How chamber air actually gets tested

1. What the laboratory actually measures

A breathing-air test is a defined parameter set, not a general "is this air clean" verdict. One A2LA-accredited laboratory publishes its parameter set and its sampling media per test. For medical air purity plus CGA G-7.1 Grade D, the limits it reports against are total hydrocarbons 25 ppmv (methane included), halogenated hydrocarbons 2 ppmv, carbon monoxide 10 ppmv, carbon dioxide 500 ppmv, water 1,544 ppmv, dew point 35 °F, oxygen 19.5–23.5 %, and oil mist plus particulate 5 mg/m³ (Trace Analytics, medical air and gas testing).

Two details on that page are worth carrying into a conversation with a laboratory. Dew point is reported at 50 psig, not at atmospheric pressure — a dew point figure without a pressure reference is not a specification. And the sampling media are named per test: SB is a source bottle, F is a filter, DT is a water-vapour detector tube, DT2 is sulphur dioxide and NOx detector tubes. The Grade D medical air test uses "SB + F," which is why a single small package can carry both the gas and the particulate result.

On analytical method, a second laboratory states plainly that it "uses a gas chromatograph (GC) as the most appropriate method for the analysis of gaseous breathing air components," describes itself as an AIHA-accredited laboratory, and states that "all instrumentation calibration is performed per ISO/IEC 17025" (TRI Air Testing). For oxygen specifically, paramagnetic and laser analyzers are described as "very accurate and reliable when calibrated correctly," with no numeric accuracy published, and they detect only the identification and strength of oxygen (Trace Analytics). If you want a stated accuracy figure for the oxygen result on your report, you will have to ask for it — it is Not published.

2. How the sample is physically taken

A field sampling kit, not a site visit. The contents of one published low-pressure kit are an adaptor (SCUBA or ¼-inch NPT), a flowmeter, a bottle holder, a needle replacement tool, spare needles, tubing, and instructions (AirCheck Kit K201). The same page states samples may be drawn from the compressor, the storage, or the outlet, and names typical users as commercial diving, industrial airline respirators, and low-pressure compressors with a refrigerated dryer or no dryer — which is exactly the configuration class a chamber supply falls into.

For hyperbaric facilities the same laboratory names a specific kit: K902 "is used to obtain routine samples for Medical Air and CGA Grade D testing for Class A or Class B hyperbaric chambers." It states that samples "can be taken by facility maintenance personnel," that they are "simple and straight-forward," that reports are emailed within 1–3 business days of receipt, that a failing sample triggers a phone or email notification, and that hyperbaric-facility reports are accessible online (Trace Analytics).

An independent description of the identical workflow appears in NFPA's own ballot record for the 2024 hyperbaric TIA: "The Restore Hyper Wellness locations can order an air sampling kit from a laboratory. The air sampling kit is attached to the compressor and obtains a sample of the air. The sample is then sent back to the laboratory…" (NFPA 99-2024 TIA Log 1735 ballot record). That is a useful corroboration that the postal-kit method is the normal practice in this industry, described in an NFPA document, rather than a shortcut.

3. What accreditation to look for

This is the part of the transaction where the money buys credibility rather than hardware, so it is worth being specific. One laboratory publishes that it is A2LA-accredited, complying with ISO 17025, certificate #0322-01 (Trace Analytics). Another publishes AIHA-LAP accreditation and ISO/IEC 17025 calibration compliance, with a stated 24-hour standard analysis turnaround (TRI Air Testing). Ask which accreditation body, ask for the certificate number, and ask whether the specific test you are buying is inside the accredited scope. A laboratory that answers those three in writing is giving you a report an inspector can lean on.

4. What it costs

Kit prices are published by at least one vendor. Read the caveat under the table before you budget from it.

The useful framing: on a six-month sampling regime you are buying roughly two kits a year, in the low hundreds, against a capital cost in the thousands for the supply system and its monitoring. Testing is the cheapest line in the whole programme and the only one that produces evidence. If a budget conversation ever puts sampling on the chopping block, it is cutting the wrong thing.

5. The interval — and this is where most pages get it wrong

6. What OSHA does mandate instead

It is worth reading, because the duties are real and they are engineering duties rather than paperwork ones (29 CFR 1910.134(i)):

• (i)(4)(iii) — cylinder moisture content such that dew point does not exceed −50 °F at one atmosphere.
• (i)(5)(ii) — compressors shall "minimize moisture content so that the dew point at 1 atmosphere pressure is 10 degrees F (5.56 deg. C) below the ambient temperature."
• (i)(5)(iii) — "suitable in-line air-purifying sorbent beds and filters."
• (i)(5)(iv) — a tag at the compressor showing the most recent change date and the signature of the person authorised to certify the change.
• (i)(6) — for non-oil-lubricated compressors, carbon monoxide in the breathing air shall not exceed 10 ppm.
• (i)(7) — for oil-lubricated compressors, "the employer shall use a high-temperature or carbon monoxide alarm, or both." If only a high-temperature alarm is used, the air supply must be monitored at intervals sufficient to keep CO below 10 ppm.

Notice the shape of that list: a dew point tied to your ambient temperature rather than to a fixed number, filtration, a signed tag, and continuous alarms. It is a design-and-monitor rule. A twice-yearly laboratory sample satisfies none of it, and none of it substitutes for the sample. You need both, and the monitoring instruments page covers the alarm side.

7. An honest word about how settled this is

From NFPA's own ballot record on the 2024 hyperbaric TIA, one balloter objected that "no requirements for the quality of the compression gas are specified. … How do you determine if the air is safe if no standard is mentioned?" and another wrote that the proposal "does not list any requirements for air compressor piping, active sample monitoring or quality of the air to be used for chamber pressurization" (NFPA 99-2024 TIA Log 1735 ballot). Those are individual committee members' recorded comments during a ballot, not code text and not formal NFPA interpretations.

We include them because the regulatory picture in this corner of the market is genuinely less settled than most vendor pages imply, and pretending otherwise would make everything else on this site less believable. The six-month interval and the Grade E limit are the two firm things. Build your programme on those, document what you did, and hand the reports to your authority having jurisdiction.