Reduction of the methodological error in measuring the frequency characteristics of the impedance of LAN cables

Abstract

A brief overview of the standards required for the serial production of LAN cables of various categories is given. It is shown that for twisted pairs LAN cables of categories 5e and higher, it is necessary to finalize the procedure for measuring key frequency characteristics (in particular, determining the frequency dependence of the actual value of the wave resistance), regulated by the current GOST R 54429-2011 “Symmetrical pair communication cables for digital transmission systems. General technical specification”. The reason for the errors of network analyzers – the influence of an extended cutting section – is revealed, and the hypothesis that it is impossible to eliminate it within the framework of existing models is substantiated. A hypothesis is formulated about the interference of signals reflected from the beginning of the twisted pair and from the place of its entry into the shell as the main cause of distortion of the results of practical measurements. Experimental data confirming the correctness of the proposed assumptions are presented. The modernization of the calibration procedure of the measuring system is proposed without updating the hardware and software of the vector circuit analyzer. The desired result is achieved by using a distributed calibration load, individual for each type of cable, based on a segment of a twisted pair, which is manufactured in a production laboratory. Using calculated data and measurement results, it is shown that the use of such a load reduces from 14 to 3 % the methodological error in measuring the impedance of widespread category 5e cables. The results obtained are invariant to the type of measuring system and can be used in the production testing of modern LAN cables of high categories.