The development of magnetic technologies employing microfabricated magnetic structures for the production of integrated electronic components is a driving topic in the electronic industry. Despite the large amount of work reported in the literature towards the production of magnetic devices that can be integrated into conventional silicon technology, the published research has only achieved moderate success. The research presented in this thesis was conducted with the aim to progress towards the production of a magnetic MEMS relay based on electro-deposited NiFe that combines magnetic and electrostatic actuation and that can be integrated in a standard IC processing chain. This work includes a comprehensive design study for the proposed MEMS device and presents the development of the manufacturing processes required for its fabrication. As the theoretical performance of the device is found to be crucially reliant on the mechanical and magnetic properties of the microformed structures, a series of novel test methodologies has been devised and implemented with the aim of acquiring knowledge on the behaviour of the NiFe films. Novel mechanical test routines employing microfabricated test structures are presented and applied to build a systematic and robust system for the characterisation of the electrodeposited films. The quantitative mapping of residual stress at the wafer level using microfabricated test structures has been demonstrated for the first time and applied to optimise processes and tools. A complete fabrication process flow for manufacturing the designed magnetic MEMS switch has been proposed and the fabrication of the actuated section of the switch has been demonstrated, comprising all the functional electric and magnetic components. The fabricated magnetic devices have been tested to monitor their response to an external magnetic force and prove their viability for use in MEMS actuators. Additional work was finally conducted towards the development of a reliable and robust process aimed at increasing the device yield and thus facilitating the eventual commercialisation of magnetic MEMS switches.

Chan, Ngai Shing. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2007. / Includes bibliographical references (leaves 67-70). / Abstracts in English and Chinese. / Abstract --- p.I / 摘要 --- p.III / List of Figures --- p.VI / List of Tables --- p.VIII / Chapter 1 --- Introduction --- p.1 / Chapter 1.1 --- Background Information --- p.2 / Chapter 1.1.1 --- Project Background --- p.2 / Chapter 1.1.2 --- Literature Review --- p.4 / Chapter 1.1.3 --- Background on Gear Hobbing --- p.10 / Chapter 1.1.4 --- Traditional gear hobbing machines --- p.12 / Chapter 2 --- Design and Testing of the MMT system --- p.15 / Chapter 2.1 --- Specifications of the MMT system --- p.16 / Chapter 2.1.1 --- Overall Configuration --- p.18 / Chapter 2.1.2 --- Linear Actuation --- p.18 / Chapter 2.1.3 --- Main Spindle Assembly --- p.19 / Chapter 2.1.4 --- Tool Plate Assembly --- p.20 / Chapter 2.1.5 --- Motion Control --- p.22 / Chapter 2.2 --- Main Features --- p.24 / Chapter 2.2.1 --- Mechanically Decoupled Gear Hobbing --- p.24 / Chapter 2.2.2 --- Single Setup for Non-planar Gears --- p.26 / Chapter 2.2.3 --- Quality Assurance by Computer Simulation --- p.27 / Chapter 2.3 --- Turning Test --- p.28 / Chapter 2.3.1 --- Experiment Results --- p.29 / Chapter 2.3.2 --- Tornos' Performance --- p.30 / Chapter 2.3.3 --- Estimation of Cutting Force and Workpiece Deflection --- p.32 / Chapter 2.4 --- Synchronization Test --- p.33 / Chapter 2.4.1 --- Experimental Results --- p.34 / Chapter 2.5 --- Gear Hobbing Test --- p.36 / Chapter 3 --- Diagnostic Tool: Gear Hobbing Simulation --- p.40 / Chapter 3.1 --- Simulation Model --- p.41 / Chapter 3.2 --- Simulations with Process Defects --- p.44 / Chapter 3.2.1 --- Asynchronous motion between tool and workpiece spindle --- p.44 / Chapter 3.2.2 --- Pitch error of the cutter hob --- p.45 / Chapter 3.2.3 --- Tool spindle run-out error --- p.47 / Chapter 3.2.4 --- Combination of process defects --- p.49 / Chapter 3.3 --- Experiment Validation --- p.50 / Chapter 4 --- Technical know-hows --- p.55 / Chapter 4.1 --- Premature Part Break-off --- p.55 / Chapter 4.2 --- Tool Alignment and Centering --- p.58 / Chapter 4.2.1 --- Two-turns Aligning Algorithm --- p.59 / Chapter 5 --- Conclusion and Future Work --- p.63 / References --- p.67 / Publication Record --- p.71 / Appendix --- p.72

This thesis presents a) a scheme for selecting the most suitable flexure hinge compliance equations, and b) a simple methodology of deriving kinetostatic models of micro-motion stages by incorporating the scheme mentioned above. There were various flexure hinge equations previously derived using different methods to predict the compliances of circular flexure hinges. However, some of the analytical/empirical compliance equations provide better accuracies than others depending on the t/R ratios of circular flexure hinges. Flexure hinge compliance equations derived previously using any particular method may not be accurate for a large range of t/R ratios. There was no proper scheme developed on how to select the most suitable and accurate hinge equation from the previously derived formulations. Therefore, the accuracies and limitations of the previously derived compliance equations of circular flexure hinges were investigated, and a scheme to guide designers for selecting the most suitable hinge equation based on the t/R ratios of circular flexure hinges is presented in this thesis. This thesis also presents the derivation of kinetostatic models of planar micromotion stages. Kinetostatic models allow the fulfillment of both the kinematics and the statics design criteria of micro-motion stages. A precise kinetostatic model of compliant micro-motion stages will benefit researchers in at least the design and optimisation phases where a good estimation of kinematics, workspace or stiffness of micro-motion stages could be realised. The kinetostatic model is also an alternative method to the finite-element approach which uses commercially available software. The modelling and meshing procedures using finite-element software could be time consuming. The kinetostatic model of micro-motion stages wasdeveloped based on the theory of the connection of serial and parallel springs. developed based on the theory of the connection of serial and parallel springs. The derivation of the kinetostatic model is simple and the model is expressed in closed-form equations. Material properties and link parameters are variables in this model. Compliances of flexure hinges are also one of the variables in the model. Therefore the most suitable flexure hinge equation can be selected based on the scheme aforementioned in order to calculate the kinetostatics of micro-motion stages accurately. Planar micro-motion stages with topologies of a four-bar linkage and a 3-RRR (revolute-revolute-revolute) structure were studied in this thesis. These micromotion stages are monolithic compliant mechanisms which consist of circular flexure hinges. Circular flexure hinges are used in most of the micro-motion stages which require high positioning accuracies. This is because circular flexure hinges provide predominantly rotational motions about one axis and they have small parasitic motions about the other axes. The 3-RRR micro-motion stage studied in this thesis has three-degrees-of-freedom (DOF). The 3-RRR stage consists of three RRR linkages and each RRR linkage has three circular flexure hinges. A Pseudo-Rigid-Body-Model (PRBM), a kinetostatic model and a two-dimensional finite-elementanalysis (FEA) model generated using ANSYS of micro-motion stages are presented and the results of these models were compared. Advantages of the kinetostatic model was highlighted through this comparison. Finally, experiments are presented to verify the accuracy of the kinetostatic model of the 3-RRR micromotion stage. / http://proxy.library.adelaide.edu.au/login?url= http://library.adelaide.edu.au/cgi-bin/Pwebrecon.cgi?BBID=1289361 / Thesis (Ph.D.) -- University of Adelaide, School of Mechanical Engineering, 2007

In mesoscale manufacturing desired dimensional and surface characteristics are defined, but edge conditions are not specified in design. The final edge conditions that exist in mesoscale objects are created not only by the manufacturing process but, because of their size, also by part handling procedures. In these parts, the concern is not only with burrs, which can be formed by some mesoscale manufacturing processes, but also with the shape and size of the edge. These properties are critically important as the edge can constitute a large percentage of the smallest features of mesoscale objects. Undefined edge geometry can result in measurement, assembly, and operational difficulties. Due to the potential problems caused by edge conditions, it is desirable to have the ability to measure and characterize the edge conditions of parts. This thesis considers mesoscale measurement tools to provide an edge measurement tool recommendation based on edge size and properties. A set of analysis techniques is developed to determine the size and shape of the measured edge, locate any local inconsistencies such as burrs or dents, and track trends in calculated parameters as a function of edge position. Additionally, a standard method for communicating design requirements is suggested in order to differentiate between acceptable and unacceptable edges.

In order to apply to S-Band (1-4.5 GHz) of wireless communication system, we designed and fabricated a high-insolating RF-MEMS switch by surface micromachining technology in this study. In terms of the micro switch, we performed the structural design, high frequency simulation, components process integration and high-frequency measurement in this study. Especially for making components be high-isolation, low-loss and low-driving voltage, we proposed the following three methods: (i) adjusting the space and width of the transmission lines to improve the RF performance; (ii) applying the stress imbalance, by using dual metal composite top electrode, to form a arched contact electrode and reduce the drive voltage efficiently; (iii) using non-isometric spring structure to stabilize the electrode movement of the components. Besides, we did the optimizing simulation for this study, which were supported by Ansoft-HFSS and ADS, in terms of the micro switch which has different structural design as mentioned above. The size of the optimized RF micro-switch which we developed for this study is only 145 £gm ¡Ñ 205 £gm. Switched from on-state to off-state, the component needs 36.5V drive voltage only. According to the result of the commercial network analyzer in 1-4.5GHz frequency range, the isolation rate of the components reaches -59.721dB while off-state; the insert los reaches -1.625dB while on-state.

Miniaturized devices are the key producing next-generation microelectro-mechanical products. The applications extend to many fields that demand high-level tolerances from microproducts and component functional and structural integrity. Silicon-based products are limited because silicon is brittle. Products can be made from other engineering materials and need to be machined in microscale. This research deals with predicting microtool failure by studying spindle runout and tool deflection effects on the tool, and by measuring the cutting force that would fail the tool during microend-milling. End-milling was performed using a tungsten carbide (Ø1.016 mm dia., 2 flute) tool on SS-316L material. Tool runout measured using a laser was found to be less than 1 µm and tool deflection at 25000 rpm was 20 µm. Finite element analysis (FEA) predicts tool failure due to static bending for a deflection greater than 99% of tool diameter. Threshold values of chipload and cutting force resulting in tool failure were found using workdone by tool. Threshold values to predict tool failure were suggested for axial depth of cut in between 17.25% - 34.5% of cutter length. For a chipload greater than 20% of cutter diameter, the microtool fails instantly for any radial depth of cut.

This thesis presents a study into the enhancement of productivity in micromilling processes by considering a fundamental treatment of tool path trajectory generation techniques and process optimization strategies that account for the impact of scale effects present in high-speed, high-precision micromachining operations. Micromilling is increasingly applied to the production of a wide variety of micro components, due to its high precision and flexibility. However, the productivity of micromilling is limited by the low feedrates necessitated by the inherent high precision and small feature size. In this study, several scale effects present at the microscale are identified, in particular the increase of the ratio of tool size to feature size, and the corresponding impact on trajectory generation and process optimization is investigated. The scale effects are shown to cause increased geometric error when the standard method of VF-NURBS is applied to microscale feedrate optimization. The method of Enhanced Variable-Feedrate NURBS (EVF-NURBS) is proposed and shown to successfully compensate for the scale effects leading to reduced geometric error. A key contribution of this study is the construction and experimental validation of the Variable-Feedrate Intelligent Segmentation (VFIS) method for increased feedrates and improved stability. The VFIS method provides a cutting time reduction of more than 50% in some cases, while effectively constraining geometric error. Two tool size optimization schemes are presented for maximizing productivity and minimizing geometric error while accounting for dynamic effects uniquely present at the microscale. Finally, the development of a low-cost, high-precision micro-mesoscale machining center (mMC) is presented.

No description available.

Micromachined needles have been shown to successfully transport biological molecules into the body with minimal invasiveness and pain, following the insertion of needles into the skin. The aim of this research is to demonstrate that micromachined hollow polymer needle arrays fabricated using UV lithography into micromolds, a potential batch-manufacturable process, can exhibit comparable insertion and injection performance to conventional hypodermic needles for drug delivery into skin. A dual-exposure-and-single-development process flow is proposed for the above-mentioned UV lithography into micromolds approach to construct a pyramidal-tip hollow microneedle array with an integral baseplate and fluidic manifold. The developed process ultimately resulted in the ability to fabricate a 10×10 array of hollow SU-8 microneedles measuring 825 μm in height, 400 μm in width, and possessing a lumen of 120 μm in diameter. The tip diameter of the microneedles ranges from 15 μm to 25 μm. The insertion force of single needles characterized using excised porcine skin as a substrate is 2.4±1.2 N. Nevertheless, the high insertion force of 2.4 N per needle may cause a significant concern when a large number of needles are required to insert into skin for drug delivery. Conventional hypodermic needles have two key structural characteristics: a sharp beveled tip and a large side-terminated lumen. Integration of these two key characteristics of hypodermic needles into microneedle design can potentially enhance microneedle performance. To reduce the insertion force and to incorporate the two key characteristics of hypodermic needles into the design of microneedles, a new needle tip design, namely the hypodermic-needle-like design, is presented. A 6×6 array of hypodermic-needle-like microneedles of 1 mm in height, approximate 350 μm in width, and with a lumen of 150 μm in diameter is demonstrated with successful insertion of the needle array into skin and an 85% lumen openness yield. The insertion force is significantly reduced by an order of magnitude with the new needle tip design and is 0.275±0.113 N per needle, comparable to that of hypodermic needles, i.e., 0.284±0.059 N. The hypodermic-needle-like microneedles exhibit a margin of safety of 180 for successful needle insertion into skin prior to needle fracture. A successful manual fluid injection into skin using single microneedle is demonstrated. The micromachined hypodermic-needle-like polymer needle arrays presented in this dissertation are fabricated using UV lithography into micromolds, a potentially batch-manufacturable process, and exhibit comparable insertion performance to conventional hypodermic needles. Injection capability into skin is also demonstrated with a hypodermic-needle-like microneedle, illustrating the utility of these devices.

This thesis presents design, modeling, and fabrication of an RF MEMS adjustable impedance matching network. The device employs the basic triple stub matching technique for impedance matching. It has three adjustable length stubs which are implemented using capacitively loaded coplanar waveguides. The capacitive loading of the stubs are realized using the MEMS switches which are evenly distributed over the stubs. There are 40 MEMS bridges on each stub whichare separated with &amp / #955 / /40 spacing making a total of 120 MEMS switches in the structure. The variability of the stub length is accomplished by closing the MEMS switch nearest to the required stub length, and making a virtual short circuit to ground. The device is theoretically capable of doing matching to every point on the Smith chart. The device is built on coplanar waveguide transmission lines. It has a center operating frequency of 10GHz, but because of its adjustability property it is expected to work in 1-40GHz range. It has dimensions of 8950 &times / 5720&micro / m2. This work is the continuation of the first national work on fabrication of RF MEMS devices. The device in this work is fabricated using the surface micromachining technology in the microelectronic facilities of Middle East Technical University.