The myth of connecting hardware power-sum performance claims

Vendors who claim that their connecting hardware is "power-sum" compliant appear to offer a performance advantage over those vendors who specify their hardware`s worst-case pair-to-pair near- end crosstalk (NEXT) performance. Although perfectly applicable to the specification of cabling and cabling performance, some power-sum claims for connecting hardware are ill-founded and could result in end-users getting less than what they bargained for.

Sep 1st, 1998

Valerie Rybinski

The Siemon Co.

Vendors who claim that their connecting hardware is "power-sum" compliant appear to offer a performance advantage over those vendors who specify their hardware`s worst-case pair-to-pair near- end crosstalk (NEXT) performance. Although perfectly applicable to the specification of cabling and cabling performance, some power-sum claims for connecting hardware are ill-founded and could result in end-users getting less than what they bargained for.

Historically, the rationale behind specifying power-sum near-end crosstalk (PS-NEXT) requirements for connecting hardware was based on the thinking that if power-sum channels are going to be specified, then power-sum components should be specified as well. Although there is some logic behind the statement, significant supporting data exists to demonstrate that a worst-case power-sum measurement is difficult, if not impossible, to apply consistently and accurately to connecting hardware qualification. To make matters worse, because a test algorithm for worst-case power-sum data collection does not exist, different manufacturers` power-sum claims may mean different things.

For example, a company can easily claim power-sum Category 5 performance for a connector based on data collected at one frequency point (such as 100 megahertz), measured in one test orientation (from the plug orientation only), with a high-end test plug rather than the full test-plug range specified in the TIA/EIA-568a Commercial Building Telecommunications Cabling Standard, or after testing in only one termination mode, such as the differential mode.

No basis in standards

A manufacturer must analyze hundreds of test configurations before reasonably claiming measured power-sum performance. Even then, there is no assurance that the worst-case test configuration has been evaluated. The PN-2948 Connecting Hardware Task Group of the Telecommunications Industry Association (Arlington, VA) recently addressed these concerns by specifying a worst-case pair-to-pair next characterization for connecting hardware that is sufficent to ensure channel and basic link ps-next performance. In fact, until such time as a power-sum test algorithm is defined, claims of "power-sum compliant" connecting hardware are not supported by industry standards.

The good news is that you can easily calculate an equivalent worst-case pair-to-pair next margin that correlates to Category 5 power-sum performance. By assessing the statistical relationship between pair-to-pair next values in connecting hardware, the tia PN-3727 utp Systems Task Group has determined that a common and differential mode pair-to-pair next margin of 3 decibels over Category 5 requirements is sufficient to ensure Category 5 power-sum performance.

In specifying connecting hardware, be sure to note worst-case guaranteed pair-to-pair next performance and verify that measurement margins are based on testing with and without common-mode terminations. Although power-sum claims may sound appealing, the power-sum algorithm tends to mask worst-case pair-to-pair performance for components that may prove cumulative--for example, two worst-case pair-to-pair combinations that are in a series and interact adversely in a channel configuration. Claims of connecting hardware power-sum performance are not sufficient to ensure proper channel near-end crosstalk performance.

More in Home