Abstract: Methods of forming power semiconductor devices having merged split-well body regions include the steps of forming a semiconductor substrate containing a drift region of first conductivity type (e.g., N-type) therein extending to a first face thereof. First and second split-well body regions of second conductivity type (e.g., P-type) may also be formed at spaced locations in the drift region. First and second source regions of first conductivity type are also formed in the first and second split-well body regions, respectively. A central body/contact region of second conductivity type is also formed in the drift region, at a location intermediate the first and second split-well body regions. The central body/contact region preferably forms non-rectifying junctions with the first and second split-well body regions and a P-N rectifying junction with the drift region at a central junction depth which is less than the maximum well junction depths of the split-well body regions.

Abstract: An integrated circuit inverted thin film resistor structure and method of manufacture having interconnect defining resistor contacts and leads resident within and coplanar with a supporting layer, resistive material uniformly overlaying the supporting layer and contacts, the resistive material diffused into the resistor/interconnect contact region.

Abstract: A digital-to-analog converter (DAC) includes a plurality of current sources on a substrate operable in a predetermined sequence of use for generating an output current based upon a digital input, and a connection network for establishing the predetermined sequence of use for the current sources based upon the actual current values and to increase performance of the DAC. For example, the predetermined sequence of use can be set to reduce integral non-linearity error of the DAC. The connection network may be provided by a plurality of fusible links selectively connected to set the predetermined sequence of use. The current sources may include a first group of most significant bit (MSB) current sources for a predetermined number of MSBs. In addition, the plurality of current sources may include a second group of mid-most significant bit (mid MSBs) current sources for a predetermined number of mid MSBs.

Abstract: A class D modulator 100 is self oscillating. It includes an input stage 110, an operational amplifier 120, an integrator 130 and a pulse width modulator 140. A start-up generator 22 generates signals ENABLE and MUTE. The MUTE signal is connected to the input stage 110 to prevent any unwanted, initial sound. The ENABLE signal connects a start-up voltage source to the integrator 130 to set the initial voltage for integration. The ENABLE signal is coupled to a dead time control circuit 16 which controls the operation of the output power transistors 42, 44 to assure that the low side power transistor 44 is turned on before the high side transistor 42.

Abstract: A bonded wafer has a first handle wafer 12, a device layer 10', an interconnect layer 14, and a number of vias filled with conductive material that extends between the surfaces 6, 8 of the device layer 10'. the interconnect layer 14 has conductors that connect internal device contacts to the conductive vias. A second handle wafer 40 of glass is bonded to the interconnect layer 14 and the first handle wafer is removed. Bottom, external contacts 36 are formed on surface 6 of device layer 10'.

Abstract: Semiconductor devices 340 are formed in semiconductor wafer 300. Contact pads 332 are formed in each die 330. An interconnect connects the contact pads 332 to die surface contact regions 210, 212. Scribe trenches 348 are formed in device wafer 300; corresponding trenches 358 are formed in cover wafer 360. The cover wafer 360 is thinned to open scribe trenches 348. Conductive vias 310-313 connect the contact pads 210, 212 to external surface bump contacts 333.

Abstract: A complementary multiplexer with low disabled output capacitance and method in which a plurality of switched buffers are packaged together to avoid the capacitance of a plurality of switched buffers applied to the printed-circuit board transmission lines.

Abstract: A method of fabricating a MOS controlled thyristor (MCT) semiconductor power device which reduces process time, reduces cell size, and increases the density of turn-off channels. The method uses a single, dopant-opaque mask to form MCT structure above the bottom N and P layers, including the upper portions of PNP and NPN transistors which form the MCT and On-FETs and Off-FETs which operate the MCT. The single mask may also be used to fabricate floating field rings for the device. The method may also be used on both sides of the device to provide a Fast Turn Off (FTO) device with both On- and Off-FETs on one side and at least an Off-FET on the other side.

Abstract: A trench process for establishing a contact for a semiconductor device with trenches such as the trench and planar MOSFETs (UMOS), trench and planar IGBTs and trench MCTs which reduces the number of masks and eliminates the need for lateral diffusion into the trench channel region. Improved control of the parasitic transistor in the trench MOSFET is also achieved. The cell size/pitch is reduced relative to conventional processes which require source block and P+ masks.

Abstract: A sensing circuit 100 includes two resistors 11 and 12, two pnp transistors 13 and 14, and a current mirror 15 and 16. Resistors 11 and 12 convert the high voltages present in the bridge into a proportional current. The current mirror, mosfets 15 and 16, compares I.sub.1 and I.sub.2. If I.sub.2 is greater than I.sub.1 the voltage at point A is high. Otherwise this voltage is low. Resistor 12 is chosen smaller than 11 so that under normal operation, when FET 22 turns on, I.sub.2 is greater than I.sub.1 and the voltage at point A is high. During an overcurrent event, the drop across the FET 22, Von, is so great that I.sub.2 is less than I.sub.1 and the voltage a point A stays low.

Abstract: A class D amplifier 100 has an integrator 10, a comparator 12, and a frequency compensation and gain control (FCGC) circuit 40. The FCGC circuit 40 senses the output and reduces the gain in order to keep the sampling frequency high enough to avoid audio artifacts.

Abstract: A circuit and method for synchronizing multiple transmitting devices in a multiplexed communication system in which transmitting nodes generate bit transitions on the multiplexed bus based on the time elapsed since the last received bit transition, and in which the synchronization is dependent only on the last received bit transition and on no other synchronization signal. The circuit and method may be used to prevent inadvertent bit transition transmission within a predetermined period of time of receipt of a bit transition.

Abstract: An analog-to-digital converter (ADC) includes a plurality of capacitors formed on a semiconductor substrate and having actual capacitance values statistically related to desired capacitance values, and a gain stage comprising an amplifier and capacitors selected to provide a more accurate gain for the gain stage. A first at least one capacitor is connected between an input and an output of the amplifier defining a feedback capacitance, and a second at least one capacitor is connected between the input of the amplifier and an input of the at least one gain stage defining an input capacitance. In addition, the ADC includes a connection network selectively connecting the first at least one capacitor and the second at least one capacitor from among the plurality of capacitors to provide a desired ratio of feedback capacitance to input capacitance based upon the actual capacitance values. Accordingly, a gain can be set for the gain stage that is more accurate than would otherwise be obtained.

Abstract: A semiconductor device includes at least one device active region formed in a first surface of a semiconductor substrate, an electrical contact layer on a second surface of the semiconductor substrate, and at least one resistivity-lowering body positioned in a corresponding recess in the substrate and connected to the electrical contact layer. The body preferably comprises a material having an electrical resistivity lower than an electrical resistivity of the semiconductor substrate to thereby lower an effective electrical resistivity of the substrate. The device active region may be an active region of a power control device, such as a MOSFET or IGBT, for example. The body may preferably comprise an electrical conductor such as copper, aluminum, silver, solder, or doped polysilicon. The at least one recess and associated resistivity-lowering body preferably defines a proportion of the semiconductor substrate area adjacent the device active region greater than about 0.

Abstract: In combination with a transistor designed to drive an inductive load, there is included a network connected between the output electrode (e.g., drain) and the control electrode (e.g., gate) of the transistor for limiting the overshoot and controlling the waveshape of the signal produced at the output electrode of the transistor, when the transistor is being turned-off. The network includes a series string of zener diodes with one or more by-pass capacitors connected across the zener diodes closest to the control electrode of the transistor for shaping the output signal produced at the output electrode of the transistor and for reducing electromagnetic radiation. The network also includes unidirectional conducting elements for discharging each bypass capacitor each time the transistor is turned-on. The zener diodes and the "discharging" unidirectional conducting elements of the network may be formed as integral parts of the same integrated circuit (IC).

Abstract: A high voltage MOSFET with low on-resistance and a method of lowering the on-resistance for a specific device breakdown voltage of a high voltage MOSFET. The MOSFET includes a blocking layer of a first conductivity type having vertical sections of a second conductivity type or the blocking layer may include alternating vertical sections of a first and second conductivity type.

Abstract: A method for fabricating a field effect transistor using the supporting substrate as a device layer in accordance with the present invention comprises several steps. To fabricate the field effect transistor, an epitaxial layer is grown on one surface of a substrate having an opposing surface, the epitaxial layer forming a drain for the transistor. Once the epitaxial layer is grown, the substrate is thinned to an appropriate device thickness and then a gate and a source for the transistor are formed on the opposing surface of the substrate. In an alternative embodiment, a polysilicon layer is used instead of the epitaxial layer. A field effect transistor fabricated from the method described above includes an epitaxial layer formed on one surface of a substrate having an opposing surface, the epitaxial layer forming a drain for the transistor and a gate and a source for the transistor formed on the other surface of the substrate.

Abstract: In a semiconductor device, a trench is etched into a surface of a semiconductor body comprising, from the surface down, a highly doped first (source) region; a moderately doped second (body) region; and a lightly doped third (drain) region. The trench walls are then oxidized. For reducing the effects of etching rate and oxide growing rate variations which occur at the junctions between regions of differing concentrations, the trench is first formed by etching and the trench walls then oxidized prior to the formation of the first region. Trenches having straighter walls and more uniformly thick oxides are thus formed.

Abstract: A power module includes a semiconductor substrate and spaced power transistor circuits formed in the substrate, each having respective anode and cathode portions. A power module housing substantially encloses the substrate. Flexible cathode straps and flexible anode straps are connected to the respective anode and cathode portions and extend out of the power module housing spaced from each other. At least two anode straps and cathode straps fold over at least a portion of the power module housing to overlap each other and form phase straps. The remaining cathode and anode straps fold over at least a portion of the power module housing such that respective cathode and anode straps overlap each other, thereby lowering inductance and reducing voltage overshoots at turn off.

Abstract: An integrated sample-and-hold S/H circuit includes a subthreshold conduction current compensation circuit for reducing undesired effects of subthreshold conduction current in a first field-effect transistor (FET) during the holding time. More particularly, the S/H circuit may include a substrate, a sampling capacitor formed on the substrate, and the first FET. The first FET has a first conduction terminal for receiving the input signal, a second conduction terminal connected to the sampling capacitor, and a control terminal. The control terminal is responsive to control signals for connecting the input signal to the sampling capacitor during a sampling time and for disconnecting the input signal from the sampling capacitor during a holding time. The first FET preferably further includes a body which unfortunately creates a parasitic diode connected to the sampling capacitor.