Thesis (M.Phil.)--Hong Kong University of Science and Technology, 2005. / Includes bibliographical references (leaves 101-106). Also available in electronic version.

Submitted towards the Degree of Master of Technology in Mechanical Engineering at the Cape Peninsula University of Technology / A Small Punch Test is a non-destructive technique for evaluating mechanical behaviour. The main advantage of this testing technique is the fact that material can be extraxted from a component in service due to the small dimensions of the speciments.Typical test specimens cut from components are similar in size to a normal human fingernail.

During its service life, a structure may be subjected to various environmental and loading conditions. However, in general, the properties determined under one set of conditions may not be used to determine the behaviour of the material under a different set of conditions. For example, it is well known that concrete is a strain rate sensitive material; therefore, its properties determined under conventional static loading cannot be used to predict the performance of concrete subjected to high strain rates. The problem is serious because these high strain rate loadings are associated with large amounts of energy imparted to the structure in a very short period of time, and concrete is a brittle material. Since the strain rate sensitivity of concrete prohibits the use of its statically determined properties in assessing its behaviour under dynamic conditions, high strain rate tests are required. Impact tests were carried out on about 500 concrete beams. An instrumented drop weight impact machine was used. The instrumentation included strain gauges mounted in the striking end of the hammer (called 'the tup'), and also in one of the support anvils. In addition, three accelerometers were mounted along the length of the beam in order to obtain the beam response, and also to enable the inertial correction to the observed tup load to be made. Two different concrete mixes, normal strength with a compressive strength of 42 MPa, and high strength with a compressive strength of 82 MPa, were tested. The effect of two types of fibres, high modulus steel, and low modulus fibrillated polypropylene, in enhancing concrete properties was investigated. In addition, tests were also conducted on beams with conventional reinforcement. Hammer drop heights ranging from 0.15m to 2.30m were used. Static tests were conducted on companion specimens for a direct comparison with the dynamic results. In general, it was found that concrete is a very stain rate sensitive material. Both the peak bending loads and the fracture energies were higher under dynamic conditions than under static conditions. Fibres, particularly the steel fibres, were found to significantly increase the ductility and the impact resistance of the composite. High strength concrete made with microsilica, in certain circumstances, was found to behave in a far more brittle manner than normal strength concrete. High speed photography (at 10,000 frames per second) was used to study the propagation of cracks under impact loading. In general, the crack velocities were found to be far lower than the theoretical crack velocities. The presence of reinforcement, either in the form of fibres, or of continuous bars was found to reduce the crack velocity. A model was proposed based on a time step integration technique to evaluate the response of a beam subjected to an external impact pulse. The model was capable of predicting not only the experimentally observed non-linear behaviour of concrete under impact loading, but also the more pronounced brittle behaviour of high strength concrete. / Applied Science, Faculty of / Civil Engineering, Department of / Graduate

It is well known that transportation means high and varying loads for products as well as packages. To develop corrugated boxes with optimal design and efficient use of raw materials is crucial. Vibrations and shocks acting on pallets during transportation are transferred to the corrugated boxes and considerably reduce the integrity and life time of the boxes. The development of experimental and analytical tools for measurement and prediction of the influence of dynamic loads on the box performance, such as stacking strength and conservation of stacking pattern would therefore be of large practical importance. In order to develop such tools, it is important to know the load distribution between different boxes. This master thesis presents a technique for investigating these stresses based on a pressure sensitive film, which gives many data points. A series of tests using random and sinusoidal vibration testing have been done utilising this technique and results are presented for different positions on the pallet and for different box filling methods. Investigations performed on the vibrations of the boxes also demonstrate a pitch type of motion. A level-crossing study on the forces existing between the boxes shows a Rayleigh force distribution. A mathematical model is also proposed for simulation of a stacking system. Advantages and disadvantages with this technique and with the model are described. Comparison between the experimental and numerical results shows a proper correlation. Using the pressure sensitive film as a quantitative sensor and applying the recorded data for the statistical study of the contact forces existing in a stack of boxes gives useful and important results for further analysis of the fatigue life and vulnerable positions of boxes.

A thesis submitted to the Faculty of Engineering, University of the Witwatersrand. Johannesburg, in fulfilment of the requirements for the degree of Doctor of Philosophy / Hydraulically placed cemented backfill is increasingly being used as a means of stope support in South African hard-rock mines. The addition of binder provides backfill with a primary compressive strength. the property for which the material has traditionally been utilised. Binder-induced cohesion, however, is not the only factor determining the quality of cemented backfill and the material's utility can be enhanced significantly by applying all of its attributes to the task of stope support. The main purpose of this thesis is, therefore, to examine those factors which influence the performance of cemented backfill, and to provide relevant and useful information for the design of improved cemented backfill mine support for tabular mining excavations. In an extensive laboratory investigation, a number of relevant factors were tested for their influence on the loading behaviour of cemented backfill. Twenty two cemented backfills. grouped according to aggregate type, water content, binder content and binder type, were analysed to determine the influence of composition on their material properties. This study was augmented by several test series on the effects of curing conditions on cemented backfill quality. In a second major investigation, the effects of geometrical parameters, including sample volume and sample width to height ratios were analysed. A further study dealt with the effect of spatially separating sample backfill ribs under normal compression, as well as at high closure rates. It is concluded, that by co-ordinating the composition design of cemented backfills with the spatial configuration of backfill support elements, it is possible to modify all phases of the cemented backfill loading response. This implies the control of the binder-induced compressive strength at low strains, the large-scale yielding behaviour of backfill support, as well as the stiffness of the backfill body at high stresses. In the light of stope support requirements. particularly under rockburst conditions, the capacity of backfill support to yield and absorb rapid stope closure and then to decelerate the hangingwall by the rapid strain-hardening of the, now, large width to height ratio backfill mass, has the potential of substantially increasing mine safety in tabular stopes, / Andrew Chakane 2020

Optimization of dynamic systems involving complex non-hermitian matrices is often computationally expensive. Major contributors to the computational expense are the sensitivity analysis and reanalysis of a modified design. The present work seeks to alleviate this computational burden by identifying efficient sensitivity analysis and approximate reanalysis methods. For the algebraic eigenvalue problem involving non-hermitian matrices, algorithms for sensitivity analysis and approximate reanalysis are classified, compared and evaluated for efficiency and accuracy. Proper eigenvector normalization is discussed. An improved method for calculating derivatives of eigenvectors is proposed based on a more rational normalization condition and taking advantage of matrix sparsity. Important numerical aspects of this method are also discussed. To alleviate the problem of reanalysis, various approximation methods for eigenvalues are proposed and evaluated. Linear and quadratic approximations are based directly on the Taylor series. Several approximation methods are developed based on the generalized Rayleigh quotient for the eigenvalue problem. Approximation methods based on trace theorem give high accuracy without needing any derivatives. Operation counts for the computation of the approximations are given. General recommendations are made for the selection of appropriate approximation technique as a function of the matrix size, number of design variables, number of eigenvalues of interest and the number of design points at which approximation is sought. / Ph. D.

This dissertation is a study of vertical penetration of right-cylindrical projectiles into dry granular soil (Ottawa sand). Rigid projectiles were impacted into the soil at velocities less than 1000 ft/sec. Past theories for predicting penetration in soils have almost exclusively been based upon the impact velocity and the corresponding maximum penetration. In this research, the motion of the projectile during the penetration process was obtained by utilizing microwaves. The motion characteristics were obtained by "bouncing" microwave signals from the bottom of the projectile as it moved downward, and recording the information on an Ampex CP 100 magnetic tape recorder. It was found that penetration of the projectiles into the granular media could successfully be predicted by expressing the resisting forces on the projectile by -M dv/dt = Av² + Cz + D v > v_c and by -M dv/dt = Bv + Cz + D 0 < v < v_c where M = mass of projectile V = velocity of projectile at time t z = depth of penetration at time t A,B,C,D = constant coefficients. v_c is a critical velocity at which the "flow'' of the soil changes from one regime to the other, and is believed to be related to the energy necessary for comminution of the sand particles. The static coefficients, C and D can be found from the strength properties of the soil (c, φ and y) or from a static penetration test. A linear relationship was found between the static resistance, C , and the dynamic resistance coefficient, B. The value of A is believed to be related to the comminution energy, and is not greatly affected by density changes in the sand. The stability of the projectile was found to be an important factor in the values of the penetration forces. / Ph. D.

An important problem confronting the designer of machines and structures which may be subjected to suddenly applied loads is the determination of the actual forces which the members must withstand. This investigation has been confined to a study of impact forces on beams in simple flexure. Margerum (American Society of Testing Materials, 1922) suggested a method whereby the maximum load might be found by obtaining indentations on a calibration bar attached to the moving head of the impact machine, and suggested using this in conjunction with an autographic stress strain diagram. Tests have been made upon both notched and unnotched bars, but the results were not satisfactory. It has been proposed that an accelerometer be designed with which the acceleration could be measured directly. After the acceleration is known the force producing the acceleration can be calculated. The acceleration of a point can be determined by a double differentiation of a space-time diagram. This method is being employed in this thesis. / M.S.

This paper presents the results of testing to determine the appropriateness of the seismic load duration factor and to investigate the possible effect of previous load history from cyclic loading on connection reserve capacity and ductility. The single shear nail and bolt connection types tested represent common connection geometries used in wood construction in the United States. The results of two methods of fully reversing cyclic loading of connections are presented. The first method was a load-controlled test with the applied cyclic loads acting at specified percentages above current nominal design values. The sets of specimens were then ramped to failure and the results were compared to a monotonic control set of specimens to determine if any reduction in connection capacity or ductility had occurred as a result of the cyclic loading. From the load-controlled cyclic testing, it was found that previous cyclic loading at load levels as high as twice current nominal design loads did not adversely affect connection capacity. The second cyclic loading method was a displacement-controlled test that involved successive phases at increasing displacement levels. Each phase consisted of a peak displacement, followed by a series of three decay cycles, then by a series of three cycles at the original peak displacement, the third of which is used to determine the stabilized system. The process is repeated at increasing incremental levels of displacement. The phased displacement stabilized load-displacement curve was fit to an equivalent energy elastic-plastic system for determination of connection parameters. Results from the two cyclic test methods, and from monotonic testing, indicate that the current load duration factor for wind and seismic loading is justified. / Master of Science / incomplete_metadata

A damage identification method incorporating the use of inductive learning is presented. Inductive learning is the process of learning from examples. The method utilizes as much dynamic-response data as is available, ordering this information to find the best data with which to discriminate among a set of damage states available for dynamic testing. This method takes into account the inherent variabilities in the damage identification problem. These inherent variabilities include but are not restricted to sensor noise, changes in environmental conditions, slight changes in boundary conditions, and manufacturing differences. The method statistically isolates changes in the dynamic-response characteristics due to damage from these inherent variances. This method is model-independent and can be used to accommodate any sensors, actuators, and data type. In order to demonstrate the method, an experiment was performed on a 12” x 12” x ⅛” aluminum plate hung horizontally from the corners to simulate free-free boundary conditions. The plate was sensed and actuated by two piezoelectric patches mounted diagonally symmetric from one another. A small test mass (2% of the mass of the plate) was placed at four discrete locations, changing the physical properties of the structure. The structural impedance-responses were measured for all of the damage cases for both sensors. This information was processed by the damage identification algorithm to generate rules to which a small amount of data, extracted from a single set of structural impedance-response information, can be applied. The method was able to successfully discriminate all of the damage states from one another as well as to detect the existence of a change in physical properties due to a damaged state of which there was no prior knowledge. / Master of Science