Gas testing process flowchart (permit-to-work atmospheric testing)

What the gas testing process flowchart (permit-to-work atmospheric testing) process is

Gas testing is the measurement step inside a permit-to-work system. A permit is raised for work in an area where the atmosphere could be oxygen deficient, oxygen enriched, flammable or toxic, and the permit says an atmospheric test is required before the work is authorised. Everything after that is this process: an authorised tester is found, an instrument is proved fit to be believed, the atmosphere is sampled in a fixed order and at the levels where each gas actually collects, the readings are judged against limits somebody wrote down in advance, and the result becomes a dated record attached to the permit. The chart then follows the part most procedures cover badly, which is what happens after the first clear reading: whether the job runs on continuous personal monitoring or on periodic retests, what the person holding the monitor does when it alarms, and how the test record and the instrument datalog are filed when the area is handed back.

This is the testing process, not the entry process and not the hot work process. A confined space entry chart covers the register, the isolations, the rescue arrangement, the attendant and the entrant tally, and treats gas testing as one box inside it. A hot work permit chart covers combustible clearance, detection isolation, the fire watch and the post-work watch period, and treats the gas test as a single precondition before the torch. This chart zooms into that box and gives the instrument, the tester's authorisation and the retest regime lanes of their own, because that is where the failures live: an out-of-calibration detector, a tester certificated for the wrong scope, a reading taken at the hatch of a vessel that is stratified, a clear result reused hours after it stopped meaning anything. Use it alongside those charts rather than instead of them. It is a starting point to adapt under your own procedures, the regulations that apply where you operate, and competent-person review; it does not make anyone compliant and it does not replace a site-specific safe system of work.

Four decisions carry the process. 'Atmospheric test required?' belongs to the permit issuer, not the tester, because the trigger comes from the risk assessment and the permit type. 'Tester authorised for this scope?' sits in the issuer's lane too, since authorisation to test before entry, to clear an area for hot work and to stand watch on live hot work are certificated separately and the issuer is the person checking. 'All readings inside the stated limits?' sits with the tester and carries three ways out, because refusing to declare an area gas free is a legitimate answer and needs somewhere to go. 'Alarm or changed conditions?' sits with the entrant or attendant rather than the supervisor, because the person wearing the monitor is the one who has to act on it and waiting for permission is how people are overcome.

What this flowchart covers

In this template

• Five swimlanes (Permit issuer, Work party / performing authority, Authorised gas tester, Entrant / attendant and Instrument technician) across six phases: test request, tester and instrument, test the atmosphere, record and issue, monitor the work, and hand back and file
• An "Atmospheric test required?" decision owned by the permit issuer, with a route straight to work under the permit's other controls when the risk assessment does not call for a test, so the chart covers the no-test case honestly instead of assuming one
• A whole phase on instrument readiness that most gas testing procedures compress into a single line: "Detector calibration still in date?", "Zero in fresh air and bump test" and "Sensors and alarms respond?", with a failed check routed to the instrument technician for a full calibration before the detector is used
• The test sequence itself as two ordered steps, "Read oxygen first, then LEL, then toxics" and "Sample every level through the full depth", so the order of gases and the stratification of the space are both on the canvas rather than in a footnote
• A three-way limits decision: within limits goes to "Issue the gas test record for the permit", outside limits loops through ventilation and a full retest, and an atmosphere that will not clear ends at "Declare the area not gas free and stop" in the issuer's lane
• The monitoring half: a "Continuous monitoring required?" decision that splits personal monitors from a fixed retest interval, an "Alarm or changed conditions?" call owned by the entrant or attendant, a "Work finished or test expired?" gate that sends an expired test back for retesting, and closure on the filed datalog

When to use this template

• You are writing a gas testing procedure and need one picture of who requests the test, who is competent to take it, who records it and who acts on an alarm
• Your permits say gas test required with no gases, no sample points and no limits written on them, and testers are reaching different conclusions about the same area
• Detectors are going into service overdue for calibration or without a daily bump test, and you need the instrument checks drawn as a gate rather than assumed
• A job overran its permit and the work carried on against a gas test taken that morning, so the validity period and the retest interval need to be explicit and owned
• An auditor, insurer or regulator has asked how you demonstrate an atmosphere was safe when the work started and stayed safe while it ran

How it works

1. Rename the lanes to your roles

   Replace Permit issuer, Work party / performing authority, Authorised gas tester, Entrant / attendant and Instrument technician with the titles your procedures actually use. On many sites the tester and the attendant are the same person and the instrument technician sits in a workshop off site; merge or rename those lanes rather than drawing handoffs that never happen.

2. Write your trigger rule onto the first decision

   List the work types and locations where an atmospheric test is mandatory before authorisation, and say who decides. Keep it concrete: vessels and tanks, sewers and pits, line breaks on process systems, hot work near flammables. Anything left to judgement on the day should say whose judgement it is and what they look at.

3. Set your gas list, sample points and limits

   Replace the placeholder wording on the fourth step with the gases you actually test for, the points or levels each sample is taken from, and the acceptable value for each. Cite the limits that apply where you operate rather than copying a number from a template, and record who has authority to declare an area gas free.

4. State your instrument regime

   Put your own calibration interval, your bump test rule and your out-of-service rule onto the instrument phase. Say who holds the calibration records, what happens to a detector that fails its check, and whether a bump test station or a manual check is used. A detector nobody can prove was in calibration produces a reading nobody can defend.

5. Decide continuous monitoring against periodic retests

   Write the rule that drives the monitoring decision. Which jobs need a personal monitor worn in the breathing zone throughout, which are covered by retests at a fixed interval, and what that interval is. Set the low and high alarm points, and name who is authorised to stop the work when either sounds.

6. Fix the validity period and the record

   State how long a test result remains valid, what ends its validity early, and what the test record must carry: each gas and reading, the level or point sampled, the instrument and serial number, the time and the tester. Decide where the record and the instrument datalog are stored and for how long they are kept.

7. Walk it against a real gas test

   Take two or three recent permits, one that ran cleanly and one where the atmosphere failed or an alarm sounded, and trace them through the chart. Any step people describe that is not drawn, and any box that is drawn but skipped in practice, is the finding worth fixing before you publish the procedure.

Frequently asked questions

What are the steps in a gas testing process?

A permit is raised for work in an area with a possible atmospheric hazard and the issuer decides whether a test is required. If it is, the permit names the gases, the sample points and the acceptable limits, and the issuer confirms the tester is authorised for that scope. The tester checks the detector calibration is in date, zeroes it in fresh air, bump tests it and confirms the alarms respond; a failed check sends the instrument for full calibration first. The atmosphere is then sampled with oxygen read first, then flammables, then toxics, at every level through the depth. Within limits produces a gas test record for the permit; outside limits triggers ventilation, removal of the source and a full retest; an atmosphere that will not clear is declared not gas free and the job stops. The record is attached with a validity period, the work runs on continuous monitoring or timed retests, an alarm clears the area, and closure files the record and the datalog.

What is the difference between a bump test and a calibration?

A bump test, also called a function check, applies a challenge gas at a concentration and for long enough to activate every alarm setting. It answers one question: does gas reach the sensor and do the alarms work. It does not measure accuracy. A calibration check compares the reading against the known concentration on the cylinder and tells you whether the response sits inside the manufacturer’s acceptable range. A full calibration goes further and adjusts the instrument so its reading matches a certified test gas. OSHA guidance on portable direct-reading monitors is that a bump test or calibration check should be carried out before each day’s use in line with the manufacturer’s instructions, and that an instrument failing either should be fully calibrated before it is used again. Calibration intervals themselves follow the manufacturer rather than one universal figure, so put your own interval on the chart.

Why is oxygen tested before flammable and toxic gases?

Because the instrument depends on it. OSHA procedures for atmospheric testing state that oxygen is tested first because most combustible gas meters are oxygen dependent and will not give reliable readings in an oxygen deficient atmosphere; combustible gases are tested next, then toxics. A flammable reading taken in a depleted atmosphere reads low and reassuring, which is the wrong error to make. The same logic explains why enrichment matters as well as deficiency: an atmosphere above the normal range widens the flammable range of anything present and makes materials ignite more readily. Where the sample is drawn through a hose with an aspirator pump, the second discipline is patience. Purge the line and wait out the stated response time for the length of tubing in use, because a reading taken before the sample has arrived is a reading of the air in the hose.

Who is allowed to carry out a gas test?

A competent person, authorised for the specific kind of test being done. In Great Britain the Confined Spaces Regulations 1997 and the supporting HSE approved code of practice, L101, require testing of the atmosphere by a competent person using a suitable and correctly calibrated gas detector, and continuous monitoring where the risk assessment shows conditions may change. In the oil and gas sector the OPITO Authorised Gas Tester standard splits the role into three levels: testing for oxygen, flammable and toxic gases up to and including confined space work, testing for flammable gas in preparation for hot work, and providing safety watch by ongoing monitoring of a hot work site. The point common to both is that authorisation is scoped rather than general, which is why this chart makes the issuer check the tester's scope before the instrument is even picked up.

What limits count as a safe atmosphere for entry?

Limits come from the regulations that apply where you operate and from your own risk assessment, so the chart says stated limits rather than carrying numbers of its own. As a reference point, the acceptable entry conditions in the United States for permit-required confined spaces under 29 CFR 1910.146 are an oxygen concentration between 19.5 and 23.5 percent, flammable gas, vapour or mist below 10 percent of its lower flammable limit, and any substance with a published exposure limit below that limit; hydrogen sulphide sits in OSHA Table Z-2 with an acceptable ceiling of 20 ppm. Those figures are jurisdiction specific and many operators set tighter internal limits, particularly for hot work, so write your own onto the permit and onto the decision box. Whatever numbers you choose, a limit means nothing without the level the sample was taken at and the time it was taken.