Q: I am currently working on a consolidated lodging facility for an Air Force Reserve project with the Army Corps of Engineers. They want to follow all ANSI/TIA/EIA-specified standards. I am having difficulty finding any information regarding lodging facilities in either the standards or the two Telecommunications Distribution Methods Manuals (TDMMs) published by BICSI (www.bicsi.org).
Currently specified in each guest room, the faceplate has two outlets-one for voice and one for data (modem). Would a guest room in a hotel/motel be considered a "work area" and, therefore, require a minimum of two cable runs? Or, would one Category 5E cable run be enough? Using one pair for the voice and another pair for the modem would leave the two extra pairs for future use.
William Fauver
Rothe Development Inc.
Minneapolis, MN
A: While neither the TIA standard nor BICSI's TDMM specifically addresses hotel/motel cabling, the TIA does define a work area as a building space where the occupants interact with telecommunications terminal equipment-certainly applicable to the guest rooms in your project.
Speaking from personal experience, taking care of business from the road is demanding. You have to stay connected to the office and accessible by family. To a business traveler, the in-room work area is equally as important as plenty of towels and a good mattress. Cozy work areas with large desks, task-appropriate lighting, ergonomic desk chairs, and plug-and-play laptop connectivity for simultaneous voice and high-speed data access (not a modem line) are what we seek.
The meaning of the term "high-speed" continues to evolve, and one Category 5E channel accommodating all services seems unlikely. My recommendation: install a minimum of two telecommunications outlet/connectors for each guest room and consider additional connections for larger or special-use guest rooms. And remember the "TIA golden rule": never split a cable between connectors. One connector always equals one cable.
Q: We are in the midst of a controversy regarding the installation of Category 5E four-pair cables in cable tray. The building owner wants the cables "combed," or installed in neat rows. My co-worker and I believe that the cables should be laid randomly in the tray to minimize the possibility of crosstalk. We have been unable to find anything in writing on this subject in the TDMM or in TIA/EIA documents. What is your opinion on this subject?
Marvin D. Hengst, RCDD
D&E Networks
Lancaster, PA
A: The TIA defines bundled cable as an assembly of two or more cables continuously bound together to form a single unit. I contend that you do not need a hydroscopic thread to "bind" cables together. Having them all hand-formed to lay next to their neighbor-cable for the entire length of the cable tray will perform exactly the same function-lots of "eye candy" for the owner to brag about... and performance degradation. Yes, there will be a penalty for all that neatness.
For cables bundled prior to installation-sometimes referred to as loomed, speedwrap, or whip cables-ANSI/TIA/EIA-568B.2 says that the power-sum near-end crosstalk loss for any disturbed pair and all pairs external to that pair's jacket but within the bundled cable must be 3 dB better than the specified pair-to-pair near-end crosstalk loss. Translation: bundling either costs you 3 dB of performance, or it costs you the additional dollars to purchase a better-performing cable.
If the owner wants to see a neat-and-tidy installation and does not want to pay the premium for cabling that such an installation practice requires, then put a lid on it. On the cable tray, that is. A cover on the cable tray would hide the unsightly randomness and provide additional physical protection.
Q: Is there a standard wiring configuration for a 50-pin Amphenol cable that would be used to connect T-1 lines from telecommunications equipment to an 8-pin modular patch panel that is T568B-compliant?
Greg Brophy
Oakland, CA
A: Yes, actually, there are two standard 50-pin connector configurations for use with 1.544-Mbit/sec digital access lines. Namely, RJ-48M, for up to eight 4-wire connections and RJ-48H, for up to twelve 4-wire connections.
There are also two standard 8-pin modular connector configurations for use with 1.544-Mbit/sec digital access lines. RJ-48C is a single 4-wire connection and RJ-48X is a single 4-wire connection equipped with a shorting bar. A T568B connector can be terminated as an RJ-48C, but not as an RJ-48X. So, if you do not need the loopback feature associated with the shorting bar, then you could use your T568B patch panel.
For additional information on these and other connector configurations, Committee T1-Telecommunications publishes Technical Report No. 5, entitled Network and Customer Installation Interface Connector Wiring Configuration Catalog. The most recent edition was released in August 1999.
Technical Report No. 5 describes connector wiring configurations used at the network-to-customer installation interface and within the customer installation that are specified in Part 68 of the Federal Communications Commission rules and regulations, carrier technical references, and ANSI-approved standards. The report is available from Global Engineering Documents (www.global.ihs.com).
Don't leave home without it
I recently received an urgent plea from an installation contractor working on a large project (valued at about $3 million), in which the Category 6 (draft 5) cabling specified by the engineer is not passing any testing above 200 MHz.
The cabling manufacturer claims compliance with an earlier draft of the proposed Category 6 standard but since the building owner is also the cabling manufacturer, all fingers are pointing at the installation contractor-who is now desperately seeking the unpublished working committee drafts.
Draft 5, dated May 26, 1999, was an update to Draft 2a, dated April 9, 1998. Draft 5 was where reference to 250 MHz first appeared in the document. The plan was that Category 6 channels would have power-sum attenuation-to-crosstalk ratio (ACR) that is greater than 0 at 200 MHz, and that all transmission parameters would be specified up to 250 MHz.
The component specification text reads something like this: "For all frequencies from 1 MHz to 250 MHz, [a given parameter shall] meet the values determined using equation [#]. The values in Table [#] are provided for information only." Draft 5, however, was riddled with the dreaded TBDs-to be determined.
As for field testing, Draft 5 refers the reader to TIA/EIA TSB-67 and TIE/EIA-568A-5 for the procedures to measure wire map, length, attenuation, worst pair-to-pair near-end crosstalk, power-sum near-end crosstalk, return loss, worst pair-to-pair equal-level far-end crosstalk, power-sum equal-level far-end crosstalk, propagation delay, and delay skew. That is how to test, but what passes and what fails would still be based on the TBDs in Draft 5.
The moral of this story: do not accept a contract that requires compliance to a document that you do not have in your company library. I am not saying that you have to wait until a standard is published. You simply can ask the owner or designer to provide you with a copy (usually at your expense) of any reference document not in the public domain. Or, the owner can draft his or her own.
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 protected].