6A’s final hurdle: testing (But not the kind you think)

Committee members optimistically look at December as a potential publish date.

Committee members optimistically look at December as a potential publish date.

by Patrick McLaughlin

If all goes according to plan, the final chapter of the Telecommunications Industry Association’s (TIA; www.tiaonline.org) standardization of Augmented Category 6 (Category 6A) specifications will be written before this calendar year ends. The group has its sights set on a meeting the second week of December, at which all outstanding issues may be resolved and the standard approved for publication.

“The standard is in great shape,” says Val Rybinski, global sales engineer with Siemon (www.siemon.com) and newly-re-elected chair of the TIA’s TR-42.7 Telecommunications Copper Cabling Systems Committee. “The transmission numbers have been firm for a long time; they have not changed in two years,” she adds. In addition to establishing link and channel performance requirements, TIA standards also specify the performance of components within those systems. And it’s some of those component specifications-for connecting hardware, in particular-that TR-42.7 must finalize before the standard is complete.

Measuring component compliance?

From a standards-process viewpoint, the only part of the standard that is still under review-and, therefore, still has the potential to change-is the procedure for measuring connecting-hardware component compliance. “We froze the entire body of the document except a few minor technical changes” related to the component-measurement procedure, Rybinski states. So, it is safe to say the link and channel specifications are indeed final, as they have been for two years.

The most recent category-rated TIA specification, Category 6, specified performance levels to 250 MHz. Category 6A doubles that frequency to 500 MHz, which is in sync with the maximum frequency of the Institute of Electrical and Electronics Engineers’ (IEEE; www.ieee.org) 10GBase-T specifications. Such high frequencies have been new territory for the group, which has been challenged to make some minute measurements of connectors to 500 MHz without using ASTM International (www.astm.org) tests as references, because no such tests exist.

“We can make measurements with a certain amount of inaccuracy-say 1 dB, for example,” Rybinski continues. “What we’d like to do in the standard is get that inaccuracy reduced, and the way to do that is to develop better test fixturing. We believe the new measurement fixture made specifically for connecting hardware will allow us to make very accurate measurements.”

Effectively isolating a connector and measuring that connector’s performance presents numerous challenges from technical and practical standpoints, and TR-42.7 formed a task group-headed by the committee’s co-chair, Sterling Vaden-to develop and refine a test fixture for the purpose.

Among the technical challenges the group has faced are maintaining good impedance around 100 Ω at 500 MHz, and one of the biggest practical challenge is isolating the twisted pairs from the measurement. Historically, the characteristics of the test leads used in the process have been subject to change during the testing process, due at least in some part to the leads’ movement while testing is underway. The fixture recently developed and currently being used by the test-fixture task group keeps the test leads very precisely located, thereby allowing them to be subtracted from the measurement and allowing as pure a connector measurement as possible.

If the connector fits…

The next step, one that is set to be taken as this article is going to press, is a series of round-robin testing in which multiple manufacturers’ connectors are tested using a single fixture. Hugo Draye, product manager for certification products with Fluke Networks (www.flukenetworks.com), points out that such testing is critical to arguably the most important characteristics of standard-compliant products-interoperability and backward compatibility.

“Manufacturers of connecting hardware are ready to conduct round-robin testing of each component,” says Draye. “That happens late in the standards process,” and he notes that the group is close to a final document. “At some point, jacks and plugs need to be defined as components. Link and channel performance reports must be based on individual components.”

Draye recalls that similar challenges faced Category 6, when early pre-standard plugs and jacks were not interoperable among vendors, nor necessarily backward-compatible with lower-category hardware. Back then, technology progressed and connecting-hardware manufacturers refined their components to perform within the TIA-established parameters that would ensure both interoperability and backward compatibility. With Category 6A, TIA is on the threshold of finalizing the means for measuring connectors’ ability to perform within those parameters.

Somewhat ironically, it has been another form of Category 6A testing-alien crosstalk and the means of field-testing for it-that has generated about 99% of the pre-standard buzz. On that topic, Rybinski offers two facts that may surprise many:

  • Everything related to field-testing procedures for Category 6A have been closed. The specifications are complete.
  • Standards have never made field testing mandatory. Category 6A is no exception.

Sampling undefined

To that end, Draye observes, “The TIA and IEEE have said, ‘Here’s how you measure, here are limit lines, et cetera.’ The standards clearly define what to measure and how to do it to ensure valid results. But they do not address the sampling method.”

In other words, the standards are silent and users are left to make their own decisions about testing all, some, or no Category 6A circuits for alien crosstalk.

As Rybinski pointed out, this is nothing new.

PATRICK McLAUGHLIN is chief editor of Cabling Installation & Maintenance.

More in Home