Twists and turns of twisted pairs
Q: I just received a shipment of Berk-Tek Category 5e plenum cable. I cut off a piece about 18 inches long and removed the conductors from the outer jacket. I noticed that the blue and green pairs have a substantially tighter twist than the orange and brown pairs
Q: I just received a shipment of Berk-Tek Category 5e plenum cable. I cut off a piece about 18 inches long and removed the conductors from the outer jacket. I noticed that the blue and green pairs have a substantially tighter twist than the orange and brown pairs. The same seems to be true of Category 5 cable from a different manufacturer. Is what I am seeing an optical illusion, or is it real? If it is real, what is the rationale?
Fort Myers, FL
A: No, what you are seeing is not an illusion. It is very real. Remember in elementary school when the music teacher would divide up the class and have certain groups sing in rounds? My most memorable was Row, Row, Row Your Boat, with everyone singing their loudest while holding their fingers in their ears-to keep from getting confused by hearing the part of the song that the other groups were singing. Believe it or not, the situation with the twist rate is similar, except the conductors don't really have ears. I will try to explain.
Twisted-pair as we know it consists of two copper conductors, typically 22 or 24 AWG, each covered with insulation. The two wires are twisted around each other such that each is equally exposed to interfering signals picked up from their environment. This is especially handy for differential transmission, where information is communicated by the difference in voltage potential between the two conductors.
For example, if the signal voltage applied to one conductor is V1 volts, and the signal voltage applied to the other is V2, then the signal being transmitted is V1V2. And when an externally induced noise voltage Vn is coupled to the conductors, the twisting of the conductors will cause Vn to be equally coupled onto both conductors, so that the voltage on one conductor will be V1+Vn and the voltage on the other conductor will be V2+Vn. The receiver will see the difference between the voltages, which is V1+Vn(V2+Vn), which equals V1V2 because the noise voltage subtracts out of the equation.
Nice, but so far we only have two conductors making one pair and we need four pairs of conductors for our cable. If we use four of our V1V2 pairs, then not only will each have Vn to contend with, but also the other three V1V2 pairs.
But what if the rate of twist of each pair, relative to each of the other pairs in the cable, was different? By varying the twist rate, or "lay" as it is termed in most catalogs and standards, it is possible to greatly reduce interference between the transmissions on the various pairs within a cable or binder group. This interference is typically called "crosstalk." And this is why cables that contain multiple twisted pairs within the same overall sheath are required to have the pairs twisted together with varying lay lengths. This is also why when you test a four-pair cable for length you get four different answers-none of which will match the length numbers stamped on the jacket.
And what about delay skew? Well, if all the horses leave the starting gate at the same instant but a couple of them have to run a longer course around the track, it is unlikely that they all will arrive at the finish line at the same time. The difference between the fastest and the slowest arrival is the "delay skew." And just to keep it interesting, not all manufacturers have the same color pair with the tightest twist.
A bit of wireless advice
I have been receiving an increasing number of inquiries regarding wireless networking. Wireless is a wonderful thing, which is why the commercial wireless-networking industry will sell an estimated 6.2 million units this year-almost double last year's sales. Instead of being tied to the outlet, wireless users are free to move about their work environments. And with home-network users buying an estimated 2.3 million network kits this year, they are free to move about their homes as well.
But according to the Wireless Ethernet Compatibility Alliance (WECA-www.wirelessethernet.org), the biggest problem with wireless networks is that users don't correctly install software designed to improve security. These are the same folks who will tell you that "breaking into a wireless network is difficult and requires very expensive and complicated equipment."
Hackers find it easy to infiltrate wireless networks because they are, well, wireless. Wireless networks use radio signals that are transmitted at the same frequency as most cordless telephones. You know, the ones that are so easily eavesdropped with a police scanner or even a baby monitor.
To access most wireless networks, all a hacker needs is a laptop, an antenna, and a little computer code-and they can be just another member of your team, sharing all the same files on the same servers. You get the idea.
At a recent hackers' gathering, using downloadable software and off-the-shelf equipment, Ian Goldenberg, chief scientist at Zero Knowledge Systems, demonstrated how to capture outgoing messages and send e-mail over what could well be your wireless network.
WECA is the organization that tests wireless LAN products to determine interoperability among vendors. Wireless Fidelity, or Wi-Fi, is the trademarked name that WECA uses to signify wireless LAN product interoperability. WECA performs tests on wireless LAN products, and those that meet the interoperability standards are awarded the Wi-Fi logo.
An ad reads: "In your office, or in your car, at the job site, you can connect. Get rid of all those troublesome cables and wirelessly access the information you need when and where you need it."
But be very mindful that you are not the only one on the planet with a wireless access card. Make security a priority, not an afterthought.
And finally, to the dozen or so folks who asked about a color code for optical fiber, according to ANSI/TIA/EIA-598:
Donna Ballast is a communications analyst at The University of Texas at Austin and a BICSI registered communications distribution designer (RCDD). Questions can be sent to her at Cabling Installation & Maintenance or at PO Drawer 7580, The University of Texas, Austin, TX 78713; e-mail: email@example.com.