Important revisions to the original TIA/EIA 568 standard may include new and previously issued TSBs, and maybe even a protocol for testing UTP horizontal subsystems.

Revisions to one of the major voice/data telecommunications standards are in the process of being made and it's important that you know about them. For one thing, the original TIA/EIA 568 standard is being updated to reflect certain performance enhancements to the use of unshielded twisted pair (UTP) wiring for high-speed data networks. Also, the new standard, TIA/EIA 568-A, may include previously issued Technical System Bulletins (TSBs) as well as other important addenda.

The new standard, originally appearing in the form of the SP-2840 draft, is scheduled for another (hopefully final) round of balloting this month. Nevertheless, it's important that you're aware of these proposed revisions now. The following discussion provides a complete review of them.

What's included in the revised standard

The addenda for UTP wiring and connecting hardware issued after the original 568 standard was released is to be included within the new standard. This material will no longer exist as addenda. Also to be included within the 568-A standard are the following.

* TSB 36, Additional Cable Specifications for Unshielded Twisted Pair Cables (November 1991).

* TSB 40, Additional Transmission Specifications for Unshielded Twisted Pair Connecting Hardware (August 1992).

* A separate addendum for shielded twisted pair (STP) wiring.

* A change in fiberoptic cabling connecting hardware from ST-type to SC-type connectors.

* The beginnings of a standard for "UTP link performance."

The big news

In the course of discussions with the 568 committee chairman, questions were raised about the status of a procedure for testing UTP, which appears in the annex of the new 568-A standard (i.e., in the SP-2840). Referred to as Annex E, Unshielded Twisted Pair (UTP) Link Performance, this annex attempts to establish a standard procedure for field testing the performance of UTP wiring in horizontal subsystems (i.e., UTP wiring between wiring closets and work spaces).

UTP link performance refers to the performance of a complete end-to-end UTP. Performance, in this context, refers to a cabling scheme's ability to reliably support a range of transmissions at different frequencies. The ability to do so successfully or not is of great concern to end users and installers alike when it comes to determining whether or not a cable plant is properly installed. The problem has been that, in spite of all of the industry standards for telecommunications and cabling, there are no standards to go by when it comes to how UTP cabling in the horizontal subsystem should be tested.

The operative phrase on the subject is "end-to-end." This is meant to include testing of not just the cabling, but the connecting hardware at both ends of a circuit as well. After all, cabling systems are always installed with terminations at both ends of a circuit (i.e., an outlet at one end and connecting hardware at the other). The use of cabling, therefore, always involves the transmission of signals through the entire link, including outlets and connecting hardware. Thus, it would seem appropriate to test not just the individual components but the entire link on an end-to-end basis.

Unfortunately, all of the industry standards to date (especially the 568 standard) have focused on performance parameters for individual components, but not on their performance in concert with one another (i.e., in the form of a complete link).

Annex E of the new 568-A standard addresses this issue, but only in the form of a nonbinding, "informative" manner. While this is helpful, it only begs the question of exactly how the industry should handle this important matter.

And so the really big news on the horizon is that, while the 568-A standard will go no further than the informative nature of its Annex E did in the SP-2840 draft, there is a new standard in the works that will address the issue of the UTP testing in a definitive way. The new standard will be referred to as follows: TIA/EIA Technical Systems Bulletin 67 (TSB-67), Link Performance Specifications for Field Testing of Unshielded Twisted-Pair Cabling Systems. There is a draft version of the TSB now in circulation, dated December 16, 1994.

The fact that we now have an addendum in circulation for a standard which, itself, has not yet been finalized speaks for itself. It's no wonder that keeping up with the standards process can sometimes feel like a wild goose chase. Nevertheless, it's pleasing to see that we finally have some definitive movement on the subject of testing. Without consistent guidelines for testing, there has really never been a truly meaningful way of determining whether or not complete end-to-end systems have been properly installed, or in fact, meet the intended applications. Let's take a closer look at the content of the new TSB-67 draft.

The proposed TSB-67

First of all, TSB-67 concerns itself with only 100-ohm UTP cabling and connecting hardware in the horizontal subsystems of structured cabling systems. The "link" of interest, therefore, generally includes outlets in faceplates at workstations, the UTP wiring itself, and terminating or connecting hardware inside wiring closets. In addition, TSB-67 allows for a transition connector close to the workstation, which might take the form of a junction box or some other cross-connection device used to extend horizontal UTP wiring to the workstation. (Note: this practice was sometimes used in the past for nonpermanent wiring, but is strongly discouraged for new installations of structured cabling systems.)

In terms of specific configurations defined by the new standard, TSB-67 actually defines two possibilities. The first type is referred to as a channel and consists of the outlet, the transition connector, the UTP wiring itself, connecting hardware in the wiring closet, and dedicated line or patch cords on each end. See Fig. 1 for a view of this first configuration.

Table

Frequency

(MHz)

1-31.25

31.26-100

The second configuration is referred to by the new standard as a basic link, as shown in Fig. 2, and is much more consistent with contemporary systems design. It basically assumes that there are no intermediate cross-connections or splices in the horizontal UTP run, and includes the outlet device at the workstation end and the connecting hardware at the closet end. The basic link also allows for a line or patch cord at each end, but recognizes that these ancillary cords are independent of the permanent horizontal subsystem and will vary depending on the types of systems in use at any given time.

Per the parameters of the 568 standard, the new TSB-67 standard also assumes that in any configuration, the length of the horizontal UTP wiring does not exceed 90 m, and the combined length of the line and patch cords at either end of the link/channel does not exceed 10 m.

The new standard defines five specific tests to be performed on the UTP link.

* Wire map

* Length

* Attenuation

* Near end crosstalk (NEXT)

* Propagation delay

The attenuation and NEXT values are to be taken using swept/stepped frequency voltage measurements, or equivalent methods, using field test equipment. The step size for NEXT tests are as shown in Table 1. The step size for attenuation tests should not exceed 1 MHz.

Wire map test. The purpose of the wire map test is to determine whether or not the continuity and polarity of pairs and conductors have been properly installed. TSB-67 further states that for each of the 8 conductors in a 4-pair drop, the wire map test examines the following.

* Proper pin termination at each end.

* Continuity to the remote end.

* Shorts between any two or more conductors.

* Crossed, reversed, split pairs, or any other miswires.

Length test. The length test is just as it sounds: a measure of the length of the link to determine compliance with 568 specifications. The test, however, is conducted in two slightly different ways, depending on which of the two test configurations is in use. Since the channel type configurations will, by definition, include terminated line and/or patch cords at both ends, the length of the link could be up to the full 100 m allowed by the 568 standard.

In the case of the basic link configuration, however, the cable plant is assumed to be of the structured, permanent type, and should, therefore, be limited to 90 m. Added to the 90 m then are line cords on each end to serve as leads to the testing devices; such lead should be no more than 2 m. This results in a total length of 94 m in the case of the basic link length test.

Attenuation rest. The attenuation test focuses on signal loss throughout the link or channel. It takes into account the cumulative attenuation produced by all components in a link, including each plug/jack combination, patch and/or line cords, and of course the UTP cable itself. Attenuation values when measured in accordance with the proposed TSB-67 standard for basic links are summarized in Table 2. (Note: values for channel type links are omitted since most contemporary cable plants are of the basic link type.)

NEXT test. NEXT loss is a measure of signal coupling from one pair to another. As in the case of the attenuation test, the measures are derived from swept/stepped frequency voltage measurements. The test calls for measurements to be taken on all pair combinations. When performed in accordance with the proposed TSB-67 standard for basic links, the results should conform to the values summarized in Table 3. (Note: values for channel type links are omitted since most contemporary cable plants are of the basic link.)

Propagation delay test. The TSB-67's treatment of this test is quite brief and reads as follows.

The propagation delay can be determined from phase angle measurements of the output signal relative to the input signal. Alternately, the propagation delay can be determined (approximately) from TDR measurements.

The test calls for the measurement of propagation delay on all pairs, with the worst case (longest delay) to be reported.

Helpful annexes

Like most of the TIA/EIA standards, TSB-67 includes a series of annexes that provide supporting documentation on procedures (Annex A), accuracy guidelines for test instruments (Annex B), interpretive explanations for tests that fail (Annex C), and electrical length measurement methods (Annex D). Annexes A and B are considered to be part of the formal standard, while Annexes C and D are provided for informative purposes only.

When to expect TSB-67 ratification

As far as when the proposed TSB-67 will be ratified into formal existence, it's impossible to say. If experience is any indicator, however, it should be no sooner than next summer. But even with TSB-67 in place, you should still recognize that there is much more to a structured cabling system than solely the horizontal subsystem. And there are certainly more cable types to consider than simply UTP. Until such time as the TIA/EIA expands the scope of its testing standards to include these other subsystems and cable types, testing methods and, therefore, metrics in the industry will still be inconsistent and, therefore, somewhat arbitrary.

SUGGESTED READING

EC&M articles:

"The EIA/TIA 568 Cabling Standard" October '93 issue. "The Do's and Don'ts of UTP Wiring" June '94 issue. "Testing Methods for UTP and Fibre" August '94 issue. For copies, call 913-967-1801.

Mark W. McElroy is Senior Manager with Peat Marwick's Management Consulting Practice Malvern, PA. He is professionally certified by BICSI as a registered communications distribution designer (RCDD).