Microelectromechanical single-pole, double-throw (SPDT) switches with predominantly planar configurations have been proposed for use in radio-frequency (RF) circuits. These switches would have overall dimensions of the order of a few millimeters and would be fabricated by micromachining techniques compatible with those used to fabricate microelectronic circuits; as such, these switches would be attractive for incorporation into inexpensive, highly integrated electronic circuits. Moreover, because of their fundamentally electromechanical character and in contradistinction to semiconductor switches, these switches are expected to exhibit minimal intermodulation distortion and very low power dissipation.

The simplest switch of this type (see Figure 1) would be an assembly of (1) a T junction between an input signal path and two output signal paths and (2) two electrostatically actuated microscopic reed contacts- one on each output path - close to the junction. In a normal operational condition, one reed contact would be closed ("on") and the other open ("off"). In this condition, the switch would provide an excellent dc and low-frequency block between the input line and the "off" output line, and an excellent conductive path between the input line and the "on" output line. However, as determined in a computational simulation, the isolation between the input line and the "off" output line at a frequency of 30 GHz would be only about 28 dB - a level that is considered mediocre in the industry.

Figure 1. Microelectromechanical SPDT Switches configured generally like this one would offer advantages over semiconductor switch

The placement of a second reed contact directly after the first on the output lines would improve the isolation, but only by about 15 dB across most of the band from dc to the maximum design frequency (e.g., 30 GHz). A much more significant increase can be obtained by placing that second contact at a quarter wavelength of 30 GHz; this would yield an improvement over the single contact case of more than 30 dB across the entire band.

The increase in isolation at 30 GHz can clearly be attributed to the impedance transformation of the second-contact open circuit to a short directly after the first contact. However, the high isolation across the rest of the band is the result of the unique circuit created by the topology. At frequencies between dc and the maximum design frequency, the open-circuited contacts and section of transmission line between them act as two very small capacitors in series with a low-pass filter between them. This low pass filter characteristic provides the high isolation at the intermediate frequencies.

Figure 2. Multiple Contacts at Properly Designed Intervals would provide superior isolation between the input line and the "off"-side output line.

Following similar reasoning, one could increase the isolation further by placing a third set of contacts at a quarter wavelength from the second contact (see Figure 2). Yet rather than placing the third contact at a quarter wavelength of the maximum fundamental frequency, if one desires to suppress second harmonic leakage, both the second and third contacts can be placed at a quarter wavelength of the second harmonic frequency to achieve an ultra-broadband, low-loss SPDT switch. This can be done with minimal effect on the performance at the fundamental frequency. For the dc-to-30-GHz case, computer simulations reveal 84 dB isolation at 30 GHz and 68 dB at 60 GHz; insertion loss at 30 GHz is predicted to be about 0.5 dB. This level of performance would exceed any other known RF SPDT switch.

This work was done by Sam Valas of Caltech for NASA's Jet Propulsion Laboratory.

In accordance with Public Law 96-517, the contractor has elected to retain title to this invention. Inquiries concerning rights for its commercial use should be addressed to

Technology Reporting Office
Mail Stop 122-116
4800 Oak Grove Drive
Pasadena, CA 91109
(818) 354-2240

Refer to NPO-20663