Linear Technology Corporation

Indigo Systems Corporation

Linear Technology Holding LLC

Analog Devices International Unlimited Company

Jeffrey Lynn Heath

David McLean Dwelley

Kirk Su

Heath Stewart

John Arthur Stineman, Jr

Ryan Huff

Andrew J Gardner

Michael Paul

Harry Joseph Kleeburg

Trevor Wayne Barcelo

William J Parrish

Naseem Y Aziz

David Stover

Theodore R Hoelter

Jeffrey D Frank

Joseph Kostrzewa

Robert Peter Jurgilewicz

Jacob Herbold

Clayton Reynolds Stanford

Steve Robbins

Kaung Zin Htoo

Eugene Timothy Fitzgibbons

George H Poe

John Stineman

Ryan Charles Huff

Kirk Tzukai Su

David M Stover

David Mclean Dwelley

John Arthur Stineman

An ideal diode circuit is described which uses an NMOS transistor as a low-loss ideal diode. The control circuit for the transistor is referenced to the anode voltage and not to ground, so the control circuitry may be low voltage circuitry, even if the input voltage is very high, referenced to earth ground. A capacitor is clamped to about 10-20 V, referenced to the anode voltage. The clamped voltage powers a differential amplifier for the detecting if the anode voltage is greater than the cathode voltage. The capacitor is charged to the clamped voltage during normal operation of the ideal diode by controlling the conductivity of a second transistor coupled between the cathode and the capacitor, enabling the circuit to be used with a wide range of frequencies and voltages. All voltages applied to the differential amplifier are equal to or less than the clamped voltage.

Self-biasing ideal diode circuit

One or more Power Sourcing Equipment (PSE) are coupled to points in a network of interconnected nodes. Each node has a first port and a second port. Assume the first port of a first node is receiving DC power from the PSE. The first node, at its second port, then detects an electrical signature from a first port of an adjacent second node. If the proper electrical signature is presented by the adjacent second node, the powered first node closes a switch to pass power between its first port and second port to power the second node via the first port of the second node. All nodes in the network are then sequentially powered up in this manner. If there is a fault between the first node and the second node, the second node will be powered by another node connected to the second port of the second node.

Fault-tolerant power network

An ideal diode circuit is described which uses an NMOS transistor as a low-loss ideal diode. The control circuit for the transistor is referenced to the anode voltage and not to ground, so the control circuitry may be low voltage circuitry, even if the input voltage is very high, referenced to earth ground. A capacitor is clamped to about 10-20V, referenced to the anode voltage. The clamped voltage powers a differential amplifier for the detecting if the anode voltage is greater than the cathode voltage. The capacitor is charged to the clamped voltage during normal operation of the ideal diode by controlling the conductivity of a second transistor coupled between the cathode and the capacitor, enabling the circuit to be used with a wide range of frequencies and voltages. All voltages applied to the differential amplifier are equal to or less than the clamped voltage.

A composite transconductance amplifier is formed using a single transconductance amplifier with its output connected to a load via one or more resistors in series. The single transconductance amplifier has a linear transconductance (gm). As the current through the series resistors is increased, the voltage drops across the nodes of the resistors increase. Control terminals of separate drive circuits are connected to the various nodes and successively turn on as the current from the single transconductance amplifier slews more positive. Thus, the effective gm of the composite transconductance amplifier is based on the gm of the single transconductance amplifier and the currents contributed by the successively enabled drive circuits. Therefore, the gm is nonlinear. Pull-down drive circuits are also connected to the resistor nodes to successively pull down the current as the output from the single transconductance amplifier slews negative. The composite transconductance amplifier has low quiescent current.

Transconductance amplifier with nonlinear transconductance and low quiescent current

In a method performed by a PoE system, a PSE provides data and operating voltage over Ethernet wires to a PD. Before the full PoE voltage is supplied, the PSE generates a low current signal received by the PD. A circuit in the PD, connected across its input terminals, has a characteristic analog response to the PSE signal corresponding to the PD's PoE requirements, such as whether the PD is a Type 1 or Type 2 PD. The circuit may be a certain value capacitor, zener diode, resistor, or other circuit. The PSE may generate a fixed current, fixed voltage, or time varying signal. Upon the PSE sensing the magnitude of the analog signal response at a particular time, the PSE associates the response with the PoE requirements of the PD. The PSE then applies the full PoE voltage in accordance with the PD's PoE requirements.

Detection and classification scheme for power over Ethernet system

The invention pertains to systems where DC power is supplied by a PSE to a PD, such as over differential data wire pairs. IEEE standards require a minimum current to be drawn from the PD in order for the PSE to continue supplying the DC voltage. If the PD is in a low power mode, the PSE will normally discontinue supplying the DC voltage, which then requires a new detection and classification routine for powering up again. To avoid this, a 'maintain power signature' controller provides a periodic current pulse by a current source connected between the PD input and the PD's full bridge rectifier. Any droop in the DC voltage that reverse biases the full bridge rectifier while the PD is in its low power mode will not affect the current pulse, so the PSE continues to supply the DC voltage.

Maintain power signature controller at power interface of PoE or PoDL system

A PoDL system includes a PSE connected via a wire pair to a PD, where differential data and DC power are transmitted over the same wire pair. Typically, low voltage/current detection and classification routines are required upon every powering up of the system to allow the PD to convey its PoDL requirements to the PSE. Various techniques are described that simplify or obviate such start-up routines or enable increased flexibility for the PoDL system. Such techniques include: ways to specify a particular PD operating voltage; ways to disable the PD's UVLO circuit during such routines; using opposite polarity voltages for the two routines; using voltage limiters or surge protectors to convey the PoDL information; detecting loop resistance; using a PSE memory to store previous results of the routines; and powering the PD communication circuit using the wire pair while the PD load is powered by an alternate power source.

Power over data lines system providing a variable voltage to powered device

Power over data lines system using voltage clamp in PD for detection or classification

In one embodiment, a PoDL system includes a PSE that uses high side and low side circuit breakers that uncouple the PSE voltage source from the wire pair in the event that a fault is detected. Faults may include a temporary short to ground, or to a battery voltage, or between the wires. The breakers perform an automatic retry operation in the event the fault has been removed. The voltages on the wires in the wire pair may be monitored to determine whether the voltages are within a normal range or indicative of a fault condition. Other embodiments are disclosed.

Circuit architectures for protecting against PoDL wire faults

Circuits and techniques are described for detecting a ground fault leak between the PSE and the PD. Prior to PoDL voltage being applied to the PD, a test switch is temporarily closed for sensing a voltage drop in a loop between the positive terminal of the PSE voltage source and any ground leakage path between the PSE and the PD. If the resistance of the ground leakage path is below a certain threshold, a fault is declared. A similar test may be performed without a test switch by supplying a known test current through the loop and sensing the voltage drop. Another test is to connect the positive terminal of the PSE voltage source to the loop and sense the resulting current. After the full PoDL voltage is applied to the PD, a ground fault may be detected by sensing the equivalence between the source and return PSE currents.

Detecting ground isolation fault in ethernet PoDL system

Detection and classification scheme for power over ethernet system

Power over data lines detection and classification scheme

An anemometer and method for analyzing fluid flow is described. In one embodiment, a transistor sensor is heated by applying power to cause its base-emitter junction to rise from an ambient first temperature to a second temperature. The power is removed, and the Vbe is measured at intervals as the junction cools. The Vbe equates to a temperature of the junction. The temperature exponentially decreases, and the time constant of the decay corresponds to the fluid flow velocity. A best fit curve analysis is performed on the temperature decay curve, and the time constant of the exponential decay is derived by a data processor. A transfer function correlates the time constant to the fluid flow velocity. The transistor is thermally coupled to a metal rod heat sink extending from the package, and the characteristics of the rod are controlled to adjust the performance of the anemometer.

Method of manufacturing an anemometer used for determining a fluid flow

A Power Over Data Lines (PoDL) system includes Power Sourcing Equipment (PSE) supplying DC power and differential Ethernet data over a single twisted wire pair to a Powered Device (PD). Due to start-up perturbations, PD load current variations, and other causes, dV/dt noise is introduced in the power signal. Such noise may be misinterpreted as data unless mitigated somehow. Rather than increasing the values of the passive filtering components conventionally used for decoupling/coupling the power and data from/to the wire pair, active circuitry is provided in the PSE, PD, or both to limit dV/dt in the power signal. Such circuitry may be implemented on the same chip as the PSE controller or PD controller. Therefore, the sizes of the passive components in the decoupling/coupling networks may be reduced.

PoDL system with active dV/dt and dI/dt control

Novel system and methodology for adjusting a current limit threshold in a Power over Ethernet (PoE) system in accordance with requirements of a Powered Device (PD). A system for supplying power to a PD over a communications link has a requirement determining circuit for determining a PD's requirement, and a control circuit for setting a parameter restricting an output signal of the PSE in accordance with the determined PD's requirement. The control circuit may set a current limit threshold of the PSE and/or the PD in accordance with the determined PD's requirement, such as a power requirement.

Growing community of inventors

Santa Barbara, CA, United States of America

Jeffrey Lynn Heath