SUGAR: The SPICE for MEMS
Professor Jason V. Clark, Electrical and Computer Engineering and Mechanical Engineering Depts., Purdue University
In this seminar, I will present some design, modeling, and simulation features of a computer aided engineering tool for microelectromechanical systems (MEMS) called SUGAR.
Results: For a test case, I will use a microdevice that is difficult to simulate with conventional MEMS software. I will show that the relative errors of the lumped models are less than 3% of finite element analysis; that the computational costs are much less than 1% of finite element analysis; and that simulation of the test case fairly agrees with experiment.
Attributes: Features of SUGAR include: a flexible SPICE-like netlist language for MEMS design; a simple modeling framework for computationally efficient lumped models; an extensible architecture to which users can add features; and the ability to display 3D circuits together with deflected electromechanical structures. Since SUGAR is programmed in MATLAB, many MATLAB functions and toolboxes may be used with SUGAR. Such attributes facilitate the exploration of design spaces and feature modifications.
Last, I will discuss a new direction toward modeling and simulating constrained multidisciplinary systems micro/nanosystems.
Jason Clark is an Assistant Professor of Electrical and Computer Engineering, and of Mechanical Engineering, at Purdue University. He is a faculty member of the Birck Nanotechnology Center, and of the Network for Computational Nanotechnology. Dr. Clark received his Ph.D. in Applied Science from the University of California at Berkeley, fall 2005. His CAD/E efforts lead to one of the first nodal analysis software packages for MEMS, called SUGAR (i.e. a SPICE for MEMS). And his micro-metrology efforts lead to the first comprehensive on-chip methods for measuring micro-scale geometry and material properties, called electro micro-metrology (EMM). Prior to professorship, he held positions at Lawrence Livermore National Lab, Berkeley National Lab, Berkeley Biomedical Microdevices Center, Berkeley Sensor and Actuator Center, and Coventor.