Abstract: A four-quadrant multiplier uses a CMOS digital-to-analog converter (DAC) and just one operational amplifier. The back gates of the CMOS switches in the DAC are biased in the "off" condition during a substantial voltage swing at the output of the DAC. In one embodiment, the back gates of the CMOS switches are held at about -5 V with respect to the output lines, and the logic low level to the off switch also is set at -5 V relative to the output lines. The DAC connections are "reversed" so as to receive the analog input across the terminals intended as the DAC's output, with the inputs of the operational amplifier being connected across the reference voltage terminal and a feedback or output terminal of the DAC.

Abstract: Apparatus and method for performing low-frequency noise measurements on active devices such as operational amplifiers, instrumentation amplifiers and the like. A device under test is settled thermally in a short, controlled interval. The device under test is placed in a test fixture within a closed housing which shields against air flow. A substantial heat sink is urged against the device under test while a pre-heat circuit preheats the device for about 1-2 seconds by forcing it to consume a predetermined amount of energy. The voltage output of the device under test is supplied to a band-pass filter which includes a high-pass filter having a single dominant pole whose frequency may be varied. A control circuit smoothly moves that pole from a first, higher frequency of about 50 Hz to a second, lower frequency of about 0.1 Hz in an interval of only 3 seconds or so, rapidly settling the filter.

Abstract: A monolithic integrated circuit generates a programmable time delay under control of a digital word. The delay is generated by comparing a ramp signal to a threshold determined by the value of the digital word and appears as a time difference between a trigger pulse and a pulse generated when the value of the ramp voltage equals the value of the threshold voltage. The ramp voltage is generated by a simple resistance/capacitance charging circuit whose time constant can be adjusted by the user. The threshold voltage is set by a digital-to-analog converter (DAC) and resistor circuit which converts the digital control word into a variable voltage. In order to stabilize the device against changes in temperature and power supply variations, a voltage coupling circuit forces the threshold voltage to track changes in the ramp voltage caused by temperature and power supply variations.

Abstract: In a semiconductor device fabrication process, the SILO (Sealed Interface Local Oxidation) field oxide formation process is used to provide essentially vertical sidewalls between the field oxide surface and active regions. After field oxide formation and doping of active regions, the device is conformally coated with an oxide layer, which is patterned by a conventional photomasking process to define contact holes. Contact holes are then anisotropically etched through the oxide layer to the active regions. Conformal coating of the vertical sidewalls insures that an oxide sidewall spacer remains where the contact holes intersect the field oxide. Finally, a metal contact layer is deposited in the contact holes. The sidewall spacer automatically spaces the metal contact from the edges of the active region, thereby preventing leakage to the substrate.

Abstract: Apparatus for sampling a repetitive input signal includes a sampling bridge responsive to a sampling strobe for sampling the input signal at an equivalent time point during its cycle, an error-sampled feedback loop for providing a sampling loop output representative of the input signal sample amplitude by determining and storing the difference between the input signal sample and the attenuated loop output, and a timebase circuit for generating the sampling strobe. The timebase circuit includes a circuit for repeating the sampling strobe at least twice at each equivalent time point during succesive cycles of the input signal for reduction of errors in the sampling loop output.

Abstract: A three-state output buffer delivering digital signals to a multi-line bus when in the data state, and presenting a high-impedance to the bus in the third state. The buffer output includes a two-transistor totem pole. Individual control transistor drivers are provided to switch the output transistors off when switching to the third state. The control transistors are actively driven both on and off. One of the output transistors includes an inverted-mode auxiliary collector which reduces base drive and saturation in that transistor, and which serves to hold off the other output transistor. Common control circuitry for all the buffer stages includes special means for reducing saturation effects to speed up control signals.

Abstract: A circuit for measuring the leakage current of an electrometer amplifier 10 determines the leakage currents at both amplifier input terminals simultaneously by a method that eliminates the effects of the amplifier offset voltage as well as of the offset voltages and leakage currents of the amplifiers (22 and 40) used in the measurement. With no amp. under test present in the circuit, measurements are made of the outputs of the two test-circuit amplifiers (22 and 40). This results in quantities that indicate the required compensation for the offset voltages and leakage currents of the test-circuit amplifiers. From then on, measurements are made of the amplifier output voltages with an amp. under test present and the leakage currents at both input terminals of said amplifier 10 under test can be inferred from these second measurements by using the information obtained from the first set of measurements.

Abstract: A guided-wave electrooptic analog-to-digital converter utilizes a multiple wavelength optical source as a sampling source to minimize the number of interferometers needed for conversion of an analog signal with a given resolution. A reduction in the number of interferometers reduces the capacitive impedance of the analog signal input and facilitates driving the converter with conventional R.F. amplifiers. The multiple-wavelength signal consists of a combination of a plurality of signals with wavelengths which are substantially binary multiples of the shortest wavelength. The signals are passed together through a conventional Mach-Zehnder interferometric modulator and the interferometer output is split back into a plurality of output beams each with a single wavelength. Each of the output beams represents a bit of the digitized signal. By increasing the number of different wavelengths in the sampling signal, higher resolutions in the output signal can be obtained with a single interferometric modulator.

Abstract: A digitally programmable infinite impulse response (IIR) filter particularly useful as a temporal averager. The filter comprises a voltage mode multiplying digital/analog converter (12) and two sample-and-hold circuits (10, 16). The first sample-and-hold circuit receives the input signal to the system and supplies its output signal to the reference voltage input terminal of the DAC. The second sample and hold circuit has its input terminal connected to the voltage output terminal of the DAC and has its output connected to the analog ground terminal of the DAC. A digital input code (D) supplied to the DAC controls its gain and the degree of noise rejection provided by the filter, by altering the frequency response of the filter.

Abstract: A system for controlling the flow of semiconductor products and their components through a production facility including assembly and final testing of a large number of different products with multiple product grades. The system stores information on the demand and inventory of all product grades together with grade distribution data giving the yield of all co-products of a product family from testing the common component of that family. The system includes means for calculating the number of common components to be tested to cover the requirements for all co-products of a family. The distribution data also gives the yield of all by-products of a test for particular products, and the system operates to adjust the inventory status for other products corresponding to such by-products, to avoid excess production of such other products in meeting projected demand.

Abstract: A method is disclosed for matching the sensitivities of different-sized resistors to changes in resistance due to changes in width resulting from a systematic manufacturing error. In order to produce sets of resistors which can be deployed in predetermined ratios of resistance, the sensitivities of a matching resistor and a reference resistor are equalized by forming the matching resistor as a plurality of parallel strips as opposed to a unitary rectangular section.

Abstract: A voltage reference source for supplying one or more reference voltages to load circuitry. The reference source comprises an ungrounded voltage reference cell (20) floated between two power supply levels (Vcc and Vee), and an operational amplifier (22) whose non-inverting input (terminal 9) may be connected to sense the voltage of the load circuit's ground node. The inverting input of the op amp (terminal 10) may be connected to various nodes (6, 8, 11) in the reference cell; the voltage on the selected node is forced to match the sensed ground voltage and that node thus becomes an internal ground reference point. The high input impedance of the op amp (22) limits the current in the ground sensing path to a very low value, such as about 10 nA. Consequently, very little voltage is developed in the leads from the load circuit's ground node to the op amp input. The floating reference cell (20) has a resistive divider (48, 50) across its output.

Abstract: A voltage/current source includes a loop controller (24) that digitally determines the control signals that must be applied to a driver amplifier (34) to achieve the desired load voltage or current. The analog outputs of current- and voltage-sensing amplifiers (38 and 68) are converted by analog-to-digital converters (52 and 74) to digital feedback signals that the loop controller (24) uses in determining what control signals to generate. The loop controller (24) keeps the driver-amplifier output voltage equal to the load voltage until switch contacts (30) connect the source to the load so that connection-caused transients are minimized. The loop controller (24) includes read-write memory (25) in which it stores program instructions and operational parameters, so the source can readily change its feedback characteristics.

Abstract: A band-gap reference circuit having a pair of transistors operated at different current densities to produce a positive temperature coefficient (TC) signal proportional to the .DELTA.V.sub.BE of the two transistors and combined with a negative TC voltage derived from the V.sub.BE of one of the transistors to produce a composite signal substantially invariant with temperature. The .DELTA.V.sub.BE signal component is increased in magnitude by connecting resistor string bias circuit to each of the transistors, to effectively multiply the V.sub.BE of each transistor, and thereby multiply the .DELTA.V.sub.BE signal. The composite signal is sensed in the emitter circuits of the two transistors, so that it is unnecessary to access the collectors of the transistors, thereby making it readily possible to use the circuit with CMOS IC devices.

Abstract: A single-chip 8-bit DAC with bipolar current sources, an output buffer amplifier for developing an output voltage, a regulated reference for producing a calibrated output, and operated by a single-voltage supply, e.g. +5 volts. The buffer amplifier includes means providing for driving the output voltage virtually to ground level when the DAC output is zero. The current sources comprise a single-transistor cell driven by an I.sup.2 L flip-flop circuit, and the reference supply is merged with the reference transistor circuit regulating the DAC current levels, both aiding in reducing required chip area. A highly efficient bias network is utilized to supply the high-level bias currents required.

Abstract: A technique for enclosing microelectronic circuit elements in hermetically sealed packages comprising a planar ceramic substrate with a box-like ceramic cover sealed thereto by a fused glass coating. The glass sealant is applied to the substrate in the form of a paste which thereafter is fired at high temperature and cooled to produce a smooth glass coating. With the cover in place on the substrate, the glass coating is remelted by heat developed by infra-red radiation impinging on all sides of the package structure from heaters in an infra-red furnace. A reflective shield on top of the cover reduces the inflow of heat through that surface, and a heat sink beneath the substrate removes heat, thereby to reduce the temperature rise experienced by circuit elements in the package interior.

Abstract: A universal trigonometric function generator which is selectively programmable by pin-strapping to generate any of the standard trigonometric functions (sine, cosine, tangent, cotangent, secant and cosecant). The circuit includes two identical sine-function generating networks each of which produces an output signal proportional to the sine of a corresponding angle input. These networks are so interrelated that the composite output signal is proportional to the angle input of one network and inversely proportional to the angle input of the other network, producing an output ##EQU1## where A is a controllable amplitude, .theta..sub.1 -.theta..sub.2 is the angle input to one network, and .phi..sub.1 -.phi..sub.2 is the angle input to the other network. By selectively connecting the input terminals for .theta..sub.1, .theta..sub.2, .phi..sub.1, .phi..sub.2 to an angle control signal and reference voltages corresponding to 0.degree. and 90.degree., any one of the standard trigonometric functions can be generated.

Abstract: A method of laser trimming thin film resistors on semiconductive substrates wherein the laser is set to a frequency equal to or less than E.sub.g /h, where E.sub.g is the optical band-gap energy of the doped semiconductor substrate, and h is Planck's constant.

Abstract: This invention relates to a method of trimming an element of an integrated circuit with a laser beam. A reflective, or otherwise optically opaque, template-like layer is placed over a portion of the circuit element to control the trimming operation. The laser beam is directed along the sharp edge of this layer in such a way that the maximum energy portion of the beam strikes a part of the circuit element, while low energy portions of the beam are prevented from reaching other parts of the element where they might adversely affect its characteristics.

Abstract: An IC chip having a Zener diode with a subsurface breakdown junction to assure stable operation. The diode is formed by a triple diffusion process compatible with conventional bipolar processing. A deep p.sup.++ diffusion first is applied, reaching through the epitaxial region to the buried n.sup.+ layer; next, a shallow p.sup.+ diffusion is formed over the deep p.sup.++ diffusion and extending laterally beyond that diffusion; finally, a shallow n.sup.+ diffusion is applied over the p diffusions, to form a subsurface breakdown junction therewith. The topology of the mask windows is selected to provide concentration profiles which insure that the breakdown occurs at the subsurface junction, and that other desirable diode characteristics are achieved.