What to Do When Your UPS Alarms: A Step-by-Step Emergency Response Guide

A UPS alarm demands an immediate, structured response. Whether you're an IT support officer, a facility manager, or an overnight security operator with limited electrical knowledge, the actions you take in the first five minutes can determine whether your load stays protected or goes down hard.

This guide walks through a practical response sequence, covers the most common alarm conditions across Eaton, APC/Schneider, Vertiv, and CyberPower systems, and explains what information to have ready before you call a technician.

Step 1: Establish Whether the Load Is Still Protected

Before anything else, determine the UPS operating mode. Most modern UPS systems display their current mode on the front panel LCD or via a status LED array.

Normal (online double-conversion): The inverter is active, the load is fully conditioned and protected. An alarm in this state is a warning, not an emergency. You have time to assess.

On battery: The UPS has lost mains input and is running from battery. This is time-sensitive. Check the estimated runtime on the display. If it shows less than 10 minutes, begin controlled shutdown procedures for non-essential loads immediately.

Bypass mode: The load is connected directly to mains with no UPS protection. Any mains disturbance will pass straight through to your equipment. Treat this as an unprotected state and escalate quickly.

Fault/shutdown: The UPS has removed itself from service. The load may be dead or on a separate source. Verify server room power status directly.

If you cannot read the display because the unit is alarming loudly and showing a red fault indicator, note the fault code or alarm description shown. Most units display a text string or numbered code. Write it down before you do anything else.

Step 2: Silence the Audible Alarm (Without Clearing the Fault)

Audible alarms exist to get attention, not to be dismissed. Most UPS systems allow you to mute the buzzer without clearing the underlying fault condition. On Eaton 9PX and 9SX units, a short press of the alarm mute button silences the tone while keeping the fault active on the display. On APC Smart-UPS SRT units, the same function is typically accessed through the front panel button or the NMC web interface.

Do not press any button that says "clear fault" or "reset" until you understand what caused the alarm. Resetting a fault without understanding it can mask a developing failure.

Step 3: Identify the Alarm Condition

The table below covers the most frequent alarm conditions across the brands UPS Services Australia services regularly.

Eaton (9PX, 9SX, 93PM, 93E)

• Battery low / EOD (end of discharge): Battery runtime is exhausted or near zero. If on battery, load shutdown is imminent.
• Replace battery: Internal battery test has failed. The UPS remains operational but runtime is compromised. Schedule replacement within 30 days.
• Overload: Connected load exceeds UPS rated capacity. Identify and shed non-essential loads.
• Output fault: Inverter output is outside tolerance. The UPS may have transferred to bypass. Do not reset without technician assessment.
• Fan fault: Cooling fan has failed. The unit will derate or shut down under thermal load. Ensure ambient temperature is within spec (typically below 25°C for full rating).
• Communication fault: Network management card or SNMP module has lost contact. Not a power emergency, but monitor closely.

APC / Schneider Electric (Smart-UPS SRT, Symmetra, Galaxy)

• On battery: Mains has failed or is out of tolerance. Check incoming supply at the distribution board.
• Battery disconnected: Internal battery connector has failed or battery module is not seated. Common after a battery replacement.
• Site wiring fault: The UPS has detected a wiring issue on the input circuit, often a missing neutral-earth bond or reversed polarity. Do not ignore this; it indicates an installation problem.
• Overload / near overload: Load is at or above rated capacity. Galaxy series will alarm at 85% load as a warning threshold.
• Bypass active: The unit has transferred to static bypass due to an internal fault. Load is unprotected.
• Temperature alarm: Internal temperature has exceeded threshold. Check room cooling.

Vertiv / Liebert (GXT5, APM, EXL S1, Trinergy Cube)

• Input voltage out of range: Mains voltage has drifted outside the UPS input window. The unit will transfer to battery if mains does not recover.
• Battery test failed: Scheduled self-test has returned a fail result. Runtime figures on the display should no longer be trusted.
• Module fault (APM/Trinergy): In modular systems, a power module has faulted and removed itself. The system may still be operating at reduced capacity. Check N+1 redundancy status.
• EPO activated: Emergency power off has been triggered. This requires a deliberate reset sequence and investigation of why EPO was activated before restoring.

CyberPower (OLS, OL Series)

• Utility failure: Standard on-battery alarm. Check mains supply.
• Battery capacity low: Battery state of charge is below 25%. If on battery, shutdown is approaching.
• UPS fault: General internal fault. The unit will typically display a fault code alongside this alarm. Note the code.
• Overload: Load exceeds rated output. Shed load immediately.

Step 4: Check the Incoming Mains Supply

If the UPS is alarming for an input-related reason (on battery, input out of range, utility failure), verify the mains supply before assuming the UPS has failed. Check:

• The main distribution board for tripped breakers
• Whether other equipment on the same circuit is functioning
• Whether the building has experienced a wider power event

A UPS that has transferred to battery because of a genuine mains failure is doing exactly what it should. The alarm is informational. Your job is to monitor runtime and manage the load until mains is restored.

Step 5: Operating the Maintenance Bypass

If the UPS itself has faulted and you need to keep the load energised while waiting for a technician, the maintenance bypass allows you to route mains power directly to the output without UPS protection. This should only be used when the UPS has failed or must be taken offline, and only if mains supply is stable.

The correct sequence for a manual maintenance bypass on most systems:

• Confirm mains input to the UPS is present and stable.
• Locate the maintenance bypass switch or handle, usually a rotary switch or a dedicated breaker panel on the side or rear of the cabinet.
• On systems with a static bypass already active, the transfer to manual bypass is a single switch operation. On systems still in normal mode, first transfer to static bypass via the front panel, then engage the manual bypass.
• Verify the load has transferred by checking that output voltage is present and equipment remains operational.
• The UPS input and output breakers can now be opened to fully isolate the UPS for servicing.

Do not attempt this sequence on three-phase systems without referring to the specific manufacturer wiring diagram for that installation. Three-phase bypass involves multiple breakers and an incorrect sequence can cause a momentary outage or a phase fault. If you are not certain, leave the UPS in its current state and call for emergency service.

Step 6: When to Call for Professional Service

Call a UPS technician immediately for any of the following:

• The UPS has faulted and the load is unprotected
• The unit is on battery and runtime is under 15 minutes with no mains restoration in sight
• An EPO has activated and you do not know why
• The display shows an output fault, inverter fault, or rectifier fault
• You can smell burning or see any sign of thermal event
• A battery module is showing a fault on a lithium-ion system
• The maintenance bypass sequence is unclear or the bypass hardware is unfamiliar

For non-urgent conditions (replace battery warning, fan fault with stable temperatures, communication faults), schedule a service call within 48 to 72 hours. Do not let a replace battery warning sit for months; a battery that has failed its internal test will not deliver rated runtime when you actually need it.

Step 7: Information to Have Ready for Your Technician

A technician who arrives with full context resolves the fault faster. Before calling, gather:

• UPS make, model, and serial number: Usually on a label on the front or side panel
• The exact alarm text or fault code: Photograph the display if possible
• Current operating mode: Normal, on battery, bypass, or fault
• Approximate load percentage: Shown on most displays as a bar graph or percentage
• Battery age and last replacement date: Check your maintenance records or the date stamp on the battery label
• Whether a recent event preceded the alarm: Power outage, load change, maintenance work, or environmental change
• Site access details: After-hours entry procedures, parking, and who to contact on arrival

For 24/7 emergency callouts, having this information ready at the point of the call reduces diagnostic time on site by a measurable margin.

Preparing Before the Next Alarm

The best time to understand your UPS alarm system is before an alarm occurs. Most UPS manufacturers publish alarm code reference guides for each model. Printing the relevant pages and keeping them in the server room or plant room means that anyone on site, including overnight security, can identify an alarm condition without needing to log into a system.

Scheduled preventive maintenance under AS IEC 62040 includes alarm verification testing, which confirms that fault conditions are correctly detected and reported. If your UPS has never had its alarm outputs tested, you cannot be confident they will behave correctly under a real fault.

UPS Services Australia provides emergency callout across Brisbane, Sydney, and Melbourne, along with scheduled maintenance programmes that include alarm testing and battery condition assessment. For more information on emergency response services or to discuss a maintenance agreement, visit [https://ups.services](https://ups.services).