Use a systematic six-step electrical fault finding process: safe isolation, gather evidence, dead tests, live tests only where justified, diagnose and repair, then verify and document. Every step follows BS 7671 sequencing and the Electricity at Work Regulations 1989. Instrument detail on insulation resistance, continuity, Zs and RCD testing follows below.

TL;DR:

Proper fault finding requires following a six-step process: safe isolation, evidence gathering, dead testing, justified live testing, diagnosis and repair, then verification.

Safe isolation involves proving the circuit is dead with a GS38-compliant voltage indicator and locking off before testing or repairs.

Insulation resistance tests are performed at 500 V DC with a minimum of 1 MΩ for a healthy circuit, using lower voltage only for sensitive electronics.

The three main diagnosis techniques are half-split testing for long circuits, elimination for specific appliances, and experience-based judgment based on pattern recognition.

Documentation, photography, and recording test results on site are crucial for compliance and protecting the engineer in case of disputes.

Delta First

Keep Commercial Electrics Operating Safely

Delta First provides electrical maintenance, compliance and installation support for commercial buildings across Essex, Suffolk and Greater London.

Visit Delta First

What is the correct electrical fault finding process?

The electrical fault finding process is not a single test. It is a sequence, and the order matters as much as the readings themselves. Skip a step and you risk a false result, or worse, a live conductor you believed was dead.

Our engineers work through six stages on every commercial call-out: safe isolation, evidence gathering, dead testing, justified live testing, diagnosis and repair, then verification. The sequence exists because each stage protects the one after it. Proving dead before you touch a terminal protects the technician. Gathering evidence before you test protects your time. Testing in the correct order, as set out in BS 7671, protects the validity of every reading that follows. Get this order wrong and a continuity fault can masquerade as an insulation fault, sending you down the wrong branch of the circuit entirely.

This guide walks through each stage in the depth a working electrician needs, from the prove-dead sequence to ring final circuit interpretation and EICR coding.

Safety and legal duties before fault finding

Fault finding almost always starts with a circuit that might be live, might be faulty, and might be both. The Electricity at Work Regulations 1989 place a legal duty on employers and self-employed engineers to ensure work is carried out in a way that prevents danger, and safe isolation is where that duty starts. BS 7671 expects the same discipline: prove the circuit dead before any hands-on diagnostic work begins.

The prove-dead sequence has a fixed order:

Identify the correct circuit at the distribution board and isolate it.
Prove your voltage indicator on a known live source, ideally a proving unit, before and after testing.
Test the isolated circuit is dead using a GS38-compliant approved voltage indicator, checking line to neutral, line to earth and neutral to earth.
Lock off the isolation point and attach a warning label naming the engineer and the reason for isolation.

A risk assessment should confirm whether a second competent person needs to be present, particularly on three-phase distribution boards or where isolation cannot be visually confirmed from the working position. Appropriate PPE, insulated tools and a documented method statement round out the legal picture; a written justification is what separates a defensible live test from a breach of Regulation 14.

Pro Tip: Keep a spare proving unit in the van. A dead battery in your only proving unit at 8am on a hospital ward is not a problem you want to solve on the spot.

Toolkit: essential instruments for the fault finding process

A multifunction tester (MFT) covers most of the dead and live test sequence: continuity, insulation resistance, loop impedance and RCD trip timing all sit within one instrument on a modern unit. Beyond the MFT, a competent toolkit typically includes:

A dedicated insulation resistance tester, useful when the MFT is tied up elsewhere on a larger job.
A clamp meter for current readings without breaking the circuit, particularly useful on suspect loads.
A standalone RCD tester where independent verification of trip times is required.
A GS38-compliant approved voltage indicator (AVI) for safe isolation, kept separate from your general-purpose meter.
A thermal imaging camera for identifying overheating connections and unbalanced loads under normal operation.
Non-contact voltage detectors, useful as a first screening tool but never a substitute for an approved voltage indicator.

Insulation resistance testing normally runs at 500 V DC, with 250 V permitted where sensitive electronics, dimmers or data equipment are on the circuit, since the lower voltage protects components while still giving a usable reading. Every tester needs proving on a known live source before use, and probes, leads and calibration dates deserve a quick visual check before you rely on a reading in a report.

Gather evidence and initial assessment: what to ask and record

Professional technicians know that fault finding is not guesswork. Gathering the who, where, when and why often narrows the probable cause of electrical problems before an instrument leaves the case, cutting both diagnostic time and unnecessary live testing.

Before touching a distribution board, work through these questions with whoever reported the fault:

Who first noticed the fault, and what were they doing at the time?
Does it happen at a specific time of day, under a particular load, or with certain equipment running?
Is the fault permanent or intermittent, and has anything changed recently, such as new equipment, a refurbishment or recent electrical works?
What does the previous EICR, Minor Works certificate or asset register show for that circuit?

An intermittent fault that will not reproduce on demand should be recorded as an FI observation rather than guessed at. Plan a return visit with a data logger fitted to capture voltage dips or current spikes over a working day. Predictive approaches that log conditions over time, similar to those described in predictive maintenance planning, can flag the pattern before it becomes a full failure.

Plan your approach: choosing a fault-finding method

Once the evidence points somewhere, choose the locating technique that fits the fault, rather than defaulting to the same method every time.

The six-step sequence our engineers follow in practice is: safe isolation, evidence gathering, dead tests, justified live tests, diagnosis and repair, then verification. It exists because skipping a stage tends to cost more time than it saves.

Within diagnosis, three techniques cover almost every scenario:

Half-split: test the midpoint of a circuit run and use the result to discard half the remaining search area each time, ideal for long radial runs or ring finals with no obvious symptom location.
Elimination: disconnect suspect appliances or sections one at a time and retest, best suited to fault symptoms tied to specific loads rather than the fixed wiring.
Experience-based: jump straight to the most probable cause based on symptom pattern and installation age, useful on familiar building types but only with a clear stop condition if the hunch proves wrong.

Pro Tip: Set yourself a time limit before switching from experience-based guessing to a proper half-split search. Twenty minutes chasing a hunch is reasonable; an hour is a sign you should have started splitting the circuit from the outset.

Visual inspection and quick screening checks

A good visual inspection catches faults that no instrument will find faster. Before any test leads come out, walk the installation and look for the obvious.

Check the consumer unit for scorched terminals, loose glands or signs of a previous repair done in a hurry. Inspect cables for damaged insulation, chafing at clips, and rodent damage in ceiling voids. Look at accessories for cracked faceplates, discoloured sockets or switches, and any sign of water ingress around outdoor fittings or roof penetrations. Overheating leaves its own signature: browning on a faceplate, a faint burning smell, or a terminal that feels warm to the touch after normal use.

Quick screening steps often resolve simple faults without further testing:

Isolate any appliance under suspicion and see if the symptom disappears.
Reset tripped protective devices once and note whether they hold or trip again immediately.
Swap a suspect load with a known-good one of similar rating to confirm whether the fault travels with the appliance.
Photograph and note every visual finding before moving to dead testing.

Where a visual defect presents an immediate danger, isolate the circuit and make it safe as a temporary measure before continuing the fault finding process.

Dead tests sequence: continuity, ring finals and polarity

BS 7671 sets a fixed order for dead testing, and the reason is practical: each test relies on the validity of the one before it. Testing insulation resistance before confirming continuity, for instance, risks a misleading result on a circuit with a broken conductor.

The dead test sequence runs:

Continuity of protective conductors, confirming the circuit protective conductor (CPC) is intact end to end.
Continuity of ring final circuits, measuring end-to-end resistance of line, neutral and CPC (r1, rn, r2), then cross-connecting to calculate R1+R2 at each socket.
Insulation resistance, covered in detail below.
Polarity, confirming line, neutral and CPC are correctly connected at every point, including switches and socket outlets.

On a ring final circuit, an elevated r1+r2 reading at one or two sockets compared with the rest of the ring usually points to a poor connection, a high-resistance joint, or a break in one leg of the ring that has been wired as a spur without anyone realising. A reversed polarity result on a socket or lighting point is a C1 or C2 observation depending on the risk it presents, and it demands immediate correction before the circuit is left energised again. Record every reading against its location, not just against the circuit number, so the next engineer can see exactly where the anomaly sat.

Statistic callout: Insulation resistance testing at 500 V DC with a 1 MΩ practical minimum, and continuity testing before it, forms the backbone of the BS 7671 dead test sequence used across UK commercial and domestic installations alike.

Insulation resistance testing: voltages and thresholds

Insulation resistance (IR) testing checks whether the insulation between conductors, and between conductors and earth, is still doing its job. BS 7671 sets 500 V DC as the standard test voltage, with 1 MΩ treated as a practical minimum for a healthy result.

A reading below 0.5 MΩ signals a serious defect that needs locating before the circuit goes back into service. Where a circuit feeds sensitive electronic equipment, such as building management systems, dimmer racks or IT infrastructure, drop the test voltage to 250 V rather than risk damaging the equipment, accepting that the lower voltage is slightly less sensitive to some fault types.

When a low IR reading comes back on a distribution circuit feeding multiple points, isolate systematically:

Disconnect sockets, luminaires or appliances one section at a time.
Retest after each disconnection to see where the reading recovers.
Note the location and reading at every step, not just the final result.

Compare every figure against the previous EICR or installation certificate. A steady decline in IR over successive tests on the same circuit often flags ageing cable insulation long before it fails outright.

When and how to re-energise: justified live testing

Some readings simply cannot be taken dead, and BS 7671 accepts that live testing is sometimes necessary. Current guidance, though, pushes the other way: minimise live testing wherever a dead test can verify the same thing, and document the Regulation 14 justification whenever a live test does go ahead.

Three live tests come up repeatedly in fault finding:

Earth fault loop impedance (Zs), measured live at the point of use because it depends on the actual supply characteristics, not just the wiring layout. Compare against Ze plus the calculated R1+R2, applying temperature correction where the standard tables require it.
Prospective fault current (PFC), checked to confirm the installed protective devices can safely interrupt the fault current available at that point. A higher than expected PFC may mean a device needs upgrading before the circuit is signed off.
RCD testing, using a 50% test to confirm the device does not nuisance-trip, then a test at 1× IΔn with a maximum 300 ms trip time as the pass criterion, with a 5× test available as an optional overload check. Record every trip time against the device rating, not just a pass or fail.

Delta First’s approach to RCD testing under BS 7671 reflects this same discipline: justify the live test in writing, keep it as brief as the reading requires, and never treat a live test as the default option when a dead test would do.

Pinpointing and repair: half-split, substitution and ring finals

Locating the fault is where the earlier planning pays off. On a long radial circuit, half-split works cleanly: test the midpoint junction box, and depending on the result, discard either the upstream or downstream half of the run. Repeat at the midpoint of whichever half remains faulty, and each pass roughly halves the search area.

For appliance faults, substitution settles the question quickly:

Disconnect the suspect appliance and retest the circuit in isolation.
If the fault clears, swap in a known-good appliance of the same type and rating.
If the fault returns only with the original appliance connected, the appliance itself is the problem, not the fixed wiring.

Ring final circuits need a slightly different approach. Mid-point testing at a socket roughly halfway around the ring narrows the search quickly, and correctly identifying the CPC at each terminal matters more than on a radial circuit, since a mis-identified CPC on a ring can mask a break entirely. Once the fault is isolated, re-terminate to the correct torque setting and confirm the connection visually before closing up.

Verify, document and report after fault finding

A repair is not finished until it is proven. Repeat the relevant dead tests, continuity, insulation resistance and polarity, then retest Zs and RCD trip times live, and run a functional check under normal load before calling the job complete.

Any remaining observation gets coded to the EICR standard: C1 for danger present, needing immediate action; C2 for a potentially dangerous defect requiring urgent remedial work; C3 for improvement recommended but not required; and FI where further investigation is needed without delay. A C1 finding demands verbal notification to whoever ordered the work before you leave site, followed up in writing.

Repeat continuity, IR, Zs and RCD tests after repair, under normal load where practical.
Code any residual observations correctly and flag C1/C2 items immediately, not in a follow-up email days later.
Provide digital test records, a remedial quote where needed, and a Minor Works certificate for the completed repair.

Digital testing software that records results on site and checks them against tabulated maxima removes a lot of transcription error from this final step, and gives the client a cleaner audit trail than a handwritten certificate.

Delta First practical notes and field tips from our engineers

Our engineers run a consistent on-site checklist regardless of the building type: photograph the fault location and the isolation point, keep the MFT printout or digital export against the job reference, and confirm the lock-off is photographed before work starts. It sounds procedural, but it is what protects both the engineer and the client if a finding is ever queried later.

Commercial premises tend to throw up the same handful of faults on repeat: high-resistance joints in older distribution boards, ring finals wired as unintentional spurs during past alterations, and RCD nuisance tripping caused by accumulated leakage current across multiple circuits rather than one clear fault. Reporting these findings promptly, with a clear remedial priority, tends to prevent a repeat call-out within the following quarter.

Photograph the fault and the isolation point on every visit.
Retain MFT printouts or digital exports against the job reference.
Recommend a PPM inspection interval where repeat faults suggest ageing infrastructure rather than a one-off defect.

Pro Tip: If the same circuit generates two call-outs inside twelve months, treat that as a signal to recommend a full EICR rather than another isolated repair.

Why method beats guesswork

The engineers who diagnose fastest are rarely the ones who jump straight to a meter. They are the ones who ask the right questions first, then let the test sequence do the work it is designed to do.

A disciplined process pays for itself the first time a “random” intermittent fault turns out to follow a load pattern nobody mentioned until you asked directly. Skipping evidence gathering in favour of an immediate live test on a ring final once cost a colleague a wasted afternoon chasing a symptom that a five-minute conversation with the site manager would have explained. Documentation matters just as much as the diagnosis itself: a clear record of what was tested, in what order, and why, protects the engineer if a finding is ever disputed, and gives the client confidence the job was done properly rather than patched.

Stick to the standards. They exist because someone, somewhere, learned the hard way what happens when a step gets skipped.

— Ashley

Need help with fault finding, maintenance or compliance?

Chasing an intermittent fault across a live commercial building without the right instruments, standing team or spare hours is a different proposition to a controlled workshop repair. Delta First gives facilities managers and estates teams one accountable contractor for electrical fault finding, remedial EICR work, and the planned preventive maintenance that stops the same fault recurring every few months.

Regionally deployed engineers across several areas mean a faster response when a fault presents itself, and full compliance documentation on every visit, including test results, EICR coding and Minor Works certificates. Our single-point-of-contact model, built around a directly employed core team, with specialist subcontractors managed by us where needed, keeps quality and compliance consistent across every visit. Whether the immediate need is a one-off diagnostic call-out, commercial EICR testing with remedial works to follow, or a rolling maintenance contract that catches faults before they cause downtime, request a quotation or a planned maintenance survey from Delta First and get a straight answer on scope and timescale.

Need help with your electrical installation or compliance? Delta First’s directly employed electricians provide testing, maintenance, installations and upgrades for commercial and residential properties across Essex, Suffolk and London, with 24/7 emergency cover. Call 01206 237422 or request a quote.

FAQ

What are the five steps to fault finding?

Most guidance actually describes a six-step sequence: safe isolation, gather evidence, dead tests, justified live tests, diagnose and repair, then verify and document. Some sources compress this to five by merging diagnosis and repair into one stage, but the underlying BS 7671 test order stays the same either way.

How much does an electrician charge to fault find?

Fault-finding charges vary widely depending on the complexity of the installation, access, and whether the fault reproduces on the first visit. Delta First does not publish a fixed fault-finding rate; current pricing is available on request through a quotation request.

What is the 6-point technique in fault finding?

The six-point technique refers to the systematic sequence: safe isolation, evidence gathering, dead testing, live testing where justified, diagnosis and repair, and final verification. It is designed so each stage protects the validity of the one that follows it.

What are the methods of fault finding?

The three main locating techniques are half-split, which narrows a search area by testing the midpoint of a circuit; elimination, which isolates suspect appliances or sections one at a time; and experience-based diagnosis, which jumps to the likely cause based on symptom pattern. Each suits different fault types, and competent technicians switch between them depending on what the evidence points to.

Do I need to isolate a circuit before fault finding?

Yes. Safe isolation under the Electricity at Work Regulations 1989 is a legal requirement before hands-on dead testing, and it must include proving the circuit dead with a GS38-compliant voltage indicator, locking off, and labelling the isolation point.

Talk to Delta First

Delta First Building Services is a Colchester-based M&E contractor working across Essex, Suffolk and London, with commercial work in Norfolk. Our directly employed electrical team is supported by vetted specialist subcontractors where extra certification is needed, and we offer genuine 24/7 emergency cover. Get in touch to arrange a survey or quote.

6 step electrical fault finding to BS 7671 for UK technicians