User login

On-off control for full-gap positioning of parallel-plate electrostatic MEMS

Publication Type:

Conference Paper


Proceedings of IMECE 2006, 2006 ASME International Mechanical Engineering Congress and Exposition, Nov. 5-10, Chicago, USA, ASME, Chicago, Illinois, USA (2006)




Technical Publication

On-Off Control for Full-Gap Positioning of Parallel-Plate Electrostatic


Lukas Mol, Delft University of Technology

Luis A. Rocha, University of Minho

Edmond Cretu, University of British Columbia

Reinoud F. Wolffenbuttel, Delft University of Technology


Electrostatic parallel-plate actuation is classically limited to
displacements up to 1/3 of the gap due to the pull-in effect [1,2].
This limits the use of electrostatic parallel-plate actuation in
many applications. In order to overcome this limitation several techniques
have been proposed: geometry leverage [3], series feedback capacitor
[4], current drive methods [5,6] and closed-loop voltage control
[7]. Stable displacement over the full available range has not been
achieved with these approaches, except for the geometry leverage
technique, which is limited by the higher voltage levels required
and the larger dimensions.

The method presented in [8] is recently introduced to overcome these
difficulties. It is based on the comparison of the measured momentary
actuator displacement with the desired displacement. The voltage
applied to the actuator is changed between two values (unlike traditional
feedback): between a high level, if the measured displacement is
lower than the reference, and a low level, if the actuator displacement
is higher than the reference value.

The method relies strongly on the nonlinear dynamic behavior of the
MEMS devices. By changing the applied voltage a shift occurs between
stable and unstable trajectories. This implies that the device must
be overdamped or critically damped in order not to be in the oscillatory
operating regime. At very small gaps (i.e. large displacements),
the damping forces are huge due to the rarefaction effects. These
damping forces slow even further the structure displacements, improving
the dynamic device response when operated with the on-off method.
Due to the feedback topology, there is a finite time delay that results
in a small ripple of the electrode displacement. Two important parameters
are available for reducing the remaining ripple. Although the voltage
levels are not critical for proper operation, adjusting the high
and low level is affecting the device response and can be used to
improve the performance. Furthermore the feedback loop delay must
be reduced as much as possible, for example by integrating the displacement
measurement circuits with the actuator.

Careful selection of device geometry, gas type, pressure and actuation
voltage levels, can yield greatly reduced position ripple, while
maintaining bandwidth and quick response time. This could effectively
reduce an existing device size with a factor three for the same operating
displacement range.

References [1] H.A.C. Tilmans, and R. Legtenberg, Sensors and Actuators,
A 45 (1994) 67-84. [2] L.A. Rocha, E. Cretu and R.F. Wolffenbuttel,
J. Microelectromech. Syst. 13 (2004) 342-354. [3] E.S. Hung and S.D.
Senturia, J. Microelectromech. Syst. 8 (1999) 497-505. [4] E.K. Chan
and R.W. Dutton, J. Microelectromech. Syst. 9 (2000) 321-328. [5]
R. Nadal-Guardia, A. Deh{\'e}, R. Aigner and L.M. Casta{\~n}er, J. Microelectromech.
Syst. 11 (2002) 255-263. [6] J.I. Seeger and B.E. Boser, J. Microelectromech.
Syst. 12 (2003) 656-671. [7] Y. Sun, D. Piyabongkarn, A. Sezen, B.J.
Nelson and R. Rajamani, Sensors and Actuators, A 102 (2002) 49-60.
[8] L.A. Rocha, E. Cretu and R.F. Wolffenbuttel, J. Micro-electromech.
Syst., 15 (2006), pp. 69-83.

Session: MEMS-10 Micro and Nano Sensors and Actuators