Search

Non-destructive testing (NDT) of concrete members has been widely used for characterisation of material and assessment of functional structures without impairing their functions and performances. This thesis focuses on addressing critical challenges related to the practical implementation of NDT techniques based on wave-propagation approaches for characterisation of concrete members used in civil infrastructures. Specially, this research aims to achieve three interdependent objectives related to developing NDT techniques with piezoceramic-based transducers: monitoring of very early-age concrete hydration process, detection, and monitoring of cracking in concrete members of different complexity under loading. The concept of piezoceramic-based Smart Aggregate (SA) transducers is central to this research. Embedded SA transducers with an active sensing method have shown great potential for characterisation of construction materials such as concrete and concrete-steel composites. Based on the developed SA based approaches, an active sensing approach with appropriate arrangement of SAs in and on concrete members, and analysis of the received signal using the power spectral density, total received power and damage indexes is developed and applied in this thesis. To confirm its applicability for characterisation of very early-age concrete, a systematic investigation is performed into concrete specimens with different values of water-to-cement ratio due to slightly different initial water amounts, and different separation distances between the embedded SAs. For the detection and monitoring of cracking in concrete members under loading the mounted SA based approach is proposed and applied. It is shown that NDT systems, based on this approach, provide detection and monitoring of cracking in a variety of concrete members under loading, including relatively simple concrete beams and reinforced concrete beams under bending, and reinforced concrete slab as a part of a complex composite member under cyclic loading. Comparisons are provided between the proposed system and conventional load cell and strain gauge systems with each tested member.

Degradation of concrete members exposed to sulphuric acid environments is a key durability issue that affects the life cycle performance and maintenance costs of civil infrastructures. Groundwater, chemical waste, sulphur bearing compounds in backfill, acid rain in industrial zones and biogenic acid in sewage systems are the main sources of sulphuric acid affecting concrete structures. In this research, as part of an ongoing research on development of novel concretes for special applications, an acid resistant mortar (ARM) with current applications in lining and repair purposes was converted to acid resistant concrete in the laboratory and investigated for structural applications in acidic environments. Mechanical properties of the initial acid resistant mortar material, this novel acid resistant concrete (ARC) and a type of conventional concrete (CC), as the control, have been studied in the laboratory subjected to an accelerated test, 7% (by volume) sulphuric acid. The studied mechanical properties included compressive strength, modulus of elasticity (MOE), modulus of rupture (MOR) and indirect tensile strength tests. Apart from acid resistance experiments, other important properties for a structural concrete such as drying shrinkage and concrete performance subjected to high rate strain loads and elevated temperatures were also evaluated for ARC and CC. Structural performance of reinforced concrete (RC) flexural members made of this new concrete (ARC) and CC was assessed before and after different periods of continuous immersion in 7% sulphuric acid solution through static and cyclic loading under four-point bending tests to detect the effects of acid attack on structural performance of RC beams. Load- deflection behaviour, curvature- moment resistance at mid span, ultimate load capacity, ductility factor, stiffness degradation, dissipated energy and damping ratio were the main variables studied in these experiments. Application of ARC in beam-column joints, as another application for this concrete was also investigated due to possessing higher ductility than conventional concrete in mechanical properties tests aiming at reduction of transverse reinforcing bars in such members and the potential for seismic applications. Structural elements (i.e., beams and joints) were also modelled by using FE software ATENA to analyse the experimental results numerically. Microstructural characterisation was also performed on ARC and CC samples before and after acid exposure using scanning electron microscopy (SEM), energy dispersive X-ray spectroscopy (EDS), X-ray mapping (XRM) and X-ray diffraction (XRD) to gain a better understanding regarding the change of microstructure of materials after exposure to acid. ARC showed superior performance than CC after exposure to acid in terms of loss of mechanical properties. Structural performance of ARC has been comparable to CC before exposure to acid and after a long period of exposure to acid it showed better performance than CC, particularly in terms of load bearing capacity. The application of ARC in beam-column joints allowed reducing transverse reinforcing bars in these joints (50% compared to CC). Microstructural characterisation also revealed significant facts regarding the deterioration mechanism in both types of concretes and their effect on mechanical properties.

Composite steel-concrete beams are common structural elements formed by connecting a concrete slab to the supporting steel beam with shear connectors. This type of construction is considered favourable for bridge design and is extensively used in high rise buildings. These beams are also used frequently in situations where axial forces are introduced. Such situations may occur through the installation of post-tensioning cables or through the inclination of the beam such as in stadia or ramps. In these situations, the beam may be subjected to any combination of flexure and axial load. However, modern steel and composite construction codes currently do not address the effects of these combined actions. The primary objective of this thesis is to study the behaviour and strength of composite steel-concrete beams subjected to combined flexure and axial load. The research comprises two main parts: a thorough experimental program and an analytical model. The experimental program consists of 24 full-scale composite beams. Each specimen is tested to failure to determine the behaviour, ultimate strength and failure mode. The analytical model is based on cross-sectional analysis (CSA) using a strategy of successive iterations. Results derived from the model show an excellent agreement with the experimental results. Using the analytical model, a parametric study is conducted to investigate the effect of axial load on the flexural strength of composite beams with full and partial shear connection (PSC). The study shows the hogging bending capacity of a member is unaffected by a relatively low axial tensile force. For beams subjected to sagging bending, the tensile capacity of the section is unchanged by moments up to approximately 40% of the sagging moment capacity. The application of axial compression to a beam is immediately detrimental to the hogging moment capacity of the section. This is due to initiating an earlier onset of buckling. The strength of a specimen subjected to combined sagging bending and compression is shown to be crucially dependent on the degree of shear connection. With full shear connection, the CSA and rigid plastic analysis (RPA) both show that the ultimate sagging moment capacity remains unchanged for axial compression loads less than approximately 60% of the squash load. However, significant additional stiffening is required in order to achieve axial compression forces greater than the squash load of the bare steel section. The RPA is shown to be a feasible design tool for axially loaded composite beams. In most cases, the RPA produced results similar to the CSA and the experiments and is generally more conservative. Some overestimation of the flexural strength occurs particularly for the sagging bending and axial compression case with low degrees of shear connection. Design models are proposed for estimating the flexural strength of an axially loaded member with full or PSC. These design models are based on results from the CSA, RPA and experiments as well as a finite element analysis performed in a concurrent study. It is hoped that the findings herein provide an improved understanding of the complex behaviour of composite steel-concrete beams and contribute to the continuing evolution of composite steel-concrete building structures.

Composite steel-concrete construction has been widely used around the world. Techniques for the ultimate load analysis and design of composite steel-concrete beams are also well-established and solutions can be obtained with relative ease. However, beams under combined actions such as edge or curved in plan beams subjected to torsion and flexure can be difficult to ascertain their strengths and behaviour due to their complex stress state. These effects of combined actions are not currently addressed in the Australian Standard AS2327.1 (Standards Australia, 2003) or any their international codes. Therefore, this thesis will present two full-scale experimental series to investigate the behaviour of straight and curved in plan composite steel-concrete beams subjected to torsion and flexure with the influence of partial shear connection. Push-out and material tests were carried out to determine the material properties of each component of the composite steel-concrete beams.

Reactive Powder Concrete (RPC) is a relatively new ultra high strength concrete with a wide range of capabilities. Currently it is used in high performance applications where either impact resistance, high durability and/or member size are governing factors in design. The relatively high cost, required production control, and lack of industry knowledge have generally precluded its widespread use in more common engineering applications. Accordingly, the major component of this Thesis aims to contribute new knowledge with respect to the understanding of the material behaviour and damping characteristics of RPC. To achieve this goal a staged experimental study was undertaken encompassing the following (1) the production and material behaviour including a parallel study on the durability of RPC in an aggressive environment (2) the static performance and damping characteristics of RPC when utilised in beam elements. The initial component of the investigation involved several trial mix programmes to ascertain the viability of RPC production using local Queensland materials. A further large scale production programme of RPC was carried out utilising two mixer types: (1) vertical pan mixer representing a precast mix and (2) a mobile inclined drum mixer representing an onsite mix. The results indicated that the onsite application of RPC was viable and resulted in only a minor reduction in characteristic strengths when compared to precast operations. A parallel experimental study investigating the relative performance of RPC immersed in acidic solutions indicated that the range of RPC’s investigated were more susceptible to sulphuric acid attack when compared to a standard high strength concrete (HSC) used in the investigation. Following verification of RPC characteristic strengths, twenty three test beams were produced as part of a large scale RPC production programme. These test beams were incrementally loaded to failure under simply supported conditions during which static measurements were observed both under load and upon the removal of load. The results indicated that the fibre reinforced (FR) RPC test beams exhibited a higher load capacity, an increase in first cracking loads, significantly reduced crack widths and an overall reduction in shear behaviour when compared to the non-fibre reinforced (NF) RPC and HSC test beams. Concurrent free vibration tests through the use of an impulse force hammer were carried out during the static loading of each of the twenty three test beams in an effort to ascertain the damping characteristics throughout the various levels of test beam damage. A new program using analytical and theoretical techniques was developed to analyse the excitation response histories (ERH) obtained at each load increment, the program was found to be more efficient when compared to previous analysis methods. A method of assessing the current and previous analysis methods was also incorporated within the program enabling the user to compare the damping ratio determined through the various analysis methods.

Composite steel concrete beams are formed by connecting concrete slabs to a supporting structural steel beam. In the early 1900’s, composite beams were considered favourable for bridge design but in recent decades, composite steel and concrete structures are employed extensively in modern high rise buildings. The flexural strength of composite beams is greatly influenced by the strength and ductility of the shear connectors between the structural steel beam and the concrete slab. The behaviour of the shear connectors is important in understanding the shear force transmitted and the degree of slip which occurs at the interface of the steel and concrete. Composite steel-concrete beams are becoming increasingly popular in multistorey buildings due to their higher span/depth ratio, reduced deflections and increased stiffness value. However, their performance is highly dependent on the load-slip characteristics of the shear connectors. More recently, trapezoidal profiled slabs are becoming increasingly more popular for high-rise buildings when compared with solid slabs because they can achieve large spans with little or no propping and they require less concrete and plywood formwork. However, the profiles used to achieve these savings can have a detrimental effect on the shear connector behaviour. The issues emphasised in this thesis are: the effects of steel fibres as a strengthening system in composite steelconcrete beams: the effects of elevated temperatures on the behaviour of headed stud shear connectors for composite steel-concrete beams: the long term effects on the behaviour of the composite steel-concrete beams: the effects of strain regimes on the behaviour of the composite steel-concrete beams: the effects of the combination of axial tension and shear loading on the behaviour of composite steel-concrete beams. This thesis consists of many of numerical studies. They include the mechanical behaviour and various loading conditions of the behaviour and strength of headed stud shear connectors for composite steel-concrete beams. The finite element package known as ABAQUS was used to compare the existing experimental studies by other researchers. The outcome of the numerical analysis is very satisfying and also contains design recommendations. Furthermore, a methodical parametric study is undertaken using the model that has been properly calibrated. These parametric studies broaden the range of application for design guidance. It is fervently hoped that the findings herein are useful in contributing further insight to the continuing evolution of composite steel-concrete structures.

In many reinforced concrete structures built in the 1970's and earlier, bottom beam reinforcement is not continuous and if a column support is lost due to terrorist attack or other unexpected action, the structure could be vulnerable to progressive collapse. The use of CFRP material may provide a solution for rehabilitating such structures. CFRP materials cannot develop full tensile capacity unless they are properly anchored to the reinforced concrete structure. The intent of this study is to find an effective method of anchoring CFRP material to a reinforced concrete beam so that the ultimate tensile strength of the CFRP is realized. In this study, ten reinforced concrete beams rehabilitated using different configurations of anchors were tested to assess the effectiveness of the anchors. Both CFRP anchors and CFRP U-wraps were investigated. The rehabilitated beams were loaded until failure of the CFRP material or anchor occurred. Different failure modes, strengths and deformation capacities of the rehabilitated beams were observed depending on the configurations of anchors. The maximum capacity was obtained using a combination of CFRP anchors and U-wraps.

The inclusion of trapezoidal types of steel decking in the shear connection of composite beams has been found to significantly reduce their maximum strength and ductility by causing premature concrete-related failure modes. In order to investigate the complex behaviour and various load-transfer mechanisms that can occur in composite beams incorporating this type of connections, a total of 91 carefully-designed push-out tests were performed. Specific failure modes in conventionally reinforced specimens were initially induced by varying critical parameters. Specimens incorporating specific stud reinforcing devices were subsequently tested to suppress the undesirable failure modes. The concrete reinforcing and stud performance-enhancing devices, which included novel waveform-type reinforcement elements and spiral wire or ring components surrounding individual studs in secondary composite beams and special haunch reinforcement in primary beam applications, significantly delayed the onset and reduced the effect of the premature concrete-related failure modes. Hence, they increased the ultimate strength and ductility of the shear connection. The findings of the small-scale push-out tests were also verified in two full-scale composite beam tests which showed good agreement in shear connection behaviour and failure mode. Most of the design approaches currently used around the world take into account the weakening effect of trapezoidal types of decking by applying a reduction factor to the nominal strength that the same connection would have in a solid slab. From the test results, it is evident that not every shear connection incorporating steel decking, and within the limits of the associated standards, can be classified as ductile. A new and more reliable design approach is proposed which also incorporates the application of the various stud reinforcing devices. The key element of this design approach is to classify the anticipated connection behaviour, in regards to its deformation capacity, into ductile or brittle connections, hence ensuring satisfactory shear connection behaviour where the new types of trapezoidal steel decking are used. A reliability analysis of the new proposal is presented which enables the application of this new approach in accordance with AS 2327.1 (Standards Australia 2003). It is calibrated to provide a reliability index similar to stud applications currently in use. Simple strength reduction factors for the types of trapezoidal steel decking available in Australia are also provided which can be applied to the current solid slab shear connection strength for a fast and simplified design.

The objective of this research project was to create a Finite Element Routine for the Linear Analysis of Post-Tensioned beams using the program CALFEM® [20] developed at the division of Structural Mechanics in Lund University, Sweden. The program CALFEM and our own made files were written in MATLAB, an easy to learn and user-friendly computer language. The approach used in this thesis for analyzing the composite beam consists in embedding the steel tendons at the exact location where they intersect the concrete parent elements, without moving the concrete parent element nodes. The steel tendons are represented as one dimensional bar elements inserted into the concrete parent elements, which at the same time are represented as 8 node Iso-parametric plane elements. The theory presented in Ref. [4] served as basis for the modeling of the post-tensioned beams; however it only explained the procedure for modeling simple reinforced concrete beams, due to this we needed to make the appropriate adjustments so we could model post-tensioned beams. Assembly of the tendon stiffness into the concrete elements will depend on the bond interface between the steel and concrete, this bonding effect will be modeled using link elements; the stiffness of this link element used in the concrete-tendon interface will be the change in cohesion (between the grout or duct and the steel tendon) at the interface due to the relative slip between the concrete and the steel elements nodes. Loads (Distributed, Concentrated or Post-Tensioning) are applied directly into the concrete parent elements, and then from their resultant displacement the displacements and forces of all the steel tendon elements are obtained, this is done consecutively for all the post-tensioned tendons at every load increment. Four examples from different references and software programs are solved and compared with our results: (1) A simply reinforced cantilever plate. (2) A reinforced concrete beam, under the effect of a vertical concentrated load at mid-span. For this problem the force distribution along the steel reinforcement is obtained for two conditions, perfectly bonded and perfectly un-bonded, our results are compared with the ones obtained with the program SEGNID. (3) Consists of a continuous un-bonded post-tensioned beam with two spans, without stress losses on the tendon. The reactions at the supports and the concrete stress distribution at the location of the mid-support are obtained after the post-tensioning force is applied at both ends. (4) Consist on a un-bonded post-tensioned beam with stress losses on the tendons due to friction, wobbling and anchorage loss, under gradual loading and consecutive post-tensioning of two tendons, the results are compared with the ones reported using the program BEFE [5] developed at the University of Technology Graz, Austria. The results obtained using our program are very similar to the ones obtained with the other programs, including the more powerful curved embedded approach used by BEFE [5].

The objective of this study is to develop a fundamental understanding of the flexural behavior of Reinforced Concrete (RC) beams strengthened using externally bonded Fiber Reinforced Polymeric (FRP) laminates. An experimental program was carried out where reinforced concrete beams were fabricated, strengthened with Carbon Fiber Reinforced Polymeric (CFRP) laminates, and tested. In addition, material characterization and interface bond tests were conducted. Also, an analytical program was carried out to model the full response of beams under bending, using a mechanics approach. The interface between the FRP laminate and concrete was modeled by two different approaches to define the stress distributions at this level. Failure of the concrete cover was modeled using an energy approach. It was found that interfacial bond failure on the concrete side is the most important type of failure of reinforced concrete beams strengthened by externally bonded FRP laminates. The interfacial bond strength is adversely affected by environmental conditions, such as freeze-thaw cycles. The proposed analytical model, based on equilibrium and strain compatibility, was able to accurately predict failure modes, such as concrete compression crushing and FRP tensile rupture. It was observed that a uniform shear flow assumption is acceptable as an approximate model interfacial bond failure. The predicted magnitude of the stresses at the interface epoxy level is significantly reduced when the material nonlinearity is taken into account. The onset of the interfacial failure is limited by an upper and a lower bound. The strength of the adhesive layer to withstand a biaxial state of stresses (tension, shear) represents the upper bound. This failure does not seem to occur for the current materials used. The strength of the concrete under biaxial state of stresses (tension, shear) represents the lower bound. The methodology proposed in this study, which takes into consideration fracture energy and bonded area of FRP laminate, leads to a reasonable prediction of the occurrence of concrete cover failure, or tooth failure.

This thesis presents a theoretical and experimental study into the local buckling of steel plates that have been mechanically attached to the sides of reinforced concrete beams in a retrofit procedure. The study has considered the elastic buckling of glued side plates as well as the elastic, inelastic and elastic post-buckling response of bolted side plates. The buckling capacity of bolted side plates, when subjected to elevated temperatures, has also been determined. A buckling design example, which illustrates a culmination of the theoretical and experimental work developed in this thesis, is finally presented. Reinforced concrete structural elements can deteriorate with time, primarily through concrete degradation or reinforcement corrosion and the like. They can also be damaged from seismic activity, or their use as a structural element may change and as a consequence so will their strength and stiffness requirements. A relatively simple and cost effective solution to upgrade unserviceable reinforced concrete elements is to retrofit them by attaching steel plate to their sides. The initial stage of this thesis demonstrated the beneficial effect side plates have on the overall strength, stiffness and ductility of the resulting composite beam. When the retrofitted beams are subjected to their applied actions, load is transferred to the steel plates via the mechanical shear connection. As a result, in-plane stresses are produced in the steel plates and this can precipitate local bucking. The local buckling mode is unilateral, since the concrete beam constrains the steel to buckle locally away from the concrete core, and this is referred to as a contact problem. The Rayleigh-Ritz method was utilised in this thesis to study the stability of unilaterally constrained steel plates. The Ritz displacement function consisted of two dimensional simple or orthogonal polynomials multiplied by an additional polynomial describing the plates geometric and kinematic boundary conditions. The restraining medium was modelled as a tensionless foundation and this restraint condition was introduced through a penalty method formulation. Extensive comparative, convergence and parametric studies have been undertaken by considering a wide range of unilaterally constrained plates. Such plates possessed varying kinematic boundary conditions and loading scenarios that were general and not all strictly germane to the side plated beam problem. Based on the theoretical results, slenderness limits were proposed and these were found to be useful in the design of side plated reinforced concrete beams against premature failure from local plate buckling. An extensive experimental programme was conducted which utilised local buckling push-tests and full scale side plated reinforced concrete beams. A comparison was made between the results obtained from the experimental programme to the theoretical predictions and a good agreement was found to exist. The theoretical results, where applicable, were then calibrated from the experimental findings to account for initial imperfections and residual stresses. In conclusion, it was found that the developed theoretical and experimental programmes were applicable for studying the behaviour of composite structures in general where a thin steel plate has been juxtaposed with a rigid concrete core. Recommendations for future research have also been given which will serve to further increase understanding into the retrofit of reinforced concrete structures.

Abstract: "Spring, 1984."

Research was conducted to apply the Spectral-Analysis-of-Surface-Waves (SASW) method to the identification of internal cracking in concrete structural elements such as beams and columns. The SASW method is a nondestructive technique which involves the propagation of surface stress waves along the exposed face of a structural element. By yielding the relationship between wave velocity and wavelength of the surface wave energy, the SASW method provides information about how the stiffness of a material varies with depth. SASW measurements were made on intact and cracked concrete structural elements. Surface waves propagated through cracked concrete at consistently lower velocities. These velocity reductions corresponded to the presence of visibly detectable cracking and, in some cases, occurred in the absence of visible cracking. This result demonstrated the ability of the SASW method to detect otherwise hidden damage. When the cracks were reclosed by the application of a compressive load, the surface waves propagated at velocities comparable to those of uncracked concrete. In addition to performing velocity measurements, material damping measurements were made on intact and cracked concrete structural elements. These measurements revealed that the presence of cracking can be qualitatively assessed through an increase in observed surface wave material damping, with material damping ratios ranging from less than 1% in undamaged concrete to around 3% in damaged concrete

This thesis examines the design and overall behaviour of haunched composite beam frames braced against sidesway. It has been compiled as a result of both a theoretical and an experimental study. A detailed design method is presented, which incorporates options for plastic or elasto-plastic analysis at the Ultimate Limit State. A method for checking the susceptibility to lateral instability in the hogging regions of composite beams is also presented, and it has been developed by using elastic energy theory. The design method is intended to be applied in conjunction with the British Standard for the design of steel-framed building structures, BS5950 Parts 1 and 3, although the engineering principles involved could be used with other countries' codes. A detailed design example is also included to demonstrate the application of the method. A review of other relevant experimental and theoretical work has also been undertaken and is critically appraised where appropriate. The experimental programme carried out as part of this work involved two full-scale tests, and these are described. Particular attention was given to the stiffening details at the shallow end of the haunches, and several options were tested. The most successful proved to be that where full-depth web stiffeners were fitted at this position. Analyses of the tests are included and they show that the theoretical predictions are generally slightly conservative. A finite element model of the hogging region of a haunched composite beam was also developed, and this was applied to one of the tests. The results of this are presented. It is intended that this particular work will enable further computer simulations to be carried out. To enable early dissemination of the ideas developed in this thesis, during the course of this work an interim design guide, written jointly by this author and his industrial supervisor, Dr R M Lawson, was published by the Steel Construction Institute.