Treat every measurement in AM2E-style practice as a decision, not a number. Classify each reading as satisfactory, investigate, or unsatisfactory, and attach the reason: the criterion used, the conditions of the test, and what you checked before deciding. That classification habit, rehearsed until it is automatic, is the transferable skill for scenario-based assessment.
Turning a number into a verdict: satisfactory, investigate, or unsatisfactory
Measurement assessment rewards classification with justification. A three-way verdict, each option backed by a stated criterion and the test conditions, is more defensible than a reflexive pass or fail.
A binary pass/fail habit breaks down whenever a reading sits near a limit or the test conditions were imperfect. Build a three-way model instead: satisfactory means the value meets the criterion under valid conditions; unsatisfactory means it fails a clear criterion under valid conditions; investigate means the conditions were invalid, the value is borderline, or two credible interpretations conflict. The middle category is a distinct technical outcome leading to different actions: retest with equipment disconnected, compare against a calculation, or check the method itself.
Trace a contrast to make the habit concrete. Suppose a test yields a value fractionally above a stated minimum threshold. Recording it as unsatisfactory without comment ignores the possibility that a connected electronic load influenced the result, while recording it as satisfactory ignores that the margin is thin. The defensible entry records the value, notes that the criterion is not met and that sensitive equipment was left connected, and states the follow-up: retest with those items isolated from the circuit.
Why dead tests come first: sequencing follows safe isolation
The order of tests is a safety logic, not a memorised list: verify isolation before dead tests, complete dead testing before energising, and leave live measurements until the circuit is proven safe to energise.
Reason through the chain rather than reciting it. Continuity and polarity checks on de-energised conductors are meaningful, and safe, only once the circuit is proven dead with a verified voltage indicator and proving unit, plus locking-off and warning procedures. Insulation resistance testing applies elevated voltage between conductors, so it also belongs in the dead phase, with sensitive equipment disconnected and the test voltage matched to the circuit's rating. Only when dead testing is complete does energising for live measurements become a defensible step.
A sequencing error is instructive to trace. If a test voltage is applied with electronic equipment still connected, the equipment can be damaged and its components distort the reading, so the recorded value describes the equipment, not the fixed wiring. If a circuit is energised before polarity is confirmed, an inverted line-neutral fault may go undetected until something is plugged in. In both cases the recorded numbers look plausible while the evidence behind them is invalid, which is exactly what a classification habit is designed to prevent.
Continuity of protective conductors: R1+R2 and the long-lead method prove different things
R1+R2 measured line-to-protective-conductor at the furthest point tests the combined path; a long-lead method isolates the protective conductor alone. Choosing the method that answers your specific question is the skill.
These methods differ in what fault they can reveal. A line-to-protective-conductor reading at the circuit's far end exercises the whole path including the line conductor, so a high value can mean a resistive joint in either conductor. A long-lead measurement of the protective conductor alone removes the line conductor from the equation, so a discrepancy between the two readings localises the problem to one conductor. Method choice is therefore a diagnostic decision, not a preference.
Apply the distinction with a short example. A ring final circuit gives a combined reading roughly double what the conductor sizes and lengths predict, while the long-lead reading on the protective conductor alone matches the calculation. The differential points to excess resistance in the line conductor path, perhaps a loose connection at an accessory, rather than a corroded earth. Recording only the combined value produces a number that fails expectation but no located fault and no justified classification.
Insulation resistance: the reading depends on what you left connected
Insulation resistance measures leakage between conductors and between conductors and earth. Anything electronic left connected, plus damp, drags readings down, so disconnect-and-retest is a legitimate decision path, not a workaround.
The variables that move an insulation reading are mostly physical: surge protective devices, dimmer modules, electronic control gear, and appliances all present leakage paths that are not faults in the fixed wiring. Long cable runs lower readings simply because cable area is large, and damp in external or unheated areas does the same. A commonly taught benchmark for low-voltage circuits is a minimum of roughly one megohm, with healthy circuits typically reading far higher, so a value near the floor signals a need to examine conditions before condemning the installation.
Worked scenario one. On a practice circuit, a candidate measures line-to-earth insulation resistance on a lighting circuit and obtains 0.9 megohms. The mistake: recording it as unsatisfactory and recommending rewiring. The better decision: notice the electronic dimming modules and surge protective device, disconnect them, retest, and find the fixed wiring comfortably above the benchmark. The classification becomes satisfactory with a note that connected equipment suppressed the initial value. Why it matters: the first verdict triggers needless destructive work, while the second correctly localises the reading to the devices. The reverse case is equally real: if the value stays low after disconnection, damp or degradation in the fixed wiring now stands confirmed.
Earth fault loop impedance: when a low reading is a warning, not a success
Measured loop impedance includes parallel paths through bonding and other earthed metalwork. A value far below the expected external figure plus circuit resistance signals those paths, so compare against expectations before recording.
Build the comparison around two reference points. The external loop impedance is a supply-side property; adding the circuit's line and protective conductor resistances predicts roughly what a far-end measurement should show. Tabulated maximum values assume conductors at operating temperature, and resistance rises with temperature, so a value above the tabulated maximum is unsatisfactory even when measured cold; correction reasoning runs the other way, judging whether a cold reading below the maximum would still comply at operating temperature. A value far below the external figure plus calculated resistances points instead to parallel paths.
Worked scenario two. A candidate measures impedance at the far end of a circuit and obtains a value dramatically lower than the external figure plus the calculated conductor resistances. The mistake: recording it as an excellent result and moving on. The better decision: recognise that parallel paths through bonding are dominating the reading, note that the measurement does not represent the circuit alone, and use the calculated resistances as the supporting evidence for the circuit's own path. Why it matters: a low reading can mask a high-resistance joint in the circuit's protective conductor, turning a hidden fault into a clean-looking record.
| Measurement | Dead or live | What it establishes | Common misreading of the result |
|---|---|---|---|
| Continuity (R1+R2 or long lead) | Dead, after isolation | A continuous, low-resistance protective path exists | Accepting a plausible value without comparing against conductor size and length |
| Insulation resistance | Dead, sensitive equipment disconnected | Leakage between conductors and earth is within limits | Attributing a low value to the wiring when connected devices or damp caused it |
| Polarity | Dead, then confirmed live | Line, neutral and protective conductors terminate correctly | Skipping it because other results looked normal |
| Earth fault loop impedance | Live | The fault loop is low enough for timely disconnection | Treating a very low value as automatically good without checking for parallel paths |
| RCD operation | Live | The device trips within its required time at set multiples | Recording trip time without noting the test current multiple used |
Documentation: a reading without its conditions is incomplete evidence
Schedules and certificates must record the value, the method, and any departure or limitation so another competent person can verify the decision. Honest recording under imperfect conditions is a professional-standards issue, not paperwork.
A schedule of test results earns its value when a stranger can reconstruct your reasoning: the measured value alongside the criterion it was judged against, the method where it matters, any equipment disconnected for the test, and any investigate classification with its planned follow-up. In scenario-style assessment the record is often the only window into your thinking, so an entry holding just a number and a tick discards the very judgement the task is designed to demonstrate.
The professional-standards dimension is direct: never adjust a record to fit a limit, and never present a retest taken under different conditions as if it were the original. A genuine departure from expected practice is recorded as a departure with its implications stated, and a result affecting safety is escalated as part of the task. Practise writing one-sentence justifications for each verdict during study, because stating a reason on paper keeps the classification honest when conditions are imperfect and time is short.
Case drill: a ten-reading classification exercise with a self-check rubric
Practise with fictional readings, classify each in one written line, then score yourself against a rubric covering classification, stated reason, and sequencing. Repeat the drill until borderline cases feel routine rather than stressful.
The exercise: write ten fictional results across the measurement types in the table above, deliberately including three borderline cases, such as an insulation value just below a benchmark with devices connected, a loop impedance far below the external figure plus calculated resistances, and a missing polarity entry on an otherwise clean sheet. Classify each as satisfactory, investigate, or unsatisfactory with a one-line reason. Expected observations: the borderline items all land in investigate with different reasons, sound values pass without hesitation, and the missing polarity entry is classified as incomplete evidence requiring the test, not silently ignored.
Score yourself on a four-point rubric per reading: correct category; correct reason naming the criterion or condition; a stated follow-up where investigate was chosen; and no invented certainty, meaning you did not upgrade a conditional value to a firm pass. A self-check milestone worth aiming for is eight of ten with all four points, sustained across two question sets you write yourself. This score is a learning benchmark for the drill, not a prediction of any assessment outcome.
An adaptable preparation sequence. First, map each measurement to its purpose and its dead-or-live position until you can state both from memory. Second, run the classification drill with values near limits. Third, convert verdicts into schedule-style records with one-line justifications. Fourth, run full scenarios under time pressure, sequencing isolation through live testing. Readiness checks: state every measurement's purpose and position without notes, justify a verdict near a limit in one sentence, produce a complete record unprompted, and explain why each step of the isolation-to-live-test sequence precedes the next. For administrative details, the issuing body's own site is the authority.
- Step one: build a purpose map linking each measurement to the fault it detects and its dead-or-live position.
- Step two: drill ten-reading classification sets with deliberate borderline values.
- Step three: convert verdicts into schedule-style records with one-line justifications.
- Step four: run full scenarios under time pressure, sequencing isolation through live testing.
- Step five: retest yourself against the rubric on a fresh self-written set before declaring readiness.
References and further reading
Use these references to explore the concepts and check the latest information from the relevant organizations.