Combustible Gas Detector

What is a Combustible Gas Detector

A combustible gas detector is a safety instrument that continuously samples the surrounding atmosphere and converts the concentration of flammable gas or vapour into an electrical signal, an alarm contact, and a numeric reading. Unlike a process gas analyser, whose job is measurement accuracy, a combustible gas detector exists to prevent a fire or explosion, so it is engineered to fail toward a safe and visible state rather than to read silently. It sits in the Safety and Protection family alongside flame detectors, toxic gas detectors and oxygen monitors, and in most plants it is wired into a gas detection panel that drives alarms, ventilation, and emergency shutdown.

The reading is expressed as a fraction of the lower explosive limit, written percent LEL. The lower explosive limit is the minimum concentration of a gas in air that will propagate a flame if ignited. Below the LEL the mixture is too lean to burn, and above the upper explosive limit (UEL) it is too rich. A combustible gas detector therefore does not need to read the full flammable range. It only needs to give early, reliable warning as the atmosphere climbs from clean air toward the LEL, which is why the standard scale runs from 0 to 100 percent LEL, where 100 percent LEL is the LEL itself, not a 100 percent gas atmosphere.

Typical action thresholds follow a widely used convention: a first alarm at around 20 percent LEL and a second, higher alarm at around 40 to 60 percent LEL, with automatic shutdown of ignition sources and isolation of the gas source on the high alarm. For confined-space entry with portable instruments the limits are stricter: many entry procedures permit work below 10 percent LEL, sound a warning at 10 percent LEL, and require evacuation at 25 percent LEL or above. These numbers are policy choices set by the site and its regulator, but the detector must be fast and stable enough to make them meaningful.

The hazard the detector guards against is severe. A methane-air mixture only needs to reach about 5 percent by volume to become ignitable, and a confined ignition of a few cubic metres of gas can destroy a building. Because the consequence is catastrophic and the gas is often invisible and odourless at industrial scale, combustible gas detection is treated as a layer of protection in its own right, governed by functional safety standards and subject to mandatory periodic testing. A detector that quietly stops responding is more dangerous than no detector at all, which is why fail-to-safe design, poison resistance, and a disciplined calibration record dominate engineering selection.

Four engineering attributes determine the quality of a combustible gas detector across its service life: the sensing technology and its immunity to poisons and oxygen loss, response time (T90), long-term stability between calibrations, and hazardous-area and functional-safety certification. These four drive the total cost of ownership far more than the purchase price. A cheap pellistor that must be bump tested every month and replaced every two years can cost more in labour and downtime than an infrared detector that holds calibration for a year and lasts a decade.

Detector Types and Form Factors

Combustible gas detectors are classified first by deployment form factor, because that decides who is protected and how the instrument is powered, mounted and maintained. The four mainstream form factors are fixed point detectors, portable and personal detectors, open-path beam detectors, and sample-draw (aspirated) systems. Choosing the wrong form factor is a system-level mistake that no sensor quality can compensate for: a single portable instrument cannot protect an unmanned compressor station, and a row of point detectors cannot economically watch a 150 metre tank-farm boundary.

Fixed point detectors are permanently installed near a credible leak source such as a flange, pump seal or valve, and monitor the atmosphere by diffusion. They are the backbone of plant gas detection: continuously powered, wired to a control panel, and certified for the hazardous area. A fixed detector typically combines a sensor head with a flameproof or intrinsically safe transmitter that outputs 4-20 mA, Modbus, HART or a fieldbus signal. Honeywell Sensepoint XCD and Dräger Polytron are representative fixed-point platforms that accept either catalytic or infrared sensor cartridges.

Portable and personal detectors protect people rather than places. A worker carries or wears the instrument, which alarms locally with sound, light and vibration. Single-gas LEL units and multi-gas instruments (combining LEL with oxygen, carbon monoxide and hydrogen sulphide) are the standard tools for confined-space entry, hot-work permits and leak surveys. Because they move with the wearer, they need fast response, robust batteries and frequent bump testing, but they cannot provide unattended area coverage.

Open-path beam detectors project an infrared beam between a transmitter and a receiver, or to a retroreflector and back, and measure the total gas along the line of sight in LEL.metres. One beam up to about 200 metres can replace a line of point detectors along a perimeter or pipe rack, and is far more likely to intersect a drifting cloud outdoors. The Honeywell Searchline Excel is a widely installed open-path detector with selectable path lengths from roughly 5 to 200 metres and a beam T90 of under 3 seconds.

Sample-draw systems use a pump to pull a continuous gas sample through tubing from a point that a diffusion detector cannot reach, such as a duct, a sealed cabinet or a sub-floor void, into a remote sensor cell. They add response delay equal to the transport time through the tubing and require filters and flow monitoring, but they make otherwise inaccessible spaces measurable. The choice among these four form factors is driven by whether the hazard is a known fixed source, a moving worker, a large open area, or an enclosed inaccessible volume.

Sensing Technologies

Beneath the form factor sits the sensing technology, and this choice determines which gases the detector can see, whether it needs oxygen, and how it fails. Three technologies dominate combustible gas detection: the catalytic bead pellistor, the non-dispersive infrared (NDIR) point sensor, and the open-path infrared beam. Semiconductor (metal-oxide) sensors appear in low-cost consumer alarms but are rarely specified for industrial safety. The table below compares the three industrial technologies on the metrics that decide selection.

Catalytic bead (pellistor) sensors work by catalytic oxidation. A platinum coil is embedded in a ceramic bead coated with an oxidation catalyst and heated to roughly 500 degrees C. When flammable gas reaches the active bead it burns on the catalyst surface, the heat of combustion raises the bead temperature, and the platinum coil resistance changes. A matched inert bead with no catalyst forms the other half of a Wheatstone bridge, so the bridge output is proportional to gas concentration and compensated for ambient temperature and humidity. The pellistor responds to almost any flammable gas, including hydrogen, which makes it broadly useful and inexpensive.

The pellistor has two structural weaknesses that drive its maintenance burden. First, because it works by combustion, it requires oxygen and cannot operate in inert or oxygen-displaced atmospheres, and a sustained exposure well above 100 percent LEL can overheat and permanently damage the bead, often making it read low afterward. Second, it is vulnerable to poisoning by silicones, sulphur compounds, lead and phosphates, which coat the catalyst so the sensor under-reads or goes blind while still appearing powered and healthy. These two failure modes are why pellistors demand routine bump testing and why poison-resistant formulations and a flame arrestor (which also gives the flameproof rating) are standard.

Non-dispersive infrared (NDIR) point sensors exploit the fact that hydrocarbon molecules absorb infrared light at characteristic wavelengths, typically around 3.3 to 3.4 micrometres for the carbon-hydrogen bond. An infrared source shines through the gas to a detector behind an optical filter at the absorbing wavelength, and a second reference wavelength that the gas does not absorb compensates for dirt, ageing and fog. The ratio of the two gives a stable, drift-resistant concentration reading. Because nothing is consumed, an infrared sensor does not need oxygen, is immune to poisoning, survives exposure above 100 percent LEL, and typically holds calibration far longer than a pellistor.

The infrared technology has one hard limitation: it can only detect gases that absorb infrared, which excludes hydrogen and other diatomic molecules. It also responds differently to different hydrocarbons, so an NDIR detector calibrated for methane will read inaccurately on propane unless cross-calibrated. Representative infrared point detectors include the Honeywell Searchpoint Optima Plus and the Dräger PIR 7000 sensor used in the Polytron 8700 transmitter, both designed for hydrocarbon monitoring in oil, gas and offshore service.

Open-path infrared applies the same absorption principle across a long beam instead of a small cell. A transmitter projects collimated infrared toward a receiver tens to hundreds of metres away, and the instrument integrates absorption over the whole path, reporting the result in LEL.metres so that, for example, 1 LEL.m could be 100 percent LEL over one metre or 10 percent LEL over ten metres. This integration is exactly what makes open-path strong for area coverage and weak for pinpointing a source. Open-path shares infrared's immunity to poisons and oxygen loss and its blindness to hydrogen, and adds a dependence on a clear, stable line of sight.

Target Gases, LEL Data and Standards

A combustible gas detector is always calibrated to a specific target gas, and selecting the wrong calibration gas, or applying one gas calibration to a different gas, is a frequent and serious error. The lower explosive limit varies widely between gases, so 100 percent LEL means a very different absolute concentration for hydrogen than for propane. The table below lists published flammability data for common flammable gases. Values are nominal at atmospheric pressure and room temperature and follow NFPA 497 and IEC 60079-20-1; always confirm against the detector maker chart and the safety data sheet for the actual gas, because published figures vary slightly between sources.

The data above explains two practical rules. First, hydrogen and any non-absorbing gas can only be detected by a catalytic bead or an electrochemical hydrogen sensor, never by infrared, so a hydrogen hazard forces the technology choice. Second, the wide span of LEL values means a detector calibrated for one gas misreads another: a methane-calibrated infrared detector exposed to propane gives an incorrect percent LEL because the absorption strength and the LEL both differ. When a process can release several flammable gases, engineers either calibrate to the most conservative gas, fit gas-specific detectors, or use a pellistor whose broad response covers the mixture, accepting reduced accuracy.

Performance is not a matter of opinion: it is defined by IEC 60079-29-1, the international standard for the performance requirements of detectors for flammable gases, adopted as EN 60079-29-1 in Europe and as ANSI/ISA-60079-29-1 (which superseded the older ISA 12.13.01) in North America. This standard sets the accuracy, response time, repeatability, and environmental test regime a certified detector must pass, and crucially it treats lower flammable limit (LFL) and lower explosive limit (LEL) as synonyms, as it does upper flammable limit and upper explosive limit. A detector that carries third-party 60079-29-1 certification has been tested against a defined, repeatable benchmark rather than a maker claim.

Three companion standards complete the framework. IEC 60079-29-2 covers selection, installation, use and maintenance, including calibration discipline. IEC 60079-29-3 guides the functional safety of fixed gas detection systems, and IEC 60079-29-4 sets performance requirements specifically for open-path detectors, which is why beam instruments are tested differently from point detectors. Layered on top, the broader IEC 60079 series defines the explosion protection of the enclosure itself (flameproof Ex db, intrinsic safety Ex ia and others), and IEC 61508 governs the functional safety integrity (SIL) when the detector forms part of a safety instrumented function. A complete specification cites all the parts that apply, not just one.

Key Specification Parameters

Reading a combustible gas detector datasheet is a core procurement skill. A datasheet may list twenty or more lines, but only a handful truly drive the safety case and the selection decision: measuring range and scale, accuracy, response time, drift and long-term stability, operating temperature and humidity, hazardous-area protection, functional safety integrity (SIL), and output signal. Each is explained below, with the typical values a certified industrial detector should meet.

Measuring range and scale is the first line to confirm. Point detectors normally read 0 to 100 percent LEL of a named gas; open-path detectors read 0 to 5 LEL.metres; and detectors used for inerting or pipeline work may instead read 0 to 100 percent by volume. Mixing these scales is dangerous, so the datasheet must state both the range and the unit, and the target gas the calibration assumes. A 0 to 100 percent LEL methane detector and a 0 to 100 percent volume methane detector look identical on a label but differ by a factor of twenty.

Accuracy is typically quoted as a percentage of full scale, with values around plus-or-minus 2 to 3 percent LEL common for catalytic and infrared point detectors when measured under the IEC 60079-29-1 regime. Accuracy interacts with the alarm threshold: a detector accurate to plus-or-minus 3 percent LEL with a 20 percent LEL alarm has comfortable margin, but the same accuracy matters far more near a low entry limit. Always check whether the quoted accuracy is at calibration conditions or across the full temperature range.

Response time is given as T90, the time to reach 90 percent of the final reading after a step change of gas, as defined in IEC 60079-29-1. Diffusion pellistor point detectors typically reach T90 in 20 to 30 seconds, modern infrared point detectors in under 10 seconds, and open-path beam detectors in about 3 seconds. Splash guards, dust filters and sintered flame arrestors lengthen the installed T90 beyond the bare-sensor figure, so confirm the value applies to the detector as it will be installed, with its weather protection fitted.

Drift and long-term stability separate technologies sharply. Catalytic beads drift and may be poisoned, with annual zero and span drift often a few percent LEL per year, which is why they are bump tested every 30 to 90 days. Infrared sensors are far more stable, frequently holding calibration for 6 to 12 months. Stability directly drives the maintenance labour cost over the detector life, so it belongs in the total-cost calculation, not only the safety case.

Hazardous-area protection and functional safety are mandatory for fixed plant detectors. The enclosure carries an explosion-protection marking such as Ex db (flameproof) or Ex ia (intrinsically safe) under the IEC 60079 series, certified by ATEX (EU 2014/34/EU), IECEx, NEPSI in China, or FM and CSA in North America, with a defined gas group and temperature class. Where the detector is part of a safety instrumented function, it carries a SIL rating assessed to IEC 61508, with SIL 2 common for fixed combustible gas detection. The certificate must cover the exact target gas and range, not merely the hardware family.

Output signal and interface link the detector to the gas detection panel or control system. Common options are listed below:

• 4-20 mA: Two- or three-wire analogue current loop, the most widespread interface, with fault and inhibit signalled by current levels outside the 4 to 20 mA band.
• 4-20 mA + HART: Adds digital configuration, diagnostics and asset management over the same current loop.
• Modbus RTU (RS-485): Digital bus letting many detectors share one cable pair into a controller, common on packaged gas detection systems.
• Relay / alarm contacts: Dedicated dry contacts for local alarm, ventilation start and shutdown, often in addition to the analogue output.
• Fieldbus / Ethernet (PROFINET, EtherNet/IP, HART-IP): Used on large integrated safety systems and digital plants.

Selection Decision Factors

To convert the preceding chapters into a specific model, follow the decision sequence below. Most selection mistakes come not from a single wrong line on a datasheet but from deciding details before the higher-level questions are settled. These eight steps double as a fixed RFQ template for combustible gas detection.

1. Target gas and hazard: First identify every flammable gas or vapour that could be released and its LEL. A hydrogen or other non-absorbing gas forces a catalytic bead or dedicated sensor; a hydrocarbon allows infrared. Where several gases are possible, decide whether to calibrate to the most conservative, fit gas-specific detectors, or accept broad pellistor response.
2. Form factor and coverage: Decide fixed point at a known source, portable for personnel, open-path for large outdoor areas, or sample-draw for inaccessible volumes. Coverage is a system question that sensor quality cannot fix.
3. Sensing technology: Choose pellistor for hydrogen, broad gas coverage and low cost where oxygen is present and poisons are absent; choose infrared for hydrocarbons, oxygen-free or oxygen-displaced atmospheres, poison-laden environments, and low maintenance. Confirm the technology can survive credible over-range exposure.
4. Range, scale and alarm thresholds: Confirm 0 to 100 percent LEL, 0 to 5 LEL.m, or 0 to 100 percent volume, and set first and second alarm levels (commonly 20 percent and 40 to 60 percent LEL) per the site risk assessment and regulator.
5. Hazardous-area certification: Match the area classification to the protection method (Ex db or Ex ia), gas group and temperature class, certified under ATEX, IECEx, NEPSI or FM/CSA. Verify the certificate covers the actual target gas and range.
6. Functional safety (SIL): If the detector is part of a safety instrumented function, specify the required SIL (SIL 2 is common) with an IEC 61508 assessment and a safety manual, including proof-test interval and diagnostic coverage.
7. Environment and ingress protection: Confirm operating temperature and humidity, ingress rating (commonly IP66 or IP67 for outdoor and washdown), and resistance to wind, fog, rain or snow for open-path beams along the intended line of sight.
8. Output, panel and total cost of ownership: Match the output (4-20 mA, HART, Modbus, relays or fieldbus) to the gas detection panel, then add bump testing, calibration gas, sensor replacement and downtime to the purchase price. A low-cost pellistor with monthly bump tests can exceed the lifecycle cost of an infrared detector that holds calibration for a year.

One dimension that purchasing often overlooks is serviceability and the calibration regime: how easily the sensor cartridge can be replaced in the field, whether calibration gas and regulators are readily available, how the maker supports proof testing, and whether a calibration and bump-test log is maintained. Many regulators and insurers treat a missing calibration record as an uncontrolled hazard, so a detector that is cheap to buy but awkward to verify can quietly become non-compliant. Established platforms from Honeywell (Sensepoint, Searchpoint Optima, Searchline Excel), Dräger (Polytron, PIR 7000), MSA Safety, Det-Tronics and Crowcon maintain spare-cartridge supply and documented calibration procedures, which is why they recur on large project specifications.