top of page
Battery Incident Response Before Fire Starts

Battery Incident Response Before Fire Starts

2 days ago
6 min read

A lithium-ion battery incident rarely begins with visible flames. In a BESS enclosure, UPS room or EV charging area, the first signs can be hydrogen, VOCs and electrolyte vapours released as cells degrade internally. A capable battery incident response plan is built around that reality: recognising abnormal conditions early enough to protect people, isolate the risk and preserve critical operations before thermal runaway takes hold.

For asset owners and safety leaders, this changes the question from “How do we respond to a battery fire?” to “What decisions must happen in the minutes, or potentially hours, before a fire?” That earlier window is where detection technology, control logic and trained site personnel have their greatest value.

Why conventional fire response is not enough

Conventional fire protection remains essential, but it is the final layer of a much broader safety strategy. Smoke detection, heat detection and suppression systems are generally designed to respond once combustion or significant heat is already present. By that stage, a lithium-ion battery event may be escalating rapidly, with potential for toxic gases, reignition, equipment loss and prolonged site disruption.

Thermal runaway is a self-heating failure process within a battery cell. It may be initiated by manufacturing defects, mechanical damage, overcharging, electrical faults, elevated ambient temperatures, cooling failures or progressive cell ageing. As the cell chemistry breaks down, it can release flammable and harmful gases before smoke or flame appears.

This does not mean every gas alarm is a thermal runaway event. Gas concentrations can be influenced by ventilation, battery chemistry, enclosure size and adjacent equipment. It does mean an early off-gas alarm deserves a defined operational response, rather than being treated as a nuisance alarm or a maintenance task for the next shift.

Battery incident response starts with early warning

An effective response framework separates an emerging battery fault from a confirmed fire. It establishes what the site will do at each level of escalation, who has authority to make decisions and how operators verify that the danger is reducing rather than progressing.

Early-stage detection should monitor indicators associated with battery failure, including hydrogen and electrolyte vapours, VOCs, humidity shifts and temperature changes. In larger installations, these signals should be visible through the existing building management system, SCADA platform or site control room where practical. Relay outputs and Modbus RTU compatibility can support automatic alarms, ventilation actions, equipment isolation and remote notification.

The aim is not to automate every decision without judgement. The aim is to give operators timely, credible information while response options still exist. A detector located close to the credible source of off-gassing can provide a materially different warning window from a ceiling-mounted smoke detector that only responds after the incident has advanced.

Define alarm levels before an event

A clear alarm philosophy prevents confusion under pressure. The exact setpoints and sequence should be developed for the battery chemistry, room configuration, ventilation design, operating profile and risk assessment. However, most sites benefit from a staged approach.

At an initial advisory level, the system may notify the control room, log the event, increase ventilation where that is safe and initiate an investigation by authorised personnel. At a higher alarm level, the response may include stopping charging or discharging, isolating affected battery strings or containers where designed to do so, restricting access and escalating to emergency contacts.

At the emergency level, personnel should evacuate in accordance with the site emergency plan, emergency services should be contacted and responders should be given accurate information about the battery type, location, state of charge, system isolation status and any active gas or temperature alarms. The response must account for the potential for delayed escalation and reignition.

The operating decisions that protect people and uptime

A battery incident response plan should be practical enough to use at 2 am, by an operator who may not be a battery specialist. Long procedures are valuable for engineering review, but the live response process needs clear actions, clear ownership and unambiguous escalation thresholds.

First, protect people. Prevent unnecessary entry into a battery room, container or enclosed charging area after an off-gas alarm. Personnel should not enter simply to confirm an odour, reset a detector or inspect a rack without assessing the risk. Depending on the hazard assessment, appropriate gas monitoring, ventilation, electrical isolation procedures and emergency response controls may be required.

Second, stabilise the asset where the system design permits. This can include ceasing charge and discharge activity, preventing automatic restart, isolating faulty equipment and confirming the status of HVAC or exhaust systems. Isolation needs careful engineering. Switching equipment may reduce electrical energy input, but it will not necessarily stop an internal cell failure that is already progressing.

Third, preserve evidence and operating data. Alarm timestamps, detector readings, battery management system records, inverter data, thermal images and CCTV can help technical teams identify whether the cause is an isolated cell, a module-level fault, a cooling issue, an electrical abnormality or a wider installation problem. This information also supports insurer notification, regulatory obligations and a more reliable return-to-service decision.

Detection location and integration matter

A detector is only as useful as its installation and alarm pathway. In BESS applications, gas sensor placement should consider where gases are likely to accumulate or be transported by airflow, as well as access for inspection and functional testing. A poorly placed detector can produce delayed warning, while an overly exposed position may create false alarms or unnecessary maintenance burdens.

Battery rooms, UPS spaces, charging bays and electrical enclosures all have different airflow patterns and failure modes. There is no single placement rule that fits every site. Engineering teams should assess enclosure volume, HVAC operation, pressure relief pathways, cable penetrations, likely fault locations and the specific battery technology in use.

For industrial applications, fixed off-gas detection systems such as the Evikon E2673 can form an engineered early-warning layer by monitoring hydrogen, VOCs, electrolyte vapours, humidity and temperature indicators. When connected to SCADA or building controls, the alarm can become an actionable site event rather than an isolated local warning. Compact installation and maintenance-free operation are particularly valuable where equipment rooms are constrained or difficult to access.

Integration should also be tested as a complete chain. It is not enough to confirm that a sensor powers on. Sites should verify that the alarm reaches the control room, the correct relay or Modbus point changes state, notifications are delivered, response instructions are available and duty personnel understand what the alarm means.

Training turns a plan into a response capability

Battery response procedures often fail at the handover point between engineering, operations, security and emergency services. Each group may have partial information, but no one person has the full operational picture. A planned exercise can expose those gaps before an actual event does.

Training should include realistic scenarios: an early off-gas alarm with no visible smoke, a high alarm during peak charge, loss of ventilation, a detector fault, and a suspected thermal runaway event requiring evacuation. Teams should practise identifying the asset, isolating it where safe, communicating status and preventing an unauthorised reset or restart.

Contractors need particular attention. Electrical contractors, maintenance personnel and cleaners may be first to encounter an alarm in a plant room or charging space. They need to know that invisible off-gassing is a potential battery hazard, where to report it and when not to enter. This is especially relevant at facilities with changing work crews or remote monitoring arrangements.

Recovery requires more than resetting alarms

An incident is not over when the alarm clears. Before returning a battery system to service, the responsible engineering team should determine why the event occurred and whether the remaining system is safe to operate. That may involve battery supplier support, inspection of modules and connections, review of battery management logs, thermal imaging, electrical testing and replacement of affected components.

The recovery decision should balance continuity against consequence. A data centre or critical infrastructure site may face substantial commercial pressure to restore capacity quickly. Yet returning a compromised battery asset to operation without understanding the initiating fault can create a second, potentially more severe event.

After every alarm, even one that proves non-critical, review the response. Were alarm thresholds appropriate? Did ventilation operate as expected? Did the right people receive the alert? Was the control room able to identify the affected zone? Small improvements after a minor event can prevent major consequences later.

A battery incident response plan earns its value before emergency crews arrive. Early detection of off-gassing gives site teams time to make calmer, safer decisions: protect people, isolate risk, inform responders and preserve the infrastructure that keeps the operation running. When seconds matter, the most valuable warning is the one that arrives before smoke and fire.

 
 
 

Comments


bottom of page