The Evolution of Category 6 Field Testing
By Mark Johnston, RCDD, FLUKE Networks
Turn your cabling costs into assets by assuring functionality with testing.
This article explores some of the lesser-known issues surrounding field testing of advanced structuring cabling systems, particularly as they relate to using hand held field testers. Rather than trot out some of the same tired phrases you’ve probably read before (keep the twists tight, no sharp angles, don’t pull tie wraps so hard they compress the cable, etc), we’ll use a Q&A format. BICSI’s own Richard Dunfee posed many of these questions, and I’ve added a few based on my 21 years developing and supporting field test equipment. Think you know all the answers? Read on and find out!
How can you claim to test Cat 6 when it is not yet a standard?
This is a question we have often received. Field tester manufacturers are commonly called upon to test next generation cabling systems the moment they arrive on distributor’s shelves often years before standards are completed. Customers buy products with new, higher performance claims, and they want proof those claims are being met. Typically field testers are designed to allow for changes in limit lines and test requirements via flash software updates. Written detailed claims are always that the tester meets a specified draft of a standard. In fact, if you look at results screens they usually say something like “Category 6 Draft 9”. Many manufacturers who have confidence in their tool often go a step further and guarantee that the field tester will be compliant with the final standard. Given the glacial pace of standards evolution, this can take a while! For example, the first ‘Category 6 field tester’ on the market was the Microtest OMNIScanner in June of 1998 over 4 years ago! Today, over 10 drafts of the standard later, owners of original OMNIScanners can download software that brings them up to date with today’s requirements.
Do different testers take different approaches?
The reporting requirements of the standards mean that all credible testers will provide the same measurements and provide a similar-looking report output. Of course, there are different approaches (time domain, frequency domain, wideband, narrowband, vector, scalar) but these really become important when you get a * or a FAIL. When you get a PASS, everyone is happy (assuming the unit was accurate and the autotest speed was reasonable). Failures are where differences in approaches can make the difference between quickly highlighting a fault with a high degree of confidence, or wasting time guessing and finger pointing. Be sure when evaluating a tester you see a demonstration of how it can find NEXT and return loss faults, and whether it can clearly differentiate cable from connecting hardware failures.
Another key difference can be in the software. As an example, one of an installer’s biggest irritations can be *’s in the results. An asterisk means the result is closer to the limit line than the accuracy of the measurement, and thus the result is uncertain. Most customers will not accept * results, and so the installer ends up fiddling with terminations, re-running tests, or doing whatever is necessary to push the result to a PASS. Well, one way some testers allow you solve this problem is to simply ‘turn off’ the * function. However standards require manufacturers to ALWAYS show stars; we’re not supposed to allow them to be disabled. Customers should insist that stars always be enabled in reports they receive.
Another difference can be found in link adapter design. As we will see later, full interoperability has not yet been achieved with Category 6 connecting hardware from all suppliers. This has forced field tester suppliers to embark on an interesting journey, full of twists and turns and surprises (as we learn things about patch cord variability, how to calibrate out cord effects to meet permanent link requirements, and other things). Ask your test set supplier to contrast their approach to the others. There are essentially three different approaches and they are completely different! Get two suppliers in the same room and you’re guaranteed real entertainment!
Is today’s “Cat 6” test the same as yesterday’s?
A Category 6 test today is in fact quite different from one run a couple of years ago. Changes include:
Permanent link vs. Basic link the permanent link moves the test reference point out to the end of the test cable, which reduces available NEXT margin by about 2 dB at 250 MHz.
Limit line changes these changes have not been significant, but tweaks in the limits for most measurements have occurred over time.
Frequency changes after draft 3 the test bandwidth for Category 6 was extended from 200 MHz to 250 MHz at the request of the IEEE.
3 dB rule when insertion loss is less than 3 dB (links ~ < 15m) you cannot fail return loss regardless of the result.
4 dB rule (ISO 11801 2nd Edition only, not TIA Cat 6) when insertion loss is less than 4 dB you cannot fail NEXT regardless of the result.
Adapters We’ve learned a lot about plug interoperability issues and patch cord return loss variability. Today’s testers provide much better stability and reliability.
As this shows, testers need to evolve and YOU need to keep up! Fortunately all these changes are easily accommodated with software updates or new accessories your tester investment remains intact.
For more information on these changes and generic testing information, see cabletesting.com
What are the most common troubleshooting problems, and how do the new test sets make it easier in the field?
Wiremap still the most common failure in the field. New wiremap designs are not pair-based, so they can isolate individual conductor opens. They also identify 568a/b mix-ups by name, and can detect split pairs. They can also now show you at which end you will find the short or open.
While wiremap is the most common, NEXT and return loss failures are still the toughest to solve. However new techniques for displaying faults in the time domain such as HDTDX and S-Bands make it much simpler to find and correct these problems.
What are some of the most overlooked capabilities of a test set that can be utilized in the field for troubleshooting? How do they work?
One useful feature in most certification tools is the ability to measure length without using the remote unit. You can measure the length of cable left on a spool this way. Another useful application is to measure pair lengths from a patch panel prior to terminating the far end. If any pairs are zero length that alerts you to a local termination issue.
Re-certification is a way to use PC software to re-check prior autotest results against adjusted or new standards. This can help protect against additional labor costs when standards change.
Other useful features include tone generation that will work with any inductive probe; office locator functions that identify up to 12 different remote probes to match ‘unknown’ jacks in the wiring closet with remote offices; traffic monitors that can activate a hub or switch port and verify the presence of traffic; and adapter use counters, that help you keep track of the level of wear and tear on adapters. Often overlooked are the latest channel adapters, which employ full vector correction to truly isolate the channel from the tester’s mated connection and provide a good view of the performance your network will really see. This is critical since the performance of your network will often depend upon the weakest link patch cords that are rarely tested.
Can the test results be archived and retrieved for later as-built documentation?
Today’s tools can save full plot data and store the information either internally or on removable memory cards. However, this can come at a cost. Not all data is equally compressed. If you are a serious user and saving full plot data you should ask how much space 100 full sweep Category 6 autotests take on a hard disk. There are significant differences in the amount of storage needed by tools that provide the ‘same’ information.
What troubleshooting problems cannot be detected by a certification test set? What other specialized equipment can be used?
Layer 2 and 3 network and packet layer/application problems such as duplicate IP addresses, incorrect switch configurations, and the like are outside the scope of physical layer test tools. Some of these problems can be found with embedded network management systems, while others require the use of a network analyzer such as a NetTool or OptiView.
Another issue you might think would be easy to identify is still impossible with field testers: when you see a pair reversal or crossover, which end has the wiring error?
There are conditions not explicitly covered in standards requirements, which may cause problems. These include noise (of different types), alien crosstalk, reflected FEXT, and common mode NEXT coupling. Field testers will usually detect these problems via a failed NEXT measurement, for example, but troubleshooting becomes more challenging unless you have a very high level of expertise.
What is the biggest thing installers can do to increase their productivity with certification testers?
This answer is easy: people don’t keep their field testers up to date with the latest free software releases. 9 out of 10 calls to our technical support center could have been prevented if units had current software and tests were done with the correct adapters.
As much as manufacturers would like to help, to a great extent our hands are tied. Since these tools are sold through distribution, we do not know who the owners are unless they return registration information. Even though we offer financial incentives for people to do so, a significant percentage of users do not. If we know who you are, and you are willing to accept email product information updates, we will let you know automatically when new software is available. But it is the responsibility of the installer to keep up to date. If we don’t have registration data we can’t advise you. Countless thousands of useless autotests have been run and time wasted because the tester was running an obsolete set of limit lines for the selected test. Don’t let this happen to you!
The second big issue is the use of correct adapters for Category 6 testing. Because we are not yet at a point of universal Category 6 interoperability, and manufacturers use different generations of plugs & jacks, you must match to the system under test. Fortunately this situation is improving, but we won’t be at a point of universal compatibility until sometime next year.
Finally, users are often unaware that free training tools exist. Computer based training is great way to learn the key features and functions of your test tool while working at your own pace at the time and place of your choice.