibre reinforced plastic (CFRP) composite sample are compared, by image contrast signal-to-noise ratio estimates, with those obtained using conventional incandescent flash and lock-in excitation sources. The LED arrays had to be mounted on heat sinks with active cooling in to prevent them exceeding their thermal tolerance. Despite this cooling the LED arrays were still found to emit some IR radiation, although far less than conventional incandescent light sources.

Swing describes the lateral deviation of a cricket ball in its trajectory towards the batsman. Conventional swing is effective with a new, or well-preserved, ball, and the fluid dynamics governing this phenomenon was first explained in 1957. In 2012, many test-match fast bowlers are able to swing, at high speed, an older ball in the reverse direction. This reverse swing of a ball aged under match conditions has never been explained fully. A cricket ball is asymmetric with six seams of 80–90 encircling stitches, protruding approximately 1 mm proud of the surface. Both conventional and reverse swings are a consequence of asymmetrical flow separation leading to a skewed wake and a net pressure force on the ball perpendicular to the flight trajectory. Here, experimental evidence is presented for the first time showing that the formation of a laminar separation bubble is the prominent flow feature creating the flow asymmetry for reverse swing. A new flow visualisation technique to capture the fluid dynamics of boundary-layer separation using an infrared camera is also introduced here.

Eddy-current induced thermography (induction thermography, hereon referred to as eddytherm) is an active thermographic method which is capable of rapid and non-contacting detection of out-of-plane cracks in electrically conductive parts. In an eddytherm inspection, the part is induction heated; cracks cause localised changes in the induced eddy-current flow and the associated Joule heating is imaged at the surface of the part with an infrared camera. In this study the detectability of fatigue cracks in steel, titanium and Waspaloy is quantified by novel but simple image processing routines which are specifically applicable to eddytherm inspection. The quantitative detection data is then input into a cumulative log-normal probability of detection model to estimate the probability of detecting the fatigue cracks as a function of crack length. a90,95 (i.e., the crack length which can be detected 90% of the time with 95% confidence) is found to be 0.60 mm for steel, 0.78 mm for titanium and 1.50 mm for Waspaloy (a nickel-based superalloy), showing eddytherm to be an extremely sensitive method.

Advanced composite materials are finding increasing application in aerospace, marine and many other industries due to their performance and structural efficiency. Maintenance inspection of these light-weight structures is a relatively new and difficult task for Non-Destructive Testing (NDT), which need robust methods to be applicable in industrial environments. In this paper, a new numerical–experimental procedure to detect size and depth of flat bottom holes in metallic and laminated composite structures by digital shearography (DISH) is proposed. The flaw detection capabilities of DISH have been evaluated by measuring the dynamic response of defects to applied stresses. Vibration dynamic loading is used to reveal flat bottom holes made with different sizes and placed at different depths in CFRP laminates. The shearographic methodology is based on the recognition of the (0 1) resonance mode per defect. A simplified model of thin circular plate, idealized above each flaw position, is used to calculate the natural frequency of vibrating defects. Then, the numerical difference between experimental resonance frequencies and those computationally obtained is minimized using an unconstrained optimization algorithm in order to calculate the defect depth. Considering the simplicity and rapidity of this technique, the laser shearography methodology is evaluated reliable as NDT method.

Pulsed, lock-in and frequency modulated thermography are three alternative nondestructive evaluation techniques. The defect imaging performance of these techniques are compared using: matched excitation energy; the same carbon fiber composite test piece and infrared camera system. The lock-in technique suffers from “blind frequencies” at which phase images for some defects disappear. It is shown that this problem can be overcome by using frequency modulated (chirp) excitation and an image fusion algorithm is presented that enhance phase imaging of defects. The signal-to-noise ratios (SNRs) of defect images obtained by the three techniques are presented. For the shallowest defects (depths 0.25 and 0.5 mm, 6 mm diameter), the pulsed technique exhibits the highest SNRs. For deeper defects the SNRs of the three techniques are similar in magnitude under matched excitation energy condition.

Work on three thermographic NDE techniques that have been developed to locate and image cracks in metallic components is reviewed. These techniques are: thermosonics (sonic IR, ultrasonically stimulated thermography); laser spot thermographic imaging; and pulsed eddy current stimulated thermography. The three techniques are explained and examples are given of their performance in imaging well-characterised cracks. Factors concerning the practical and reliable implementation of the techniques are discussed.

Surface breaking defects are detected using a scanning laser source to deposit heat into a sample surface. Any lateral flow of heat is disturbed by such a defect, with a change in thermal spot shape being detected by an infrared camera. An apparent increase in temperature as the laser passes over a defect is due to a localised change in emissivity, allowing defects to be distinguished from surface markings. Defects have been detected in both stainless steel and aluminium.

A new crack imaging technique is presented that is based on second derivative image processing of thermal images of laser heated spots. Experimental results are shown that compare well with those obtained by the dye penetrant inspection method. A 3D simulation has been developed to simulate heat flow from a laser heated spot in the proximity of a crack. A ‘ghost point’ method has been used to deal efficiently with cracks having openings in the micometre range. Results are presented showing the effects of crack geometry and system parameters on thermal images of laser heated spots.

There has been a growing interest in the use of composites especially in structural application ranging from aerospace to automotive and marine sectors. However, their performances under impact loading represent one of the major concerns as impacts may occur during manufacture, normal operations and maintenance. This paper presents two novel NDT techniques, thermosonics and digital shearography (DISH) to detect and assess barely visible impact damage (BVID) produced on a stiffened composite wing panel by unknown low energy impacts. Thermosonics is based on synchronized infrared imaging and ultrasonic excitation. Despite the apparent simplicity of the experimental setup, thermosonics involves a number of factors, e.g. acoustic horn location, horn crack proximity, horn-sample coupling etc., that significantly tend to influence both the degree and the period of the excitation. Then, a numerical-experimental procedure for the assessment of the size and depth of delamination by digital shearography (DISH) is proposed. The flaw detection capabilities of DISH have been evaluated by measuring the dynamic response of the delaminated area to applied stresses. The shearographic methodology is based on the recognition of the (0 1) resonance mode per defect. A simplified model of thin circular plate, idealized above each impacted area, is used to calculate the natural frequency of vibrating delamination. The numerical difference between experimental resonance frequencies and those computationally obtained is minimized using an unconstrained optimization algorithm in order to calculate the delamination depth. The results showed that thermosonics is a quick and effective method to detect and localize BVID damage while the combined shearography and optimization methodology was able to size and localize delamination due to low velocity impacts.

The objective of this work was to image the presence of impact damage by monitoring the nonlinear response of damaged carbon/epoxy composite samples. The presence of microcracks, debonding, delamination, etc… induce the material to behave in a nonlinear elastic fashion highlighted by the presence and amplitude of harmonics in the spectrum of the received signal when the sample is periodically excited at one of its resonance frequencies. The sensitivity of a second harmonic imaging technique (SEHIT) based on material nonlinear elastic effect known as second harmonic generation (SHG) was investigated. The proposed imaging process was used to detect barely visible impact damage (BVID) due to low velocity impact (

Thermographic inspection with eddy current (EC) excitation is an emerging integrative NDT&E method with the ability to inspect for defects over large areas. The resultant surface heat distribution from direct EC heating and diffused heat can be obtained easily with a thermal camera, but techniques for the determination of heating mechanisms around a particular defect for quantitative defect characterisation are required. In this paper, numerical modelling and experimental studies are applied to understand EC stimulated thermography on simple discontinuity defects, including transient EC distribution and heating propagation for slots and notches. This fundamental understanding of transient EC distribution and heating propagation will aid in the development of feature extraction and pattern recognition techniques for the quantitative analysis of EC thermography images and defect characterisation.

Swing, the sideways deviation of the cricket ball in flight, is a phenomenon which can be explained in terms of fluid dynamics. Conventional swing has been used since the beginning of the twentieth century and is effective with a new or well-preserved ball. A controversial form of swing emerged in Pakistan in the 1980s, featuring an older worn ball which, at a high speed, swung in the reverse direction. A cricket ball is asymmetric because of the presence of a seam, which is made up of rows of prominent encircling stitches. For conventional swing, this seam trips into turbulence the boundary layer adjacent to one hemisphere of the ball which remains attached to a greater angle (about 120°) than does that on the other side (about 80°) where no seam is present to trip the laminar boundary layer. The result is asymmetrical separation, leading to a skewed wake and a net pressure force on the ball perpendicular to the flight trajectory. Reverse swing is thought to be a consequence of the fact that asymmetry inverts at a high speed if the seam thickens the turbulent boundary layer on one side of the ball. Although the fluid dynamics causes of both conventional and reverse swing are well known, this paper demonstrates clearly, by means of flow visualization and pressure measurements, the inversion of this pressure asymmetry at Reynolds numbers greater than 170×103.

The defect detection capabilities of transient thermography and shearography have been compared using optimum excitation methods for each technique: short pulse heating for thermography and vibration excitation using a piezoelectric transducer for shearography. A signal-to-noise ratio and limit of detection analysis has been performed on defect images obtained by the two techniques using the different excitation methods. Test samples considered in this paper are flat-plate samples made from aluminium, mild steel, stainless steel, CFRP and thermoplastic, containing flat-bottomed hole artificial defects of 20mm diameter at depths ranging from 0.5mm to 3.0mm. The technique of flash thermography involves using a short duration (~1ms) high intensity flash of light to heat the surface of a test piece. The test piece surface temperature is recorded by an infrared camera and computer system as it decays due to heat being conducted into the part after its deposition on the surface. Sub- surface defects reduce the conduction of heat away from the surface and therefore reduce the surface cooling rate compared to that occurring over non-defective regions. Consequently, a surface temperature contrast develops over a defect that can be used to detect a defective region. Shearography, otherwise known as shearing speckle interferometry, is a technique that uses interferometry to measure the out of plane displacement gradient of a sample's surface. The presence of defects will alter the way in which a sample reacts to an applied stress (in this case created by vibrating the test sample) and this change can be observed using shearography thereby inferring the presence of the defect. In shearography a speckle pattern is applied to the surface of the test sample and the sample is illuminated using expanded collimated (i.e. laser) light. The light is reflected diffusely from the surface of the sample and passes through a lens and shearing device to be recorded by a video camera and processed. The shearing device acts to slightly change the path of half of the rays of light reflected by the speckles, causing the reflected light from neighbouring speckles to overlap and produce an image sheared in the shearing direction. A reference sheared image is stored with the sample in an unstressed condition and the sample is then stressed. The stressing can take many forms, mechanical strain (e.g. with a vacuum hood or by applying a bending load with a clamp), thermal stress (generated by heating or cooling the sample), or vibration excitation (using a piezoelectric transducer as is the case in this testing). A second sheared image is recorded with the sample in the stressed state, and the interferometric superposition of these two images creates an interferogram that represents the phase difference of neighbouring speckle sources. From this phase difference the gradient of the surface displacement can be calculated.

There is a need for fast and efficient techniques to inspect engineering structures and complex components such as aircraft turbine blades to identify potential sites of failure. Pulsed eddy current (PEC) thermography is a new inspection technique which allows the user to capture the eddy current distribution in a component or structure using infrared imaging and detect defects over a relatively wide area. The technique is applicable to materials with a reasonable level of electrical conductivity and has the ability to detect defects under coatings. However, PEC thermography has received relatively little attention compared to other thermographic inspection techniques. In this paper, the design, development and optimisation of a PEC thermography inspection system is detailed, including coil design for global and local heating of samples, optimisation of excitation parameters (frequency, power, pulse duration etc) and camera selection. The system is used to inspect several real-world samples, using different coil designs, and the results are assessed using newly developed feature extraction techniques. The work shows that with judicious coil design and selection of excitation parameters, PEC thermography can be used to obtain quantitative information for defect characterisation through analysis of the surface heating pattern and the transient temperature change.

This paper examines whether emergent behaviour is observed in dielectric composites, i.e. mixtures of two phases with zero conductivity and of permittivity ϵ 1 and ϵ 2 . The model developed enabled the study of random capacitor networks which contain capacitors of two different magnitudes (C 1 and C 2 ), which represent the two individual dielectric phases of a composite. The effective capacitance of the network was examined as a function of the individual capacitor magnitudes and their volume fraction. The capacitor networks exhibit an emergent region where there is low variability between individual random networks of a particular composition. The logarithm of the network capacitance scaled linearly with the logarithm of the capacitance ratio (C 1 and C 2 ), indicating that under particular conditions a logarithmic mixing rule can be used to determine the properties of dielectric mixtures.

Porous lead magnesium niobate-lead titanate (PMN-PT 90:10) relaxors were impregnated with water to provide a model conductor-insulator mixture, to study their power law frequency dependency of ac conductivity, permittivity and phase angle. Relaxor materials with a range of open porosity filled with water created composites with conductor volume fractions ranging from 8.2% to 22.2%. The use of a high relative permittivity PMN-PT (∼ 8000) enabled the power law dispersion to be observed at relatively low frequencies (∼ 2 kHz). Good agreement was obtained between experimental data and predicted results based on a logarithmic mixing rule with a strong correlation between the power law exponent and conductor-insulator fraction. The model and numerical methods presented are considered a simple approach to interpret and predict and the frequency dependent properties of materials which similar heterogeneity.

The defect detection capabilities of pulse transient thermography and lock-in thermography were compared using equal excitation energies. A signal-to-noise ratio analysis was performed on defect images obtained by the two techniques. The test piece imaged by both techniques was a carbon fibre composite plate containing back-drilled flat-bottomed hole artificial defects of 4, 6 and 12 mm diameters at depths ranging from 0.25 to 3.5 mm. Similar limits of defect detection were found for the two techniques. Lock-in thermography phase images were found to exhibit anomalous switches between positive and negative phase values for different modulation frequencies and for different combinations of defect diameter and depth. This effect resulted in values of defect phase response that differed substantially from the one-dimensional thermal wave interpretation of lock-in thermography imaging. The consequences of this anomalous effect and its physical origins are discussed.

The use of pulsed thermography as a non-destructive evaluation tool for damage monitoring of composite materials has dramatically increased in the past decade. Typically, optical flashes are used as external heating sources, which may cause poor defect definition especially for thicker materials or multiple delaminations. SMArt thermography is a new alternative to standard pulsed thermography as it overcomes the limitations on the use of external thermal sources. Such a novel technology enables a built-in, fast and in-depth assessment of both surface and internal material defects by embedding shape memory alloy wires in traditional carbon fibre reinforced composite laminates. However, a theoretical model of thermal wave propagation for SMArt thermography, especially in the presence of internal structural defects, is needed to better interpret the observations/data measured during the experiments. The objective of this paper was to develop an analytical model for SMArt thermography to predict the depth of flaws/damage within composite materials based on experimental data. This model can also be used to predict the temperature contrast on the surface of the laminate, accounting for defect depth, size and opening, thermal properties of material and defect filler, thickness of the component, and intensity of the excitation energy. The results showed that the analytical model gives good predictions compared to experimental data. This paper is one of the first pioneering work showing the use thermography as a quantitative non-destructive tool where defect size and depth could be assessed with good accuracy.

A surface crack, close to a small, well-defined heated spot, impedes lateral heat flow and produces alterations in the shape of the thermal spot image that can be monitored by thermography. Spot heating has been achieved using both cw and pulsed laser beam illumination on stainless steel and titanium samples and the technique has been found to be successful for determining the location of fatigue cracks. When using a pulsed laser beam one can also simultaneously generate wideband ultrasonic signals in the sample; this can be used to detect the presence of surface and sub-surface defects in the sample. Results are presented that have been obtained using a fixed camera and cw laser beam position with the sample being moved through the field of view of the camera. Results are also presented using a fixed camera and sample, with a raster scanned cw or pulsed lasers beam moving across the sample. A demonstration of how a non-contact ultrasonic measurement can be performed simultaneously is presented. Thermal i...

Transient thermography images of artificial defects in a carbon fibre composite test piece obtained using a cooled focal plane array (FPA) infrared camera are compared with those obtained using an un-cooled microbolometer array camera. Signal to noise ratios of each defect image are computed to provide an objective comparison of the performances of the two cameras. The cooled FPA infrared camera exhibits a higher ultimate defect detection capability than the un-cooled microbolometer array camera but this camera has the advantages of much smaller size and lower cost.

This study investigated methods of reducing the electrical power required to detect defects by thermosonics, or vibro-thermography, and thus to reduce surface damage that can occur at the exciter attachment point during testing. The surface temperature rise over excited defects has been modelled to determine the heating requirements to visualise damage. A long pulse, low power excitation method has been found to produce satisfactory impact damage images whilst eliminating damage at the exciter attachment point. Thermosonics has been found to detect impact damage in composites more reliably than optically stimulated thermal NDE methods.

A new analytical model for predicting the compressive fatigue limit strain of composite laminates which contain barely visible impact damage (BVID) is presented. The model represents the complex damage morphology as a single, circular delamination, and calculates the strain at which thin-film buckling of the circular region of delaminated plies occurs. The fatigue limit strain is defined as the strain at which the strain energy release rate for a thin post-buckled strip of the delaminated plies is equal to the critical Mode I value ( G 1C ) for the resin. The model predicts a “critical” depth at which propagation of damage during fatigue is likely to occur. Results obtained using the model are compared with two sets of experimental results, and show agreement of fatigue limit strain to within 4% of the experimental value.

A photoinduced giant dielectric constant of ∼106 has been found in impedance spectroscopy measurements of lead halide perovskite solar cells. We report similar effects in measurements of a porous lead zirconate titanate (PZT) sample saturated with water. The principal effect of the illumination of the solar cell and of the introduction of water into the pore volume of the PZT sample is a significant increase in conductivity and dielectric loss. This is shown to exhibit low frequency power law dispersion. Application of the Kramers-Kronig relationships show the large measured values of permittivity to be related to the power law changes in conductivity and dielectric loss. The power law dispersions in the electrical responses are consistent with an electrical network model of microstructure. It is concluded that the high apparent values of permittivity are features of the microstructural networks and not fundamental effects in the two perovskite materials.

In previous work, electrical networks of randomly distributed resistors and capacitors driven with a periodic voltage have shown emergent scaling behaviour: a logarithmic scaling rule is found to dictate the conductance of the bulk network as a function of frequency. This feature has also been shown in other bi-materials and is thought to be independent of the materials involved and, hence, “universal” in nature. The present article examines the potential universality of these scaling laws by considering a mechanically loaded random network. In particular, the mechanical network under consideration is a statically loaded truss. The mechanical system shows an emergent region and percolation paths, much like its electrical counterpart, further suggesting that random, networked systems may have a universal character.

The frequency dependent conductivity and permittivity of a ceramic composite are modelled using electrical networks consisting of randomly positioned resistors and capacitors. The electrical network represents a heterogeneous microstructure that contains both insulating (the capacitor) and conductive regions (the resistor). To validate model results, a model ceramic conductor–insulator composite was designed consisting of a porous lead zirconate titanate impregnated with different concentrations of water. Excellent agreement between experimental and model data was achieved with a strong correlation with many other ceramics, glasses and composites. It is proposed that the 'universal' dielectric response of many materials is a consequence of microstructural heterogeneity. The modelling approach could be used as a simple and effective method for microstructural design of ceramics and other materials with tailored dielectric properties.

Thermal non-destructive testing (NDT) is commonly used for assessing aircraft structures. This research work evaluates the potential of pulsed thermography (PT) for certain applications. In particular, real-time monitoring was obtained using PT. In some cases, thermal modelling or other non-destructive testing and evaluation techniques (ie acoustography, ultrasonic wheel array) were also used with the intention of providing supplementary results. The following features were studied and presented: □ Through-skin sensing on aluminium (Al) alloy and carbon fibre reinforced plastic (CFRP) structures. □ Drilling-induced defects on multi-ply laminates of CFRP. □ Impact damage on CFRP fibre reinforced plastic panels and honeycomb sandwich structures. It is concluded that pulsed thermography is a rapid large-area NDT technique that can find adequate use on aircraft - aerospace materials and structures.

Very large networks of randomly positioned resistors and capacitors have been used to simulate the microstructures of real two-phase (conductor–insulator) materials. These networks are found to exhibit fractional power law frequency dependences of dielectric properties and ac conductivity, of the type reported for a wide range of materials. The network results are related to the resistor and capacitor values by a simple logarithmic mixing rule. The same mixing rule is used to model the electrical characteristics of two-phase electrical composites. The results are tested using water impregnated lead zirconate titinate (PZT) ceramics samples that have a microstructure that forms a complex interconnected random array of conducting (water) and insulating regions. Excellent agreement is obtained between the experimental data and the modelling predictions based on the network simulation results. The power law exponents for ac conductivity and relative permittivity are found to be equal to the proportions of the composite occupied by the insulating and conducting phases, respectively. Studies of conducting polymer impregnated PZT are also presented which show less good agreement with modelling predictions.

Thermosonics, or ultrasound stimulated thermography, has been shown to reveal structural defects such as impact damage in CFRP that are not detectable by conventional optically stimulated thermography. However, there are concerns that the large amounts of ultrasonic excitation energy employed to make thermosonic inspections may cause damage, particularly by heating at the attachment point of the exciter. The objectives of this study were to investigate methods of reducing the electrical power required to detect defects and thus to reduce surface damage during testing. The magnitude of the heating at the surface over ultrasonically excited defects was modelled, to determine the heating requirements for different materials. Thermosonic images were collected whilst measuring test piece vibrations, using strain gauges. A long pulse, low power excitation method has been found to produce satisfactory impact damage images whilst eliminating damage to the test piece caused, in other methods, by heating of the ultrasonic excitation horn. Images of impact damage in 4 mm, 8 mm and 16 mm CFRP plates have been obtained using only I W of electrical power applied to an ultrasonic excitation horn.

Thermal wave phase measurements are reported on the drying of wet paint films on aluminium substrates. Measurements of the change in thickness as the paint dries have also been obtained using a differential focussing technique on an optical microscope. By including the optical microscope measurements of the drying paint film thickness together with estimates for the density and thermal properties of the drying and cured paint, predictions have been made of the thermal phase/thickness relationships for the wet, dry, curing and cured paints. It is concluded that a phase measurement on the wet paints could be used to predict a final cured paint thickness with an accuracy of approximately ±2 μm. Errors in predicting cured film thickness from a wet film thermal phase measurement arises principally from uncertainty over the solvent content of the wet paint film, the state of cure, and the consequent uncertainty over the paint density and thermal properties.

The dc conductivities of ZrO 2 -Y 2 O 3 ceramic alloys (in the range 3-12 mol% of Y 2 O 3 ) have been obtained from ac impedance measurements at temperatures between 250° and 370°C. The Almond-West ac conductivity model has been applied to evaluate hopping rates in this system. The migration enthalpies were evaluated and shown to increase with yttria concentration, but all values determined were shown to be lower than the corresponding activation enthalpies for conductivity. The association enthalpies thus calculated were shown to be very small in 3 mol% Y 2 O 3 -ZrO 3 and to increase with yttria concentration until the yttria contents were high enough to form fully stabilized cubic zirconia. For these samples the association enthalpies are about 0.19 eV, and no longer sensitive to yttria content. The low hopping rate at high yttria concentration might be attributed to low entropy in the system, which might be attributed to the formation of vacancy clusters and/or an ordering of the structure.

A phenomenological overview of the buckling and postbuckling of fully and partially delaminated struts is developed using a simple four-degree-of-freedom nonlinear Rayleigh-Ritz formulation. Bifurcation analysis indicates that instability occurs in general at an asymmetric point of bifurcation. After bifurcation, realistic solutions are seen to follow the stable part of the postbuckling path, with the unstable branch being denied by contact between the contributing laminates. Depending on the geometry and position of the delamination, thoroughly stable (thin-film), effectively neutral (overall), or potentially unstable (mixed-mode) buckling can occur in the postbuckling range. Numerically obtained equilibrium solutions of the model are found to compare well with those obtained using the finite element code ABAQUS. Critical delamination depths, where the response undergoes a change in form, are discovered at positions of secondary bifurcation. The multiplicity of equilibrium solutions that arise at such points are seen to cause possible problems of path selection for standard finite element routines.

In this work, an investigation has been made of the potential of pulsed-transient thermography for identifying the location of fixtures beneath aircraft skins to facilitate accurate automated assembly operations. Representative test structures, comprising of aircraft skin (i.e. aluminium (Al) or carbon fibre reinforced plastic (CFRP)) positioned over a thick strut fixture of Al or CFRP, were investigated experimentally and analysed using finite difference thermal modelling software, taking into account the size and depth of the features, as well as the thermal properties of the investigated materials. The ability of the technique to detect a subsurface fixing and to provide information about its location was analysed. Results from the modelling and the experimental analysis are presented and discussed.

Transient thermography (TT) was used for the non-destructive assessment of repaired aircraft carbon or boron composite panels, as well as of composite patches enclosing a fibre optic across their width. The panels were assessed using two different thermographic evaluation systems. In all situations the subsurface features, i.e. defects and/or the fibre optics, were positioned intentionally. After detecting the subsurface features, representative thermal images obtained from the investigation underwent quantitative image analysis. Finally, mathematical thermal modelling was also attempted in order to obtain information about the defects in space and in time.

The prerequisite for more competent and cost-effective aircraft has led to the evolution of innovative testing and evaluation procedures. Smart methods for assessing the integrity of an aircraft structure are essential to both reduce manufacturing costs and out-of-service time of aircraft due to maintenance. Nowadays, thermal non-destructive testing (NDT) is commonly used for assessing aircraft composites. In this work, certain applications of transient thermal NDT relating to the assessment of aircraft composites are presented. In particular, the following features were studied: (a) Notches under multi-ply composite patching (bonded with FM73 adhesive film to the surface of Al 2024-T3) and a simulated delamination between two plies on a multi-ply composite repair. (b) Drilling induced defects on multi-ply laminates of HEXCEL AS4/8552 carbon fibre composites. (c) Impact damage on carbon fibre reinforced plastic (CFRP) panels and honeycomb sandwich structures (bonded with AF-163-2U.03 adhesive film). (d) Through skin sensing assessment of CFRP. Real-time monitoring of all features was obtained using pulsed thermography. However, in the composite repairs cases thermal modelling and pulsed-phase thermography were also used with the intention of providing supplementary results, whilst in the case of through skin imaging thermal modelling was also used in order to demonstrate the importance of thermal contact resistance between two surfaces (skin and strut). The thermal NDT approaches used in this work gave satisfactory results in all instances.

Transient-pulsed thermography was used as a through skin imaging technique with the intention of locating anchoring points beneath the outer skins of aircraft structures. A representative test structure, comprising a thin skin positioned over a thick strut fixture, was investigated experimentally and analysed using finite difference thermal modelling software. Both carbon fibre reinforced plastic (CFRP) and aluminium structures were investigated. The ability of the technique to detect a subsurface fixing and to provide information about its location was analysed. Results indicate a precision of better than 0.5 mm to be achievable.

Transient thermography was employed in the inspection of defects in various aircraft composite panels. Three different categories of defects were investigated in the laboratory; notches on aluminium alloy panels under carbon or boron composite patching, a simulation of delamination on a boron composite patch, and fibre breakout on carbon composites. In all situations, the defects were artificially created. After detecting the defects, quantitative analysis concerning the contrast of the detected defects was carried out. Finally, information about the size of the artificial delaminated panel in relation to transient time was also performed.

The real-time monitoring of damage in composite materials during fatigue loading has been made possible by an emerging new ultrasonic imaging technology, acoustography. The successful integration of acoustographic technology and a servo-hydraulic test machine has resulted in a new measurement system which can be used for the in-situ imaging of impact damage in composite specimens during long-term fatigue tests. Results show that damage area growth during constant load fatigue occurs in three stages. After an initial small enlargement, damage grows at a constant rate until the last stage is reached when damage grows with an increasing rate to final failure.

There are growing concerns about the effects of accidental impact damage on the structural integrity of aerospace composites and about the possible growth of the damage due to in-service fatigue. There has been some success in the use of established methods (ultrasonic C-scan, thermography, X-rays) to monitor damage development during fatigue experiments by interrupting a test and removing the specimen for damage inspection but this stop-and-restart test procedure is far from satisfactory. Real-time damage monitoring in composite materials during fatigue has now become possible by the emergence of a new ultrasonic imaging technology, acoustography. The successful integration of acoustography and a servo-hydraulic fatigue test machine has resulted in a new measurement system which can be used for the in situ monitoring in real time of damage growth in composite specimens during long-term fatigue tests. Results are presented which show damage-area growth during fatigue cycling under high compressive loads. After an initial small enlargement (stage 1), damage grows at a constant rate (stage 2) until the third stage is reached when there is further growth at an increasing rate to final failure. However, a ‘fatigue limit’ has also been observed. At stresses below this fatigue limit, a zero damage-growth regime has been found in studies of >10 6 fatigue cycles. The results obtained have important implications for the understanding of the effects of damage on fatigue life and for the design of ‘safe’ damage-tolerant structures.

Simulations of the electrical noise in large networks of randomly positioned resistors and capacitors show a network noise power spectral density that varies as 1/fα(0

The integration of an acoustography ultrasonic imaging system with a mechanical testing machine is described. The apparatus is designed for real-time ultrasonic imaging of impact damage growth in composite material samples subjected to either static or fatigue loading. Experimental results are presented of the damage growth to failure of an aerospace standard carbon-fibre composite sample subjected to compressive cyclic loading. Experimental results for static loading show a reversible increase and decrease in damage area, indicating an opening and closing of delamination defects, which points to the possibility of an under-estimation of defect size by the conventional ultrasonic testing of unloaded components.

Simulations of the AC electrical characteristics of 2D square networks randomly filled with resistors or capacitors exhibit many features in common with experimental dielectric responses of solids. These include the `universal' fractional power law dispersions in permittivity and dielectric loss characterized by the Cole-Davidson response function. Simulations are presented of networks containing different proportions of resistors and capacitors which show that the power law frequency response is accounted for well by the logarithmic mixing rule. Limiting high and low frequency characteristics are found to be controlled by percolation paths of either resistors or capacitors. It is suggested that the power law response of a solid could be an indication that it is microscopically inhomogenous, containing an effective microscopic random network of conducting and dielectric insulating islands.

The ac electrical characteristics of a 2-D square randomly filled RC network are shown to exhibit many features in common with those of a typical ionic conductor. These include the ‘universal’ power law frequency responses of ac conductivity and permittivity. It is suggested that a random network of ion conduction channels and insulating islands may provide a realistic model of the conduction paths within a bulk ionic conductor. DC ionic conductivity is identified with a resistive percolation path through the network and the higher frequency power law component with that of the random network of ac conducting linked capacitive and conductive regions. The dielectric loss peak associated with ionic conductors is also found to be a random R-C network characteristic. The effect of the insulating islands having a Debye-like capacitance is modelled and ZrO 2 12 mol% Y 2 O 3 data are presented which exhibit the effects predicted.

Constant-stress fatigue data for five carbon-fibre-reinforced plastics and one glass-reinforced plastic laminate have been used to evaluate possible artificial neural network architectures for the prediction of fatigue lives and to develop network training methods. It has been found that artificial neural networks can be trained to model constant-stress fatigue behaviour at least as well as other current life-prediction methods and can provide accurate (and conservative) representations of the stress/R-ratio/median-life surfaces for carbon-fibre composites from quite small experimental data-bases. Although their predictive ability for minimum life is less satisfactory than that for median life, and is non-conservative, the procedures developed in this work could nevertheless be used in design with little further modification. Some success has been achieved in modelling fatigue under block-loading conditions, but this problem is more difficult and requires much more effort before ANNs could be used with confidence for variable-stress conditions.

DC conductivity in ionic conductors is proportional to the hopping rate of the charge carriers. The hopping of the charge carriers can result in a dielectric loss peak and the DC conductivity is also proportional to the loss peak frequency. In the present study, the AC conductivity of 12 mol.% Y2O3 doped ZrO2 has been measured and the relationship between the hopping frequencies characteristic of the AC conductivities and dielectric loss peak frequencies has been examined. In addition, the peak frequencies in the complex impedance plane were also measured, to compare with the hopping frequencies. It has been demonstrated that two sample constants can be used to connect the three frequencies. It is suggested that the three frequencies are a consequence of a common phenomena and that all result from the hopping process of the charge carriers.

Measurements of the bulk ac electrical response of the fluorine ion conductor (1− x )CdF 2 x ErF 3 , x =0.15, are shown to be composed of separable dipolar dielectric and hopping conductive components. The dielectric component has a relaxation frequency of ∼4 kHz and power law, Cole–Davidson frequency dependencies of real and imaginary parts. The conductive component is frequency independent up to ∼100 kHz and decreases in a Drude fashion at higher frequencies. Ion hopping, characterised by a narrow exponential distribution of activation energies, is shown to generate the measured electrical responses. A model is proposed in which the dielectric component is identified with ions associated with immobile counter-ion or defect clusters and the conductive component with ions at lower activation energy sites between the clusters. It is found that a quantitative account of the conductivity can be obtained by assuming it to arise from a percolation fraction of the ions.