Electricity: electrical systems and devices – Control circuits for electromagnetic devices – For relays or solenoids
Reexamination Certificate
2001-03-27
2003-12-23
Jackson, Stephen W. (Department: 2836)
Electricity: electrical systems and devices
Control circuits for electromagnetic devices
For relays or solenoids
C361S152000
Reexamination Certificate
active
06667873
ABSTRACT:
BACKGROUND OF INVENTION
The present invention is in the field of reconfigurable wiring manifolds, and in particular, relates to electrical manifolds whose wiring patterns can be electrically reconfigured using an arrangement of pre-fabricated interconnection assemblies and switchbox devices employing multi-stable MEMS switches.
A wiring assembly is a multi-conductor configuration, which is used within an electronics circuit or between electronic circuits to form electrical power and signal connections. Simple examples include an ordinary printed wiring board (such as the motherboard of a personal computer), a cable (such as the printer cable in a computer or more elaborately a cable connecting two electronics boxes in an aircraft), or even an electrical package (which routes external pin terminal connections to the bond pad terminals of components within the package). Such wiring configurations are usually completely fixed, although it is possible to use jumpers or mechanical switches on the assembly to change a few connection patterns. It is also possible to rectify or modify connections in an assembly crudely, by physically severing wires or adding jumper wires to break or make, respectively, a new desired connection.
Relays have of course existed for well over one hundred years. The variety of relay referred to as “latching” have the property of remaining in an open or shorted (“off” or “on”) state until changed, and they were the basis of the first crude memory devices. These devices can in fact be used to electrically reconfigure a wiring pattern, and they have been used for this purpose, but the traditional relays are bulky and cumbersome in this role. Hence, outside of telecommunications applications and power switching applications in which large cabinets can be dedicated to a mass switching function, the notion of using reconfigurable wiring harnesses within embedded or portable systems has not been pursued seriously.
An adaptive manifold involves a multi-conductor electrical wiring assembly, e.g., a package, board, or harness, with one or more programmable connections. Such a wiring assembly can be altered into a new wiring pattern without physically moving the wires themselves. The intent is to permit this to be done under automated and even remote control, to effect the ability to re-wire a system in situ.
The primary reason no adaptive manifolds have been built before is that relays are too cumbersome. Due to this impracticality, little thought has been given as to how it might be possible to exploit hundreds or thousands of latching relays within packages, circuit boards, and wiring harnesses. If one were to assume that it were possible to build a relay the size of a human hair, for example, then one might further contemplate schemes for exploiting such devices, which is essentially the original basis leading to the proposed invention. From this standpoint, one can exploit the rich basis of knowledge of graph theory and graph routing problems, which is the basis of field programmable gate arrays (FPGAs). The routing/switch-box problems are very complex computationally (referred to as non-deterministic polynomial time or “NP”-complete), but a number of excellent algorithmic approximations have been studied. In the design of FPGAs, transistors are used as switches, and a great deal of thought has been given as to how to arrange wires and switches to provide the greatest incremental flexibility. This field of endeavor is diametrically opposed to the large relays and wiring harnesses in that the dimensions are vastly smaller, and the number of devices involved are large (millions). Similarly, though, little thought has been given to the application of these techniques outside of a field programmable gate array, which is a monolithic piece of silicon, to an ordinary wiring harness.
Previous realizations of this concept of adaptive manifolds involved using transistors (FETs) as the switch elements. There are a number of disadvantages to this approach. The logical configuration of the manifold is lost when power is removed from the manifold. The insertion losses associated with a FET switch can be large, e.g., the on resistance can be high (~1 k&OHgr;). Similarly the coupling losses associated with a FET switch manifold are large since the gate channel circuit used to turn the FET on and off must be physically adjacent to the source and drain circuits of the FET. This results in coupling between the switched circuit and the switching circuit. In addition, for the gate voltage to control the channel resistance, the absolute value of the gate voltage must be near the voltages at the source and drain contacts. As the FET functions by controlling field penetration from the source to the drain, near means the absolute value of the gate voltage is near the absolute voltage of either the drain or source circuit.
It is one object of the present invention to overcome these disadvantages by employing MEMS switches. A MEMS switch is a mechanical switch that is truly bistable, once set, it is independent of external power. The on resistance of a MEMS switch is 5 orders of magnitude smaller (~10 m&OHgr;) than that of an FET. Furthermore, the control circuit of the MEMS switch is physically remote/removed from the switched circuit, which reduces coupling to a minimum. In the MEMS implementation of an adaptive manifold, the control circuit is both physically and electrically isolated from the switched circuit. This allows the control voltages to be orders of magnitude different from the switched voltages. It also allows the chaining of successive switches without limitation due to the absolute value of the switching voltages relative to the switched voltages.
SUMMARY OF INVENTION
The adaptive manifold of the present invention is comprised of: (a) one or more bistable MEMS switches on one or more integrated circuit chips; (b) a monolithically integrated or off-chip control component set that can program any of the switches; and (c) a means of programming any switch using commands issued on a serial-port based boundary scan method. In addition, this adaptive manifold can be extended by adding additional copies (identical copies or variants of this configuration) programmed through the same single port as the first device by daisy-chaining the boundary scan chain. Included is a means of verifying (reading) the configuration of each switch in the entire array through commands issued on the (c) interface. Alternatively, the control interface can be a bussed or multi-drop configuration wherein the control lines are in parallel for each device.
REFERENCES:
patent: 5757319 (1998-05-01), Loo et al.
patent: 5852617 (1998-12-01), Mote, Jr.
patent: 6191663 (2001-02-01), Hannah
Broyles Ren H.
Lyke James C.
Wilson Warren G.
Callahan Kenneth E.
Demakis James A
Jackson Stephen W.
Skorich James M.
The United States of America as represented by the Secretary of
LandOfFree
Adaptive manifold does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Adaptive manifold, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Adaptive manifold will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3138298