Shielded cabling systems - DTX Alien Crosstalk

If installed correctly, you should see hardly any alien crosstalk on a shielded cabling system. According to IEC 14763-2, if you test three disturbed (victim) links and the margins are better than 5,0 dB, you can stop testing. A really good shielded cabling system can result in nothing showing on the graphs in the AxTalk Analyzer Software. So practically speaking, we're talking about testing only three disturbed (victim) links on a shielded cabling system - assuming it passes of course. Depending on the bundle size, that's no more than a couple hours of testing at the most.
 
Given that "Alien crosstalk is always superior on a shielded system", you could argue that there is no need to test for alien crosstalk because the performance is always going to be substantially better than the cabling standards requirements.......but again....so long as it is installed correctly.

FACT: An installed shielded cabling system can FAIL the alien crosstalk requirements for Category 6A/Class E_A.
 
In the cabling community, the widespread view is that even if installed incorrectly, the alien crosstalk is still going to be far superior compared to an unshielded cabling system.  Even if the shield is not correctly attached to the connector at each end of the link, it should at least provide enough immunity to stop any significant alien crosstalk from happening. As with any cabling, what seems to be theoretically impossible seems to happen in the field.

Here is an alien crosstalk result from an installed Category 6A rated shielded cabling system: (customer identification removed of course)

It goes without saying that everyone was surprised to see this, having been told specifically by experts in the cabling industry that this could not happen. There is a troubleshooting tool called Disturber Ranking within the AxTalk Analyzer Software. You can use this feature to diagnose which disturber links are contributing the most to the overall power sum alien crosstalk.

Looking at the Disturber Ranking for the link above, we see:

If this was a "cable issue", we would expect to see the contribution spread evenly over the links. However, the contribution seems to be focused on the immediate adjacent links to our disturbed (victim) cable ID 60. That suggests a connector issue.

Just incase the shield was not connected, the DTX CableAnalyzer was used to run a Wire Map test from panel to panel (data center application). The result was as expected, a PASS:

However, in a data center application where you are running from one grounded panel to another grounded panel, the shield can be open (not connected) on a cable and yet the Wire Map will pass. The DTX CableAnalyzer is looking for a simple d.c. continuity between the shield on the main unit and the shield on the remote unit. The d.c. signal will look for anyway it can to get to the remote unit. Using the common building ground is one way, shown below:

** The DSX-5000 CableAnalyzer would report this shield as open since it uses a new measurement system to overcome this issue.

There are a number of ways the shield in the connector can be installed incorrectly. The most common way is to clamp down on the non conductive side of the foil in the cable. It is critical that you obtain the installation instructions from the connecting hardware vendor and ensure your entire team understands how to terminate the shield correctly.

Still think alien crosstalk testing is not required for shielded cabling systems? Take a look at this article on sample testing alien crosstalk.