<p>A research programme is presented in which 14 HSS Vierendeel joints were tested. The specimens tested were all unequal width connections and two distinct b/h values were used (0.60 and 0.83). Four specimens were unreinforced while the remaining ten had three distinct reinforcement types. These were branch member flange stiffening plates, chord member top flange stiffeners and branch member offcut haunches.</p> <p>Each specimen was loaded to failure and load-deflection and moment-rotation curves were plotted. In addition, the joint modulus (J) for each joint type was calculated.</p> / Master of Engineering (ME)

<p>With the current pressures to conserve energy and protect the environment, research to take advantage of the cementitious nature of vitreous pelletized blast furnace slag (a byproduct of the steel industry) should yield both economic and technical advantages to the Canadian construction industry.</p> <p>Three specific areas were chosen for study.</p> <p>1. The development of a test method to quantify the degree of vitrification achieved in the quenching of blast furnace slags (a property known to have fundamental influence on its reactivity) and the evaluation of new and existing test methods with regard to simplicity and accuracy for potential use as a basis for commercial quality control.</p> <p>2. The study of the incorporation of slag cement in autoclaved binders having a wide range of compositions in the ternary system containing slag, portland cement and ground quartz (silica flour).</p> <p>3. The effects of variations in the physical and chemical properties of slag cement on autoclave reactivity.</p> <p>Within the previously defined areas, a few of the findings are given.</p> <p>1. From analysis of glass content determinations, the method adopted is critical to the results obtained. The McMaster Individual Particle Analysis was found to give the most accurate and reliable glass content values when compared to a QXRD standard procedure and when used to predict strength potential.</p> <p>2. Optimum strengths were obtained with ternary slag-portland cement-silica flour binders containing 60 to 75 per cent slag. When autoclaved 4h at 185°C, high strengths corresponded to the presence of semi-crystalline C-S-H resulting in fine pore size distributions, and mixtures of C-S-H and ∝C₂SH. It was also found that slag was activated by silica flour alone resulting in high compressive strengths, high tensile to compressive strength ratios and relatively well crystallized tobermorite.</p> <p>3. The degree of vitrification of slag was found to have the most significant influence on strength of autoclave cured pastes but the trend was not clear at high glass contents. The CaO and MgO contents were significantly related to strength when combined with the degree of vitrification. However, for evaluating slag hydraulicity, it is suggested that only physical strength testing has the authority for engineering decision making.</p> / Doctor of Philosophy (PhD)

<p>Although panel buildings have been constructed since before World War II, there is very little known about their ultimate capacity when subjected to lateral loads. The existing methods of analysis and design are based on tests conducted on isolated vertical and horizontal joints, and empirical relationships derived from these experiments.</p> <p>Very little attention has been paid to the influence of joint behaviour on the response of an assembly of panels. The studies of the overall behaviour of panel buildings which have been done by constructing mathematical models for panel assemblies, have not in general devised realistic models for the joints.</p> <p>Since the joints are weaker than the panels they connect, and because failure in almost every case occurs in the joints, a realistic joint representation is very important if a true prediction of lateral load response of panel buildings is to be provided.</p> <p>In this work an attempt has been made to develop a realistic model for both the horizontal and the vertical joints of the assemblies of panels forming the shear walls in panel buildings. In determining the joint model, joint details as well as existing experimental information and on-site construction techniques (and possible imperfections) were taken into account. The mathematical model was based on the finite element method of analysis. Due to practical limitations of computer storage and computation time, modifications had to be made to reduce the total number of elements used to simulate the panel assemblies.</p> <p>The results confirmed the immense importance of the realistic modelling of joints, and also illustrated the susceptibility of horizontal joints to failure due to the redistribution of gravity loads during lateral loading. It was shown that the method is capable of predicting the ultimate lateral load capacity of panel buildings.</p> <p>Recommendations were made regarding an acceptable limit for drift to height ratio (Δ/H) for panel construction. Factors of safety against collapse were also established based on the maximum allowable lateral loads to which an assembly of panels could be subjected without compromising the overall integrity of the structure, or incurring minor local failures.</p> / Doctor of Philosophy (PhD)

<p>Empirical, analytical, and semi-rational methods for the stability analysis and design of underground openings are generally not adequate to consider the major design parameters for tunnels in squeezing ground. Tunnel construction in squeezing ground and the increasing North American use of the precast concrete segmental liners were the prime motivation for developing an elastic-plastic-creep simulation of tunnel excavation and liner placement.</p> <p>Based on the finite element method, the simulation method incoporates: the primary state of stress; excavation in stages; liner placement delay; and determination of the elastic-plastic-creep response for each stage of construction, including ground-interface-support system interaction for both plane strain (two-dimensional) and advancing face (three-dimensional) conditions. Construction sequences of excavation and support placement are simulated using 'deactivation' and 'reactivation' operations on the stiffness matrix terms corresponding to the selected rate of face advance. Incorporation of the inelastic behaviour of both ground yielding and time-dependent deformations is based on the initial stress and initial strain methods, respectively. The three-dimensional stress analysis near the advancing face is based on an axisymmetric approach and Fourier series approximation that allows the non-symmetric radial and axial loadings due to the primary state of stress to be considered. The axisymmetric approach is also extended to include inelastic ground behaviour.</p> <p>A comprehensive design for a tunnel in a formation that may exhibit squeezing is given. Using this typical field problem, several parametric studies are presented. It is shown that the elastic deformations are only significant near the advancing face and their influence is very minor at a distance of about two times the tunnel diameter. The consideration of the actual advancing face condition, lining placement delay, and the use of a soft backing can significantly reduce the design stresses developed in the liner.</p> <p>The simulation method presented can be used to develop improved ground characterization through the monitoring of tunnel convergence followed by back analysis.</p> / Doctor of Philosophy (PhD)

<p>Many lakes act as receiving waters for polluted runoff and other effluents. Water quality problems in lakes depend for their management on numerical lake transport models. Lake transport models have recently become quite detailed and realistic. However, a difficulty still facing this field is the determination of the interfacial transports at the boundaries and related model parameters. Numerical modelling of lake physics and bio-chemistry requires special treatment of the boundaries at the air-water interface, sediment-water interface at the bottom of the lake and horizontal densimetric fluxes of active materials across the lake boundaries into and out of the lake. This latter problem is particularly complex when two interconnected lake basins of dissimilar characteristics exchange flow via communicating channel.</p> <p>In this study, the following important parameters were derived and related to lake and flow conditions: surface drag coefficients, vertical eddy diffusion coefficients and bottom roughness.</p> <p>At the air-water interface, momentum transfer is studied and the over-water wind stress is determined using the logarithmic wind velocity distribution and Von Karman's integral equation for turbulent flow over a rough movable surface of variable roughness. Thus the wind drag coefficient is determined as a function of wind-and-wave characteristics.</p> <p>At the lake bottom, shear stress due to a very rough lake bed, with weed growth, is determined as a function of local parameters such as weed length and distribution. The approach is particularly suitable for shallow lakes with considerable roughness. Also, vertical transports at the sediment-water interface and in the water column above the sediments is studied using two natural tracers: temperature and Radon-222 isotope. Vertical eddy diffusivity in the lower layers of a lake is determined as a function of depth and time by using the observed temperature and Radon profiles.</p> <p>At the lateral boundary, exchange flow between two interconnected stratified bodies of water is investigated. The local exchange is modeled through a proper treatment of stratified flow at the open boundaries at the two ends of the "connecting channel" and proper presentation of the effect of the adjacent basins.</p> <p>Numerical models are developed to study the above processes and boundary conditions. The computational algorithms are tested for theoretical and computational performance. Numerical models are tested against available field observations at Valens reservoir, Hamilton harbour, Baldeggersee (Switzerland) and Lake Erie.</p> <p>A three-dimensional hydrodynamical model of lake currents is applied to two, non-stratified and stratified, lakes during summer periods where field observations were carried out: Valens reservoir and Hamilton harbour.</p> / Doctor of Philosophy (PhD)

<p>Composite steel-concrete beams often form part of multi-story frames. However, very little information is available on which to base the design of composite beam-to-column connections. As a result, these connections are currently designed on the basis of the connecting steel elements. Most of the information available is limited to continuous composite beams subjected to gravity loads which may be used as a part of braced multi-story frames. Limited information is available on the behaviour of composite beams is unbraced multi-story frames subjected to combined gravity and lateral loads.</p> <p>The object of this thesis is to study the behaviour of composite beams with ribbed metal deck in an unbraced multi-story frame. Of particular importance in this investigation is the effective slab width which is to be used in evaluating the strength of the composite beam at the connection and the effective slab width which is to be used in evaluating the stiffness of the composite beams in the positive and negative moment regions of an unbraced frame subjected to combined gravity and lateral loads. A combined analytical and experimental investigation was conducted.</p> <p>A method of analysis has been devloped using a combination of the finite difference and finite elements methods to model the composite steel-concrete beams in the various moment regions. The analytical model was used to study the parameters affecting the effective slab widths, for strength and stiffness, and consequently the behaviour of composite beams in an unbraced frame. Curves showing the variation of the effective slab width for strength and the effective slab width for stiffness with the side ratio (L/b) and the column width to slab width ration (c/b) in the case of complete interaction and a method for evaluating the effective slab width for stiffness for any degree of interaction are presented.</p> <p>The analytical model was used to study the effective slab widths for composite simple beams subjected to a central point load, and the results were compared with the different code formulae. The results showed that the AIJ formula agrees very well with the analytical results of the effective slab width for strength while the CSA S16.1 formulae provides a good estimate for the effective slab width for stiffness especially for (L/b) greater than 4.</p> <p>In the experimental program, 10 composite beam-to-column connections were tested under positive moment conditions to stimulate that portion of the composite beam in an unbraced frame between the point of contraflexure and the column face and subjected to positive moment. The test results showed that the strength, stiffness and ductility depend on the (L/b) and (c/b) ratios, steel beam size, slab thickness and the existence of lateral support. The test results showed also that the composite beam-to-column connections possess sufficient rotattion capacity to enable plastic design to be applied to unbraced frams with composite beams.</p> <p>The test beams were analysed using the analytical model and very good agreement was obtained between the experimental and analytical results.</p> / Doctor of Philosophy (PhD)

<p>This dissertation presents an inelastic analysis of the behaviour of composite beams with ribbed metal deck.</p> <p>A layered finite element model is used to allow for any variation in material properties through the thickness. An incremental and iterative technique is adopted using the tangent modulus stiffness approach.</p> <p>The dependability of the model is checked by means of comparison with some experimental results, obtained from testing composite beams with solid and ribbed slabs.</p> <p>A study of the effect of the type of loading and the transverse moment in the slab on the deformation and the ultimate capacity of composite beams with ribbed metal deck, is presented.</p> <p>The effective width of composite beams with ribbed metal deck, subjected to uniformly distributed load, is investigated. The effective width in the inelastic stage and at the ultimate load are also studied.</p> <p>Finally, a study of the longitudinal cracking of composite beams with ribbed metal deck, is presented. Some design recommendations are presented to account for the longitudinal cracking of composite beams with ribbed metal deck, subjected to a uniformly distributed load over the entire slab.</p> / Doctor of Philosophy (PhD)

<p>The lack of understanding the behaviour of concrete masonry and the complex interaction existing between its components (block, mortar, and grout) at failure may be the cause of the continued use of the code's working stress method. This approach could underestimate the potentials of masonry as a construction material. It is the main objective of this investigation to provide a better understanding of concrete masonry behaviour under different in-plane load conditions (compression, tension, shear and biaxial stresses) considering the anisotropic nature of masonry as a composite material. This understanding was gained through a combined experimental and analytical investigation.</p> <p>In the experimental study, 323 masonry assemblages were tested under compression normal and parallel to the bed joints, splitting tension at different orientations from the bed joints, shear along the bed joint with different levels of precompression, and off-axis compression and tension to produce biaxial states of stresses along the bed and head joints. The test material variables were mortar type, grout strength, and bed joint reinforcement.</p> <p>Analytical strength formulas, based on a "strength" approach, are proposed to express, in quantitative terms, the assemblage compressive strength normal to the bed joints, tensile strength normal, diagonal, and parallel to the bed joints, and shear strength along the bed joint with and without precompression. The applicability of the failure theories for both isotropic and composite materials to masonry were examined. Failure criteria are proposed to predict the strength and the failure mode of concrete masonry under biaxial stresses taking into account the anisotropic nature of masonry as a brittle composite material. Design code provisions (CSA S 304) for plain masonry are reviewed.</p> / Doctor of Philosophy (PhD)

<p>No rigorous explanation exists for the behaviour of reinforced concrete joints. The lack of understanding the complex interaction may account for continued use of inadequate joint details particularly for knee joints subjected to opening moments. The object of this research was to provide a better understanding of joint behaviour through a combined experimental and analytical investigation.</p> <p>In the experimental program, 6 specimens with different types of joint details were tested under opening moment. These details reflected recent recommendations by other investigators and some adaptations of their ideas. The results served as evidence of the ability to produce effective and simple details as well as serving as a basis for evaluating the reliability of the analytical model.</p> <p>A method of analysis has been developed using the finite element method to model the behaviour of reinforced concrete including the interactions between concrete and steel in the form of bond and dowel forces. One of the principle features of this method of analysis is the built in ability to trace the propagation, location and orientation of cracking. This feature was tested by comparing predicted behaviour of beams failing in shear and flexure with test results. Very good agreement with the well documented phenomena of various modes of cracking and failure provided evidence of the validity of the analytical method.</p> <p>The analyses of the joints reproduced cracking patterns and sequences as well as capacities which are quite close to the test results. The analytical information helped to identify the primary causes of failure of premature failure and thereby provided a more rational basis for suggesting alternate joint details which will behave much better than inadequate details which are currently recommended.</p> / Doctor of Philosophy (PhD)

<p>Fiber reinforced polymer (FRP) - confined concrete-steel composite columns have been recently introduced as a retrofit technique for structural steel columns. This technique involves placing an FRP tube around an in situ steel column and subsequently filling the void between the steel section and the FRP tube with concrete to create a composite column. The composite action occurs due to the concrete encasing the steel section and the FRP confining the concrete. An experimental investigation has been undertaken to evaluate: the effect of adding a steel W section to confined concrete; the effect of using a split tube system as a practical application technique; and the effect of shrinkage reducing admixture in confined concrete. A total of eighteen stub columns, consisting of six different column types were tested to meet these objectives. It was found that the load-displacement response of confined concrete that includes a steel W section is similar to the response of confined concrete plus the elastic-perfectly plastic steel contribution. Experimental findings demonstrate the enhanced performance of the proposed split tube retrofit technique. An analytical model was used to predict the response of the different column types. A method is proposed to incorporate the increased stiffness of the FRP jacket in the split tube system and is shown to be an improvement on the original model. A parametric study was also performed on the revised model.</p> / Master of Applied Science (MASc)