Article

Higher Order Noise-Shaping Filters for High-Performance Micromachined Accelerometers

Southampton Univ., Southampton
IEEE Transactions on Instrumentation and Measurement (Impact Factor: 1.71). 11/2007; DOI: 10.1109/TIM.2007.904477
Source: IEEE Xplore

ABSTRACT Micromachined inertial sensors that have been incorporated in sigma-delta force-feedback loops have been proven to improve linearity, dynamic range, and bandwidth, and also provide a direct digital output. Previous work mainly focused on using the sensing element only to form a second-order single-loop sigma-delta modulator (SigmaDeltaM); however, the advantages of higher order single-loop electromechanical SigmaDeltaM have not been fully explored. High-performance inertial sensors require higher signal-to-quantization noise ratio (SQNR). This paper presents topologies for higher order single-loop electromechanical SigmaDeltaM with optimal stable coefficients that lead to a better SQNR. The topologies have good immunity to fabrication tolerances, which was verified by Monte Carlo analysis. The topologies are applicable not only to accelerometers but also to other inertial sensors such as gyroscopes.

0 Bookmarks
 · 
122 Views
  • [Show abstract] [Hide abstract]
    ABSTRACT: High performance MEMS gyroscopes can be implemented by embedding in a closed loop controller. For these gyroscopes, the electrostatic controlled force on the proof mass is inherently nonlinear. This study reports a MEMS gyroscope with control capability in which comb drive electrodes are employed as the controlling mechanism to linearize the electrostatic controlled force. To avoid interaction of drive and control comb drive electrodes, another mass has been added to the structure. Consequently, proposed gyroscope consists of a 2-DOF sense mode oscillator and a 1-DOF drive mode oscillator. The prototype with an overall size of 0.46 × 0.635 mm is designed to be fabricated in surface Micromachining process with 3 μm structural layer thickness. The device operation has verified by Finite Element Method simulations showing a raw mechanical sensitivity of 32 nm for drive direction oscillation amplitude of 120 nm and angular rotation of 50 [°/s].
    Electrical Engineering (ICEE), 2013 21st Iranian Conference on; 01/2013
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: The major challenge of multi-domain application design is that the different domains often closely interact with each other. So a separate design of each component and afterwards combining them to the application often does not result in the expected application function. Application modeling at transistor level is not feasible in terms of simulation duration and data handling. This work proposes a new approach for Application Hardware Modeling which addressing the risks of performance degradation caused by interferences between the components through modeling component performances and through virtually validating the application performance in the system environment. The simulations consider all the relevant components that could have an impact on the performance at application level to predict them. The fundamental of Application Hardware Modeling for right-on-first-pass applications is to select the relevant models for a given component, along with the creation and validation of models through equivalence checking, which are presented in this article.
    Semiconductor Conference Dresden-Grenoble (ISCDG), 2012 International; 01/2012
  • [Show abstract] [Hide abstract]
    ABSTRACT: In this paper a variety of innovative control systems for MEMS inertial sensors are discussed. These control interfaces have considerable potential to improve important characteristics of MEMS inertial sensors, such as linearity, bandwidth, dynamic range and susceptibility again fabrication tolerances. All control systems are based on the incorporation of the micromachined sensing element in a sigma-delta modulator loop forming an electro-mechanical sigma-delta modulator (EMSDM). A very effective design methodology for such EMSDM based on a Genetic Algorithm is introduced. Subsequently, a high order, single loop EMSDM and a Multi Stage Noise Shaping (MASH) EMSDM are discussed for an accelerometer. For MEMS gyroscopes, results from a bandpass EMSDM and a novel quadrature EMSDM are presented.
    Frequency Control Symposium (FCS), 2012 IEEE International; 01/2012