In this study, different surface pretreatments were applied to clean and activate titanium alloy surfaces. The samples were subjected to grit blasting treatments using two different pressures and afterwards, a UV/Ozone treatment was applied at different times to study the wettability and surface oxidation of the titanium samples. Scanning electron microscopy and laser confocal microscopy showed the surface morphology and the increased roughness with grit blasting pressure. X-ray Photoelectron Spectroscopy revealed that titanium was increasingly oxidized with increasing UV/Ozone treatment time, which leads to a reduced contact angle and a better adhesive performance in a butt tension test proving the effectivity of this surface treatment for titanium. Furthermore, the addition of sol-gel AC-120 and corrosion inhibition primer BR 6747 showed to be an additional improvement in the initial adhesion and after different degrees of aging by exposure to salt-spray, making the surface treatment techniques used in this research, a promising environmental friendly alternative to improve adhesive bonding performance.
Single-lap shear (SLS) joints are straightforward to manufacture. This makes them especially attractive for testing polymer composite welded joints. Owing to local heating, which is characteristic of composite welding processes, the production of more geometrically intricate joints (such as double-lap or scarfed joints) or bigger joints (such as end-notched flexure or double cantilever beam) typically entails significant complexity in the design of the welding process. Testing of SLS joints is also uncomplicated and, even though, owing to mixed-mode loading and uneven stress distribution, it does not provide design values, it is widely acknowledged as a valuable tool for comparison. Even so, comparing different aspects of composite welded joints through their corresponding SLS strength values alone can be deceptive. This paper shows that comparison of different welding processes, adherend materials, process parameters or different types of joining techniques through SLS testing is only meaningful when strength values are combined with knowledge on other aspects of the joints such as joint mesostructure, failure modes and joint mechanics.
This article is part of a discussion meeting issue ‘A cracking approach to inventing new tough materials: fracture stranger than friction’.
The purpose of this research study is to analyse the effect of the design of mechanical fastener on their effectiveness as Disbond Arrest Features (DAFs). This was done by means of a sensitivity study using a 3D model using the Virtual Crack Closure Technique (VCCT) to model the disbond growth of a single lap shear specimen comparing the Strain Energy Release Rate (SERR). Different fastener material stiffnesses, shaft radii, head radii and head geometries were compared. The sensitivity study showed the Mode I SERR was completely independent of the fastener design and the Mode II was suppressed more strongly for fasteners with a lower flexural flexibility. Fasteners with a countersunk head showed less reduction of the Mode II SERR, this was caused by the head rotating as a result of the fastener adherend interaction, reducing the effectiveness of the load transfer.
This research focuses on the possible causes of variation in the tensile strength and stiffness of witness specimens that are used in the manufacturing of aerospace grade selective laser sintered parts. Using specimens and data obtained from Materialise of the PA2241FR samples, the crystallinity and porosity of the samples was determined. The effect of the flame-retardant additive was also investigated. A relationship was found between the degree of crystallinity and the tensile strength, and between the porosity and the tensile strength. Unfortunately, no relationship could be found between the crystallinity and stiffness. The effect of the FR-additive could not be tested using NMR and as a result no data on the influence was available. When analysing the other factors, they seemed to indicate the significance of an unknown parameter on the mechanical properties and especially on the stiffness. A variation of the halogenated flame-retardant additive seems to fit this effect.
Meta-biomaterials offer a promising route towards the development of life-lasting implants. The concept aims to achieve solutions that are ordinarily impossible, by offering a unique combination of mechanical, mass transport, and biological properties through the optimization of their small-scale geometrical and topological designs. In this study, we primarily focus on auxetic meta-biomaterials that have the extraordinary ability to expand in response to axial tension. This could potentially improve the longstanding problem of implant loosening, if their performance can be guaranteed in cyclically loaded conditions. The high-cycle fatigue performance of additively manufactured (AM) auxetic meta-biomaterials made from commercially pure titanium (CP-Ti) was therefore studied. Small variations in the geometry of the re-entrant hexagonal honeycomb unit cell and its relative density resulted in twelve different designs (relative density: ~5–45%, re-entrant angle = 10–25°, Poisson’s ratio = -0.076 to -0.504). Micro-computed tomography, scanning electron microscopy and mechanical testing were used to respectively measure the morphological and quasi-static properties of the specimens before proceeding with compression-compression fatigue testing. These auxetic meta-biomaterials exhibited morphological and mechanical properties that are deemed appropriate for bone implant applications (elastic modulus = 66.3–5648 MPa, yield strength = 1.4–46.7 MPa, pore size = 1.3–2.7 mm). With an average maximum stress level of 0.47 σy at 106 cycles (range: 0.35 σy– 0.82 σy), the auxetic structures characterized here are superior to many other non-auxetic meta-biomaterials made from the same material. The optimization of the printing process and the potential application of post-processing treatments could improve their performance in cyclically loaded settings even further.
Since welded joints in structures are known to be critical regions for crack initiation and growth, study of fatigue crack growth behavior of welded joints under various loading scenarios are essential. In present study, two groups of experiments were conducted. At first, fatigue crack growth behavior of Al5083 butt welded joint subjected to mixed mode loading with constant amplitude has been studied. Afterwards, fatigue crack growth behavior of butt welded joints subjected to mixed mode single overload has been investigated. The sensitivity of the crack growth retardation behavior to post-welding heat treatments was also studied. Considering the effect of residual stress on fatigue crack growth, a modification to the Wheeler model was proposed to improve the accuracy of retardation prediction after applying a mixed mode overload.
Welding is a joining method for thermoplastic composites (TPCs) that offers multiple advantages over the more traditional methods of mechanical fastening and adhesive bonding. A particularly promising welding technique is ultrasonic welding, which features very short process times as a result of the high heating rates that can be achieved. This spot welding technique is hypothesized to have potential for improved damage tolerance compared to more commonly used continuous welding techniques: in a multi-spot welded joint, evolving damage will need to re-initiate in subsequent spots. The fact that damage initiation will need to occur multiple times might delay overall damage evolution through the joint compared to a continuous welded joint, where damage initiation needs to occur only once. This work is a first exploratory step into the domain of fatigue of multi-spot welded joints in TPCs. Existing research on the fatigue behavior of four-spot welded steel joints in various layouts served as the main reference throughout this research: its methodology was transferred to four-spot welded joints in TPCs. By comparing fatigue behavior across both materials, it was evaluated to what extent existing knowledge and design rules for steel could potentially be transferred to TPCs. Differences were observed in the results obtained for TPC and steel joints. Most notably, in steel joints the dominant failure mode was seen to change from spot fracture to sheet fracture at higher fatigue lives. In TPCs, joints consistently showed spot fracture across all load levels. A different interrelation between layout performances was seen in the steel and thermoplastic composite joints, assumed to be a result of localized material strengthening in the steel joints from interference of adjacent heat-affected zones. These results indicate that existing knowledge on multi-spot welded joints in steel cannot be readily transferred to TPCs, as failure modes and material mechanisms may differ. It was discovered that, when one spot failed prematurely as a result of existing damage in the joint, the remaining layout no longer seemed to have an effect on fatigue life performance. This was attributed to asymmetry in the remaining joint layout, meaning one spot would always become a preferred location for damage initiation and subsequent evolution. Therefore, subsequent damage evolution would only be restricted by a single spot up to the point where the shear strength of the joint was exceeded.
Surface morphology of adherends is an important factor to take into consideration when studying and improving the performance of an adhesive bonded joint. In this study, the adhesion performance three different surface morphologies of Selective Laser Melted (SLM) Ti6Al4V was studied. The three surface morphologies were created by manufacturing the adherends with different build directions (0, 45 and 90°). Scanning electron microscopy and laser confocal microscopy were used to assess the obtained morphology and roughness of the printed surface areas to be bonded. Those surfaces were subjected to 40 min of UV/Ozone treatment to remove organic contamination traces on the surface which lead to a reduced apparent contact angle and improved adhesive strength. The samples printed at 45°, which showed the highest surface roughness, presented the best adhesive performance during the tensile tests. The addition of sol-gel AC-120 and corrosion inhibition water-based primer BR 6747-1 showed an effective improvement in aging behaviour after 6 weeks of salt spray exposure.
To expand the use of additive manufacturing in aerospace towards more critical applications, it is required to design parts in a damage tolerant context. Therefore, the damage tolerance of additive manufactured multiple load path structures is assessed by analysing the fatigue life and damage propagation of components with increasing redundancy. An experimental approach is chosen, whereby specimens with 1, 9 and 81 parallel struts are tested. A decreased fatigue life is found for the specimens with more but thinner struts. This decrease is attributed to manufacturing related effects that occur upon producing smaller elements. The failure of the multiple load path structures showed a step-wise pattern. Due to this, the decreased variation in fatigue life and decreased sensitivity to initial damage, multiple load path structures are more damage tolerant. However, in design a balanced decision should be made upon applying these structures, due to the decreased fatigue life.
The application of strain gauges as recommended by the ASTM standards provides accurate strain measurements in isotropic materials. However, their use in composite materials becomes more challenging due to their anisotropic nature. In this study, we hypothesized that the use of the distributed sensing system and the three-dimensional digital image correlation, which can average strain along a line and surface, respectively, may account for strain variability in composite materials. This study shows an investigation on the mechanical properties of unidirectional, cross-ply, and angle-ply carbon-epoxy specimens using strain gauges, distributed sensing system, and digital image correlation. The Bhattacharyya distance method was used to provide a preliminary evaluation of the closeness of the three different measurement techniques while the B-basis statistical method was used to analyze the experimental data in order to obtain a more conservative and reliable material parameter compared to the conventional averaged value, recommended by ASTM standards. Finally, a finite element model was created in Ansys Workbench™ as a means of evaluating the implication of a single point strain gauges measurement, versus a line or a surface strain measurement. The finite element analysis investigation was performed at a laminae level using the measured experimental elastic modulus and at a lamina–lamina level in which the elastic modulus of the unidirectional case was used as input in all the laminate configurations. The former analysis showed good agreement between the finite element analysis and all the strain measurement systems with an averaged percentage difference below 5%. The latter analysis showed a higher discrepancy in the measured percentage difference. A comparison between the finite element analysis and the strain gauges measurements showed an overall percentage difference between the range of 10% and 26%. Distributed sensing system and three-dimensional digital image correlation measurements provided an overall percentage difference below 10% for all the specimen configurations with a maximum percentage difference recorded for the longitudinal angle-ply case of approximately 9%.
Traditional design methods are generally unsuitable for optimally designing organic shapes made possible by additive manufacturing. In this study, a simple Genetic Algorithm (GA) optimisation routine was developed for a relevant engineering design problem – the optimisation of thickness distribution for a crenelated fuselage skin panel. The basis for this optimisation is the damage tolerance behaviour of the panel in the presence of a fatigue crack. The results demonstrated that crossover and mutation are inherently more similar than expected, thus questioning whether it is not more important to design a set of search heuristics through better understanding of the fitness space, rather than the application of a flawed, nature-inspired standard crossover and random mutation. Through these insights, this research contributed to ongoing research in understanding GAs, which, if better understood, could assist engineers in finding improved designs of additively manufactured components.
Although aerospace traditionally has always had a multidisciplinary approach to engineering and design, the increasing complexity of aircraft and spacecraft and the rapid digitization within the aerospace industry has led to a large number of related engineering and scientific disciplines such as electrical engineers, computer scientist to work much more directly within the aerospace domain than before. Next to that there is a shortage of highly trained engineers worldwide to meet demand. As a result, there is a clear need to provide basic knowledge to non-aerospace engineers working in the field and to motivate and attract more people to engineering and aerospace in particular. This paper details how the creation of a Massive Open Online Course (MOOC) at an introductory level in Aerospace Structures and Materials provides an efficient and fit-for-purpose tool to achieve both aims. The paper will discuss the course design, the course set up, the course evaluation and how the course fits within the online learning philosophy of Delft University of Technology. It will use learning analytics to analyze our learners and their needs.
To create a more sustainable future for aviation, new, lighter-weight structures and materials will need
to be engineered. It will also be critical that damage tolerance and safety are not compromised in
the process. Lattice materials represents one avenue of exploration; however, two key challenges
arise: limited experimental work has been conducted to date regarding tensile mechanical response
and lattice materials are generally considered to be less tough than traditional aerospace materials.
Advancements in additive manufacturing in recent years creates the opportunity to rapidly produce
high-quality complex geometries, allowing for both challenges to be more easily investigated. To address
the issue of toughness and damage tolerance, nature is a source of inspiration, as all of nature’s
toughest materials derive this characteristic from creating structural hierarchy using intrinsically weak
building blocks.
Two sets of lattice structures were fabricated using stereolithographic (SLA) 3D printing and tested
under quasi-static tensile loading. Two sets of lattices were fabricated: lattices with uniform strut thickness,
or relative density, and mixed-relative density lattices which create structural hierarchy. Using a
novel method to track lattice deformation during loading, lattice stiffness-displacement response has
been correlated with beam elongation and rotation behavior and the deformation of individual cells.
The stiffness-displacement response of uniform lattices can be classified by relative density as either
an elastomeric, elastoplastic, or hybrid response. In hierarchical lattices, cell deformations occurring
in different relative density regions are directly correlated to features of the stiffness-displacement response.
Aspects of the mechanical response of hierarchical lattices, particularly fracture toughness and fracture
pattern, are heavily influenced by the exact configuration of structural hierarchy, spurring a discussion
of what characteristics are most important in the pursuit of increased lattice damage tolerance. While
none of the lattices represent an optimal solution, each displayed characteristics which, if combined to
form a hybrid structure, could substantially improve lattice damage tolerance.
Understanding and characterizing crack growth is central to meeting the damage tolerance and durability requirements delineated in USAF Structures Bulletin EZ-SB-19-01 for the utilization of additive manufacturing (AM) in the sustainment of aging aircraft. In this context, the present paper discusses the effect of different AM processes, different build directions, and the variability in the crack growth rates related to AM Ti-6Al-4V, AM Inconel 625, and AM 17-4 PH stainless steel. This study reveals that crack growth in these three AM materials can be captured using the Hartman–Schijve crack growth equation and that the variability in the various da/dN versus ΔK curves can be modeled by allowing the terms Δ K thr and A to vary. It is also shown that for the AM Ti-6AL-4V processes considered, the variability in the cyclic fracture toughness appears to be greatest for specimens manufactured using selective layer melting (SLM).
Fatigue is a major cause of failure in several industries, and in many practical cases, local mixed-mode conditions prevail at the crack front. The effect of plane mode mixity on the crack growth rate and crack growth direction has been investigated. Fatigue crack growth experiments have been conducted on aluminum alloy Al5083-H111 for several mode mixities. A fixture was manufactured in order to apply the different combinations of mode I and II by changing loading angle. Afterward, three-dimensional simulations have been implemented using the Zencrack software. Based on numerical simulations, new relations are proposed to estimate stress intensity factors for compact tension shear geometry by modifying Richard’s equations [1].
Redistribution of residual stress during crack growth in butt welded joint of Al5083-H111 was studied. Furthermore, the effect of post weld heat treatment and removing the weld reinforcements on fatigue crack growth rate was investigated. Three dimensional simulations of welding process and fatigue crack growth were conducted. It was shown that removing the weld reinforcements causes an increase in the residual stress. A correlation between the fatigue crack growth rate data of welded joints can be obtained with that of the parent metal by considering the weld geometry and redistribution of residual stresses in calculation of stress intensity factor.
The effect of mixed mode overload on the retardation of pure mode I cyclic loading has been investigated experimentally on Al5083-H111. Two models for extension of the Wheeler retardation model are proposed in order to calculate retardation of fatigue crack growth subjected to mixed mode overload. The first proposed model can predict retardation with high precision. The second proposed model is a simplified form of the first one and does not contain more experimental constants than the Wheeler model so, it may be more practical. The application of the proposed models has been examined for another material and testing conditions.
A new certification approach for bonded primary Fiber Metal Laminate (FML) structures is investigated: using bolts as Disbond Arrest Features (DAF)s to contain the growth of bond line damages so that they can be found and repaired by inspection before becoming critical. By fatigue testing with coupon specimens and model analysis, it has been demonstrated that reducing the Mode I Strain Energy Release Rate (SERR) is the main driver for arrest. The peak stress associated with a disbond front can initiate adherend fatigue cracks during slow growth. The effect of adherend fatigue cracks on the arrest of disbond growth could not be determined and must be investigated in future work. In the process, a novel quasi-analytical disbond growth model has been developed and validated. An algorithm is developed and verified that utilizes the strain field measured by Digital Image Correlation (DIC) to locate the disbonded region.
Complex structural shapes can be produced with additive manufacturing. The geometrical complexity that can be achieved translates into an increase in possible designs. Designing these structures with traditional methods is difficult. A design process with computational optimization will enable engineers to use the geometrical freedom offered by these manufacturing methods. This study explores how topology optimization can be used to design structures that are fatigue tolerant. Two optimization algorithms for fatigue tolerance were developed in this thesis. One algorithm minimizes the stress intensity factor, whereas the other one maximizes the fatigue crack growth life. Both algorithms use a resource constraint to limit the total amount of material, an enriched finite element method to analyze the crack growth performance and the method of moving asymptotes to incrementally improve the design. Example problems showed that the algorithm dramatically improves the fatigue resistance.
Additive manufacturing allows material structuring, supporting the fabrication of multiple-level structures or metamaterials. Through the lens of classical stress reduction, nature’s cellular solid structures feature stress-homogenizing nodal topologies. Avian long bones are an example. Research into the mechanics of open cell cellular solids seems focused on the effectiveness of unit cell architecture and neglects the detailed behavior of constituent nodes. Several specimen series were printed on the nodal- and cellular solid-levels of analysis, all with varying nodal topologies. A discussion of force-displacement and digital image correlation experimental data is had; the cellular solid deflection rigidity seems highly sensitive to nodal topology under quasi-static compression. It is thought that bioinspired profiles successfully homogenize stress and improve load transfer, mitigating nodal softening: peak stresses and the propagation of nodal torsion into adjoining strut deflection decreased. This sensitivity is relevant for lightweight strain energy absorption and stiffness provision, and demands further research.
The present study reports on the evaluation of residual stress field formation and distribution in Central Cut Plies (CCP) specimens. Real-time measurements were performed using a distributed sensing fiber optic system based on Rayleigh Backscattering, which was successfully able to capture strain distribution inside the adhesive layer at every 0.65 mm during the entire curing cycle, for both unidirectional and cross ply laminate configurations. A finite element analysis was also performed to cross-correlate with the experimental residual strain distributions in the proximity of the severed central cut plies. The results outlined in this study demonstrate the presences of tensile residual stresses within the adhesive layer for both configurations. A full field strain distribution and the significance of these findings in relation to the use of the CCP test for fracture mechanics testing will be discussed. Results of this study have shown that residual stresses arise after the curing process for which the amount of longitudinal and transverse residual stresses for the unidirectional CCP laminate are 61% and 19% of the total strength of the adhesive respectively, while for the cross-ply CCP laminate are 72% and 71%, respectively.
In previous work, single-spot ultrasonically welded joints were found to feature similar load carrying capability in shear but significantly low capability in peel as joints with a representative single-mechanical fastener. This leads to questioning welding as an appropriate solution for the commonly-used single-lap joint configuration. The present paper investigates the mechanical performance of spot welded single-lap joints in thermoplastic composites in comparison to their mechanically fastened counterparts. Single-row joints, double-row joints with varying inter-row distance and multi-row joints with varying number of rows were investigated in this study. The results showed that, owing to higher joint stiffness and hence lower secondary bending and peel stresses, the load carrying capability of the spot welded joints was comparable to that of the mechanically fastened joints in all considered cases. Likewise, the effects of increasing the inter-row distance and of increasing the number of rows were similar for both types of the joints.
The lack of a widely-accepted test standard for characterizing the mode II fatigue disbond growth behavior of adhesively bonded interfaces is a challenge to the research community in terms of producing consistent and repeatable results. Typically, researchers apply the End Notch Flexure specimen, which is already used for static delamination studies. However, the needs for static and fatigue disbond growth characterization are not the same, resulting in some undesirable effects in such specimen. This study looks at a particular mode II test configuration known as the Central Cut Plies (CCP) specimen. A critical evaluation of the suitability of this specimen, including the influence of geometry, disbond measurement approaches and the stability of the disbond growth is carried out through a combination of numerical and experimental investigations. A distributed strain sensing system based on Rayleigh Backscattering provided a surface strain profile from which disbond growth rate data was obtained. A finite element model was used to verify the experimental results and determine the disbond length from the strain profiles. Results of this evaluation have shown that the CCP specimen is a promising specimen configuration for characterizing fatigue disbond growth; however, it also presents several challenges that require consideration in its application.
This paper presents two major elements of a course redesign with the aim to strengthen the connection between engineering design and engineering analysis. The course, Aircraft Structural Design and Analysis, had previously been delivered with a heavy focus on mathematical analysis and solving complex problems. It was observed, however, that in later design projects within the curriculum, students were unable to apply these skills in a less constrained design context. To combat this, two course elements were introduced. The first element was a design tutorial session that ran in parallel with the course and interfaced with real design activities being carried out within the AeroDelft Dream Team at Delft University of Technology. This session attempted to have students apply the skills they had learned in class to a less constrained design problem with more freedom than traditional practice problems, focusing on design thinking rather than reproducing an expected answer. The second element was a design-based final exam, where all of the questions within the exam were interconnected by a single design context. The first iteration of these design elements, including lessons learned and an analysis on their impact on student success, will be presented within this paper.
Recent findings have highlighted the potential of a 3D-printable high-strength Liquid Crystal Polymer, whose anisotropy can be fostered for topology optimization intents. The mesostructure of a 3D-printed liquid crystal polymer is studied: the observation of interlayer features under the form of regular notches or spiraling patterns swirls is reported on optical microscopy of cross-sections. A formation mechanism is proposed: interlayer features may be formed as a result of an offset in placement of material. Another question is raised by the observation of these crenelated shapes: by providing mechanical interlocking between layers, they are expected to enhance interlaminar shear strength of a part. Short-beam shear tests indicate that when interlayer features are tall with respect to the layer height, and oriented perpendicular to the shear loading direction, the interlaminar shear strength of the 3D-printed part is enhanced by up to 112%. Microscopic evidence further indicates the crack-arrest ability of these features.
Mechanically fastened joints are susceptible to the presence of multiple-site damage (MSD) cracks in the critical fastener row. Different from the MSD growth in joints consisting of metallic substrates, the two coupled metal crack growth and interfacial delamination propagation failure mechanisms in Fibre Metal Laminates (FMLs) make the prediction of fatigue behaviour in FML joints with MSD scenario burdensome and impractical when considering all factors influencing the fatigue performance. This paper presents a theoretical study on the MSD crack growth behaviour in mechanically fastened FML joints with a focus of modelling the effects of bearing and bypass loads. The proposed model in this paper is built upon analytical models dealing with MSD growth in flat FML panels and single crack growth in FML panels subjected to a combined tension-pin loading case. This model would be particularly useful for symmetric FML joints where no secondary bending effects present. A deliberately designed symmetric FML joint was tested to validate the proposed model. The model captures the rapid crack growth in the vicinity of fastener holes due to bearing stresses and crack acceleration due to the interaction of cracks. It is identified that the load redistribution between intact fastener rows and the cracked fastener row accelerates crack growth with crack length. The effects of secondary bending stresses in FML joints on the crack growth behaviour is extensively discussed. The performance of the proposed model for single lap FML joints is also examined using test data from open literature. It is found that the proposed model provides a conservative prediction for the tested single shear lap FML joint from open literature.
A challenge in developing an in-depth understanding of the crack growth resistance of ALM materials is the fact that mechanical properties of additive manufactured materials have been shown to be both process and part-geometry dependent. Up to now, no studies have investigated the influence of off-axis (beyond the three orthogonal build orientations) orientations on the fatigue crack growth behaviour of selective laser melted Ti-6Al-4V. Furthermore, the widespread use of compact tension specimens for investigating the material behaviour generates data more suitable for plane-strain conditions, rather than the plane-stress state which is more applicable to many lightweight aerospace structures. To address this gap in knowledge, a comprehensive study was carried out to investigate the influence of off-axis build direction inthin SLM Ti-6Al-4V plates, with a focus on the influence of microstructure anisotropy on the fatigue crack growth behaviour. It was found that although an anisotropic grain structure is visible on the specimens, it had no discernible influence on the crack growth resistance when the specimen had undergone a stress relieving heat treatment.
In this study, we tried to quantify the isolated and modulated effects of topological design and material type on the mechanical properties of AM porous biomaterials. Towards this aim, we assembled a large dataset comprising the mechanical properties of AM porous biomaterials with different topological designs (i.e. different unit cell types and relative densities) and material types. Porous structures were additively manufactured from Co-Cr using a selective laser melting (SLM) machine and tested under quasi-static compression. The normalized mechanical properties obtained from those structures were compared with mechanical properties available from our previous studies for porous structures made from Ti-6Al-4V and pure titanium as well as with analytical solutions. The normalized values of elastic modulus and yield stress were found to be relatively close to each other as well as in agreement with analytical solutions regardless of material type. However, the material type was found to systematically affect the mechanical properties of AM porous biomaterials in general and the post-elastic/post-yield range (plateau stress and energy absorption capacity) in particular. To put this in perspective, topological design could cause up to 10-fold difference in the mechanical properties of AM porous biomaterials while up to 2-fold difference was observed as a consequence of changing the material type.
Additive manufacturing (AM) techniques enable fabrication of bone-mimicking meta-biomaterials with unprecedented combinations of topological, mechanical, and mass transport properties. The mechanical performance of AM meta-biomaterials is a direct function of their topological design. It is, however, not clear to what extent the material type is important in determining the fatigue behavior of such biomaterials. We therefore aimed to determine the isolated and modulated effects of topological design and material type on the fatigue response of metallic meta-biomaterials fabricated with selective laser melting. Towards that end, we designed and additively manufactured Co-Cr meta-biomaterials with three types of repeating unit cells and three to four porosities per type of repeating unit cell. The AM meta-biomaterials were then mechanically tested to obtain their normalized S-N curves. The obtained S-N curves of Co-Cr meta-biomaterials were compared to those of meta-biomaterials with same topological designs but made from other materials, i.e. Ti-6Al-4V, tantalum, and pure titanium, available from our previous studies. We found the material type to be far more important than the topological design in determining the normalized fatigue strength of our AM metallic meta-biomaterials. This is the opposite of what we have found for the quasi-static mechanical properties of the same meta-biomaterials. The effects of material type, manufacturing imperfections, and topological design were different in the high and low cycle fatigue regions. That is likely because the cyclic response of meta-biomaterials depends not only on the static and fatigue strengths of the bulk material but also on other factors that may include strut roughness, distribution of the micro-pores created inside the struts during the AM process, and plasticity.
A challenge in developing an in-depth understanding of the crack growth resistance of Additively Manufactured materials is the fact that their mechanical properties have been shown to be both process and part-geometry dependent. Up to now, no studies have investigated the influence of off-axis (beyond the three orthogonal build orientations) orientations on the fatigue crack growth behaviour of selective laser melted Ti-6Al-4V. Furthermore, the widespread use of compact tension specimens for investigating the material behaviour generates data more suitable for plane-strain conditions, rather than the plane-stress state which is more applicable to many lightweight aerospace structures. To address this gap in knowledge, a comprehensive study was carried out to investigate the influence of off-axis build direction in thin SLM Ti-6Al-4V plates, with a focus on the influence of columnar grain orientation on the fatigue crack growth behaviour. It was found that although a macroscopic columnar grain structure is visible on the specimens, it had no discernible influence on the crack growth resistance when the specimen had undergone a stress relieving or HIP heat treatment.
In the context of the prevalence of thin-walled metallic aerospace structures, the added resistance to crack propagation offered by a built-up structure is desirable from a damage tolerance standpoint. The analysis of failure in such structures, however, is limited by the lack of crack opening solutions. This paper develops analytical models that calculate crack opening displacements (CODs) for a more general cracking scenario, i.e. non-symmetric cracks. The proposed models are based on the Westergaard stress functions. It is then found that the COD solution of one model is particularly accurate. The potential significance of the obtained solutions lies in analysing failure in built-up structures containing non-symmetric cracks. The crack opening solution is particularly useful in estimating the load transfer between cracked body and intact bridging structures in built-up structures using the principle of displacement compatibility.
Most universities have taught on-campus courses for decades and although many have also provided for professional (and life-long) learners by means of seminars and short courses taught on-site on set topics, few universities had a set program to offer professional learners dedicated courses based on original on-campus courses.
After the widely-publicized success of the Massive Open Online Course (MOOC) on Artificial Intelligence by Thrun and Nordvic from Stanford in 2011 with over 160,000 enrolled, the endless possibilities of online learning started to reach the world of STEM education. At the Faculty of Aerospace Engineering, the largest aerospace faculty in Western Europe with an enrollment of almost 3,000 BSc, MSc and PhD students, of Delft University of Technology in the Netherlands started to develop its own array of online courses ranging from MOOC to blended campus courses and paid on-line MSc courses (Groot-Kormelink et al., 2013 and Saunders-Smits et al, 2014).
Initially, the paid online MSc courses were intended for the working professionals as well as for our own on-campus students, but experience quickly showed that the needs, interest and priorities of a working professional are very different than that of an on-campus learner. A need arose for a new type of online courses: the so-called ProfEd – Professional Education, aimed at working professionals in the field, teaching at academic level, taking into account the specifics of these learners.
This paper will outline how two Aerospace Engineering MSc courses were transformed into two successful ProfEd courses run via the online platform of TU Delft (onlinelearning.tudelft.nl). The courses highlighted as case studies are:
Fibre Metal Laminates (FMLs) are a hybrid metal-composite laminate technology known for their superior resistance to fatigue crack growth compared to monolithic metals. This crack growth behaviour has been the subject of many studies, resulting in numerous empirical and analytical models to describe the complex damage growth phenomenon in the material. This study builds upon the analytical Alderliesten crack growth prediction methodology for FMLs, extending it from a tension loaded plate to a case of a combined tension-pin loaded plate. This new loading case is a more representative case to utilise for predicting fatigue crack growth behaviour in mechanically fastened joints. Development of the model extension and validation through experimental testing are detailed within this paper.
Composite mechanical characteristics can be heavily influenced by impact damages; however, this influence can be reduced by choosing a correct stacking sequence and constituents materials. In this paper, the influence of placing a metal layer within the stacking sequence of a carbon/epoxy laminate on impact resistance was studied. Impacts were simulated by means of Quasi Static Indentation tests.
Fibre metal laminates (FMLs)were developed and refined for their superior crack growth resistance and critical damage size that complimented the damage tolerance design philosophy utilized in the aerospace sector. Robust damage tolerance tools have been developed for FMLs. However, they tend to focus on the evolution of an isolated crack. There is also a risk that they will be invalidated overtime as a result of the occurrence of multiple cracks within one structure (one form of widespread fatigue damage). To combat another failure due to widespread fatigue damage, the airworthiness regulations were revised to include the concept of a Limit of Validity (LOV) of the damage tolerance analyses. Consequently, it is crucial to examine fatigue crack growth (FCG) in FMLs containing Multiple-site Damage (MSD) cracks despite their superior damage tolerance merits. The focus of this thesis therefore is to analyse MSD crack growth in FML structures. Mechanically fastened FML joints are potentially weak structural designs that are susceptible to MSD due to the stress rising contributors such as secondary bending, pin loading and open holes subjected to bypass loading. In this thesis, predictive models were developed to address several key mechanisms that affect FCG in FML joints containing MSD, and validated with corresponding experimental work. Then the predictive models were systematically integrated and implemented for FML joints. It was identified that the nature of fatigue in FMLs led to the load redistribution mechanism as the key factor to be modelled in predicting MSD growth in FMLs. The structural stiffness reductions caused by the presence of multiple cracks resulted in load redistribution from the other cracks to the single crack to be analysed, exacerbating the total stress intensity factor (SIF) experienced at the tips of the single crack, increasing the crack growth rate (CGR). The load redistribution mechanism was first substantiated by investigating FCG in FMLs containing discretely notched layers. The prediction model fairly captured the load redistribution mechanism by idealizing the notches in the metal layers as removals of metal strips. The crack acceleration over a major portion of the crack propagation was well predicted with the model; however, the surge in CGR over roughly 3 mm crack length prior to the link-up was underestimated since the plasticity interaction was not accounted for. The capability of modelling the load redistribution mechanism allows the states of multiple cracks to be analysed one by one. It was found that the load redistribution could not be symmetric for every crack and non-symmetric crack configurations therefore developed in FMLs with finite width. Hence, non-symmetric crack growth in FMLs was also investigated in this work. It was also found that both crack tip non-symmetry and delamination shape non-symmetry affected the crack growth in the metal layers. The model for non-symmetric crack growth in FMLs was validated with experimental data. Good correlation was observed. The model for MSD growth in FML panels sequentially analyses each crack state. The other cracks are idealized as removals of metal strips when analyzing the state of a single crack. This non-physical idealization of the cracks led to consistently conservative prediction results in comparison with the test data. Nevertheless, the prediction model provided good predictions of the evolution of MSD configurations. Additionally, it was proven that a very non-conservative predicted fatigue life could be obtained if the load redistribution mechanism was not considered. The effects of pin loading on FCG in FMLs were also investigated. The test data showed very rapid growth of the crack in the vicinity of the pin loading. The CGR decreased with increasing crack length. The model applied the principle of superposition to split the non-symmetric tension-pin loading into simpler tensile loading and a pair of point loads acting on the crack flanks. The SIFs for the simpler loading cases were derived and superposed to obtain the total SIF as a result of the tension-pin loading. The predicted CGR and equivalent delamination shape correlated with the measurements very well, but the model failed to predict the crack path and the measured delamination shape which were trivial issues for this work. The relevance and applicability of the developed models in this thesis for predicting the MSD behaviour in mechanically fastened FML joints was examined. The predicted results captured the trends of the measured CGR in FML joints containing MSD cracks, although there were some discrepancies. The discrepancies are mainly due to the two major shortcomings of the model which are neglecting the load redistribution over multiple fastener rows and neglecting the effects of secondary bending stresses.
The in-plane and out-of-plane mechanical behaviour of both ultrasonically spot-welded and mechanically fastened joints was investigated by double-lap shear and pull-through tests, respectively. Spot-welded specimens showed comparable onset failure load and significantly higher joint stiffness compared to mechanical fasteners when carrying shear load. The failure modes and the damage within specimens were analysed after mechanical tests. Intralaminar failure and very limited damage on the out-most ply were found for welded specimens, whereas catastrophic through-the-thickness failure was observed for mechanically fastened joints. Based on the experimental outcomes, the mechanical performance and failure mechanisms of spot-welded joints were critically assessed in comparison to the mechanical fasteners.
This master thesis investigates the effect of surface morphology of Selectively Laser Melted Ti-6Al-4V on fatigue performance. The literature study performed here iden- tifies roughness as one of three major causes for the decreased performance of the additive manufactured alloys (along side porosity and residual stresses). In order to eliminate the two other causes for poor fatigue, and isolate the surface roughness, a hot isostatic pressing (HIP) treatment was performed on every sample. Luckily, a first set of tests proved that chemical treatments, specifically electropolishing and plasma- polishing, greatly improve the surface quality. A second series tests (rotating bending fatigue tests) concluded that, although specific roughness parameter such as Ra reach lows of 0.2μ there was no statistically sound relation between fatigue performance and an improved surface roughness. Upon further investigation of the specimens, it was discovered through CT scans, that their porosity was in fact much closer to that of non HIP treated specimens. This master thesis illustrates once more that the effect of surface roughness on SLM Ti-6AL-4V alloys cannot be properly identified if the bulk material shows internal defects.
Today, the application of adhesive bonding technology for primary aerospace structures is limited due to the certification regulations. State of the art is the widely used “chicken rivet” as crack arrestor which is limiting the benefits of bonding technology, particularly in composite bonded joints.
In this paper results from fatigue testing of novel design approaches for damage tolerant high load transfer (HLT) joints as e.g. Panel Joints or large bonded repairs will be discussed for CFRP and for Fiber Matel Laminates´(FML) adherents.
Results from fatigue testing with Wide Single Lap shear (WSLS) specimen will be presented for different configurations to proof the crack stopping behavior of sate of the art fasteners as reference crack arrestor concept.
Additively manufactured (AM) porous metallic biomaterials are considered promising candidates for bone substitution. In particular, AM porous titanium can be designed to exhibit mechanical properties similar to bone. There is some experimental data available in the literature regarding the fatigue behavior of AM porous titanium, but the effect of stress ratio on the fatigue behavior of those materials has not been studied before. In this paper, we study the effect of applied stress ratio on the compression-compression fatigue behavior of selective laser melted porous titanium (Ti-6Al-4V) based on the diamond unit cell. The porous titanium biomaterial is treated as a meta-material in the context of this work, meaning that R-ratios are calculated based on the applied stresses acting on a homogenized volume. After morphological characterization using micro computed tomography and quasi-static mechanical testing, the porous structures were tested under cyclic loading using five different stress ratios, i.e. R = 0.1, 0.3, 0.5, 0.7 and 0.8, to determine their S-N curves. Feature tracking algorithms were used for full-field deformation measurements during the fatigue tests. It was observed that the S-N curves of the porous structures shift upwards as the stress ratio increases. The stress amplitude was the most important factor determining the fatigue life. Constant fatigue life diagrams were constructed and compared with similar diagrams for bulk Ti-6Al-4V. Contrary to the bulk material, there was limited dependency of the constant life diagrams to mean stress. The notches present in the AM biomaterials were the sites of crack initiation. This observation and other evidence suggest that the notches created by the AM process cause the insensitivity of the fatigue life diagrams to mean stress. Feature tracking algorithms visualized the deformation during fatigue tests and demonstrated the root cause of inclined (45°) planes of specimen failure. In conclusion, the R-ratio behavior of AM porous biomaterials is both quantitatively and qualitatively different from that of bulk materials.
This paper proposes an analytical model for predicting the non-symmetric crack growth and accompanying delamination growth in FMLs. The general approach of this model applies Linear Elastic Fracture Mechanics, the principle of superposition, and displacement compatibility based on the understanding of deformation behaviour in eccentrically cracked metal panels. The non-symmetric crack growth behaviour of two crack tips and accompanying asymmetric load transfer from the eccentrically cracked metal layers to the intact bridging fibres are successfully predicted with the model. The predicted crack growth rates and delamination evolution are compared to test data, good correlation is observed.
This paper presents a theoretical and experimental study on Multiple-site Damage (MSD) crack growth behaviour in Fibre Metal Laminates (FMLs). The prediction model is developed based on a simplified analysis of the effects of load redistribution on a single crack in FMLs containing multiple cracks. Test results show that the crack growth accelerates as cracks grow towards each other. The tests also show non-symmetric crack growth behaviour and non-symmetric interfacial delamination propagation in case of multiple cracks. The prediction model successfully captures the crack growth acceleration and non-symmetric growth behaviour.
Keywords: MSD; Fibre metal laminates; Load redistribution mechanism; Non-symmetric crack growth
Meta-biomaterials are porous biomaterials created by additive manufacturing techniques such as Selective Laser Melting. These materials are built up form a repeating unit cell, resulting in a porous structure that can be applied for bone implants or prosthetics. The mechanical properties of these meta-biomaterials can be tailored by variations in unit cell type and strut thickness, resulting in different porosity values. This makes it possible to create a material with mechanical properties similar to bone, which prevents negative side effects of conventional bone implants such as stress shielding. The fatigue behavior of these meta-biomaterials has been studied before, but only at a single stress ratio of R=0.1 This study investigates the fatigue behavior at different stress ratios which result in an increased mean stress. A cylindrical porous structure that is built up from a diamond unit cell is tested at stress ratios of R=0.1, R=0.3, R=0.5, R=0.7 and R=0.8. Two samples types are made of Ti-6Al-4V ELI powder, resulting in a theoretical porosity of 80% and 10%. Also an experimental DIC method is developed, to visualize the deformation behavior during the fatigue tests.
The primary objective of this research study was to evaluate the capabilities for measuring strain of a composite UAV wing with an embedded optical fibre connected to a Rayleigh Backscattering distributed sensing system. This research paper summarizes the manufacturing procedure used during the instrumentation of the composite UAV wing. In addition, a Finite Element Model was developed in order to verify the strain distribution of this complex structure under static and dynamic loading conditions. The use of strain gauge data as a means for verification is presented as part of this research. Finally, fatigue tests were carried out to determine the longevity of the embedded fibre during the design life of the structure. The results demonstrate the ability of distributed sensing system to obtain complex and accurate strain distributions on a single non-grated fibre. In addition, the findings demonstrate current limitations in the system for capturing accurate strain profiles in dynamic loading test cases.
Damage is found in resin rich bead corner radii of RTM 6 epoxy-based composite ribs due to in-service thermal-mechanical loading after an aircraft inspection. The same types of damage are obtained through pure thermal cycling of a single bead. However, thermal cycling is a time-consuming process. Therefore, a faster way of damage investigation is required. A potential way to achieve this objective is by mechanically cycling a coupon specimen with simpler geometry than the bead. To investigate potential solutions to the problem a literature review is performed. The scope of the literature review covers several topics as follows. Laminate fatigue damage modes and their impact on the laminate is researched. It is discovered that it is typical for RTM 6 epoxy-based laminates to build-up high matrix residual tensile stresses after manufacturing. Several differences between thermal and mechanical cycling are discovered. It is found that the most common way to test composites is by the use of ASTM standards. However, there is little available information about the fatigue behavior of laminates with resin rich areas. Investigation of the fatigue behavior of laminates with resin rich areas is performed by the use of FEA and physical tests. Four specimen types labeled from A to D are manufactured. Type A is a dog-bone pure RTM 6 specimen. Types B to D are all composite specimens with the same in-plane geometry and different layups and manufacturing processes. All specimens are tested statically and in fatigue. In the fatigue testing session, fractography of the damage occurring at different test conditions for different specimen types is performed. In addition, two FE models are created. The first model is of the bead. The second model is a harmonized model applicable to all composite specimen types with required layup readjustments for each specimen type. It is discovered by FEA that the maximum principal stress in the resin rich area is perpendicular to the matrix cracks in the bead and the composite specimens. It is also discovered that the maximum principal stress cycle of the matrix at the resin rich layer interface with the fabric is similar in the bead and the specimens. The similar fatigue parameters are the R-ratio and the stress amplitude. The parameter similarity suggests they could potentially drive the resin rich layer fatigue damage initiation. Moreover, a positive through-the-thickness stress gradient is discovered, which suggests the cracks are likely to initiate bellow the resin rich layer surface. This hypothesis is further supported by fractographic observations of cracks not reaching the laminate free surface. Static and fatigue tests are performed. The static test provides the UTS of all specimen types, based on which cyclic load levels are selected. In the specimen fatigue tests several results are observed. In the first place, damage similar to the bead damage is found, namely cracks and delamination. In the second place, the damage is observed to penetrate through the laminate thickness and to be dependent upon the laminate compaction. However, this penetration depth dependency on the compaction might be influenced by the second curing cycle of specimen type C, in which the damage was observed. Finally, reduction in the specimen stiffness is observed due to fatigue damage accumulation in time. Based on the results recommendations are formulated. For design purposes resin rich area formation should be avoided both inside and outside the laminate. If their formation is inevitable, at least the laminate should be kept well compacted and the resin rich area location should be kept only at the surface. For future research, two topics are identified as requiring such. First, is the laminate stiffness reduction. Second, is the influence of the second curing cycle of the well compacted specimen type C.
This paper provides an experimental investigation into the pin loading effects on the crack growth behaviour in Fibre Metal Laminates. The pin loading effects and bypass loading effects are incorporated in two different tested joints. The analysis of the test results shows that pin loading dominates the crack growth only in the vicinity of the pin hole and the superposition method for analysing stress intensity factor in FMLs with pin loading effects can be applied.
Large classroom sizes are a reality university educators need to contend with, particularly in the first year of a given cohort within a degree programme. Activating and engaging students in these large classroom environments present numerous sets of challenges. These challenges are exacerbated by student learning development needs in the early stages of the degree programme. In their first year, students are still adapting to a new learning environment and are developing new study skills and practices. Early success and failure in courses will shape intrinsic and extrinsic factors that will motivate the student in the remainder of their degree. Thus the perceived challenge of activating large classrooms early in a degree programme goes beyond simple engagement; beneath the core learning objectives of the course are implicit learning objective about developing effective motivation and study skills.
This paper examines the efforts to reorganize a first year Mechanics of Materials course taught in the Bachelor of Engineering Programme within the Faculty of Aerospace Engineering at Delft University of Technology University to address this need using a Blended Learning approach.
This paper presents an analytical model for fatigue crack growth prediction in Fibre Metal Laminates (FMLs) containing discretely notched layers. This model serves as a precursor in the development of a simplified prediction methodology for modelling the effect of load redistribution on a single crack in FMLs containing Multiple-site Damage (MSD) scenario. The model mainly focuses on capturing the influence of load distribution around discretely notched layers on the growth behaviour of an adjacent crack in a FML panel. The utilized approach in the model is the use of linear elastic fracture mechanics (LEFM) in conjunction with the principle of superposition and displacement compatibility. The proposed model is also validated using experimental data.
Aerospace structures have been long dealing with the safety versus weight issue. Lighter airplanes are cheaper to operate, however, they may face a safety issue because of the reduced fatigue life. Consequently, a heavier/safer structure is designed. Adhesive bonding is a joining technique that offers potential for improvement in the fatigue behavior of a structure, resulting in reduced weight. However, predicting the fatigue behavior of a bonded joint for its use in a damage tolerance design philosophy still remains a problem with no satisfactory solution. Often, the joint is subjected to a combination of peeling and shearing stresses. Hence, one of the most important factors influencing the fatigue behavior of an adhesively bonded joint is the Mode Ratio. The objective of this investigation was to study of the Mode Ratio on the fatigue behavior of a bonded joint. First, the fatigue disbond mechanisms were investigated throughout the entire Mode Ratio range and compared to fatigue delamination mechanisms. After the mechanisms were identifed, a parameter related to the mechanisms was chosen as similitude in the Paris relation and the Mixed-Mode fatigue disbond model was developed. Later, the model was evaluated on a different adhesive and on a condition of variable Mode Ratio. The fatigue disbond mechanisms study identified the local principal stress as the driving force for the micro-crack formation and growth, and the Mode Ratio was identified as the controlling parameter for coalescence between the micro-cracks. Based on these findings, a parameter directly related to the principal stress was proposed as a similitude parameter. Additionally, a linear interpolation between Mode I and Mode II parameters of the Paris relation was proposed to predict the Mixed-Mode fatigue behavior. Thus, the model predicts the fatigue behavior for the entire Mode Ratio range using only pure Mode I and pure Mode II as inputs. The evaluation of this model revealed that it presents good predictions for Mode Ratios in the range of 0% to 50% and conservative predictions in the range of 50% to 100%. The model also seems to be valid in a variable Mode Ratio condition. The limitations and shortcomings of the model along with the limitations of using a damage tolerance philosophy on adhesive bonding were discussed. Despite these issues, the model is an improvement over the models available in the literature because it captures some of the phenomena involved in the Mixed-Mode fatigue disbond. Additionally, the model also reduces the amount of empirical data required for its implementation.
Currently, Strain Energy Release Rate (SERR), or a function of it (Gmax, DG among others) is used to predict Fatigue Disbond Growth (FDG). However, Fig 1.a shows that the use of such variable does not completely describe FDG at Mixed-Mode conditions. Pascoe et al [1] proposed a new approach to improve disbond phenomenon understanding. This manuscript expands the theory developed in [1], which focus on the stress ratio effect, for FDG under mixed-mode conditions. Figure 1.b presents a preview of the results. This figure shows clearly the collapse of all FDG data into a single trend. Despite the fact that Ucyc can only be calculate a posteriori, the use of cyclic strain energy can contribute for the Mixed-Mode FDG understanding and be later developed into a predictive model.
In this study, mode II fatigue crack growth utilizing Central Cut Plies (CCP) specimens is considered, in order to assess the durability of bonded composite repairs. Fatigue tests were performed with unidirectional carbon-epoxy specimens and adhesive film co-cured. A back-face strain technique was used to obtain strain data with a fiber optics distributed sensing system based on Rayleigh Backscattering. Crack growth rates were obtained from the strain profiles and the results show a good correlation with other measurement techniques. The technique also indicated an unequal crack growth behavior as observed in the ultrasonic inspection. As a consequence, it was verified that the common CCP geometry used for static tests is not ideal for fatigue tests and thus, this type of specimen requires further improvement for fatigue assessment of bonded composite repairs.
A model was developed for delamination growth in bonded repair patches under constant amplitude fatigue loading. The model used the finite element method, employing the virtual crack closure technique, to determine the strain energy release rate (SERR) as a function of delamination length. Interaction effects between multiple delaminations, and the effect of delamination shape was also investigated. Fatigue cycling of coupon specimens was performed in order to find a relation between the SERR and the delamination growth rate. A power law (Paris-type) relation was established. Using this relation and the relation between SERR and delamination length, delamination growth predictions were produced. This predictions agreed well with the results of the coupon tests. A further validation by tests on more representative patch repair specimens was inconclusive due to the lack of delamination growth in the patch repair specimens.
Fibre Metal Laminates (FMLs) are a family of hybrid metallic-polymer matrix composite materials with superior fatigue and damage tolerance behaviour. Although technically a composite material, they share many behaviour features more in common with their metallic constituents. This chapter will introduce the FML material concept and discuss the static and fatigue design considerations needed in the design of bolted FML joints.
Solid rivets are widely used as mechanical fasteners in airframe applications due to their relative low cost and good fatigue performance. Although rivet installation is known to influence this fatigue performance, variabilities in hand riveting practices make exploiting rivet installation as a design variable difficult. Developments in riveting technology have led to force-controlled rivet squeezers and fully automated riveting gantries, improving the consistency of rivet installation and providing the opportunity to exploit its influence on
fatigue.
This dissertation describes a research program undertaken to examine the influence of rivet installation on the fatigue performance of riveted lap joints and identify what aspects can be exploited during design. A combination of finite element analyses and experimental techniques were used to investigate the role of rivet installation on the formation of residual stresses and on secondary bending stresses in a loaded joint, two aspects established as critical to the fatigue performance of riveted lap joints. Crack growth reconstructions of fracture surfaces marked using a special marker fatigue spectrum were also completed in order to quantify the effects of these residual and secondary bending stresses on fatigue performance. Additionally, variations in the effects of rivet installation on traditional monolithic aluminum sheet materials and hybrid aluminum-fibre glass laminates known as GLARE were also investigated.
Results from these investigations provided new insights into the role of rivet installation on fatigue. The radial expansion mechanism to which residual stress formation during riveting is typically attributed was observed to be a secondary mechanism relative to the through-thickness compression of the joined sheets. The location and magnitude of peak secondary bending stresses were found to be directly influenced by rivet head geometry. In
certain cases, shifts in the location of peak secondary bending stress were found to negate the benefits of residual stresses formed during riveting. Finally, differences in the influences of rivet installation on GLARE and monolithic aluminum sheet were also identified. The presence of the fibre layer in GLARE alters the residual and secondary bending stress states, indicating a potential difference in joint designs optimized for aluminum and GLARE.
GLARE (GLAss REinforced) laminates are part of a family of metal-composite hybrid materials being used for airframe applications. One critical design element in such an application is the riveted lap joint. Machine countersinking, necessitated by the use of flush-head rivets, produces a knife-edge in some of the GLARE plies. The resulting stress concentrations likely accelerate crack nucleation in such joints, diminishing fatigue performance. Furthermore, the superior fatigue performance of GLARE relative to monolithic aluminum permits reductions in skin thickness, increasing the risk of a through-thickness knife-edge.
This thesis examines the potential of dimple countersinking as an alternative method to machine countersinking in thin GLARE laminates. Using simple coupons, the dimpling process and crack initiation behaviour of dimpled GLARE is investigated. Additionally, wide lap joint fatigue specimens were tested to compare the relative performance of machine countersunk and dimpled GLARE lap joints. Initial results showed that dimpled GLARE joints have an inferior fatigue performance; however, these results are based on dimpling tools and processes optimized for aluminum. Delamination damage was observed during the dimpling process which may be responsible for the poor fatigue performance. Through optimization of the dimpling tool geometry, such damage may be avoidable, allowing the true potential of dimpling to be investigated.