摘要:
A programmable logic device (PLD) includes a delay circuit and a body-bias generator. The delay circuit has a delay configured to represent a delay of user circuit implement in the PLD. The body-bias generator is configured to adjust the body bias of a transistor within the user circuit. The body-bias generator adjusts the body bias of the transistor in response to a level derived from the signal propagation delay of the delay circuit.
摘要:
A programmable logic device (PLD) includes a delay circuit and a body-bias generator. The delay circuit has a delay configured to represent a delay of user circuit implement in the PLD. The body-bias generator is configured to adjust the body bias of a transistor within the user circuit. The body-bias generator adjusts the body bias of the transistor in response to a level derived from the signal propagation delay of the delay circuit.
摘要:
A programmable logic device (PLD) includes mechanisms for adjusting or setting the body bias of one or more transistors. The PLD includes a body-bias generator. The body-bias generator is configured to set a body bias of one or more transistors within the programmable logic device. More specifically, the body-bias generator sets the body bias of the transistor(s) so as to trade off performance and power consumption of the transistor(s).
摘要:
An integrated circuit having an enhanced on-off swing for pass gate transistors is provided. The integrated circuit includes a core region that includes core transistors and pass gate transistors. The core transistors have a gate oxide associated with a first thickness, the pass transistors having a gate oxide associated with a thickness that is less than the first thickness. In one embodiment, the material used for the gate oxide of the pass gate transistors has a dielectric constant that is greater than four, while the material used for the gate oxide of the core transistors has a dielectric constant that is less than or equal to four. A method for manufacturing an integrated circuit is also provided.
摘要:
A method for fabricating a transistor on a semiconductor substrate includes varying a polysilicon doping level near a first and second edge of a diffusion region with a polysilicon doping level of a center region of a polysilicon region.
摘要:
Integrated circuits with memory elements are provided. The memory elements may be arranged in a memory block. The memory block may include cross-coupled inverters that store data. The stored data may be used to program pass transistors. Transistors in the memory block may be stressed. Depending on the type of stress-inducing layer used, a tensile stress or a compressive stress may be built in into the transistors. Stressed transistors may help improve the routing speed of the memory block. Stressed transistors may be implemented using dual gate-oxide process.
摘要:
Integrated circuit antifuse circuitry is provided. A metal-oxide-semiconductor (MOS) transistor serves as an electrically-programmable antifuse. The antifuse transistor has source, drain, gate, and substrate terminals. The gate has an associated gate oxide. In its unprogrammed state, the gate oxide is intact and the antifuse has a relatively high resistance. During programming, the gate oxide breaks down, so in its programmed state the antifuse transistor has a relatively low resistance. The antifuse transistor can be programmed by injecting hot carriers into the substrate of the device in the vicinity of the drain. Because there are more hot carriers at the drain than at the substrate, the gate oxide is stressed asymmetrically, which enhances programming efficiency. Feedback can be used to assist in turning the antifuse transistor on to inject the hot carriers.
摘要:
An integrated circuit (IC) includes mechanisms for adjusting or setting the gate bias of one gate of one or more multi-gate transistors. The IC includes a gate bias generator. The gate bias generator is configured to set a gate bias of one gate of the one or more multi-gate transistors within the IC. More specifically, the gate bias generator sets the gate bias of the transistor(s) so as to trade off performance and power consumption of the transistor(s).
摘要:
A semiconductor device and method for electrostatic discharge protection. The semiconductor device includes a first semiconductor controlled rectifier and a second semiconductor controlled rectifier. The first semiconductor controlled rectifier includes a first semiconductor region and a second semiconductor region, and the second semiconductor controlled rectifier includes the first semiconductor region and the second semiconductor region. The first semiconductor region is associated with a first doping type, and the second semiconductor region is associated with a second doping type different from the first doping type. The second semiconductor region is located directly on an insulating layer.
摘要:
A process for fabricating field effect transistors with lightly doped drain (LDD) regions having a selected width includes a method of optically detecting the width of spacers used to mask the LDD regions during the source and drain implant and a method of electrically determining (confirming) the width of the LDD regions. In the optical method, reference structures are formed concurrently with the fabrication of the gates for FETs, a spacer material is formed on the substrate, the gates and the reference structures, the spacer material is etched away and the width of the spacers is optically detected by aligning the edges of spacers extending from two reference structures separated by a known distance. In the electrical method, the width is determined by defining a test area with known dimension, forming both N.sup.+ and N.sup.- regions in the test area, measuring the resistance across the test area, calculating the resistance of the N.sup.+ and N.sup.- regions, and calculating the width of the N.sup.- region from the resistance of the N.sup.- region. The electrically determined width is compared with the desired LDD region width, and the difference between the electrically determined width and the desired width is used to adjust the distance between the reference structures for a subsequent processing run.