A catalytic bead combustible gas sensor has a dangerous failure mode: it can go partly or fully blind to gas while still looking perfectly healthy in clean air. This is poisoning, and it happens when certain substances degrade the catalyst the sensor relies on. Because a poisoned sensor reads a normal zero and gives no obvious warning, it can silently fail to detect a real release. This guide explains what poisons a pellistor, why the failure is so quiet, and how it is caught and prevented.
Catalytic sensor poisoning in one line: Catalytic sensor poisoning is the loss of sensitivity in a catalytic bead (pellistor) combustible gas sensor when its catalyst is degraded by contaminants such as silicones, lead, and sulfur compounds, or temporarily suppressed by inhibitors. A poisoned sensor can still read a normal zero in clean air yet respond weakly or not at all to real gas, making the failure silent. Regular bump testing with gas is the reliable way to catch it, and infrared detectors are immune.
A catalytic bead sensor works by burning flammable gas on a catalyst-coated heated bead and measuring the resulting temperature rise. Its whole measurement depends on that catalyst staying active. Poisoning happens when a substance reaches the bead and permanently degrades the catalyst, so that even when flammable gas arrives, less of it oxidizes and the temperature rise, and therefore the reading, is reduced. Over enough exposure the sensor can lose most or all of its ability to respond to gas.
The classic poisons are silicones, which are found in many sealants, greases, cleaning agents, and personal-care products, along with lead compounds and certain sulfur and phosphorus species. Inhibition is a related but often temporary effect, where a substance such as some sulfur compounds or halogens suppresses the catalyst while present and the sensor may partly recover once the contaminant clears. Whether the effect is permanent poisoning or temporary inhibition, the immediate consequence is the same: the sensor under-reports flammable gas.
What makes this so hazardous is that it is silent. A poisoned pellistor sitting in clean air still reads zero, exactly as a healthy one does, because there is no gas to expose the loss of sensitivity. Nothing on the display or in the baseline signal reveals that the sensor would fail to alarm on a real release. Unlike a sensor that fails to an obvious fault, a poisoned sensor keeps up a convincing appearance of normal operation while providing little or no protection.
Because poisoning does not show up in clean air, the only reliable way to catch it is to challenge the sensor with gas. A bump test does exactly this: a known concentration of gas is applied and the technician confirms the detector responds and reaches the expected level. A poisoned sensor fails this challenge by responding weakly or not at all, which is why regular bump testing is the frontline defense against silent poisoning. The more contamination a sensor might see, the more frequently it should be tested.
Full calibration goes further by quantifying the loss, revealing through the as-found reading how much sensitivity the sensor has lost against a known gas. Tracking that as-found value over time turns a series of tests into a trend that shows a sensor gradually being poisoned, often before it fails outright. A sensor whose response keeps dropping test after test is a candidate for replacement even if it still technically passes, because the trend points toward an imminent silent failure.
Infrared point detectors avoid the problem entirely because they do not use a catalyst. An infrared sensor measures how much infrared light hydrocarbon gas absorbs, a physical measurement with nothing to poison, so silicones and sulfur do not degrade it. This immunity is a major reason infrared detectors are chosen in areas where poisons are likely, or where a silent loss of sensitivity would be intolerable. The trade-off is that infrared cannot detect hydrogen, so catalytic and infrared technologies are matched to where each fits best.
Beyond testing, poisoning is mitigated by keeping poisons away from the sensor and by design choices. Controlling the use of silicone-based products near detectors, being careful with sealants and greases during construction and maintenance, and specifying poison-resistant sensor elements where available all reduce exposure. In areas known to carry poisons, choosing infrared detection for the combustible hazard sidesteps the risk altogether. These measures do not remove the need for testing but they slow the rate at which sensors are attacked.
The management side is about not trusting a catalytic sensor by its display alone. Since a poisoned sensor looks healthy, the discipline is to verify it with gas on a defined schedule and to act on declining response rather than waiting for a hard failure. Sensors in poison-prone service warrant tighter test intervals, and a run of weakening as-found readings should trigger replacement. The whole point is to replace a poisoned sensor on the strength of test data, before it is called on during a real release.
This is where a maintenance record and monitoring platform help most. Logging bump and calibration results with as-found sensitivity, flagging detectors as their test dates come due, and trending each sensor's response lets a team see which catalytic detectors are drifting toward poisoning across many sites. Merobix keeping that history, alongside live detector fault status, means the silent nature of poisoning is countered by a visible trend, so a weakening sensor at a remote location is caught in the data rather than during an incident.
The classic poison is silicone, found in many sealants, greases, cleaning products, and personal-care items, along with lead compounds and certain sulfur and phosphorus species. Some sulfur compounds and halogens act as inhibitors that suppress the sensor temporarily. All of them degrade or block the catalyst the pellistor relies on, reducing its ability to respond to flammable gas.
Because the failure is silent. A poisoned catalytic sensor still reads a normal zero in clean air, so nothing on the display suggests a problem, yet it would respond weakly or not at all to a real gas release. Without a functional test using gas, an operator has no way to know the sensor has lost sensitivity and would fail to alarm when it matters.
Yes, for the combustible hazard they cover. Infrared detectors measure how much infrared light hydrocarbon gas absorbs, a physical measurement with no catalyst involved, so silicones, sulfur, and other poisons do not degrade them. This immunity makes infrared a common choice in poison-prone areas, though it cannot detect hydrogen, so catalytic sensors still have their place where hydrogen detection is needed.
Merobix reads your field devices into a cloud SCADA - the real thing behind these terms, live in days from any browser.