Why Fluke Results Get Rejected — and How to Build a Test Pack That Survives Review

The email nobody wants: the consultant has reviewed your test results and rejected the pack. Cue arguments about retest costs and a handover date that just slipped a week. Here is the uncomfortable truth from the other side of that review — most rejected results were doomed before anyone pressed AutoTest.

The failures that happen before you test

In our experience, the same handful of setup errors accounts for the bulk of rejections:

• Wrong limit line — Cat6A links tested against a Cat6 limit pass easily and prove nothing
• Calibration certificates out of date, invalidating every result after expiry
• Worn permanent link adapters throwing marginal results on perfectly good links
• A bad reference set on fibre — negative loss is physically impossible, and reviewers know it
• Cable IDs that do not match the drawings, so results cannot be tied to links
• *PASS results submitted as if they were clean passes

How a consultant actually reads your pack

Nobody reviews two thousand results plot by plot. The reviewer loads your file into LinkWare and sorts. First check: the limit line — is it the standard the specification calls for? One wrong selection fails the entire pack in thirty seconds. Second: calibration dates on every tester used. Third: a filter for marginal results, which surfaces every *PASS instantly. Fourth: a cross-check of IDs against the drawing schedule. Only then do they sample individual plots — and if headroom swings wildly between adjacent links, they dig deeper.

In other words, the review is automated pattern-matching before it is engineering judgement. Packs fail on hygiene long before anyone questions the installation.

The review is pattern-matching before it is engineering judgement — packs fail on hygiene, not workmanship.

Set up like you will be audited — because you will

Before the first link is tested: confirm the limit against the specification, not against habit — an ISO Class EA permanent link limit and a TIA Cat6A limit are not the same test. Set the correct NVP for the cable actually installed. Fill in the operator, site and project fields — anonymous results invite questions. Check the insertion count on your permanent link adapters; they are rated for thousands of cycles, not infinite ones, and a worn adapter quietly eats headroom. On fibre, set references every morning and after every lead change, then sanity-check the first result — if loss reads negative, stop and re-reference.

*PASS is a warning, not a pass

An asterisked pass means the result sits inside the tester's measurement uncertainty — technically compliant, but too close to call. One or two across a large pack is normal physics. A cluster of them is a message: a worn adapter, a marginal termination habit, or damaged leads. Reviewers read clusters exactly that way, which is why we treat *PASS as a retest trigger on site rather than a debate to have by email three weeks later. Fix the cause, retest the links, and submit a file with a story you can defend.

The daily discipline

Testing that survives review is not heroics at handover. It is six habits, run every shift:

• Verify limit line and NVP at the start of every shift
• Photograph calibration certificates into the job folder before mobilising
• Reference fibre sets each morning and after any lead change
• Name results to the drawing ID as you test — never in a renaming session afterwards
• Review the day's marginals before leaving site
• Back the results file up daily — a lost tester should cost hardware, not evidence

The walls close up, the ceiling tiles go in, and the only visible product of a cabling package is the test file. On our jobs — from commercial works at Canary Wharf to fibre deployment in Germany — the pack is treated as the deliverable, because to the client, it is. If your last one came back covered in red ink, we should talk. Ask us about testing and certification