71results about How to "Reduce load capacitance" patented technology

The present invention relates to a timing system including an integrated circuit having an oscillator that provides both high and low frequency clock signals from a single high frequency crystal without the necessity of a tuning fork crystal. The low frequency signal is available for time-keeping applications, with low power consumption during “idle” periods. The high performance high frequency signal is available on demand for clock and frequency reference use. The oscillator of the present invention provides improved time-keeping accuracy, whilst size, cost and component count is reduced. Furthermore, phase noise and other critical parameters of the high frequency oscillator are not compromised. Shock vulnerability, a known problem for tuning fork crystals, is reduced.

The CMOS image sensor according to the present invention includes a sample hold section 3A1 for retaining an analog input signal voltage and a ramp wave signal voltage; and a comparing section 3A2 for taking an output from the sample hold section 3A1 as an input to compare it with a reverse level of itself, in which the sample hold section 3A1 applies a stabilization promoting voltage to a terminal of the sampling capacitance element so that an electric potential level of the terminal of the sampling capacitance element is promoted to become stabilized at a predetermined voltage, when the analog input signal is applied to the terminal of the sampling capacitance element.

A self-aligned contact structure and a method of forming the same include selected neighboring gate electrodes with adjacent sidewalls that are configured to angle toward each other. The angled surfaces of the gate electrodes can be protected using a liner layer that can extend the length of the contact window to define the sidewalls of the contact window.

There is provided an image display device operating in response to the input of digital picture signals, in which the occupied area of a signal line driver circuit thereof is reduced, and the parasitic capacitance and resistance of input transmission lines of the digital picture signals are reduced. The device includes both a unit for directly inputting the digital picture signals to shift registers and for performing series parallel conversion, and a unit for causing n (n is a natural number not less than 2) signal lines to jointly own storage circuits and D/A converter circuits in the signal line driver circuit. One horizontal scan period is divided into n periods, and the storage circuits and the D/A converter circuits perform a processing to signal lines different in each of the divided periods.

The invention provides a input buffer circuit for high-speed pipeline analog-to-digital converter (ADC). The input buffer circuit comprises an input pair transistor, a current mirror load, a gain-boosted amplifier and a compensation capacitor, wherein the input pair transistor consists of a first MOS (metal oxide semi-conductor) transistor and a second MOS transistor of a source negative-feedback resistor; the current mirror load consists of a third MOS transistor and a fourth transistor which are connected to each other by a gate; the positive input terminal of the gain-boosted amplifier is connected with the drain of the first MOS transistor, the negative input terminal thereof is connected with the drain of the second MOS transistor, and the output terminal thereof is connected with the gate of the third MOS resistor; and one terminal of the compensation capacitor is connected with the gate of the third MOS transistor, and the other terminal thereof is connected with the drain of the third MOS transistor. The input buffer circuit of the invention is capable of meeting the requirements for the input of high-speed signals, effectively reducing the inputted load capacitance and increasing the outputted impedance.

A high speed low powered 4-2 compressor according to the present invention performs an XOR/XNOR operation of input data by using a single input type NAND/NOR logic circuit and a dual input type NAND/NOR logic circuit. Thus, delays to generate complementary signals are avoided. In addition, the 4-2 compressor uses a single railed multiplexer instead of a dual railed multiplexer, so that gate drive nodes and internal load capacitance are reduced. As a result, circuit area and power consumption of the 4-2 compressor are reduced.

A method for manufacturing a heterojunction bipolar transistor is provided. An intrinsic collector structure is formed on a substrate. An extrinsic base structure partially overlaps the intrinsic collector structure. An intrinsic base structure is formed adjacent the intrinsic collector structure and under the extrinsic base structure. An emitter structure is formed adjacent the intrinsic base structure. An extrinsic collector structure is formed adjacent the intrinsic collector structure. A plurality of contacts is formed through an interlevel dielectric layer to the extrinsic collector structure, the extrinsic base structure, and the emitter structure.

A semiconductor device having a thin film formed by atomic layer deposition and a method for fabricating the same, wherein the semiconductor device includes a liner layer formed on an internal wall and bottom of a trench, gate spacers formed on the sidewalls of gate stack patterns functioning as a gate line, a first bubble prevention layer formed on the gate spacers and the gate stack patterns, bit line spacers formed on the sidewalls of bit line stack patterns functioning as a bit line, and a second bubble prevention layer formed on the bit line spacers and the gate stack patterns and at least one of the above is formed of a multi-layer of a silicon nitride layer and a silicon oxide layer, or a multi-layer of a silicon oxide layer and a silicon nitride layer, thereby filling the trench, gate stack patterns, or bit line stack patterns without a void.

A display panel drive circuit having a plurality of first and second electrodes for connecting to a display panel, a first drive circuit for driving the first electrodes, and a second drive circuit for driving the second electrodes. The second drive circuit is connected to drive all or a part of a plurality of the second electrodes, or interrupted to increase output impedance.

The present invention provides a method for producing semi-conductor. The inventive method includes the steps of: forming a plurality of patterns on a substrate, wherein the patterns are formed by stacking and patterning a first conductive layer, a silicon nitride mask layer and a metal mask layer on the substrate; depositing a first silicon oxide layer along the profile containing the patterns; etching the first silicon oxide layer for forming a silicon oxide spacer with a height lower than a top part of the silicon nitride mask layer so as to partially expose a top part of lateral sides of patterns and simultaneously etching the metal mask layer to expose the silicon nitride mask layer, wherein the metal mask layer prevents losses of the silicon nitride mask layer; forming a silicon nitride spacer on a surface of the silicon oxide spacer and the lateral sides of the patterns; forming a second silicon oxide layer on an entire structure in which the silicon nitride spacer is formed; etching selectively the second silicon oxide layer to expose silicon nitride layer spacer and forming a self-align contact hole that is partially expanded to the top portion of the patterns; and forming a self-align contact structure by filling the self-align contact hole with a second conductive layer.

According to the invention, in a semiconductor device that needs a refresh operation for storing data, data of memory cells selected in response to a row address is read to main amplifiers through bit line pairs, sense amplifiers and data line pairs in a page-mode read operation. Thereafter, while outputting the data held in the main amplifiers to the outside, connecting transistors are turned off so as to disconnect the main amplifiers from the memory cells, and thus, the memory cells can be precharged. Also, in a page-mode write operation, while writing externally supplied input data in the main amplifiers, the memory cells can be precharged.

The widths of those portions of a semiconductor layer 5 and a drain line 6a overlapping with it which cross an edge line of a gate electrode 2 are made smaller than the channel width of a thin-film transistor. With this measure, the overlap area of the gate electrode 2 and a drain electrode 6 is reduced. As a result, a variation of the above overlap area due to alignment errors in a photolithography apparatus used in patterning the gate lines 2, the drain electrodes 6, and source electrodes 7 can be reduced and the frequency of occurrence of display defects can be decreased.

The invention discloses a pipeline successive-approximation analog-to-digital converter used for solving the technical problem that the existing pipeline successive-approximation analog-to-digital converter is high in power consumption. The technical scheme is as follows: a first stage circuit adopts a 6-bit half-gain MDAC (Multiplying Digital To Analog Converter) circuit, and a second stage circuit adopts an 8-bit full-reference voltage multiple-segment capacitor SAR ADC (successive-approximation analog-to-digital converter). The whole capacitor network of the 8-bit full-reference voltage multiple-segment capacitor SAR ADC is divided into a high 4-bit binary weighted capacitor network and a low 4-bit binary weighted capacitor network. Since a sampling capacitance of the 8-bit full-reference voltage multiple-segment capacitor SAR ADC is significantly reduced, thus a load capacitance of a residue amplifier in the 6-bit half-gain MDAC circuit is significantly reduced. For the same error requirement, current consumed by the residue amplifier in the 6-bit half-gain MDAC circuit can be reduced, so that the power consumption of the pipeline successive-approximation analog-to-digital converter can be reduced.

A second insulating layer is formed on a first insulating layer. A plurality of stacks each including a bit line and a bit line mask are formed on the second insulating layer. A third insulating layer is formed overlying the second insulating layer to fill gaps between the plurality of stacks. A hard mask layer is formed on the third insulating layer. A photoresist pattern is formed on the hard mask layer. The photoresist pattern has an opening region that intersects the plurality of stacks. The hard mask layer and the third insulating layer are sequentially etched, using the photoresist pattern as an etching mask, thereby forming a hard mask pattern and forming a recess in the third insulating layer. The recess exposes a portion of upper sidewalls of the bit line mask. Spacers are formed on the exposed upper sidewalls of the bit line mask.