Black Box Explains... 10GBASE-T testing.
10GBASE-T testing.
In June 2006, the IEEE approved the standard for 10-Gbps Ethernet, or 10GBASE-T (10-GbE). 10-GbE transmission requires a bandwidth of 500 MHz.
Alien Crosstalk
Alien Crosstalk (ANEXT) is a critical and unique measurement in 10-GbE systems. Crosstalk, used in 10/100/10000-BASE-T systems, measures the mixing of signals between wire pairs within a cable. Alien Crosstalk is the measurement of the signal coupling between wire pairs in different, adjacent cables.
The amount of ANEXT depends on a number of factors, including the promixity of adjacent cables and connectors, the cable length, cable twist density, and EMI. Patch panels and connecting hardware are also affected by Alien Crosstalk.
With Alien Crosstalk, the affected cable is called the disturbed or victim cable. The surrounding cables are the disturbers.
Testing
Field certification for 10-GbE consists of two phases. The first phase is to certify the transmssion capability and quality of each individual link. The 10-GbE test limits are identical to CAT6 and ISO 11801, but the frequency range is extended from 250 MHz to 500 MHz.
The parameters are insertion loss, return loss, pair-to-pair near-end crosstalk (NEXT), power-sum NEXT (PS-NEXT), pair-to-pair equal-level far-end crosstalk (ELFEXT), power-sum ELFEXT (PS-ELFEXT), propagation delay, length, delay skew, and wire map.
The second phase is to field certify the cabling system for compliance with Alien Crosstalk (ANEXT), which is the between-channel parameter. This should include sample testing of some links in a bundle to verify compliance with ANEXT.
How to test
Typically laboratory measurements of PS-ANEXT and PS-AELFEXT are based on cables in a “six-around-one” configuration. The central cable is the victim cable, and all the adjacent cables are the disturbers. This test configuration provides a worst case scenario. A total of seven links of equal length are connected to each other at previously defined distances. Every circuit is measured against the other so there are 96 individual measurements.
At this point, it's time- and cost-prohibitive to test all possible wire-pair combinations in the field for ANEXT. One strategy is to use a sampling technique to select a limited number of links for testing. The chosen links should be those most likely to fail, such as the longest links, or shorter links with the shortest distance between connectors. You should also limit testing to links that are bundled together.
Accuracy
Level III testing equipment is designed for measurements to 250 MHz. Level IV testers certify accuracy up to 600 MHz. Manufacturers of test equipment are conforming to the changes in standards with firmware updates.
Got Questions?
