Ii Partial Differential Equations, M. Magnus and S. Nayfeh, Perturbation Methods, J. Thompson and H.
Beightler, Don T. Wilde, Foundations of Optimization M.
Table of Contents
Davis and R. Vinter, Stochistic Modeling and Control R. Deneufville and D. Marks, Systems Planning and Design, W. Dixon and F. Haith, Environmental Systems Optimization, W. Hillier and G. Lieberman, Operations Research M. Cheney, D. Kincaid, Numerical Mathematics and Computing, C. Stoer and R. Forsythe, C. Golub, Charles F. Van Loan, Matrix Computations, A. Strang, Linear Algebra, 2 nd Ed R. Wilkinson, And C. Reinsch, Linear Algebra J. Bellman, R. Kalaba, Quasilinearization R. V4, P , A. Becker, G. Carey, J. Oden, Finite Elements, Vol. I - Vi, K-J. Connor, C. Desai and J.
Crouch, A. Gupta, And Others Approximation of Functions. Kalos and Paula a. Whitlock, Monte Carlo Methods J. Physics General R. Lindsay, H. Margenan, Foundations of Physics, J. Norwood Twentieth Century Physics,. Abraham, J. Marsden, Foundations of Mechanics, 2 nd Ed.
Theses and Dissertations Available from ProQuest | Theses and Dissertations | Purdue University
Goldstein, Classical Mechanics, 2 nd Ed. Hertz, The Principles of Mechanics, M. Jammer, Concepts of Force, L. Lifshitz, Mechanics, 3 rd Ed. Meirovitich, Methods of Analytical Dynamics, N. Neil Rasband, Dynamics, , Wiley A. Sommerfeld, Mechanics, Vol. Stoker, Nonlinear Vibrations D.
Wells, Lagrangian Dynamics, , Schaum E. A Philosophical Treatment Electromagnetism A. Quantum Theory of Atoms and Radiation J. Jackson, Classical Electrodynamics, ; [2 nd Ed. Schiff, Quantum Mechanics, 2 nd Ed. Newell, Nonlinear Wave Motions J. Roseau, Asymptotic Wave Theory U. Sommerfeld, Optics, Vol. Callen, Thermodynamics, E. Fermi, Thermodynamics, J. Gibbs, Scientific Papers, Vol. I: Thermodynamics, G. Keenan, Principles of General Thermodynamics, J.
I and Ii W. Planck, Treatise On Thermodynamics, 7 th Ed. Tr A. Truesdell, Rational Thermodynamics, 2 nd Ed. Prigogine, Dissipative Structures I. Wood, Thermodynamics of Fluid Systems I. Grossman, Thermodynamics and Statistical Mechanics, K. Huang, Statistical Mechanics E. Continuum Mechanics General W. Marsden and T. Hughes, Mathematical Foundations of Elasticity G.
Mase, Continuum Mechanics, , Schaum. Green, W. Zerna, Theoretical Elasticity, L. Math V5, P , I. Solnikoff, Mathematical Theory of Elasticity, A. Timoshenko and J. Fluids General C. Chirgwin and C. Plumpton, Elementary Classical Hydrodynamics, H. Lamb, Hydrodynamics, 6 th Ed. Loitsyanskii, Mechanics of Liquids and Gases, L. Prandt and D. Tietjens, Fundamentals of Hydro- and Aeromechanics, L. Milne Thomson, Theoretical Hydrodynamics, 5 th Ed. Chandrashekhar, Hydrodynamic and Hydromagnetic Stability, P.
Drazin and W. Morton, M. Brines, Eds.
Numerical Methods for Fluid Dynamics S. Peyret, T. Schlicting, Boundary Layer Theory, Tr. Kestin, 6 th Ed. Anderson, Modern Compressible Flow, A. Chapman, W. Waler, Introduction Gas Dynamics, H. Liepmann, A. Shko, Elements of Gas Dynamics, A. Ii L. Ocean Currents Donald T. Greenspan, Rotating Fluids M.
Harbaugh, Introduction To Hydrology, 7. Buckingham, Analytical Mechanics of Gears D. Fuchs and R. Hall, A. Holowenko, H. Laughlin, Machine Design, , Schaum N. Lang, Machine Design Fundamentals, D. Leech and B. Meriam, Dynamics, 2 nd Ed. Sandor, Engineering Mechanics: Dynamics, F. Den Hartog, Mechanics, A. Den Hartog, Mechanical Vibrations, 4 th Ed. Patton, Materials In Industry, 3 rd Ed. Nejat Veziroglu H. Brown, Decentralizing Electricty Production J. Economics, Operation and Maintenance M.
M Dandekar and K. Sharma, Waterpower Engineering J. Green, Solar Cells D. Le Gourieres. Fraas and M. Ganapathy, Applied Heat Transfer Robert a. Gross, Fusion Energy, , Wiley Ted a. Li and a. Potter, Power Plant Theory and Design. Cohen, G. Rogers, Gas Turbine Theory, M. Obert, Internal Combustion Engines, 3 rd Ed. Sorensen, Energy Conversion Systems, C. Warnick, Hydropower Engineering, J.
Weisman, R. Duffie, W. Howell, R. Bannerot, G. Johnson, Wind Energy Systems, P. Seinfeld, Air Pollution, J. Abbott and H. Balschizer and Michael R. Samuels, Engineering Thermodynamics, William Z. Black and James G. Hartley, Thermodynamics, John R. Haberman and James E. John, Engineering Thermodynamics, F. Huang, Engineering Thermodynamics, B. Karlekar, Thermodynamics for Engineers, Dwight C. Look and Harry J. Foster, Thermodynamics: Principles and Applications, W. Reynolds, Thermodynamics, 2 nd Ed. Reynolds, H. Perkins, Engineering Thermodynamics, R. Kenneth Wark, Thermodynamics, 3 rd Ed.
Brown and Rudolph R. Seader, Et Al F. William Payne, Ed. Graphics Engineering and Design Kenneth H. Giesecke, Et Al. Peterson, Mechanics and Thermodynamics of Propulsion J. Pending their investigation and advice, the particular helicopter operator concerned is sponsoring research work to develop an NDI technique to inspect the bond line. The area must be kept clean to facilitate the visual inspection, and the use of a magnifying glass is suggested. The grip was manufactured from an aluminium alloy forging, the hardness being consistent with a T6 temper.
The failed grip is shown in Fig. The macroscopic features of the fracture surface clearly indicate that fatigue had initiated at numerous sites around the circumference of the bore at the root of the thread run-out. Although fatigue cracking progressed outward around the entire circumference of the bore, cracking progressed more rapidly in two directions and formed two semi-elliptical fronts.
The fracture surface is shown in Figs 9. Progressive positions of the crack front were highlighted by the grease deposits, indicating that the grease infiltrated the crack as the crack front extended through the stud holes, prior to final failure. Yellow chromate-containing paint which was present in the pitch horn keyway also appeared to h8ve penetrated into the crack before final failure. The presence of the paint on the fracture surface indicates that the crack front had intersected the root of the pitch horn keyway prior to reassembly at the last overhaul.
The threaded bore of the second grip from the accident aircraft was subjected to a rigorous inspection. A possible crack in the root of the first thread from the thread run-out was indicated by a fluorescent post-emusified dye penetrant inspection. However, eddy current inspection using a miniature probe located in a holder shaped to follow the internal thread indicated that two separate cracks were present in the first thread.
Sections through these cracks are shown in Fig. The two cracks were broken open and were found to be similar, in location, angle, direction of growth and surface characteristics, to the fatigue fracture in the broken grip. The maximum depth of the crack extending toward the pitch horn keyway was 6. Since this was the first known occurrence of this type of cracking in the entire Bell 47 operating history nearly 40 years , it was at first thought to be related to the severity of this aircraft's particular operational role cattle mustering.
However, subsequent inspection of a number of other grips, some of which had been retired from service as fatigue-life-expired, and many of which were still in service, showed that a large proportion contained fatigue cracks. Crack sizes ranged from microscopic, through to massive ie almost at the point of failure , after times in service ranging from hours to the hours retirement life.
These grips came from helicopters in all operational roles, showing that the problem was not confined to cattle mustering. It is interesting to note that, from the known world-wide active Bell 47 fleet size, the known average annual utilisation, and the hour retirement life, the manufacturer estimated that the whole fleet of aircraft should have consumed approximately grips in the previous 5 years.
The number actually delivered, from all sources, totalled During the course of the Australian investigation. Department of Aviation NDI specialists developed an effective and reliable eddy current inspection technique, which is detailed in the Appendix. This required the use of a special probe to fit the internal thread of the grip, as shown in Fig. By implementing eddy current inspections at hours, and repeating at intervals of 6U0 hours, Australian operators are assured of crack-free grips for the full hour life.
A number of other Airworthiness authorities have taken similar action. The cause of the failure is still under investigation. It is possible that the fin separated first and was struck by the tail rotor. Since this was the second report of this type of defect the first was received in , this Department issued an urgent Airworthiness Directive to institute regular inspections of the channels.
At the same time, the manufacturer and the FAA were notified. Typically, the fatigue cracks initiate at multiple origins on the outer surface of the aluminium alloy sheet material, near the edges of the anchor nuts, Fig. No particular stress concentrations are apparent at the crack initiation sites. Microscopic examination of the fracture surfaces has shown that crack growth occurred under variable amplitude spectrum loading conditions. The service bulletin requires a one-time inspection for cracks and a mandatory modification. However, it is felt that the fatigue effects of normal aerodynamic loads cannot be ruled out without more detailed investigation.
Further, there are some reservations as to whether a manufacturer's modification can be relied upon to give adequate protection against future fatigue cracking, especially on those aircraft which had accumulated significant, but undetectable, fatigue damage at the time of its installation. It is therefore intended that the hour repeat inspections specified in the Australian AD will remain until the concerns outlined above are satisfied. The blade failed because of a fatigue crack which initiated from a mechanically produced flaw, and subsequently propagated through the spar cap, as shown in Fig.
The blade failure provides an interesting insight into fatigue related airworthiness practices. With an appropriate scatter factor, the safe life is around hours, in excellent agreement with the present retirement life of hours. To provide adequate airworthiness control for the problem area, the blades in service are now being inspected every 50 hours, up to the compulsory retirement life for the complete blade. Inspections are being made for spar cracking using a low frequency eddy rurrent LFEC technique. Use of this technique is in contrast to proposals from other aviation regulatory authorities which are prescribing a high frequency eddy current technique to detect early skin cracking.
Particular care is being taken to ensure that the LFEC field equipment being used is fully responsive to the sought spar cracking. Mann - ARL Cold expansion of bolt holes was one of the techniques used to improve the fatigue lives of the wing main spars of the KMMr noiittuo mu iiutitet aiicraii. Consequently, a programme of research was undertaken which included finite element and experimental investigations of the stress and strain fields around cold-expanded holes in thick sections of aluminium alloy complemented by fatigue testing, crack propagation studies and fracture analysis.
The particular process studied was the split-sleeve cold expansion process marketed by Fatigue Technology Inc. Aluminium alloy A7- U45G equivalent to was used as the test material. Figure 9. This is shown in Fig. Measurements of the z-direction out-of-plane displacements at the mandrel inlet and outlet faces have indicated, for 15 mm thick specimens, that these are considerably less at the inlet face than the outlet face, Fig.
In a multiple- stack hole cold-ex pans ion situation the z-direction displacements depend on the materials in contact. Furthermore the z-direction displacements are significantly less for 8 mm thick specimens than for those of 15, 23 and 30 mm. In these joints the aluminium alloy in the centre was 15 mm thick. The increase in fatigue life as a result of cold expansion is significantly less for 8 mm thick specimens compared with those of the other thicknesses - against relative to non-cold-expanded holes.
There a clear evidence that, ii the latter group, cracking propagates more rapidly at the mandrel inlet face of the specimen than at the outlet face in the letter stages of crack growth. Compared with non cold-expended holes the life ratio was between 6 and 7. This is very surprising in view of the severe deformation adjacent to the ridge along the hole. Much smaller increases in life a factor of about two were found for the bolted-joint specimens, and the differences for the two types of specimen were attributed to the detrimental effects of fretting in the latter.
Little difference in total lives to failure was found between specimens with holes cold-expanded before fatigue testing and those cold-expanded after the development of small fatigue crack. There is also some preliminary evidence that, for similar failing loads, the fatigue cracked areas of specimens with cold-expanded holes are considerably greater than those with non-cold-expanded holes. These findings are of some importance in damage and durability analysis and are to be investigated in more detail. Jost, R. Carey - ARL, G. Steven - UOS The application of finite element analysis to cold-expanded holes, reported in the previous Review, has been continued and expanded.
At ARL a hole in a large circular aluminium alloy plate having a bilinear stress-strain relationship has been cold expanded first to that point 1. Examples of graphical output are shown in Figs 9. The work is being repeated, in the first instance, for the three-dimensional case where the thickness is equal to the hole diameter. In a parallel investigation at the University of Sydney which included substantial software development, finite-element analysis is being used to determine stress and strain fields in a rectangular plate containing a cold-expanded hole.
In this case the Ramberg-Osgood stress-strain characteristic closely represents that of an aluminium alloy. Immediate next stages in this study will include the influence of remote loading and pin fastener loading in conjunction with neat-fit and interference-fit fasteners in the cold-expanded hole. Heller, R. Jones - ARL, J. Williams - UOM In aircraft wings the aerodynamic loads are transferred from the wing skin to the spars by means of bolted and rivetted joints. These joints are one of the main causes of cracking and often determine the fatigue life of the structure. Accordingly, this project is investigating the on-site repairability of holes containing existing, real cracks by bonding into the hole a stiffer steel sleeve using epoxy resin and avoiding the current necessity of having to return the structure to a major depot for reworking.
The use of a sleeve, of course, minimises the opening and closing of the crack under repetitive fatigue loading and also allows the rivet or bolt to be inserted or removed as desired. The reason for doing this was to establish that the method will work for holes containing existing cracks and to establish the maximum crack size that can be successfully repaired by this method. Secondly, a detailed stress analysis of the repair procedure is being undertaken to identify the key parameters with the aim of optimizing the method.
As expected, the stress analysis, using finite element techniques, has indicated that adhesive bonding significantly reduces both the local stress concentration at the hole and the stress intensity at the crack tips thus retarding the growth rate of potential fatigue cracks. Williams - DOM, R.
- Division Collections;
- Educating English Language Learners: A Synthesis of Research Evidence;
- Bibliography for Tin Whiskers.
- Slave Narratives (Library of America, Volume 114).
However, moat failures are three-dimensional in nature. Several methods are now available for evaluating the variation of S and K around the crack front. These methods are being investigated and compared to experimental and numerical values for a typical surface flaw. Methods for calculating the energy release rate, G, in three-dimensions, now appear to require a critical overview. It seems that the value of the local G may depend on the way in which it ia evaluated, namely, should G be calculated over a finite but small area or, as is currently done, at a point?
The instrument was designed to detect the small temperature changes occurring during cyclic loading of a specimen. The thermoelastic constant relates these temperature changes to the changes in the sum of the principal stresses. A theoretical explanation of the mean stress effect has been developed This dependence suggests a potential application of the thermoelastic effect in the measurement of residual stresses.
Recent work has concentrated on improving the data transfer rate to the main frame computer and development of analytic programs. These optical methods are complementary to the use of direct measurements of the displacement of fine grids, a technique developed for the study of the plastic strains in the vicinity of crack tips and cold-expanded bolt holes see Section 9. The method involves the application of a very fine square microgrid pattern typically 25 pm line spacing to the area of interest.
The local plaatic strain distributions, produced by either static or cyclic loading, are determined from replicas of the deformed grid pottern taken whilst the specimen is under load. The grid point displacements are measured from the replicas using an optical microscope end image analysing system, and are converted to strains using a computer which outputs the data in a variety of graphical or numerical forms. This microgridding and replicating technique has been used in several areas of research into the fatigue and fracture behaviour of aluminium alloys used in the aircraft industry.
This parameter is a measure of the intensity of the plastic stress and strain field at the crack tip. The Jjq value for a medium strength aluminium alloy, T, was determined using a theoretical model for representing crack tip plastic stress and strain fields, and the strain values determined from replicas.
Fig 9. This Jjq value was compared with the values obtained from two numerical methods and the ASTM standard test method. Agreement between these values was good, Table 9. C values are due to material variability and experimental scatter. An important part of the ASTM standard Jjq test method is the measurement of the amount of crack extension as a function of toad. The usual method of marking crack fronts in aluminium alloys for subsequent measurement is to fatigue cycle after the initial static crack extension. This method was more effective than the usual fatigue cyclic method for the aluminium alloys examined in this work.
This relationship is independent of the R-value. Work is in progress to use the crack tip strain data to model fatigue crack growth in metals and components. Effects such as delayed retardation in growth rates as a result of overloading will be included in the model. These surveys indicate significant biaxial load effects on the fatigue and fracture behaviour of both types of materials. Cox - ARL Continuing concern about the presence of fatigue cracks in critical components in F-lll aircraft of the RAAF has led to a need for information to be acquired about crack growth rates and characteristics.
The components of interest are large forged, welded and heat-treated wing pivot fittings which attech the wings to the carry-through box in the fuselage centre-section, Fig 9. These fittings are manufactured from a high-strength steel which by virtue of its low fracture toughness can tolerate cracks only a few millimetres in depth.
Inspection programs have been introduced which will detect the presence of cracks with surface lengths of only a fraction of a millimetre, although the size and complexity of this component makes development and implementation of NDI programs particularly difficult. It is essential to determine the growth rates of cracks which have been detected if appropriate inspection intervals ere to be established. This information has been acquired by the use of quantitative fractography - by correlating the fracture surface fatigtie markings with the known load history of the individual aircraft, the crack size and shape may be established at a number of points and times corresponding to known events in the flight load history.
Typical results obtained by matching fatigue progression markings with flight numbers is shown in Fig. Four sequences, each at two stress scaling levels, were applied to the specimens.
The basic load sequence was a flight- by-flight fighter sequence and, after range-pair counting, three other sequences were generated each giving the same range-pair table. One reconstituted sequence was fully random, a second aimed to accelerate crack growth, while the third aimed to retard crack growth. Over the four sequences and two stress scale levels the order of increasing predicted crack growth life is practically the same, namely Unretarded, Wheeler, Modified Willenborg, Willenborg, Crack Closure, as illustrated in Fig. The Wheeler model is preferred overall because : a it gives the best predictions of crack growth for the more random random and flight-by-flight sequences, and is therefore the most suitable for many practical applications.
None of the models correctly predicts the trends in experimental crack growth obtained over the four sequences. In broader terms the same can be said of the two more-ordered sequences, acceleration and retardation. However, experimentally, the acceleration sequence gave the longest crack growth life and the random sequence gave the shortest, yet alt of the models predict the reverse trend.
Previous work had shown that, for bolted specimens, there was no difference in total life for sequences reconstituted to either accelerate or retard crack growth. Centre-cracked specimens of T aluminium alloy were tested under the original flight-by-flight sequence, a fully random reconstituted sequence, a sequence to accelerate crack growth sequence A2 and a sequence to retard crack growth sequence R2.
Three specimens were tested under each sequence at each of two stress levels The conclusions that may be drawn from these results are as follows : 1. At long lives about 40 programmes or 20, flights at At shorter lives about 10 programmes or flights at Crack growth lives under the flight-by-flight and random sequences are practically identical.
Crack growth under the more-structured sequences A2, R2 is significantly slower than under the less-structured more-fluctuating sequences random and flight-by-flightX 4. The maximum spread in average crack growth lives is between those from the acceleration and random sequences at both stress levels and is an average of 1. The dominant factor producing the small variations in crack growth life was the frequency of load level changes, and the sequences designed for crack growth extremes were deficient in this respect.
Preliminary frectographic analyses indicate that local retardations and accelerations in crack growth did occur under sequences R2 and AZ but they tended to average out over the lengthy sequences used. Ford - ARL A computer program is being developed at ARL to predict the fatigue life distribution of a structure with several interacting critical regions.
For each of these regions the fatigue process is divided into pre- and post-cracking stages, and existing cracks affect fatigue in other regions through stress redistribution. The life distribution follows from a reliability analysis of the range of possible crack-growth histories of the structure.
Its implementation has exposed several numerical problems but also allowed significant generalisations which would be intractible in an analytical treatment. The generality of the program, which reflects the variety in the fatigue process, means that more information must be fed in to obtain results. In present practice much of the same information is supplied at different stages of an analysis but important effects can be missed thereby. Additional input specific to the program describes stress models used, the interaction between cracks and strength or fracture in the presence of several cracks.
It was found that the contributory models in the program could be related to loads, S-N data and crack growth. Information under these headings is supplied for each crack and the use made of it depends on the compatible choice of user-supplied or internal models. The general philosophy of internal modelling is to fit a robust model to the supplied data and then fit bicubic splines to residual effects. For this purpose optional internal models are included for low cycle fatigue and crack growth. It was also found necessary to devote considerable attention to range pair data in order to adequately smooth load sampling.
The program has special provisions for repeated use of the same or similar data at several crack sites. Interaction between cracks implies that the development of failures can only be followed dynamically. In the current version of the program short-term local randomness of cracks is neglected and the mathematical problem reduces to a first order continuity equation for crack length probability.
This is not determined explicitly, but the characteristics are followed using standard Adams-Bashforth integration over equal time intervals. To reduce fatigue-type computation special starting procedures were developed, avoiding partial steps but starting with second derivatives in a one-step predictor-corrector routine. Inspections can increase this number. The generalisations with computational solutions are the inclusion of inspections and the confluence of cracks to form fewer, more dominant failures. Obviously the geometric end stress-dependent details for these must be user-supplied.
Other required subroutines describe the stress fields with crack interaction, the corresponding strength, repair schemes and inspections. In principle, direct stress analysis and fracture mechanics could be included but it is felt that summarised information would be more practical and efficient. Ford - ARL In fatigue, range-mean-pair or rainflow counts correspond to closed hysteresis loops so that their distributions affect initial lives in low cycle fatigue.
They are also useful empirically in general life and crack rate prediction. Many fatigue problems arise from random loading of linear structures for which the stress response is often stationary and Gaussian. Examples are oil rig platforms, vehicle suspensions and turbulence loading of aircraft. This implies that the relation between range-pair counts and the power spectrum or autocorrelation of a stationary Gaussian process is a central problem in fatigue prediction which has been addressed at ARL. In this the continuous process is treated as the limit of a discrete process. However time or frequency information also requires a third crossing to define a recurrent event.
The general procedure is therefore to predict the frequency of triple crossings equivalent to the range-pair exceedance. The length of each such event consists of two first passages from one defining level to the other. With suitably conditioned probabilities these are again established from recurrent event theory. A Pascal program is being developed for estimating exceedances for range-pair tables. Baker - ARL Developments have continued in crack patching technology for fatigue or stress corrosion cracked metallic components. As first described in the Australian ICAF review in , crack patching technology is generally based on boron-epoxy patches and structural film adhesives.
The approach taken to estimate minimum patch thickness, illiatrated in Fig. The composite patch is increased in thickness one ply about 0. In this design approach no attempt is made to design to a specified stress intensity. The analysis accounts for residual stress Oy resulting from expansion mismatch between the patch and cracked component and curing temperature range AT.
The results of an example analysis are provided in Fig. It should i retard the reinitiation of the crack, and ii reduce the rate of crack growth once growth resumes. Essentially, retardation is associated with the formation of a plastic zone at the crack tip during loading. The plastic zone acta aa an oversize inclusion, resulting in a compressive residual-stress from the surrounding elastic material when the external loads are removed. However, the opening of the crack tip will be very much less than would normally occur at K R due to the compressive residual stress field.
The onset of further plasticity at the crack tip will similarly be greatly reduced. Even under service variable loading, some retardation effects would be expected, depending on the stress level imposed on the component immediately prior to patching. Following patching, and after retardation effects on growth are exhausted, the reduction in the rate of crack growth reflects the reduction in stress intensity range.
Finally, two potentially important complicating factors must also be considered. The first is the presence following patching of the thermally induced residual stress and the second is the influence of heat-treatment following patching on the behaviour of the cracked component. Crack propagation studies Edge-notched specimens, as depicted in Fig. In this work two similar specimens are simultaneously tested while joined together to form a honeycomb panel, with the patched sides facing outwards. The aim of this configuration is first to minimise curvature, caused by the residual stress following patching: patches are usually bonded at the same time as the panels are bonded to the honeycomb core.
The second aim of the configuration is to minimise bending of the panels which would otherwise occur during testing; the moments which cause bending arise from the displacement of the neutral axis of the metal panel by the patch. The support provided by the honeycomb panel configuration is considered to be a reasonable simulation of the support that would be provided in typical military aircraft structure. Crack growth in the patched panels was monitored through the patch using an eddy current procedure and, following testing, the panels were subjected to X-ray radiography using an X-ray absorbent fluid, tetrabromoethane to detect disbonding or delamination of the patch.
In most panels no evidence of significant disbonding or delamination was found - this finding was later confirmed by metallographic observations on prepared sections. Some of the findings and conclusions from these crack propagation studies are as follows: i Patching efficiency is very high, as shown in Fig. This difference in behaviour between the two crock sizes can be explained as follows: the degree of retardation of crack-growth following patching depends on the reduction in stress intensity range fromAK a to 4K R. Since the minimum stress intensity range AK R is expected to be independent of crack size, the reduction in stress intensity range for the large crack and hence the degree of retardation is much greater than that for the small crack.
This behaviour is evident in Fig. Further reference to Fig. Hoskin, A. Baker, R. Whilst ARL has considerable experience in the use of bonded composited repairs for metal aircraft structures, the present development it more advanced than any undertaken in the past, primarily because these doublers are much thicker than any used previously. Extensive finite element design studies have been undertaken and a detailed evaluation of the materials engineering aspects has been made.
Procedures for applying the doublers to the wing in situ have been developed and particular attention hat been given to thermal mismatch problems. Using a rig, Fig. It has been confirmed that substantial reductions in the strains in the fatigue critical regions are achieved with the doublers. This development is still in progress. Watters - ARL A 2. The cross-section is shown in Fig. The main aim of the test is to examine the growth of barely visible impact damage in the composite skin, including the effects of moisture absorption and temperature excursions.
The loading configuration is three-point bending causing compression in the composite skin. Prior to fatigue testing, the mid-span region of the composite skin was extensively strain gauged and a static strain survey was performed. The composite skin was then deliberately damaged at a total of seven locations in the mid-span region by impacting it with a free-falling mass.
Impact energies ranged from 12J to J9J and impact locations were over spars, adjacent to spars, and midway between spars. In all cases the damage fell into the barely visible category with only small indentations being visible on the outer surface. The extent of sub-surface damage measured by C-scan is shown in Fig. After being damaged, the box-beam was subjected to one lifetime of fatigue loading 30 blocks of the FALSTAFF sequence or hours under laboratory ambient conditions. No damage growth was detected. Following moisture conditioning fatigue loading will be resumed.
This second phase of fatigue loading will be performed under a controlled temperature spectrum which includes cold, room and high temperature segments and a significant number of freeze-thaw cycles. Under high load level and high temperature, load dwells will be included to simulate real-time loading rates.
The segment of the FALSTAFF sequence to be used for high temperature testing will be chosen to maximise the correlation between high loads and high temperature. Williams, T. Tay - UOM It is well known that in fibre composite materials the location of maximum stresses, and the failure mechanisms associated with these stresses, occur in quite different regions and by different modes than is predicted by conventional theories for isotropic, homogeneous materials. In this case, "bearing" failure is not to be associated with the normal compressive overload situation in isotropic materials.
The Tsai-Hill and the Tsai-Hoffinan failure criteria are being applied to the problem and will be compared to results from a variety of experimental tests. One aim of this investigation is to examine the effect of edge proximity on the stress distribution surrounding the hole. The Bibliography can be used by physicists, scientists, and materials engineers to gain access to a wide variety of books, papers, and research on the above subject. We are always looking for ways to improve customer experience on Elsevier. We would like to ask you for a moment of your time to fill in a short questionnaire, at the end of your visit.
If you decide to participate, a new browser tab will open so you can complete the survey after you have completed your visit to this website. Thanks in advance for your time. Skip to content. Search for books, journals or webpages All Pages Books Journals. View on ScienceDirect. Authors: J. Imprint: Pergamon. Published Date: 1st January Page Count: