The Line That Blocked Every Few Months in the Same Place

Why this matters

Three shops cleared this line, all three did the work correctly, and it came back on the same schedule every time. That is the interesting part. When the repair is right and the fault returns, the diagnosis was never about the blockage; it was about whatever keeps rebuilding it. The recurrence interval is a measurement, it is free, it is already sitting in the service history, and almost nobody reads it. Read properly it separates an accumulation fault from a biological one before anybody puts a camera in the ground, and it tells you whether the next cleaning will change anything or just reset the clock.

Before the cap comes off

A cleanout cap on a line that has blocked is holding back whatever is standing above the blockage, and the head can be several feet of water in a multi-storey building. Crack the cap from the side of the fitting with your body out of the discharge arc, never over it, and have a bucket and a route for the volume before the last thread lets go.

Ask whether anyone has poured a drain product down the line in the past week, and look under the sinks rather than accepting a no. Standing caustic or acid above a blockage does not decay, and it arrives at the same moment the water does. Where that is a yes or an unknown, the opening is a chemical splash exposure at face level: sealed goggles, a face shield and chemical-resistant gloves rated for the product class named on its safety data sheet, and no mixing of a second product into the line under any circumstances.

Do not put your face over the opening to look or to smell. A sewer atmosphere is discussed in full in the companion article on separating a belly from roots from a partial collapse; the short version that matters here is that hydrogen sulfide deadens the sense of smell at concentrations well below the ones that hurt you, so an odour that fades is not an atmosphere that improved.

The building and the record

Four-unit walk-up, two storeys, one kitchen branch serving all four units into a 4 inch line to the street. The complaint, for two years: the ground-floor kitchen sink backs up, always the ground floor, never the upper units.

The service history, pulled from three shops' invoices:

Five intervals: 4, 3, 3, 5 and 3 months. Mean 3.6 months. One of them, the five, does not sit with the others, and that outlier turned out to be the whole diagnosis.

Reading the interval before reading the pipe

An interval that is stable within a month, over two years, is telling you the mechanism is an accumulation running at a steady rate against a fixed geometry. Something is being deposited at a constant rate and the line blocks when the deposit reaches a threshold. Random blockages do not produce a stable interval. A one-off obstruction does not produce five of them.

The question that separates the two candidate mechanisms is what the interval scales with:

• An accumulation fault scales with usage. Less discharge, slower build, longer interval. It does not care what month it is.
• A root intrusion scales with the season. Root activity is highest in the growing season, so intervals shorten in spring and lengthen in winter, independent of how much the building is used.

So look at the outlier. The five-month interval ran from January to June, which spans the spring. If roots were the mechanism, that is the interval that should have been the shortest of the five, not the longest. It went the wrong way for roots by the largest margin in the record.

Then look at what else was different. The property manager confirmed the second-floor rear unit was vacant from early February to late March, about two months of the five. Four units down to three is a 25 percent reduction in kitchen discharge for roughly 40 percent of that interval. If deposition rate tracks discharge, the expected stretch is on the order of 1.15 times the normal interval, and the observed stretch was about 1.5 times the 3.3-month average of the surrounding intervals. Consistent in direction, and not precise enough to call a measurement: occupancy is not a controlled variable and one vacancy is one data point. It is corroboration, not proof, and it is worth saying so on the ticket rather than dressing it up.

Two independent reads, usage-scaling present and season-scaling absent, both point at accumulation against a fixed geometry.

Making "the same place" mean something

The three tickets that recorded a footage at all gave three numbers: 41 feet, 38 feet and 38 feet. That looks like a discrepancy and it is not.

A cable or camera footage counter carries two error terms and they behave differently. The zero offset is where the operator started counting: at the reel, at the cleanout face, or at the point the tool entered the pipe. That is a fixed systematic offset belonging to one setup, so it cancels when you difference two footages taken in the same setup, and it does not cancel at all when you compare two shops. The counter's own error is a percent-of-reading term that grows with distance out.

Reconcile before concluding. The shop that logged 41 feet zeroed at the reel, which sat about 3 feet back from the cleanout face. Correcting to the cleanout face: 41 minus 3 equals 38 feet. All three readings are the same location, within the counter error, and the record now supports a single fixed point rather than a wandering one.

That reconciliation is the step everyone skips, and skipping it produces the opposite conclusion: three different footages reads as an intermittent problem moving up and down the line, which points at grade or roughness rather than at a discontinuity. One footage points at a discontinuity, and a discontinuity is a physical feature you can go and find.

What the camera confirmed rather than discovered

At 38 feet, with the line freshly cabled and free running, standing water across the invert, deepening toward 38 feet and clearing beyond it. No root mass. No fracture, no offset ring, no soil visible. The pipe was intact and the profile was not.

A depression in the invert holds water permanently. Water standing in a section is water at zero velocity, so there is no shear on the invert there and anything arriving settles. Above that section the line is at grade and everything moves; below it the line is at grade and everything moves. The companion reference on what a belly does to solids transport carries the geometry and the storage arithmetic; the point for this diagnosis is that the deposition rate at a fixed depression is steady, and a steady deposition rate against a fixed volume is exactly what produces a stable recurrence interval.

Every cleaning had been correct. Every cleaning had emptied a container that immediately started refilling.

Deciding what to do with a stable interval

Three options, and the honest comparison is what each does to the interval, not what each does on the day.

Keep cabling reactively. Interval unchanged at about 3.5 months, so roughly 3.4 visits a year. The hidden cost is that every one of them is an emergency call: it arrives with a backed-up sink, it displaces scheduled work, and an after-hours clear on this line has been running about twice the hours of a planned one once you count the callback and the mop-up. A comparable number of visits to the planned option below, roughly double the hours, and a tenant complaint each time.

Convert to planned maintenance at a cadence inside the interval. Cable it every 3 months, deliberately, before it blocks. The interval does not change; the emergency does. Visit count stays at about four a year, hours per visit drop to the planned figure, and the tenant impact goes to zero. This is a legitimate answer where the excavation is genuinely impractical, and it should be sold as what it is: a permanent maintenance obligation, written into the record so the next owner inherits the reason and not just the schedule.

Correct the geometry. A spot repair on a few feet of pipe. This is the only option that changes the interval rather than resetting the clock, and after it the correct expectation is that the line joins the rest of the building's drainage in needing nothing on a schedule. Excavating to reach it is trench work: a protective system is required once the excavation reaches the depth set in 29 CFR 1926 Subpart P, and no one enters an unprotected trench to look at a pipe, however quick the look.

A fourth thing gets proposed and is worth naming so it can be declined: jetting the line. Jetting cleans better than a cable and it will lengthen the interval, because it removes the film the cable leaves behind. It does not remove the depression. It moves a 3.5-month interval to something longer and leaves the mechanism in place, so it is a better version of the option that changes nothing.

Confirming the call afterwards

The confirmation for this fault is not a clear camera run on the day. Every one of the six visits ended with a clear camera run.

The confirmation is the interval. After a geometry correction, the first meaningful data point arrives at about 3.5 months, when the line would have blocked and does not. Put a calendar note at four months and one at eight, and record a non-event at each. Two clean intervals past the historical mean is a repair; one is a coincidence, because a fresh cabling alone would have carried you to about 3.5 months anyway.

Log the interval itself as a field, not as prose in a note. This diagnosis was available to the second shop, and to the third, and none of them had it because nobody's record put six dates next to each other. Interval, footage with its zero point named, and occupancy events. Three fields, and they turn a service history into an instrument.

References

• 29 CFR 1926 Subpart P (excavations), for any trench opened to reach a spot repair
• Safety data sheet for any drain-cleaning product identified on the line before a cleanout is opened
• See related: What a Belly in a Line Does to Solids Transport; How to Separate a Belly From Roots From a Partial Collapse; Sewer Camera Inspection Reference