SPE Redefines PoE

Image Source: xiaoliangge/Stock.adobe.com

By Bill Schweber for Mouser Electronics

Published June 21, 2022

Power over Ethernet (PoE) is just what the name implies: a way to send substantial and valuable amounts of power over the same copper cabling that carries the Ethernet data. It enables remote, Internet-addressable devices to receive their operating power and leverage the benefits of Ethernet connectivity. This dual power/data capability is useful for Internet of Things (IoT) applications, such as sensors and HVAC controllers, building security systems (cameras, elevators, fire panels, smart locks, and entry/exit), and essential office functions (enhanced area lighting, even basic voice telephones).

PoE is getting an increasing amount of attention and design-in from smart building and industrial-automation engineers for many reasons, three of which stand out:

• First, with the formal approval of the third-generation standard for PoE, it can supply a substantial amount of power – approximately 100W – to support a wide range of applications, peripheral devices, and loads.
• Second, the approval of PoE's single-pair Ethernet (SPE) standard means that PoE no longer requires more costly multipair cabling and somewhat fragile connectors and now is a good installation, performance, and cost match for smart buildings and industrial applications.
• Third, component vendors have introduced the needed ICs and passive components, supported with validated reference designs and embedded firmware so that OEMs can design and manufacture the required PoE interface systems and circuitry.

Note that PoE is not unique in combining power and data over a single wire or cable, as there’s a long history of engineers doing similar combinations in other settings. For example, USB has become both a data bus bit and a DC power bus with complicated signaling, handshaking, and sequencing. As a result, USB, in its latest iteration, can provide significant power to a load, independent of its high-speed data function.

In the pre-digital days, engineers devised a way to send power to the preamplifier built into many types of microphones by using the same cable which carried the audio signal back to the main amplifier, in a technique called “phantom power.” Similarly, the DC power for the low noise block (LNB) RF preamplifier mounted on an outdoor small-aperture dish antenna is provided via the same coaxial cable, which sends the boosted RF signal back to the system’s RF front end.

What are the benefits of using PoE for smart buildings and industrial applications? It comes down to basics: it saves on wiring cost and complexity, it reduces the number of distinct “networks” needed (one for data, one for power); it is a safe, low voltage (48VDC nominal); it does not require an electrician to install or have to meet building codes for higher-voltage wiring, and it eases the path to widely dispersed devices in a residential or commercial building or plant. It’s a better alternative to installing many local AC/DC supplies and their AC wiring, batteries alone, or extra DC-only power runs.

Standards Define the PoE World

Key to the viability of PoE as Ethernet connectivity and power delivery link are standards defined by the Institute of Electrical and Electronics Engineers (IEEE). The first iteration of PoE, designated IEEE 802.3af, could deliver only about 15W. At the same time, a subsequent version a few years later, informally called PoE+ but formally IEEE 802.3at, increased this to 25W. The most recent version, known as PoE++ before being formalized as IEEE 802.3bt, increased the power to 90W.

Each step up requires a somewhat more complicated type and cable-pair arrangement for the multipair cable (Cat 3, Cat 5, Cat 5e, or Cat 6 A) and associated RJ-45 connector and its wiring. These three PoE standards are upward compatible, so the lower-power connections and loads can be used on the higher-power cabling and connections; downward power/connector compatibility is not supported.

A recently approved PoE standard IEEE 802.3cg-2019 (IEC 63171-6) for SPE is even more promising for smart buildings of all types. This dramatically simplifies and reduces the cost of the cabling and connectors while supporting data rates up to 10Mbps and is adequate for many dispersed or outlying applications.

PoE: Internally Sophisticated

It’s one thing to say that it is a good idea to get double duty from a single copper interconnect. It’s another thing to figure out how to do it while not corrupting or slowing down the data, while simultaneously ensuring the integrity and performance of the PoE system in the event of a problem, fault, or even a short circuit at one of the powered devices. Imposing DC power onto the copper wires at a source end and then having a load device extract it at the other end requires a complicated and sophisticated arrangement, especially as this must be transparent to the user who is connecting and disconnecting PoE-enabled devices.

For PoE, the three essential components needed are:

1. The equipment that delivers power to the cable (often referred to as a PSE for power sourcing equipment). The PSE may either be an "Endspan” (a PoE switch) or a “Midspan” (a PoE power injector) located along the Ethernet cable, closer to the loads.
2. The device that receives its power from the cable (the powered device, or PD). Typical PDs include IP-enabled cameras, wireless access points, lighting units, security devices, and HVAC controllers, for example.
3. The cable itself includes various "CAT-X” designations to support the different PoE standards.

The PSE would usually be a PoE-enabled network switch or a midspan power injector, patched in to add PoE capability to a non-PoE network switch channel, or similar (Figure 1).

Figure 1: A PoE network switch can deliver power to a PD, such as an IP camera, over its data cable;A midspan PSE can be used to add PoE capability to standard network switch channels. (Image source: Veracity/UK)

This high-level diagram is just that: a “big-picture” overview. A PoE system requires a significant amount of circuitry (ICs and passive components) within the PSE and suitable CAT cabling to make it all look like a "plug and play" arrangement to someone who plugs in a PoE-enabled powered device and expects it to function. This circuitry must integrate the supplied DC power with the Ethernet data signals at the PSE and then separate them for use at the PD (Figure 2).

Figure 2: A PoE link requires considerable support circuitry in addition to the physical-layer IC (PHY including transformers to combine data and power at the PSE and split them at the PD, along with bridges to rectify the AC power into the DC rails needed. (Image source: Veracity/UK)

A complicated back-and-forth identification, query, and response protocol are needed between the PD and the PSE to initiate, manage, and maintain the PoE link. It takes many state and timing diagrams to fully describe the many normal operating sequences and accommodate all the “abnormal” but possible conditions. These sequences inform the PSE and the PD of the system status, their status, what the PD needs, what the PSE can provide, and checks for out-of-bounds conditions.

The status and power needs of the PD must be considered in the larger PoE system context. For this reason, the PoE specification includes additional features for efficiency, reliability, and safety, and “classification” is one of these features. As part of the start-up process, when a PoE connection is made, the PD can “advertise” its power class to indicate how much power it requires to operate to ensure the PSE can provide it. There are five power classes in total, but only three different bands of power level into which devices can fall (Table 1). The difference between what the PSE can deliver in each classification and what the PD needs to operate is due to unavoidable I²R power loss in the cable.

Table 1: PSE and PD power limits by class (Image source: Veracity/UK)

Fortunately, the PoE system designer does not need to delve into the PoE operational details as ICs are now available that embed and implement them. These ICs manage the power flow, sequencing, protocol, and other considerations which need to be executed and monitored for an effective and reliable PoE power and data link.

For example, the Analog Devices ADIN1110 is an ultralow power, single port 10BASE-T1L transceiver design for industrial Ethernet applications (Figure 3). It is compliant with the IEEE 802.3cg-2019 Ethernet standard for long reach, 10Mbps SPE. This IC is designed for edge node sensors and field instruments deployed in building, factor, and process automation. It operates from a single power supply rail of 1.8V or 3.3V and is available in a 40-lead, 6 mm × 6 mm lead frame chip package (LFCSP).

Figure 3: The Analog Devices ADIN1110 MAC-PHY is compliant with the IEEE 802.3cg-2019 Ethernet standard for long reach, 10 Mbps single pair Ethernet, and is designed for edge node sensors and field instruments. (Image source: Analog Devices)

What about SPE?

SPE provides new options for implementing a PoE system, as it is a better match to the needs of smart buildings. It has advantages in this setting compared to traditional Ethernet and the various CAT cables. Among these are thinner, less costly cables and connectors that are easier to manage and pull, reduced space requirements and lower weight, smaller bending radii for easier installation in cramped areas, and reduced use of resources such as copper and insulation.

SPE is not designed to replace the existing Ethernet infrastructure. At present, the maximum speed for SPE is 10Mbps (longer distance) to 1Gbit/sec (shorter distance) and only up to a distance of forty meters. In contrast, 8-wire Ethernet can extend up to one hundred meters at ten and more Gbit/sec (Figure 4). 

Figure 4: SPE greatly reduces the Ethernet cabling BOM and installation costs, but with a reduction in maximum available throughput and data rate. (Image source: HARTING)

[Note:  SPE was initially developed to meet the demands of the automotive industry — smaller, lighter cables and connectors for transmitting data from the numerous sensors involved in adaptive cruise control, parking assist, autonomous driving, and other onboard systems. The lower data rate, tightly constrained, and shorter distances of automotive applications were a good fit for early SPE developments.]

Unlike conventional Ethernet, where the individual user ports radiate from a central switch, SPE supports a daisy-chain topology. Adding more ports to a conventional Ethernet arrangement requires more ports and a larger switch, while maximum latency – the time lag and uncertainty for data between the port and the switch – is limited and known in advance.

In contrast, daisy-chaining allows for easy expansion to handle more ports, but latency increases as the chain are extended, especially for ports and devices far along the chain (Figure 5). However, for typical SPE applications, the latency is still short enough, while the connectivity benefits are worth it. The ability to easily extend the link to new ports via this topology is a good fit with SPE's target installations.

Figure 5: While “conventional” Ethernet requires that each port be connected directly to a switch (right), SPE allows for daisy-chain interconnection and extension to new ports. (Image source: Texas Instruments)

SPE does not define a single solution with the same structure for every application (automotive versus smart building versus industrial). Instead, it is a foundation technology in which different combinations of parts can be selected and integrated to create an optimized solution for what is also called Power over Data Line (PoDL). Different SPE protocols with different transmission speeds, link lengths, and topologies can be selected for varying applications. These are SPE variations defined by 10BASE-T1S, 10BASE-T1L, 100BASE-T1, and 1000BASE-T1 standards.

Another part of what makes SPE attractive for its target applications is the rugged, low-cost connector options. For example, HARTING T1 connectors for building cabling have an SPE mating face as defined by IEC 63171-1 (Figure 6). These connectors have a MICE rating (for Mechanical robustness, Ingress IP degree, Climatic resistance, and Electromagnetic safety) of M1I1C1E1, which is suitable for the environment of an office building.

Figure 6: The SPE connector pair shows the simple, rugged implementation which the two-wire standard supports. (Image source: HARTING)

Filling out the PoE BOM

It takes much more than sophisticated ICs and obvious connectors to make PoE possible: it also takes passive components of many types. It's somewhat ironic that a transformer – among the most "ancient" of electrical components – is needed for PoE. These transformers superimpose the high-frequency Ethernet data signal onto the same cable, injecting the PoE DC power (typically 48VDC) at the PSE end, while complimentary transformers at the PD end separate the Ethernet data from the DC power and direct the power to the PD circuitry.

One example is the Bourns SM13143EL, a dual-port LAN 10/100/1000 Base-T PoE++ transformer for all classes of Ethernet applications (Figure 7). It contains eight 1:1 transformers that meet all relevant ratings, with a footprint of about 11 × 28mm and 8mm height.

Figure 7: This multichannel transformer provides critical isolation and coupling for eight PoE wire pairs. (Image source: Bourns, Inc.)

As they couple power and signal, the transformers also provide galvanic (ohmic) isolation, so problems or failures in the port do not induce a ripple effect of problems in the network. For example, this can include a PD failure or even a malicious act if someone shorts out the connector with a paper clip.

The PoE standards and good engineering practice also require protection against line surges due to lightning or nearby motors starting/stopping. One external protection device (which is separate from used within the PSE or PD for other types of protection) is the ESP-100-POE Network Lightning/Surge Protector from Laird Connectivity (Figure 8).

Figure 8: PoE standards also call for protecting the system and circuitry against lightning and other volage surges; this protection can be provided using this RJ-45-connector unit. (Image source: Laird Connectivity)

The all-metal cast enclosure and shielded RJ-45 connectors provide good EMI noise suppression with integral mounting feet that can be screwed to a metal chassis to provide proper; as an alternative, there is an integral ground wire provided which can be connected to a ground for those systems that require a single-point ground connection. The devices are compatible with 100Mbps data and 48V POE systems, with protection on all eight PoE pins (7.5 V clamping on data pins 1, 2, 3, and 6 plus 70V clamping on POE pins 4, 5, 7, and 8).

Conclusion

Implementing a PoE design and providing a PSE or PD circuit for PoE is complex. Fortunately, the needed components and support are now widely available, thus allowing smart-building and industrial applications to take advantage of the benefits of PoE. PoE saves on wiring costs due to the lower voltage and single unified data/power cable, but the installation is physically more manageable and can be done by non-electricians. For dispersed, lower-throughout applications, the SPE version of PoE provides an even more cost-effective way to get many to the benefits of PoE, along with new options for topology and connectors.

---

Bill Schweber is an electronics engineer who has written three textbooks on electronic communications systems, as well as hundreds of technical articles, opinion columns, and product features. In past roles, he worked as a technical web-site manager for multiple topic-specific sites for EE Times, as well as both the Executive Editor and Analog Editor at EDN.

At Analog Devices, Inc. (a leading vendor of analog and mixed-signal ICs), Bill was in marketing communications (public relations); as a result, he has been on both sides of the technical PR function, presenting company products, stories, and messages to the media and also as the recipient of these.

Prior to the MarCom role at Analog, Bill was associate editor of their respected technical journal, and also worked in their product marketing and applications engineering groups. Before those roles, Bill was at Instron Corp., doing hands-on analog- and power-circuit design and systems integration for materials-testing machine controls.

He has an MSEE (University of Massachusetts) and BSEE (Columbia University), is a Registered Professional Engineer, and holds an Advanced Class amateur radio license. Bill has also planned, written, and presented on-line courses on various engineering topics, including MOSFET basics, ADC selection, and driving LEDs.