There are many reasons why acceptance measurements on network data links can fail. These reasons range from measurement settings that have not been set correctly, to more than a dozen individual measurement parameters that can generate errors, triggered by inadequacies in the system, to impairments on the part of the measurement equipment used as the cause of faulty measurements. How to check the condition of your measuring system to eliminate this potential source of error will be briefly described here.
Measuring plugs as the culprit
A measurement system for certifying copper networks generally consists of the two measurement devices themselves and the measurement accessories. These are primarily the measuring cables with the measuring plugs that establish the connection to the system under test.
In our consideration here, we assume that no defect or software error in the devices has ruined our measurements. It is helpful to have a valid factory calibration, which already states that the devices are in a condition that guarantees measurements in the permitted tolerance range and thus excludes errors in the evaluation in the device.
The main cause of measurement errors emanating from the measurement system is therefore wear effects on the measurement cables, more precisely on the RJ45 measurement connector. Wear that affects the measured values is caused by repeated plugging of the measuring cables. RJ45 plugs are specified with a number of mating cycles of 750, but this is a laboratory value. In practical use, test plugs may well break down earlier, depending on how they are handled and also on the mechanical compatibility between the socket in the DUT and the test plug. However, since the degree of wear and the current influence on the measured values cannot be derived directly from the number of mating cycles, an external reference standard is required against which the condition of the test plug can be evaluated.
Reference ring as "referee"
The best way to be able to check the condition of the measuring system at any time, even in the construction site environment, is to take along a, let's call it a "reference ring", including a measurement record of the same.
In order to obtain a comparable reference to the currently commonly laid links of performance class EA, i.e. with 500 MHz bandwidth, the reference link should also be at this performance level. Such a reference link is therefore usually constructed from a high-quality laid cable, nowadays typically at least category 6A terminated with components also according to category 6A. Careful assembly of the link is also important. Subsequently, this ring is measured and the measurement protocol is packed with it. It is important that this "reference measurement" is carried out with the measuring device when it is in perfect condition, i.e. freshly calibrated and equipped with current firmware and the measuring adapters, measuring cables and measuring tips used are new or as good as new in order to achieve the optimum measuring success. This procedure is even described in the measurement standard IEC 61935-1 as one of the methods to obtain reliable measurement results.
If one notices a continuous deterioration of the values for ohmic loop resistance and the high frequency parameters during the measurement of a cabling system, a check measurement should be carried out against this brought reference ring. If the measured values now turn out to be significantly worse compared to the "reference values", one should carry out a replacement of the measuring tips or the measuring cable (depending on the system) in order to remain within the specifications for these measurements. It is best to also pack a new set of measuring cables or tips with the reference ring in order to be able to continue work without interruption.
More on the subject
This article is only intended to briefly highlight a possible cause of error in failed acceptance measurements. If you want to know more about the topic, we offer a wealth of tools dedicated to troubleshooting. These include various white papers, especially on high-frequency faults, which deal in detail with the possible causes of faults using examples and which we also make available here successively. Furthermore, we offer a series of live webinars, or recordings of webinars we have already held, which explicitly deal with troubleshooting when certifying on data links. If you have more in-depth questions, just contact us directly. We'll be happy to help, with detailed advice and a wealth of good tips from the field.
Softing IT Networks