Sensor mounting height is dictated by gas density relative to air (1.204 kg/m³ at 20 °C): heavier-than-air gases such as chlorine (2.99 kg/m³) and sulfur dioxide (2.66 kg/m³) pool at low level and call for sensors at 0.2-0.5 m above the floor, while lighter-than-air gases such as hydrogen (0.08 kg/m³) and ammonia (0.72 kg/m³) accumulate at ceiling level within roughly 0.3 m of the highest point [S2][S4].
Near-neutral-density toxic gases, including carbon monoxide (1.16 kg/m³) and hydrogen sulfide (1.43 kg/m³), mix throughout the occupied volume, so the breathing zone of 1.5-1.8 m above the floor is the recommended mounting band for personnel protection [S2]. For comparison, semiconductor-grade toxicants span 0.08 kg/m³ hydrogen to 12.37 kg/m³ tungsten hexafluoride, a 150x density range that drives the placement logic [S4].
Mounting Height by Gas Density Class
Heavier-than-air toxic gases pool laterally across floors and accumulate in low-lying regions, which is why toxic gas detection systems for chlorine, nitrogen dioxide, and sulfur dioxide should be installed 0.2-0.5 m above the surface, well below typical breathing level [S2]. Industry guidance for propane and gasoline also places sensors within inches of the floor to catch settling plumes before they migrate into occupied zones [S1]. A practical floor-level band of 6-12 in (15-30 cm) is widely applied for refrigerants and heavier hydrocarbon vapors [S8].
Lighter-than-air gases such as hydrogen, methane (0.67 kg/m³), and ammonia rise rapidly and collect in ceiling voids, roof apexes, and beam intersections; detectors for these species should sit within roughly 0.3 m of the ceiling to intercept accumulation before dilution [S2][S4]. Hydrogen and methane detectors are commonly specified at approximately 12 in (30 cm) below the ceiling in fixed-instrument quick-reference tables [S1].
Near-neutral-density toxic gases (CO, HCN 1.13 kg/m³, NO 1.25 kg/m³, H₂S, HCl 1.39 kg/m³, propylene 1.75 kg/m³, CO₂ 1.84 kg/m³) do not stratify cleanly, so placement tracks human exposure: the breathing zone of 1.5-1.8 m above floor, equivalent to 4-6 ft in imperial practice [S1][S2]. When in doubt, treat the gas as breathing-zone and verify with a smoke-tracer or computational fluid dynamics (CFD) check, especially where temperature gradients or forced ventilation distort the natural density behaviour.
Horizontal Spacing, Source Coverage, and Dead Zones
Horizontal spacing sets the area each detector can guard. In open spaces with low air velocity, a single fixed detector typically covers a 5-7 m radius, giving a useful grid baseline of roughly 80-150 m² per point [S2]. For personnel-safety monitoring near a probable leak, detectors should be sited no more than 5 m (about 16.5 ft) from a credible release point such as a valve, pump, compressor, or flanged joint [S3].
Source-oriented placement, with detectors installed approximately 1 m from high-probability leak points, is consistently recommended over uniform grid spacing because it shortens the path between a release and the sensor head [S2]. Perimeter placement along escape routes and ventilation pathways complements source coverage, and structural obstructions (beams, racking, large machinery) must be mapped because they create stagnant zones where plumes can pool unseen [S2]. A practical rule is to add detectors wherever airflow modeling or smoke visualization reveals eddies or dead-air pockets adjacent to the primary grid.
Site Assessment, Environmental Limits, and IP Rating
Before mounting, walk the site to identify storage, processing, and generation points for the target toxic gas, then catalog probable discharge points including valves, pipes, fittings, and ignition sources, because these are where early detection delivers the most value [S1]. Sensors need an unobstructed line of sight to those hazard sources; placing a head behind large equipment or storage containers can leave a release entirely unmonitored until concentrations reach dangerous levels in the breathing zone [S1].
Environmental qualification is non-negotiable: install only where temperature, humidity, and airborne particulates stay inside the instrument's published spec, and verify the enclosure's Ingress Protection (IP) rating against dust and water exposure on site [S3]. Avoid mounting near ventilation intakes, doorways, or known drafts, since airflow can sweep a passing plume past the sensor without raising the reading [S1][S3]. A practical acceptance criterion is to require a documented IP rating (commonly IP65 or higher for industrial plants) plus a temperature band matched to the worst-case seasonal condition in the monitored room.
Wiring, Sampling Lines, and Extractive-System Pitfalls
Wiring and tubing must follow the product manual's maximum cable and tube lengths and physical dimensions, because exceeding those limits degrades signal integrity or adds transport delay on extractive systems [S3][S4]. Sampling-line best practice is to keep tubing runs short and direct, slope horizontal runs gently, and keep vertical runs straight, which together minimize both response delay and condensation pooling [S4]. Honeywell Vertex, Midas M, Midas S2, and ACM150G2 instruments feature an automated line-integrity test that detects leaks or blockages in the sample line, a useful safeguard where tubing traverses long distances [S4].
Extractive systems introduce two well-known failure modes: sampling delay, which can be reduced by minimizing transport distance, and moisture build-up, where condensation in the tubing drifts sensor readings and shortens sensor life [S4]. Regular maintenance checkpoints should confirm that seals are intact and that no condensation has collected at low points, because a small leak or a water slug can mask a real release and silence an alarm that should have fired.
Bump Testing, Calibration, and Commissioning
Every fixed toxic gas detector must be bump-tested and calibrated with a certified gas before being placed in service, and this is true for both initial commissioning and any return-to-service event after sensor replacement or board repair [S5]. A standard sequence is: power on and allow the manufacturer-stated warm-up, expose the sensor to a known concentration of certified span gas, confirm the reading reaches the target threshold, then lock calibration and record the result against the unit's serial number and date [S5].
For personnel-safety monitors, the bump-test gas concentration should be high enough to trigger the configured alarm setpoints but within the sensor's specified range, and the response time should fall inside the manufacturer's published T90 figure [S5]. The commissioning record should capture gas type, concentration, response, calibration date, and the next due date; without that audit trail, a sensor may stay in service long past its calibration interval and silently drift out of spec. A practical acceptance criterion is a post-calibration reading within the manufacturer's tolerance band (commonly ±10% of applied gas) and an alarm-trip test at the configured thresholds.
Coverage Comparison: Density Class, Mounting Band, and Spacing
For quick selection, the three density classes differ on three decision criteria: mounting height, horizontal spacing, and the dominant placement strategy [S1][S2][S3]. Heavier-than-air gases (Cl₂ 2.99 kg/m³, SO₂ 2.66 kg/m³, NO₂ 1.91 kg/m³, CO₂ 1.84 kg/m³) mount at 0.2-0.5 m above floor with a 5-7 m radius and source-oriented placement near valves and flanges, because pooling at low level is the dominant failure mode [S2][S4].
Lighter-than-air gases (H₂ 0.08 kg/m³, NH₃ 0.72 kg/m³, CH₄ 0.67 kg/m³) mount within 0.3 m of the ceiling with a 5-7 m radius, with extra attention to ceiling voids, beam intersections, and roof apexes where stratification can hide a release from a low-mounted sensor [S2][S4]. Near-neutral-density toxic gases (CO 1.16 kg/m³, H₂S 1.43 kg/m³, HCN 1.13 kg/m³, NO 1.25 kg/m³) mount in the breathing zone of 1.5-1.8 m, with a 5 m maximum distance from probable leak points, and the dominant strategy is personnel-exposure coverage along escape routes and occupied work areas [S2][S3].
When Not to Install, and When to Escalate
Do not install a fixed toxic gas detector in locations where environmental conditions exceed the instrument's published specification, where the IP rating is insufficient, or where physical damage from routine operations is likely, because in each case the head cannot deliver the reliability the safety case requires [S3]. Also avoid mounting directly in ventilation intake paths, in dead-air pockets behind large equipment, or in spaces where the operator cannot see or hear the alarm, since detection without a clear human path to action is functionally inert [S1][S3].
Escalate to a specialist installer, an instrument engineer, or the manufacturer whenever the monitored space has unusual geometry, strong forced ventilation, elevated temperature, or a gas whose density sits very close to 1.0 kg/m³, because under those conditions neither the simple density rules nor the standard 5-7 m radius is reliable on its own [S2]. For broader safety-system context, the heat detector selection spec-first guide and the heat detector certification checklist for electrical rooms cover the parallel fixed-detector discipline for thermal hazards, while the oxygen, hydrogen, and cryogenic gas temperature limit compatibility specs address adjacent cryogenic-and-oxygen exposure cases that often share a detector panel with a toxic-gas loop.
Two trackable signals close this out: the next bump-test or calibration due date on every commissioned unit, recorded against serial number, and a quarterly airflow review (smoke pencil or CFD re-check) at any detector location whose mounting band is closer than 1 m to a ventilation supply, a doorway, or a structural pocket, because those are the locations where the published 5-7 m radius most often fails in practice. For a broader look at the wider gas-detection family that includes combustible gas detector and fixed gas detector hardware classes, the encyclopedia entries cover the toxic gas detector baseline, while the related gas detector overview ties the toxic-only rules back into the full sensing portfolio.