Abstract
A methodology is developed statistically to make MEMS devices robust to process variations to improve manufacturability and yield. Two approaches are applied to discuss the effects of multi-process variations. Comparisons have been made between the proposed method and Monte Carlo simulations, which confirm the robustness of the proposed one with performance error less than 4%. Experiments on beams and comb-drive resonator verified the effectiveness of the methodology and it is useful for practical device designs to be more robust to process variations and yield enhancement realization.
Publisher
Trans Tech Publications, Ltd.
Subject
Mechanical Engineering,Mechanics of Materials,General Materials Science