Wireless seems like a strange topic for a cabling installation magazine, but it is not. You see, a wireless LAN is not exactly wireless, it's...well...less wired.
Donna Ballast is BICSI's standards representative, and a BICSI registered communications distribution designer (RCDD). Send your questions to Donna via e-mail: [email protected].
A wireless access point must have either a data cable or a power cable, or both, to function. Unlike a wired network, which can be designed at 30,000 feet from construction drawings and a spec book, the success of a wireless network rests heavily on the accuracy of the site survey.
All the wireless systems design guides I have seen stress the need for a thorough site survey. For example...Step one: Identify the space to be covered, the devices that will be used, and the interface between the wireless and the wired LAN. Step two: Survey the site to determine the placement of access points and antennas that will maintain the maximum amount of coverage and signal strength for optimal performance.
This is excellent advice, provided you have a site constructed that you can survey.
Most of the new building plans that I have reviewed in the last few years have conduit for wireless access points shown in various areas throughout the buildings. But when I ask how these locations were determined, answers varied from, "One per column bay; we need to cover everything," to, "We only wanted to cover the main conference room."
802.11 is not the same old Ethernet, with 1s and 0s traveling down a conductor. 802.11 is radio, from the wireless access point to the wireless network interface card. If you have ever experienced poor reception on your car radio when you pass by a tall building or through a tunnel, then you can appreciate the importance of locating the wireless access points to provide a strong signal throughout the area to be covered.
But it isn't simply a question of installing more access points to provide more performance or greater coverage. The limited channel capacity of 802.11-based wireless LANs does not allow for an infinite number of access points, overlapping their antenna coverage within a given area.
The antennas are usually chosen based on the size and shape of the area requiring coverage, available mounting options, and aesthetics. You may need to install a splitter and use directional antennas to isolate the boardroom network from the breakroom users while blending them into the boardroom décor.
If the access points are not appropriately placed and the correct antennas are not accurately positioned in relation to the environmental stumbling blocks and rival radio signals, then poor coverage and even worse data throughput will occur.
Wireless phones can interfere with the wireless LAN operating in the same spectrum; commercial microwave ovens can create radio frequency (RF) interference; and certain materials-even people-can create an RF-absorbent "wall" that soaks up the signals like a sponge. None of this is readily apparent before the building is actually constructed and occupied.
Wireless-if it truly were wireless, system design would be a lot less complicated. At least during the "blueprint phase" of a building, where it currently is almost impossible.
Q: A question has come up during the construction of one of our elementary schools concerning under-slab, in-conduit cabling. I have searched my BICSI manuals and have been unable to find clarification.
I have a situation in which our main crossconnect is located at the rear of a computer lab housing about 40 systems. The cabling runs through conduit under slab to floor monuments.
My question is whether or not this should be considered a "wet" location, requiring the use of wet-location-rated cable. Our cabling contractor on this project proposes to install plenum cable. The California State Division of Architects has ruled that this is, in fact, a wet location requiring, under Title 24 Minimum Requirements, that cable rated for wet locations be used.
John A. Frailey
Elk Grove Unified School District
Elk Grove, CA
A: It is usually safe to assume that sometime after the slab was poured and before the building was "dried-in," it rained. That rain is still in the conduit. But that is not all the moisture that will collect there. Over time, the inside and the outside of the conduit will be exposed to different temperatures, and condensate will form inside the conduit. CMP and CMR cable jackets and conductor insulation make nice wicks to absorb this leftover rain and continually-forming condensate, but they are certainly an expensive way to keep a poorly designed pathway system useable. Periodically (in Austin, TX, about every six months) you must change the cables.
This is why TIA/EIA-569A paragraph 4.1.7 states that "Intrabuilding horizontal pathways shall be installed in 'dry' locations that protect cables from moisture levels that are beyond the intended operating range of 'inside' premises cable (see ANSI/TIA/EIA-568A). For example, 'slab-on-grade' designs wherein pathways are installed underground or in concrete slabs that are in direct contact with earth, are considered to be 'wet locations.' See ANSI/NFPA-70, Article 100 for definitions of damp, dry, and wet locations." BICSI felt so strongly about this issue that the exact same text appears in the ninth edition of their Telecommunications Distribution Methods Manual (TDMM).
Translation: Do not use this type of pathway for distribution of telecommunications cabling inside a building. Too late in this case. From experience, I can assure you that CMP cable installed in conduits that are either in or under concrete slabs that are in direct contact with the earth will be a problem. One solution is to use a cable rated for wet locations. But this creates yet another problem. Typically, cable designed to be installed in wet locations does not meet NEC Article 800. The contractor will need to obtain a waiver from the AHJ before the cable is installed. Another solution is to abandon the conduit system and route the cables overhead.
Given that you and I (at the very least) have now witnessed this design problem at least once, it would seem appropriate that much stronger language be in the next revision of TIA-569, as well as in the BICSI TDMM, warning architects, engineers, and designers of the problems associated with this type of design.
I have forwarded this request to Ray Keden, chair of the TR-42.3 Subcommittee, and Ray Craig, chair of the BICSI TI&M for consideration by their respective committees.
Let's hear your questions
Have you read the new National Electrical Code 2002 from cover to cover yet? Anything in there you find especially puzzling or problematic? If so, please forward your comments to me before August 15. I have been asked to serve as an alternate to a Panel 16 member for the upcoming code cycle, and would like to see if you are concerned about the same areas that I am. In the October issue, I will publish a synopsis of the comments collected, including my own pet peeves.
I would also like to encourage you to submit any proposed changes to the Secretary of the Standards Council at NFPA headquarters. Forms and instructions are included in your NEC 2002.