Additive manufacturing (AM) has the ability to fabricate components of high geometric complexity that are difficult or near impossible to be produced by traditional manufacturing technologies. Selective laser melting (SLM) is a commonly used AM technology for metallic fabrications. SLM offers the opportunities to customize the characteristics of the as-build part produced, by adjusting the laser settings. However, high strength aluminum (Al) alloys presents an obstacle for SLM production due to the low alloying content, which increases the alloys’ probabilities to form cracks due to thermal stress induced by the SLM build process. The current study focuses on the study of surface roughness and dimensional accuracy of SLM fabrication of Al6061 and AlSi10Mg. Using design of experiment (DOE), wide ranges SLM process parameters were experimented with, and their individual effect along with their interactive effects on the fabricated parts’ quality were evaluated. The quality characteristics studied are: microstructures, microhardness, tensile strength (ultimate tensile strength, and yield strength), density, surface roughness, and dimensional accuracy. Regression models were created for each quality characteristics, and the combination of density, surface roughness, and dimensional accuracy results was used to create processing window for SLM that ensures the production of high-quality parts. The work aims to not only be used as-is, to help with the selection of SLM process parameters for Al6061 and AlSi10Mg that will reduce the post- processing time, but also to set a foundation for future development for numerical models that could better predict and describe the relations between SLM process parameters and the part’s fundamental qualities. / Thesis / Master of Applied Science (MASc)

The idea underlying the present study was to apply twisting in order to introduce different levels of transverse pressure. The modified structure affected both the bonding level and the evolution of the damage in the yarn. In order to isolate this effect in a broader context, additional parameters were included in the experiment design, namely effects of loading rate, specimen length and filament diameter (directly linked to the fineness of the yarn). These factors have been studied in various contexts by several authors. Some related studies on involved factors will be briefly reviewed.

Robot design and development has been the backbone of industrial automation and is in the forefront of accelerated development across all areas. Robot designers have been using simulations for reducing product development lead times. With growing demand for faster, precise and efficient robots, the requirements on computerized simulation for stress analysis has become stringent. While the product structure is mostly designed and developed from scratch, some components are sourced from suppliers, leaving a gap in the knowledge for modelling an entirety of a robot. This thesis applies a structured method to develop a grey-box model of the compact gears, which provides the robots its dexterity in a compact form factor. The method utilizes experimental strain measurements as a basis for building the model. Design of experiment is used as the guide for conducting FE analysis on robot links with unit load case, followed by scaling of stresses to actual load case. Simulated and measured stress plots are compared to conclude on optimum modelling approach. Further, the thesis proposes an alternative method for stress analysis of robot links by omitting the compact gear embodiment. While the method applies well on the robot links considered during the study, its validation across other links and robot architecture is yet to be performed. Finally, recommendations for implementation of proposed method and areas for expanding this thesis work are proposed. / Robotdesign har varit ryggraden inom industriell automation och är i framkanten av utvecklingen inom alla områden. Robotdesigners använder sig av simuleringar för att korta utvecklingstider. Med växande krav på snabbare, effektiva och noggranna robotar, har kraven på datorsimuleringar ökat. Medan huvuddelen av strukturen är utvecklad från grunden, är några komponenter köpta från leverantörer vilket skapar ett glapp i kunskapen för att kunna modellera en hel robot. Det här examensarbetet använder en strukturerad metod för att utveckla en "grey-box" modell av en kompaktväxel, vilken tillåter robotens mångsidighet i ett kompakt format. Metoden använder töjningsmätningar till grund för att bygga modellen. Experimentell design används som en guide för att utföra FE analyser på robotdelar med enhetslastfall, vilka skalas till verkliga lastfall. Simulerade och uppmätta spänningar jämförs för att optimera modelleringen. Det föreslås även en metod där kompaktväxlarna inte modelleras som solider. Metoden fungerar väl för robotarmar som ingick i denna studie, dock bör den verifieras på andra modeller och delar. Slutligen föreslås implementering av metoden som tagits fram i denna rapport samt ytterligare arbete för att verifiera metoden.

This thesis focuses on the use of a design of experiment approach to examine the significance of process factors and interactions on the fabrication of micro- textured surfaces. The micro-textured surfaces examined contain pillar and hole features ranging from 80 – 2 micrometers in diameter. The processes examined are the deep reactive ion etching of silicon wafers for the production of silicon mould inserts and the micro-injection moulding of polypropylene, high density polyethylene and 316LS stainless steel replicate samples of the silicon mould insert. During the deep reactive ion etching of the silicon wafers the design of experiment approach was used to determine the significant of platen power, C4F8 gas flow and switching times to the presence of pillar undercut of 10 x 10, 5 x 5 and 2 x 2 micrometer pillars. Undercuts occur when the pillar base has a smaller cross-section than the apex of the pillar. Switching times was found to be the only statistically significant parameter for both 10 x 10 and 5 x 5 micrometer pillars. The design of experiment approach is used in the micro-injection moulding of polypropylene, high density polyethylene and 316LS stainless steel replicates to examine the significance of mould temperature, cooling time, holding pressure and injection speed on the part and buffer mass of the produce samples, the height and width of pillar on the replicate surfaces and the variation of the replicated pillars height and width from the original silicon mould insert. Examination of the high density polyethylene replicates found that mould temperature was the most significant factor regarding pillar dimensions (and variation from the silicon mould insert) across the range of pillar sizes. Upon examination of the polypropylene replicates it was found that the factor of most significance on pillar dimensions varied across the different pillar sizes. Holding pressure was identified as the most significant factor with regards to the 53 x 29 and 19 x 80 micrometer pillars. Injection speed was found to be most significant for the 25 x 25 and 19 x 29 micrometer pillars. Cooling time was found to be most significant with regards to the 30 x 10, 25 x 10, 20 x 10 and 15 x 10 micrometer pillars. While ii mould temperature was found to be most significant for the 20 x 20, 15 x 15 and 10 x 30 micrometer pillars. The interaction between mould temperature and injection speed was also found to be the most significant factor with regards to the 43 x 29 and 25 x 30 micrometer pillars. Examination of the 316LS replicates found that mould temperature was the most significant factor regarding pillar dimensions for 80 x 80 and 19 x 80 micrometer pillars. While holding pressure was found to be most significant to the 29 x 29 micrometer pillars and injection speed was identified as most significant to the 53 x 80 micrometer pillars. The samples produced during the design of experiment investigations were then used to examine the effect of surface texturing on droplet behaviour. Droplet contact angles were examined on polypropylene, high density polyethylene and silicon samples structured with 10 – 2 micrometer pillar. Initial droplet contact angles were found to be higher on the polypropylene samples than the high density polyethylene or silicon samples. With the lowest initial contact angles being found for the silicon inserts. Droplet ‘channelling’ and evaporation were examined on silicon, polypropylene, high density polyethylene and 316LS samples structured with micro-channel surface pillars and holes ranging from 80 – 2 micrometer in diameter. Contact pinning of the droplet to the surface via the three- phase contact-line was noted during observations of droplet ‘channelling’. This pinning effect was observed at all sample tilt angles (30 - 90 o ). With regards to droplet evaporation, the droplets were noted to evaporate evenly (with no or limited contact pinning) on all unstructured surfaces and the surfaces structured with hole features. On the surfaces structured with pillar features, the droplets appeared too evaporated along the surface gradient from the smallest pillars to the largest.

The continuous high demand on permanent magnets in industries opened new research plateau to develop alternative magnetic material. The current used permanent magnet materials in the market still suffer from high cost and insufficient magnetic or thermal properties. The central focus of this dissertation work is the optimization of cobalt carbide based nanomaterial by means of modifying polyol synthesis assisted by nucleation agent and systematic statistics using JMP software tool. In most existing literatures, producing cobalt carbide (Co2C or Co3C) lack reproducibility and consistency resulting in nonsolid magnetic properties results. The practical requirements for cobalt carbide to be used as permanent magnet are high coercivity (Hc), high magnetization (Ms), resulting in a high-energy product (HcxMs). Previous literatures have shown coercivities of 1.5 to 2.5 kOe for cobalt carbides under aggressive temperatures conditions (300oC) or after aligning the particles under magnetic field. A statistical guided method performed a sequence of experiments toward producing high coercivities using surface response design. Primarily, the statistical study to optimize cobalt carbide was made by analyzing experimental condition to fulfill high magnetic properties with tuned conditions as much as possible. Therefore, having the advantage for superior control on process variable when shifting cobalt carbide for scale up production in flow chemistry set up using microreaction system (MMRS). The optimization is based upon selecting the most important conditions in polyol reaction to produce cobalt carbide (Co2C or Co3C) and feed JMP software model e.g. reaction temperature, reaction time, and or precursor concentration…etc. These factors called (effects) used to design experiments and generate tables to run minimum experiments. Points of each effect (levels) are selected based on previous knowledge and experience with the synthesis. The output called (response) can be any of the magnetic properties of our interest e.g. magnetization (Ms), coercivity (Hc), or energy magnetic product (HcxMs). In the first model fit of cobalt carbide magnetic was studied in a polyol reaction to increase its magnetic energy product and optimize the experimental conditions. The results disclosed increase in magnetic energy product (6.2 MGOe) when validating the prediction model conditions suggested by JMP: shorter reaction time, and lower precursor concentration conditions at maximum reaction temperature. Finally, to my knowledge studying the effect of the nucleating agent to alter cobalt carbide growth have not been studied so far. Therefore, statistical study design using central composite design (CCD) to investigate the nucleating agent effect of silver nitrite on cobalt carbide coercivity was made. The importance of nucleating agent on coercivity is vigorous to attain and control the growth direction of cobalt carbide nanoparticles. This is due to the shape anisotropy contribution to enhance coercivity unlike weak shape anisotropy attributed to agglomeration of nanoparticles demonstrated in previous studies. Enhancement of coercivity reached 3 kOe with aspect ratio control as a function of silver nitrite concentration under lower reaction temperature. The continuous high demand on permanent magnets in industries opened new research plateau to develop alternative magnetic material. The current used permanent magnet materials in the market still suffer from high cost and insufficient magnetic or thermal properties. The central focus of this dissertation work is the optimization of cobalt carbide based nanomaterial by means of modifying polyol synthesis assisted by nucleation agent and systematic statistics using JMP software tool. In most existing literatures, producing cobalt carbide (Co2C or Co3C) lack reproducibility and consistency resulting in nonsolid magnetic properties results. The practical requirements for cobalt carbide to be used as permanent magnet are high coercivity (Hc), high magnetization (Ms), resulting in a high-energy product (HcxMs). Previous literatures have shown coercivities of 1.5 to 2.5 kOe for cobalt carbides under aggressive temperatures conditions (300oC) or after aligning the particles under magnetic field. A statistical guided method performed a sequence of experiments toward producing high coercivities using surface response design. Primarily, the statistical study to optimize cobalt carbide was made by analyzing experimental condition to fulfill high magnetic properties with tuned conditions as much as possible. Therefore, having the advantage for superior control on process variable when shifting cobalt carbide for scale up production in flow chemistry set up using microreaction system (MMRS). The optimization is based upon selecting the most important conditions in polyol reaction to produce cobalt carbide (Co2C or Co3C) and feed JMP software model e.g. reaction temperature, reaction time, and or precursor concentration…etc. These factors called (effects) used to design experiments and generate tables to run minimum experiments. Points of each effect (levels) are selected based on previous knowledge and experience with the synthesis. The output called (response) can be any of the magnetic properties of our interest e.g. magnetization (Ms), coercivity (Hc), or energy magnetic product (HcxMs). In the first model fit of cobalt carbide magnetic was studied in a polyol reaction to increase its magnetic energy product and optimize the experimental conditions. The results disclosed increase in magnetic energy product (6.2 MGOe) when validating the prediction model conditions suggested by JMP: shorter reaction time, and lower precursor concentration conditions at maximum reaction temperature. Finally, to my knowledge studying the effect of the nucleating agent to alter cobalt carbide growth have not been studied so far. Therefore, statistical study design using central composite design (CCD) to investigate the nucleating agent effect of silver nitrite on cobalt carbide coercivity was made. The importance of nucleating agent on coercivity is vigorous to attain and control the growth direction of cobalt carbide nanoparticles. This is due to the shape anisotropy contribution to enhance coercivity unlike weak shape anisotropy attributed to agglomeration of nanoparticles demonstrated in previous studies. Enhancement of coercivity reached 3 kOe with aspect ratio control as a function of silver nitrite concentration under lower reaction temperature.

The surface quality is identified by surface roughness parameters. The average surface roughness (Ra) is used in this study, as it is the most commonly used roughness parameter in the industry. A particular curved cavity of a forging die is selected for the experimental study. Different milling methods are tested. The comparison studies are conducted between 3-axes and 5-axes milling, linear and circular tool path strategies and down and up milling. According to the experimental study, appropriate method is determined for the milling of a particular curved cavity of a forging die. The experimental analysis based on design of experiments (DOE) has been employed by considering cutting speed, feed per tooth and stepover parameters. Multiple linear regression technique is used by which a mathematical formula has been developed to predict the Ra values for milling parameters. The results of the mathematical formula are controlled by conducting test experiments and good correlations are observed between the results of the formula and the results of test experiments.

Induction heating is one of the efficient and modern technique for heating raw materials for hot forging process. The induction heating furnaces use electro-magnetic field to transfer energy to the metal workpiece and heat is generated inside the material. The magnetic field can be provided by using induction coil. The power supplied to induction coil, the moving speed of the billet that is called conveyor speed and the coil box hole diameter are the factors affecting the resultant temperature of the heated billet. In this study, AISI 1045 type steel billets with a diameter of &Oslash / 30 mm and length of 100 mm have been heated in a particular induction heater. During heating, effects of different levels of power, conveyor speed and the coil boxes with different hole diameters are investigated. The 125 KW 3000 Hz induction heater which is available in METU-BILTIR Research and Application Center Forging Laboratory is used in experiments. The heating experiments are designed according to 23 Factorial Design of Experiment Method. Multiple linear regression technique is used to derive a mathematical formula to predict the temperature of the heated billet. A good correlation between the measured temperatures that are the results of different sets of induction heating parameters and the predicted temperatures that are calculated by using temperature prediction formula has been observed.

A bascule bridge is a type of movable bridge that can be opened or closed to facilitate the movement of water-borne traffic such as ships and yachts. Trunnion-Hub-Girder (THG) assembly plays a role as a fulcrum in the bascule bridge. To make the fulcrum, the trunnion is shrink-fitted into the hub, and then the trunnion-hub assembly is shrink-fitted into the girder. Hundreds of thousands of dollars could be lost due to failures during this step. Crack formations in the hubs of various Florida bascule bridges during assembly led the Florida Department of Transportation to commission a project with USF professors to investigate.Finite elements method (ANSYS package) is employed to model the THG assembly procedure and solve for the critical crack length and critical stress in this transient thermal structural problem. Design of experiments (DOE) is used with different cooling processes and the geometrical dimensions of the THG assembly to find the sensitivity of these parameters on the outputs.The influence of the hub outer diameter and the radial interference (between the trunnion and hub) is at different levels on the critical crack length and the stress ratio as it is dependent on fulcrum geometry. If we include four staged cooling methods as follow Type 1: liquid nitrogen Type 2: dry-ice/ alcohol bath followed by liquid nitrogen Type 3: refrigerated air chamber followed by liquid nitrogen Type 4: refrigerated air chamber followed by dry-ice/alcohol bath and then by liquid nitrogenthe cooling type factor contributes the most to both critical crack length (up to 79%) and the stress ratio (up to 84%) in the TH assembling procedures in all three considered bascule bridges. The staged cooling procedure type 2, which is, immersing the TH assembly into the dry ice medium, followed by immersing the TH set into the liquid nitrogen, give larger critical crack length (up to 400%) and stress ratio (up to 87%) compared to the case that used only liquid nitrogen.

abstract: In mixture-process variable experiments, it is common that the number of runs is greater than in mixture-only or process-variable experiments. These experiments have to estimate the parameters from the mixture components, process variables, and interactions of both variables. In some of these experiments there are variables that are hard to change or cannot be controlled under normal operating conditions. These situations often prohibit a complete randomization for the experimental runs due to practical and economical considerations. Furthermore, the process variables can be categorized into two types: variables that are controllable and directly affect the response, and variables that are uncontrollable and primarily affect the variability of the response. These uncontrollable variables are called noise factors and assumed controllable in a laboratory environment for the purpose of conducting experiments. The model containing both noise variables and control factors can be used to determine factor settings for the control factor that makes the response "robust" to the variability transmitted from the noise factors. These types of experiments can be analyzed in a model for the mean response and a model for the slope of the response within a split-plot structure. When considering the experimental designs, low prediction variances for the mean and slope model are desirable. The methods for the mixture-process variable designs with noise variables considering a restricted randomization are demonstrated and some mixture-process variable designs that are robust to the coefficients of interaction with noise variables are evaluated using fraction design space plots with the respect to the prediction variance properties. Finally, the G-optimal design that minimizes the maximum prediction variance over the entire design region is created using a genetic algorithm. / Dissertation/Thesis / Ph.D. Industrial Engineering 2010

abstract: This work aimed to characterize and optimize the variables that influence the Gas Diffusion Layer (GDL) preparation using design of experiment (DOE) approach. In the process of GDL preparation, the quantity of carbon support and Teflon were found to have significant influence on the Proton Exchange Membrane Fuel Cell (PEMFC). Characterization methods like surface roughness, wetting characteristics, microstructure surface morphology, pore size distribution, thermal conductivity of GDLs were examined using laser interferometer, Goniometer, SEM, porosimetry and thermal conductivity analyzer respectively. The GDLs were evaluated in single cell PEMFC under various operating conditions of temperature and relative humidity (RH) using air as oxidant. Electrodes were prepared with different PUREBLACK® and poly-tetrafluoroethylene (PTFE) content in the diffusion layer and maintaining catalytic layer with a Pt-loading (0.4 mg cm-2). In the study, a 73.16 wt.% level of PB and 34 wt.% level of PTFE was the optimal compositions for GDL at 70 °C for 70% RH under air atmosphere. For most electrochemical processes the oxygen reduction is very vita reaction. Pt loading in the electrocatalyst contributes towards the total cost of electrochemical devices. Reducing the Pt loading in electrocatalysts with high efficiency is important for the development of fuel cell technologies. To this end, this thesis work reports the approach to lower down the Pt loading in electrocatalyst based on N-doped carbon nanotubes derived from Zeolitic Imidazolate Frameworks (ZIF-67) for oxygen reduction. This electrocatalyst perform with higher electrocatalytic activity and stability for oxygen reduction in fuel cell testing. The electrochemical properties are mainly due to the synergistic effect from N-doped carbon nanotubes derived from ZIF and Pt loading. The strategy with low Pt loading forecasts in emerging highly active and less expensive electrocatalysts in electrochemical energy devices. This thesis focuses on: (i) methods to obtain greater power density by optimizing content of wet-proofing agent (PTFE) and fine-grained, hydrophobic, microporous layer (MPL); (ii) modeling full factorial analysis of PEMFC for evaluation with experimental results and predicting further improvements in performance; (iii) methods to obtain high levels of performance with low Pt loading electrodes based on N-doped carbon nanotubes derived from ZIF-67 and Pt. / Dissertation/Thesis / Masters Thesis Mechanical Engineering 2016