ABB ACS800 SHORT CIRC 2340: External Load vs Power-Stage Investigation

Why a pre-power workflow is essential

A SHORT CIRC event may leave damage in more than the part that visibly failed. A shorted motor cable may have damaged a previously healthy bridge; a failed bridge may have damaged the driver; an unsuitable output-side capacitor or surge device may make replacement hardware fail again. The repair must therefore identify the initiating cause and the consequential damage boundary.

The ACS800 provides phase-leg and inverter-module localisation in applicable configurations. Use that information to target inspection, but do not let localisation replace the broader cause review.

Phase 1 — classify the event

Establish whether the failure happened during acceleration, steady operation, a restart after repair, or immediately on enable. Record connected motor and cable condition, output accessories, recent mechanical events and whether a prior repair already changed an IGBT or driver assembly.

Phase 2 — verify external and internal boundaries

The motor and motor cable must be treated as potential causes, not merely loads. Follow qualified test procedures to evaluate phase-to-phase and phase-to-earth integrity before exposing a repaired converter to them. Confirm the output circuit does not include power-factor correction capacitors or surge absorbers incompatible with inverter output.

For internal hardware, tie any phase-leg localisation to the affected driver and switching section. Inspect the driver/module assembly, associated interconnection, thermal contact and signs of explosive or carbonised damage. Replacement of a semiconductor assembly without review of its driver boundary is not a complete repair decision.

Phase 3 — controlled commissioning gate

A drive is ready for a controlled test only when the external initiating causes have been cleared or disproved, internal replacement hardware is exactly identified, the associated driver/protection boundary has been reviewed and the test is planned with suitable containment and measurement. A successful no-load or staged test is evidence to record, not permission to omit production-load validation.

Stop and escalate when localisation is inconclusive, replacement revisions are uncertain, the drive has suffered repeated failures or testing resources are insufficient for the energy rating of the unit.

Commissioning record and risk release

The controlled return-to-power stage should be recorded with the same discipline as diagnosis. Document the replacement assembly, external load condition, protective test arrangement, initial energisation observation, first enable result and any subsequent load validation. A repair that survives one energisation but fails during acceleration has not passed the same test as a drive verified through its intended operating condition.

Where the application is production-critical, recurrent failure data should feed an asset decision: maintain stocked proven spares, engage a qualified repair service with load-testing capability, or evaluate replacement/modernization. The repair workflow is therefore also the first layer of an industrial asset-management dataset.