Electrical origin is among the most frequently asserted and least frequently proven conclusions in fire investigation. The assertion is easy because electrical equipment is everywhere and is almost always damaged. The proof is hard because the damage is usually a consequence of the fire rather than a cause of it.
Two different mechanisms
Arcing and overheating are distinct phenomena with distinct signatures, and conflating them is a common source of error. An arc is a luminous discharge across a gap, intense and localised, capable of igniting adjacent combustibles almost instantaneously. Overheating is a resistive process: a poor connection, an undersized conductor, or a loaded circuit dissipating more heat than the installation can shed, producing elevated temperatures over minutes to hours.
The two leave different physical records and imply different failure narratives. Overheating at a termination suggests a workmanship or maintenance issue with a long incubation; an arc suggests an insulation failure or mechanical damage event. Investigations that identify melted copper and stop there have not yet distinguished between them.
What fire patterns actually indicate
Fire pattern analysis, as set out in NFPA 921, is a legitimate and necessary tool. It is also routinely over-read. Patterns record where the fire burned longest and hottest with the ventilation available, which correlates with origin only when ventilation was relatively uniform and the fuel load relatively even.
Neither condition is common. Ventilation-controlled compartment fires generate intense localised damage at openings and at the point where fresh air entered, which can produce a convincing pattern with no relationship to origin. Post-flashover conditions degrade pattern reliability substantially. An origin hypothesis resting on pattern alone, in a compartment that reached flashover, is weakly supported regardless of how clear the pattern appears.
Arc mapping and its assumptions
Arc mapping — surveying a circuit for arc damage and using the distribution to constrain where fire first attacked energised conductors — is one of the more useful physical techniques available. Its value is that it depends on the electrical system's behaviour rather than on thermal pattern interpretation.
It carries assumptions worth stating. The circuit must have been energised, the protective device must not have operated so early as to de-energise the run before meaningful damage, and the survey must be reasonably complete. Partial surveys produce partial maps that can point confidently in the wrong direction. Where a panel schedule is unavailable or the wiring was modified, the mapping is only as good as the circuit reconstruction underlying it.
Cause versus victim
The central discipline in electrical fire work is distinguishing arcing that caused a fire from arcing that resulted from one. Fire attacking a cable's insulation will produce arcing between conductors; the resulting bead is physically similar to one produced by a pre-existing fault. Metallurgical examination of arc beads has been studied extensively and can be informative, but the literature does not support treating bead morphology alone as decisive.
This is where corroboration matters. An electrical origin conclusion is materially stronger when the electrical evidence, the pattern evidence, the fuel geometry and the witness or alarm timeline agree, and materially weaker when it rests on any one of them.
Negative corpus and its proper use
NFPA 921 addresses the practice of concluding an electrical cause by eliminating other hypotheses without affirmative evidence for the electrical one. Used carelessly this becomes reasoning from ignorance, and it is a frequent basis for challenge.
The distinction that matters is between eliminating alternatives to support a hypothesis that also has affirmative physical support, and eliminating alternatives in place of such support. The first is ordinary scientific method. The second tends not to survive examination, particularly where the scene was incompletely documented.
What to preserve
Electrical evidence degrades in specific and avoidable ways. Conductors should be recovered with their spatial relationships recorded rather than pulled free; overcurrent protective devices should be preserved without being operated, since their as-found position is data; and appliances or equipment suspected of involvement should be removed intact rather than disassembled on scene.
Radiographic examination before disassembly is frequently worthwhile for enclosed devices, because the internal arrangement is itself evidence and is easily disturbed. Once a device has been opened outside a documented protocol, its evidentiary value is difficult to restore.
Independent timelines
Physical evidence is considerably stronger when corroborated by data generated independently of the investigation. Alarm and detection system logs, building management records, utility interval metering, and increasingly the telemetry of connected equipment can establish when a circuit de-energised, when detection first activated, and in what order compartments were affected.
This data is perishable in its own way. Panels are reset, systems are replaced during restoration, and retention periods on utility and monitoring data are often short. Requesting preservation early costs little and occasionally settles a question that the physical evidence alone would leave open.
Where these determinations are challenged
Challenges to electrical origin conclusions cluster around a few themes: that the pattern analysis was performed on a post-flashover compartment where patterns are unreliable, that arc mapping was based on an incomplete circuit reconstruction, that arcing was a consequence rather than a cause, and that alternative hypotheses were listed rather than genuinely tested.
Anticipating those challenges tends to improve the underlying work. An examination that documents why each alternative was excluded, and on what physical basis, is more useful than one that reports only the conclusion reached.
This article is general technical orientation, not a failure analysis, an engineering opinion, or advice on any specific matter. Determining the cause of a particular incident requires hands-on examination by a credentialed expert.