How to Test an Automatic Transfer Switch Safely: A Qualified-Person Workflow
A useful ATS test starts by defining what must be proved. It then controls both sources, protects the load, follows the exact manufacturer and site method, records the full transfer-and-return sequence, and ends only after automatic readiness is verified.
Test the complete sequence—not just the generator
First decide whether you need evidence of source starting, automatic transfer, real-load support, retransfer, alarm reporting, or coordinated system operation. Then use the exact ATS model manual, approved drawings, site operating procedure, risk controls, and acceptance criteria.
Qualified personnel execute the approved manufacturer/site method. Observers compare the real source, switch, load, alarm, and timing sequence with the expected sequence. No one improvises by opening a breaker, removing a fuse, bridging terminals, defeating an interlock, or changing settings.
A passing result also includes restoration. The team must confirm the intended source, ATS position, automatic mode, remote status, alarms, temporary control states, and every affected load before the job is closed.
Seven levels that are not interchangeable
Credit only the evidence that the approved topology and test scope actually produced. A more dramatic test is not automatically a better test.
Closed-door baseline
Review nameplate, normal indicators, event logs, prior defects, restrictions, and documented maintenance status without removing barriers.
Does not prove that the alternate source will start or that the ATS will move.No-transfer exercise
Shows that the alternate source starts and reaches the controller's configured acceptance state during this test.
Does not prove ATS movement, contact path, or facility-load support.Transfer and retransfer
Shows the configured sequence actually observed: source call, acceptance, transfer, return logic, indications, and alarms.
The report must state whether the connected load moved and which loads were included.Live-load transfer
Transfers a defined facility load through the ATS to the alternate source and back. It may interrupt service.
Proves the ATS route plus only the downstream loads included in the approved plan.Generator capability
Applies controlled test load to the generator. The connection diagram decides whether current passes through the ATS.
If the load bank bypasses the ATS, it does not prove the ATS main path or facility loads.Coordinated system behavior
Exercises the approved interaction of source, ATS, protection, alarms, remote monitoring, selected loads, and recovery roles.
Requires a multidisciplinary plan, witnesses, communications, and clear abort authority.Offline maintenance and diagnostics
Contact resistance, insulation resistance, internal mechanical service, dielectric work, and similar tests require controlled isolation and exact OEM limits.
This page gives no instrument settings, test voltages, connection points, jumpers, or internal service sequence.Transition type changes the risk plan
Confirm the installed switching technology, transition method, bypass arrangement, normal and alternate sources, and critical-load path before writing the test.
| System feature | What it means during testing | Planning question | Important boundary |
|---|---|---|---|
| Open transition | One source is disconnected before the other source is connected. | Which loads can tolerate the expected interruption and restart behavior? | LOAD IMPACT |
| Delayed transition | An intentional off or neutral interval may be part of the approved sequence. | Do motors, drives, controls, and process equipment need a specific coast-down or restart plan? | DELAY-SPECIFIC |
| Closed transition | Acceptable sources may be momentarily paralleled under synchronizing permissives. | Are source-owner or utility approval, protection, synchronization, and permitted parallel time confirmed? | NOT A SHORTCUT |
| Bypass/isolation ATS | The load can remain on a bypass path while the main mechanism is isolated for defined work. | Which parts and load paths remain energized in each equipment state? | BYPASS ≠ SAFE |
| Static transfer system | Solid-state switching uses a different product and test framework from an electromechanical ATS. | Does the exact STS manual and applicable IEC 62310, UL 1008S, or project framework govern? | SEPARATE PROCEDURE |
Do not confuse closed-transition ATS with STS. IEC 60947-6-1:2026 covers low-voltage transfer switching equipment but expressly excludes static transfer switches covered by IEC 62310. UL 1008 likewise separates solid-state static transfer equipment. Their procedures and acceptance evidence are not interchangeable.
Do not start until six readiness questions have clear answers
The test plan must match the installed assembly and the real operating impact. If a required document, role, source, communication link, or recovery path is missing, postpone the test.
- Exact equipment: ATS, controller, options, transition type, bypass/isolation arrangement, source equipment, and firmware or drawing revision are identified.
- Correct documents: approved one-line, control drawings, OEM manuals, setpoint record, prior defect list, and site switching procedure are current and available.
- Qualified roles: operating authority, test leader, electrical worker, source or generator specialist, load owner, safety lead, witnesses, and stop-work authority are named.
- Load impact: planned interruption, process response, life-safety or critical loads, communication method, and mission-recovery actions are approved.
- Every source: normal, generator, second utility, PV, UPS, BESS, stored energy, backfeed paths, and remote start/stop controls are included in the boundary.
- Recovery: expected final source, ATS position, automatic mode, alarms, remote status, temporary controls, and escalation contacts are defined before testing begins.
Seven stages from test objective to automatic-ready closeout
These stages describe management, observation, and evidence. The exact controller commands, switching actions, measurements, and internal access remain in the approved OEM and site procedure.
Define the proof objective
State whether the required evidence is start-only, functional transfer, selected-load transfer, generator load bank, integrated operation, or offline maintenance.
Write measurable acceptance criteria and exclusions before choosing the test method.Identify the installed system
Confirm ATS or STS technology, exact model/controller/options, source and load labels, transition type, bypass arrangement, and critical downstream systems.
If markings and drawings disagree, stop and correct the system record first.Approve risk and recovery
Set roles, communications, permitted interruption, source-owner coordination, access control, stop triggers, restoration authority, and expected final state.
Closed-transition and integrated tests may require utility, generator, controls, process, and safety coordination.Capture the closed-door baseline
Record source indications, ATS position and mode, alarms, event log status, remote points, restrictions, and alternate-source readiness without creating exposure.
An “available” lamp shows configured controller logic—not an independent calibrated power-quality result.Execute only the approved method
Authorized qualified personnel use the exact OEM/site procedure. Observers compare actual source, switch, load, alarm, and timing events with the approved sequence.
Never improvise a source failure, bypass a permissive, or change a setting to make the test continue.Observe transfer and return
Record each requested event, source acceptance, switch position, included-load response, retransfer condition, cooldown or source stop logic, and every alarm.
Use actual timestamps and evidence sources; do not judge with generic seconds from another controller.Restore and close the evidence gap
Verify intended source, ATS position, automatic mode, alarms, remote points, temporary states, affected loads, exceptions, owner, and retest requirement.
A test is not complete until automatic readiness and the report are independently confirmed.Record the full source-to-load-to-source sequence
The expected result comes from the approved drawings, controller settings, equipment manual, project requirements, and system operating basis—not from a universal timing table.
Normal state
Source indication, position, mode, alarms, remote status, restrictions.
Approved test call
Who initiated it, configured mode, test start time, expected scope.
Alternate available
Controller acceptance and any qualified independent evidence required by the plan.
Transfer state
Observed position, transition behavior, indications, alarms, included-load response.
Emergency period
Selected loads, protection, source stability, controls, alarms, remote points.
Retransfer logic
Normal-source qualification, return event, load response, cooldown/stop sequence.
Automatic ready
Final source, position, auto mode, no unexplained alarm, all affected loads restored.
Know what each test does not prove
This comparison prevents a common reporting error: calling any generator exercise a complete ATS or emergency-power-system test.
| Test | Strongest valid evidence | Does not automatically prove | Key record |
|---|---|---|---|
| Start-only exercise | Alternate source started and met the controller's configured available state. | ATS movement, main contact path, real-load support, or retransfer. | Configured mode and source-available evidence. |
| Functional transfer | The exact configured transfer and return sequence that was observed. | All building loads, full source capacity, or untested alarm/remote functions. | Event sequence, position, included load, alarms, timestamps. |
| Selected-load transfer | ATS power path plus only the downstream load included in the test. | Excluded loads, full facility operation, or every failure scenario. | Load list, interruption impact, measured or observed results. |
| Generator load bank | Generator capability under the stated controlled load and duration. | ATS main path or facility loads when the load-bank topology bypasses them. | Connection diagram, load profile, source results. |
| Integrated system test | Coordinated response of the systems and scenarios included in the approved plan. | Scenarios, modes, loads, or failure paths that were excluded. | Scenario matrix, witnesses, dependencies, deviations. |
| Offline diagnostics | Condition data for the isolated components and exact tests performed. | Automatic operating sequence or live-load system performance. | Isolation boundary, instrument, OEM limits, results, return-to-service checks. |
Reporting rule: write “not tested” when a function was outside scope. Do not convert a controller indication, a source start, a load-bank result, or a remote status point into broader evidence than it provides.
Stop the test when the real system leaves the approved plan
Enter the approved recovery plan. Do not continue “to see whether it clears.” Record the last known safe state and notify the responsible operating and technical authorities.
- The normal, alternate, ATS, bypass, or isolation state cannot be positively identified.
- Source quality is outside the approved acceptance band or is unstable.
- An unexpected transfer, retransfer, trip, source start, or protective-device operation occurs.
- The actual load impact differs from the approved plan.
- Field position and controller or remote indication disagree.
- An unexplained alarm, smoke, odor, abnormal heat, noise, vibration, or chatter appears.
- Required communication, witness coverage, or access control is lost.
- A guard, interlock, protection function, or setting would need to be bypassed or changed.
- A real utility or emergency event supersedes the test, or the recovery path is no longer available.
- Final automatic-ready status cannot be proved.
Three boundaries that prevent unsafe shortcuts
Closed-door observation means doors, deadfronts, and barriers remain installed. Work that creates electrical exposure needs a separately authorized qualified-person procedure.
No generic fault simulation
Do not disconnect the normal source, open a breaker, pull a fuse, bridge a control terminal, defeat an interlock, or change a setpoint from an online sequence.
The approved procedure must state the exact safe method and the authority to use it.No generic energized measurement
Opening the enclosure, removing a deadfront, probing voltage or frequency, working near exposed energized parts, internal manual operation, and bypass/isolation actions are not public steps.
PPE is one control; it does not convert an unqualified person into a qualified worker.No field recreation of type tests
Contact resistance, insulation resistance, dielectric or overpotential work, internal maintenance, and product qualification tests need exact OEM limits and controlled isolation.
Never apply a generic megger, hipot, torque, lubrication, cleaning, or retightening instruction.When the observed sequence does not match
Stop, stabilize, record, and escalate. A failed test is useful evidence when the team protects the system and preserves the last known state.
| Observed condition | Safe immediate response | Evidence to preserve | Do not do |
|---|---|---|---|
| Alternate source does not start | Stop under the recovery plan and separate the source-start issue from ATS transfer performance. | Command time, controller status, source alarms, prior restrictions, remote points. | Do not bridge start contacts or defeat source protections. |
| Source starts; ATS does not transfer | Keep the approved safe state and escalate for model-specific review of acceptance logic, mode, inhibits, and permissives. | Source-available indication, actual mode, alarms, event log, position, timestamps. | Do not force the mechanism or change setpoints to make it transfer. |
| Load result is wrong | Protect the affected process and compare the approved one-line, included-load list, and actual switch/feeder status. | Which loads changed, protection events, ATS position, source state, remote status. | Do not assume a position lamp proves every downstream circuit is energized. |
| No retransfer or no source stop | Use the approved recovery path and record normal-source stability, configured delays, modes, alarms, and observed logic. | Return-source indication, event times, controller mode, cooldown/stop status. | Do not bypass delays or manually operate an unknown mechanism. |
| Heat, odor, noise, smoke, or chatter | Abort, restrict access, place the system in the approved safe state, and escalate immediately. | Time, operating state, visible external evidence, alarms, load and source state. | Do not open the cabinet, retighten, clean, lubricate, or keep cycling. |
| Remote and field status disagree | Do not accept the test until the physical state and indication path are reconciled by qualified personnel. | Field indication, controller display, remote point, timestamps, communication alarms. | Do not treat remote-only status as proof of power-path position. |
A useful test report says what happened—and what was not tested
Keep evidence tied to the exact asset, configuration, method, and scope. A pass/fail box without the sequence, acceptance basis, deviations, and final state is weak evidence.
Need an ATS model or BOM review?
SENTOP supports product matching for generator, control-panel, equipment-manufacturer, and distribution projects. Send the project data below. Site commissioning, energized testing, and approval remain with the project's authorized qualified team.
Planning a complete panel package? Review SENTOP's ATS manufacturing support, control-cabinet wiring components, digital panel meters, surge protective devices, and customer application cases.
ATS testing questions, answered clearly
These answers define safe boundaries. They do not replace the exact manufacturer manual, site procedure, employer electrical-safety program, or local authority requirements.
Can I test an ATS by pressing its Test button?
Does a generator exercise prove that the ATS works?
Is a generator load-bank test the same as an ATS load-transfer test?
How often should an automatic transfer switch be tested?
Can a non-electrician perform an ATS test?
Do I need to open the ATS enclosure or measure live voltage?
What should I do if the ATS does not transfer?
What proves that the ATS test passed?
Are UL 1008 and IEC 60947-6-1 field test procedures?
Does this workflow also apply to a static transfer switch?
Use the current requirement that actually governs
Standards define scope and requirements. Manufacturer manuals define product operation. The authority, employer, project, and installed-system documents determine what applies at the site.
