Parametric tools could significantly increase efficiency in the design process within the construction industry, especially with prefabricated elements. Efficiency is also searched for in material use. An element that uses its material very effectively is a structural insulated panel (SIP). However, the synthetic polymers used in already existing panels are not very sustainable. The new climate agreements push producers into making more sustainable choices. Due to this, Kingspan Unidek is experimenting with a bio-based insulation material integration into the SIP. A new structural insulated panel with OSB/3 faces and a Steico wood fibre polyurethane resin insulation core is being developed. In this thesis, the panel’s behaviour is analysed for construction purposes under axial and transverse loading experimentally, analytically and through modelling. The behaviour of the panel is analysed for residential building applications, and the possibility of using parametric modelling tools in the design process for SIPs is assessed.
The analysis is performed by using four main methods – full-scale experiments under eccentric axial loading by Kingspan, models on Karamba3D and RFEM, and analytical calculations with Timoshenko beam theory, shear beam equations and rod in compression equations. The failure methods are analysed with hand calculations from sandwich panel theory. The Karamba3D model is developed by using shell and beam elements since it is not yet possible to model layered materials. The material is defined by the user and the core is modelled as beams which are scaled to match the Timoshenko beam equation results.
The faces exhibited a separate behaviour during the experiments under low eccentric axial loads suggesting a delamination failure or a core failure. The failure occurred during an eccentric 18 kN/m axial loading which caused the faces to behave separately. The core has very low strength properties compared to other conventionally used SIP insulation materials which are at least twice as strong as the Steico insulation board. Additionally, the analytical calculations showed that the panel has very low transverse loading capacity as a result of core shear failure – a maximum distributed load between 0.1 kN/m to 2.5 kN/m can be applied depending on the analysis method. These findings confirm that the panel could only be used with the application of through-thickness stiffeners. The developed Karamba3D model successfully reproduced the experimental results, but the model should not be used for failure analysis since local failures or interface failure is not described. Timoshenko beam theory and the RFEM model showed a good correspondence for transverse loading, but a bending test should be performed to confirm the results. The RFEM model could not show the SIP behaviour under axial loading but did exhibit a trend in deflection results towards where a local failure could occur. The conventional sandwich panel theory overestimated the panel resistance based on the experimental results and the modelling results.
The panel in this form could only be applied if through-thickness stiffeners are used to strengthen the panel or if the core material is significantly improved. However, there is a possibility that the separate behaviour by the faces could be avoided by avoiding eccentric loading and only applying centric loading. It would be necessary to test the panel for failure to get a better understanding of the failure modes and to further validate the model values. Additional testing to confirm the failure method would be testing the material properties of the Steico material, and the bond strength of the interface between the faces and the core. The model has the potential to be used for analysing all different sandwich structures, but would need to be further verified with different materials.

The transverse stability of timber terraced houses can be provided by CLT stability walls where decoupled timber terraced houses are preferred to meet acoustic requirements leading to rather large thicknesses for the CLT stability walls and many large connections for anchorage to the foundation. The main goal of this research was to investigate the transverse stability behaviour of timber terraced houses, using a case study provided by WSP.

Models to assess the transverse stability of timber terraced houses are simple schematizations with hand calculations of the decoupled and coupled timber terraced houses and FE models of decoupled timber terraced houses with and without T-section and coupled timber terraced houses with and without T-section. The coupling of the houses can be achieved by 4 Straviwood Modulink 6.0 kN connections per floor level. The T-section can be realised with double screws with diameter Q = 12 ZZ with a spacing of ` = 100 ZZ.

Simple schematizations with hand calculations of the decoupled and coupled timber terraced houses result in overdimensioning of the connections since bending of the floors due to deformation of the stability wall, providing resistance to the deformation of the stability wall, is not considered. This resulted in 6 hold downs per CLT stability wall.

Compared to the FE model of decoupled timber terraced houses without T-section, all FE models have a positive influence on the design of connections in terms of reduction of number of hold downs and improvement of the total tension support reaction. Coupling of the houses is better compared to application of a T-section for decoupled houses. An improvement of the total tension support reaction of 83% and 67% for respectively the house on the left and the house on the right, compared to the decoupled timber terraced houses without T-section, can be achieved by coupled timber terraced houses with T-section. This also resulted in a reduction of 5 hold downs per CLT stability wall and the possibility to reduce the thickness of the CLT stability wall.

In timber connections, screws have a benefit over dowels or nails, as due to their withdrawal capacity they can transfer loads both perpendicular and parallel to the screw axis. Screws display their largest stiffness and load-bearing capacity in the direction parallel to the screw axis. These properties can be taken advantage of in case timber elements are connected with screws inserted under an angle with respect to the shear plane. Although a calculation method for the translational stiffness of connections with inclined screws is currently not included in Eurocode 5, multiple methods are discussed in the literature. This work focuses on the rotational stiffness of connections with inclined screws, which remains
a considerable knowledge gap to date.

An extensive literature study is carried out, in which methods for the calculation of translational stiffness of inclined screws as well as methods for the calculation of rotational stiffness of dowel-type fasteners are studied. Combinations of these methods are used to put forward fourteen calculation methods for the rotational stiffness of inclined screw connections. Rotational stiffness tests of connections with inclined screws have been carried out prior to the start of this thesis project. The calculated rotational stiffness values of each of the fourteen methods are compared to the experimentally obtained values. Two prototypes of a novel truss concept developed at Karlsruhe Institute of Technology, in which inclined screws are implemented in the chord-diagonal connections, are modelled in Rhinoceros Grasshopper. The influence of the rotational stiffness of the chord-diagonal connections on the serviceability and ultimate limit state behaviour is studied.

A correction for friction occurring between the timber members as a result of the specific sequence of the application of the loads in a subset of the rotational stiffness experiments is proposed. After this correction, the best method shows good similarity with experimental results, yielding a coefficient of determination of 0.79 for a total of 91 tests. With regard to the truss model, it is concluded that the rotational stiffness has no influence on the deflections of the truss, however, the moments in the connections increase for higher rotational stiffness values. Individual screws in the connections are found to receive an additional load of 5-10 % in their main loading direction as a result of the moments in the connections in the large-scale truss prototype. The model can be used in the preliminary design of these trusses to quantify moments in the connections and the forces per screw as a result of these moments.

Summary

Numerous ancient historical constructions worldwide depend primarily on an extensive array of wooden foundation piles, as they are subject to loading conditions governed by the superstructure above. Wooden foundations transfer loads through a combination of compression and lateral resistance. The inherent strength of wood handles compressive forces, while stiffness and soil friction counteract lateral loads. Proper arrangement and maintenance ensure even load distribution. Careful design, wood quality, depth, and protective treatments are essential for longevity and load-bearing efficiency.

Amsterdam, the Netherlands' capital city, renowned for its rich artistic heritage, intricate canal infrastructure, and slender architectural dwellings, originated as a modest fishing hamlet that underwent remarkable development into a prominent global European city. During this urban transformation, less visible engineering elements, such as wooden foundation piles, were overlooked, despite their critical significance. In Amsterdam's historical core, the majority of structures including buildings, bridges, and quay walls, rely on these wooden supports. Noteworthy, the city estimates that 12 million such piles are still active. These structural components have consistently demonstrated economic efficiency and reliability. Nonetheless, the aging process affecting these foundations, with some dating back up to 500 years, introduces complexities when assessing their current load-bearing capacities and the ensuing reliability of the structures they support.

The lack of knowledge and inspection techniques of the mechanical and physical properties of these timber piles hinders a proper evaluation of the remaining life span of the foundations which could lead to possible irreplaceable structural damage to these structures. This body of research evaluates the physical and mechanical properties such as the actual moisture content, density distribution, compressive strength, and modulus of elasticity through the cross section of Spruce (Picea abies) foundation piles. Therefore, the overarching research question has arisen:

“How do the variations of mechanical and physical attributes manifest across the cross-sectional profile of both degraded and non-degraded spruce foundation piles and how can micro-drilling techniques be utilized to assess these characteristics?“

This will be achieved by means of small-scale compressive experimental testing of five prisms extracted from each cross-section (3 separate locations along the length of the pile) of foundation piles never driven into the soil and piles that were retrieved under bridges in the historical centre of Amsterdam that were planned to be demolished. These aforementioned retrieved piles had a service life between 100 years and 300 years, always under the water table, presenting mechanical degradation due to loading over time and in addition possible bacterial degradation of the cross-section peripheral regions.

Initially, micro-drilling techniques were employed to ascertain the drilling amplitude. This step served to assess the initial quality of the wood under examination. Additionally, it aided in identifying specific points of interest for specimen extraction, including degraded wood in the peripheral regions, sound wood in the internal section, and the pith. Subsequently, the acquired data underwent thorough analysis. This analysis, combined with the micro-drilling measurements, enabled an assessment of the potential applicability of drilling amplitude in predicting the mechanical and physical properties of the pile. This sequential approach ensured a systematic and scientifically rigorous evaluation of the wood's characteristics and its implications for pile performance. The investigation was conducted to enhance the understanding of the structural performance and material characteristics of spruce foundation piles, while also evaluating the applicability of micro-drilling methods as a predictive tool in engineering assessments...

Self-tapping screws are popular tools used in timber engineering. The axial withdrawal strength of self-tapping screws could be used in timber structures as an active or passive application. An example of an active application is the use as a fastener in wall to wall connections. An example of a passive application is the use as reinforcement perpendicular to the grain, to prevent cracking along the grain. Existing methods to calculate the withdrawal strength of self-tapping screws are mainly based on tests done on softwood. Some screw manufacturers offer calculation methods for the use of their screws in hardwood from temperate regions in so called European Technical Assessments (ETAs). But the Eurocode and the ETAs offer no calculation methods for high density tropical hardwood species. Tropical hardwood species are generally used in The Netherlands for hydraulic and outside structures such as lock-doors, bridges, weirs and jetties. Tropical hardwood species are used because of their strength and durability in outside conditions. There are no well-established methods to calculate the withdrawal strength of self-tapping screws in tropical hardwood perpendicular to the grain and the purpose of this thesis is to gain more knowledge on this topic. The central question of this thesis is: What is the withdrawal strength of self-tapping screws in tropical hardwood species perpendicular to the grain and is it possible to develop a verification model for engineering purposes?
To find answers to the main question first a literature review has been conducted. Literature on timber anatomy, self-tapping screw properties and available calculation methods have been reviewed. This is followed by a literature review on possibilities to drive self-tapping screws into high density woods. This is problematic because screws could fail already during insertion or cause splitting of the timber. Screw modifications such as cutter tips could be used to reduce the insertion moment, but also pre-drilling of the timber could be done to avoid torque failure of screws or the splitting of timber. The literature study is finalized with an investigation to the influence of material properties on the withdrawal strength. Tropical hardwood species are often used in outside or wet structures so literature on the influence of moisture content has been investigated. In research done by Ringhofer et al. [22] it was found that if the moisture content increases above 12 % the withdrawal capacity significantly decreases. Pre-drilling of the timber helps to insert self-tapping screws more easily into timber, but this also removes material and could therefore reduce the withdrawal strength. Research done by Brandner, Ringhofer, and Reichinger [5] found that if the pre-drilling diameter is smaller or equal to 80 % of the nominal screw diameter pre-drilling had no influence on the withdrawal strength. The mentioned research however has been done on temperate hardwood and the question is whether this is the case for tropical hardwood as well. According to EN 1995-1-1:2011 pre-drilling is necessary if the characteristic density of the timber is larger than 500 kg/m³. With increasing timber density the withdrawal strength also increases (Hubner, Rasser, and Schickhofer [13]). Regarding the screw diameter, with increasing diameter the withdrawal strength (f_ax, MPa) decreases as was found by for example Hubner, Rasser, and Schickhofer [13]. With increasing diameter the withdrawal capacity (F_ax, kN) also increases. With increasing effective length of self-tapping screws inserted in wood the withdrawal strength (f_ax, MPa) does not increase or decrease (Xu et al. [30]). The withdrawal capacity (F_ax, kN) however increases, a linear relationship between the effective length withdrawal capacity can be observed in research done by Xu et al. [30].
Findings in the literature study were the basis of how the test series were designed. EN1382:2016 describes how to test and determine the withdrawal strength of screws in timber. This standard also gives rules on the dimensions of the test piece. The width of the available tropical hardwood pieces was to small according to EN1382:2016 which states that a minimal edge distance of 5d has to be used. Methods from the EN 1995-1-1:2011 and ETA-11/0030:2020 from Rotho Blaas have been used to make a selection of screws that will be used in testing. The goal is to have withdrawal as main failure mode. The following screw diameters have been selected: 6, 8 and 11 mm. The diameter 11 mm screw has been tested in a pre-testing series with pre-drilling diameters: 0,7d and 0,8d; with insertion depth 90 mm; Also three tropical hardwood species with the highest density were selected to test the limits of the materials. It was found that the smaller edge distances did not cause any problems. Further because of pre-drilling no damages occurred on the timber or screw, and there were no problems during screw insertion. No significant difference in withdrawal capacity was found between pre-drilling 0,7d and 0,8d so this was changed to pre-drilling 0,8d and 0,9d in the main test series. Also screw failure was observed for insertion depth 90 mm, this was changed to a maximum of 70 mm in the main test series to make sure screw withdrawal is the main failure mode.
For the main test series 6 tropical hardwood species were used (Kanda, Lati, Longhi, Tali, Limbali and Mukulungu), and also the softwood Spruce as a reference material. The diameter 6 mm, 8 mm and 11 mm screws were used with pre-drilling diameter 0,8d and 0,9d. Conditioning of the wet test pieces has been done by putting the timber blocks underwater. On every test piece or timber block four tests were done: two withdrawal strength tests (according to EN 1382:2016) and two stiffness tests (according to EN 26891:1991). Also the insertion moment was measured. After testing the density of every test piece was determined together with the moiture content.
The main test series provided a broad range of data. The following was measured: the density, the moisture content, the withdrawal capacity, the insertion moment and slip modulus. All results can be found in annex A. The test piece geometry was not according to EN 1382:2016, in general larger edge distances were required. Nevertheless did this not lead to any problems, no cracks or screw failure occurred during the insertion of the self-tapping screws nor was it any problem to insert the screws at all. This had all to do with the pre-drilling diameters of 0,8d and 0,9d what was used. Because of this very low insertion moments were measured. Also during and after withdrawal tests no significant damage was observed. Even in extreme situations that were tested during pre-testing there was no evidence that a larger edge distance was needed. In the analysis it was found that with increasing density the withdrawal strength also increases, this is in line with literature. It was also found that the species does influence the withdrawal strength, as was found by calculating the withdrawal strength to density ratio per species. With increasing screw diameter the withdrawal strength decreases as was found in this analysis, and also in line with literature. Only in the case of pre-drilling 0,9d for the diameter 8 mm screw this was not the case when comparing it to the diameter 6 mm screw, this probably was a result of pre-drilling accuracy. Regarding the influence of pre-drilling it was found that on average the withdrawal strength is reduced by a factor of 0,68 when the pre-drilled hole is 0,9d instead of 0,8d. In literature it was found that the effective screw length does not influence the withdrawal strength. No clear influence of effective length has been found, except for screw diameter 6 mm and 8 mm when pre-drilled 0,9d. The diameter 6 mm and 8 mm screws are HSBH screws with a larger inner diameter and it might be that the withdrawal strength is not optimal for these types of screw when inserted in 0,9d pre-drilled holes. During testing wet and dry pieces were tested. It was found that the moisture uptake rate is very different for the various tropical hardwood species that were tested. It was also found that as expected the withdrawal strength is negatively influenced by an increased moisture content. First of all it has to be mentioned that the scatter in data is larger when compared to the test data from withdrawal tests, this has to be taken into account. In the analysis it was found that with increasing density the slip modulus also increases. Also the timber species influences the slip modulus just like in the case of the withdrawal strength. No clear influence of screw diameter was found on the withdrawal strength. Just like for the withdrawal strength is the slip modulus influenced negatively when pre-drilling 0,9d has been used instead of 0,8d. This negative influence increases with increasing diameter. With increasing effective screw length also the slip modulus increases, this would be expected and was also found in the test results. Also a higher moisture content has a negative influence on the slip modulus, this was also as expected. A larger influence was observed in the case of timber species with the highest densities measured (Tali and Mukulungu), while for these species the lowest moisture content was measured in wet conditions compared to other species.
Finally two models have been proposed to better predict the withdrawal strength of self-tapping screws perpendicular to the grain in tropical hardwood. For the first model withdrawal strength $f_{ax}$ parameters are determined per species, per timber condition and per pre-drilling diameter. The second model is one formula based on the density, screw diameter and effective screw length for all tested tropical hardwood species. Both proposed models show a good fit to the test data. 

For the renovation of the Suurhoff bridge, Arup decided to design and propose a new, innovative strengthening scheme, which improves the fatigue performance of the bridge deck and extends the design life of the bridge by at least 15 years. In this strengthening scheme, a steel plate is placed on top of the existing deck plate with a layer of epoxy in between. Preloaded injection bolts are also used to connect the strengthening plate with the deck plate. This strengthening techniques has clear advantages over the current alternatives with regard to weight, execution time, risks and flexibility in the design.
In order to better understand the behaviour of the renovated bridge deck, verify the effectiveness of the strengthening scheme and check the accuracy of the numerical models, a monitoring scheme is desirable. This is an important step in the development and optimisation of the strengthening approach, especially when the goal is to apply the scheme more often on future bridge renovation projects. To achieve this, the two research questions of this thesis are formulated as follows:
What is the effectiveness of strengthening an orthotropic steel deck with a bonded & bolted strengthening plate?
How can the behaviour of the bridge be numerically modelled to accurately capture the improved fatigue resistance?
This question will be answered through a combination of monitoring and finite element modelling. First, a monitoring scheme is set up with 16 strain gauges that are installed on the deck plate, troughs and cross girder. Quasi-static load tests are executed using a truck with known weight, both before and after the application of the strengthening scheme.
The load tests were successfully and accurately carried out and the results show a large reduction in the stress cycle. Stresses in the troughs are alleviated by 45-55% and stresses in the deck plate are reduced by 85-90%. This is largely in line with what was expected during the design.
Furthermore, the used FE models are validated so that more confidence can be gained in the design decisions. The full influence line is simulated, with the truck positioned at more than 120 longitudinal locations. The numerical modelling was able to accurately predict the shape and magnitude of the influence line generated by the truck loading. A difference in peak value between experimental and numerical results of no more than 25% was observed. The largest differences are observed for local bending in the deck plate of the unstrengthened bridge, but this is largely explained by the large sensitivity to the exact wheel position. For the strengthened bridge, a very good match is obtained in almost all locations. A difference between numerical and experimental results of no more than 10% is observed when not considering the area close to the bolts.
Close to the bolted connection (±200 mm), no accurate results can be obtained with a simplified modelling approach that uses shear springs to model the bolts. However, the obtained numerical results are conservative compared to the experimental results. Some simple modelling adjustments have been applied but are unable to significantly improve the results. A detailed modelling approach has successfully been applied in which the bolt has been modelled fully in solid elements. The preload is implemented through a dynamic relaxation phase, so the bolt force is transferred through friction of the plates without any relevant increase in computation time. Therefore, this modelling technique can relatively easily be applied in a global FE model. This modelling technique is more accurate, and analyses have successfully managed to reduce the error by 50%. However, stresses close to the bolt are still overestimated even with this advanced modelling approach and more research is needed for a complete match in this area.
In conclusion, the largely matching results reinforce the decisions from the design report. More confidence is gained in the static and fatigue design life, acknowledging the potential of this new strengthening scheme for future applications.
As recommendations for future research, more testing and design work could be carried out to further optimise the design of the strengthening scheme. Furthermore, the temperature loading can be investigated through testing and monitoring in order to reduce this critical load case. Lastly, more detailed local FE modelling around the bolted connection can help understand the behaviour in this area. This can further increase the potential of the strengthening scheme when bolts are applied in close proximity of critical fatigue details.

Timber usage in construction works has increased over the past years due to its better environmental performance. Applying timber elements instead of steel or concrete decreases the emissions impact of the built environment. Architect firm ZJA and Iv-consult are interested in investigating a timber railway bridge as part of the Norwegian infrastructure plan. A kilometre-long double-track railway bridge with 50 m spans is used as a case study to investigate the possibilities of timber application. The objective is to design and assess a timber beam bridge with steel sections at the support.

The aim of this research is to design the sections and their connection. The design requirements and calculation methods used are based upon the Eurocodes. The design is considered to be part of the preliminary design phase of a project. Unity checks of the cross-section and deflection are calculated by use of linear mechanics. An FE model with beam elements is created to determine the force distribution of the bridge and calculate deflection. The strength and stiffness of the connection are calculated using empirical and plastic mechanic methods given in the standards. The resulting stiffness is used in the model to determine the final deflection when accounting for the semi-rigid joint.

The design consists of timber and steel box girders that only carry a single track. On top of the support, a slender stiffened steel box girder of 10 m length is situated, 5 m to both sides. Between the steel sections, there are timber box girder sections of 40 m that have very thick flanges and webs. On top of the structure lies a slab track that is only considered as weight. A dowel-type connection is used with double slotted-in steel plates in all flanges and webs of the timber section. The slotted-in plates have a slightly smaller width than the timber elements and have a depth of around 2.7 m. The steel fasteners are simple dowels.

This thesis has found that a timber and steel beam bridge with box girder cross-sections is governed by the creep deflection, shear strength of the timber, fatigue strength of the steel, and the dimensional stability of the connection. Except for the last issue, each of these limits can be satisfied by increasing the dimensions of each cross-section. Dimensional stability is the timber size change due to moisture content variation. The perpendicular-to-grain stresses become too large due to the configuration of the steel in the connection. The use of modified timber, which decreases the dimensional stability effect drastically, is only suggested as a solution.

An architecturally pleasing structure is designed by adding very limited visual elements but mainly practical and structural elements that have two functions, shown in the figure below. The desired continuity and bridge rhythm is accomplished via a line concept. The abrupt material cross-section changes are partly covered by continuous timber elements. From the practical and architectural adjustments, new cross-section dimensions were determined. The governing unity checks of the new design were calculated and showed the possibility of the design under the assumption the connection shrinkage issue is resolved.

Thus a timber and steel railway bridge has potential. It should be stressed that methods and factors from the Eurocodes are assumed to be applicable to the large timber box girder sections. This assumption is recommended to be checked by means of experimental research on large timber specimens. Additional research on the semi-rigid is advised, as introducing slotted-in steel plates in the web and flange plane can significantly improve the moment capacity.

Finally, two alternative suggestions are given to avoid solving the dimensional stability issue. The first is to design a bridge with lower loading demands, like a road bridge. Secondly, the bridge can be made of only timber sections. Alternating spans would have two hinged connections to form a continuous cantilevered bridge. While the first alternative creates easier circumstances to solve for dimensional stability, the second alternative completely removes the issue by having a connection with lower demands. It is recommended to incorporate connections early on in timber design.

Human-induced rhythmic loading is an increasingly critical aspect in the design process of civil engineering structures such as sports stadiums and floors accommodating gym and aerobic classes. There are two main reasons for this trend. First, structures are becoming more slender with improvements in materials and construction techniques and modern trends in architectural designs. And second, crowds are getting livelier than previously was the case, their activities can become better synchronized due to the presence of various auditory and visual stimuli at above mentioned events.

In recent years there is also a growing discussion on the strength and stability of event structures. A very common type of structure is an event deck structure. Visitors gather on top of these decks during festivals and in some cases festival organizers place bars underneath them to efficiently use the structure. Personnel working in these bars experience the movement of the structure due to a dancing crowd and tend to feel uncomfortable. Engineers at Tentech are aware of this phenomenon through contacts with clients.

Nowadays, in the Netherlands, event deck structures are designed to withstand a vertical static load of 5 $kN/m^2$, prescribed by Dutch design codes. The amplitude of this static load is based on a dense static crowd. But according to existing literature, a synchronically jumping crowd can cause a vertical load which far exceeds the design load prescribed by design codes. This provides a reason to further investigate the extreme load case of a synchronically jumping crowd on an event deck structure.

A missing element in the standard design of structures subjected to a synchronically jumping crowd is the consideration of dynamic interaction between human and structure. In this research the focus is on how human-structure interaction (HSI) can influence the human-induced loading and the internal stresses caused by that loading in the structure. This is done by building a 3D finite element model in Abaqus.

A 3D finite element model is used to easily vary in the position of a jumper on the structure. Modelling group effects such as the coordination factor is also easier when using a 3D model. And in the case of an event deck structure a 3D model will result in more detailed results compared to 2D models because mechanical properties of an event deck can vary over the third dimension.

A mass-spring system is suggested to represent a jumping person. To simulate a synchronically jumping crowd, multiple mass-spring systems are used. By assigning an initial velocity to the mass in the mass-spring system, it is possible to simulate a similar mechanism as a jumping person colliding with a structure. By analysing the force in the spring over time it is possible to draw conclusions on the influence of human-structure interaction on the impact peak force. And by comparing the impact forces with the reaction forces which are measured at the base of the structure, the effects of structural vibrations on the internal stresses are determined.

It is found that taking HSI into account does influence the impact peak force and the internal stresses of the structure. But for an event deck structure with a natural frequency of 20 Hz or higher, this only accounts for an individually jumping person. It is found that the more people jump on a structure, the less significant the effect of HSI will be, even when a crowd tries to jump synchronically. This is caused by the natural time lag between each jumping person.

To optimize steel halls for this thesis, an optimization tool is created. This tool creates a parametric model and optimizes this model to its costs. The connections in the tool are limited to beam-column connections.
The optimization starts with the input variables of the model, the most important input variable is the type of the beam-column connection. This can vary between a hinged, semi-rigid or fully rigid connection. Other inputs are the list of profiles that need to be considered, the loads acting on the hall and the main dimensions and topology.
With all of these inputs the parametric model can be created. This model is then put into the finite element software RFEM, which calculates all the internal forces and deformations in the structure. Then with python code created for this thesis the strength and deformations for the structure are checked according to NEN-EN 1993. In case the structure is not sufficient the profile sizes are increased. This keeps increasing until the structure is sufficient. For the connection design used in this study a database is created with all possible bolted end-plate connections of the four different connection designs used in this research. Of all these possible connections the stiffness and bending moment resistance were calculated with the component method. These values are added in the database.
With the selected connection input a list is created from this database with all the connections that have a stiffness within the range of § 10% of the wanted stiffness. Then for each connection in this list the cost is calculated. This list is then sorted on the costs from lowest to highest cost. After the complete connection list is created, the first connection of this list is checked in the component based finite element software IDEA StatiCa for the deformations and strength. In case the connection is sufficiently strong according to the Eurocode, the stiffness of the connection with the actual forces is checked with the component method. This is a python script created for this tool. In case the stiffness is not within the wanted range it checks the second connection from the list. After the connection loop all the results of the optimized structure are saved. When all the results are saved, the loop is repeated for different topologies and different connection types. With these results the most cost-efficient structure design can be found including the connection design.

Additive manufacturing (AM), or 3D printing, is emerging as a technology for different applications made of steel. It is expected that in the coming years the construction industry will benefit from the free and lightweight forms that can be fabricated and the possible material savings. A few experimental results of 3D printed stainless steel are published, while no data is available on the material properties of wire and arc additively manufactured (WAAM) low carbon steel nor on specific connections. Carbon steel is widely applied in the construction industry and is less expensive, compared to other types of steel, as stainless, which was investigated for WAAM in other studies, at the TU Delft and abroad. Therefore, WAAM low carbon steel plates with thicknesses of 3 and 6 mm, produced by the company MX3D, were investigated thoroughly in this research. Tensile coupon tests were performed to determine the strength, stiffness and ultimate strain. The surface roughness, effective thickness and the influence of the thickness and the printing direction on the material properties were investigated by experiments and an evaluation of 3D scanning and Digital Image Correlation (DIC). The Archimedes’ principle was used to establish an effective thickness. The experiments were conceived to predict the tear-out failure behaviour of 3D printed plates. This is one of the basic failure modes in bolted connections. The conducted experiments provide the necessary evidence for the behaviour of single bolts and pins that interact with the WAAM material. A design formula for bolted and pinned connections is proposed based on experimental results. The results of the tensile and tear-out tests were compared with conventionally produced carbon steel and existing findings on WAAM stainless steel. The influence of the end distance of a double lap bolted connection is evaluated by comparing the experimental results with the existing standards and studies on rolled carbon steel. This research was performed to assess the applicability of current design standards for conventionally produced carbon steel to WAAM low carbon steel connections. The reliability of this method was checked by a statistical analysis. A new design factor for the tear-out strength of single bolts and pins in a WAAM low carbon steel plate is recommended. This factor reduces the design resistance of a tear-out connection and includes the ultimate tensile strength of the WAAM low carbon steel, the end distance, effective thickness of the plate and printing direction, compared to the applied force. The reduction factors that originate from the effective thickness determination and that reduce the ultimate tensile strength due to the surface roughness effect have to be incorporated for WAAM material as well. The starting hypothesis was confirmed by the investigation and following is concluded: WAAM carbon steel is a suitable material for structural applications, due to its lower costs and higher stiffness compared to the stainless steel alternatives. The experimental results of this research are meant to be used for design purposes and as input for Finite Element Modelling, so the resistance of more complex geometries can be studied accordingly. It is expected that with topology optimisation (TO) of connections, considering the printing direction, and the consequential material efficiency, full potential of 3D printing can be yielded for the manufacturing of structural parts.

Timber structures are experiencing a revelation in the build environment due to new technologies and production techniques. However, the longest spans created with timber structures still don't compete with the span widths of structures made of steel. This thesis provides a preliminary design for the long-span football stadium roof structure of The New Feyenoord stadium, which expends far beyond the existing maximum spans of timber structures. The aim is to provide insight in the feasibility of a long-span structure in timber. New Feyenoord stadium The city of Rotterdam is planning to built a new football stadium for the football club Feyenoord with a magnificent roof structure. This stadium is used to determine initial boundary conditions for the preliminary design in this thesis. A perfect bowl with an elegant flat, almost floating, roof that allows unobstructed viewing is desired. The playing field needs to be exposed to the weather conditions for the natural quality of the grass. The stadium will consist of three tiers with a total height of 40 metres, that need to be covered by a roof structure with a width of 205 metres, and length of 245 metres. Structural timber for a grand roof Timber has a high strength-to-weight ratio that is beneficial for long-span structures in which the self-weight of the structure is a significant part of the load. It has natural durability, a good performance in fire conditions, ease of workability, and a high ratio of prefabrication. The latter two result in an ease of construction and less human induced errors. Engineered wood products make it possible to create complex timber structures with more reliability than wood in its natural shape. The most promising are laminated veneer lumber (LVL) and glued laminated timber (Glulam), where LVL made of beech hardwood has the highest strength properties. Furthermore, new types of connections diminish the impact that connections traditionally have on the structural performance of timber structures. The enormous size of this stadium means that multiple connections are necessary due to dimensional restrictions for transport. Which reduces the amount of prefabrication. Structural systems for long-span stadium roofs Promising configurations for a long-span stadium roof structure are the single span, stress ribbon, shell roof system, and tension / compression ring. First, several existing stadium roof structures are presented to gain better insight on the possibilities of these systems. These stadiums are De Kuip, preliminary design of The New Feyenoord stadium by RHDHV, the Allianz Riviera stadium, the Wanda Metropolitano stadium, the Estadio Municipal, and the Tokyo National stadium. Initial designs are made for each structural system applicable to the case of the New Feyenoord stadium. These initial designs are explored on their structural behaviour and benefits for the case of the New Feyenoord stadium in combination with the material behaviour of timber. Structural forms for long-span timber structures Several structural forms are applied in long-span timber structures showing their applicability. These forms are the arch, box girder, truss, shell structure, space frame, and stress ribbon. This is explored by means of respective reference projects. These projects are the Multi-use Arena in Lisbon, Trade Fair Hall 11 in Frankfurt, the Anaklia-Ganmukhuri bridge in the Georgian Republic, the Maicasagi bridge in Nord-du-Quêbec, the geodesic domes in Brindisi, the Elephant House in Zurich, the Allianz-Riviera stadium in Nice, the Grandview Heights Aquatics Centre in Surrey, and the Essing bridge in Essing. These existing long-span timber structures mainly span only half of the required length of the New Feyenoord stadium. Possibility to create an efficient structural design for a long-span stadium timber roof structure After a rough assessment of the design concepts it is found that a combination between a tension / compression ring structure and a stress ribbon configuration shows the greatest potential for an efficient roof structure for the New Feyenoord stadium. Their are still many uncertainties for this type of structure for such a long span. There is no reference project and thus there can not be learned from mistakes. Also, the system is mathematically very complex because of its non-linear behaviour. Consequently, the structural system is modelled and verified in the parametric FEM environment of Grasshopper and Karamba3D. The found structural behaviour is verified by the strength verifications for timber structures. Tension / compression ring with radial stress ribbons The design consists of a triangular truss for the outer ring and a flat truss for the inner ring with radial stress ribbons in between. The structural design shows a feasible solution for the highest downward load, namely loads instigated by wind. The strength verification is performed on the maximum occurring force combination within a element group. Only the strength verification of shear stiffness in the bottom chord at the inside perimeter of the outer truss ring is not met. Which is a very localised effect and hence can be strengthened. Furthermore, the element groups of the rings show a undesirable utilisation distribution due to the non-circumferential perimeter of The New Feyenoord stadium. Large cross sections are required to provide stiffness to the entire structural system. The ribbons make use of their inherent bending stiffness to carry the loads in the long and short straight sides of the stadium. An elegant and stiff tensile force flow is found for the ribbons in the corners. At last, the designed connections for the ribbons which are attached to the timber fulfil the strength verifications. These connections consist of self-drilling dowels with slotted in steel plates, self-drilling screws, and glued in steel rods. Many steel fasteners are used to increase the efficiency factor of the joint. Suggested connections for the complex nodes in the ring trusses consist of slotted in steel plates with bolts and dowels, glued-in rods, and cast steel parts. The structural system is supported on roller bearings. The presented structural design only takes downward loading into account. Potential solutions for the severe upward loading are increasing the weight of the structural elements in the loaded area, tensile ties in the long and short side of the stadium, or tensile cables attached to the bottom side of the ribbons in the long and short side. An more in depth analysation of its stability against uplift forces and asymmetrical loading is needed to verify the proposed solutions. Recommendations are made to improve the design for The New Feyenoord stadium, for general possibilities for the chosen structural system, and for improvements and advice on the feasibility of special timber structures. It is concluded that a timber long-span stadium roof structure consisting of the chosen structural system shows potential to be a feasible solution for the New Feyenoord stadium. It will be a grand architectural statement that makes a stadium iconic, being the only timber tension / compression ring stress ribbon roof structure spanning with an exceptional distance.

There are several methods to perform fatigue assessment as described by the Eurocode 3 and the International Institute of Welding (IIW). The codes establish the relation between stress ranges and their respective number of cycles until failure of the detail. This method is called S-N curves, however this is based mainly on nominal stresses. A different approach is analyzed in this project, the hot-spot stress method. The hot-spot stress is used to analyze stress distribution caused by geometrical discontinuities on a welded connection. Finite element modelling (FEM) is used in order to ascertain and calculate the hot-spot stress for different details. This method consists on performing a stress extrapolation based on read-out points to avoid any peak stress caused by the finite element analysis itself.
This project focuses on a project-specific welded connection in a rail track for a movable bridge located in Tallinn, Estonia, designed by the company Witteveen+Bos. Initially, the company performed a model of the complete rail track used in the movable bridge, where the welded connection is located. This project takes a more specific scenario and develops the analysis of the local model of the connection itself. The hot-spot stress approach is taken to analyze this structural detail by means of two different finite element software, RFEM and ABAQUS, to perform a validation between these programs. In this report, a comparison is performed between modelling using shell elements and solid elements as well as mesh refinement. The extrapolation of the hot-spot stress is performed by taking the normal stresses at the surface of the element. Based on the results obtained from the analysis, this project provides recommendations when performing this type of analysis on welded connections. A design check is also performed to establish if the detail design is sufficient against fatigue, caused by the motion of the movable bridge.

The fatigue strength of the pin connection is assessed in the research project presented in this thesis. The fatigue strength is quantied by an estimated fatigue lifetime, which is the expected time to failure of the most critical point in the connection. The lifetime is estimated based on the yearly accumulated fatigue damage. A computer model of the structure has been constructed, with which time signals of the internal forces in the structure are calculated, based on the environmental loads. The pin connection was not included in this model, so internal forces in the steel tubular members have been translated manually to forces in the pin connection. The connection was decomposed in its basic components and nine potential critical locations have been identied. Stress concentrations are expected at these locations, which make them sensitive to fatigue. The internal forces were combined with the dimensions of the components, so that nominal stresses could be calculated. The hotspot stress at the potential critical locations was estimated with stress concentration factors (SCFs), which are found in literature. Relationships between the internal force and the hotspot stresses are derived in this way, and time signals of the hotspot stresses are obtained. Most of the identied locations could be assessed with hand calculations, except for two locations. They are analyzed with nite element software (ANSYS). The time signals of the hotspot stresses have been used to calculate the expected fatigue damage with the Palmgren-Miner rule. The damage is calculated for all wave directions and based on real wave statistics.

The building industry is fragmented. In the case of steel structures, two parties are involved in the design. The engineering firm designs the structure, and at a later phase the steel contractor designs all the detailing. This fragmentation restrict the possibilities of creating a cost optimal design. Currently, engineers often optimise on total weight. The result is that at a later phase, the steel contractor will need to design more complex joints, in order to create sufficient resistance in the joints of the slender structure. This can lead to higher total cost of the steel structure, compared to when joint design has been considered at the early phase. This situation leads to the following research question: “How can engineers optimise the production costs of steel trusses with welded connections?”. In this research, I will focus on trusses with welded connections which comply with transport dimensions. To answer this question, a parametric optimisation model is developed. This model combines three software packages Grasshopper, Karamba3D and IDEA Statica Connection . Grasshopper creates the wireframe geometry that resembles the structure. Karamba3D performs the structural analysis and defines cross-sections through size-optimisation. Finally, IDEA Statica Connection will design all joints, check the joints plates for the strain percentage, and optimise the welds according to the directional method.
The model shows that selected beams from the size optimisation in Karamba3D often fail the joint analysis of IDEA. This is because the strain criterion is exceeded. This failure can be resolved in the model either by modifying the cross-sections, changing the geometry of the structure, or applying stiffeners in the joint. When the truss does not show failure in the plates of the joints, the weld volume can be optimised. The optimisation model will automatically optimise the welds of each joint in the structure by using the directional method in IDEA. Optimising the welding volume, creates a significant reduction in welding volume compared to the full strength method. However, in the current state, deformation capacity in the joint cannot be guaranteed. This can prevent a plastic hinge to form and may cause premature brittle failure of the structure. To be able to use the welding volume optimisation, it should be complemented with an extra check to define whether the yield strength of the connected parts is lower than the rupture strength of the welds. Further research is recommended, to complement the weld volume optimisation with the needed additional check. Additional research into the cost-optimisation of steel structures with bolted connections and moment-resisting connections is also encouraged.

The complex mechanical behaviour of timber makes it hard to predict the failure modes in connections made of timber in Finite Element Models (FEM). The combination of various failure modes (brittle and ductile) , anisotropic behaviour, contact of steel elements and large deformations that can occur in a timber joint challenges the use of FEM. To this date no widely used approach is available for the modelling of timber connections. This knowledge gap impedes the use of large timber connections for high rise buildings in seismic regions like New Zealand. For tall seismic resilient structures a profound understanding of the various failure modes of a connection is needed to guarantee a safe design. In this thesis a new model approach with the use of cohesive elements to simulate cracking is investigated for the prediction of the mechanical behaviour of connections. An embedment test simulation is a logical step towards this connection model.

Timber can be characterised by its strong longitudinal fibres and the lignin that forms the bonding between the fibres. This anisotropic structure of the material results in a strong and stiff parallel and a weaker perpendicular behaviour of the material. Timber reacts ductile to compression loading and brittle in tension and shear loading. A typical crushing action of the timber (with micro cracking and densification of the timber) occurs when the maximum compression parallel to the grain stress is reached.

The specific manufacturing process of Laminated Veneer Lumber (LVL) reduces the inhomogeneous character of timber. This improves the strength and the predictability of the material. The cracks that occur in tension and shear can cause four different brittle failure modes in a dowelled connection (row shear, group tear out, failure of the net cross section and tensile splitting). A brittle failure mode can be prevented when minimum end or edge distances and spacing between fasteners are satisfied. In that case a ductile failure is expected with plastic deformation of the dowel and crushing of the timber underneath the dowel.

FEM is a powerful tool that is able to solve complex partial differential equation problems. Its basis lies in the linear formulation of small elements that are linked by coinciding nodal degrees of freedom to form a structure. The linear formulation has limited validity and a failure criteria is needed to define the onset of nonlinear behaviour. Multiple nonlinear approaches are available to accurately simulate the complex behaviour of timber in connections. The most promising approach is the use of cohesive elements at the locations where cracks are expected. The anisotropic nature of wood makes the prediction of crack locations in connections possible. The cohesive elements have a damage formulation to simulate strength and stiffness loss after the material strength is reached. This softening model hinders the solution procedure and therefore special solution techniques (e.g. line search, automatic stabilization and viscous regularization) are employed.

A first model is made to simulate the embedment behaviour in LVL. In the embedment tests conducted by Franke and Quenneville a steel dowel is pushed in a timber block with a pre-drilled hole. In the translation of this test to an accurate FEM model three nonlinear phenomena are simulated (cracking, crushing in compression and contact). The cracking behaviour in tension and shear is modelled with cohesive elements with a damage formulation. These cracks are inserted at the location of potential crack growth. The remaining timber has a trilinear isotropic plastic hardening formulation to accurately predict the deformations in the LVL under compression loading. The last nonlinear phenomena is contact between the steel and the timber. This is simulated as "hard" contact in normal direction and frictional contact in tangential direction.

The implicit solver encountered difficulties in converging due to contact alterations (chatter) and the softening behaviour in the cohesive elements. The automatic time incrementation algorithm reduced the increment size to overcome these difficulties. The analysis resulted in a load displacement curve that had good agreement with the experimental curve. A parameter study proved that the small difference can be related to the natural variation of material properties.

The approach of the embedment FEM was implemented in a more complex connection model. The connection tests conducted by Ottenhaus et al. that is simulated consists of 4 dowels that connect two outer LVL blocks with an inner steel plate. The spacing was chosen in such a way that a ductile failure mode was expected with brittle failure modes at large deformations. In the connection model plasticity in the steel dowels, the size of the specimens and the inclusion of tension parallel cracks increased the complexity of the model. This increased the convergence difficulties and the analysis ceased (at 0.43 mm) before the maximum load was reached .

A study was made to improve the stability of the numerical solution procedure. The impact of changing the formulations of cohesive elements, contact and the solution procedure on the convergence is tested. The viscous regularization and the initial dummy stiffness of the cohesive elements had the most influence on the convergence. With increased viscous regularization the implicit solver becomes more stable and computes more displacement increments (up to 6.91 mm). However, viscous regularization introduces artificial forces that significantly decreased the damage evolution. This prevented the formation of brittle failure mechanism.

By reducing the initial dummy stiffness of the cohesive elements (down to 2 times the timber element stiffness) the convergence improved significantly. With this initial cohesive stiffness the global softening behaviour (up to 10.08 mm) and failure development that are observed in the experiments could be simulated. The failure development consisted of the formation of plastic hinges in the dowels, tensile splitting and finally row shear failure that completely removed the supporting action of the timber under the dowels.

The decrease of cohesive element stiffness has impact on the effective stiffness of the adjacent timber elements and decreases the accuracy of the model. The model needs to be improved to make the predictions of the brittle failure development more accurate. With arc-length control, an explicit solver or the sequential linear analysis method the convergence might be increased, without the accuracy loss that is attributed to cohesive stiffness decrease. Further research is needed to improve this connection model approach.

Wire and arc additive manufacturing (WAAM), a relatively new metal 3D-printing technology, allows for fabrication of shapes that were not possible to produce a few years ago. Using 6-axis industrial robots with a gas metal arc welding (GMAW)-head, Dutch start-up MX3D prints metal in every direction mid-air. By welding small parts of material at a time along a laser guided - CAD-generated path, self-supporting cross-sections of very complex shapes, such as internally reinforced non-prismatic hollow sections can be manufactured fully automatically. Unlike powder-bed based manufacturing, this technology is not bounded by size limits of a final product, which creates opportunities for fabrication of large-scale structural components and finally entire structures. It is MX3D’s aim, in collaboration with key industry partners, to 3D-print world’s first fully functional bridge spanning 12 metres, to be installed in the city centre of Amsterdam in the course of 2019. The bearing elements of the bridge consist of tubular shaped elements, together building up to a complex-shaped free-form steel structure. These tubular members mainly behave as axially compressed bearing elements, nonetheless it is yet not known how the stability of WAAM-members can be assessed. Several standards for testing (and characterisation) of additively manufactured materials are published by different national and international committees, e.g. ISO/ASTM 529 and CEN-CT 438, but a comprehensive set of international standards has not been achieved yet. It is widely recognised that there is a strong link between manufacturing process parameters and material properties, which requires special attention in this research area. So far there is a lack of sufficient test results to cover a specific manufacturing process. In this research the material properties of MX3D’s dot-by-dot and continuously printed members are studied. This pioneering research provides insight in material and geometrical properties relevant for the stability of (ER)308LSi stainless steel tubular columns. A tubular column is used as existing basic structural element to assess the stability of wire and arc additively manufactured steel members. Advanced 3D-laserscanning technology is applied to characterise and analyse geometrical imperfections. A clear understanding of bending stiffness and buckling behaviour is established by performing four-point flexural bending tests and flexural buckling tests on tubular columns, diameter 33.8mm, thickness 3.7 mm. Tensile properties of milled WAAM specimens are analysed both in - and perpendicular to - their print direction for dot-by-dot and continuously printing. It is verified whether or not the structural properties are in compliance with the existing steel standard (EC3). Additional Vickers hardness measurements and metallographical analyses are performed to examine the microstructure of column samples. Finally, a buckling design graph is proposed based on own experiments, aiming towards a safe model suitable for stability calculations of additively manufactured tubular columns.
It is concluded from experiments that the material properties of 3D-printed steel differ from that of conventionally produced steel. Following the thermal gradient of the weld pool, large columnar grain structures are detected giving rise to anisotropy. Stiffness, strength and ductility prove to be dependent on the print direction.
Whereas the tensile strength top expected values, the obtained stiffness is significantly lower than that of commonly applied steel grades. Inaccuracies of the printing process result in local wall thickness variations and a relatively high out-of-straightness, both negatively affecting the buckling capacity of printed tubular columns.
Yet it should be acknowledged that due to rapid advancements in improving the printing process, the geometrical imperfections are soon expected to reduce drastically. It is recommended to apply active cooling during manufacturing to enhance mechanical properties even further. By combining topology optimisation and WAAM, an optimal material layout can be found ánd manufactured, consequently an excellent material efficiency can be achieved. A promising prospect for future construction projects.

Transparency and translucency are essential features of modern architecture. Glass products have been widely used in facade applications and recently their use has expanded to the load-bearing structure due their large in-plane compressive strength. However, the fragile nature of glass requires special treatment and attention to detail to avoid fracture caused by tensile stresses. Connections between glass components are very critical to the structural integrity of the system. Bolted connections, which are often used to transfer forces between glass elements, require drilling the glass. This process may cause additional flaws and stress intensifications around the holes and thus reduces the bearing capacity of glass elements. Adhesive connections offer an alternative approach for the connecting joints and enable a more uniform distribution of stresses. Laminated connections have recently been developed that combine high strength and transparency. This work focuses on the Transparent Structural Silicone Adhesive (TSSA), produced by Dow Corning, that is used for the realization of laminated connections. TSSA connections have been used in several projects worldwide; however, the hyperelastic and viscoelastic nature of the material has not been fully investigated. In this work, the mechanical response of TSSA laminated connections under static and cyclic loading is investigated by means of experimental, analytical and numerical studies.

Firstly, the shear behaviour of TSSA laminated circular connections is characterized by means of monotonic and cyclic loading tests at different frequencies. The hysteretic behaviour of the material under loading cycles, which is caused due to its viscoelastic nature, is analyzed. The adhesive exhibits significant stress-softening under repeated cycles that becomes more severe as the maximum load increases. Energy dissipation analysis is conducted to understand the nature of this phenomenon with the aim to simplify the cyclic behaviour of the adhesive for modeling purposes.

Secondly, TSSA laminated circular connections are subjected to monotonic and cyclic tensile loading of increasing maximum load. The development of the whitening phenomenon is studied for both cases. The stress level when whitening appears for the first time and the way it propagates to the adhesive surface show some consistency both for the cases of static and cyclic loading. The occurrence of the stress-softening phenomenon is also recorded, in order to observe whether it appears within the working limit of the connection.

Thirdly, the deformation behaviour of the adhesive is described analytically based on hyperelastic prediction models. Conventional test set-ups, such as uniaxial and biaxial tension tests, are combined with the simple shear test results obtained within the framework of this thesis, for the material characterization of TSSA. The hyperelastic material parameters are calibrated by a simultaneous multi-experiment-data-fit based on the nonlinear least squares optimization method. The softening behaviour observed in shear tests is modeled based on a simplified pseudo-elastic damage model, which is supported by most finite element software. A first attempt is also made to model the actual softening response of the adhesive. A less conservative approach proposed by Guo, also based on the theory of pseudo-elasticity, proved to give a good approximation of the actual cyclic response of the adhesive.

Finally, TSSA laminated connections on the edge of the glass are experimentally and numerically investigated. The edge bonded specimens are tested in shear and the stress distribution of the adhesive is analyzed by means of a three-dimensional finite element model. The distribution of stresses in the adhesive is non-linear showing significant stress peaks towards the free edges of the adhesive. A parametric study is conducted to relate the magnitude of the stress peaks with the eccentricity of the applied load.