What Signals Should a UPS or Generator Send to a BMS?

A BMS can show that a generator is running or that a UPS has a common alarm. Those points are useful, but they may not answer the questions an incident team actually has: Is the critical load on backup power? How much runtime remains? Is redundancy degraded? Who needs to respond now?

Good integration starts with operational decisions, then selects signals. It does not begin by importing every available register into one crowded alarm page.

Quick answer: which signals are essential?

At a minimum, the BMS should distinguish normal state, loss of normal supply, equipment running or supporting load, bypass or transfer state, common warning, common shutdown and communications failure. Critical sites normally benefit from selected load, battery-runtime, fuel and temperature measurements as well.

Separate state, alarm and measurement

State

A state describes the current operating position: UPS on inverter, UPS on battery, ATS on generator, generator running or equipment in manual mode.

Alarm

An alarm identifies an abnormal condition requiring attention: low fuel, battery fault, failed start, overload or high temperature.

Measurement

A measurement adds context: load percentage, kW, voltage, frequency, battery charge, runtime estimate, fuel level or temperature.

Keeping these types distinct makes displays clearer and avoids treating a normal operating state—such as generator running during a planned test—as a fault.

Recommended UPS status points

1. UPS available/healthy
2. normal operation on inverter
3. on battery
4. on static or maintenance bypass, where available
5. battery charging or low battery
6. output supplying load
7. redundancy lost or module unavailable for modular systems
8. remote shutdown active or UPS not in normal auto configuration

Recommended UPS alarms

• common warning and common critical/shutdown alarm
• battery fault or replace-battery condition
• low battery or low runtime
• overload or capacity exceeded
• high temperature or fan fault
• rectifier, inverter or charger fault
• input, bypass or output supply abnormal
• communications failure

The exact alarm map is model-specific. Where the BMS cannot import detailed codes, retain access to the UPS event log and define what the common alarm means operationally.

Useful UPS measurements

1. input and output voltage and frequency
2. output kW, kVA, current and load percentage
3. battery charge percentage and estimated runtime
4. battery voltage or string data where supported and appropriate
5. internal or battery temperature
6. module load and redundancy state

Measurements are valuable when they drive thresholds or trends. Avoid displaying precision that the sensor and protocol do not genuinely support.

Recommended generator status points

• auto mode available
• start command present
• engine running
• generator available at acceptable voltage and frequency
• generator breaker closed or load supplied, where monitored
• mains failure
• ATS normal-source or generator-source position
• cool-down or stopping state where useful

'Generator running' and 'critical load on generator' are different facts. Include breaker or ATS position if the response depends on knowing whether power has actually transferred.

Recommended generator alarms

• failed to start or failed to stop
• common warning and common shutdown
• low fuel and critically low fuel
• low oil pressure and high coolant temperature
• battery or charger fault
• over/under voltage or frequency
• overload or breaker trip
• fuel-transfer or day-tank fault
• not in auto mode
• communications failure

Useful generator measurements

• voltage, frequency and current by phase
• kW, kVA, power factor and load percentage
• engine speed, oil pressure and coolant temperature
• running hours and start count
• fuel level and, where derived reliably, estimated autonomy
• starting-battery voltage and charger state

Do not forget the ATS and distribution

The ATS connects source availability to load supply. Normal source available, emergency source available, switch position, transfer failure, manual mode and common alarm can be as important as the generator points. Critical breakers or downstream distribution may need their own status.

Dry contacts, Modbus or network monitoring?

Dry contacts

Simple relay contacts are robust and easy to understand. They suit a small number of essential states or alarms but provide limited detail. Decide whether the contact is normally open or normally closed and how cable or power failure will be detected.

Modbus

Modbus can expose many states and measurements. The integrator needs the correct register map, data type, scale, byte order, unit address and firmware context. Polling too many unnecessary points can create noise and maintenance burden.

SNMP or vendor network interface

Network interfaces can support richer remote monitoring and notifications. They require ownership, cyber-security controls, time synchronisation, access management and firmware maintenance.

Design communications-loss alarms

A silent equipment point can look healthy if the BMS only alarms on an active value. Use a heartbeat, watchdog or protocol-quality state so loss of data is visible. Define how quickly a missed update becomes an alarm and avoid nuisance events during planned network work.

Prioritise alarms by action

• critical: immediate risk to load, safety or available runtime
• urgent: resilience degraded and engineer response required
• warning: condition needs investigation or planned action
• information: state change or planned test event

Each priority should have a recipient, response expectation and escalation route. A BMS that raises an alarm nobody owns is only an archive.

Test from equipment to person

Commissioning should prove the point at the source, its correct label and state on the BMS, timestamp, priority, notification and response record. Test normal-to-alarm and alarm-to-normal transitions, plus communications loss.

Retest after controller, interface, firmware, BMS or network changes. During generator and UPS maintenance, confirm that planned test events are handled correctly and that equipment returns to normal configuration.

Keep a controlled signal schedule

For every point, record source equipment, protocol address or terminal, name, units, normal state, alarm priority, delay, recipient and response. Include firmware/register-map version where applicable. This schedule is the contract between equipment supplier, controls integrator and operator.

Frequently asked questions

Make monitoring answer the incident team's questions

The strongest signal list is short enough to operate, detailed enough to diagnose and tested all the way from equipment to the responsible person.

Contact P&I Group to review UPS, generator and ATS monitoring, remote alerts and maintenance integration.