PoE Plus, ambient temperatures, and the National Electrical Code

The circuits described in the National Electrical Code can be divided into four broad groups: electric light, power, signaling, and control and communications.

The circuits described in the National Electrical Code can be divided into four broad groups: electric light, power, signaling, and control and communications. As I pointed out in my November 2006 column, the standards-based telecommunications cabling systems that we design and install today are primarily used as Class 2 signaling circuits and not communications circuits within the NEC.

A key point to remember is that all cables are designed to perform within a certain operating-temperature range. For listed category unshielded twisted-pair (UTP) cable, this range typically is -40º C to +60º C (-40º F to +140º F). A byproduct of applying current to a conductor is heat.

PoE Plus and the NEC

Work focusing on cabling issues for Power over Ethernet Plus (PoE Plus) continues in IEEE 802.3at Power Over Ethernet Plus Task Force, and TR-42.7 Telecommunications Copper Cabling Systems.

Because their target media for this new application is the installed base of cabling, the IEEE has asked TR-42 to develop ampacity limits of Category 5e, 6, and 6A cabling. Why have the consumer dollars going into new cabling when you can use the installed base and sell them more hardware? Sound business model, right there.

The NEC defines ampacity as the current, in amperes, that a conductor can carry continuously under the conditions of use without exceeding its temperature rating. Ambient temperature is a condition of use and must be considered. So, a conductor with insulation rated at 60º C and installed near a boiler where the ambient temperature is continuously maintained at 60º C has no current-carrying capacity. Any current flowing through the conductor will raise its temperature above the 60º C insulation rating. Therefore, the ampacity of this conductor, regardless of its size, is zero.

According to its introduction, the NEC is all about safety: “… the practical safeguarding of persons and property from hazards arising from the use of electricity.” The Code further goes on to caution us that, “Hazards often occur because of overloading of wiring systems by methods or usage not in conformity with this Code. This occurs because initial wiring did not provide for increases in the use of electricity.”

In the case of PoE Plus, a hazard may arise from line-powering Class 2 circuits over the installed base with no correction factors for ambient temperature or limitations on the number of cables in a bundle. The bigger the bundle, the more heat generated and the more difficult that heat is to dissipate. Some numbers I have seen being kicked around are indicating cable surface-temperature rise between 8º C (46.4º F) and 22º C (71.6º F) for cables deep within the bundle.

With the current and conductor size fixed, there are only three logical options:

  1. Increase the operating temperature range of the cabling;
  2. Reduce the ambient temperature;
  3. Mitigate the heat.

The first option would require that a new, higher-rated cable be installed, which is not in keeping with the IEEE business model of using the installed base.

Option two would require closer control over the ambient temperature over the length of the existing cable; exactly how to do that would be a mechanical engineering and maintenance nightmare.

The third option would require unbundling and separating the cables to allow heat transfer into the surrounding air. It brings to mind visions of oven thermometers nestled among the cables and little cooling fans clamped to the sides of cable trays. (Line-powered, Class 2, plenum-rated cooling fans, of course.) OK, daydream over … must have been all that eggnog and cookie-baking.

In October 2006, a Task Group was formed under TR-42.7 to assemble a list of recommendations concerning the application of DC power over Category 5e, 6, and 6A cabling. The Task Group will also address the heat generated and its effect on the cabling’s life cycle and performance.

According to an October 2006, liaison letter to IEEE 802.3at, the TR-42.7 Task Group is studying limiting a maximum cable surface temperature rise of 10º C over ambient, which would result in roughly a 4% increase in channel insertion loss and a 50% reduction in the life of cabling materials.

TR-42.7 plans to provide an update by the IEEE meeting this month; IEEE plans to publish PoE Plus in June 2008.

At this point, the NEC 2008 is pretty much set in stone. Proposals to change the NEC 2008 are due to NFPA before November 2008 for the changes to be effective in NEC 2011, which leaves us to struggle with what we have for at least the next four years.

The bottom line is, running PoE Plus on Class 2 circuits causes problems that are not currently addressed in the NEC. The two biggest are: heat caused by application of current (almost 1 amp per conductor pair in a four-pair cable, for a total of 4 amps per cable); and arcing and sparking, which will occur at the connectors when unplugging the plug/jack interface.

Currently in the plan is a sort of closed-loop control on the application of current to the conductor; how to remove the current from the conductor before unplugging the connector is not yet resolved.

Brother, can you spare some time?

In a recent e-mail dialogue, a reader pointed out that, “Maybe someone needs to tell NFPA about convergence, since they are stuck in the ’80s. It is a shame that NFPA has trouble getting in step with the real world.”

Paid ‘volunteers’

I agree, but in their defense, the NFPA does not write the NEC, it only publishes the document. The NEC is developed by Panels and Task Groups of “volunteers” who actually do the work. Most of these volunteers are paid by their employers to be at the table and voting for or against proposed changes based on the benefit to those employers. Translation: The entire process is the product of a consensus reached by a bunch of special-interest lobbyists. We, as users of the Code, just do not currently have a paid “dog in that fight.”

So, is this really compromising the usefulness of the NEC for low-voltage applications? The answer depends on whom you ask, and there are several camps:

  • The status-quo camp says, “No. The NEC is fine as it is. Do not change anything.” That translates to, “We have worked very hard to get the Code to say exactly what we want, and we do not want to risk having to alter our current business plan.”
  • The new-product-opportunity camp says, “Yes. The NEC must change to address this problem and embrace my new product.” There are a lot of these guys sitting at the meeting table pushing their agendas.
  • Then there is the “yes” camp to which I belong, whose mantra seems to have become, “Someone should reformulate the NEC low-voltage cabling sections to be media-based and not application-based, so we can more quickly and accurately address the ‘real-world’ changes in how services are delivered in the Code. But I don’t have the time.”

At the 2008 Report on Proposals and Report on Comments meetings, I was asked to champion a group that would undertake the project, but that is a lot to take on as a volunteer. Hopefully, one of the paid lobbyists will hear our plea and take on the cause.

DONNA BALLASTis BICSI’s standards representative, and a BICSI registered communications distribution designer (RCDD). Send your question to Donna at: dballast@swbell.net

More in Home