Test and measurement — instrument selection, safety rating and misreading
Test Instruments and the Measurement Errors That Mislead You
- Applies to
- Multimeters, clamp meters, insulation testers, voltage indicators and thermal cameras used on generators, UPS systems, inverters, drives and installations
- Difficulty
- intermediate
- Competence required
- qualified electrician
- Diagnosis complexity
- Not a fault diagnosis. This is the instrument knowledge every other guide assumes, and the source of a surprising share of wrong conclusions.
- Electrical system
- All systems — AC 240 V / 415 V 50 Hz, DC battery and bus systems, PV arrays, drive outputs
- Safety classification
- live electrical
- Author
- EmersonEIMS Engineering
- Technical review
- Mr. Kararaho
- Last reviewed
- 2026-07-27
Scope — read this before relying on the guide
This guide sets out diagnostic method, ranked causes and safety requirements. It deliberately does not state equipment-specific numbers — test voltages, resistance and component values, torque figures, acceptance windows and pin assignments are referred to the manufacturer's documentation for your exact model rather than given here, because a plausible-looking figure for the wrong machine is more dangerous than no figure at all.
Direct technical answer
Two instrument errors produce more wrong diagnoses than any other, and both give confident, plausible readings. The first is using an averaging meter on a distorted waveform: inverter outputs, UPS outputs, drive outputs and generators feeding non-linear loads are not clean sine waves, and an averaging meter reads them wrong while displaying a perfectly believable number. Use a true-RMS instrument for anything downstream of power electronics. The second is the ghost voltage: a modern high-impedance meter will read induced voltage on a long, disconnected cable running alongside live conductors, which looks exactly like a live circuit and has caused people to condemn healthy installations — and, far worse in the other direction, to distrust a genuine reading. Use an instrument with a low-impedance mode, or a two-pole voltage indicator, when the question is whether a conductor is actually live. Beyond those, three rules cover most of the rest: select an instrument whose measurement category and voltage rating suit the installation, not just the voltage; use a DC clamp for DC, because an AC clamp reads essentially nothing on a battery circuit; and never use a multimeter as your means of proving dead when a two-pole indicator and a proving unit are the correct tools.
01Symptom description
Controller / display
- Readings that do not agree between two instruments on the same point
- A voltage present on a conductor that should be dead, with no explanation
- Current reading of zero on a circuit that is obviously working
- Insulation reading that changes markedly between tests on the same winding
- A meter reading that contradicts the equipment display
Indicators
- Instrument category rating lower than the installation being tested
- Damaged, cracked or repaired test leads
- Calibration date long expired, or absent
- Fuse blown in a meter's current input
Sounds
- Continuity buzzer sounding on a circuit that is not continuous, indicating a lead or range problem
- Nothing — most instrument errors are silent, which is precisely the problem
Smells
- Burnt smell from a meter after a measurement, indicating it was used beyond its rating
- Any burnt instrument must be taken out of service, not simply refused
Behaviour
- A reading that disappears when a second meter is connected — the signature of a ghost voltage
- Readings that vary with how the leads are routed, indicating induced voltage or poor lead condition
- An AC clamp reading nothing on a battery or DC bus circuit
- Voltage measured on a drive output that bears no relation to what the motor is receiving
- Insulation resistance that improves as the test continues, which is normal absorption rather than a fault
Visible
- Instrument category and voltage rating printed on the meter and on the leads
- Lead insulation, shrouding and probe tip condition
- Fuse rating in the meter, and whether it is the correct type
- Calibration label and date
- Battery condition in the instrument itself
02What the fault means
In plain language
Instruments do not simply report the truth. They report what they are designed to measure, in the way they are designed to measure it, and if the situation does not match those assumptions the number on the screen can be confidently wrong. Knowing where each instrument misleads is as important as knowing how to use it, because a wrong reading sends you diagnosing a fault that does not exist.
Technical explanation
Measurement error in field work arises mostly from three mismatches. The first is waveform: an averaging meter is calibrated to display the RMS value of a sine wave by scaling a rectified average, so it is only correct for a sine wave. Downstream of any power electronics — inverter, UPS, variable-speed drive — or on a generator feeding significant non-linear load, the waveform departs from sinusoidal and an averaging instrument reads incorrectly while giving no indication that it has. A true-RMS instrument computes the heating-equivalent value and remains valid on distorted waveforms. The second is input impedance: modern digital multimeters present a very high input impedance, which is desirable because it does not load the circuit, but it also means capacitive coupling between a long disconnected conductor and adjacent live cables can develop a readable voltage at that high impedance. This ghost voltage is real in the sense that it exists, and misleading in the sense that it can source essentially no current. An instrument with a selectable low-impedance mode loads the circuit slightly and collapses the ghost while leaving a genuine supply unaffected, which distinguishes the two definitively. The third is the physical quantity: an AC clamp meter senses the changing magnetic field of alternating current and reads essentially nothing on direct current, so battery, DC bus and PV string measurements require a Hall-effect DC-capable clamp. Layered over all of this is safety rather than accuracy: IEC 61010 defines measurement categories describing the transient overvoltage energy an instrument is built to survive at its location in an installation, and using an instrument of too low a category is dangerous in a way that is invisible until a transient arrives.
03Common causes, ranked
These are ordered by likelihood. Presenting every possible cause as equally probable is a failure of diagnosis, not thoroughness.
Most likely
start here- Averaging meter used on a distorted waveform downstream of power electronics
- Ghost voltage read on a disconnected conductor by a high-impedance meter
- AC clamp used on a DC circuit, reading nothing
- Measuring at a convenient point rather than at the point that answers the question
Possible
check next- Instrument measurement category or voltage rating unsuited to the installation
- Damaged leads or poor probe contact producing intermittent or low readings
- Meter fuse blown in the current range, so current reads zero
- Wrong range or function selected
- Insulation reading taken without allowing for temperature or absorption
Less common
after the above- Instrument out of calibration
- Thermal camera emissivity setting wrong for the surface, giving a misleading temperature
- Instrument battery low, affecting accuracy
- Measuring drive output voltage with an instrument that cannot interpret the switched waveform meaningfully
- Clamp jaws not fully closed, or a conductor not centred, on a current measurement
Model specific
verify per unit- Category and voltage rating are printed on the instrument and its leads and must both suit the installation — the leads are rated separately and are frequently the weak link
- Low-impedance mode is present on some instruments and not others, and is invoked differently
- Insulation tester output voltages are selectable and must suit the equipment under test
- Clamp meters differ in whether they are AC-only or AC/DC capable
- Thermal camera emissivity handling differs between instruments
Environmental
site conditions- Long parallel cable runs, which promote capacitive coupling and ghost voltages
- Humidity and surface contamination affecting insulation readings
- Temperature, which affects insulation resistance substantially and must be accounted for
- Strong magnetic fields near clamp measurements
- Bright sunlight making displays hard to read, which causes transcription errors
Installation related
built in- No accessible, safe test point at the place a measurement is actually needed
- Unlabelled circuits, so the wrong conductor is tested
- Test points that force awkward working positions, encouraging shortcuts
Maintenance related
deferred work- Instruments never calibrated or checked against a known reference
- Leads never inspected before use
- Proving units not carried, so proving dead is done improperly
- Instruments kept in service after being subjected to an over-range event
Component level
electronics- Test lead damaged internally beneath intact insulation
- Meter input fuse blown
- Clamp jaws damaged or contaminated, affecting accuracy
- Instrument internal damage from an over-range event
04Safety requirements
Isolation
- Where the work permits it, isolate and prove dead rather than measuring live — the safest measurement is the one you do not need to take live
- Where live measurement is genuinely required, treat it as live working with the corresponding controls
- Prove dead with a two-pole voltage indicator and a proving unit, not with a multimeter
Lockout and tagout
- Where measurement follows isolation, secure the isolation with your own lock before working
- Do not rely on a measurement taken before someone else could have reinstated a supply
PPE
- Arc-rated protection appropriate to the prospective incident energy at the measurement point
- Eye protection
- Insulating gloves where the task requires them
- Leads with adequately shrouded probes, and finger guards in place
Stored energy
- An insulation tester charges the winding or cable capacitance to a high voltage; ALWAYS discharge after testing
- Capacitive circuits can retain charge after a measurement
- DC bus capacitors must be verified discharged before any contact, regardless of what other measurements show
Specific hazards
- MEASUREMENT CATEGORY IS A SAFETY RATING, NOT AN ACCURACY ONE. It describes the transient energy the instrument can survive at its position in the installation. An under-rated instrument works perfectly until a transient arrives, and then fails in the operator's hand. Select the category and voltage rating for the installation, and check the LEADS as well as the meter.
- Never use an insulation tester on a circuit containing electronics — it will destroy drives, controllers and inverter boards. Disconnect them first.
- Never use a multimeter as the sole means of proving dead. It has no proving mechanism and a failed meter reads the same as a dead circuit.
- Never assume a non-contact voltage detector proves anything. It is an indicator only and can miss a live conductor behind screening.
- Never open-circuit a current transformer secondary to insert a meter while primary current flows — short the secondary first
- Inspect leads before every use; damage beneath intact insulation is common and invisible
Stop and call a qualified professional if
- You do not have an instrument of the correct measurement category for the installation
- The measurement requires live working and you are not competent and equipped for it
- Readings are inconsistent and you cannot establish which is correct
- The instrument has been subjected to an over-range event — take it out of service
- There is no safe means of accessing the point that would actually answer the question
05Tools required
| Tool | Why it is needed |
|---|---|
| True-RMS digital multimeter with low-impedance mode | The default instrument; true-RMS is essential downstream of power electronics, and low-impedance mode is what distinguishes a ghost voltage from a real supply |
| Two-pole voltage indicator and proving unit | The correct instruments for proving dead — a multimeter is not, because it cannot be proved in the same way |
| AC/DC clamp meter (Hall effect) | DC circuits — batteries, DC bus, PV strings — read essentially nothing on an AC-only clamp |
| Insulation resistance tester with selectable test voltage | Winding and cable insulation, at a voltage suited to the equipment |
| Low-resistance ohmmeter | Winding balance and joint resistance, which a standard multimeter cannot resolve adequately |
| Thermal camera with adjustable emissivity | Loose joints and hot components; emissivity must match the surface or the temperature is misleading |
| Power quality analyser with logging | Intermittent faults and waveform problems that no spot reading can capture |
| Spare fuses of the correct type and rating for the instrument | A blown current-range fuse reads zero current, which is easily mistaken for a fault |
06Diagnostic decision tree
1. Is the instrument category and voltage rating suitable for this installation — meter AND leads?
Yes → Continue
No → Stop. This is a safety rating, not an accuracy one, and it fails without warning.
2. Are the leads and probes undamaged, and the fuse intact?
Yes → Continue
No → Replace them. Damage beneath intact insulation is common and invisible.
3. Is the measurement downstream of an inverter, UPS or drive, or on a generator with non-linear load?
Yes → A true-RMS instrument is required — an averaging meter will read wrong and give no warning
No → Continue
4. Is the circuit AC or DC?
Yes → Match the instrument — an AC clamp reads essentially nothing on DC
No → Establish which before selecting the instrument
5. Is the question "is this conductor live?"
Yes → Use a two-pole indicator with a proving unit, or low-impedance mode — not a plain high-impedance multimeter
No → Continue
6. Does a voltage reading collapse under low-impedance mode?
Yes → It was a ghost voltage — induced, not a supply
No → Treat it as a genuine live supply
7. Does the circuit contain electronics and are you about to insulation-test it?
Yes → STOP. Disconnect the electronics first or you will destroy them.
No → Proceed, and discharge afterwards
07Step-by-step diagnosis
Step 1Select the instrument for the installation, not just the voltage
- Inspect
- Measurement category and voltage rating on the meter AND on the leads
- Where
- Printed on the instrument and lead insulation
- Instrument
- Visual inspection
- Expected result
- Both suited to the position in the installation being tested
- If the result is abnormal
- Category describes the transient energy the instrument can survive, not its accuracy. Leads are rated separately and are frequently the weak link in an otherwise adequate set-up.
- Next
- Step 2
Safety: An under-rated instrument works perfectly right up until a transient arrives, and then fails in your hand.
Step 2Inspect leads, probes and fuse before use
- Inspect
- Lead insulation, probe tips, shrouding, finger guards and the current-range fuse
- Where
- The whole lead set and the meter inputs
- Instrument
- Visual inspection and a continuity check of the leads themselves
- Expected result
- Undamaged leads, intact fuse
- If the result is abnormal
- A blown current-range fuse reads zero current, which looks exactly like a dead circuit. Internal lead damage beneath intact insulation gives intermittent readings and is invisible.
- Next
- Step 3
Step 3Match the instrument to the waveform
- Inspect
- Whether the measurement is downstream of power electronics or on a distorted supply
- Where
- The circuit being measured
- Instrument
- True-RMS instrument where distortion is present
- Expected result
- True-RMS used for anything after an inverter, UPS or drive
- If the result is abnormal
- An averaging meter is calibrated for sine waves only. On a distorted waveform it reads incorrectly and displays a completely believable number with no indication that anything is wrong, which is why this error survives so long undetected.
- Next
- Step 4
Step 4Match the instrument to AC or DC
- Inspect
- Whether the circuit is AC or DC, and whether the clamp is AC-only or Hall-effect
- Where
- The circuit and the instrument specification
- Instrument
- AC/DC clamp for battery, DC bus and PV measurements
- Expected result
- Correct instrument type for the quantity
- If the result is abnormal
- An AC clamp on a battery circuit reads essentially nothing, which is easily and confidently misread as "no current flowing".
- Next
- Step 5
Step 5Distinguish a ghost voltage from a real supply
- Inspect
- Whether a voltage on a supposedly dead conductor persists under load
- Where
- At the conductor in question
- Instrument
- Low-impedance mode, or a two-pole voltage indicator
- Expected result
- A genuine supply holds; an induced voltage collapses
- If the result is abnormal
- High-impedance meters read capacitively coupled voltage on long disconnected cables running alongside live conductors. It is real but can source almost no current. Never conclude a circuit is live or dead on a high-impedance reading alone.
- Next
- Step 6
Safety: Do not let ghost voltages train you to disbelieve readings. Resolve them properly — the habit of dismissing an unexpected voltage is how people get hurt.
Step 6Use the correct instruments for proving dead
- Inspect
- Whether a two-pole indicator and proving unit are being used
- Where
- At the point of work
- Instrument
- Two-pole voltage indicator and proving unit
- Expected result
- Indicator proved before and after the dead test
- If the result is abnormal
- A multimeter cannot be proved in the same way, and a failed multimeter reads exactly like a dead circuit. This is a procedure point, not a preference.
- Next
- Step 7
Step 7Protect electronics before insulation testing
- Inspect
- Whether the circuit contains drives, controllers, inverters or other electronics
- Where
- The circuit under test
- Instrument
- Insulation tester, after disconnection
- Expected result
- Electronics disconnected before the test voltage is applied
- If the result is abnormal
- An insulation tester applies a high voltage that destroys electronic equipment. This is a common and expensive mistake, particularly when testing motor cables that remain connected to a drive.
- Next
- Step 8
Safety: Discharge the winding or cable after testing — the tester leaves it charged.
Step 8Interpret the reading in context, not in isolation
- Inspect
- Temperature, absorption behaviour, emissivity and history as they apply to the measurement
- Where
- Alongside the raw reading
- Instrument
- Recorded conditions and prior baselines
- Expected result
- A reading interpreted against conditions and history
- If the result is abnormal
- Insulation resistance varies substantially with temperature and rises during a test as absorption occurs — neither is a fault. A thermal camera with the wrong emissivity setting gives a confidently wrong temperature. Context is what turns a number into evidence.
- Next
- Record the reading WITH its conditions so it means something to the next person
08Repair procedure
Instrument condition
component replacement- Replace damaged leads rather than repairing them; a repaired lead is not rated
- Replace meter fuses with the correct type and rating only — a substitute fuse defeats the instrument protection
- Take out of service any instrument subjected to an over-range event, even if it still appears to work
Instrument selection and setup
configuration- Select measurement category and voltage rating for the installation position, checking leads as well as meter
- Select insulation test voltage appropriate to the equipment under test
- Set thermal camera emissivity to suit the surface being measured
- Use low-impedance mode where the question is whether a conductor is genuinely live
Calibration and competence
manufacturer level- Maintain instruments on a calibration schedule and keep the records
- Check instruments against a known reference between calibrations where the work justifies it
- Where an instrument gives readings you cannot reconcile, verify against a second instrument before trusting either
09Post-repair validation
- Confirm the instrument category and voltage rating suit the installation, meter and leads
- Confirm a true-RMS instrument was used wherever the waveform may be distorted
- Confirm any unexpected voltage was resolved as genuine or induced, not simply dismissed
- Confirm electronics were disconnected before any insulation test, and the circuit discharged afterwards
- Cross-check a critical reading against a second instrument before acting on it
- Record readings WITH the conditions they were taken in — irradiance, temperature, load, so they remain meaningful
- Confirm instruments are within calibration and leads undamaged at the end of the work as well as the start
10When not to repair
- Instruments outside their calibration period where the result matters — the reading cannot be relied on
- Damaged leads, which must be replaced rather than repaired
- Instruments subjected to an over-range event, which may have degraded protection that is not visible
- Any situation where the correct-category instrument is unavailable — the work should wait rather than proceed under-rated
11Prevention
- Buy instruments rated for the highest category and voltage you actually work on, and check the leads carry the same rating
- Carry a proving unit, and treat proving dead as a procedure rather than a habit
- Default to true-RMS instruments; the situations where averaging is adequate are shrinking as power electronics spread
- Inspect leads before every use — this takes seconds and catches the failure mode that is otherwise invisible
- Keep instruments on a calibration schedule and retain the records
- Record measurement conditions alongside readings, so a future comparison is possible
- Where a reading surprises you, resolve it rather than dismissing it — the habit of dismissing anomalies is itself the hazard
12Questions engineers actually ask
Why does my meter show voltage on a cable I have disconnected?
Almost certainly a ghost voltage. A modern meter has very high input impedance, and a long disconnected conductor running alongside live cables picks up a capacitively coupled voltage that the meter faithfully displays. It is real but can source almost no current. Switch to low-impedance mode, or use a two-pole voltage indicator: an induced voltage collapses, a genuine supply does not. What you must not do is get into the habit of dismissing unexpected voltages.
Does it matter if my meter is true-RMS?
Downstream of any inverter, UPS or variable-speed drive, or on a generator feeding significant non-linear load, yes — an averaging meter is calibrated for sine waves and reads distorted waveforms incorrectly while displaying an entirely believable number. There is no warning on the screen, which is why this error persists undetected and produces confident wrong diagnoses.
What is a measurement category and does it really matter?
It is a safety rating, not an accuracy one. It describes the transient overvoltage energy the instrument is constructed to survive at its position in an installation, so an under-rated meter measures perfectly right up until a transient arrives, and then fails in your hand. Check the rating on the leads as well as the meter, because leads are rated separately and are often the weakest part of an otherwise adequate set-up.
Can I megger a motor cable without disconnecting the drive?
No. An insulation tester applies a high voltage that will destroy the drive output stage, and this is a common and expensive mistake. Disconnect the drive and test the motor and cable on their own. Discharge the cable and winding afterwards, because the tester leaves them charged — that stored charge has injured people who assumed the test was over.
Standards and references
- IEC 61010-1 and IEC 61010-2-030 — safety requirements for electrical measurement equipment, which define the measurement categories referred to throughout
- IEC 61010-031 — safety requirements for hand-held probe assemblies, which rate the leads separately from the instrument
- IEEE 43 — recommended practice for testing insulation resistance of electric machinery, including temperature correction and absorption behaviour
- IEC 60364-6 — low-voltage installations: verification
- The instrument manufacturer's documentation for the specific meter and leads, which states the category, voltage rating, fuse specification and calibration interval that apply
This guidance is written from engineering principle and is not a substitute for the manufacturer's model-specific documentation. Where a figure is model-specific, confirm it against the service data for your unit before acting on it.
Need this diagnosed properly?
Send us the make, model and any fault codes shown, and photographs of the controller display if you have them. Our mobile workshop covers all 47 counties.