MODELLING AND SIMULATION OF MICROACCELEROMETER
YAO YUFENG
YAO YUFENG
Citations
Altmetric:
Alternative Title
Abstract
Micro-accelerometer systems have become more common due to the increasing demands in the market. In order to reduce the cost and shorten the development cycle, there is a need for modelling and simulation of micro-accelerometer. To realise this
goal, this thesis presents a work on modelling and simulation of micro-accelerometer; and provides an application to verify the works developed. This thesis analyses five spring configurations and derives the spring constants using energy method. The spring configurations include fixed-fixed flexure, folded flexure, "U" folded flexure, extended folded flexure and extended "U" flexure. The natural frequency is also derived using Rayleigh's energy method. The equivalent mass which includes the spring mass and proof mass is obtained. The dynamic response of the micro-accelerometer shows that the motion of the device is
proportional to the motion of the base. The acceleration which the accelerometer experienced is approximately equal to the acceleration of base. The electrical modelling of the sensing elements which detect the capacitance is another important part developed in the thesis. This thesis analyses the differential capacitance, electrostatic force, electrical spring constant, resolution, sensitivity and ratio metric error. It also considers the effect of the fringe fields. The deformation of the fixed fingers is so large that it should not be neglected. The electromechanical modelling presents the model with mechanical and electrostatic coupling.
The macro-model developed for the micro-accelerometer provides a method to simplify the model and reduce the order of the dynamic equations by using Galerkin's method. This study makes it possible to simplify the dynamic models. A low-g micro-accelerometer has been developed as an application to verify the modelling and simulation results. It has been successfully fabricated and tested, and the test results agree with the design. From the test, we have obtained the results of
natural frequency, sensitivity and ratio metric error of the fabricated microaccelerometer as 1.69kHz, 36.2fF/g and 4.5% respectively. Comparing with the estimated natural frequency (2.04kHz), sensitivity (35.2fF/g) and ratio metric error
(4.99%) of the micro-accelerometer, the error is 17.1 %, 2.84% and 9.82% respectively. These errors are mainly caused by the dimensional errors of the fabricated microaccelerometer. This shows that modelling and simulation of micro-accelerometer
developed in this thesis is valid and effective.
Keywords
Source Title
Publisher
Series/Report No.
Collections
Rights
Date
2000
DOI
Type
Thesis