Wafer Scale Factories Using Assembled MEMS Robots

This paper presents recent research progress in the manufacturing of MEMS based robotic positioning technology. The goal is to configure a nanofactory consisting of multiple microrobots that combine high precision with high throughput along with other application-specific requirements such as strength, dexterity, and work volume. This requires design to satisfy part mobility and dexterous manipulation with high precision. The first microrobot called the Articulated Four Axes Microrobot (AFAM) is a 3D dexterous micromanipulator robot, and we describe nano indentation experiments using AFM tips mounted on the microrobot. By combining positioning data obtained using laser interferometers and SEM imaging of nano indentation data, precision metrics such as accuracy, repeatability and resolution of the AFAM robot are determined. Preliminary reliability findings are presented. The second robot type is the ARRIpede microcrawler, and we present recent progress in microrobot precision evaluation including motion resolution and repeatability. Using these two microrobots as basic positioning and manipulation units, a nanoassembly module concept for hybrid nano assembly applications is proposed. In this paper we present recent progress in fabricating, assembling and evaluating the basic microfactory modules.

Rupture and Reformation of Ultra-Thin Surface Films

In lubricated contact conjunctions film ruptures close to the exit boundary. This significantly affects the load carrying capacity and can lead to direct surface interactions. Nano-scale films (several molecular diameters of the lubricant) are no exception, a fact that has been observed using ellipsometry studies for ultra-thin film conjunctions representative for high storage capacity hard disk drives. Immediately beyond the film rupture an area of cavitation occurs and the continuity of flow condition is breached. It has been shown that for molecularly smooth surfaces solvation effect becomes dominant. This means that the contact exit is subject to discrete drainage of lubricant and may be devoid of a sufficient lubricant for film reformation to occur. This can be a stumbling block in an increasing quest to increase the data storage density of hard disk drives. Wear can become a problem as well as non-uniformity of free surface film at the inlet meniscus. It has been noted that peaks of lubricant can gather in some places, a phenomenon referred to as lubricant mogul. These localized piles of lubricant can exceed the nominally aimed for lubricant film thickness necessary for a given data storage level. This paper carries out an in-depth prediction of ultra thin film lubricant behavior through the contact. Hydrodynamic as well as near surface effects and intermolecular interactions responsible for the supply, formation, cavitation and reformation of thin films in the slider-disk conjunction have been considered.

Modeling and Analysis of an Optically-Actuated, Bistable MEMS Device

Bistable microsystems have drawn considerable interest from the MEMS/NEMS research community not only due to their broad applicability in commercial applications, such as switching, but also because of the rich dynamic behavior they commonly exhibit. While a number of prior investigations have studied the dynamics of bistable microsystems, comparatively few works have sought to characterize their transient behavior. The present effort seeks to address this through the modeling and analysis of an optically-actuated, bistable MEMS switch. The work begins with the development of a distributed-parameter representation for the system, which is subsequently reduced to a lumped-mass analog and analyzed through the use of numerical simulation. The influence of various system and excitation parameters, including the applied axial load and optical actuation profile, on the system’s transient response is then investigated. Ultimately, the methodologies and results presented herein should provide for a refined predictive design capability for optically-actuated, bistable MEMS devices.

Development of a Driving Simulator for Virtual Evaluation of Vehicle Dynamics

This paper delineates the conceptual algorithms of a driving simulator which is intended for vehicle performance evaluation and to act as a virtual platform for research studies and therefore eliminates the cost and dangerous of field testing. A virtual proving ground for vehicle testing has been devised through which virtual handling, traction and ride tests can be performed. A fully instrumented simulator cabin combining the driver and the vehicle simulation package is developed. Different vehicle configurations are simulated during typical sever manoeuvres which reflects the robustness and fidelity of the devised simulator.

A Geometric Parametric Analysis of a Magnetorheological Engine Mount

Engine vibration has caused a great deal of research for isolation to be performed. Traditionally, isolation was achieved through the use of pure elastomeric (rubber) mounts. However, with advances in vehicle technology, these types of mounts have become inadequate. The inadequacy stems from the vibration profile associated with the engine, i.e. high displacement at low frequency and small displacement at high frequency. Ideal isolation would be achieved through a stiff mount for low frequency and a soft mount for high frequency. This is contradictory to the performance of the elastomeric mounts. Hydraulic mounts were then developed to address this problem. A hydraulic mount has variable stiffness and damping due to the use of a decoupler and an inertia track. However, further advances in vehicle technology have rendered these mounts inadequate as well. Examples of these advances are hybridization (electric and hydraulic) and cylinder on demand (VCM, MDS & ACC). With these technologies, the vibration excitation has a significantly different profile, occurs over a wide range of frequencies, and calls for a new technology that can address this need. Magnetorheological (MR) fluid is a smart material that is able to change viscosity in the presence of a magnetic field. With the use of MR fluid, variable damping and stiffness can be achieved. An MR mount has been developed and tested. The performance of the mount depends on the geometry of the rubber part as well as the behavior of the MR fluid. The rubber top of the mount is the topic of this study due to its major impact on the isolation characteristics of the MR mount. To develop a design methodology to address the isolation needs of different hybrid vehicles, a geometric parametric finite element analysis has been completed and presented in this paper.

Organic Rankine Cycle for Waste Heat Recovery in a Hybrid Vehicle

The Rankine cycle is used commercially to generate power in stationary power plants using water as the working fluid. For waste heat recovery applications, where the temperature is lower, water is typically replaced by a carefully selected organic fluid. This work is based on using the waste heat in an automobile to generate electricity using the Organic Rankine cycle (ORC) with R245fa (1, 1, 1, 3, 3 penta-fluoropropane) as the working fluid. The electricity thus generated can be used to drive the accessory load or charge the battery which in any case helps improve the fuel economy. A simple transient numerical model has been developed that is capable of capturing the main effects of this cycle. Results show that exhaust heat alone can generate enough electricity that is capable of bringing about an improvement to the fuel economy under transient drive cycle conditions. Power output during EPA Highway drive cycle is much higher than EPA City due to higher exhaust mass flow rate and temperature. Time needed to reach operating conditions or in other words, the warm-up time plays an important role in the overall drive cycle output. Performance is found to improve significantly when coolant waste heat is used in conjunction with the residual exhaust heat to pre-heat the liquid. A sizing study is also performed to keep the cost, weight, and packaging requirement down without sacrificing too much power. With careful selection of heat exchanger design parameters, it has been demonstrated that the backpressure on the engine can be actually lowered by cooling off the exhaust gas. This lower backpressure will further boost the fuel economy gained by the electricity produced by the Rankine bottoming cycle.

Optimal Plug-In Hybrid Vehicle Design and Allocation for Minimum Life Cycle Cost, Petroleum Consumption and Greenhouse Gas Emissions

Plug-in hybrid electric vehicle (PHEV) technology has the potential to help address economic, environmental, and national security concerns in the United States by reducing operating cost, greenhouse gas (GHG) emissions and petroleum consumption from the transportation sector. However, the net effects of PHEVs depend critically on vehicle design, battery technology, and charging frequency. To examine these implications, we develop an integrated optimization model utilizing vehicle physics simulation, battery degradation data, and U.S. driving data to determine optimal vehicle design and allocation of vehicles to drivers for minimum life cycle cost, GHG emissions, and petroleum consumption. We find that, while PHEVs with large battery capacity minimize petroleum consumption, a mix of PHEVs sized for 25–40 miles of electric travel produces the greatest reduction in lifecycle GHG emissions. At today’s average US energy prices, battery pack cost must fall below $460/kWh (below $300/kWh for a 10% discount rate) for PHEVs to be cost competitive with ordinary hybrid electric vehicles (HEVs). Carbon allowance prices have marginal impact on optimal design or allocation of PHEVs even at $100/tonne. We find that the maximum battery swing should be utilized to achieve minimum life cycle cost, GHGs, and petroleum consumption. Increased swing enables greater all-electric range (AER) to be achieved with smaller battery packs, improving cost competitiveness of PHEVs. Hence, existing policies that subsidize battery cost for PHEVs would likely be better tied to AER, rather than total battery capacity.

Optimal Design of a Micro Series Elastic Actuator

We propose using series elastic actuation (SEA) in micro mechanical devices to achieve precise control of the interaction forces. Using μSEA for force control removes the need for high-precision force sensors/actuators and allows for accurate force control through simple position control of the deflection of a compliant coupling element. Since the performance of a μSEA is highly dependent on the design of this compliant coupling element, we employ a design optimization framework to design this element. In particular, we propose a compliant, under-actuated half-pantograph mechanism as a feasible kinematic structure for this coupling element. Then, we consider multiple design objectives to optimize the performance of this compliant mechanism through dimensional synthesis, formulating an optimization problem to study the trade-offs between these design criteria. We optimize the directional manipulability of the mechanism, simultaneously with its task space stiffness, using a Pareto-front based framework. We select an optimal design by studying solutions on the Pareto-front curve and considering the linearity of the stiffness along the actuation direction as a secondary design criteria. The optimized mechanism possesses high manipulability and low stiffness along the movement direction of the actuator; hence, achieves a large stroke with high force resolution. At the same time, the mechanism has low manipulability and high stiffness along the direction perpendicular to the actuator motion, ensuring good disturbance rejection characteristics. We model the behavior of this compliant mechanism and utilize this model to synthesize a controller for μSEA to study its dynamic response. Simulated closed loop performance of the μSEA with optimized coupling element indicates that force references can be tracked without significant overshoot and with low tracking error (about 1.1%) even for periodic reference signals.

Compensator Design for a MEMS Gyroscope With Quadratic Optimal Control

A number of control algorithms have been reported to adopt force balancing scheme into MEMS vibratory gyroscope systems. In practice, however, many algorithms are difficult to implement with electronic circuits. This paper designs and analyzes a lead compensator for a MEMS gyroscope via the Linear Quadratic Regulator (LQR) technique. LQR optimizes and balances the control effort and system response swiftness. Simulation shows the gyroscope achieves high linearity, wide dynamic range, and high robustness to fabrication uncertainties with this efficient compensator design. The closed-loop scale factor uniformity error is 0.7% under ±10% parameter perturbations. The compensator designed in this paper exhibits comparable outstanding performance compared to other reported control algorithms. The method reported in this paper is proved to be effective and can be used in a wide range of applications.

Parametric Excitation of Flexural Vibrations of Micro Beams by Fringing Electrostatic Fields

We report on an approach for efficient excitation of large amplitude flexural out-of-plane vibrations of micro beams and present results of theoretical and experimental feasibility study of the suggested principle. An actuating electrode is located symmetrically at the two sides of the beam and is fabricated from the same layer of the wafer. The electrostatic force is engendered by the asymmetry of the fringing fields in the deformed state and acts in the direction opposite to the deflection therefore increasing the effective stiffness of the system. The time-varying voltage applied to the electrode results in the modulation of this electrostatic stiffness and consequently in the parametric excitation of the structure. The device may exhibit large vibrational amplitudes not limited by the pull-in instability common in close-gap actuators. In contrast to previously reported devices excited by the fringing fields, the force considered here is of distributed character. The reduced order model was built using the Galerkin decomposition with linear modes as base functions and the resulting system of nonlinear differential equations was solved numerically. The electrostatic forces were approximated by means of fitting the results of three-dimensional numerical solution for the electric fields. The devices fabricated from a silicon on insulator (SOI) substrate using deep reactive ion etching (DRIE) based process were operated in ambient air conditions and the responses were registered by means of Laser Doppler Vibrometry. The experimental resonant curves were consistent with those predicted by the model. Theoretical and preliminary experimental results illustrated the feasibility of the suggested approach.