How chamber air is tested
What a laboratory sample measures, how the sample is drawn, what accreditation to look for, and what the interval really is.
Guide
What has to alarm, at what number, and what the instrument costs
A twice-yearly laboratory sample tells you what the air was like on one morning in March. Continuous monitoring is what tells you about the other 364 days. Here are the published setpoints and the instruments that meet them, with prices where the vendors publish them.
1. The setpoints, and where each one comes from
| Parameter | Setpoint | Applies to | Source | Tier |
|---|---|---|---|---|
| Oxygen, high | Alarm above 23.5 % | Chamber atmosphere | NFPA 99 §14.2.9.4.2.1, cited as HBOV 10.1 (UHMS, 4th Ed.) | Code requires |
| Oxygen, range | Below 19.5 % or above 23.5 % is a hazardous atmosphere | Occupied spaces generally | 29 CFR 1910.146; the same 19.5–23.5 % band is the oxygen limit on a Grade D medical air report (Trace Analytics) | Code requires |
| Combustible gas | Alarm at 10 % of the lower explosive limit | Chamber atmosphere | Accreditation criterion HBOV 8.1 (UHMS, 4th Ed.) | Body recommends |
| Carbon monoxide | Not to exceed 10 ppm in breathing air | Supplied-air respirators, non-oil-lubricated compressors | 29 CFR 1910.134(i)(6); also the CO limit on a Grade D report | Code requires |
| CO or high temperature | An alarm on one or both is mandatory | Oil-lubricated compressors | 29 CFR 1910.134(i)(7) | Code requires |
| Dew point | 10 °F below ambient at 1 atm for compressor air; −50 °F for cylinder air | Supplied-air respirators | 29 CFR 1910.134(i)(5)(ii) and (i)(4)(iii) | Code requires |
| Carbon dioxide | 500 ppm in the supply air on a Grade D medical air report | Supply air, not chamber atmosphere | Trace Analytics | Body recommends |
| Ventilation rate | Class A 3 ft³/min per occupant; Class B 1 ft³/min | Chamber ventilation — the control that keeps CO₂ down | NFPA 99 §§14.2.4.1.1 and 14.2.4.4.1 (UHMS, 4th Ed.) | Code requires |
One definition worth having to hand, because it is what the 23.5 % figure protects against: NFPA 99 §3.3.137 defines an oxygen-enriched atmosphere as an atmosphere containing more than 23.5 % oxygen by volume, per the same accreditation manual.
A withdrawn criterion, so you do not budget for the wrong thing
The UHMS Fourth Edition deleted HBOV 12.1, which had required continuous monitoring "for volatized hydrocarbons and carbon monoxide at a location downstream from the oil filter when the compressor is running" (UHMS accreditation manual, 4th Ed.). Continuous downstream hydrocarbon monitoring is therefore not a current accreditation criterion. Continuous CO monitoring on an oil-lubricated compressor remains an OSHA duty under 1910.134(i)(7) for respirator service. If a proposal quotes HBOV 12.1 at you, it is quoting a withdrawn item.
2. Carbon monoxide monitors
The two published options at opposite ends of the price and capability range.
| Instrument | Range | Accuracy | Alarms | Sample handling | Price |
|---|---|---|---|---|---|
| ENMET CO-GUARD, P/N 03481-005 | 0–50 ppm CO | ±10 % of reading | Three: 5, 10 and 20 ppm, adjustable across 1–50 ppm | Internal regulator brings supply down to 55 psi; sample flow about ½ L/min | Not published |
| GfG RAM 4021, P/N 76334-812 | CO, airline monitor | Not published on the listing | Not published on the listing | Requires a 0.5 LPM flow regulator, sold separately | $2,470.22 |
| GfG RAM 4021 panel version, P/N 76334-816 | As above, panel-mounted | Not published | Not published | As above | $8,727.79 |
Sources: ENMET CO-GUARD literature; VWR listing for the GfG RAM 4021. Note the ENMET unit's three-level alarm structure with a factory 5/10/20 ppm arrangement — a warning well below the 10 ppm limit, the limit itself, and an escalation. That is the right shape for a supply-air alarm, and it is worth asking any vendor whether their single-setpoint monitor can do the same.
The recurring cost nobody budgets for. Published from the same listing: calibration kit $460.13; CO calibration gas at 20 ppm in a 34-litre cylinder $94.22; CO calibration gas at 10 ppm, also $94.22; 0.5 LPM flow regulator $266.05 (VWR). An electrochemical CO sensor with no calibration history is a decoration. Budget the gas and the regulator alongside the instrument, and put the calibration on the same schedule as the six-monthly laboratory sample so both happen in one visit.
3. Multi-parameter medical air monitors
If you would rather buy one enclosure than four instruments, this class exists. One published unit monitors all four parameters that matter here (ENMET MedAir 2200 datasheet):
| Carbon monoxide | 0–50 ppm, alarms at 10 and 20 ppm |
|---|---|
| Oxygen | 0–30 %, alarms at 19.5 % and 23.5 % — the same band as the code and the laboratory report |
| Carbon dioxide | NDIR sensor, 0–5,000 ppm, alarms at 500 and 1,000 ppm |
| Dew point | −112 to +68 °F, alarms at −40 °F and +39 °F |
| Listing | UL 60601-1 |
| Accuracy, each channel | Not published on the datasheet |
| Price | Not published |
The 500 ppm CO₂ alarm matches the Grade D supply-air limit exactly, and the oxygen alarms match both 19.5 % and 23.5 %. That alignment is the argument for this class of instrument. The missing accuracy figures are the argument for asking before you buy — ask for the accuracy on the CO and O₂ channels in writing, since an alarm at 23.5 % means little if the channel accuracy is ±2 %.
4. Dew-point transmitters
A representative published instrument: typical measurement range −70 to +60 °C (−94 to +140 °F) dew point, accuracy ±2 °C (±3.6 °F) in air or nitrogen, operating pressure 0–50 bar(a) / 725 psia, and measurement in air, nitrogen, hydrogen, argon, helium or oxygen (Vaisala DMT143 specification table). Four caveats from the manufacturer's own pages decide whether it suits you: below 0 °C it outputs frost point rather than dew point, it is not certified for hazardous or explosive atmospheres, it was not leak-tested for small-molecule gases, and its alarm capability is limited to an "LED alarm for exceeded dew point level" (specification table; DMT143 product page). That last point matters: it is a measurement device, not a supervisory alarm, so it does not by itself satisfy an alarm duty. Price is Not published.
Put that ±2 °C next to the OSHA duty. If your ambient is 75 °F, "10 °F below ambient" is 65 °F, and ±2 °C is ±3.6 °F — comfortable margin. If you are chasing the −48 °C practical figure from the society literature, ±2 °C is a large fraction of your remaining headroom. Match the instrument to the number you actually have to hit, not to the best number on the market.
5. Carbon dioxide monitors, and a worked example of the wrong instrument
A published unit: NDIR diffusion sampling, range 0–50,000 ppm, accuracy ±300 ppm or ±3 % of measured value, two user-adjustable alarm relays rated 2 A at 24 V, a 90 dB horn with LED, a 4–20 mA output, weekly automatic recalibration to ambient CO₂ with no manual calibration required, a stated 10-plus-year sensor life and a 3-year warranty, at $1,205.00 (PureAire CO₂ Safety Monitor).
This is a good instrument for the wrong job — and the arithmetic shows why
The supply-air CO₂ limit on a Grade D report is 500 ppm. An instrument with a ±300 ppm accuracy floor cannot usefully police a 500 ppm limit: a true reading of 500 could display anywhere from 200 to 800. Compare the NDIR channel on the multi-parameter monitor above, which alarms at 500 and 1,000 ppm and is scaled 0–5,000 ppm — ten times narrower.
Where the 0–50,000 ppm unit is appropriate is monitoring the chamber atmosphere, where accumulated exhaled CO₂ operates in thousands of ppm and ±300 ppm is a small relative error. Same parameter, two jobs, two instruments. We include this example because "we monitor CO₂" is a claim you should always follow with "at what accuracy, and on which side of the compressor."
6. What a monitoring package actually costs
Adding up only the figures published above, a CO-monitoring installation on a lubricated compressor starts around $2,470 for the instrument plus $460 for the calibration kit, $266 for the flow regulator and roughly $94 per calibration gas cylinder — call it about $3,290 before installation, with the panel-mounted version of the same monitor at $8,727.79 instead (VWR). Adding chamber-atmosphere CO₂ at $1,205 (PureAire) brings a two-instrument arrangement to roughly $4,500 in published hardware.
Those totals are our arithmetic on published list prices, not a quotation, and they exclude the instruments whose prices are Not published — the ENMET CO-GUARD, the ENMET MedAir 2200 and the Vaisala transmitter — any of which could change the picture substantially. They also exclude installation, wiring, and any relay integration to shut the supply down rather than merely sound. Treat the range as a floor for planning, and get quotations for the unpriced options before deciding.
The honest summary: monitoring is a meaningful fraction of a supply system's cost, it is mandatory on a lubricated machine, and it is the only part of the installation that can tell you something is wrong while a session is running. It is also the part most often left out of a low quotation. If you are comparing two proposals and one is dramatically cheaper, check the monitoring line first.
Keep reading
Related pages
Contamination and filtration
Where bad air comes from — intake siting, oil carryover, moisture, carbon monoxide — and the filtration stages that address each.
Oil-free vs lubricated
The code permits both. An even-handed read of flow per dollar against failure-mode elimination.