Gas Detector Calibration & Bump Test Procedure
SWMS variants reference your state’s WHS legislation. Instant download after payment.
Every atmospheric decision made at a wastewater confined space rests on one thing: a number on an instrument screen. The entry permit is issued because the reading was clear. Ventilation is stopped because the reading came down. The entrant descends because the meter said it was safe. If the instrument is wrong, none of those decisions mean anything — an uncalibrated or unbump-tested gas detector is not a control, it is a false sense of security. Part 4.3 of the WHS Regulation 2025 (NSW) and AS 2865 require atmospheric testing by a competent person using equipment suitable for the purpose, and an instrument that has not been bump tested before use or calibrated to the manufacturer's instructions is not suitable for the purpose. This procedure is the instrument discipline that sits behind the permit.
The document covers bump testing and calibration of portable multi-gas instruments used for atmospheric testing and continuous monitoring in wastewater confined spaces — sewers, maintenance holes, wet wells, pump station chambers, treatment plant tanks, channels and digesters — for oxygen, hydrogen sulphide and flammable gas (LEL) sensors and any additional sensors fitted. It sets out the definitions people conflate (bump test, calibration, zero/fresh air set, span), the frequency table, an eight-step bump test procedure, a six-step calibration procedure, the wastewater-specific sensor cautions, and three working logs. The scope boundary is stated plainly in the document itself: the **manufacturer's instructions are the primary authority** for calibration intervals, gas concentrations and procedure — where this document and the manual differ, the manual is followed. Authored for New South Wales. Regulator: SafeWork NSW.
Hazards identified
14 hazards covered, sorted by priority.
The most dangerous failure mode in a sewer — the space reads safe, the permit is issued, and the entrant descends into a lethal atmosphere with no warning at all
Nothing confirms the sensors respond or that the alarms sound; a dead sensor and a clean atmosphere look identical on the display
Readings drift with no indication on screen; the number is precise, repeatable and wrong, and every decision built on it is wrong with it
The baseline is set on the contamination, so every later reading is understated by that amount for the whole shift
H2S is heavier than air and pools at the base — a clear reading at the lip says nothing about the invert where the worker is going
The instrument works perfectly and still fails to warn, because it is alarming at the wrong number
The sensor under-reads afterwards with no obvious symptom, so a flammable atmosphere is not detected
Catalytic sensors need oxygen to work — the LEL reading is falsely low in exactly the atmosphere where it matters most
Shock or poisoning can shift a sensor permanently; assuming it recovered is assuming the sensor is honest about its own damage
A known-faulty instrument is talked back into service at the point of maximum consequence, with no record that it ever failed
The bump test or calibration is meaningless, and a sensor that never responds is signed off as passed
A failed sensor carries a current calibration label — the paperwork is clean and the instrument is not
Readings on unrelated sensors are distorted, and the crew has no cross-sensitivity table to interpret them against
The faulty meter is picked up by the next crew, who have no way of knowing why it came back
Control measures
Hierarchy-of-controls order: elimination → substitution → isolation → engineering → administrative → PPE.
- 1Bump test before each day's use, and again before any entry where the instrument has been out of service or handled roughly — if it has not been bump tested today, it is not used today.
- 2Treat bump testing and calibration as different duties: a bump test confirms the instrument works and the alarms sound; only calibration confirms that 9 ppm means 9 ppm. Passing a bump test does not make a reading accurate.
- 3Calibrate in accordance with the manufacturer's instructions — commonly every 6 months, more frequently for hard-service instruments and for some H2S sensors — with the manual as the primary authority over this procedure wherever the two differ.
- 4Calibrate immediately after any failed bump test, before the instrument returns to service, and after any high-concentration exposure, drop, impact, water ingress or suspect reading.
- 5Perform the zero / fresh air set at each start-up in known clean air only — never at the chamber opening, near plant exhaust or near vehicle exhaust — and confirm a stable baseline of approximately 20.9% O2, 0 ppm H2S and 0% LEL before the instrument leaves the yard.
- 6Check the calibration due date, charge, physical condition and sensor service life before the instrument goes to site; a passed due date means it does not go.
- 7Verify the test gas before use: correct mix for the sensors fitted, within expiry, sufficient pressure, correct regulator and tubing, and applied at the manufacturer's specified flow rate.
- 8Confirm every sensor responds toward the certified concentration and that the audible, visual and vibrating alarms all activate, then confirm the alarms clear on return to a clean-air baseline.
- 9Verify alarm set points against your entry criteria at each calibration — O2 low and high alarms, H2S set against the workplace exposure standard, LEL at or below 5%.
- 10Tag out and remove from service immediately on a failed bump test or an out-of-tolerance sensor, and record it in the out-of-service register — never re-try a failed instrument on the day at the chamber.
- 11Replace the sensor or remove the instrument from service where it cannot be brought within tolerance; never adjust a sensor into apparent compliance.
- 12Apply the wastewater sensor cautions in the field: track H2S sensor age as well as calibration date, verify catalytic LEL sensors by calibration after any sulphide-rich atmosphere, read oxygen before interpreting any LEL reading, know your instrument's cross-sensitivity table, sample the full depth with a sample line because H2S pools at the base, and test before ventilating as well as after so you know what the space generates and what it returns to.
- 13Ensure the persons carrying out testing and calibration are competent for the specific instrument, and that all workers hold a current White Card (CPCCWHS1001) where the work is on a construction site, together with confined space entry, standby person, permit issuer and gas detection competencies as applicable.
- 14Consult workers on WHS matters affecting them per Section 47 of the WHS Act 2011 (NSW), record the consultation, and review this procedure whenever the instruments, sensors, manufacturer's instructions or exposure standards change, after any instrument-related incident or suspect reading, and at minimum every 12 months — with SWMS relying on it reviewed under Section 302.
Applicable Codes of Practice
The benchmark for atmospheric testing and continuous monitoring by a competent person using equipment suitable for the purpose — the duty this procedure exists to satisfy.
The management of hydrogen sulphide as a hazardous chemical, including the duty to keep exposure below the workplace exposure standard that the H2S alarm set point is referenced to.
The technical standard requiring atmospheric testing with suitable equipment by a competent person, which this procedure operationalises.
Cited in the document as the guidance for selection, use and maintenance of flammable gas and oxygen detectors, alongside the manufacturer's instructions.
The values the H2S alarm set point is referenced to, and the reason the document flags set points for review ahead of the transition.
Stated in the document as the primary authority for calibration intervals, gas concentrations and procedure — where the document and the manual differ, the manual is followed.
High-Risk Construction Work triggered
Calibration and bump testing are not construction work in themselves — they are carried out in a clean, well-ventilated area away from the space. The category applies to the entry work this procedure supports: the sewers, maintenance holes, wet wells, chambers, tanks, channels and digesters the instrument is taken into.
The atmospheres this instrument is relied on to characterise routinely contain hydrogen sulphide, may be oxygen-deficient and can contain flammable methane. The instrument reading is the evidence on which that category is assessed, which is why its accuracy is a legal question and not a housekeeping one.
This procedure is a supporting document, not a SWMS, and calibration itself is not high risk construction work. Its duties sit elsewhere and apply regardless of construction status. Part 4.3 of the WHS Regulation 2025 (NSW) requires atmospheric testing and monitoring for every confined space entry, carried out by a competent person using equipment suitable for the purpose — an instrument that has not been bump tested before use or calibrated to the manufacturer's instructions does not meet that description, so the testing duty is not discharged no matter how carefully the permit was filled in. Separately, hydrogen sulphide is a hazardous chemical under Part 7.1 and a PCBU must ensure no person is exposed above the workplace exposure standard; an alarm set point that is wrong, or a sensor that reads low, means the PCBU has no reliable basis for asserting that duty was met. Where the entry work is high risk construction work under Section 291 the SWMS required by Section 299 will reference this gas detection regime, and reviewing that SWMS under Section 302 reaches this procedure with it. An H2S exposure, an asphyxiation or a confined space incident causing death or serious injury is prosecuted as a Category 1 or Category 2 offence under the WHS Act 2011 (NSW), and an instrument log is the first record a regulator asks for.
Who this is for
- →Local councils and water utilities whose crews rely on portable multi-gas instruments for sewer, pump station and treatment plant entry.
- →Confined space entrants, standby persons and permit issuers who make an entry decision on an instrument reading.
- →Store, workshop and depot personnel responsible for calibrating, bump testing, tagging and issuing gas detection equipment.
- →Supervisors accountable for instruments being in calibration, test gas being in date, and failed instruments being tagged out and removed.
- →WHS managers and HSE advisors who have to demonstrate to SafeWork NSW that atmospheric testing was carried out with equipment suitable for the purpose.
What you receive
- ✓A purpose and scope section tying the procedure to Part 4.3 of the WHS Regulation 2025 (NSW), AS 2865 and AS/NZS 60079.29.2, covering O2, H2S and LEL sensors and any additional sensors fitted, across sewers, maintenance holes, wet wells, pump station chambers, treatment plant tanks, channels and digesters.
- ✓A definitions table that settles what people conflate — bump test (function check), calibration, zero / fresh air set and span — and states plainly that an instrument which passes a bump test can still read badly wrong.
- ✓A frequency table covering bump testing, full calibration, calibration after a failed bump test, calibration after an event (high concentration, drop, impact, water ingress, suspect reading) and the zero / fresh air set, each with the trigger and a practical note.
- ✓An eight-step bump test procedure from pre-check and clean-air power-up through test gas verification, sensor response, alarm activation (audible, visual and vibrating), tolerance check, baseline recovery and logging — ending with tag-out on failure.
- ✓A six-step calibration procedure covering competent-person conditions, the fresh-air zero, certified span gas application and adjustment, alarm set point verification against your entry criteria, gas certificate and due-date recording and labelling, and the rule that a sensor which cannot be brought within tolerance is replaced, never adjusted into apparent compliance.
- ✓Seven wastewater-specific sensor cautions written for sewer conditions: H2S sensors as consumables that read low at end of life, poisoning by high concentration, catalytic LEL poisoning by silicones and sulphides, LEL under-reading in oxygen-deficient atmospheres, cross-sensitivity, sampling the full depth because H2S pools at the base, and testing before ventilating as well as after.
- ✓Section 7 — a bump test log with columns for date, instrument serial, O2 pass, H2S pass, LEL pass and initials.
- ✓Section 8 — a calibration log (date, instrument serial, gas certificate / lot, next due, calibrated by) and Section 9 — an out-of-service register (date, instrument serial, reason, action taken / returned to service).
- ✓A responsibilities and review section splitting the duties of the instrument user, the supervisor and the PCBU, with the review triggers and a note on the 1 December 2026 Workplace Exposure Limit transition — supplied as an editable Microsoft Word (.docx) file, unbranded, ready to carry your own document control and issue details.
Worked example
A NSW council crew is called to a wet well at a sewage pump station. They have a confined space SWMS, an entry permit book and a four-gas detector, and they use all three. The permit is filled in properly. The meter is powered up at the truck beside the chamber while the pump station vents — so the fresh air zero is set on the sewer's own atmosphere, and every reading that day is understated by whatever was in the air at that moment. It has not been bump tested for six weeks, and nobody tracks sensor age, only the calibration sticker: the H2S sensor is past its service life and reading low. Three weeks earlier the same instrument saw a high H2S concentration during a blockage clear and went straight back in the ute, un-calibrated, on the assumption it recovered. The crew tests from the surface, gets a clear reading, and issues the permit. The number was precise, repeatable and meaningless, and the entrant descends into a stratified chamber that a probe at the lip never sampled anyway. Every one of those failures is caught here: the fresh air set in known clean air away from plant exhaust, the bump test that morning or the instrument does not leave, sensor age tracked alongside calibration date, mandatory calibration after a high-concentration event, and a sample line down the full depth. The crew was not short a SWMS or a permit. They were short the discipline that makes the number on the screen worth acting on.
Related legislation
- Work Health and Safety Act 2011 (NSW) — Section 19 primary duty of care, including the duty to provide and maintain suitable plant and equipment; Section 47 consultation; Sections 35-38 notifiable incidents.
- Work Health and Safety Regulation 2025 (NSW) — confined spaces (Part 4.3): atmospheric testing and continuous monitoring by a competent person using equipment suitable for the purpose, applying to every confined space entry independently of construction status.
- Work Health and Safety Regulation 2025 (NSW) — hazardous chemicals (Part 7.1) and the workplace exposure standard for hydrogen sulphide, transitioning to Workplace Exposure Limits from 1 December 2026, against which the H2S alarm set point is referenced.
- Work Health and Safety Regulation 2025 (NSW) — Section 291 (high risk construction work) and Section 299 (preparation and content of a SWMS), with review under Section 302, where the entry work this procedure supports is construction work.
- AS 2865 (Confined spaces) and AS/NZS 60079.29.2 (gas detectors for flammable gases and oxygen — selection, use and maintenance), together with the manufacturer's instructions for the specific instrument as the primary authority on calibration intervals and procedure.
Frequently asked questions
We bump test every morning. Isn't that the same as calibrating?
No, and the document opens with the distinction because it is the one that gets people killed. A bump test is a brief exposure to a known test gas to confirm the sensors respond and the alarms activate — it confirms the instrument works. Calibration adjusts the instrument against a certified concentration so the reading is accurate — it confirms the instrument is correct. An instrument can pass a bump test every morning and still read badly wrong all week. Bump testing tells you the sensor is alive and the alarm sounds; only calibration tells you that 9 ppm means 9 ppm.
Does buying this make us compliant?
No. This is a documentation set, not compliance. It is a procedure — you still need instruments that are actually in calibration, certified test gas that is in date, competent persons doing the testing, the logs actually filled in, and the manufacturer's manual for your specific instrument, which the document names as the primary authority over itself on intervals, gas concentrations and procedure. What you get is the documented instrument discipline, authored to NSW law, so the remaining work is implementation rather than drafting. A procedure in a folder and a meter with a dead sensor is exactly the false sense of security the document warns about.
Our meter reads clear at the maintenance hole. Why does the procedure keep pushing on sampling?
Because of where the gas is and where the meter isn't. H2S is heavier than air and pools at the bottom of a chamber, so a reading taken at the opening can be safe while the base of the space — where the worker is going — is lethal. The procedure requires a sample line and testing at multiple depths for that reason. It also requires the zero / fresh air set to be done in known clean air, never at the chamber opening or near plant or vehicle exhaust, because a zero set in contaminated air makes every later reading wrong by that amount, in the direction that reads safe.
The instrument saw a big H2S hit last month but seems fine now. Do we really have to calibrate it?
Yes. The document is explicit that high-concentration exposure can poison a sensor, and that after an instrument has seen a high H2S concentration it is calibrated before it is used again — do not assume it recovered. The same applies after a drop or impact, water ingress, or any suspect reading, because shock and poisoning can shift a sensor permanently with no symptom on the display. Catalytic LEL sensors carry the same risk from silicones and sulphides common in wastewater: they under-read afterwards without any obvious sign, so a sensor that has seen a sulphide-rich atmosphere is verified by calibration, not by looking normal.