摘要:
A voltage translator uses an n-channel translator transistor to translate an input voltage at its drain to an output voltage at its source. The gate and substrate of the translator transistor are each biased by charge pumps. A reference transistor is also biased by the charge pumps. A reference input voltage is translated to a reference output voltage by the reference transistor. The reference output voltage is compared to a target output voltage by comparators. When the reference output voltage is below the target, the gate charge pump is turned on, raising the gate voltage to both the reference and translator transistors. The higher gate voltage VGATE raises the output voltage VOUT since VOUT=VGATE-VT for a transistor in saturation. When the reference output voltage is above the target, the substrate charge pump is turned on, pulling the substrate bias voltage below ground. The body effect causes the transistor threshold VT to increase as the substrate is pumped. The higher threshold lowers the output voltage. Once the reference output voltage reaches the target, the charge pumps turn off. The input voltage can toggle high and low since the reference transistor sets the gate and substrate voltages.
摘要:
A bus switch has an n-channel bus-switch transistor that connects an input-bus signal to an output bus. A gate protection circuit prevents undershoots on the inputs from coupling to the output when the bus switch isolates the buses. The gate of the bus-switch transistor is driven to ground during isolation mode. When a high-to-low transition of the input-bus signal is detected, a pulse generator generates a pulse. The pulse disconnects the gate from ground. A connecting n-channel transistor with its gate connected to ground connects the gate to the input-bus signal when the undershoot pulls the input-bus signal below ground. Internal circuitry is isolated from the below-ground gate by an isolating n-channel transistor that has its gate driven by the input-bus signal during the pulse. A substrate bias generator is used for N-well processes, but P-well processes use a well protection circuit. The P-well under the bus-switch transistor is disconnected from ground during the generated pulse. Another n-channel connecting transistor with its gate grounded connects the P-well to the input-bus signal when the undershoot occurs. The protection circuits are only enabled when the bus switch transistor is in the isolation mode and a low-going transition of the input-bus signal is detected.
摘要:
A bus switch uses both n-channel and p-channel transistors in parallel to connect two busses. The bus switch can be used on a network card to be plugged into a running network. During hot or live insertion of the network card into a live or hot bus, the network card and a bus switch are in a powered down state. Although n-channel transistors are normally off when the power is off, p-channel transistors can conduct. The hot bus could be disturbed when the bus switch is first connected since the p-channel transistor conducts when its gate is powered down to zero volts. A p-n junction from the p-channel transistor's drain to its substrate can become forward biased, drawing current from the hot bus. These problems are avoided by an isolation circuit that operates without power from a power supply. Instead, a high voltage from the hot bus is routed to the gate of the p-channel transistor, keeping the p-channel transistor turned off during hot insertion. The high voltage frog the hot bus is also routed to the substrate or N-well under the p-channel transistor, preventing the p-n junction from forward biasing. Thus a full CMOS bus switch can be used for live insertion, even when powered down.
摘要:
A bus switch is constructed from an n-channel transistor. The gate terminal of the n-channel transistor is boosted above the power supply (Vcc) to increase current drive and reduce the channel resistance of the bus switch. The gate terminal is connected to a boosted node. When the bus switch is turned on, a pulse is generated to drive the boosted node from ground to Vcc. The boosted node is also an input of a delay line. After a delay through the delay line, the pulsed pull-up is turned off. Feeding the boosted node to the delay line allows the pulse to be self-timed. The delay line then drives the back-side of a capacitor from ground to Vcc. This voltage swing is coupled through the capacitor to the boosted node, driving the boosted node about 1.3 volts above Vcc. A small keeper transistor supplies a small current to the boosted node to counteract any leakage. This leaker transistor is connected to a charge pump, and the delay line that enables this keeper transistor is also connected to the charge pump. A precise sequence of events boosts the gate voltage of the bus switch above Vcc without drawing large currents from the charge pump.
摘要:
A synchronizer that has reduced latency is used for synchronizing and conditioning a clock-enable signal to a free-running clock. Once the clock-enable signal is synchronized it is used to enable and disable gating of the free-running clock to a gated clock that suspends pulsing in response to the clock-enable signal. A first-stage flip-flop is `meta-stable hardened` to reduce the probability of it becoming meta-stable. Gating on the clock and the clear inputs reduces the chance that simultaneous inputs will violate the timing of the flip-flop and thus cause metastability. A clear pulse is generated to clear the flip-flop. The clear pulse skews the flip-flop to be more likely to become metastable for one edge of the asynchronous input than for the other edge. The settling time to the second stage flip-flop is then adjusted to account for this skew in metastability. Settling time in the second stage is increased for the edge that is more likely to become metastable. Clock-enable conditioning to prevent partial output pulses is merged with the synchronizing function to further reduce latency.
摘要:
A network switch uses a simple switch core of analog MOS transistor switches. The switch core is surrounded by many media-access controllers (MAC's) which buffer the data through the switch core. Multiple connections through the switch core may be made between different pairs of MAC's. Just one signal path through the switch core is needed per connection as a second path through the switch core for the clock is not needed. The clock is not encoded with the data, so PLL's are not needed for clock recovery. Data skew is instead measured for each packet transmitted through the switch core. A start flag is added to the packet by a source MAC as a packet header before being transmitted through the switch core. The start flag is a unique sequence which is detected by the destination MAC and triggers measurement of the data skew of the received start flag to the local clock. The measured data skew is then used to compensate for the rest of the packet. Because dynamic de-skewing is performed for each packet, any packet from any source MAC can be received. This allows freedom in MAC placement and enables construction of larger switches. The switch may have the MAC's physically separated from each other by greater distances since clocks can be locally generated. The clocks are independent from each other except for having the same frequency.
摘要:
A substrate bias generator for an integrated circuit has a charge pump driven by an oscillator. The oscillator is enabled and disabled to save power and control the voltage-level itself for the substrate bias. An enabling circuit senses the substrate voltage and enables the oscillator when the substrate voltage rises above a bias set by a programmable reference voltage. The enabling circuit which senses the voltage on the substrate draws no active current from the substrate. The sensing circuit includes a transistor with only its bulk terminal connected to the substrate; the source, gate, and drain of this sensing transistor are not connected to the substrate. A differential comparator compares the output of the sensing transistor to the programmable reference voltage and enables the oscillator when the sensing transistor output is lower than the reference voltage. The sensing transistor attenuates large swings in the substrate voltage to provide the differential comparator with a small voltage swing which keeps the differential comparator operating near its optimum design point. Since no active current is drawn from the substrate when sensing the substrate voltage, no IR voltage drops can develop from the enabling and sensing circuit. Thus latch-up immunity is improved and substrate noise is reduced.
摘要:
A jitter attenuator receives data and a receive clock extracted from an input data stream. A transmit clock is generated for retransmitting the data. The transmit clock has less jitter than the receive clock but has the same average frequency. An elastic buffer or FIFO is necessary to buffer the data. The receive clock is divided into a series of write clocks for writing data into the elastic buffer, and the transmit clock is also divided into a series of read clocks for reading data from the elastic buffer. A series of multi-phase clocks is used to generate the transmit clock. The multi-phase clocks all have the same frequency but are offset in phase from one another. A phase selector, under control of a counter, selects one of the multi-phase clocks to be the transmit clock. The counter is incremented or decremented by a phase comparator. The phase comparator compares the phase of one of the write clocks to the phase of one of the read clocks. The counter is incremented when the phase of the write clock lags the read clock, selecting a multi-phase clock with a more retarded phase, but the counter is decremented when the write clock leads the read clock, selecting a multi-phase clock with a more advanced phase. Thus the phase of the transmit clock is adjusted by the phase comparison of the write and read clocks for the elastic buffer. The elastic buffer is forced to half-full by comparing a write and read clock that are separated by half the capacity of the buffer. The phase, rather than the frequency, is adjusted, eliminating the feedback to an external VCO, allowing the jitter attenuator to be integrated on a single silicon substrate.
摘要:
A jitter attenuator receives data and a receive clock extracted from an input data stream. A transmit clock is generated for retransmitting the data. The transmit clock has less jitter than the receive clock but has the same average frequency. An elastic buffer or FIFO is necessary to buffer the data. The receive clock is divided into a series of write clocks for writing data into the elastic buffer, and the transmit clock is also divided into a series of read clocks for reading data from the elastic buffer. A series of multi-phase clocks is used to generate the transmit clock. The multi-phase clocks all have the same frequency but are offset in phase from one another. A phase selector, under control of a counter, selects one of the multi-phase clocks to be the transmit clock. The counter is incremented or decremented by a phase comparator. The phase comparator compares the phase of one of the write clocks to the phase of one of the read clocks. The counter is incremented when the phase of the write clock lags the read clock, selecting a multi-phase clock with a more retarded phase, but the counter is decremented when the write clock leads the read clock, selecting a multi-phase clock with a more advanced phase. Thus the phase of the transmit clock is adjusted by the phase comparison of the write and read clocks for the elastic buffer. The elastic buffer is forced to half-full by comparing a write and read clock that are separated by half the capacity of the buffer. The phase, rather than the frequency, is adjusted, eliminating the feedback to an external VCO, allowing the jitter attenuator to be integrated on a single silicon substrate.
摘要:
Distortions of both amplitude and phase along a transmission line are compensated for by a trace canceller inserted between a transmitter and a receiver. The trace canceller has an equalizer that compensates for a trace length between the transmitter and the trace canceller. A variable gain amplifier between the equalizer and an output buffer has its gain controlled by an automatic gain control circuit that compares low-frequency swings of the input and output of the trace canceller. The gain of the variable gain amplifier is reduced to prevent the output buffer from saturating and clipping peak voltages on its output. Thus both the variable gain amplifier and the output buffer remain in the linear region. Training pulses from the transmitter are passed through the trace canceller without clipping of peak voltages, allowing the transmitter and receiver to adjust transmission parameters to best match the transmission line.