A firewater deluge or a facility shutdown is a big, disruptive action, so a fire and gas system is careful about what triggers it. Voting, also called coincidence detection, is the logic that provides that care. Instead of letting a single detector command a major response, the system waits until two or more detectors in the same area agree that gas is present. This guide explains how voting is applied at the fire and gas level, how it differs from a first-alarm alert, and the balance it strikes between avoiding false trips and staying safe.
Gas detector voting in one line: Gas detector voting, or coincidence detection, requires two or more detectors within a defined zone to confirm gas before the fire and gas system takes an executive action such as emergency shutdown or firewater deluge. This prevents a single faulty or contaminated detector from tripping a facility on its own. Typically the first detector to alarm annunciates a warning, and only a confirmed multi-detector alarm drives the automatic mitigation.
Gas detectors live in a harsh world. They can be splashed, contaminated, knocked, or simply fail, and any of those can produce a spurious high reading. If a single detector going into alarm were enough to fire a deluge or shut down a process area, then every faulty sensor would risk an expensive and hazardous nuisance trip. Frequent nuisance trips also erode trust in the system, tempting people to bypass detectors, which is far more dangerous than the occasional false alarm.
Voting solves this by insisting on agreement. When the logic requires two detectors in a zone to confirm before acting, a single faulty detector can still raise a warning but cannot by itself command the high-consequence output. A real gas release, being a physical cloud that spreads across an area, will typically reach more than one detector, so genuine events still get through while lone false alarms are held back from executive action.
This is the fire and gas application of the more general voting idea. Rather than describing sensor channels of a single instrument, coincidence detection here groups multiple physical detectors watching the same real-world zone and requires a minimum number of them to concur. The purpose is the same everywhere: to keep a single point of failure from either causing a spurious trip or, in other arrangements, from being able to block a needed action.
Voting only makes sense once the plant is divided into zones. A zone is an area whose detectors are grouped together because a release there should reach several of them. Defining zones is a design task informed by the layout and the coverage study, and detectors are assigned so that a credible cloud in the zone would be seen by enough of them to satisfy the vote. Get the zoning wrong, with detectors too far apart, and a real release might only reach one detector, defeating the coincidence requirement.
The logic usually distinguishes two stages. The first detector in a zone to sense gas raises a first-alarm annunciation, alerting operators and often starting low-consequence responses like extra ventilation or a status change, but it does not trigger executive action. When a second detector in the same zone confirms, the alarm becomes a confirmed gas alarm and the system is authorized to take the high-consequence steps such as isolating the source, initiating emergency shutdown, or releasing deluge.
This two-stage arrangement means nothing is ignored while nothing overreacts. Operators are told immediately when any detector sees gas, so they can respond and investigate, but the automatic actions that halt production or dump firewater wait for corroboration. The exact vote can be tuned per zone: a critical, congested area might use a tighter arrangement than a low-hazard space, reflecting how much confidence is required before the plant reacts.
Voting is a balance, not a free win. Requiring two detectors to confirm reduces spurious trips but, in theory, makes the system slightly slower or less likely to act if one of the needed detectors is out of service. If a zone needs two votes and one detector is faulted or bypassed, the zone may be unable to reach a confirmed alarm at all, which quietly weakens protection. The choice of voting arrangement therefore trades safety availability against spurious-trip avoidance, and it must account for detectors being unavailable for maintenance.
That makes the live voting state of each zone genuinely important operational information, not just design detail. Operators need to know which detectors are healthy, which are in alarm, and whether a zone still has enough working detectors to confirm a real event. A zone silently running on a single available detector is in a degraded state that should be visible and acted on, whether by prioritizing repair or by adjusting how the area is watched.
This is where remote visualization helps. A cloud monitoring platform can show per-zone voting status: which detectors have alarmed, whether the zone is in first alarm or confirmed alarm, and how many detectors are available to vote. Merobix presenting that state for scattered or unmanned sites lets a central team see not only that gas has been detected but whether each zone's voting logic is healthy enough to act, turning an abstract safety-logic detail into something an operator can actually watch.
A first alarm is raised as soon as any single detector in a zone senses gas; it alerts operators and may start low-consequence responses but does not trigger executive action. A confirmed gas alarm occurs when a second detector in the same zone also confirms gas, satisfying the voting requirement. Only the confirmed alarm authorizes high-consequence actions like emergency shutdown or deluge.
Because detectors can fail or be contaminated and produce false high readings, and acting on every single detector would cause frequent, disruptive nuisance trips. Those trips are costly and erode trust in the system, which can lead people to bypass detectors, a far worse outcome. Requiring two detectors to confirm keeps genuine releases getting through while filtering out lone false alarms from executive action.
If a zone requires two votes and one of its detectors is bypassed or faulted, the zone may not be able to reach a confirmed alarm, which weakens protection there. This is why detector availability matters so much and why bypassed or failed detectors should be tracked and repaired promptly. Monitoring per-zone voting status makes such degraded conditions visible before they matter.
Merobix reads your field devices into a cloud SCADA - the real thing behind these terms, live in days from any browser.