Until recently, the ISO Class E / CAT6 standards were the measure of all things. Now, proposals for CAT6a, ISO Class Ea, and ISO Class Fa standards are hot on the agenda. While it took four years to adopt the CAT6 and ISO Class E norms, it can be safely assumed that the new standards will be more quickly accepted.

Among technicians and manufacturers there is confusion regarding the impact of these new standards on testing equipment. This article aims to shed light on the current state of testers and where we go from here, including where the new standards will take the testing industry.

Higher Frequency Ranges
As a result of the new standards, the major difference in field-testing equipment will be an increase in bandwidth. For CAT6a and ISO Class Ea the bandwidth will be extended from 1–250 MHz to 1–500 MHz. For ISO Class Fa the measuring range will be increased from 1–600 MHz to 1–1000 MHz. This boost in bandwidth establishes a new definition of accuracy for measuring instruments. For CAT6a / ISO Class Ea there will be a new Level IIIe accuracy specification class and for ISO Class Fa a new Level IVe definition.

It is worth noting that Level IV and Level IVe are downwardly compatible to Level IIIe. In practice, this means that any currently used Level IV CAT7 test device is suitable for CAT6a or ISO Class Ea. Of course, some threshold values will change to accommodate the future 10 gigabit application function. However, this is "only" a software problem for programming the correct threshold value curves in each case.

The manufacturers of field-testing hardware will take these changes in standardization into consideration through corresponding firmware updates. Under the new design standards, so-called "CAT6" testing devices with Level III accuracy are unfortunately no longer acceptable for certification purposes. However, certification instruments for CAT6a / ISO Class Ea/Fa applications are now available.

The Other "What's Next": ANEXT and AFEXT
In standardization circles attention is heavily focused on the new test parameters for acceptance tests with field-testing devices. Their practical use and market acceptance are still far from certain.

For laboratory testing purposes the measurements of ALIEN NEXT and ALIEN FEXT – also known as ANEXT and AFEXT – are being discussed as new test methods.

ANEXT and AFEXT, or rather the Power Sum values of ANEXT and AFEXT, describe the crosstalk of adjacent cables on one cable, and play an important role in CAT6a and ISO Class Ea as well as 10 Gbit/E applications. For the sake of repetition ANEXT is used in this article to describe all other "Alien" parameters. The aim of ANEXT parameters is to define the influence of foreign interference on data-grade cable. The standardization for this allows for two (2) test set-ups:

1. The "6 around 1" Method – a link is surrounded by 6 disturbers. Seven (7) equally long links are bundled together and defined along a straight run.

2. The "Drum" Method – 4 data-grade cables are wound around a drum with a defined diameter.

The idea behind both methods is to simulate a repeatable worst-case disturbance scenario. The key word here is "repeatable", since ANEXT is a highly fluctuating entity. The smallest change in the geometric set-up of the test sequence causes a variation in the effect on ANEXT. A different positioning of the cable, connector components or patch cables also generates new test values. The described 2 methods are very good for determining a worst-case maximum value as they provide excellent data on the maximum expected disturbance of the links or, alternatively, of the cables on each other. Everything that occurs later in practice can only have a weaker effect.

In practice, the above set-ups hardly apply since permanently installed cable is rarely laid using the "6-around-1" Method. This would only cover the disturbances to links among one another. A far more disruptive effect in daily use is the occurrence of electro-magnetic waves from outside. Disturbers in a building are not restricted to data cable. There are hundreds of other electro-magnetic sources in or near the building, such as air conditioning units, electric motors, electric cables, transmitter masts, microwave ovens, telephones and PCs. Even weather influences, such as thunderstorms, can have an effect. Test measurements in the field are therefore very dependent on daily conditions and are not easy to simulate. The measurements depend on how many external disturbance sources are simultaneously switched "on" or are within range. For example, if a technician is taking test measurements in a normal office block, he will register different values on the weekend than he would during the week. Why? Because in most offices the majority of electrical appliances are switched off over the weekend and the air conditioning is either off or running on low.

The influence of data-grade cables on each other must also be considered when field-testing. This involves an extremely complex, time-consuming series of measurements. A test series for ANEXT is several dimensions greater in complexity than simple CAT6 or CAT7 tests with hand-held meters. With technicians already working under pressure this would lead to an unacceptable additional burden in time and money.

On the issue of field-testing equipment, ongoing discussions on standardization are centered around several field testing models aimed at dealing with the above-mentioned phenomenon. The regulatory bodies have as yet not even been able to agree on a joint model (i.e., a test method). The reason for this is that the known models are still a long way from suitability for any practical application and that the problem of the effects of external electro-magnetic influences cannot be given consideration any longer. There is still plenty to do with regard to standardization procedures before a workable solution is found. Indeed, there is even a certain consensus of opinion against measuring these parameters in the field due to the high costs this would involve.

The most plausible solution for the ANEXT problem is to define a suitable interference resistance for CAT6a products "ex works" so that, at least in the context of ANEXT and associates, a smooth 10 gigabit Ethernet operation in the field is guaranteed. When cables, panels, connectors, and patch cable are resistant to ANEXT from the word go, it is no longer necessary to verify this effect in the field.

What Will Happen to Existing CAT6 Systems?
Special cases will need to be defined for standardization here. For the TSB155, for example, cases will be identified in which 10 gigabits could even run on short CAT6 links. However, it still remains to be seen whether these special cases are suitable for everyday practical purposes.

Test Cabling and Test Connectors Are Growing in Significance
As things stand, manufacturers can only guarantee CAT6a capability for the whole channel link, since standardization designs for components are still at the development stage. The mix-and-match idea from the world of CAT6 cannot be applied here any longer. We would be well-advised to pay close attention to the maker's instructions regarding the components to be used. The lack in compatibility between the individual manufacturers creates a special challenge for measuring hardware. The RJ45-connectors on test equipment are the interfaces to the link under test. For CAT6a products, test plugs and connectors at the link are still mechanically compatible RJ45 connections. But what happens when these interfaces are no longer electrically compatible with the link to be tested, as with the CAT6 line? Then the test connectors or test cables will also no longer be electrically compatible and will falsify test results.

In practice the outcome is as follows: if you do not use the test connectors and cables of the corresponding manufacturer, you will no longer have any guarantee that your test results are accurate. In the worst case a wrong test cable would falsify the result to such an extent that your test equipment would evaluate a link as defective even though it may in reality be just the opposite, simply because the test connector and the link do not match.

The ISO Class Ea has left one small backdoor open: at present it only defines channel measurements. This means that you can always use the right test cable by applying the manufacturer's patchcord. The TIA in the USA set new benchmarks here a long while ago and in its CAT6a designs also defined permanent link values. It is only a matter of time before the ISO Class Ea follows suit and also defines permanent link values.

So What Can You Do in the Case of Permanent Link Test Measurements?
As the mix-and-match principle identified above no longer works, it is strongly recommended that the test cable for permanent link measurements comes with a matching connector.

The simplest way to do this is to employ field testing devices which can also use patch cords for permanent link measurements. This type of test equipment is adjusted to the test cable to be used by means of a simple zero adjustment and can then always make use of suitable patch cords.

Incidentally, the application of patch cords for permanent link measurements also has additional advantages - not only for CAT6a measurements, but also for 100 MHz or 250 MHz measurements. One advantage of this is that patch cords produce an absolutely realistic test result for the measurement, since the patch cord type preferably selected for test procedures is also the type used later for operating the plant. The second advantage is that measurements with patch cords are also extremely cost effective. It is a well-known fact that test cords are expendable. The test connectors wear out in the course of time and have to be constantly replaced. If you want to use expensive special cable or connectors you'll have to dig deep into your pocket to maintain your test device in good working order. In contrast, patch cords are exceptionally cheap components and help the user to reduce the test costs by a considerable margin.

Conclusion
The establishment of the CAT6a and ISO Classes Ea and Fa is moving forward at a steady pace. This raises hopes that, at least in the case of standards for field testing, there will soon be clarity about what to do and what not to do. One thing is already apparent: devices of 500 MHz and more are well-equipped to face this new challenge.

About the Author
Konstantin Hüdepohl is Market Development Manager at IDEAL INDUSTRIES. His areas of responsibility in Europe include LAN cable testing procedures. Mr Hüdepohl is also an active member of the standardization panels for DKE 715.3.2 and IEC TC46WG9. For more information, contact IDEAL INDUSTRIES at www.idealindustries.com.

What is your experience with this? Tell you fellow readers now!