Search

We report record-high performance from a fused-silica micro Precision Shell Integrating (PSI) gyroscope. The PSI gyroscope uses a fused-silica resonator with a diameter of 1 cm and a mechanical quality factor (Q) of 5.2 million. The gyroscope has an angle random walk (ARW) of 0.00016 deg/√hr and short-term in-run bias instability of 0.0014 deg/hr without any temperature compensation. This performance for such a small shell gyroscope compares well with the larger mHRG and HRG devices reported previously. This performance, in a device this small, is achieved by optimizing the fabrication of the shell resonator so it provides high Q, large mass, high frequency resonance, and low as-fabricated frequency mismatch.

This paper reports the design, fabrication technology and initial testing results of an all fused-silica 3D shell (hemispherical) resonator and capacitive drive/sense electrodes made from a second shell, for use in high-performance gyroscopes. A single fused-silica shell is patterned to form multiple electrodes that follow the curvature of the shell resonator. 10 mm diameter Shell resonator is then integrated inside a larger Shell electrode (SiS). Our approach enables, for the first time, the fabrication of curved electrodes at the micro-scale. The SiS device provides large overlap area and nearly uniform gap with large capacitance for better frequency tuning capability, reduced operating voltages, improved capacitive sensitivity and reduced temperature sensitivity as compared to other electrode architectures. Successful fabrication and testing of two prototype SiS devices are demonstrated. Quality factor as high as of 0.4 Million and frequency split as low as 1.3 Hz is obtained at operating frequency of ~ 6000 Hz.

This paper presents a modular and generic micromachined oven-control system for use with miniature micro-electro-mechanical system (MEMS) transducers. The micro-oven-controlled off-the-shelf commercial six-axis MEMS inertial measurement unit (IMU), Invensense MPU-6050, provides the lowest reported temperature-induced root of sum of squares bias errors of 62.71°/h and 1.920 mg from -40°C to 85°C for three-axis gyroscopes and three-axis accelerometers, respectively. The micro-oven control system provides thermal isolation from the surrounding environment using a micro-machined isolation platform, vacuum-sealing, and a metal package. In addition, a CMOS temperature sensor, a proportional-integral-derivative-based temperature control scheme, and least mean square and random forest compensation algorithms are utilized to reduce temperature-induced bias drifts of IMUs. The most stable axes achieve peak-to-peak bias drifts of 12.78°/h and 665.2 ug during a thermal-cycle test for gyroscopes and accelerometers, respectively. The oven's heater power consumption is

The fabrication of shell resonators for vibratory gyroscopes is often accompanied by a split between drive and sense mode frequencies. To tune these frequencies and reduce the frequency split, a voltage can be applied between the shell and electrodes placed around it. This paper investigates electrostatic frequency tuning in shell resonators as a function of voltage, capacitive gap between the shell and electrode, electrode span angle and height, and electrode placement and configuration using a numerical approach. The results of this investigation show that conformal electrodes placed around the shell rim with a uniform gap provide the best tuning capability. Considering fabrication challenges and shell aspect-ratio, other electrode configurations might be more practical in some devices, although they have worse tuning capability.

This paper presents the design and experimental results of a microvibratory actuation and sensing platform to provide on-chip physical stimulus for in situ calibration of long-term scale factor drifts in multiaxis microelectromechanical systems (MEMS) inertial sensors. The platform consists of a three degrees-of-freedom micromotion stage that can provide piezoelectric actuation for X / Y -tilting reference stimuli, compensation of undesired off-axis motion, integrated sensing of applied periodic stimulus, and electrostatic position lock-down for shock protection. A dual-axis MEMS gyroscope is mounted on top of the microplatform, and its electrical interconnects are provided through microfabricated highly flexible parylene cables with virtually zero-loading. The piezoelectric stage is measured to provide up to 280°/s angular ac excitation to a 25-mg inertial sensor payload at an expense of

This paper describes a generic, ultra-low-noise transimpedance amplifier (TIA) for capacitive MEMS sensors. The TIA obtains both very low input referred current noise (5.12 fA/VHz) and wide dynamic range (123 dB) by employing a digitally-controlled floating resistor, a voltage-and-temperature tolerant current generator, and low-noise operational amplifier (op-amp). The floating resistor allows the amplifier to obtain excellent output linearity and obtain a wide dynamic range. An integrated digital controller of the current generator lowers the TIA noise by reducing the amount of fluctuation in the resistance of the floating resistor by eliminating the transmission of control voltages for the floating resistor through a bonding pad. A beta-multiplier current reference circuitry with a resistor with low temperature coefficient of resistance (TCR) generates bias current with very low sensitivity to variations in supply voltage and temperature. A two-stage Miller-compensated op-amp with optimized transistor dimensions is designed. The prototype chip was fabricated using a 0.18 μm standard CMOS process with 1.8 V supply voltage. The circuitry is evaluated with a high performance vibratory MEMS gyroscope and excellent bias stability (0.04 deg/h) is measured.

We report the latest experimental results from a fused-silica Birdbath Resonator Gyroscope (BRG) with a quality factor of 419k and a resonant frequency of 9030 Hz. The BRG and its readout/control system achieve an angle random walk (ARW) of 0.00126 deg/√hr and a bias stability of 0.0391 deg/hr. These results were obtained using a force-rebalanced control architecture without temperature control or additional compensation. The ARW and bias stability are further improved by driving the resonator at displacements near 10% of the nominal electrostatic gap. The low gyroscope noise is attributed to a relatively large measured scale factor of 100 mV/deg/s.

This paper reports characterization of a micro actuation and sensing platform for on-chip scale factor measurement and calibration of generic MEMS inertial sensors. The platform has a unique 6-DOF-motion capability that is used to deliver calibration stimuli with minimum off-axis motion, simultaneous excitation of multiple calibration axes, and on-chip measurement of acceleration sensitivities of the tested IMU. The microsystem is demonstrated to provide rotational calibration references with less than 15–30 ppm short-term instability for characterization of scale factor of a commercial gyroscope load in its full sensing range.

In this paper, we propose a comparator-based switched-capacitor (CBSC) architecture using a multi-level input tracking preset voltage scheme. The CBSC is used to compensate for technology scaling and to reduce the power consumption of a 2.8 MS/s 10-bit cyclic analog-to-digital converter (ADC). A multi-level preset voltage tracks the input voltage in order to improve the conversion rate without consuming additional power. Additionally, a comparator, current sources, and a feedback capacitor are shared to reduce the power and area of this cyclic ADC. Near the Nyquist-rate, a prototype implemented in 0.18 μm CMOS technology has a signal-to-noise and distortion ratio of 53.69 dB and a spurious-free dynamic-range of 62.36 dB, while consuming 0.73 mW of power.

We present a 3-D fused-silica micro-scale shell gyroscope, called the birdbath resonator gyroscope (BRG). The BRGs axisymmetric geometry leads to a good frequency and Q symmetry. The birdbath resonator can be fabricated with a good structural symmetry because its anchor is self-aligned to the rest of the structure. The BRG has n=2 wine-glass modes at 10.5 kHz and has a large frequency separation between the n=2 wine-glass modes and the closest parasitic mode (|fparasitic-fn=2|/fn=2=0.3), which will potentially lead to a low vibration sensitivity. The equations of motion for 3-D shell gyroscopes are derived and the effective mass and angular gain of the BRG is estimated using finite element method (FEM). The BRG is fabricated using a 3-D micrometer-blow-torching process and assembly on an electrode substrate made with the silicon-on-glass process. The BRG is operated in the force-rebalance mode at vacuum at room temperature and has a scale factor of 27.9 mV/(deg/s), a full-scale range , an angle random walk of 0.106 deg/√h, and a bias stability of 1 deg/h. A large angular gain (0.317) is measured, which is close to the estimated value of 0.25 obtained via FEM.

Identification of gamma rays and neutrons is a fundamental technique used in elementary particle physics experiments, such as the search for dark matter, double beta decay, etc. The pulse shape discrimination (PSD) methods with liquid and crystal scintillators, a well-known technology, have been under development for several decades. The PSD methods with crystal scintillators are known to be less stable and less effective than PSD methods with liquid crystals. Here, an improved PSD technique with a CsI crystal scintillator based on the well-known PSD methods with liquid scintillators is introduced. In this study, we found the possibility that the proposed PSD with a CsI crystal scintillator provides a performance to discriminate gamma rays and neutrons equivalent to that of a PSD with a liquid scintillator.

We propose a low-voltage low-power clock and data recovery (CDR) circuit which incorporates a relaxation-based voltage-controlled oscillator and clock-edge modulation, which eliminates the need for an external reference clock without allowing harmonic locking. This CDR supports input data rates between 200 kbps and 10 Mbps at 0.7 V and operates up to 24 MHz at 1.0 V. The proposed design consumes 8 at an input data rate of 10 Mbps and achieves 0.8 pJ/bit of energy per bit even though the circuit is implemented in a 0.18- μm CMOS technology.

A capacitive sensor design is introduced with an associated method for constructing a trusted threshold crossing angle (θ TH ) from the capacitance profile to perform precise measurement of the out-of-plane (X and Y) tilting angles of a 6 degree of freedom (DOF) piezoelectric actuation stage for in situ calibration of micro electromechanical (MEMS) gyroscopes. The capacitance vs. tilting angle relationship of the sensor is calibrated with the assist of an external Laser Doppler vibrometer (LDV) and a closed-form model of the sensor. The θ TH are extracted by detecting the crossing location of the reconstructed capacitance trajectories using measurement from LDV or on-chip capacitive sensor, the comparison shows close agreements. The experimental results demonstrate: (a) the on-chip capacitive sensor is able to track the tilting motion with high accuracy; (b) the θ TH extracted from on-chip capacitive sensor can serve as an invariant reference point for scale factor calibration of MEMS gyroscopes.

This paper introduces a digital pulse width modulation technique for MEMS oven-control system. The proposed method generates digital, pulse width modulated signals from analog signals produced by a proportional-integral-derivative controller. This technique, therefore, provides good digital properties, such as better design scalability and higher power efficiency. These advantages are confirmed with an oven-control system for a commercial 6-axis inertial measurement unit. As a result, the oven-control system is fabricated on a single printed circuit board measuring 42mm×42mm×10mm and the power efficiency is increased to 94.4%. The entire system shows ±1.8ppm frequency drift during a thermal-cycle test, from −40°C to 85°C with the rate of the change 1°C/min.

Power gating is able to counter subthreshold leakage in low-power nanometer technology circuits without sacrificing performance. But mode transitions in power-gated circuits are accompanied by large inrush/discharge currents causing inductive bounce noise on the power supply and ground rails. This issue has been addressed by gradually turning on the sleep transistor; but this introduces a fixed lower bound on the delay overhead irrespective of the duration of the sleep period, and takes no account of the effects of changes in the circuit internal nodes during wake-up on the ground bounce noise. We observed the behavior of internal nodes during the sleep-to-active mode transition and identified three distinct stages. This motivates a three-step turn-on scheme and an associated compact power-gating structure that limits the current flowing through the sleep transistor only while the gated block is metastable, but quickly boosts the power supply rail when there are no short-circuit current paths in the logic. This strongly suppresses power gating noise, and also reduces wake-up time. Simulation results of 16-bit arithmetic logic units in 65-nm CMOS technology show that the proposed technique offers the advantage of a wake-up time that scales with the discharged value (during sleep) of the virtual power rail.

The performance of a commercial 6-axis (3-axis accelerometer and 3-axis gyroscope) MEMS inertial measurement unit (IMU) has been improved by a factor of >300x by utilizing a low-power ovenized microsystem. The IMU is thermally isolated from the external ambient by mounting it on a custom-designed micro-machined glass platform and packaging it in vacuum. In the present study, a microcontroller and voltage-controlled current source are assembled together with the thermally-isolated packaged IMU on a printed circuit board. The entire system is thermal-cycled over a temperature span from −40°C to 85°C. Both temperature control (i.e., ovenization) and compensation are used to reduce bias drift due to temperature change. The measured frequency drift of the IMU is improved by a factor of 950x and stabilized to ±2ppm, and the bias drift of the IMU is reduced to 60 °/hr for one of the gyroscope axes, and 1.7 mg for one of the accelerometer axes.

Low-dielectric-constant SiOC(–H) thin films were deposited on p -type Si(100) substrates using plasma enhanced chemical vapor deposition (PECVD) from vinyltrimethylsilane (VTMS; CH 2 = CHSi(CH 3 ) 3 ) and oxygen gas as precursors. To improve the structural, mechanical and electrical characteristics, SiOC(–H) films deposited using PECVD were post-treated by ultraviolet (UV) irradiation for various time intervals. Carbon content of the SiOC(–H) films increased before 240 s of UV irradiation time. But carbon-bonded functional groups of the SiOC(–H) film, in case of 480 s UV irradiation time, is replaced with Si–O bond. Because the Si–CH n bond groups are broken due to UV irradiation, Therefore, the films are formed with Si–O bond rich in the Si–O–C(–H) structure. The lowest relative dielectric constant, leakage current density, the elastic modulus and the hardness of SiOC(–H) with 240 s of UV irradiation time were about 2.07, 2.1 × 10 −7 A/cm 2 , 43 GPa, and 3.68 GPa, respectively. The results indicate that the SiOC(–H) films exposed by UV irradiation improve the structural, mechanical, and electrical characteristics.

In this brief, we present a time-to-digital converter (TDC) in which a single interpolator is used to improve the resolution by time stretching. The interpolator is based on a triple-slope conversion. Without slowing down the measured event, this approach extensively reduces the chip area and the corresponding power consumption, as compared with the prior arts with two parallel time interpolators. A prototype was designed and fabricated in a 0.35- μm CMOS digital process, and its core area merely occupies 0.126 mm2. Measurements show that our TDC achieves a resolution of 357 ps while consuming 1.22 mW with a 2.5-V supply. The dynamic range of the TDC exceeds 1.46 μs. The measurement rate can achieve above 400 kS/s.

Storage systems based on scanning probe microscopy incorporate an array of thermopiezoelectric cantilevers that require charge amplifiers with a wide frequency response but low power consumption. We use an improved frequency compensation method to enhance the in-band frequency response without requiring any additional power. Measurements show that the prototype chip draws the same 68 A at 1.8 V as an unmodified design but has a closed-loop frequency response that is ten times greater. The chip was fabricated in a 0.18-m standard CMOS process, and the charge amplifier core occupies 0.042 mm2.

Recent CMOS technology scaling has seriously eroded the bit-line noise immunity of register files due to the consequent increase in active bit-line leakage currents. To restore its noise immunity while maintaining performance, we propose and evaluate a 256×40-bit register file incorporating dual-V t bit-lines with a boosted gate overdrive voltage in 65 nm bulk CMOS technology. Simulation results show that the proposed bootsrapping scheme lowers leakage current by a factor of 450 without its performance penalty.

A simple method has been developed to suppress the DC term in in-line digital holographic microscope images. The technique is based on the use of the average intensity of a variable number of pixels rather than a fixed number of pixels. The DC-suppression ratio is similar to that associated with the averaging method proposed by Kreis and Juptner, while the image quality resulting from the new method is higher. This method can potentially yield significant improvements in the quality of reconstructed images.

Many customs inspection machines based on particle detector techniques have been researched and developed. An inspector with a proper scanning speed of an inspector will provide better efficiency. This report attempted to optimize the scanning parameters for container inspectors with segmented detectors for gamma rays from radioactive isotopes for custom inspections executed on higher-density objects hidden in imported iron containers. This calculation method provided good results for segmented detectors such as a resistive plate chamber (RPC), segmented photomultiplier tube (PMT), cathode stripe chamber (CSC), pixelized semiconductor detectors, etc. We estimate the proper scanning speed for the best combinations of the intensity of the gamma ray, the distance between the detector and the gamma ray source, the size of a pixel in the detector, the scanning speed, etc. to get a sufficient resolution. We used an RPC, whose pixel size is a few cm and whose gamma-ray detection efficiency is 0.18, as a reference detector for our calculation. Also we used Co emitting two kinds of gamma rays with kinetic energies of 1.17 MeV and 1.33 MeV, as a gamma ray source.

We present a power gating turn-on mechanism that digitally suppresses ground-bounce noise in ultra-deep submicron technology. Initially, a portion of the sleep transistors are switched on in a pseudo-random manner and then they are all turned on fully when VVDD is above a certain reference voltage. Experimental results from a realistic test circuit designed in 65nm bulk CMOS technology show the potential of our approach.

A 9-bit 110-MS/s pipelined-SAR ADC is proposed. To alleviate the design tradeoff between conversion rate and power consumption, the design adopts a voltage-mode open-loop amplifier and a time-interleaved SAR architecture with comparator sharing. The ADC simulated in a 65-nm CMOS technology achieves an ENOB of 8.63 bits near the Nyquist input frequency at the sampling rate of 110MS/s. The power consumption is 7.9mW, resulting in 181.3fJ/conversion-step of Figure of Merit (FoM).

Eu-doped In2O3 (EIO) thin films were deposited by radio-frequency magnetron sputtering on glass substrates with varying growth temperatures. All the EIO thin films showed a significant dependence on the growth temperature. From the figure of merit index data, the optimum growth temperature for depositing high-quality EIO thin films was found to be 300 degrees C. The ELO thin film deposited at 300 degrees C showed a highly preferential growth orientation along the (222) plane with an average particle size of 160 nm, bandgap energy of 3.94 eV, average optical transmittance of 65.2% in the wavelength range 450-1100 nm, and electrical resistivity of 2.5 x 10(-3) Ω cm. These results indicate that the electrical and optical properties of EIO thin films can be modulated by controlling growth temperature.

This paper presents the design and preliminary test results of a micro actuation and sensing platform to provide on chip physical stimulus for in situ calibration of long-term drifts in scale factor and offset of dual-axis MEMS gyroscopes. The platform consists of a 3-DOF micro vibratory stage that can provide piezoelectric actuation for X/Y-tilting reference stimuli, off-axis motion compensation, analog sensing of applied periodic stimulus, and electrostatic position lockdown for shock protection. Initial characterizations are performed with a commercial MEMS gyroscope (25 mg) mounted on top of the micro motion platform, while its electrical interconnects provided through microfabricated highly-flexible parylene cables. The piezoelectric stage is measured to provide up to 220 degree/sec angular AC excitation, while providing an analog sensing signal to determine the applied stimulus with a precision of 1.2 degree/sec. The estimated scale factor has

We present the fused-silica micromachined birdbath resonator gyroscope (μ-BRG) operating in the whole-angle (WA) mode. The key advantages of the whole angle mode operation is rotation angle measurement, large bandwidth, and full-scale range which is needed in detecting the motion of fast-moving objects. The μ-BRG is made with fused silica using a micro blow-torching process and has n = 2 wineglass modes at 10.46 kHz with a small frequency mismatch (Δf = 10 Hz) and a decay time (τ) of 2.2s. The WA-BRG achieves a stable angular gain (Ag) and a large full scale range (700°/s).

In this paper, we propose a time-to-digital converter (TDC) that uses a multi-phase cyclic Vernier delay line (VDL) to achieve the high-resolution and wide-dynamic range. Its control voltages are provided by two phase-locked loops (PLLs) to compensate for the process and ambient variations. The two PLLs share a single reference clock and have different frequency-division ratios. It also improves the resolution of the TDC. A prototype chip, designed and fabricated in 0.18μm CMOS technology with an active area of 0.40mm2, achieves a 3.4ps of resolution and an input range of 100ns at 2.5M samples/s, while consuming 32mW from a 1.8V supply.

We report a three-dimensional (3-D) micro-scale shell gyroscope made of fused silica (FS), called the birdbath resonator gyroscope (BRG). The BRG operates at 10.45 kHz, far above most environmental noise frequencies, and offers high mechanical quality factor (Qn=2 = 249 k) and low stiffness and damping anisotropy (Δfn = 2 = 14 Hz, Δ1/τn=2 = 4.1 mHz). The birdbath (BB) resonator can be fabricated with good structural symmetry because its anchor is self-aligned to the rest of the structure. The BB resonator also has large frequency separation between the n=2 wineglass modes and the closest parasitic mode (|fparasitic - fn=2|/fn=2 = 0.3), which makes the gyro less sensitive to vibration. The BRG is fabricated using a 3-D μ-blow-torching process and assembly on an electrode substrate made with the Silicon-On-Glass (SOG) process. The BRG is operated in the force-rebalance mode at 400 deg/sec, an angle random walk (ARW) of 0.106 deg/√hr and a bias stability of 1 deg/hr.

Our readout integrated circuit (ROIC) for differential capacitive sensors, such as thin-membrane transducer, uses current switching and time-domain based technique to measure the difference between the capacitance of the sensing and reference more rapidly, while maintaining accuracy. The 12-bit ROIC is designed and fabricated in a 0.35µm digital CMOS bulk technology.

This paper presents the design of a phase-locked loop (PLL) based clock and data recovery (CDR) circuit that meets fast locking and low jitter. We reduce the locking time of a CDR circuit using a new autonomously reconfigurable charge pump and loop filter in a 1.25 Gb/s CDR circuit. An experimental prototype was implemented in a 0.18 mum standard CMOS technology. A receiver that incorporates our CDR circuit has an active area of 380 mum times 350 mum.

— To comparatively evaluate various driving methods of an electronic display in respect to image sticking, a consistent and reliable quantification method is required. For proper evaluation, the entire area of a display is often monitored by using a chessboard pattern, and long-range gradual luminous variation in the background is eliminated. Estimation in terms of a single number is also preferred for simple comparison of image sticking. However, the prior method that uses the initial luminance for normalization and estimates the range-to-maximum ratio is not well-suited for the driving methods that relieve image sticking by restoring luminance uniformity. We have developed a method of extracting reference values for normalization and introduced the relative standard deviation (RSD) into our estimation. The resulting method is insensitive to the temporal change in the long-range gradual luminous variation and sufficiently indicative to allow driving methods to be compared effectively. The reference extraction method and the indicative capability of the RSD have been assessed by experiments using a real active-matrix organic light-emitting-diode (AMOLED) display cell.

A low-voltage bandgap reference (BGR) circuit is designed and fabricated in a 90 nm CMOS technology. To mitigate error resulting from the mismatch in temperature dependency of the current in the output current mirror device and that of the BGR core, an output-regulated current mirror is incorporated. Experimental results show that the output voltage is 497.2 mV at 25 ο C with a temperature coefficient of 28.3 ppm/ C between -40 ο C and 80 ο C. The circuit occupies 0.0337 mm 2 and dissipates 276.6 pW with a supply voltage of 1.2 V.

A low-complexity CMOS circuit is proposed for reading out monolithically integrated differential capacitive sensors. It directly converts the differential capacitance of a MEMS sensing device to a frequency by accumulating the charges produced by repeated charge integration and charge conservation. A prototype chip has been designed and fabricated in 0.35 μm CMOS technology. Experimental results show that differential capacitance is linearly converted to output frequency.

A new estimation method for image sticking has been developed that can effectively eliminate the influence by the difference in long-range non-uniformity pattern. An 8×8 chessboard pattern was used for selective degradation, and normalization based on extracted reference and coefficient of variation were adopted for better results.

— Early loss of image uniformity has been a critical drawback of active-matrix organic light-emitting-diode (AMOLED) displays operated in time-ratio gray-scale mode. This problem is addressed with an analog calibration technique which measures the voltage across each OLED for a given current and subsequently controls the supply voltage of pixels and the voltage drop across the driving th in-film transistor (TFT) of each OLED. The uniformity of test cells, which were aged to produce image sticking in a chessboard pattern, were improved. A measure of image sticking, called the extracted image-sticking value (EISV), was formulated, which is developed and used for the quantitative evaluation of the calibration method. OLED voltages over a range of about 0.35 V were compensated to produce more uniform OLED currents than those before aging. The variation of luminance associated with image sticking was reduced by about 40% for a full-white image after between 2 and 10 hours of accelerated aging with a constant voltage of 8 V across an OLED.

We propose a new hybrid pixel driving method that combines conventional digital driving with analog compensation for the declining electrical performance that occurs as OLEDs age. Image uniformity, and thus display lifetime, is improved by directly measuring the variation in OLED voltage at each pixel for a given OLED current level and then controlling the voltage drop across its driving TFT of an individual pixel.

An 8-bit coarse digital-to-analog converter (DAC), which adopts both array and tree-type decoders, is combined with a 2-bit fine interpolation DAC to reduce RC time delay and die area of a column driver for LCD-HDTV applications. Error amplifiers drive a pair of column lines in the output buffer to realize rail-to-rail voltage swing with a high slew rate. The design has been fabricated in 0.3 μm LV-HV CMOS technology.

A new fast-locking scheme is applied to a clock and data recovery (CDR) circuit based on a phase-locked loop. Locking time is reduced by using an autonomously reconfigurable charge pump and loop filter. A 1.25 Gbit/s prototype CDR circuit has been implemented in a 0.18 mum CMOS technology.