Abstract: A ferroelectric transistor (72) is programmed with a gate voltage that shifts a threshold voltage of the ferroelectric transistor (72). The shifted threshold voltage generates a correction current (.DELTA.I.sub.(N-1)) in a combination circuit (52) that trims an output voltage of a DAC trim circuit (50). A ferroelectric material (32) of the ferroelectric transistor (72) stores a charge in accordance with a programming voltage and allows a dynamic adjustment of the correction current (.DELTA.I.sub.(N-1)) that is used to modify the output voltage of the DAC trim circuit (50).

Abstract: The amplifier (200) includes an input stage (220) coupled to two output transistors (281, 282) having a common terminal at the output terminal (206) of the amplifier. Class AB operation of the output transistors (281, 282) is possible at a comparatively low supply voltage. In order to obtain such operation, measurement transistors (271, 272) are coupled to the same control input (283, 284) as the output transistors (281, 282). These measurement transistors (271, 272) are serially coupled to a current mirror (260). The quiescent current of the output transistors (281, 282) is measured and used to produce a feedback signal which is superimposed to the control signals.

Abstract: A circuit and method produce a pump current (I.sub.p) at the output (31) of a charge pump (26). A switching transistor (32, 35) is coupled at a node (38, 39) to a current source transistor (33, 34) to produce the pump current at a specified magnitude in response to an input pulse (V.sub.PU, V.sub.PD). A charge is stored on a parasitic capacitance of the node. A charge conduction path (42, 43) is coupled to the node and enabled on one transition edge of the input pulse to alter the charge by routing to a discharge node (45) to reduce charge flowing to the output as an error current. The charge conduction path is disabled on the other transition edge of the input pulse to isolate the node from the output.

Abstract: A neural network (10) organizes the data items into a graphically oriented format by retrieving data items from a database (68) where each data item has a plurality of attributes. The neural network is organized (102) such that data items having similar attributes are assigned to neurons located closer together. The neurons of the neural network are matched (104) with the data items from the database and stored in a cross reference table. The cross reference table is displayed (106) on a computer screen (108) in a graphical format so that user visually relates the food items and sees the similarities and differences in their attribute data by the proximity of the data items to one another. The graphic format allows easy visual interpretation of the data items. For large databases, multiple neural networks (110, 112) can be organized hierarchically.

Abstract: A battery charge control circuit (10) senses the charge condition of cells in a battery pack (12, 14, 16, 18) using a measurement circuit (51). Upon detection of a single under-voltage cell, the charge control circuit is placed in a sleep mode. Pack sense circuit (240) senses when the battery pack is placed in a charger. If circuit (10) was in a sleep mode, it is awakened. If any cell is measured over-voltage, the status is checked versus the other cells. If all the cells are over-voltage, the battery is considered balanced. If one or more cells are not over-voltage, a control circuit (32) activates a discharge transistor (212, 214, 216, 218), discharging the cell within a hysteresis voltage below the over-voltage limit. Charge balancing of cells is continued until the cells are within a programmable hysteresis voltage of each other.

Abstract: A method for flip-chip bonding of two electronic components (27,28) does not use a flux material. A substrate (13) of one electronic component (28) is roughened during processing to provide an improved adhesive surface for a solder ball (12). The roughened pattern is replicated by additional conductive layers formed over the substrate or in an alternate embodiment may be formed on one of the intermediary or top conductive layers. Tacking pressure is applied to the two components so the solder ball (12) will be affixed to the roughened surface and provide a temporary bond. This bond ensures the surfaces of the two electrical components remain in contact with each other during reflow of the solder ball (12) to form a permanent bond.

Abstract: A serial data signal is synchronized to a clock signal in a synchronization circuit (10). Synchronization is accomplished by generating a plurality of delayed versions of the serial data signal using serially connected delay elements (21-27). The delayed versions of the serial data signal are sampled using a set of flip-flops (11-18). The sampled delayed data signals appearing at the outputs of each flip-flop of the set of flip-flops (11-18) are used to determine which delayed data signal is most closely aligned to the clock signal. The output of the multiplexer (40) is an aligned serial data signal. In addition, a drift correction circuit (50) continuously monitors and corrects the alignment between the clock signal and the aligned serial data signal.

Abstract: An integrated circuit and method produce a first clock signal (CLOCK) from a reference clock (REFCLK) but at a different frequency. A variable delay line (32) produces the first clock signal by introducing a variable delay in the reference clock signal that is controlled by a programming signal generated in a counter (34). The programming signal is incremented by a second clock signal (UP) while transitions of a fixed delay clock signal lead transitions of the first clock signal. When the programming signal reaches the count of a rollover code (ROLLOVER), the programming signal is reset to a zero count to begin a new sequence. A calibration circuit (36, 38, 40, 42) determines the count of the programming signal needed to produce the rollover code when the variable delay is at least as great as one period of the reference clock signal.

Abstract: A current source (20) uses a current mirror having an input coupled for receiving a first reference current. A first transistor (34) is serially coupled in the input of the current mirror. A second transistor (38) has a first conduction terminal coupled for receiving a second reference current, and a second conduction terminal coupled to an output of the current mirror. The first conduction terminal of the second transistor is coupled to common control inputs (42) of the first and second transistors. As the output voltage of the current source decreases the current mirror transistors are forced to have the same drain-source voltage and gate voltage, and operate at substantially the same point in their linear region. The tracking of the drain-source voltages of the current mirror transistors allow the current source to maintain a constant output current when operating at very low output voltages.

Abstract: Low voltage operational amplifier (10) operates in a voltage range of one to eight volts over a temperature range of 0.degree. to 70.degree. centigrade. Op amp input stage (12) uses N-channel depletion-mode MOSFETs to provide amplification of the differential input and maintain constant transconductance. Source follower MOSFET (13) provides unity gain in transferring the AC signal, STAGE-1 OUTPUT, to the base of current sinking transistor (18). Sink control circuit (14) and source control circuit (22) generate the base drive currents for in transistors (18) and (24). The signal at the output of MOSFET (13) either causes the sink transistor (18) to sink current or the signal to be transposed by means of a translinear loop (16) and causes the source transistor (24) to source current.

Abstract: A circuit and method are provided for generating a programmable clock signal VRCLK at a frequency of a reference signal VREF such that the phase of VRCLK in relation to the phase of VREF is variable. VREF and VRCLK are each coupled to a frequency comparing circuit which computes their respective frequencies. The frequency comparing circuit subtracts the respective frequencies of VREF and VRCLK to produce a frequency adjusting signal VFREQ which corresponds to the difference in the frequencies of VREF and VRCLK. VFREQ is used to adjust the frequency of VRCLK.

Abstract: A display controller (112) reduces the power consumed in displaying a graphics image in a portable wireless communications device (100) when a graphics image is smaller than the size of the display (118). The number of rows and columns used to display the graphics image is counted by a decoder (108) which is a microcontroller used to operate the communications device (100). The decoder (108) provides the reduced row or column count to the display controller (112), which reduces the frequencies of clocks (PIXEL CLOCK, LINE PULSE, FRAME PULSE) used for timing data transfers to the display (118). Power is reduced by operating the display (118) at a lower frequency while acceptable frame refresh rates are maintained.

Abstract: A monolithic clamping circuit (10) and a method of protecting a semiconductor power transistor (14) from entering avalanche breakdown. The semiconductor power transistor (14) controls the switching of an inductor (16). The monolithic clamping circuit (10) causes energy stored in the inductor (16) to be dissipated in the semiconductor power transistor (14). A voltage sense circuit (18) provides a feedback signal to a selector circuit (12) in response to a voltage at a collector terminal of the semiconductor power transistor (14). The selector circuit (12) switches the semiconductor power transistor (14) to the conductive mode when the feedback signal indicates a high voltage at the collector terminal of the semiconductor power transistor (14). Dissipating the energy stored in the inductor (16) while operating in the conductive mode prevents the semiconductor power transistor (14) from entering avalanche breakdown.

Abstract: A three dimensional packaging approach reduces the overall footprint for interconnecting multiple semiconductor die. An three-dimensional folded module (10) produces a final package having a footprint size reduced by an approximate factor of four when compared to conventional electronic packaging. The module has a protective covering such as a cap (62) or a sealant (64) as a moisture barrier. Thus, high integration using flexible appendages (15, 25, 35, and 45) attached to a rigid substrate (12) and singularly folded above the substrate (12) results in both a small footprint package and also a light package. A reel-to-reel flex tape (56) assembly provides pre-tested flex boards (16, 26, 36, and 46) resulting in a cost-effective manufacturable package for semiconductor components.

Abstract: A compensation circuit (106) corrects for nonlinearities in a sensor signal representing the physical state of a sensor (100). A transducer (102) produces a non-linear component in a transducer voltage signal. A voltage-current converter (104) converts the tranducer voltage signal to a transducer current which contains the non-linear component. A compensation circuit (106) squares the transducer current (I.sub.216) and uses a scaling current (I.sub.412) to generate a compensation current (I.sub.408) equal to the non-linear component. The current (I.sub.216) and scaled compensation current (I.sub.408) are summed at a summing junction (418) to produce an output current (I.sub.OUT) which is a substantially linear representation of the physical state of the sensor.

Abstract: A method for controlling production lots which deals with production lots comprising a plurality of discrete units distributed among a plurality of locations. A first event is triggered when a first unit from a first production lot enters a processing step. A second event is triggered when a last unit from the first production lot leaves the processing step.

Abstract: The A/D converter comprises a resistor series (30), a plurality of first comparators (1) and a plurality of second comparators (2). Resistor series (30) has a plurality of resistors(R) connected in series between two terminals to which predetermined reference voltages are applied. First comparator (1) compares a node voltage between resistors and an analog voltage signal to be compared. Second comparator (2) compares an average voltage of the two node voltages across each resistor R and the analog voltage signal. First and second comparators (1,2) are disposed alternatively.

Abstract: A memory circuit (24) includes a sense amp circuit (30) that uses multiplexers (86) in a column mux (32) for pre-charging only selected bitlines in order to limit the current during a read operation of the FLASH memory circuit (24). The sense amp circuit (30) provides the bitline with a pre-charge voltage that is set by a current reference (68) that is substantially supply independent. In the read mode the sense amp circuit (30) responds to either a voltage on the bitline that is lowered from the pre-charge voltage value by a selected programmed memory cell (40) or by a voltage that remains at the pre-charged voltage value for an unprogrammed memory cell.

Abstract: A method and a circuit for automatically adjusting a switching threshold of an input buffer circuit (100) to conform to an input signal V.sub.IN which can be from either a TTL or a CMOS logic family. A latch circuit (120) is initialized to set the switching threshold to that of one of the logic families. A level shifting circuit (130) has a switchable load which varies the switching threshold under the control of the latch circuit (120). The amplitude of the input signal V.sub.IN is detected in a threshold detector circuit (110). If the input signal V.sub.IN is a signal from the CMOS logic family, the latch circuit (120) changes state, switching the switchable load of the level shifting circuit (130) to adjust the switching threshold of the input buffer circuit (100).

Abstract: A computer implemented method provides for orientation and lay-out asymmetrical semiconductor devices. The sources of operating potential (20, 30) are identified. The PMOS transistors (22-28) are combined between the first (more positive) source of operating potential and a common node (12) into a combination block (48). The NMOS transistors (14-18) are combined between the common node and the second (less positive) source of operating potential into another combination block (52). The PMOS source terminals are coupled to more positive potentials, and the PMOS drain terminals are coupled to less positive potentials within the first combination block. The NMOS source terminals are coupled to less positive potentials, and the NMOS drain terminals are coupled to more positive potentials within the second combination block. For transmission gates (60, 62), a driving source is identified and the PMOS and NMOS source terminals are coupled to the driving source.