Is Cat6A cabling needed?

Some of the world’s premier experts on cabling looked at whether Cat 6A cables should be tested In situ, or if the ‘met by design’ standard was sufficient.

Tags: CommScope IncorporationNexansR&M Middle EastSiemon
  • E-Mail
Is Cat6A cabling needed?
By  Georgina Enzer Published  August 17, 2013

Some of the world’s premier experts on cabling looked at whether Cat 6A cables should be tested In situ, or if the ‘met by design’ standard was sufficient.

Network Middle East held the debut edition of the Network Middle East Cable Debate, featuring six top global experts in structured cabling, who debated three top cabling issues at the Marina Yacht Club in Dubai.

The first topic under debate was whether Cat 6A cables need to be tested for alien crosstalk in the field.

It is widely recognised that field-testing of AXT is time consuming and can be inconsistent. Several of the cabling standards state that Class EA, F and FA Channels is ‘met by design’ if certain coupling attenuation requirements are met. This would suggest field-testing is not required, however, the panelists had differing views.

Is alien crosstalk testing needed for Cat 6A cabling?
Alan Flatman: I would start by saying that it really depends on what cable construction is being used. If you are using screened cabling and perhaps double screened cabling like Pairs in Metal Foil individual screened pair cabling, because of the way we design and specify that, and you look at the screening effectiveness or coupling attenuation, I would say the answer is definitely not. But on the other hand, if you are using unscreened cabling, then the alien crosstalk performance is generally not as good, and I would say the answer is definitely yes.

To qualify that I would say if you are contemplating alien crosstalk testing, there are two different ways of doing it, one is by sampling, or statistical test techniques, which are specified in the standards, and the other is a 100% everything to everything testing. You don’t have enough time and money to devote to 100% testing, so it has to be statistical testing. If it is screened I would say you would not have to do it at all, it is over the top, if it is unscreened, do it carefully, do it statistically, but if things in the network change, consider redoing it.

There is a standard that says ‘met by design’ when it comes to alien crosstalk and coupling attenuation. How does that influence your answer in regard to shielded versus non shielded debate?

Alan Flatman: It is a detailed aspect of how we specify the cabling. There is a ‘met by design’ clause for coupling attenuation specifications which means that if the cabling has a measured margin of at least 10 or 15dB, depending on the class of cabling, over the coupling attenuation limit, then according to the standard, you assume that alien crosstalk is ‘met by design’, in other words you don’t have to test it.

Martin Rossbac: For screened systems we normally do lab testing of the coupling attenuation and we design the product so that the threshold is met. We as manufacturer of cabling systems do not offer UTP systems for 10G, so with our systems we mitigate alien crosstalk. But as cable manufacturer we also have a Cat 6A UTP cable, which is resold, and we advise people to do alien crosstalk testing for UTP systems. We just believe it is an unnecessary risk that customers take and they must test for any issues early on in the installation. It is a mistake to not do it.

What we have seen in the industry is that 6A UTP systems are installed and no one does alien crosstalk testing. It’s quite often ignored and we believe that is a mistake. A lot of internal noise can be cancelled away using intelligent technique on the chipset, but with alien crosstalk that can’t be done. If there is external noise, it can affect the application. We know that it is very tough to design true UTP cables that are immune to alien crosstalk.

Alien crosstalk is installation dependent and varies depending on the installation circumstances, and systems with little margin in the laboratory environment are not safe in the field and require testing on site. 10GBase-T is not on the market yet, but it would be this application only which would have alien crosstalk problems. So companies that have an alien crosstalk problem won’t see it today, they will only see it in four or five years.

How can end users validate that potential issues are met by design?
Stefan Ries: ‘Met by design’ is basically defined by standards. I suppose that most of the shielded twisted pair (STP) systems that are established on the market are met by design.

However, following the conversation here I disagree with what was already said. Of course, a ‘met by design’ system design can be realised for UTP. Provided that you are capable of designing a high end UTP system, you can simulate the worst case in the lab.

Add a Comment

Your display name This field is mandatory

Your e-mail address This field is mandatory (Your e-mail address won't be published)

Security code