Drain Odor Complaint Investigation

Purpose

Sewer odour is the drainage system's air arriving somewhere it was never meant to be, which means a barrier failed. There are five of them, and the whole job is finding out which one: the trap seals, the fixture-to-drain seals, the cleanout plugs and test caps, the piping itself, and the vent path that is supposed to release this air outdoors and away from anything that breathes it back in.

This needs a standard because odour investigations are usually run by nose, and the nose is the one instrument on the truck that fails in the direction that hurts. Hydrogen sulfide deadens the sense of smell at concentrations that still injure, so a space that has stopped smelling bad is a reason to ventilate and leave, not to relax or report the problem solved. This procedure opens with an instrument, tests the barriers in a fixed order, and releases on a reading plus an occupant confirmation.

Scope

Covers investigating a sewer or drain odour complaint in a house, a small office, a small retail space, a small restaurant or a multifamily unit, through to naming the failed barrier, correcting it where it is within a service tech's scope, and verifying the correction.

It does not cover clearing a stoppage, correcting a venting fault beyond a blocked or badly located terminal, or replacing defective pipe, each of which routes to its own SOP. It does not cover a grease interceptor, a lift station or any confined space. Odour with no drainage source, such as a dead animal, mould or a gas leak, is ruled out here and handed on.

Roles and the handoff

Procedure

Release criterion for this call: the failed barrier named and corrected, a personal multi-gas instrument reading zero for hydrogen sulfide across the window the customer identified, and the occupant confirming no odour over three consecutive days. Smell alone never releases a ticket in either direction.

1. Put the instrument in the space before you form an opinion. Acceptance: a calibrated personal multi-gas monitor bump-tested and switched on at the door, with a reading logged on arrival and at the hour the complaint occurs. Stop rule: a reading that rises while you work means everyone leaves and the space is ventilated before anyone returns; the hydrogen sulfide ceiling governing employee exposure is 20 ppm under 29 CFR 1910.1000 Table Z-2, so the instrument sets that call, not the nose. Hazard: a basement, crawl space, pit or vault with a suspected atmosphere is not entered at all - that is permit-required confined space work under 29 CFR 1910.146 with an attendant and supplied air, and it is not this procedure.

2. Map the complaint in time and place before testing anything. Acceptance: the rooms affected, the hours it occurs, what else runs at those hours, and how long since it started. Stop rule: a complaint nobody can localise gets the instrument left in place across a full day first, because testing five barriers blind costs more than a day of logging. Hazard: none at this step, it is an interview and a walk-through with the monitor already running from step 1.

3. Barrier one: dip every trap in and adjacent to the affected zone, including the ones nobody uses. Floor drains, a basement shower nobody has run since spring, a bar sink, a laundry standpipe. Acceptance: a seal depth for every trap, each between 2 in and 4 in, which is IPC Section 1002 in the edition your jurisdiction has adopted. Stop rule: a dry trap is refilled and the complaint re-checked at 24 hours before it is credited as the cause, because refilling a trap that will empty again by Friday closes a ticket that is not fixed. Hazard: an empty trap is an open sewer-gas path, so ventilate before kneeling over it and keep the monitor on your person rather than set down across the room.

4. Barrier two: test each fixture-to-drain seal. Acceptance: every water closet checked for movement under hand pressure with base and flange dry to a paper towel, every tub shoe and overflow gasket flood-tested, every tailpiece and disposer flange checked wet. Stop rule: a toilet that rocks is not simply tightened, because tightening against a failed seal pulls the bowl onto a flange already leaking air; the fixture is pulled and the seal replaced. Hazard: pulling a water closet opens the flange straight into the room, so it is plugged the moment the bowl is off and stays plugged until the bowl is set.

5. Barrier three: find every cleanout plug, test cap and capped stub, and account for all of them. Acceptance: a list of every access with its condition, including capped stubs left by a remodel behind a vanity, in an attic or under a deck. Stop rule: an access a drawing or prior ticket says exists but you cannot find stays on the open list, and the ticket does not close on the other four barriers while it is unresolved. Hazard: an uncapped stub can be at head height in a chase, so approach with the monitor in front of you and do not put your face to an opening to check whether it smells.

6. Barrier four: pressure or smoke test the piping when the first three are sound. Acceptance: the test performed, the barrier confirmed sound or the leak located and marked. Stop rule: no smoke is introduced until the fire alarm panel is on test with the monitoring company, the occupants are notified in writing, and anyone with a respiratory condition is out of the building; if any of those cannot be arranged today, the test is deferred and step 7 runs first. Hazard: smoke-test fluid is a chemical exposure and a fire-response trigger at once, so it is used under its own safety data sheet and the alarm test-mode procedure; cutting a wall to reach a located leak gets vacuum or water dust control, and on tile or masonry the silica controls at 29 CFR 1926.1153 on construction work, with 29 CFR 1910.1053 as the general-industry counterpart.

7. Barrier five: measure the vent terminal against the openings around it. Acceptance: the terminal's height above the roof, its horizontal distance to every door, openable window and outside-air intake, and its height above the top of any opening inside that distance, measured against the adopted edition's vent terminal section - IPC Section 904 in the 2018 and 2021 editions, Section 903 in earlier ones. Stop rule: a terminal inside the stated horizontal distance and below the stated height above the opening is the cause, and the correction is a relocation or extension, not anything done inside the building. Hazard: the terminal exhausts at head height for anyone kneeling beside it, so approach upwind, and where the roof calls for a personal fall arrest system it is on and anchored before you leave the ladder, with the fall trigger being 6 ft at 29 CFR 1926.501 for construction work and 4 ft at 29 CFR 1910.28 for general-industry walking-working surfaces.

8. Rule out the sources that are not sewer gas, then close what you opened and verify. Biofilm in an overflow or a disposer collar smells organic and sour rather than of sewer gas and clears with mechanical cleaning; mould, a dead animal and a fuel gas leak go where they belong. Acceptance: every trap refilled, every plug and cap reinstalled with its sealant, the corrected barrier named, a zero hydrogen sulfide reading across the customer's stated window, and a three-day occupant confirmation booked. Stop rule: no ticket closes on the day of the correction, because the barrier that failed on a schedule needs at least one cycle of that schedule to prove it holds. Hazard: this step restores the air barrier you disturbed, so the plug-by-plug walk-back is the verification and not a courtesy - one cap left off turns the repair into the next complaint.

When the site does not match this procedure

A floor drain that goes dry every few weeks. The correction is a trap seal primer valve, listed to ASSE 1018, or a barrier-type trap seal device, not a note asking the customer to pour water down it monthly. Nobody does that in month four.

The customer smells it only outside. That is the terminal, and step 7 runs first rather than sixth.

The customer will not allow a smoke test or a wall opening. Record the refusal in their words, list which barriers remain untested by name, and do not write a cause you did not establish.

The record this produces

• Instrument model, bump test, and every reading with its time
• The complaint map: rooms, hours, what else runs at those hours
• Seal depth for every trap tested, in inches
• Each fixture-to-drain seal tested and how
• Every access found, its condition, and any that remain unaccounted for
• Whether a smoke or pressure test ran, and the notifications that preceded it
• Terminal measurements: roof height, horizontal distance and height above each opening
• The barrier named, the correction, the verification reading and the confirmation date

The barrier name is the whole value. "Trap seal, master bath floor drain, dry at 0 in, primer fitted" tells the next tech what to check first if it returns; "sewer smell, resolved" sends them back to step 1.

Worked pass: small office suite, restroom odour every afternoon

Intake: two tenants report a sewer smell in the second-floor restroom from roughly midday onward, nothing in the mornings, ongoing about six weeks, no remodel.

• Step 1: monitor bump-tested and running. Zero for hydrogen sulfide at 9 a.m., rising to a peak of 3 ppm at 2 p.m. That is well under the 20 ppm ceiling at 29 CFR 1910.1000 Table Z-2, so it is a defect indicator rather than an overexposure - and the reading, not the smell, is what establishes the source is real.
• Step 2: restroom only, afternoons only, rooftop air handler runs on a schedule that starts at 11 a.m.
• Step 3: all four traps dipped between 2 in and 2-1/4 in. Barrier one sound.
• Step 4: water closet solid under hand pressure, base and flange dry to a towel; lavatory tailpieces dry. Barrier two sound.
• Step 5: every access found and accounted for, all caps in place. Barrier three sound.
• Step 6: FAILED its precondition. The building carries a monitored fire alarm and the property manager could not get the panel put on test that day. Stop rule taken: no smoke introduced, test deferred, step 7 run first.
• Step 7: on the roof with fall protection anchored, the vent terminal measured 4 ft horizontally from the rooftop outside-air intake and 1 ft below the top of that intake. Read from the adopted edition, the terminal must be at least 10 ft horizontally from such an opening, or 3 ft above its top if it is inside that distance. It fails both, and the air handler's schedule explains why the complaint starts at midday.
• Step 8: terminal extended and relocated to clear the intake by more than 10 ft horizontally. Traps refilled, nothing else had been opened. Monitor left across the following afternoon read zero for hydrogen sulfide through the 11 a.m. to 5 p.m. window, and the tenants confirmed no odour on three consecutive days. Released on the reading plus the confirmation, and the deferred smoke test cancelled as unnecessary rather than left open.

The deferral is the part worth carrying. A tech who pushed the smoke test through without the panel on test would have brought a fire response to an occupied office, on a building whose actual fault was on the roof and would have shown up in ten minutes with a tape measure.

References

• 29 CFR 1910.1000 Table Z-2 for the hydrogen sulfide ceiling that governs employee exposure, and 29 CFR 1910.146 for permit-required confined spaces, which no part of this procedure enters
• IPC Section 1002 for trap seal depth and the vent terminal section at IPC Section 904 in the 2018 and 2021 editions or Section 903 in earlier ones, in whichever edition and code family your jurisdiction has adopted
• ASSE 1018 for trap seal primer valves, with the device manufacturer's instructions for the application
• 29 CFR 1926.501 and 29 CFR 1910.28 for the fall protection triggers, split by activity, and 29 CFR 1926.1153 with 29 CFR 1910.1053 for respirable crystalline silica where tile or masonry is cut
• See related: the vent system inspection and correction SOP, the tub and shower drain service SOP, and the cleanout installation and access standard