Browsing by Author "Chen, He-Ling"

Now showing 1 - 2 of 2

A Novel Suspension Design for MEMS Sensing Device to Eliminate Planar Spring Constants Mismatch
(2012)
Jiang, Kai-Yu
;
Chen, He-Ling
;
Hsu, Wensyang
;
Lee, Yueh-Kang
;
Miao, Yen-Wu
This paper presents a new suspension design for planar MEMS sensing devices. The new design includes four coil springs which are symmetrically arranged with respect to the center of the proof mass with 90-degree orientation difference. The proposed coil spring design is found to take less area to achieve similar spring constant than the conventional serpentine spring design. Furthermore, the proposed symmetrical-to-center arrangement on suspension can reduce spring constants mismatch between X and Y axes, even under beam width deviation due to fabrication resolution. This concept is realized and verified by experimental results from two-axis capacitive accelerometers through a polysilicon-based CMOS fabrication platform.
An analytical model for calculating the pull-in voltage of micro cantilever beams subjected to tilted and curled effects
(2014-08-01)
Huang, Yuan-Te
;
Chen, He-Ling
;
Hsu, Wensyang
;
機械工程學系
;
Department of Mechanical Engineering
Pull-in is a fundamental phenomenon in electrostatic micro devices. In previous studies on modeling the pull-in voltage of suspended micro cantilever beam subject to residual stress, only curled deformation was considered. This study proposed a modified deformation function, which considered both curled and tilted deformations caused by gradient stress and mean stress, to calculate the pull-in voltage of the suspended cantilever beam with residual deformations. In order to verify the proposed analytical model, suspended poly-silicon cantilever beams with three different lengths, 260 mu m, 295 mu m and 330 mu m, are fabricated through surface micromachining process. It is shown that the residual deformations include both curled and tilted deformations, where the tilted angle and radius of curvature can be identified by white light interferometer (WLI). By comparing the analytical results with measurement results on pull-in voltages, it is found that while only considering curled effect, the average error of calculated pull-in voltage is 10.5%. On the other hand, when both tilted and curled effects are considered, the average error is reduced to 3.2%, which verifies the accuracy improvement of the proposed analytical model. (C) 2014 Elsevier B.V. All rights reserved.