The Fixture That Gurgled Only When Something Else Was Used

Why this matters

A gurgle that only fires when a different fixture is used is one of the few drainage symptoms that hands you a location for free. Two fixtures are involved, so the fault has to sit on the piping they share, and whether the noise arrives with an odor or without one tells you which way the air was moving. Techs who miss that spend the visit snaking the fixture that made the noise, which is the one place the fault is not.

This case was reconstructed from a fourteen-day occupant log rather than from a site visit, because the noise had stopped by the time anyone arrived. The log did the narrowing; the visit did the confirming.

Before anything: the two hazards this symptom carries

A gurgle is air crossing a trap seal, which means the barrier between the building and the sewer is being breached on a schedule. Treat the fixture as a live sewer-gas source until the seal is measured: keep the room ventilated with a window or exhaust fan running while you work, and do not use the smell itself as your reading. Hydrogen sulfide deadens the sense of smell at concentrations well below the ones that hurt you, so a nose that stops noticing an odor is reporting nothing useful.

If the diagnosis pushes you toward opening a cleanout on a line that may be holding water, stand to the side of the plug and back it off in stages so a surcharged line vents its head past you rather than into you, and wear splash-rated eye protection and gloves rated for wastewater contact while you do it.

Nothing in this procedure requires entering a manhole, a wet well or an interceptor. If it ever does, that is a permit-required confined space with entry permit, atmospheric testing and attendant obligations attached under 29 CFR 1910.146 in general industry, or under 29 CFR 1926 Subpart AA when the work sits on a construction site, and a would-be rescuer entering without that setup is how one casualty becomes two.

What the pairing narrows to

Call the noisy fixture the reporter and the used fixture the source. The reporter's trap is a manometer somebody installed for you: it moves when the pressure on the sewer side of it moves. The source is what created the pressure. So the fault sits somewhere on the shared path between the two, and everything outside that path is eliminated without a single tool.

That path is short in most houses. Here it was a two-story dwelling on a three-inch stack: an upstairs bath group at the top, a laundry standpipe and kitchen sink teeing into the base area on the ground floor, one vent through the roof.

        vent through roof
              |
  upstairs tub|            air pulled IN here
   ---------->|            (negative, above the slug)
              |
              |  falling water
              |
  laundry     |            air pushed OUT here
  standpipe ->|            (positive, near the base)
              |
              +---- building drain ----> sewer
                    restriction sits here

Reading the sign from the timing and the smell

Water falling down a stack drags air with it. Above and alongside the falling water the pressure goes negative; where the flow turns horizontal at the base and decelerates, it goes positive. The two ends of the same event produce opposite symptoms, and the reporter tells you which end it sits on:

• Negative at the reporter. Air is pulled from the room, down through the fixture, into the drainage system. The gurgle carries no odor at the moment it happens, because the airflow is inward. Water leaves with the air, so the seal is a little shallower after each event, and the odor complaint arrives days later or after the fixture sits unused.
• Positive at the reporter. Air is pushed from the drainage system out through the seal into the room. You get odor at the instant of the gurgle, sometimes visible movement in the bowl or standpipe, and the seal largely settles back afterward.

That is the whole discriminator, and it costs nothing to collect. A reporter above the discharge point reads negative; a reporter near the base reads positive. If the log says odor arrives with the noise, stop thinking about the upstairs vent and start thinking about what is downstream of the stack base.

What the log killed

Twenty-two gurgle events over fourteen days, about 1.6 a day, each one written down with the fixture that had just been used:

Twenty of the twenty-two entries recorded a smell at the moment of the noise. The two that did not were logged by a second household member who was never asked about odor, so they are blanks rather than negatives, and they are reported that way here rather than folded into the count.

Three candidates died on that table. A blocked roof vent would have shown its first symptoms upstairs, on the negative side, and the upstairs fixtures never reported anything. Self-siphonage of the standpipe would have needed the washer to be the source, and the washer never was. A local partial blockage in the ground-floor branch would have made the kitchen sink slow, and it was not.

What survives is the one thing consistent with a positive transient at the base of the stack that is bigger than it should be: a restriction downstream of where the stack turns horizontal, squeezing the free air area at exactly the point where the air needs to get out of the way.

The worked reconstruction

The design basis first, because the example has to be measured against it. Model plumbing codes as adopted and amended by the local jurisdiction size vents to hold pressure excursions at a trap to roughly one inch of water column, and that figure is derived for a nominal two-inch trap seal, leaving about half the seal as margin. It is a limit on the peak of the transient, not on an average.

That qualifier has to be applied here rather than quoted, because this trap was not nominal. Dipping the standpipe trap with a marked probe read a seal depth of 1-3/4 inches, not 2 inches, so the proportional allowance scales with it:

• Nominal basis: 1.00 inch w.c. at a 2 inch seal
• This trap: 1.75 / 2.00 = 0.875, so the allowance is 0.875 inch w.c.

Measurement next. A differential gauge with a 10 inch w.c. span and a stated accuracy of plus or minus 1 percent of full scale was teed into a test port above the standpipe trap. One percent of a 10 inch span is 0.10 inch w.c., and on a single instrument that is a fixed systematic offset, not a random spread, which matters twice below.

• Peak during an upstairs tub discharge: +2.6 inch w.c.
• Against the corrected allowance: 2.6 / 0.875 = 3.0 times the limit, and the 0.10 inch w.c. offset does not cancel here because this is a single absolute reading, so read it as 2.5 to 2.7 against 0.875.

The line was then cabled and jetted, the recovered debris was soft grease-bound solids rather than root mass, and the same port was re-read with the same gauge on the same instrument settings:

• Peak after clearing: +0.7 inch w.c.
• Difference: 2.6 - 0.7 = 1.9 inch w.c. of excursion that belonged to the restriction
• The gauge's fixed offset appears in both readings with the same sign, so it cancels in the difference, leaving 1 percent of the difference, 0.02 inch w.c. That is the line that makes the 1.9 trustworthy even though neither absolute reading is better than 0.10.

After clearing, 0.7 inch w.c. sits under the 0.875 inch w.c. allowance for this trap, so the seal is inside its margin without the standpipe trap being touched at all. The trap re-dipped at 1-3/4 inches on the third day and again on the tenth, which is the check that the fix held rather than the noise merely moving.

Time on the job: 1.0 hour reading the log and mapping the stack before the truck moved, 1.5 hours on site. The shop books an unbounded gurgle chase at 3.0 hours, so the reconstruction came in 0.5 hours under a 3.0 hour booking, about 17 percent under, and the saving is entirely in not snaking the reporter.

Where this reasoning would have failed

The log is only a locator if it records the source fixture, and most occupant complaints do not. A log that says "it gurgles sometimes" eliminates nothing, and the fix is to hand the customer a sheet with two columns, time and what was just used, rather than to book a longer visit.

The odor-timing discriminator inverts if the reporter's seal is already gone. An empty trap passes air in both directions with no water to move, so it will smell continuously and the timing tells you nothing; dip the seal before you trust the odor column. And a reporter that sits above the discharge but still smells at the moment of the noise is not a contradiction to explain away, it is evidence of a second event on the positive side reaching it through a shared horizontal branch, which is a different fault than this one.

Confirming it without opening a wall

Re-run the exact pairing that produced the complaint, not a general flow test. Discharge the source fixture that appeared most often in the log, stand at the reporter, and note whether the noise appears and whether odor comes with it. Then dip the reporter's seal immediately after and again after the fixture has sat unused for a week, because a seal that reads full on the day of the repair and shallow a week later means the mechanism moved rather than stopped.

Photograph the debris you recover and note its character. Soft solids at a repeatable distance and root mass at a joint point at different repairs, and the recovered material is the only part of that evidence that does not survive the drive back to the shop.

References

• 29 CFR 1910.146, permit-required confined spaces, general industry; 29 CFR 1926 Subpart AA, confined spaces in construction
• Model plumbing code vent-sizing provisions, which bind only as adopted and amended by the local jurisdiction
• See related: What a Drainage System Is Actually Doing With Air; What Siphonage Is and Why Only a Vent Stops It; How a Trap Seal Is Lost and How to Tell Which Way It Went
• See related: What Hydrogen Sulfide Does to a Sewer Above the Waterline