<p>Small scale Vertical Axis Wind Turbines (VAWTs) show potential for urban rooftop installations where they can capture the highly unstable, turbulent wind flow patterns which are typical in an urban environment. Being axisymmetric, they are omnidirectional turbines which respond well to changes in wind direction, unlike their horizontal axis counterparts. Wind tunnel experiments were performed on a small scale, 2.5m diameter by 3m high, 3 bladed H-Darrieus VAWT. The effects of blade preset pitch and blade mount-point offset were investigated. Local flow visualizations were observed on the inner surface of a blade near the lower strut by recording video of the behaviour of Mylar tufts during turbine operation for comparison with a 2-D Computational Fluid Dynamic (CFD) model. Experimental data is presented for a range of tip speed ratios, at a nominal wind velocity of 10m/s for both blade pitch and blade offset tests.</p> <p>The results of the blade preset pitch tests indicate performance increases with increasing toe-out pitch, to a limit, followed by a sharp decrease in performance as toe-out is continually increased. Performance decreases were experienced for all toe-in pitch presets. Identical behaviour was seen with experiments involving blade mount-point offsets which were geometrically equivalent to blade pitch presets. Results of tuft flow visualization tests indicate the regions during which the blade is experiencing reversed flow, and compare well with velocity vector data obtained from CFD simulations. The duration the inner blade surface experiences full stall is observed to decrease with the increase of toe-out preset pitch. Incipient stall is observed to be delayed as blade tip speed ratio is increased. Thrust curves are also presented and indicate a decrease in upwind pass and increased downwind pass perfonnance as toe-out is increased. As the tip speed ratio is increased, the interaction of the blade with its shed vortex from the upwind pass is diminished. This investigation shows that high solidity VAWTs are sensitive to relatively small changes in blade preset pitch and mount point offset and therefore care should be taken when designing turbine blades so as to avoid a toe-in condition that could adversely affect turbine performance.</p> / Master of Applied Science (MASc)

<p>To investigate the effects of mechanical loading on the rat and mouse forelimb, male Sprague Dawley rats and male C57/BL-6 mice forelimbs were cyclically loaded in axial compression. Two studies were performed, one to examine woven bone formation in both the rat and mouse forelimb, and one to examine lamellar bone formation in the rat forelimb. Both types of bone formation were induced through an acute loading event.</p> <p>Repeatable levels of fatigue damage were induced in both the rat and mouse ulna, yielding a consistent woven bone response. Rats formed more woven bone than mice. Woven bone formed by the rats was evenly distributed along the medial cortex versus woven bone formed by the mice was more abundant towards the posterior side of the medial surface.</p> <p>150 cycles of rest inserted loading to a peak load of 30 N was found to induce lamellar bone formation with the greatest apposition rate. Apposition rate was more strongly influenced by peak loading rather than number of loading cycles. At the 30 N loading level, 300 loading cycles induced woven bone formation in two of the five animals.</p> <p>Finite element models were created to examine the mechanical environment of both the rat and mouse ulna during compressive forelimb loading. Stress distribution in the rat ulna was found to be distributed along the medial surface of the rat ulna. Stress distribution was biased to the posterior side of the medial surface on the mouse ulna. These stress distributions supported the bone formation observed. Maximal stress occurred in both the rat and mouse ulna at the same location failure occurred during monotonic and fatigue to failure tests conducted on both rat and mouse ulnas.</p> / Master of Applied Science (MASc)

<p>Making large RP parts is always a challenge for common commercial RP machines or other CNC machining process. Either it is not possible to fabricate the large RP models, or can only be done at high cost.</p> <p>This thesis investigates a new method to make large RP parts in a very rapid and cost-effective way, and the part size could be theoretically unlimited. A software program for slicing 3D CAD models is developed in the Autodesk Inventor VBA environment to generate DXF files, and a common 3-axis CNC router is employed for cutting the sliced layers from sheet material by importing the DXF files from the slicing software tool. The sheet material sliced layers are laminated together to form the RP models with staircases on the laminated body surface, then epoxy resin filler is applied to cover the staircases. The surface finishing process such as sanding and coating makes the RP model closely match the original designed part shape.</p> <p>By directly slicing a 3D CAD model in Inventor with the slicing software tool, the cutting profiles are obtained and saved in the DXF file format. The errors usually caused by slicing STL files with most conventional RP machines are eliminated and hence the geometry data is more accurate for making sliced layers. The slicing process becomes quite simple with the help of the slicing software tool, and it only takes a few minutes to get the DXF files for CNC router cutting.</p> <p>Three different slicing and CNC router cutting methods are discussed in this thesis. For different shapes of 3D CAD models, an appropriate slicing method can be applied to reach the optimal results of RP model making.</p> <p>The uniform slicing method is an effective way to slice large parts and generate sliced layers with the same thickness; the adaptive slicing method is a supplement for uniform slicing and makes thinner layers to save time for post-finishing such as staircase filling and surface grinding; and the adaptive slope cutting method is developed for choosing different sloped cutters for CNC router to cut sliced layers with standard grooving router bits.</p> <p>The experiment is performed to make RP models of a sailboat keel pattern with the different slicing methods. Measurement is done for the fabricated RP models for comparing the surface deviations between the original designed 3D CAD models and the fabricated RP parts. The deviation plot made by software Geomagic shows that the RP model closely matches the original 3D CAD model, the surface deviation is mostly within lmm between the RP model and CAD model in the research experiment. The fabrication cost for RP model is also analyzed based on the experiment models in this research.<br />iv</p> / Master of Applied Science (MASc)

<p>Thermal stratification in solar energy storage tanks plays an important role in enhancing the performance of solar domestic hot water (SDHW) system. The mixing that occurs when hot fluid from the solar collector enters the top of the tank is detrimental to the stratification. Mathematical models that are used for system analysis must thus be able to capture the effects of this inlet jet mixing in order to accurately predict system performance. This thesis presents a Computational Fluid Dynamics (CFD) study of the heat transfer and fluid flow in the thermal storage tank (TST) of a solar domestic hot water system employing a vertical inlet jet geometry. The focus of the thesis is on the studying the effects of inlet jet mixing on the thermal stratification in the tarue Predictions of transient temperature profiles were assessed by comparing to experimental data from the literature. CFD was then used to study how the predicted mixing in the TST was affected by parameters such as the inlet velocity and temperature, pipe diameter and the selected turbulence model. From this study of the mixing, a one dimensional empirical model was developed to predict the temperature distribution inside the TST. The model was found to provide improved predictions of the transient axial temperature distribution in comparison to the plug-flow model which is commonly used in the broader system analysis codes.</p> / Master of Applied Science (MASc)

<p>Reproduction of hand crafted 3D shape geometry is an important industrial concept design step employed for items ranging in size and function. Adding surface art work can extend concept design to include essential product information. This thesis describes an equipment and workflow implementation for reproducing both 3D shape and art work. The hand crafted sample shape is laser digitized to obtain the XYZ geometry coordinates simultaneously with the surface gray level. The resulting point cloud is processed and converted to an STL format for plastic rapid prototyping, while the surface gray level information is treated to fit the smoothed surface, and match the reproduction ink jet printing system resolution and specification. Developed software automatically calculates the dot pattern to reproduce the art work onto the correct prototyped surface location.</p> / Master of Applied Science (MASc)

<p>p.p1 {margin: 0.0px 0.0px 0.0px 0.0px; font: 11.0px Times; color: #2d2d2d} span.s1 {color: #4c4c4c} span.s2 {font: 11.0px Helvetica}</p> <p>p.p1 {margin: 0.0px 0.0px 0.0px 0.0px; font: 11.0px Times} span.s1 {font: 11.0px Helvetica} span.s2 {font: 10.0px Verdana} span.Apple-tab-span {white-space:pre}</p> <p>The aim of this research is to enhance performance and durability of the final PM component by improving spatial density homogeneity at die filling stage of powder metallurgy process. In this research, powder die filling processes has been studied using a novel laboratory experimental set-up, with a ring-shaped die, for assessing die fill characteristics of thin, high precision components of interest to Gates Canada.</p> <p>In this investigation, usefulness of incorporating a perforated plate at the bottom of the feed shoe to improve the density uniformity in the filling condition is assessed. With this arrangement, flow and distribution of powder during the delivery stage is monitored. The powder flow pattern observed through the transparent window is utilized for obtaining full-field displacement data using an optical measurement technique. In addition to the above, several powder flow characteristics during die filling have been studied through a series of high-speed camera recordings. The role of shoe speed, and powder properties in the development of density gradient have been experimentally assessed. A series of full-scale experiments with coloured salt as a powder medium to mimic the iron powder flow have been conducted to understand flow patterns, and segregation of powder during the filling process.</p> <p>Further, a novel method of density measurement of the part in die-fill condition by heating iron powder with a polymeric resin is explored. Qualitative density data obtained from the die filling experiments, and subsequent porosity data from powder sintering in the die have been compared, to understand the reasons for density gradient in the die-cavity.</p> <p><strong></strong></p> / Master of Applied Science (MASc)

<p>Pool boiling of nanofluids has been studied in the past decade and contradicting results have been found. Several parameters have been investigated, the most popular being the nanofluid concentration. The investigations in the literature have been carried out under different conditions. The effects of method of nanofluid preparation, pH value and boiling duration on the Heat Transfer Coefficient (HTC) and nanoparticle deposition have not been thoroughly investigated.<br /> An experimental investigation has been carried out to investigate such effects and the effect of nanofluids concentration on the HTC and nanoparticle deposition. A flat copper surface with a mirror-finish (Ra = 50 - 150 nm) has bee n used as the boiling surface. Using a nanofluid prepared from a ready-made suspension showed to have a similar HTC to a nanofluid prepared from dry nanoparticles at a neutral pH value. Reducing the pH value of the nanofluids prepared from dry particles gave different responses in HTC at different concentrations. At 0.01 vol. % concentration the HTC decreased due to change in base fluid properties, at 0.1 vol. % the HTC increased due to enhanced electrostatic stabilisation, and no change took place at 0.5 vol. % due to a high intensity of nanoparticle deposition. Increasing the concentration of the nanofluids resulted in a reduction in HTC, higher nanoparticle deposition as well as a faster rate of deposition on the surface.<br /> The effect of concentration on nanoparticle deposition has been studied by boiling water on nanoparticle deposited surfaces after boiling nanofluids. This approach has shown that nanoparticle deposition from boiling a higher concentration nanofluid gives a higher HTC, which is opposite to expectations. This approach has also shown that the surface condition depends on the intensity of the deposition as well as its uniformity.<br /> An analysis has been carried out using the Rohsenow correlation to quantify the effect of nanoparticle deposition on the heat transfer. Prediction of the boiling heat transfer of nanofluids may be used by adopting a transient surface factor in the Rohsenow correlation.</p> / Master of Applied Science (MASc)

<p>p.p1 {margin: 0.0px 0.0px 0.0px 0.0px; font: 11.5px Times; color: #292929} span.s1 {color: #3e3e3e} span.s2 {color: #575757}</p> <p>The significance of online process monitoring of discrete part manufacturing usmg multivariate analysis is its ability to help the Canadian manufacturing industry compete in the global market. Process monitoring can accomplish this by: assessing the state of a machining system for unusual occurrences, moving the part quality prediction upstream, and producing higher volumes of in specification parts for improved profits.</p> <p>The focus of this research was discrete process monitoring of a turning operation in a laboratory at the McMaster Manufacturing Research Institute (MMRI) and an industrial machining cell at Glueckler Metal Incorporated (GMI). Both applications involved instrumentation of a lathe with current sensors, an accelerometer and thermocouples. Serial port communication between the machine control panel and computer was established to allow for online automated data acquisition. The multivariate latent model applied was principal component analysis to develop correlations among the machining process information. Principal component analysis was successful in identifying the occurrence of an out of balance spindle, unusual surface finish, changes in depth of cut, and a worn tool in laboratory tests, through the use of simple control plots. Industrial results validated the ability of the system to differentiate machining data from one day to another, and to isolate an unusual piece of barstock that led to slightly below average part dimensions.</p> <p>The difficulties experienced in the transitioning from laboratory conditions to industrial testing were discussed. This information will allow future researchers to continue adding new process monitoring sensors to the system.</p> <p>In conclusion, this research demonstrated the ability of online process monitoring of discrete part manufacturing in a laboratory setting; and brings the MMRI and GMI closer to having a fully implemented process monitoring system for part quality prediction and machine maintenance.</p> / Master of Applied Science (MASc)

<p>p.p1 {margin: 0.0px 0.0px 0.0px 0.0px; font: 12.0px Times} span.s1 {font: 11.5px Helvetica}</p> <p>Using an integrated Tuned Liquid Damper (TLD)-Structure in-house developed numerical algorithm that has been validated against recent rigorous experimental tests, the TLD performance was analyzed when coupled with a vibrating Single Degree of Freedom (SDOF) body representing a civil structure. The numerical algorithm solves the full two dimensional Navier-Stokes equations with no linearization assumptions. It uses the Volume of Fluid method to reconstruct the free surface, and the Partial Cell Treatment method to model the effect of any obstructions. This study investigated the structure response when coupled to a TLD with and without a screen under harmonic excitations. Structure sway was found to decrease by 71 % in the case of a TLD without a screen, and 80% in the case of a TLD with one screen. The best screen configuration was then determined for the TLD-Structure coupling under non-hmmonic excitations, taking minimal structure sway and acceleration as the deciding criteria. Eighteen different cases considering different screen locations and solidities were investigated, and the case with one screen placed in the middle with a solidity of 0.4 proved to be the best.</p> <p>The study also investigated the effect of fluid height on structure sway under a p.p1 {margin: 0.0px 0.0px 0.0px 0.0px; font: 12.0px Times}</p> <p>wide range of excitation amplitudes. Harmonic Excitations with amplitudes up to 3% of the tank length and fluid heights up to 40% of tank length were considered. The results showed better structure response with lower fluid heights in the case of low to moderate excitation amplitudes. With high excitation amplitudes, the results confirmed an opposite trend where higher fluid heights resulted in better structure response.</p> <p>The numerical code was then modified to model a Sloped Bottom (SB) TLD using the Partial Cell Treatment method. The numerical model for the SB TLD has been validated against experimental data to ensure accuracy. Numerous cases have been considered to investigate structure response under the new configuration, and to analyze how a SB TLD compares to a standard TLD. The results showed an increased damping ratio and better structure response for SB TLDs, and a significant softening spring behaviour that is important upon excitation cessation.</p> / Master of Applied Science (MASc)

<p>There is currently a lack of quality sensing techniques that can provide the required spatial and temporal resolution for use in microfluidic devices. The development of such micro sensors will allow real time monitoring and control of many processes at the micro level, and play a crucial role in expanding microfluidics to novel applications. For example, integration of sensors within the microfluidic device itself will allow active control of processes within these devices. The overall objective of this study was to develop a micro temperature and micro flow sensor for use in microfluidic devices. The specific objectives were to develop, design and micro fabricate a micro thermocouple and micro heater, and integrate these within a microchannel to show proof of concept of a micro thermal pulse flow sensor. A platinum-constantan (PT-NiCu) micro thermocouple was developed and fabricated using a three mask process. The micro fabrication protocols and procedures were developed for potentiostatically electroplating the constantan leg of the micro thermocouple. The thermocouples were characterized and the Seebeck coefficient (sensitivity) was found to be 39.04 μV/゜C and 41.75 μV/゜C for non compensated and a compensated thermocouple arrangement respectively.</p> <p>A meandering resistive type micro heater was developed. The power consumption for the 400 Å thick gold micro heaters on the silicon oxide and on the glass substrates was compared. The power required for the glass substrate was 46mW, 112mW and l60mW for 5V, 8V, 10V respectively, while for the silicon oxide was 499.5 mW, 1.27 W and 1.943 W respectively.</p> <p>The thermal flow sensor was developed by integrating the micro heater and micro thermocouple within a microchannel to show proof of concept of the sensor. The flow sensor was operated in three modes; time of flight, temperat1ll'e difference and pulsed thermotransfer calibration mode. Essentially the thetmotransfer principle occurs as the heat loss from the micro heater source to the fluid will increase with the flow rate, thereby giving greater voltage amplitude of the thetmocouple response with increasing flow velocities.</p> <p>The flow sensor performance was characterized using methanol/water as the working fluid for mass flow rate in the range of no flow to 0.7 ml/min. The device has several unique operating and physical characteristics, including the novel pulsing scheme developed that compensates against temperature drift, resulting in high repeatability.</p> <p>The flow sensor was calibrated using the thermotransfer principle for three pulse modes; single, multiple pulses with change in input voltage and multiple pulses with change in pulse duration. The comparative results showed that the multiple pulse modes generated a more detectable signal than the single pulse mode. The multiple pulse regimes allowed for a larger dynamic flow range. The flow sensor can be duplicated relatively easily so that multiple sensors can be distributed within a microfluidic device to allow simultaneous flow measurements at different locations within the device.</p> / Master of Applied Science (MASc)