Device Roles
- Power sourcing equipment (PSE)—A device that provides power through a twisted-pair Ethernet connection. The switch, through switching modules equipped with Power over Ethernet (PoE) daughtercards, functions in the PSE role.
- Powered device (PD)—A device powered by a PSE (for example, IP phones, IP cameras, and wireless access points).
Note Not all PoE-capable devices are powered from the switch. There are two sources of local power for PoE-capable devices:
- A power supply connected to the device.
- A power supply through a patch panel over the Ethernet connection to the device.
When a locally powered PoE-capable device is present on a switching module port, the switching module itself cannot detect its presence. If the device supports CDP, the supervisor engine can discover a locally powered PoE-capable device through CDP messaging with the device. If a locally powered PoE-capable device loses local power, the switching module can discover and supply power to the IP phone if the inline power mode is set to auto.
PoE Overview
Cisco PoE daughtercards support one or more PoE implementation:
– Supported with the WS-F6K-48-AF PoE daughtercard and the PoE daughtercard on the WS-X6148E-GE-45AT switching module.
– Maximum 16.80 W at the PSE.
– The IEEE 802.3af PoE standard defines a method to sense a PD and to immediately classify the power requirement of the PD into these per port power ranges at the PSE:
• Class 0: Up to 15.4 W (0.44–12.95 W at the PD; default classification)
• Class 1: Up to 4 W (0.44–3.84 W at the PD)
• Class 2: Up to 7 W (3.84–6.49 W at the PD)
• Class 3: Up to 15.4 W (6.49–12.95 W at the PD)
- Cisco prestandard inline power—10 W at the PSE.
With a PoE daughtercard installed, a switching module can automatically detect and provision a PoE-capable device that adheres to a PoE implementation supported by the PoE daughtercard. The switching module can supply power to devices supporting other PoE implementations only through manual configuration.
Only a PD connected directly to the switch port can be powered from the switch. If a second PD is daisy-chained from the PD that is connected to the switch port, the second PD cannot be powered by the switch.
Each PD requires power to be allocated from the chassis power budget. Because each PD can have unique power requirements, more devices can be supported if the system’s power management software can intelligently allocate the necessary power on a per-port basis.
You can configure ports to allocate power at a level based on the following:
- If a PD is detected, with auto mode configured:
– Information sensed from the device
– A default level
– A configured maximum level
- Whether or not a PD is present on the port, with static mode configured:
– A default level
– A configured level
Inline Power IEEE Power Classification Override
The IEEE 802.3af standard contains no provision for adjustment of the power allocation. 802.3af-compliant PDs that support CDP can use CDP to override the IEEE 802.3af power classification.
The WS-F6K-48-AF PoE daughtercard or the PoE daughtercard on the WS-X6148E-GE-45AT switching module support these inline power IEEE 802.3af power classification override features:
- Power use measurement—The ability to accurately measure the power provided by the port to the powered device.
- Power policing—The ability to monitor power usage on a port.
With power measurement and policing, you can safely override the IEEE 802.3af power classification of a device that requires a power level at the lower end of its IEEE power classification range.
PoE monitoring and policing compares the power consumption on ports with the administrative maximum value (either a configured maximum value or the port’s default value). If the power consumption on a monitored port exceeds the administrative maximum value, the following actions occur:
- A syslog message is issued.
- The monitored port is shut down and error-disabled.
- The allocated power is freed.