Abstract:
A standard package is provided in which circuit components (10) and non-integrated circuit components (20, 30, 40) are electrically connected together and then overmoulded (70). The components may be coupled together on a lead frame (60). The standard package may provide a desired electrical function by itself or it may be coupled together with other overmoulded packages or additional non-overmoulded electronic components.
Abstract:
An assembly of two or more microelectronic parts, wherein electrical and/or thermal interconnection between the parts is achieved by means of multiple, discrete, conductive nanoscopic fibrils (15) or tubules (15) fixed within the pores of an insulating film (16). Such a film is said to have anisotropic electrical conductivity, i.e., Z-axis conductivity, with little or no conductivity in the other directions.
Abstract:
An improved method of affixing spheres (4) to a foil matrix (2) is described herein. First, a cell sandwich (32) is prepared. This cell sandwich (32) includes spheres (4) mounted on a foil matrix (2) which are disposed between upper and lower pressure pads (34 and 36). The cell sandwich (32) is then heated (e.g., to about 530 DEG C). The spheres (4) are then affixed to the foil matrix (2) by directing the cell sandwich (32) through a roll press (48) which compresses the heated cell sandwich (32).
Abstract:
An integrated circuit memory device (21) includes plural input/output pins (30, 127 and others) and plural arrays of addressable storage cells (31-46). A set of circuits (51, 68, 70, 71-86, 90) provides access to a unique storage location in each array (31-46) through a given row and column address. A writing circuit (47, 68, 70, 71-86, 91-106, 131-146), designed for test, provides in parallel plural copies of a test data bit. The test data bit is applied through a single pin (30) and a common data-in lead (68), for storage in an addressed storage cell in each of the arrays. A readout circuit (110, 111, 112, 171, 127, 201-216, 131-146) is arranged for reading out the stored test data bit from the addressed storage cell in each of the arrays (31-46). The writing circuit, while in a block write test mode, stores the test data bit on the common data-in lead (68) in a block of address locations in each array (31-46).
Abstract:
Processor architecture comprising a central processing unit CPU (2, 4), a program memory (1) and a data memory (3), the central processing unit being functionally composed of a program execution portion (1, 2) and a data processing portion (4), characterised in that, in order to reduce the number of components participating in the constitution of the processor, at least one element (6, 11, 12) of the program execution portion (1, 2) is disposed in the data processing portion (4).
Abstract:
A semiconductor device having two or more p-n junctions, being in particular a bipolar transistor or a thyristor. The device has a gold ion implant (14) in a region of the device between two of or the two p-n junctions, which region is the base (2) in the case of a bipolar transistor, located away from the current carrying active region of the device. The device has a low resistance and may be turned off rapidly because the implanted gold provides recombination centres which act as a sink for carriers drawing them from the active region.
Abstract:
A field memory having a distributed architecture is disclosed and in particular a memory arranged in block, wherein each block stores a bit or bits of said word, said bits having a predetermined position within said word. The distributed architecture improves the transfer of data between the input/output buffers and the internal registers of the memory. A data cache and an improved input erasable realisation are also disclosed.
Abstract:
The invention concerns a data-processing system of the microprocessor type, with two pipeline levels comprising a device of executing an instruction sequence in a repetitive manner, comprising a program counter (74), and a program-address-start register (76) to record the number of the first instruction of the instruction sequence and a repeat counter (70) initialized at the time the first instruction of the instruction sequence is executed. The next-to-last instruction of the instruction sequence to be repeated contains an end-of-loop (EOL) code, which, when the last instruction of the loop is executed, directs that the contents of the program-address-start register (76) be loaded into the program counter (74) as long as the repeat counter is not at zero. The end-of-loop (EOL) code permits an important reduction in the circuits used and an increase in the speed of handling loops.
Abstract:
A process of removing at least a portion of a film from a substrate, such as a wafer of silicon or other similar materials, the film on the substrate typically being an oxide film, maintaining the atmosphere embracing the substrate at near room temperature and at near normal atmospheric pressure, flowing dry inert diluent gas over the substrate, pretreating the film by flowing vapor over the substrate and film flowing an anhydrous reactive gas from a source separate from the source of vapour, over the substrate and film and particularly flowing anhydrous hydrogen halide gas, namely anhydrous hydrogen fluoride gas over the substrate and film to cause the removal of portions of the film, continuing the flow of the reactive gas over the substrate film for typically 5 to 30 seconds, until a controlled amount of film has been removed, terminating the flow of reactive gas and continuing the flow of dry inert diluent gas to stop the removal of film. In the case of non-hydroscopic film on the substrate, the flow of vapor continues during the flow of the reactive gas and is terminated shortly after the termination of the flow of reactive gas.