USD

Power-over-Ethernet (PoE) - New IEEE 802.3bt Standard Boosts the Technology for IoT Applications

By Dany Haikin

Contributed By Digi-Key Electronics

Many articles and blogs have been written about Power-over-Ethernet (PoE) technology and its concept. An example is Digi-Key’s “An Introduction to Power-over-Ethernet” article.

This article analyzes and emphasizes the new features introduced by the latest IEEE 802.3bt standard to today’s IoT (Internet of Things) world where everything (“the things”) is connected, controlled and monitored via Internet.

IEEE 802.3bt enhancements

The first and the most important improvement in the 802.3bt standard is the capability to transfer much more power to edge devices (powered devices or PDs) - 71.3 W, while sending 90 W from the power sourcing equipment (PSE) side.

Secondly, it supports up to a 10 Gb/sec transfer rate of the networking devices over Cat5e cable.

These two enhancements in power delivery and signal speed enable innumerous new IoT and especially IIoT (Industrial IoT) power hungry and high-speed devices to be powered by PoE technology. The following are some examples of these new applications:

  • Professional audio
  • Digital signage
  • 5G small cell radio units (Mobile infrastructure: 3G, 4G, 5G technologies)
  • 802.11ac Wireless Access Points (WAPs)
  • High-throughput Wireless Local Area Networks (WLANs)
  • Industrial access control
  • Lighting
  • Smart home
  • Building/factory automation
  • Point of Sale (POS) terminals
  • Information kiosks
  • Outdoor IP cameras with heaters
  • Monitors/laptops
  • Digital ceiling lighting

New features of IEEE 802.3bt

The new PoE IEEE 802.3bt standard defines several new features and improvements in comparison to the older 802.3at standard that result in power savings and increased efficiency; thus, again enabling more edge devices to use PoE technology.

In principle, an IoT platform consists of four building blocks:

  • Sensing/monitoring
  • Processing (MCU)
  • Connectivity (wireless or wired)
  • Power management

The new features/improvements of the IEEE 802.3bt standard are beneficial for the power management block of IoT noted above. Four of these new features/improvements include short maintain power signature (MPS), autoclass, single/dual signature of the PD, and extended power to the PD, which are discussed further in the paragraphs below.

Short Maintain Power Signature (MPS) - MPS is the minimum power consumption drawn by a PD which will keep the PD alive and not be disconnected by the PSE. The PSE is required to remove power when the MPS is absent for at least 400 ms, ensuring that disconnected cables do not remain powered.

Additionally, almost all PDs in IoT applications have low power or sleep modes. Such PDs must draw a higher current in order to remain powered, abusing the idea of having low power standby mode. Short MPS addresses this issue by reducing the duty cycle and how long a power signature must be generated to maintain a power connection. This modification improves the minimum standby power by a factor of 10, enabling IoT edge devices to be powered with PoE and have acceptable standby power.

In IoT applications where a large number of the devices use PoE, like LED lighting for example, reducing the standby power is crucial.

Autoclass – Autoclass allows the allocation of the PSE power supply budget to the PD to be better optimized. In essence, the PSE is “measuring” the Ethernet cable losses and the power consumption of the connected PD throughout a defined period, and “knows” to provide “actual” power to this PD rather than a higher “assigned” power defined by the PD class. This allows the same PSE to power more PDs, and therefore more IoT edge devices.

Single/dual signature of the PD – The IEEE 802.3bt standard supports two PD constructions: single-signature PD and dual-signature PD. The PSE must support both single-signature and dual-signature PDs.

Dual-signature devices enable applications that require up to the same maximum power level as single-signature devices and provide the additional flexibility of different and isolated load configurations. An example of this can be an outdoor surveillance camera that needs to be powered along with a heater or cooling fan to control extreme temperature conditions. Another example could be IIoT applications with redundant circuitry used for reliability and safety purposes that are powered alternately, but not at the same time.

Figure 1 shows dual/single-signature concepts.

Diagram of single/dual-signature conceptsFigure 1: Single/dual-signature concepts (Image source: Microchip)

More technical information about IEEE 802.3bt dual-signature PDs can be found on the Ethernet Alliance (EA) website.

Extended power to the PD – The IEEE 802.3bt standard defines a maximum power of 90 W that a PSE device can send and 71.3 W that a PD device can receive. This power drop from PSE to PD takes into account the maximum loss of 19 W through the entire cable length of the maximum 100 meters defined by the Ethernet standard. With the new IEEE 802.3bt standard, the PD can measure the cable resistance, calculate the power that will be lost in the cable, and provide power high enough to compensate for the “wasted” 19 W maximum power dissipated by the 100 meter cable. If the distance from PD to PSE is less than 100 meters, more than 71.3 W can be provided to the PD. For example, if the cable length is in the range of 2 – 5 meters, the power that the PD can receive from the PSE can be close to the 90 W that the PSE is sending.

IEEE 802.3bt power efficiency improvements

Though not defined explicitly by the 802.3bt standard, but close to its ratification and in the spirit of the power management efficiency and the IoT application requirements, several leading PoE IC vendors have improved the design of their chips to address power efficiency.

Diagram of PoE block topologyFigure 2: PoE block topology (Image source: Microchip)

Before looking at Figure 2, the functionality of both the PSE and PD should be defined.

For the PSE, the functionality requirements can be briefly summarized as follows:

  • Detect a valid PD
  • Classify the power capabilities of the PD
  • Provide 4 W to 90 W of power at 44 to 57 volts to the PD
  • Perform power optimization and allocation
  • Perform fault monitoring and disconnection when necessary
  • Shutdown the power to the appropriate port if an undercurrent condition is detected
  • Provide overvoltage protection
  • Provide isolation from switch circuitry

Similarly, a summary of the functionality of a PD is as follows:

  • Provide polarity protection
  • Provide signatures for detection and correct classification
  • Perform power optimization
  • Provide isolation
  • Provide for an optional bias for DC/DC startup
  • Convert 57 V down to the required regulated supply voltage used by the application

As can be seen in Figure 2, the power from the PSE is sent through the Ethernet cable to the PD. The PDs diode bridge chip then rectifies the cable voltage. In 2-pair PoE systems, the voltage can be delivered over either the data pairs or the spare pairs, but not both. In 4-pair PoE systems defined by IEEE 802.3bt, all pairs are powered.

Therefore, two bridges are required inside the PD (Figure 3).

Diagram of two bridges inside the PDFigure 3: Two bridges inside the PD (Image source: Analog Devices/Linear Tech)

The conventional diode bridge solution has several drawbacks:

  • High power loss caused by the cable voltage drop
  • High heat dissipation
  • Requires additional thermal design considerations

Because of the above drawbacks, the usage of traditional diode bridges in many IoT applications is very problematic, if not impossible.

A more efficient solution than the diode bridge is the so-called IdealBridge, first introduced by Microsemi (now Microchip). This solution is an N-channel MOSFET-based bridge with a controller.

The differences between a dual conventional diode bridge and a single IdealBridge are shown in Figure 4.

Diagram of dual conventional diode bridge vs. a single IdealBridge™Figure 4: Dual conventional diode bridge vs. a single IdealBridge™ (Image source: Microchip)

The advantages of the IdealBridge include:

  • The fully integrated solution decreases the BOM - Saves PCB space and simplifies implementation
  • Self-driving circuitry for MOSFETs
  • Low RDS(ON), low power consumption
  • Power efficiency maximization – delivers higher output power and voltage
  • Dramatically reduces heat dissipation, eliminating thermal design issues and the need for a heatsink
  • Works with 2-Pair and 4-Pair PoE applications
  • Compatible with IEEE 802.3xx standards

Microsemi/Microchip introduced the first “IdealBridge™” solution with the PD70224. Other similar solutions from other vendors are Analog Devices/Linear Tech’s LT4321, ON Semiconductor’s FDMQ8205A 1-channel ideal diode (not a bridge), and STMicroelectronics’ PM8805 integrated solution (IdealBridge integrated into a PD IC chip).

Conclusion

The latest IEEE 802.3bt standard adds new features to PoE technology and improves the existing ones. These features broaden the range of the edge devices that can be connected using the PoE concept, thus supporting many new IoT applications.

To support non-PoE infrastructures, there are various intermediate solutions like midspans/injectors and power splitters. However, it is important to note that IEEE 802.3bt is a rather new standard and many vendors have offered products in this area prior to its ratification at the end of 2018. In order to leverage the advantages of the IEEE 802.3bt new features and maintain interoperability between the vendors, the parts and the products must be qualified for this IEEE 802.3bt standard, which should be explicitly stated in the data sheets.

Disclaimer: The opinions, beliefs, and viewpoints expressed by the various authors and/or forum participants on this website do not necessarily reflect the opinions, beliefs, and viewpoints of Digi-Key Electronics or official policies of Digi-Key Electronics.

About this author

Dany Haikin

Dany Haikin, Regional Application Engineer & Technical Support Manager at Digi-Key Electronics, has been with the company since 2012 with primary responsibility for applicational and technical support of Digi-Key customers in Israel and the Middle East. He has over 33 years of experience in the electronics industry and holds an Electrical Engineering degree from the Technion – Israel Institute of Technology.

About this publisher

Digi-Key Electronics

Digi-Key Electronics, based in Thief River Falls, Minn., is a global, full-service provider of both prototype/design and production quantities of electronic components, offering more than six million products from over 750 quality name-brand manufacturers at Digi-Key.