Alarm Monitoring System

What defines this type

An alarm monitoring system is the data-acquisition and alarm-handling layer that reads pressure, temperature, level and status signals from sensors throughout the engine room and auxiliary systems, compares each against pre-set limits, and raises an alarm the instant a limit is crossed. It is distinct from the individual protective trips built into a specific machine, such as a diesel engine's own overspeed shutdown, in that it covers the whole engine room as one integrated picture rather than one machine at a time, and it is what allows the vessel to run with a periodically unattended machinery space instead of a continuous watch. Where an automation and control system also takes protective action, such as starting a standby pump or slowing the engine, the alarm monitoring function is the layer that first detects and reports the abnormal condition.

Main components

• Field sensors and transmitters throughout the machinery spaces, feeding pressure, temperature, level and limit-switch signals into the system
• Remote I/O modules or multiplexers that gather signals locally and pass them to the central processing units over a data network, reducing the amount of individual wiring run back to the control room
• Central processing units running the alarm logic, limit comparison and event logging, usually duplicated for redundancy
• Operator stations in the engine control room and on the bridge, displaying mimic diagrams, alarm lists and trend graphs
• Extension alarm panels in crew accommodation, required so an alarm can wake an engineer during unmanned periods

Selection and sizing

• Number of monitored points required by the vessel's machinery configuration and the class unmanned machinery space notation being sought
• Redundancy of processing units and network architecture, since a single point of failure in the alarm system defeats the purpose of unmanned operation
• Integration scope with other systems: engine control, ballast, cargo, fire detection, and whether these stay as separate systems with hard-wired alarm repeat or are fully integrated on one platform
• Data logging and trend capacity for troubleshooting and for satisfying survey requirements on historical alarm records

Regulations and class

SOLAS Chapter II-1 sets requirements for periodically unattended machinery spaces, including the alarm system's capability to detect faults and alert crew in accommodation. Class societies issue an unmanned machinery space notation, terminology varies by society, only after verifying the alarm system covers all machinery required by their rules for that notation, including dead-man alarm arrangements confirming an engineer acknowledges and responds to alarms during unmanned periods. Periodic survey confirms the alarm points still match the as-built list and that extension alarms in accommodation actually function.

Typical faults

What to look for in a supplier

• Genuinely independent redundancy in processing units, power supply and network path, not just a duplicated cabinet
• Open or documented protocol for future integration with other systems rather than a closed proprietary bus
• Alarm point list and setting documentation handed over in a form the crew and future surveyors can actually audit
• Track record of software support and spare parts availability over the system's expected service life, since these systems run for decades between major upgrades

A high nuisance-alarm rate is the single biggest threat to an alarm monitoring system's purpose: a crew that mutes alarms out of habit has quietly turned an unmanned machinery space back into an unmonitored one.