[1993] Proceedings IEEE Micro Electro Mechanical Systems
DOI: 10.1109/memsys.1993.296921
|View full text |Cite
|
Sign up to set email alerts
|

Implementation of MEMCAD system for electrostatic and mechanical analysis of complex structures from mask descriptions

Abstract: We report the development of a first implementation of the MEMCAD system (version 1.0). The system is composed of three commercial mechanical CAD software packages integrated with our specialize structure generation and electrostatic analysis programs. In this paper we describe the system and demonstrate its capabilities using a comb drive example constructed directly from a CIF description of its mask set. We received the CIF file from U.C.Berkeley [l]. The analysis of the comb drive takes just a few hours, a… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
1
1

Citation Types

0
19
0

Publication Types

Select...
3
2
2

Relationship

2
5

Authors

Journals

citations
Cited by 34 publications
(19 citation statements)
references
References 9 publications
0
19
0
Order By: Relevance
“…The method used to develop these models is extendable to other test-structure geometries. Each model is ultimately based on detailed 3-D numerical quasi-static self-consistent simulation of the deformation of the test structure under the combined effect of linear elastic forces and nonlinear electrostatic forces, using the MIT MEMCAD system [10], [11]. However, because the M-Test structures are highly symmetric, indeed, nearly 2-D, it was determined that a much simpler 2-D finite-difference model, initially reported in [13] (with some important typographical errors, which are corrected here), was sufficiently precise in comparison with 3-D simulation to permit its use over a wide design space.…”
Section: B Closed-form Pull-in Modelsmentioning
confidence: 99%
See 1 more Smart Citation
“…The method used to develop these models is extendable to other test-structure geometries. Each model is ultimately based on detailed 3-D numerical quasi-static self-consistent simulation of the deformation of the test structure under the combined effect of linear elastic forces and nonlinear electrostatic forces, using the MIT MEMCAD system [10], [11]. However, because the M-Test structures are highly symmetric, indeed, nearly 2-D, it was determined that a much simpler 2-D finite-difference model, initially reported in [13] (with some important typographical errors, which are corrected here), was sufficiently precise in comparison with 3-D simulation to permit its use over a wide design space.…”
Section: B Closed-form Pull-in Modelsmentioning
confidence: 99%
“…This work presents: 1) the development of closed-form quantitative models for the pull-in of M-Test structures derived from two-dimensional (2-D) and three-dimensional (3-D) numerical simulations using the Massachusetts Institute of Technology's software package for the computer-aided design of microelectromechanical systems (MIT MEMCAD) [10], [11]; 2) an experimental procedure and associated data-reduction method, which removes geometrically correlated statistical variation in order to improve the precision of the results and uses geometric data on the test-structure dimensions (such as beam thickness and undeformed gap) to extract material properties from the pull-in data; and 3) experimental verification of the M-Test method using data from MIT's dielectrically isolated single-crystal silicon wafer-bonded process [12].…”
Section: Introductionmentioning
confidence: 99%
“…For this reason, there has been a steady effort over the last decade to develop nearly automatic approaches for generating accurate macromodels of micromachined devices starting from only layout and process descriptions. Most efforts is this area are following a three step approach [18], [11], [19].…”
Section: Numerical Macromodelingmentioning
confidence: 99%
“…Finally, the electrostatic force is approximated assuming nearly parallel plates and is given by where is the applied voltage. Spatial discretization of (17) and (18) leads to a large nonlinear system of the form (19) where is an -length state vector, in this case the vector of displacements and their time derivatives. The function , which maps an -length vector to an -length vector, represents the spatially discretized partial differential equation.…”
Section: B Model-order Reductionmentioning
confidence: 99%
See 1 more Smart Citation