Skip to content

Network Design: Power-Over-Ethernet Considerations

A power plant providing electricity to a rack of network equipment

For anyone considering a network refresh, the power-over-Ethernet (PoE) capabilities of the proposed switches is of vital consideration. Many devices that were previously powered by individual AC or DC power supplies are now built with the option – or requirement – of being powered by network switches. This development has several benefits:

  • Simplified infrastructure: With PoE, only one cable is required to provide data AND power to a device, removing the requirement to run separate AC power circuits or DC power supplies to an endpoint.
  • Improved monitoring capabilities: Using network monitoring software, the IT Department can easily tell when a PoE-powered device stops working or falls off the network. This capability is not always present in devices with standalone power supplies and analog connectivity.
  • Potential for energy efficiency: Powering devices via PoE has the potential to improve energy efficiency vs independent power supplies. This depends on several factors, including the length of cable runs and the efficiency of the power supplies in the switches. In general, switch power supplies that are rated as ‘platinum’ or ‘titanium’ will have an excellent efficiency factor.
  • Ability to power down equipment not in use: Is there a time of day/night when your PoE-powered phones or WAPs are not needed? Perhaps there is a campus undergoing modernization and the equipment does not need to stay up? One can save power by shutting down ports that are connected to PoE devices. Depending on the make and model of switch, there may be an option to create a schedule in the GUI, use a script to interface with the API, or leverage a management platform like Cisco DNA Center, Aruba Central, Juniper Mist AI, etc.

As you plan out the PoE capacity of your switches, there are several key considerations to keep in mind:

What will power consumption look like in each IDF?

  • How many PoE-enabled devices will be connected to an IDF?
  • What is the maximum power consumption of each device type?
  • Do the devices pull different amounts of power on startup vs when continuously operating? Are there different operating conditions that cause them to pull additional power?
  • Do you plan to add additional PoE-enabled devices over the next 5 years? Perhaps you are aware of a plan to migrate video surveillance cameras or air quality sensors in the near future and will need to account for those devices drawing power from the IDF switches.

How do you plan to distribute power across switches in an IDF?

  • It can be beneficial to distribute PoE devices across multiple stacked switches in an IDF, but this can cost more.
  • Having a single PoE-enabled switch per IDF saves money up front but can lead to future problems. Having the single PoE-enabled switch go down means all of your PoE-enabled equipment will fail. It can also lead to tricky cable patching, in which specific ports on the patch panel need to be fed to a single switch.

How much electricity can the switches safely pull without blowing a fuse?

interior of a network equipment room

  • The most common power receptacle type is rated for 120 volts +-10% and 15 amps. If the power receptacle has a dedicated circuit to the breaker box, this means that it can provide a total of ~1,800 watts before tripping the circuit breaker. If the electricity from the breaker is shared amongst multiple power receptacles, then that number is even lower and may change over time. The available power needs to be considered before loading an IDF with high-power switches and a large quantity of PoE devices. Ideally, the electrical circuits should be updated according to the projected PoE requirements and switch utilization.
  • Will a UPS be installed at the IDF to provide continued power in the event of a power outage or abnormal voltage fluctuation? The UPS will need to be sized according to the expected PoE budget in addition to the regular power that the switches require just to operate.

There are several iterations of PoE, with each new standard providing additional capacity per port. If you are looking at a switch refresh, which PoE standard should you go with? In general, NIC Partners advises selecting the highest PoE capacity offered for a particular make/model of switch as permitted by budget constraints. The quantity of devices being powered by the network is only increasing, and a good network design should ensure that the switches will be able to meet any known requirements for at least five years. With that in mind, let’s take a look at PoE requirements for common devices that are out today:

  • Cisco 8845 IP phone: Up to 7W
  • Verkada SV25 air quality sensor: 4W
  • Cisco 9163E outdoor WAP: Up to 25.5W or 14.0W in reduced performance mode
  • Cisco 9166 indoor WAP: Up to 30.5W or 25.5W/14.0W in reduced perf. mode
  • Ubiquiti airFiber 5 Mid-Band wireless bridge: Up to 40W
  • Meraki MV72X outdoor camera: Up to 25.5W
  • Verkada CH52-E multisensor camera: Between 19.54W and 38.68W
  • Verkada TD52 video intercom: Either 11.5W or 25.5W in extended temp. range
  • Verkada CP52-E outdoor PTZ camera: Between 23.26W and 44.04W
  • AtlasIED IP-SDMF speaker/clock: Up to 30W
  • CoolGear CG-POE-CESBT PoE-to-USB-C adapter: Up to 60W
  • PoE Texas 55” video display unit: Up to 180W (2x 90W connections)

Do you plan to deploy any of these? Are your current switches able to accommodate these requirements?

Although PoE is only one aspect of network infrastructure planning, it is a detail that could quickly turn a project upside down if sufficient capacity is not considered. To summarize: Be sure to include the following action items in your network refresh planning process:

  1. Identify the current electrical capacity (receptacle, breaker/fuse capacity) per wiring closet
  2. Identify the current PoE requirements per wiring closet. Consider power utilization from both standpoints of ongoing operations and system startup (i.e. when you reboot a switch, all of the connected PoE devices will draw their MAX power while they boot).
  3. Identify the future PoE requirements per wiring closet – if you know which devices your organization plans to purchase or deploy over the next five years.
  4. Determine how you want to allocate PoE devices across a stack of switches in the wiring closet – all connected to a single PoE switch or spread out amongst multiple switches.
  5. Determine the UPS capacity and runtime requirements based on the power consumption of everything in a wiring closet, including PoE devices pulling power from switches.

Ready to explore how Power-over-Ethernet can simplify your network infrastructure? Contact us to learn more about our PoE solutions and how we can help you design a network that meets your current and future needs.