Tech Talk: Put your cable wiring to the test

Remember the old saying, “You get what you inspect, not what you expect?” This is certainly true when it comes to your premise cabling plant.

Colleges, universities and public school officials assume that cabling installers know what they are doing and that by pulling and terminating wire, the installation is going to meet current and future requirements. This assumption can lead to problems. Although it may take more time and effort, schools and universities should include testing procedures in any wire bid specification. If it is carefully prepared, this section of your bid specification will provide protection for you and the installer.

School officials probably remember the days when they needed only a pair of pliers and screwdriver to terminate a copper cable. Today's demand for high-speed data transfer over copper or fiber requires a much more sophisticated termination. With more users comes the need for greater bandwidth and speed within your structured cabling plant.

The tighter the twist in copper, the higher the performance rating. So, maintaining that twist ratio may be the most important factor in successful copper cable termination. Casual termination practices result in excessive attenuation, crosstalk and other external influences.

Specific standards have evolved under the auspices of several organizations. Why not make sure that your cabling plant meets those standards? TIA/EIA 568A and ISO 11801 constitute the primary building telecommunications standards, laying the foundation for testing up through Category 5. Currently, TIA/EIA 568A Draft 12 represents a proposed revision and update of the entire 568A specification. It incorporates all changes to date and defines a new Category 6 that essentially will be equivalent to ISO Class E. Category 6 will be the nomenclature applied to cabling systems using RJ-45 style connectors and certified to carry 200 MHz traffic. Category 6 also will require testing to a new Level III accuracy, which will incorporate all the existing tests used in TSB95 at the Category 5e level.

Buying high-quality material is not enough — even if the cable and connectors meet specifications, improperly installed cable and connections can deliver performance substantially below the minimum required. Up to 20 percent of all installed cable runs can deliver sub-category performance if the installer is not using proper techniques. And, the effects of poor installation may not be evident immediately if the first user attached to the cable run is operating at only 10Mbps. The problem will become apparent when the user tries to switch up to 100Mbps or gig speed. That is exactly the worst time for such a problem to emerge.

Also, it is essential to use a matched system in which cable and connectivity are consistent and closely matched in terms of characteristic impedance. Therefore, require all cabling to be field-certified after it is installed and before you sign off on the installation.

Your installer should have a minimum of three years experience installing and testing structured cabling systems. All installers should have factory certification indicating they are qualified to install and test the products being used.

The installer should be responsible for furnishing all required testing equipment and for submitting acceptance documentation to the manufacturer for their certification.

Schools should have all of the installed cabling tested, and a school official should be present during all testing. Test the basic link for copper and the optical link for fiber; include cable, connectors and any splices.

Cabling must meet the indicated performance specifications: TIA 568A-A5 Category 5e or TIA 568B Category 5e; TIA 568A Category 6 Addendum, draft 5 or latest; ANSI/TIA/EIA-526-14A for multi-mode fiber, or ANSI/TIA/EIA-526-7 568-B.3 for fiber.

Testing should gather the following information for each cabling element tested:

• A wiremap that indicates the cabling has no shorts, opens, miswires, split, reversed or crossed pairs, and that end-to-end connectivity is achieved.

• Copper should be tested for attenuation, NEXT, PSNEXT, Return Loss, ELFEXT and PSELFEXT data that would indicate the worst-case result, the frequency at which it occurs, the limit at that point, and the margin.

• For fiber, test link attenuation in accordance with ANSI/TIA/EIA-526-14A. Measure optical loss on each fiber at 850 nm and 1,300 nm. Measure the loss on each fiber bi-directionally. Also, test the horizontal link and the backbone link.

Test equipment should be capable of certifying Category 5, 5e and 6 links for copper. To test single-mode and multi-mode fiber, use The WireScope 155 and Fiber SmartProbe+ test instruments.

Test equipment should include S-Band time domain diagnostics for NEXT and return loss (TDNXT and TDRL) for accurate and efficient troubleshooting.

Multimode test equipment should incorporate both 850 nm and 1,300 nm sources in same unit.

Test equipment must include DSP technology for support of advanced measurements. It must make swept frequency measurements in compliance with TIA standards and have dual-function main and remote units for bi-directional testing to eliminate the need to swap optical source and power meter.

The installer should state in writing that 100 percent of the installation meets your specifications. You should reserve the right to conduct a random retest — using installer equipment — of up to 5 percent of the cable plant to confirm results.

If your cable passes the above test, then it can be used for Gigabit Ethernet traffic without further concern. However, failing any of these tests indicates that there is something amiss with the cable.

Save the results of the testing for future reference. It is a benchmark of network performance capabilities that can be useful if a dispute arises between you and the installer. Most cable installers will guarantee their work for 15 to 25 years if you test the wires at installation.

The above requirements should be prepared and included in any wiring infrastructure bid package. If carefully conceived, this section of the bid specification will provide protection for the school and for installers.

Day is senior analyst at KBD Planning Group, Bloomington, Ind., a firm specialized in educational facilities and technology planning.

[email protected] www.kbdplanning.com

• Premise structured cabling plant is a term applied loosely to different building wiring configurations. In its most basic sense, it describes any copper or fiber wiring system designed and installed according to a fixed set of rules. In actuality, it is all of the ingredients that tie data, voice and video networks to a supporting infrastructure of cabling and components to maximize their capabilities.

• Wire mapping is used to identify wiring errors in an installation. It consists of a series of five tests that identify the status of each wire in a twisted-pair environment. A wire-map failure is most often caused by improper termination of the cable at one of the two ends.

• Attenuation or, more accurately, “insertion loss” is the loss of signal strength in a cable.

• NEXT measures crosstalk, or the noise that communications equipment creates for itself. Excessive NEXT can cause problems ranging from intermittent workstation lockups to complete network attachment failure. Maintaining the twist ratio is the most important factor in any successful unshielded twisted pair installation.

• ELFEXT measures crosstalk between pairs at the far end of the transmission line. It is similar to hearing another person's conversation when you are on the telephone.

• PSNEXT refers to the undesired coupling of signals from all other pairs into one pair. This provides a convenient format for specifying the total crosstalk performance of cabling intended to support 1000Base-T. Power Sum is a method to measure the combined effects of crosstalk on a single individual pair.

• Return loss measures the ratio of signal loss power transmitted into a system to the power reflected (returned).