Andrew Kovacs

Early-Warning Wireless Telemeter for Harsh-Environment Bearings

We demonstrate an early-warning wireless and high-temperature telemeter that continuously monitors the temperature of a roller bearing during operation. The telemeter detects imminent temperature-induced failure of the bearing four minutes before conventional thermocouple sensors respond to conditional changes, leaving ample time for appropriate actions to avoid complete failure. The telemeter includes a commercially available temperature-sensitive capacitor that operates up to 220deg C. In addition, the telemeters natural frequency changes with capacitance, which in turn is dependant on temperature. Information transmission between the bearing and a monitoring station is accomplished with a matched interrogator driven by a sinusoidal signal at 2.27 MHz. The telemeters natural frequency shifts by about 0.65 kHz/degC, which is measured as a change in voltage on the interrogator inductor.

Antibiased Electrostatic RF MEMS Varactors and Tunable Filters

This paper presents a new approach for substantially enhancing the linearity and reducing the effects of bias noise for electrostatic RF microelectromechanical systems (MEMS) devices. The proposed method relies on applying bias voltages with opposite polarities to cancel the dynamic vibration of the MEMS structures. In this paper, the method has been applied to a shunt RF MEMS varactor and a MEMS tunable evanescent-mode tunable filter. In the first case, the shunt MEMS varactor is split into two identical parts that are biased with opposite voltages. This results in almost complete cancelation of the odd-order modulation components, leading to 20-28-dB linearity enhancement depending on the noise and the design. Analytical results, a computer-aided design model and measurements validate the proposed approach. In the tunable filter case, opposite bias voltages are applied on the tuners of its resonators. Simulated and measured results are also presented in this case. Measurements show a sideband reduction as high as 13 dB. In both cases, the effectiveness of the proposed method in the presence of fabrication uncertainties are also considered.

Wireless temperature microsensors integrated on bearings for health monitoring applications

This paper reports the performance of a wireless MEMS bimorph temperature sensor integrated on a bearing for component health monitoring applications. The sensor consists of a robust array of bimorphs consisting of gold and thermally-grown oxide operable to at least 300°C. Fabrication details are included, as well as the hermetic packaging information. Speed of actuation results from a high-speed camera is included showing the actuation time is less than 600 µs. Reliability testing of the bimorph array up to 400 million thermal cycles is also shown, after which the bimorphs still yield consistent behavior. Finally, dynamic testing is performed showing actual bearing temperature values at different speeds on a real-world helicopter bearing

Estimating residual stress, curvature and boundary compliance of doubly clamped MEMS from their vibration response

Structural parameters of doubly clamped microfabricated beams such as initial curvature, boundary compliance, thickness and mean residual stress are often critical to the performance of microelectromechanical systems (MEMS) and need to be estimated as a part of quality control of the microfabrication process. However, these parameters couple and influence many metrics of device response and thus are very difficult to disentangle and estimate using conventional methods such as the M-test, static mechanical tests, pull-in measurements or dynamic mechanical tests. Here we present a simple, non-destructive experimental method to extract these parameters based on the non-contact measurement of the natural frequencies of the lowest few eigenmodes of the microfabricated beam, and knowledge of Youngs modulus and plan dimensions of the beam alone. The method exploits the fact that certain eigenmodes are insensitive to some of these structural parameters which enable a convenient decoupling and estimation of the parameters. As a result, the method does not require complicated finite element analysis, is insensitive to the gap height and introduces no contact wear or dielectric charging effects. Experiments are performed using laser Doppler vibrometry to measure the natural frequencies of doubly clamped, nickel, RF-MEMS capacitive switches and the method is applied to extract the residual stress, beam thickness, boundary compliance and post-release curvature.

Uncertainty in microscale gas damping: Implications on dynamics of capacitive MEMS switches

Effects of uncertainties in gas damping models, geometry and mechanical properties on the dynamics of micro-electro-mechanical systems (MEMS) capacitive switch are studied. A sample of typical capacitive switches has been fabricated and characterized at Purdue University. High-fidelity simulations of gas damping on planar microbeams are developed and verified under relevant conditions. This and other gas damping models are then applied to study the dynamics of a single closing event for switches with experimentally measured properties. It has been demonstrated that although all damping models considered predict similar damping quality factor and agree well for predictions of closing time, the models differ by a factor of two and more in predicting the impact velocity and acceleration at contact. Implications of parameter uncertainties on the key reliability-related parameters such as the pull-in voltage, closing time and impact velocity are discussed. A notable effect of uncertainty is that the nominal switch, i.e. the switch with the average properties, does not actuate at the mean actuation voltage. Additionally, the device-to-device variability leads to significant differences in dynamics. For example, the mean impact velocity for switches actuated under the 90%-actuation voltage (about 150 V), i.e. the voltage required to actuate 90% of the sample, is about 129 cm/s and increases to 173 cm/s for the 99%-actuation voltage (of about 173 V). Response surfaces of impact velocity and closing time to five input variables were constructed using the Smolyak sparse grid algorithm. The sensitivity analysis showed that impact velocity is most sensitive to the damping coefficient whereas the closing time is most affected by the geometric parameters such as gap and beam thickness.

Direct measurement of field emission current in E-static MEMS structures

Direct experimental evidence of field emission currents in metallic MEMS devices is presented. For the first time, high resolution I–V curves have been demonstrated for micro-gaps in MEMS-based capacitor/switch-like geometries. The I–V dependence shows a good agreement with the Fowler-Nordheim theory, supporting the hypothesis that field emission plays a significant role in charging phenomena in MEMS switches. The data has been used to extract effective values of the field enhancement factor, β, for the metallic structures fabricated under typical MEMS processes.

Anti-biased RF MEMS varactor topology for 20–25 dB linearity enhancement

A new topology for significantly improving the linearity of RF MEMS varactors is reported in this paper. A single MEMS varactor subject to a low frequency noise and/or modulating signal is the main structure under consideration. The key idea is to separate the single varactor into two varactors in parallel, which are anti-biased. This leads to out-of-phase vibration of the two beams under the influence of the low-frequency noise and/or modulating signal, which eventually leads to a nearly constant capacitance value of the entire topology. The effectiveness of the new method is demonstrated by both measured results from fabricated RF MEMS varactors and simulated results using a large-signal model of MEMS varactor in Agilents Advanced Design System (ADS). Both experimental and simulated results indicate an improvement of 20‥25 dB compared to the conventional design when the varactors are biased close to their pull-in voltage and the low-frequency signal is close to their self-resonant frequency. This underlines the potential of the proposed topology in high-power RF MEMS circuits.

Bearing cage telemeter for the detection of shaft imbalance in rotating systems

We demonstrate for the first time a low-cost battery-free wireless telemeter integrated on a dynamically balanced ball bearing operating in a test setup that mimics the operating conditions of a two-compressor turbocharger. In this method we continuously monitor the bearing cage temperature and demonstrate the ability to detect rotor imbalance during operation even at low speeds below 5,550 RPM. The telemeter is based on a simple commercially-available temperature-sensitive capacitor coupled with an inductor that is mounted on the bearing cage and is able to operate at temperatures up to 150°C. Cage temperature is sensed wirelessly through a matched interrogator tuned to the natural frequency of the cage sensor at 13.1 MHz. A 5°C temperature differential is detected at 4,700 RPM between a balanced rotor and one with an applied 2.32×10{−3} kg-m imbalance.

Residual Stress Extraction of Surface-Micromachined Fixed-Fixed Nickel Beams Using a Wafer-Scale Technique

This paper reports on the extraction of residual stress in surface-micromachined nickel thin films of electrostatically actuated fixed-fixed beams using a wafer-scale technique. The distribution of residual stress for 87 beams on a 4-in quarter wafer piece is presented. The residual stress (σ0) is determined from the best fit of the displacement-voltage curves predicted by a computationally efficient model to the experimental data. The nondestructive and automated measurements are taken at room temperature and directly at the beam itself without any additional test structures. The model employed incorporates the nonideal effects of inclined supports, nonflat initial beam profiles, and fringing fields. The extracted residual stress values vary between -12.8 and 13.6 MPa (negative values are for compressive stresses and positive ones for tensile stresses). The residual stresses for these 87 beams follow a nearly normal distribution with a mean value of -1.7 MPa and a standard deviation of 5.9 MPa, which represents the variability of the residual stresses across the wafer. Detailed uncertainty analysis has been conducted, and it reveals that inaccurate modeling of the nonideal effects will result in significant errors in the extracted residual stress. Although demonstrated on nickel thin films, this technique can be applied to other metallic thin films.

An equation-based nonlinear model for non-flat MEMS fixed–fixed beams with non-vertical anchoring supports

Anchor supports in MEMS beams are often far from the ideally assumed built-in or step-up conditions. Practical fabrication processes often result in non-vertical anchoring supports (referred to as inclined supports in the following text) which significantly influence the post-release performance of the beam. This paper brings attention to the presence of the inclined supports in surface micromachined fixed–fixed beams and models the mechanical and electromechanical effects of inclined supports for the first time. Specifically, we calculate and validate the effects of residual stress and loading on the post-release beam behavior including their nonlinear large-displacement characteristics. In addition the model accounts for non-flat beam profiles caused by residual stress and/or a non-flat sacrificial layer profile. Inclined supports are modeled as cantilever beams connected to a horizontal beam. The Euler–Bernoulli equations for all beams are simultaneously solved to calculate the axial stress of the horizontal beam and the axial, translational, and rotational compliance of the supports. Nonlinear effects due to stretching and residual stress are also included. The calculated beam displacements agree with FEM models to within 1.1% in both the linear and nonlinear regimes. Furthermore, experimentally-obtained displacements of six fabricated beams with inclined supports agree to within 5.2% with the presented model.