The industry's beginning to buzz: 10-Gigabit Ethernet over UTP?

Advances in silicon technology make 10-Gigabit Ethernet over 100 meters of UTP copper possible.

Advances in silicon technology make 10-Gigabit Ethernet over 100 meters of UTP copper possible. Companies are announcing transmission, or the development of protocols to transmit 10-GbE over UTP cabling.

IEEE 802.3 is gathering experts for a study group to determine if 10GBase-T is technically possible, economically feasible for the consumer (compared to a fiber system solution), and whether there is sufficient "market potential"—read "potential sales"—to justify the investment of time and money that such a standard and set of products requires. One can imagine, given our current economy, investment money for such R&D is not as easy to come by as it was when 1000Base-T was being developed.

A peek at the "estimated timeline" on the 10GBase-T call-for-interest agenda indicated that a 10GBase-T standard could be completed by June 2006—barring a stall at the starting line.

OK, so IEEE is looking at 10GbE over UTP—but which UTP? The installed base of Category 5 and 5e? The recently standardized Category 6? The recently proposed Category 7 UTP? That depends on whom you talk to. Historically, IEEE likes to use an installed base, which makes a lot of sense, given that most of the IEEE 802.3 participants represent network hardware manufacturers. That way, we can spend the lion's share of our IT budget on new hardware, and not on new cabling.

The Category 6 Consortium, of course, wants Category 6 to be the cable of choice (see below). Declaring that Category 6 is "what people are installing now," they are prepared to lobby heavily with data (compiled from more than 2,000 end-users surveyed), which "overwhelmingly supports that Cat 6 is the future for horizontal cabling."

Then there is the recently proposed (October) Category 7 UTP initiative—TIA TR-42.7 contribution #698 (see sidebar next page). This proposal, of course, differs from the ISO/IEC 11801 Ed2 version of Category 7, which is a shielded cable.

So, which will it be? By June 2006, we will have seen another three-and-a-half years of intense Category 6 marketingU you know, the "Cat-5-is-dead, Cat-5e-is-waning, long-live-Cat-6" routine.

But indulge me for a moment. It took the TIA from 1985 to 1995 to produce the Categories 3, 4, and 5 standards. It took them from 1995 to 2002 to produce the Categories 5e and 6 standards. Based on their previous track record, if they started today, they would not have a Category 7 UTP standard published by June 2006. This is totally aside from the fact that the IEEE follows ISO/IEC and not TIA cabling standards. Translation: Category 7 UTP is not even a contender.

Update: Static discharge

Do you recall all the debate over switch ports failing due to electrostatic discharge from the installed cabling?

Remember back in third grade, when the science teacher amazed you by rubbing an inflated balloon briskly on someone's long hair and then leaving the balloon hang there, as if by magic? That was not magic; it was static electricity. The hair lost electrons (negatively charged particles) and became positively charged. The balloon gained electrons and became negatively charged. Since the hair and the balloon were oppositely charged, they were attracted to each other. It's a harmless demonstration of how an electrostatic charge is created.

Back to 2002 and what this has to do with cabling and dead switch ports: At the molecular level, differences in materials can result in charge transfer when those materials are mechanically rubbed against each other—like maybe snaking a cable through a pathway system from the TR to the work area?

Back in January 2001, the IEEE formed an ad hoc study group, IEEE 802.3 CDE Ad hoc Committee, to address phenomena they termed "Cable Discharge Events," which was causing network equipment failures. Translation: Dead ports.

Many presentations were made during the group's three meetings. Some showed that relative humidity was more relevant to charge storage on cables than was the category of the media. Previous thinking was that failure rates were higher in Category 5e and Category 6 installations than they were in Category 5 and Category 3 installations. Other presentations showed that certain equipment was more sensitive to the problem than others. And still other presentations showed that the discharge characteristics of UTP cables are much different than the Human Body Model currently used for testing silicon devices.

The Human Body Model is a very high-impedance and low-capacitance model. But testing indicated that UTP cables discharge a very large amount of charge with a low impedance. Hence, the silicon is hit with larger amounts of energy than it is designed to withstand. Translation (again): Dead ports.

The IEEE 802.3 CDE Ad hoc Committee's final action will be to publish a test methodology for silicon devices and products. This report is expected to be published before January. Translation: This will be the concern of network equipment manufacturers, not structured cabling system designers and installers. The Committee also asked the TIA for help in understanding the range of any potential energy and voltage that can be found on the cabling, the circumstances under which this could occur, and a model that would be useful to develop testing.

In response, TIA TR-42.7 is developing installation guidelines to reduce the chances of electrostatic-discharge damage to network equipment. They are intended to provide installers of cabling and equipment specific recommendations on how to avoid static discharge into equipment. This Telecommunications Systems Bulletin is expected to be published in mid-2003.

But until then...

  • ANSI/TIA/EIA-569 recommends that the humidity in telecommunications spaces housing equipment be maintained between 30% and 55%. Do it. Low humidity is one of the primary causes of electrostatic discharge.
  • Before connecting network equipment to installed UTP cabling, discharge the cabling. You can either buy something like Anixter's LAN Static Discharge Unit, or make your own by stripping and securing all eight conductors of a piece of UTP cable to an approved grounding conductor on one end, and terminating the other end on an 8-pin modular jack.
  • Once the link is patched to a port, leave it there until you need to reuse one or the other.
  • Remember, some equipment is more sensitive than others. So, follow the manufacturer's specifications and guidelines, and heed the warnings.

Donna Ballast is BICSI's standards representative, and a BICSI registered communications distribution designer (RCDD). Send your questions to Donna via e-mail: ballast@utexas.edu.


What's the Category 6 Consortium?

Established in 2002, the Category 6 Consortium—C6C—is an association of copper cable, component, and electronics manufacturers. The C6C was formed to help promote the advantages that Category 6 cabling brings to LANs, and to create a resource where end users can learn about the technical advantages and affordability that wide bandwidth brings to LANs, and specifically to horizontal cabling applications.


A case for Category 7

Entitled "Next Generation UTP Cabling," Contribution #698 proposed that the TIA study a new category of UTP, which would be a superset of Category 5, be backward compatible with lower categories of cable, and not require a new connector. If this sounds familiar, it is. In 1995, it was called "Category 6." TR-42.7 agreed to form a study group to review the proposed Category 7 UTP. If the group agrees, we may see Category 7 UTP included in TIA/EIA-568C.2.

More in RJ45 UTP & Shielded