Power over Ethernet or PoE describes any of several standards or ad-hoc systems which pass electric power along with data over an Ethernet connection without the need for batteries or a wall outlet. This allows a single cable to provide both data connection and electric power to devices such as wireless access points, IP cameras, and VoIP phones.

How Power over Ethernet [PoE] Works

Network cables, such as CAT 5E and CAT 6, comprise eight wires arranged as four twisted pairs.  In 10 and 100BASE-T Ethernet, two of these pairs are used for sending information, and these are known as the data pairs.  The other two pairs are unused and are referred to as the spare pairs (Gigabit Ethernet uses all four pairs).

Because electrical currents flow in a loop, two conductors are required to deliver power over a cable.  POE treats each pair as a single conductor, and can use either the two data pairs or the two spare pairs to carry electrical current.

The PoE Process

1. Power Injection

Power over Ethernet is injected onto the cable at a voltage between 44 and 57 volts DC, and typically 48 volts is used.  This relatively high voltage allows efficient power transfer along the cable, while still being low enough to be regarded as safe. However, this voltage can still damage equipment that has not been designed to receive POE. In order to power a device via Ethernet, a PoE adapter is required. This device, also called an “Ethernet injector,” plugs into a standard power outlet and provides power to one or more Ethernet ports.

2. Signature Detection

Before a POE source, – PSE (Power Sourcing Equipment) – can enable power to a connected PoE device –PD (Powered Device) – it must perform a signature detection process. Signature detection uses a lower voltage to detect a characteristic signature of IEEE-compatible PDs (a 25kOhm resistance).  Once this signature has been detected, the PSE knows what voltages can be safely applied.

3. Classification

Classification follows the signature detection stage, and is an optional process.  If a PD displays a classification signature, it lets the PSE know how much power it requires to operate. This means that PSEs with a limited total power budget can allocate it effectively.

4. Enable Power

The final stage after detection and classification of a newly connected device is to enable power: the 48V supply is connected to the cable by the PSE so the PD can operate.  Once enabled, the PSE continues to monitor how much electrical current it is delivering to the PD, and will cut the power to the cable if too much, or not enough, power is drawn.  This protects the PSE against overload, and ensures that POE is disconnected from the cable if the PD is unplugged.

PoE Classification in Detail

All PoE devices must adhere to the universal IEEE 802.3af/at/bt PoE standards. This way, all PoE devices can properly communicate with one another on a network, even if some devices on the network aren’t PoE compatible. For example, if you connected a PoE Ethernet cable to a no PoE device, it would supply only data, and not electricity; since the device is not PoE compatible and cannot send a digital signature to the PSE for power, your switch will know not to send electricity down the line.

There are 4 PoE classifications;

Type 1

The first PoE type is normally referred to as PoE. It conforms to the IEEE 802.3af standard and it can supply maximum power to port of 15.4 Watts. It was an early PoE standard created back in 2003, meant to supply electricity to low-powered devices on a network, including VoIP phones, sensors, wireless access points, and simple static surveillance cameras that can’t move from side to side or up and down.

Type 2

The second PoE type is commonly referred to as PoE+ or PoE Plus. Type 2 PoE conforms to the IEEE 802.3at standard, and it can supply maximum power to port of 30 Watts. This newer standard is backward compatible, meaning it also supports Type 1 PoE devices. Type 2 PoE can power PDs such as wireless access points with six antennas, biometric sensors, LCD displays, and more advanced cameras that have pan, tilt, and zoom functionalities.

Type 3

Type 3 is the third PoE type, and it is also known as 4-pair PoE, RP PoE, PoE++, and UPoE because it uses all four pairs in a copper cable. It conforms to the IEEE 802.3bt PoE standard, and it can supply maximum power to port of 60 Watts. PoE++ has enough power per port to operate management devices and video conferencing systems.

Type 4

Commonly referred to as higher-power PoE, Type 4 also conforms to the newest IEEE 802.3bt standard, but it can supply maximum power to port of 100 Watts in order to accommodate the growing power requirements of network devices and IoT. It can even support power hungry laptops and TVs.

What Cable to Use for PoE?

IEEE 802.at and 802.3af standards uses a four-pair CAT 5E cable, but only call for power delivery from two of the pairs; either pairs 2 and 3 or 1 and 4, but not both pairs at the same time.

IEEE 802.3bt uses all of the pairs in a four-pair CAT 6 cable, which allows current to flow evenly among them. This innovation drastically improves the amount of power that can be transmitted over a single PoE cable, in addition to the higher data rate of up to 10GBASE-T.

Next week, we dive deeper into the benefits of PoE, its wide applications in the IoT, and its place in the future.

Check out our wide catalogue of Power over Ethernet (PoE) solutions, which includes Power Sourcing Equipment and Powered Devices.