Continuity Good But Circuit Dead Under Load: A Misleading-Reading Decision Tree

Why this matters

A continuity test beeps clean end to end, the ohmmeter shows a fraction of an ohm, and the tech declares the conductor good. Then the circuit dies the moment a load is applied. The continuity reading was not wrong about the wire being connected. It was wrong about the wire being able to carry current. A path that conducts a few milliamps of test current can still choke a 15 A load if there is a marginal connection in series. Knowing why a passing continuity test misleads keeps techs from condemning good devices and chasing ghosts.

Why the good reading misleads

A continuity or low-ohms test injects a tiny known current, often well under a milliamp, and measures the voltage it produces. Through a marginal connection, that microamp-level current sees almost no resistance because resistance at a loose or oxidized joint is highly current-dependent. A joint that reads 0.5 ohm on the meter can behave like 5 ohms once 15 A flows, because the contact area is too small to carry real current and the metal heats, oxidizes further, and resistance climbs.

This is the no-load versus loaded distinction again, expressed in continuity terms. The test proved a single-point connection exists; it did not prove the path can deliver working current. The only honest measurement of a conductor's current-carrying ability is a voltage-drop test taken while the circuit carries its actual load.

Symptom presentation

• Continuity and resistance tests pass on every conductor, yet the circuit will not run anything.
• Voltage present at the panel and even at the device with nothing drawing current.
• Voltage collapses the instant any load engages; the device clicks off or never starts.
• A breaker that does not trip, because the fault is high series resistance, not a short.
• Localized heat at a splice, wire nut, terminal, or device after the load has tried to run.

Quick checks

• De-energize and confirm the conductor is connected at both ends with a low-ohms reading. That tells you it is continuous, nothing more.
• Re-energize under lockout-aware conditions and measure voltage at the device with load off, then under load. A passing continuity test plus a large loaded drop is the signature of a high-resistance series fault.
• Clamp the conductor under load. If current flows but voltage at the device is low, the drop is in the path. If no current flows under a valid call, the path opened under load (a connection that separates when heated or vibrated).

Isolation tree

1. Continuity good, voltage present at rest, voltage collapses under load. Series high-resistance fault. Apply the load and walk the voltage probe upstream: device, last box, junction, breaker lug, breaker line. Voltage is correct above the fault and low below it.

2. Continuity good, voltage present, breaker trips under load. Different fault class. This is a low-impedance fault appearing only when the device's own current path closes; suspect a damaged conductor shorting under load expansion or a failing device drawing locked-rotor current. Megger the conductor to ground with the device disconnected.

3. Continuity good on the equipment ground but device case is dead-shorting protection. The EGC continuity proves the wire path, not the bonding integrity under fault current. A bootleg or undersized ground passes continuity yet cannot clear a fault. Verify ground-fault path impedance, not just continuity.

4. Continuity good, no voltage under load, no current. The path opened when loaded. A connection that is mechanically continuous cold but separates under thermal expansion. Heat the suspect joint with the load and watch the voltage; an opening joint is the fault.

Confirming diagnosis

De-energize and open the connection between the last good and first bad voltage reading. The confirming evidence is a contact that is mechanically loose, oxidized, undersized for the wire, or showing heat tint. A backstab on a 20 A circuit, a wire nut with one strand engaged, or a terminal screw that backed off are the usual culprits. Re-terminate, repeat the loaded voltage-drop test, and the drop disappears while continuity remains good. The two readings now agree because the path can carry current.

Confirming a damaged conductor

If every termination is sound but loaded voltage still sags, suspect conductor damage between boxes, a nicked strand bundle, or a partially severed run behind drywall. An insulation-resistance test will not catch a series-resistance defect; only the loaded voltage-drop comparison against expected drop for the length and gauge will isolate it.

Remediation

• Re-terminate or replace the failed connection; convert backstabs to screw or pigtail terminations.
• Torque all reworked terminations per the device or lug marking and NEC 110.14(D).
• Replace heat-damaged conductors rather than re-splicing in the damaged zone.
• Re-verify under full load and confirm drop is within NEC 210.19 informational-note guidance (3 percent branch, 5 percent total).

References

• NFPA 70, National Electrical Code, Article 110.14, Electrical Connections (termination integrity and torque).
• NFPA 70, National Electrical Code, Article 250.4, Performance of grounding and bonding (fault-current path requirements).
• NFPA 70E, Standard for Electrical Safety in the Workplace (energized-work justification and PPE).
• UL 486A-486B, Wire Connectors (current-cycling and connection-resistance behavior).