Popcorn Ceiling Removal and Skim Standard

Purpose

Answer the asbestos question with a laboratory result before anything is scraped, then remove and skim so the finished plane reads flat under the light the room actually gets. Scraping an untested ceiling can release asbestos fibers through an occupied house, which is not a callback but a health event with a paper trail starting at your invoice. Scraping it dry tears the board paper, turning a one-day removal into a full skim on the shop's time.

Scope

Removal of sprayed or troweled ceiling texture and the skim that follows, residential and small commercial, from sampling through priming the finished plane.

Which OSHA Part governs, since this SOP cites both: a residential renovation is construction work under 29 CFR 1926, and the general-industry sections here reach it that way, respiratory protection at 1910.134 through 1926.103, with step 4's electrical counterparts at 1926.416 and 1926.417, lockout and tagging of circuits.

Not covered: asbestos abatement, a licensed trade, where a positive result goes; the choice between painting, skimming over and removing (the popcorn ceiling decision article); the wall prep that follows (interior surface preparation standard); board patch repair (drywall repair and texture match standard); and RRP containment on a pre-1978 painted ceiling, which its own, stricter standard governs.

Roles and responsibilities

Procedure

1. Treat every textured ceiling as suspect until a laboratory says otherwise. Have samples collected by an accredited asbestos inspector and analyzed by an accredited laboratory using polarized light microscopy, in the number the lab or your state program specifies, not one from a convenient corner. Acceptance: a written lab report in the file naming the material, the sample locations and the result before any tool touches the ceiling, kept as long as the shop keeps its job files, because a future owner may ask. Stop rule: no result, no scrape. A positive result ends this SOP and the shop does not remove it under any protective measure, which the estimator has told the customer at sale: the work goes to a licensed abatement contractor under your state's asbestos program, which is the program that reaches a house, since the asbestos NESHAP at 40 CFR 61 Subpart M excludes a residential building of four or fewer dwelling units from its definition of facility at 40 CFR 61.141. Hazard: collecting the sample is itself a disturbance, so the spot is misted first, the collector wears respiratory protection under a written program per 29 CFR 1910.134, and the spot is sealed after collection rather than left open in an occupied room.

2. Run the pre-1978 lead gate on any painted ceiling. Establish the construction year and whether the texture has been painted. A recorded year of 1978 itself is ambiguous rather than a pass, since a county construction year is a completion date and not a paint date and a 1978 house is routinely finished from 1977 stock, so a boundary-year ceiling gets a test or a presume-lead plan. Acceptance: a stated construction year in the record and a decision on lead, either a documented negative test or a presume-lead work plan. What wrong looks like: a negative asbestos report treated as clearance for the paint too. Stop rule: pre-1978 painted ceilings run under the RRP containment standard, with its prohibited practices and its cleaning verification, not the protection here. Hazard: the two hazards are independent; clearing one says nothing about the other.

3. Build containment for water and mud, not just dust. Empty the room, cover the floor wall to wall with sheeting lapped up the walls at least a foot and taped, cover or seal ducts and vents and count them, cover every smoke and CO alarm in the room and count those too, and seal the doorway with a flapped entry. Acceptance: the floor covering holds standing water without seeping at any seam, and both counts, ducts and alarms, are written down for the restore at step 8. Stop rule: a leaking seam is re-lapped before wetting starts, not while water is on the floor. Hazard: wet sheeting is extremely slick, so the walking surface gets a non-slip covering over the plastic and nobody enters in smooth-soled boots.

4. De-energize every ceiling fixture before water goes up. Shut off the circuits feeding ceiling fixtures and fans, prove each dead with a meter proved on a known live source before and after, per NFPA 70E-2021, 120.5, with work practices at 29 CFR 1910.333(b)(2), then drop or bag them. Acceptance: every ceiling outlet in the room reads zero on a proved meter before it is opened, and each opening is bagged so water cannot enter the box. What wrong looks like: a spray bottle worked around a live fan canopy. Stop rule: a fixture that will not read dead, or a panel with no legible labels, stops the step; if nobody on site is qualified under 29 CFR 1910.332 to prove it dead, an electrician does this part. Hazard: this is water going onto an assembly that carries branch circuits, which is the one combination on this job that kills someone, so the circuits are off before the first spray and stay off through step 6.

5. Wet the texture in sections before any blade touches it. Mist a section of roughly 4 ft by 4 ft with a pump sprayer, let it absorb, then mist again to saturation, not surface-wet. Acceptance: the texture releases in soft ribbons under light blade pressure and the underlying paper stays intact; if it powders or flies, it is not wet enough. Stop rule: dust means stop, mist again and wait; nobody continues scraping a dry section faster to get it over with. Hazard: overhead spray and falling wet material go straight into upward-facing eyes, so sealed goggles rather than glasses and a brimmed cover, and the sprayer is never aimed at a bagged fixture opening.

6. Scrape at a low angle, and stop at the first torn paper. Work a wide taping knife at a shallow angle, corners rounded or taped so the blade cannot dig, taking the texture off in overlapping passes. Acceptance: the section comes off to an intact, continuous paper face, with no fuzzed or lifted paper and no gouges deeper than the surface. What wrong looks like: brown fibers showing through the paper face, which is the board's core and will telegraph through any skim. Stop rule: torn paper stops the scrape and the lead chooses between sealing the tear and skimming over it, or abandoning removal for a skim-over where a painted texture will not release at all. Hazard: scraping overhead loads the shoulders and neck and puts a blade above the operator's face, so the work runs from a stable platform at a height that keeps the elbows below shoulder level, and the blade is drawn across rather than pulled toward the body.

7. Skim the plane, sand it, and check it under raking light. Apply skim coats in the number the finish requires, feathering each beyond the last, then sand with a dust-collecting sander. Acceptance: a straightedge laid across in two directions rocks nowhere, and a work light held near-parallel to the ceiling shows no ridge, no scraper track and no paper fuzz. What wrong looks like: a ceiling that reads flat from the doorway and shows every scrape line when the afternoon sun hits it. Stop rule: any track visible under raking light gets another coat, because a ceiling in a room with a window or a wall wash is the least forgiving plane in the house and a topcoat will not hide it. Hazard: sanding joint compound overhead puts respirable dust into the breathing zone and grit into the eyes, and the compound may contain crystalline silica, so read the product SDS: where it does, 29 CFR 1926.1153 reaches this task through the alternative exposure control methods at (d), not through Table 1, which lists no drywall task, so run the sander on a HEPA vacuum, wear sealed goggles, and wear any respirator under a written program per 29 CFR 1910.134.

8. Prime, restore what you disabled, and release in writing. Prime the skimmed plane with a drywall sealer or the primer the topcoat data sheet names over fresh compound, then re-hang fixtures, re-energize the circuits opened at step 4, uncover the ducts counted at step 3, and uncover and test every alarm counted there. Acceptance: uniform sheen under raking light; ducts uncovered equals ducts counted and alarms uncovered equals alarms counted; every fixture works and every alarm passes its own test button. What wrong looks like: a fan re-hung but never tested, or an alarm still bagged behind a door. Stop rule: any item that does not restore or does not test stops the close-out and gets fixed or escalated before the crew leaves. Hazard: this is where energy goes back into a circuit that has had water near it, so the box is dry and the wiring inspected before the fixture goes back, the breaker is closed with the body to the hinge side of the panel door, and the person closing it is the person who opened it.

The record this produces

Job file, read by whoever answers for this ceiling later: the lab report naming the inspector who sampled, the sample locations and the result; the construction year and how the boundary was resolved, tested or presumed; ducts and alarms counted at step 3 against the counts restored at step 8; the skim coats applied and where; and the customer's written release.

One run, filled in

A 1978 single-story, one 14 ft by 16 ft living room ceiling, texture that looked unpainted, occupied. Step 1: three samples by an accredited inspector, all no asbestos detected, report in the file before the crew was scheduled. Step 2: 1978 is not before 1978, so RRP does not apply on its face, but 1978 is the boundary year itself, so the lead took the fork rather than the pass and carried a presume-lead plan for any painted area the scrape turned up.

Step 3 covered the floor wall to wall lapped 14 in up the walls, counted and covered three ducts and one smoke alarm, and flapped the doorway. Step 4 proved two ceiling outlets dead and bagged both boxes. Step 5 misted the first section twice; it released in soft ribbons.

Step 6 failed on the third section: the blade skidded rather than cut, the paper fuzzed at two spots, and the stop rule fired at the first torn paper. The lead found a painted skin, invisible from the floor, over roughly a third of the ceiling and took the option the stop rule allows, skimming over rather than removing, which step 2's presume-lead plan pointed at anyway. It cost one extra skim coat over that third; scraping through would have cost the paper and a level-5 finish across the whole plane.

Step 7: two skim coats on the scraped area, three over the skimmed-over third, feathered together and sanded on a HEPA vacuum; a straightedge then rocked nowhere in either direction and raking light showed no scraper track. Step 8 primed with a drywall sealer, sheen uniform at the label dry time. Restore: three ducts against three counted, one alarm against one counted and sounded on its own button, both fixtures re-hung and working, both circuits closed by the lead who opened them.

References

• 40 CFR 61 Subpart M, the asbestos NESHAP, whose facility definition at 40 CFR 61.141 excludes a residential building of four or fewer dwelling units, leaving a house to the state asbestos program.
• 29 CFR 1926.1153, respirable crystalline silica in construction, which reaches sanding of a silica-bearing joint compound through the alternative exposure control methods at (d); its Table 1 lists concrete, masonry and stone tasks only.
• 29 CFR 1910.333(b)(2) and NFPA 70E-2021, 120.5, for de-energizing and proving ceiling circuits dead, with qualified-person scope at 29 CFR 1910.332.
• 29 CFR 1910.134, respiratory protection: medical evaluation, fit testing and a written program, reaching construction work through 29 CFR 1926.103.
• See related: the popcorn ceiling paint versus skim versus remove article, and the RRP containment and cleaning verification standard.