Translated by


Technology: SCS

Power over Ethernet, or PoE is the technology allowing to transmit to the remote device together with data electrical energy through the standard twisted pair cable on the Ethernet network. This technology intends for the IP telephony, access points of wireless networks, Webcams, network hubs and other devices to which it is undesirable or it is impossible to carry out a separate electric cable.

Power over Ethernet is standardized according to the IEEE 802.3af-2003 standard. There are several options of this technology preceding this standard, but they are a little widespread.

According to the IEEE 802.3af-2003 standard (see 802.3-2005) the direct current with a rated voltage of 48 V (min. = 36 V, max. = 57 V) through two couples of conductors in the quad cable, with a maximum current 400 mA for ensuring the maximum power of 12.95 W is provided. The standard defines 5 classes of the devices fed on PoE: 0-4. To each class there correspond the parameters of power and current. The 1st class is most widespread. For it the input current is equal 120 mA, and power can vary from 0.44 to 3.84 W. the 4th class is not used and is reserved on the future.

The technology uses phantom power for transfer of power supply. Data in standards of Ethernet are transferred as potential difference between conductors in one pair (for example, between wires 1 and 2 or 3 and 6). Giving of power voltage is performed as potential difference between couples of conductors (for example, between pairs 1-2 and 3-6)

The standard does not define on what pairs it has to be supplied and also its polarity therefore the devices expected receiving PoE power supply should be expected use pairs 12-36 or 45-78 for receiving, and the correct polarity to receive using the diode bridge. This technology works with the existing cable system, including category 5 cables without the need for introduction of any modifications into existing SCS.

Equipment of PoE and principle of work

The PoE technology has no impact on transmission quality of data. For its implementation properties of the physical layer Ethernet are used:

  • Use of high-frequency transformers on both ends of the line with the central branch from windings. The constant voltage of a power supply moves on the central withdrawal of secondary winding of these transformers, and also from the central branches is removed on a receiving end. Use of the central withdrawal of signal transformers allows without cross impact, to transfer both high-frequency data and constant voltage of a power supply on one couple of wires.

  • Use of free pairs for power connection. The modern cable networks Ethernet conforming to standard 100Base-TX consist of four pairs, two of which are not involved.

Respectively, there are two options of power supply by the PSE device:

  • A — on which go on twisted pair cables data
    * B — on not used pairs in a cable

PSE devices (injectors) differ, depending on options A or B, at the same time splitters, i.e. PD devices, are universal. The PD device is obliged to be able to accept a power supply in any option including at inversion (for example when the computer — crossover cable the computer is used). That circumstance that the PSE device supplies in a cable only if the connected device is the PD device is important. Thus, the equipment which is not supporting the PoE standard and accidentally connected to the PSE device will not be put out of action. Actually, no normal equipment of Ethernet, since specification 10Base-T, can be put out of action by a certain potential difference between pairs as the standard provides obligatory galvanic isolation. The procedure of giving and power off on a cable consists of several stages.

Connection determination stage

The stage of determination of connection serves in order that define whether the device connected on the opposite end of a cable by the PD device is. At this stage the PSE device gives tension from 2.8 to 10 B on a cable and determines parameters of input resistance of the connected device. For the PD device — this resistance from 19 to 26.5 kOhm with in parallel the connected condenser with a capacity from 0 to 150 nf. Only after check of compliance of parameters of input resistance for the PD device, the PSE device passes to the next stage — a classification stage, otherwise PSE repeatedly, through time not less than 2 ms, passes "connection determinations" to a stage.

Classification stage

After a connection determination stage, the PSE device can complete a classification phase in addition. The stage of classification serves for determination of power range which the PD device can consume that then to control this power. The class from 0 to 4 will be appropriated to each PD device depending on the stated power consumption. The maximum power range has class 0. Class 4 is reserved by the standard for further development. The PSE device can remove stress from a cable if the PD device began to consume power big, than that which announced during a classification stage. Classification is executed by introduction to a cable by tension PSE device from 14.5 V to 20.5 V and measurement of current in the line.

Stage of giving of complete tension

After passing of stages of determination and classification the PSE device gives to a cable voltage of 48 V with the front of increase not quicker, than 400 ms. After giving of complete tension on the PD device, the PSE device exercises control of its work as two methods.

1) If the PD device during 400 ms consumes current less than 5 mA, then the PSE device removes a power supply from a cable.

2) The PSE device gives tension with a frequency of 500 Hz with an amplitude of 1.9 — 5.0 V to a cable and calculates input resistance. If this resistance is more, than 1980 kOhm during 400 ms, then the PSE device removes a power supply from a cable.

Besides, the PSE device continuously monitors an overload current. If the PD device consumes current during 75 ms more than 400 mA, then the PSE device will remove a power supply from a cable.

Shutdown stage

When the PSE device defines that the PD device is switched-off from a cable or there was an overload of the consumed current the PD device, there is a removal tension from a cable in time not less, than 500 ms.

Advantages of PoE technology

The technology of power supply through Ethernet is an attractive alternative method of power supply of network devices. And, its application is possible, both at the organization of new networks, and at upgrade of existing. Most often at upgrade of network installation of active equipment is required where there is no power source nearby and electric sockets. Thanks to the IEEE 802.3af standard there is a possibility of equipment installation in the places which are most suitable for this purpose, despite of lack of an electrical wiring. For example, it is possible to set Wi-Fi access point in the place of the best acceptance of a signal even if there are no electric sockets, or to place the IP camera in the place, comfortable for the overview. PoE allows not only to save significantly on the cost of power cables and other components, but also to reduce installation time of the equipment Ethernet.