A Guide to BESS Safety System Integration

A lithium-ion battery failure rarely starts with visible smoke. In many cases, cells begin releasing hydrogen, volatile organic compounds (VOCs) and electrolyte vapours as internal damage develops. A guide to BESS safety system integration should therefore begin before fire detection - with the controls, sensors and operating responses needed to identify a developing fault while intervention is still possible.

For BESS owners, EPCs and facility operators, integration is not simply about adding another detector inside a container. It is about creating a defined path from early-stage detection to a proportionate operational response: alerting personnel, isolating affected equipment, preserving evidence and escalating emergency action only when conditions warrant it. Done well, this safety layer protects people, reduces asset loss exposure and supports continuity at sites where availability matters.

Start with the BESS risk pathway

Thermal runaway is an escalating event, not a single moment. Mechanical damage, electrical abuse, manufacturing defects, overheating or cell ageing can trigger internal battery failure. As electrolyte decomposition begins, a battery may release gases and vapours before a smoke detector or heat detector reaches its alarm threshold.

That distinction matters. Conventional fire detection and suppression remain essential parts of a BESS safety design, but they are often intended to respond later in the event sequence. Early off-gassing detection gives operators more time to investigate abnormal conditions, reduce battery stress and initiate site-specific emergency procedures.

The appropriate design depends on battery chemistry, enclosure geometry, ventilation arrangement, rack layout, site occupancy and the consequences of an outage. A small commercial battery room should not be treated exactly like a utility-scale containerised system. The operating principle is the same, however: detect the earliest credible indicators and connect them to actions that are useful rather than merely noisy.

Guide to BESS safety system integration: define the response first

Before selecting alarm setpoints or mapping a Modbus register, develop a cause-and-effect matrix. This document defines what each alarm state means, who receives it and what the site will do next. It prevents a common weakness in BESS projects: an early-warning detector is installed, but its output is not connected to an accountable operational response.

A practical response hierarchy commonly has four stages:

1. Advisory condition: A low-level gas, VOC, humidity or temperature trend indicates an abnormal condition. The system logs the event and sends a notification for review.

2. Pre-alarm condition: Detection reaches a defined threshold or multiple indicators correlate. The control room receives a priority alarm, and operators assess the affected BESS block or enclosure.

3. Critical alarm condition: The likelihood of cell failure or thermal runaway has materially increased. The system initiates agreed controls, such as stopping charge and discharge, opening contactors or isolating the affected rack where that capability exists.

4. Emergency condition: Confirmed fire, severe gas escalation or other site-defined criteria trigger evacuation, emergency service notification and fire safety responses in accordance with the facility plan.

The exact sequence must be agreed by the asset owner, BESS integrator, fire engineer, electrical engineer and site operations team. Automatically shutting down an entire site on a low-level pre-alarm may reduce risk in some critical environments, but can create avoidable grid, process or contractual consequences elsewhere. Conversely, an advisory alarm that only sends an email may be inadequate for an unattended remote installation.

Choose detection that identifies the early warning window

BESS safety architecture should use complementary detection methods. Heat and smoke detection are valuable for identifying later-stage conditions. Gas detection is designed to identify the chemical evidence that may precede smoke and flame.

Industrial off-gassing detection systems can monitor hydrogen, VOCs, electrolyte vapours, humidity and temperature changes associated with failing lithium-ion batteries. These parameters should not be viewed in isolation. A rising VOC reading combined with an unusual temperature trend, for example, may provide a more meaningful operational signal than either value alone.

For BESS applications, the Evikon E2673 provides an engineered early-warning layer with relay outputs and Modbus RTU compatibility. This allows the detector to communicate with local controls, a building management system, an energy management system or site SCADA. Its compact form can also assist where equipment cabinets and battery enclosures have limited available space.

Sensor selection must account for the operating environment. Consider expected ambient temperature, dust, humidity, ventilation flow, electromagnetic interference, power supply availability and cable routing. A detector that performs well on a laboratory bench can give poor operational value if it is mounted where ventilation bypasses the sensing area or where maintenance access is impractical.

Place sensors where gases will travel

Sensor placement is one of the most consequential design decisions. Off-gases do not distribute evenly through an enclosure. They move according to air paths, mechanical ventilation, pressure changes, cable penetrations and the physical layout of battery racks.

Start with the enclosure's airflow design. Identify supply and extract points, natural high points, areas where air may stagnate and routes that could carry gases from a battery rack towards other equipment. In containerised BESS units, extraction pathways and ceiling-level zones often require close assessment, but the right position cannot be assumed from a generic layout drawing.

Where possible, use a gas dispersion assessment or computational fluid dynamics study for larger or higher-consequence projects. For smaller installations, a documented review of ventilation flow and enclosure geometry may be sufficient. The objective is to locate detection where it has the best chance of identifying a developing incident early, while avoiding false readings from external contaminants or exhaust pathways.

Do not overlook ancillary spaces. Battery rooms, inverter rooms, cable basements and service corridors may have different air movement patterns and different evacuation implications. Integration should consider the whole hazard boundary, not only the battery rack itself.

Connect detection to the systems that can act

A detector has limited value if it is visible only at the device. Safety system integration should establish both local annunciation and remote visibility. Local indicators and audible alarms assist technicians on site. SCADA integration gives operators, asset managers and control rooms a time-stamped view of conditions across the fleet.

Modbus RTU is commonly used for detector-to-control-system communication in industrial environments. The integration specification should clearly identify register mapping, alarm status, fault status, sensor readings, communications loss behaviour and polling intervals. It should also define whether a loss of communications creates a maintenance alarm, a safety alarm or both.

Hardwired relay outputs remain important even in connected facilities. A relay can provide a direct, deterministic signal to a fire indicator panel, PLC, BESS controller or shutdown circuit. This may be appropriate for critical alarm actions that should not rely solely on a higher-level network path.

The key is to avoid conflicting commands. If the BESS controller, fire system, HVAC controls and SCADA platform all respond independently, they can produce unintended outcomes. For example, ventilation may need to run to manage gas accumulation during one condition, but fire containment requirements may call for a different response after fire confirmation. These interlocks need engineering review and formal cause-and-effect testing.

Design alarms for people, not just dashboards

Alarm fatigue weakens safety. If operators receive frequent non-actionable notifications, they can become desensitised to the event that genuinely needs immediate attention. Setpoints should be based on detector capability, normal background conditions, battery manufacturer guidance, ventilation design and the site's risk assessment.

Trend information is especially useful. A fast change in gas concentration may warrant a higher priority than a stable low-level reading, even if both are within a similar numeric range. Alarm logic can also use persistence, rate of rise and multiple-sensor confirmation to reduce nuisance alarms without delaying response to credible events.

Every alarm should have an owner. Specify who acknowledges it outside business hours, what information they need to make a decision, when they dispatch personnel and when they escalate to emergency services. For remote Australian sites, where travel time can be substantial, early warning and clear remote visibility are particularly valuable.

Commission, test and maintain the safety layer

Commissioning is where a BESS safety design becomes operational. Confirm sensor placement against the final installed enclosure, not only the design drawings. Verify power supply quality, cable terminations, relay states, Modbus communications, SCADA point names, alarm priorities and event timestamps.

Functional testing should prove the full chain of response. A simulated advisory, pre-alarm and critical alarm should be visible at the detector, local panel and remote monitoring platform. It should also confirm the intended BESS control response, notification workflow and fail-safe behaviour if communications are interrupted.

Maintenance requirements should be included in the asset management plan from day one. This includes inspection intervals, functional checks, calibration or service requirements where applicable, cleaning provisions and replacement planning. Maintenance-free performance and long service life can reduce operational burden, but they do not remove the need for periodic verification that the complete detection-to-response pathway still works.

NexaGuard supports BESS projects with specialised early-stage off-gassing detection designed to identify danger before smoke and flames occur. The strongest result comes when this technology is treated as part of the site's engineered safety strategy, rather than as an isolated accessory.

A well-integrated BESS safety system gives operators something more useful than an alarm after conditions have deteriorated: time to make informed decisions, protect critical infrastructure and act while prevention is still possible.