A Fluke DSX certification report is the evidence that a structured cabling installation performs to the standard it was specified to. It is what separates a certified installation from a collection of cables that might work. Understanding what the report is actually telling you — what each parameter measures, what the numbers mean, what a marginal pass looks like, and which results actually matter for real-world network performance — is useful whether you are an installer reviewing your own results or an IT manager or building owner receiving a completed installation.

This guide explains how to read a copper cable test certificate from a Fluke DSX or equivalent certifier, what each key parameter means in plain terms, how to interpret pass and fail results, and what to look at beyond the headline pass/fail result.

What a Fluke test report certifies

A Fluke DSX CableAnalyzer — or equivalent calibrated field tester meeting the accuracy requirements of ANSI/TIA-1152-A and IEC 61935-1 — tests each installed copper link against a defined set of performance limits for the cable category and test standard selected. For a Cat6 installation, the tester checks the link against the TIA Cat6 Permanent Link or Channel limits, testing each parameter across the full frequency range of the standard. Every link either passes or fails. Every pass or fail is recorded with the measured value, the limit, and the margin between them.

The report produced by LinkWare — Fluke’s reporting software — is the formal record of the installation’s performance. It maps each test result to a physical outlet or panel port, records the test date, the tester model and serial number, the calibration date, the test standard used, and the technician ID. A complete report covers every link in the installation, not a sample. A report with only a handful of results for a 200-port installation should prompt questions.

The key principle throughout the report is that the tester compares each measured value against a limit and calculates a margin — the difference in dB between what was measured and what the standard requires. A positive margin means the link passed with headroom to spare. A negative margin means the link failed. A very small positive margin means the link passed but is operating close to the edge of the specification.

Permanent link vs channel — which test type matters

The first thing to check on any test report is which test model was used — permanent link or channel. This distinction is fundamental.

The permanent link tests the fixed installed infrastructure — the horizontal cable from patch panel port to outlet faceplate, including the keystone jack at each end but excluding the patch cords at both ends. Maximum 90 metres. This is what the installer certifies at completion, using permanent link adapters that connect the tester directly to the installed cable without introducing the patch cords. Permanent link testing certifies the infrastructure that will be in the walls for the lifetime of the installation.

The channel tests the complete end-to-end signal path including the patch cords at both ends — from switch port to device. Maximum 100 metres total. Channel testing verifies operational performance under working conditions and is useful for troubleshooting, but the channel result includes the patch cords which can change. For handover certification, permanent link is the standard test model.

A report that shows channel results rather than permanent link results is not a certificate of the installed infrastructure — it is a snapshot of end-to-end performance at one point in time with whatever patch cords happened to be connected. It is not a substitute for permanent link certification. If a test report does not state “Permanent Link” as the test model, clarify before accepting it as a project certificate.

The parameters explained

Wire map

Wire map is the first sanity check — it verifies that each of the eight conductors is correctly connected from end to end with no opens, shorts, reversed pairs, crossed pairs, or split pairs. A split pair is where individual conductors from different pairs have been accidentally connected together at a termination point — the cable will appear to pass a continuity test but will fail on crosstalk because the protective twisted pair geometry has been broken.

Wire map failures are installation errors, not cable defects. A crossed pair or reversed pair at a keystone jack termination is the most common cause. Split pairs are less common but harder to spot visually and are almost always caused by using the wrong wiring standard at one end — T568A at one end and T568B at the other, for example. Wire map must pass before the tester will proceed to electrical testing.

Length, propagation delay and delay skew

Length is reported as an electrical measurement — the time it takes a signal to travel through the cable, converted to a distance using the cable’s Nominal Velocity of Propagation (NVP). This may differ slightly from the physical cable run length because the NVP is a characteristic of the specific cable being tested. The limit for a permanent link is 90 metres; exceeding this is a straightforward failure requiring the run to be re-routed or a telecommunications room location to be reconsidered.

Propagation delay measures how long the signal takes to travel from one end of the cable to the other for each pair — reported in nanoseconds. The limit for Cat6 is 555 nanoseconds for a 100-metre channel. It is rarely a failure on its own but is included in the report as a system integrity check.

Delay skew measures the difference in propagation delay between the fastest and slowest pair in the cable. Because Gigabit and 10 Gigabit Ethernet transmit simultaneously over all four pairs, the signals must arrive at the receiver within a defined time window for correct reassembly. The limit for Cat5e and above is 45 nanoseconds. A high delay skew result — particularly in excess of the limit — can cause intermittent errors that are difficult to diagnose because the link appears to function at lower speeds. It is almost always caused by a damaged cable or an unusually long cable run where pairs of very different twist rates are used.

DC loop resistance and resistance unbalance

DC loop resistance measures the total resistance through a wire pair looped at the far end — the sum of both conductors’ resistance. The limit for Cat6 is 21.4 ohms for a 100-metre channel. High DC loop resistance affects PoE power delivery — more resistance means more voltage drop across the cable, less power reaching the device. Failures here typically indicate a damaged conductor, a poor termination contact, or CCA cable — where aluminium’s higher resistance pushes the result over the limit on longer runs.

Resistance unbalance — specifically DC Resistance Unbalance, which measures the difference in resistance between the two conductors in a pair — is the test that CCA cable fails regardless of link length. Manufacturing variation in the thickness of the copper cladding around an aluminium core means the two conductors in a CCA pair have measurably different resistance. On a correctly made pure copper cable, DC Resistance Unbalance is negligible. If a report consistently shows DC Resistance Unbalance failures or marginal results across multiple links, CCA cable is the most likely explanation.

Insertion loss

Insertion loss — sometimes called attenuation — measures how much signal strength is lost as the signal travels through the cable. It is measured in decibels across the frequency range of the test standard and compared against a limit line that becomes more stringent at higher frequencies. The limit is a maximum — the measured value must stay below the limit line across the full frequency range for the link to pass.

On the insertion loss graph in LinkWare, the limit line is the only test parameter where you want the measured trace to be below the line rather than above it. Insertion loss increases with cable length and with frequency — a longer cable or a higher test frequency will show higher attenuation. A result close to the limit on a shorter-than-expected run should prompt investigation — it can indicate damaged cable, a kinked section, or a poorly made termination. The practical consequence of high insertion loss is reduced signal-to-noise ratio, which in turn limits the achievable data rate and reliability of the link.

Return loss

Return loss measures the proportion of signal energy that is reflected back towards the transmitter due to impedance mismatches in the cable channel. Every point where the cable’s impedance changes — a connector, a kink, a section of cable with damaged pair geometry — causes a reflection. The reflected signal arrives back at the transmitter slightly delayed and at reduced amplitude, where it can interfere with subsequent transmitted signals and degrade performance.

Return loss is expressed as a positive dB value where higher is better — a return loss of 20dB means that only 1% of the signal power was reflected. The limit line is a minimum that the measured value must stay above. Poor return loss is most commonly caused by impedance discontinuities from poor quality or incorrectly installed connectors, excessive untwisting of pairs at termination, cable ties applied too tightly, sharp bends that deform the pair geometry, or mixing cable from different manufacturers in the same channel where impedance characteristics differ slightly.

NEXT — Near End Crosstalk

NEXT measures the interference induced from one pair onto an adjacent pair at the same end of the cable as the transmitter — the end where the crosstalk is most damaging because the induced noise is at its strongest relative to the received signal. It is measured as a positive dB value where higher is better — a NEXT of 40dB means the crosstalk signal is 40dB below the transmitted signal level, which is good. The limit line is a minimum.

NEXT is typically the parameter with the tightest margins in a Cat6 installation and the most common source of marginal pass results. It is measured for all six pair combinations — 1,2-3,6; 1,2-4,5; 1,2-7,8; 3,6-4,5; 3,6-7,8; 4,5-7,8 — and the worst pair combination is highlighted in the summary. Poor NEXT is almost always caused by installation issues rather than cable quality: excessive pair untwisting at termination is the single most common cause. For Cat6, TIA limits untwisting to 13mm at the termination point. Every millimetre of untwist beyond that relaxes the pair geometry and degrades NEXT at high frequencies.

Power Sum NEXT (PSNEXT) is the combined crosstalk effect of all disturbing pairs on a single victim pair simultaneously — it is a more demanding test than pair-to-pair NEXT and is the figure that matters most for Gigabit and 10 Gigabit Ethernet, which transmit on all four pairs simultaneously. On the report, PSNEXT results are presented alongside pair-to-pair NEXT and should be checked separately.

ACR-N — Attenuation to Crosstalk Ratio, Near End

ACR-N is calculated from the NEXT and insertion loss measurements — it represents the difference between the signal level received at the near end and the crosstalk noise level at the same point. It is the signal-to-noise ratio of the link, and it is arguably the single most important indicator of real-world data transmission performance. A high ACR-N means the signal is comfortably above the noise floor. A low ACR-N — even with NEXT and insertion loss individually passing — means the link is operating with a thin margin between signal and noise.

For IT managers reviewing a test report, ACR-N and its power sum equivalent PS ACR-N are the numbers that most directly answer the question “how well will this cable carry data?” They should be checked across the full frequency range, and any link where the worst margin is under 2dB should be noted as potentially marginal under demanding conditions.

ACR-F — Attenuation to Crosstalk Ratio, Far End

ACR-F (formerly called ELFEXT) measures the ratio of the received signal to the crosstalk noise induced at the far end of the cable from transmissions on other pairs — the far-end equivalent of ACR-N. It is most significant on longer cable runs where far-end crosstalk becomes a more meaningful proportion of the received signal level. Power Sum ACR-F (PS ACR-F) is the combined version and is the figure used in standards compliance.

Alien crosstalk — PSANEXT and PSAACR-F

Alien crosstalk measures interference from adjacent cables in the same bundle — not from pairs within the same cable, but from completely separate cables running alongside it. It is only required for Cat6A certification and for 10GBase-T installations. PSANEXT (Power Sum Alien Near End Crosstalk) and PSAACR-F (Power Sum Alien Attenuation to Crosstalk Ratio, Far End) are the two alien crosstalk parameters in a Cat6A report. Alien crosstalk is the most demanding test in a Cat6A certification and the parameter most likely to produce marginal results in high-density cable bundles — which is why U/FTP shielded Cat6A, with individually screened pairs that block alien crosstalk at source, consistently produces better alien crosstalk results than UTP in dense installations.

Understanding pass, fail and marginal results

Every parameter in a Fluke report produces one of three outcomes: Pass, Fail, or Pass* (marginal pass). A standard Pass means the measured value cleared the limit with a margin comfortably outside the tester’s measurement accuracy. A Fail means the value did not meet the limit. A Pass* — indicated by an asterisk — means the value passed the limit but fell within the tester’s accuracy uncertainty band. The asterisk notation is required by the standards and cannot be disabled on a compliant tester.

A Pass* is technically a compliant result — the link meets the standard from a compliance standpoint. But it means the link is operating close to the limit, and any degradation over time — increased temperature, minor mechanical stress on a connection, or the addition of a new device with different impedance characteristics — could push it to a fail. Multiple Pass* results across a single link, or a consistent pattern of Pass* results across many links in the same installation, warrant investigation before the installation is accepted.

The margin figure is the most informative number in the report. A link that passes NEXT with a margin of 8dB is a fundamentally different result from one that passes with a margin of 0.3dB, even though both are technically compliant. For installations where long-term reliability matters — and for 10G applications where the performance headroom is tighter — reviewing the worst-case margins across the installation, not just the pass/fail headline, gives a much more useful picture of installation quality.

What to prioritise when reviewing a report

Not all parameters carry equal weight for real-world performance. A useful way to think about them is in tiers:

Tier 1 — performance critical: NEXT, PSNEXT, ACR-N, and PS ACR-N are the parameters that most directly determine whether the link will carry data reliably under real network conditions. For Cat6A and 10G installations, alien crosstalk (PSANEXT and PSAACR-F) belongs in this tier. Return loss also belongs here — it affects signal integrity across the full frequency range and is sensitive to installation quality at connection points. These are the parameters to scrutinise most carefully.

Tier 2 — distance and capacity: Insertion loss and ACR-F confirm that the cable is within distance limits and that the signal-to-noise ratio at the far end is adequate. Important and must pass, but less likely to produce marginal results in a correctly installed channel within the permitted distance.

Tier 3 — sanity checks: Wire map, length, propagation delay, delay skew, and DC resistance confirm that the physical installation is correct and the cable is intact. Failures here indicate installation errors or damaged cable rather than marginal performance at the limits of the specification. They must pass, but passing them is a baseline requirement rather than a differentiator of installation quality.

What to check before accepting a certificate

Before signing off a structured cabling installation based on a Fluke test report, the following are the minimum checks:

Test standard: Confirm the correct standard is shown — TIA Cat6 Permanent Link, TIA Cat6A Permanent Link, or the equivalent ISO class. The standard determines what parameters are tested and what limits apply. A report run against the wrong standard is not evidence of compliance with the correct standard.

Test model: Confirm Permanent Link is shown, not Channel. Channel tests are useful for troubleshooting but are not a substitute for permanent link certification.

Complete records: Every outlet and patch panel port should have a corresponding test record. A report showing 150 results for a 200-port installation is missing 50 links. Missing results should be explained and tested before the installation is accepted.

Tester calibration: The report should show the tester model, serial number, and calibration date. Fluke recommends annual calibration. A tester with an expired calibration produces results of uncertain accuracy — the standards require calibration within validity for the results to be certifiable.

Marginal passes: Review the distribution of Pass* results. A small number of marginal passes across a large installation is normal — Cat6 is a demanding standard. A high proportion of marginal passes, or multiple Pass* results on the same links, suggests installation quality issues that should be investigated and remediated before the installation is accepted.

Margin statistics: For demanding projects, go beyond the headline pass/fail and look at the worst-case NEXT and ACR-N margins across the installation. An installation where the majority of links have margins above 3dB is a fundamentally better result than one where many links are scraping through at 0.5dB, even if both are technically fully compliant.

Common causes of failures and marginal passes

Frequently asked questions

What does a Pass* mean on a Fluke report?

A Pass* (asterisk) means the parameter passed the limit but fell within the tester’s measurement accuracy uncertainty band — the result is still a pass, but close enough to the limit that measurement uncertainty is a factor. The asterisk notation is required by ANSI/TIA-1152-A and IEC 61935-1 and cannot be disabled on a compliant tester. A Pass* is technically a compliant result — the link meets the standard — but it indicates the link is operating close to the specification limit and deserves investigation.

How often should a Fluke tester be calibrated?

Fluke recommends annual calibration for the DSX CableAnalyzer series. Calibration should be carried out by an authorised service centre. Results produced by a tester with an expired calibration certificate are of uncertain accuracy and may not be accepted as valid certification evidence. The calibration date should be clearly shown on the test report.

What is the difference between TIA and ISO test limits?

TIA (Telecommunications Industry Association) standards are the dominant North American standards. ISO/IEC 11801 and EN 50173 are the international and European equivalents. For most parameters the limits are similar, but the ISO standard is in some respects slightly more stringent. UK and European installations are typically certified to ISO/IEC 11801 or EN 50173 limits — referred to on Fluke reports as Class D (Cat5e), Class E (Cat6), or Class EA (Cat6A). Both sets of limits are valid for certification — what matters is that the correct standard for the project specification is selected on the tester before testing begins.

Can a link fail the Fluke test due to the patch cord rather than the installed cable?

In permanent link testing, no — the permanent link adapters connect the tester directly to the installed cable, excluding the patch cords entirely. The permanent link test result reflects only the performance of the installed infrastructure. In channel testing, yes — a poor quality or damaged patch cord at either end of the channel will affect the result. This is one reason permanent link testing is the correct model for installation certification, and channel testing is primarily used for troubleshooting operational problems.

Should I be concerned if NEXT margins are consistently low across an installation?

Yes — a pattern of consistently low NEXT margins across multiple links usually indicates a systemic installation issue rather than an isolated problem. The most common causes are a termination practice producing excessive pair untwist across the installation, a batch of substandard keystone jacks, or cable that is performing at the lower end of its specification. Individually marginal links can be remediated by re-termination. A systemic pattern needs investigation at the root cause level before it can be resolved.

Summary

A Fluke test report is the formal evidence that an installed structured cabling channel meets the performance specification it was designed and installed to. The headline pass/fail result is the starting point — the margin figures tell you how confidently it passed. Wire map, length, and DC resistance confirm the physical installation is correct. Return loss and NEXT confirm the electrical performance of the connections. ACR-N is the signal-to-noise ratio that matters most for data transmission. For Cat6A and 10G installations, alien crosstalk results are the most demanding test and the most meaningful indicator of whether the installation will perform reliably under real-world conditions. Pass* results are compliant but warrant scrutiny. Margin statistics across the full installation, not just the pass/fail summary, give the most accurate picture of installation quality.

If you have questions about cable specification, test certification, or structured cabling installation, get in touch with the DTECH team — we supply Cat6, Cat6A, and fibre cabling systems to installers and IT teams across the UK, Europe, and the Middle East.