Abstract: A micro-electro-mechanical (MEMS) switch (10, 110) has an electrode (22, 122) covered by a dielectric layer (23, 123), and has a flexible conductive membrane (31, 131) which moves between positions spaced from and engaging the dielectric layer. At least one of the membrane and dielectric layer has a textured surface (138) that engages the other thereof in the actuated position. The textured surface reduces the area of physical contact through which electric charge from the membrane can tunnel into and become trapped within the dielectric layer. This reduces the amount of trapped charge that could act to latch the membrane in its actuated position, which in turn effects a significant increase in the operational lifetime of the switch.

Abstract: A micro-electro-mechanical (MEMS) switch (10, 110) has an electrode (22, 122) covered by a dielectric layer (23, 123), and has a flexible conductive membrane (31, 131) which moves between positions spaced from and engaging the dielectric layer. At least one of the membrane and dielectric layer has a textured surface (138) that engages the other thereof in the actuated position. The textured surface reduces the area of physical contact through which electric charge from the membrane can tunnel into and become trapped within the dielectric layer. This reduces the amount of trapped charge that could act to latch the membrane in its actuated position, which in turn effects a significant increase in the operational lifetime of the switch.

Abstract: A micro-electro-mechanical (MEMS) switch (10, 110) has an electrode (22, 122) covered by a dielectric layer (23, 123), and has a flexible conductive membrane (31, 131) which moves between positions spaced from and engaging the dielectric layer. At least one of the membrane and dielectric layer has a textured surface (138) that engages the other thereof in the actuated position. The textured surface reduces the area of physical contact through which electric charge from the membrane can tunnel into and become trapped within the dielectric layer. This reduces the amount of trapped charge that could act to latch the membrane in its actuated position, which in turn effects a significant increase in the operational lifetime of the switch.

Abstract: An optical matrix switch station (1) is shown mounting a plurality of optical switch units (15, 17), each of which includes a mirror (29), moveable in two axes, for purpose of switching optical beams from one optical fiber to another. A mirror assembly (41) includes a single body of silicon comprising a frame portion (43), gimbals (45), mirror portion (47), and related hinges (55). Magnets (53, 54) and air coils (89) are utilized to position the central mirror surface (29) to a selected orientation. The moveable mirror and associated magnets along with control LED's (71) are hermetically packaged in a header (81) and mounted with the air coils on mounting bracket (85) to form a micromirror assembly package (99) mounted in each optical switch unit.

Abstract: An optical matrix switch station (1) is shown mounting a plurality of optical switch units (15, 17), each of which includes a mirror (29), moveable in two axes, for purpose of switching optical beams from one optical fiber to another. A mirror assembly (41) includes a single body of silicon comprising a frame portion (43), gimbals (45), mirror portion (47), and related hinges (55). Magnets (53, 54) and air coils (89) are utilized to position the central mirror surface (29) to a selected orientation. The moveable mirror and associated magnets along with control LED's (71) are hermetically packaged in a header (81) and mounted with the air coils on mounting bracket (85) to form a micromirror assembly package (99) mounted in each optical switch unit.

Abstract: An optical matrix switch station (1) is shown mounting a plurality of optical switch units (15, 17), each of which includes a mirror (29), moveable in two axes, for purpose of switching optical beams from one optical fiber to another. A mirror assembly (41) includes a single body of silicon comprising a frame portion (43), gimbals (45), mirror portion (47), and related hinges (55). Magnets (53, 54) and air coils (89) are utilized to position the central mirror surface (29) to a selected orientation. The moveable mirror and associated magnets along with control LED's (71) are hermetically packaged in a header (81) and mounted with the air coils on mounting bracket (85) to form a micromirror assembly package (99) mounted in each optical switch unit.

Abstract: An optical matrix switch station (1) is shown mounting a plurality of optical switch units (15, 17), each of which includes a mirror (29), moveable in two axes, for purpose of switching optical beams from one optical fiber to another. A mirror assembly (41) includes a single body of silicon comprising a frame portion (43), gimbals (45), mirror portion (47), and related hinges (55). Magnets (53, 54) and air coils (89) are utilized to position the central mirror surface (29) to a selected orientation. The moveable mirror and associated magnets along with control LED's (71) are hermetically packaged in a header (81) and mounted with the air coils on mounting bracket (85) to form a micromirror assembly package (99) mounted in each optical switch unit.

Abstract: A substantial portion of the material at the pn junction (27) of the photodiode (37, 41) having an implanted region extending to a surface thereof is selectively removed (39), leaving a very small junction region (35, 43) with the remainder of the p-type (23) and n-type (25) material of each photodiode being spaced apart or electrically isolated at what was originally the junction. In the ion implanted n-type on p-type approach, the majority of the signal is created in the implanted n-type region while the majority of the noise is generated in the p-type region. By selectively removing p-type material, n-type material or both from the pn junction of the diode or otherwise electrically isolating most of the p-type and n-type regions from each other at the pn junction and thereby minimizing the pn junction area, noise is greatly reduced without affecting the signal response of the photodiode.

Abstract: A substantial portion of the material at the pn junction (27) of the photodiode (37, 41) having an implanted region extending to a surface thereof is selectively removed (39), leaving a very small junction region (35, 43) with the remainder of the p-type (23) and n-type (25) material of each photodiode being spaced apart or electrically isolated at what was originally the junction. In the ion implanted n-type on p-type approach, the majority of the signal is created in the implanted n-type region while the majority of the noise is generated in the p-type region. By selectively removing p-type material, n-type material or both from the pn junction of the diode or otherwise electrically isolating most of the p-type and n-type regions from each other at the pn junction and thereby minimizing the pn junction area, noise is greatly reduced without affecting the signal response of the photodiode.

Abstract: Only the areas of the CdTe/HgCdTe interface of a FPA detector circuit which is coupled by an epoxy to a silicon-based integrated circuit that require interdiffusing are heated to a sufficiently high temperature or have photons of light impinging thereon for a sufficient time to cause interdiffusion of the two layers by the travel of tellurium into the HgCdTe and the travel of mercury into the CdTe. The vast majority of the wafer is masked with an aluminum thin film to greatly reduce heat gain or photon transmission. An advantage of the process in accordance with the present invention is that only a very small fraction of the HgCdTe/epoxy/silicon-based integrated circuit wafer receives incoming energy during interdiffusion whereby problems caused by the differences in coefficient of thermal expansion between silicon and HgCdTe at the epoxy interface are minimized.

Abstract: Only the areas of the CdTe/HgCdTe interface of a FPA detector circuit which is coupled by an epoxy to a silicon-based integrated circuit that require interdiffusing are heated to a sufficiently high temperature or have photons of light impinging thereon for a sufficient time to cause interdiffusion of the two layers by the travel of tellurium into the HgCdTe and the travel of mercury into the CdTe. The vast majority of the wafer is masked with an aluminum thin film to greatly reduce heat gain or photon transmission. An advantage of the process in accordance with the present invention is that only a very small fraction of the HgCdTe/epoxy/silicon-based integrated circuit wafer receives incoming energy during interdiffusion whereby problems caused by the differences in coefficient of thermal expansion between silicon and HgCdTe at the epoxy interface are minimized.