The Cast Iron Tie-In That Dropped the Stack

Why this matters

A plumber cut a section out of a 4 in cast iron soil stack in a basement to add a wye for a new basement bath, and did not clamp the stack above the cut. The work tested fine. Two weeks later a stain appeared on a second-floor ceiling, under an attic bathroom. A second tech diagnosed the leak correctly, a cracked lead-and-oakum joint at the attic bath's branch, and repaired it correctly. Three weeks after that, a new stain. Being right about the joint was not enough, because the joint was a symptom: the stack was still moving. The real repair was in the basement, and it is now the first step of the shop's tie-in procedure.

The house and the first job

A two-and-a-half story house from the 1920s with a single 4 in hub-and-spigot cast iron stack running from a base fitting on the basement floor, past a first-floor bath and a second-floor bath, up to a small attic bath, and on through the roof. Every hub joint was lead and oakum. The stack had no riser clamps at any floor anyone could find later; like many stacks of that age it stood on its base fitting and was steadied, not carried, by the framing it passed through.

The tie-in. The first tech cut out about 18 in of stack 3 ft above the basement floor with a snap cutter, slid two shielded no-hub couplings on, and dropped in a PVC sanitary wye with a short spigot above and below to fill the gap. The insert was cut about 1/2 in shorter than the opening to make it easier to swing into place. The couplings were tightened to their band torque. The new branch was water tested with the basement bath rough and held. No clamp went on the stack above the cut, before or after.

The first callback: the right diagnosis

The signal. A brown ring about 10 in across on the second-floor hall ceiling, directly below the attic bath, growing a little each time the attic tub was used.

The trace. The second tech opened a 16 in square in the stained ceiling under the attic bath, wearing a dust mask and eye protection for the plaster and lath, with a tarp below. The attic bath's toilet and tub branches tied into the stack through a cast iron sanitary tee in the joist bay. The hub joint just below that tee was wet, with a hairline crack running through the lead at the face of the hub and a dark wet line in the oakum behind it. The tech filled and drained the attic tub with a helper and watched water bead at the crack. Nothing else in the bay was wet.

The repair. Correct for what was seen: with the attic fixtures out of use, the cracked hub was cut out with the snap cutter just above and below the joint and replaced with a short length of no-hub cast iron and two shielded couplings. The lead and oakum from the old joint was bagged, the area vacuumed with a HEPA vacuum, and the tech washed up before eating; old lead joints shed lead when disturbed, and the shop treats that work under its lead program as 29 CFR 1926.62 requires for employees in construction work. The attic tub was run with a helper; the new couplings were dry. Ceiling patched.

Every word of that diagnosis was true. The joint was cracked, it was the source of the stain, and the repair stopped that leak.

The second callback: same symptom, new joint

Three weeks later the homeowner called with a new stain, this time on the first-floor ceiling under the second-floor bath, and the second tech's repair upstairs was being blamed for "shaking things loose."

The second tech came back and found the second joint: the hub just below the second-floor bath's sanitary tee had cracked the same way. Two lead joints in five weeks, on a stack that had been dry for decades, both directly below a branch tee. One cracked joint is a joint. Two in a pattern is the stack telling you it moved.

What moved, and why nobody saw it

Branches are fixed; the stack is not. Each bath's toilet is screwed to its floor through its closet flange, so the branch from the toilet to the stack is anchored at the fixture end. If the stack settles while the branches stay put, the strain concentrates at the stiffest, least forgiving point near each tee: a lead-and-oakum hub joint, which takes compression well and very little bending or pull. A small settlement is enough to crack the lead.

The basement told the rest. The lead tech went to the basement first this time, before touching the new leak.

• The couplings had slipped. The upper shielded coupling showed a bright scuffed band on the cast iron about 1/2 in long below where the coupling's edge now sat. The cast iron had slid down inside the coupling.
• The gap had closed. The 1/2 in left between the PVC insert and the cast iron on install day was gone; the cast iron above now bore on the insert's top end, inside the coupling.
• The floor collar agreed. Where the stack passed through the first floor, the paint line around the stack sat about 1/2 in below the edge of the old hole's escutcheon.

The stack above the cut weighed several hundred pounds: two and a half stories of 4 in cast iron, plus three tees and their branches. With 18 in of it removed and nothing clamping it, that weight came down on the couplings. A shielded coupling holds by its bands' grip and is not made to carry a stack's weight in tension or shear; over days, with every flush adding vibration, the cast iron crept down until it bore on the insert. The upper floors went down with it, by the amount of the gap the first tech had left. The tests passed because a water test checks for leaks at the moment of the test, not for a stack that is going to settle over the next week.

  STACK AS FOUND, BASEMENT (no clamp above the cut)

     attic bath tee      <- hub below cracked (visit 2)
          |
     2nd floor bath tee  <- hub below cracked (visit 3)
          |
     1st floor bath tee
          |              stack dropped about 1/2 in
  ========+============  basement ceiling joists
          |              (no riser clamp here)
       [coupling]  <- scuff band 1/2 in long
          |
        PVC wye --------> new basement bath branch
          |
       [coupling]
          |
     base fitting on basement floor

The repair that ended it

Stop the movement before fixing the joint. The lead tech did nothing upstairs until the stack was carried.

1. Fixtures out of use. The household was asked not to use any bath for the afternoon; the basement and first-floor fixtures were left usable after the clamps were on, and the upper baths later.
2. Riser clamps above the cut. A 4 in riser clamp went on the cast iron in the basement ceiling joist bay, its ears resting on top of 2x blocking nailed between two joists, so the stack's weight now went into the floor framing rather than the couplings. A second clamp went on at the second-floor framing, bearing on blocking in that floor, reached through the opening already cut in the first-floor ceiling for the second leak. Both were snugged to the pipe as it sat; the stack was not jacked back up, because lifting it would have re-strained the joints that had survived.
3. The insert redone. With the stack now carried by the clamps, the couplings were loosened, the PVC insert removed and replaced with one cut to fit the opening with only the clearance the coupling maker allows, and the joints remade with shielded couplings rated for the pipes they join (CISPI 310 or ASTM C1277 for cast iron to cast iron of matching outside diameter, an ASTM C1460 transition coupling where the cast iron and plastic diameters differ), re-torqued to the maker's band torque.
4. The second cracked joint replaced exactly as the first had been: the hub cut out, a no-hub section and two shielded couplings, lead debris bagged and vacuumed.
5. Every remaining lead joint inspected. Each accessible hub was looked at with a light and a mirror for a crack in the lead, and the joints in closed walls were checked by running water from each bath in turn with a helper watching the ceilings below and a moisture meter on the plaster. The first-floor bath's tee joint showed a hairline in the lead face but stayed dry under a full tub drain; it was photographed and noted for the homeowner as one to watch.

Proof. Over the next month the homeowner checked the three ceilings weekly; dry each time. On a follow-up visit the scuff marks on the upper basement coupling had not moved, and a pencil line drawn across the riser clamp and the stack on repair day still lined up.

What the shop changed

The procedure in How to Tie a New Branch into a Cast Iron Stack now opens with support: before any cut, a riser clamp goes on the stack above the cut, bearing on framing or blocking, and a second clamp or a support below if the lower section will be left hanging. That is the support IPC Section 308 and UPC Section 313, as adopted, ask for in substance: a stack carried at its base and at each floor. Only then does the cutter come out. The insert is cut to fit the opening, not short for convenience, and the coupling maker's clearance is the only gap.

A second line went into the shop's callback rule: when a lead-and-oakum joint cracks on an old stack that had been dry, ask what work has been done on that stack recently, anywhere along it, before repairing the joint. Two cracked joints below branch tees on one stack is a moving stack until a level, a tape and a look at the couplings prove otherwise.

The cost. Three visits instead of one, two ceilings opened and patched, and two joint repairs on top of the tie-in, all carried by the shop. The two riser clamps and the blocking took about 40 minutes.

References

• IPC 2021 Section 308 and UPC 2021 Section 313, as adopted by your jurisdiction: pipe support, including vertical stacks at their base and floor levels.
• 29 CFR 1926.62, lead in construction, for removing lead-and-oakum joints.
• CISPI 310 and ASTM C1277, the shielded no-hub coupling standards, and ASTM C1460 for shielded transition couplings, and the coupling maker's torque and clearance instructions.
• See related: How to Tie a New Branch into a Cast Iron Stack; Pipe Support Spacing by Material.