Marginal Passes - DSP 4x00 CableAnalyzer

This article explains the phenomena known as Marginal Passes. Many installers and end users are beginning to see * appear on their test reports, something that was not seen on previous category 5 installations. An asterisk indicates that the measured parameter value is closer to the corresponding test limit than the guaranteed accuracy of the tester. The confusion is caused by legacy category 5 installations where users were told to reject marginal passes. This was a legitimate requirement for category 5 since the components exceeded the standards so much, that any marginal passes would suggest poor installation practices.

Category 5e and in particular category 6 components do not enjoy the same margin found in category 5 components. This is not a poor reflection on the component manufacturers. Category 5e and in particular category 6 specify a much higher level of performance and are extremely difficult to meet. What this means is that from time to time you can expect a small percentage of marginal passes for category 5e installations and a larger percentage of marginal passes for category 6. The percentage depends on a number of factors including performance levels of individual components (cable or connecting hardware), installation practices, and the quality of the test lead. The high performance level and the resulting confusion around marginal passes also applies to Class D and E installations.

The screens below were taken from a DSP-4100 Cable Analyzer. The NEXT tests were conducted on the same link without moving the test lead against different test standards. What is clear is that the wide margin of 6.8dB reported for category 5 is reduced to 3.1dB for category 5e.

DSP-4100 Cable Analyzer NEXT Test Results

Category 5 NEXT                                                                  Category 5e NEXT

In 1995 when field-testing for category 5 really started, many of the components were so good that they were some 4 years ahead of their time. They unwittingly already met the future specification of category 5e agreed in 1999. Since the components met category 5e requirements, but users were only testing to category 5, marginal passes were rarely seen. There was plenty or margin to play about with. Many manufacturers who offered a warranted category 5 system actually refused to accept results with marginal passes. As the screen shots above show, there is a reduction is the margin the installer has to play with for category 5e. Anything less than perfect installation practices are more likely to show up now. It is therefore inevitable, that in the real world, some results will be marginal.

Standards dictate that any marginal passes are considered the same as a pass and therefore compliant.

What the standards dictate (In detail)
ANSI/TIA-1152 which supports category 5e, 6 an 6A installations states in section 4.2.4 Pass/Fail results.

A pass or fail result for each parameter shall be determined by the specified limits for that parameter. The test result of a parameter shall be marked with an asterisk (*) when the result is closer to the test limit than the measurement accuracy published by the field tester manufacturer for the permanent link and channel. Refer to clause 4,4 for detailed information on measurement accuracy requirements. The field test manufacturer shall provide documentation as an aid to interpret results marked with asterisks. An overall pass or fail condition shall be determined by the results of the required individual tests. Any fail or fail* shall result in an overall fail. In order to achieve an overall pass condition, all individual results shall be pass or pass*. 

NEXT Test Accuracy Region

                                                                                 Figure 1. Accuracy Region

IEC61935-1 which supports both ISO/IEC 11801 and EN 50173 Class D & E systems states in 5,4 Data reporting and accuracy 5.4.1 General - A Pass or Fail result for each parameter shall be determined by the allowable limits for that parameter. The test result of a parameter shall be marked with an asterisk (*) when the result is closer to the test limit than the measurement accuracy for the permanent link or channel

The graph (Figure 1.) shows an example of an equipment tolerance region around a NEXT pass/fail test limit. The red lines above and below the test limit are offset by a dB value equal to the specified instrument accuracy. If the measurement falls within the region below the limit, the measurement is marked with a Fail*. If the measurement falls within the region above the test limit, the measurement is marked with a Pass*. An overall Pass or Fail condition shall be determined by the results of the required individual tests. Any Fail or Fail* shall result in an overall Fail. In order to achieve an overall Pass condition, all individual results shall be Pass or Pass*.

Test equipment Accuracy
Select a tester that is repeatable, accurate and fully complies with the applicable industry standards. One way to assure yourself that a tester meets the standards in all aspects is to look for an independent certification of the tester. For instance, the Fluke Networks DSP-4000, DSP-4100 and DSP-4300 have been independently verified by both Underwriters Laboratories (UL) and ETL-Semko laboratories. These laboratories independently verify that these testers exceed the TIA Level III accuracy specifications.

Some tester manufacturers allow you to manually disable the * to hide marginal results. Disabling the * makes the test(er) non-compliant. This disabling feature is NOT possible within the DSP-4000 Series CableAnalyzers. Hiding marginal results only puts your reputation and business at risk, so why do it? The best thing you can do is select a tester with exceptional accuracy to minimize or eliminate marginal results.

Test equipment Interface adapter cables
Tester interface cables are a very significant component in the accuracy and reliability of the test results especially for the Return Loss parameter. A worn interface cable contributes itself in a very significant way to the measured Return Loss value of the link-under-test. The definition of the Permanent Link model in all the industry standards specifies that tester interface cabling cannot contribute in any way to the test results. In order to be compliant with this requirement, contracts should specify that the installation shall be tested using the Permanent Link test standard with the appropriate tester interface adapters. The Fluke Networks DSP-LIA101S permanent link interface adapters far exceed the standards requirements.

Test equipment - Diagnostics
Another significant time saving characteristic of the tester is its ability to identify the location of a possible defect that causes the link to produce a marginal test result. The DSP-4000 Series testers can be configured in the SET UP menu to report the result of the Autotest with a Pass* if any one test parameter yields a Pass* result. In this configuration of the test tool when a Pass* Autotest result appears, the technician presses the FAULT INFO button to learn which parameter(s) produced a marginal pass result as well as the wire pair or wire pair combination. In addition, the DSP Series testers identify the location of each defect in a graphical format measured in meters or feet from the main test unit. This diagnostic report allows the technician to identify a marginal connector, a suspect termination or whether the cable itself is the source of the marginal performance. Obtaining this diagnostic information immediately upon completion of the certification test (Autotest) allows the installation team to identify the problem, execute the corrective action and retest. This immediate resolution saves time and costs.

If the end-user is expected to receive a warranty from the premise-wiring manufacturer, the contractor should be fully aware of the warranty criteria and conditions. For instance, if the warranty program states that marginal passes are not accepted for any of the test parameters, the contractor may want to construct some sample or trial links and verify that the technicians obtain good results with the hardware components obtained through the distribution channel.

Lastly, the statement of work for the project, agreed to by all parties involved before any work commences, should clearly set all expectations with regard to the test procedure, the industry test standard, the accepted field test equipment, and the format for the deliverable test results.