Numerical modeling and validation of squeezed-film damping in vacuum-packaged industrial MEMS

Wajih U. Syed, Ayoola Brimmo, Owais Waheed, Alabi Bojesomo, Mohamed Hassan Ali, Ilker Ocak, Sun Chengliang, Aveek Chatterjee, Ibrahim M. Elfadel

Research output: Contribution to journalArticlepeer-review

9 Scopus citations

Abstract

Several high-performance, industrial micro-electromechanical (MEM) devices, such as gyroscopes, magnetometers, high-Q resonators and piezoelectric energy harvesters, require wafer bonding and packaging under near-vacuum conditions. One very challenging aspect of the design, verification and characterisation of these devices is to predict their performance characteristics in the presence of any residual gases post-packaging. Such gases contribute to the energy losses resulting from device surfaces squeezing or sliding against the gas films within the device cavities. In this paper, we fully expose the modelling assumptions used in commercial FEM tools to estimate the squeezed-film damping (SFD) experienced by MEM devices that are packaged under near-vacuum conditions. We also explain the various meshing options to enable the extraction of the most accurate Q factors under existing SFD assumptions. In addition, we compare the computational results across a variety of commercial FEM codes against measurements obtained under realistic vacuum conditions for an industrial high-Q magnetometer. These measurements suggest that existing computational models may deviate by as much as 25% on Q factor values for gas flow regimes under operating cavity pressures of less than 1 Torr.

Original languageBritish English
Article number075016
JournalJournal of Micromechanics and Microengineering
Volume27
Issue number7
DOIs
StatePublished - 8 Jun 2017

Keywords

  • FEM
  • MEMS
  • MEMS magnetometer
  • MEMS process variations
  • squeezed-film damping
  • vacuum-packaging

Fingerprint

Dive into the research topics of 'Numerical modeling and validation of squeezed-film damping in vacuum-packaged industrial MEMS'. Together they form a unique fingerprint.

Cite this