Tile Line and Calcium Scale Removal
Purpose
The white band at the water line is two different deposits that look identical from the deck, and picking the wrong one costs either a wasted afternoon or a customer's tile. Calcium carbonate dissolves in acid and comes off with a wet stone. Calcium silicate does not respond to acid at any strength a technician should be using, so a tech who keeps stepping the acid up etches the grout, dulls the glaze and still has the scale. This procedure guarantees the deposit is identified by test rather than by eye, the substrate is checked before any abrasive or acid touches it, and the boundary between hand removal and media blasting is drawn before the customer is quoted.
Scope
Covers the water-line tile band, the coping face where scale runs onto it, and the first foot of interior finish below the tile on residential and small semi-commercial pools: identification, hand removal with a wet stone, acid gel work, and the handoff when the deposit is silicate.
It does not cover media blasting itself, which is a lowered-water scope run by an equipped contractor and is quoted, not attempted off a service truck. It does not cover the chemistry that made the scale, which the calcium hardness and scale management reference owns, nor whole-surface acid washing, which the drain and acid wash standard owns. Acid dilution and transport rules live in the muriatic acid handling standard.
Roles and responsibilities
| Role | Owns | Hands off |
|---|---|---|
| Lead technician | The drop test, the substrate check, water level, stone and gel work | Hands the deposit identification to the service manager BEFORE any quote is given, not after a stalled visit |
| Service manager | Scope call between hand removal and blasting, the blasting subcontract, customer disclosure on glaze and grout | Hands the customer the identification result and the realistic outcome in writing before work is scheduled |
| Route technician | The follow-up chemistry that decides whether it comes back | Hands the saturation trend to the service manager if scale reappears inside one season |
Procedure
Identify the deposit with an acid drop test before choosing any method. Put one drop of dilute muriatic acid from a squeeze bottle on a dry patch of the deposit, above the water line, and watch. Accept the deposit as calcium carbonate when it effervesces visibly within about 5 seconds. No reaction at all means calcium silicate, which is the harder deposit and does not surrender to acid. Wrong is skipping the test because the band "looks like scale every other pool has": that assumption is what puts a tech three hours into a stone on silicate. Hazard is one drop of acid in a squeeze bottle at eye height, so goggles on and the bottle held below the face, never above it.
Identify the tile and grout, and gate the method on the substrate rather than the deposit. Glazed ceramic and porcelain tolerate a wet stone and a short acid contact. Glass tile, natural stone coping, and polished travertine or limestone do not: those stones are carbonate themselves and acid dissolves the finish, while a stone abrades the polish. Accept the mechanical or acid path only when the tile is glazed ceramic or porcelain and the grout is sound with no washed-out joints. Wrong looks like an etched dull halo around each tile where the acid sat. Stop and route stone or glass to a supplier-approved product. Hazard at this step is leaning over the coping to read grout: kneel on the deck, do not stand on the bond beam.
Shut the pump down and lower the water below the band, and record the starting level. Open the pump disconnect and apply a lock and tag under 29 CFR 1910.147 before dropping level, because the hazard being controlled is the pump running itself dry and the suction at an outlet you may reach into. Accept when the water is 6 to 12 in below the lowest scaled tile and the pump is confirmed off at the impeller, not at the timer. Wrong is a pump left in auto that starts on a dropped level and burns its seal in minutes. On a vinyl pool this step does not run at all: the liner shrinks and the vinyl standard owns that exclusion.
Set the exposure controls before the first stone or the first gel goes on. Splash goggles, a face shield over them, chemical-resistant gloves and boots. Work upwind and never open acid near hypochlorite residue or a floating chlorinator, because acid on hypochlorite releases chlorine gas at concentrations that drop a person at the coping. Accept when the chlorinator is out of the pool, the band is hosed clear, and a running hose is in reach. Wrong is a tech working a band with a floater bumping the wall. Where the label or safety data sheet names respiratory protection, that means a cartridge respirator under a written program per 29 CFR 1910.134 or the work does not happen.
Work carbonate off with a wet stone, starting on a test patch out of sight. Keep the stone and the tile wet at all times, because a dry stone scores the glaze and dry work puts the removed material into the air rather than into the water. Accept the test patch when the deposit lifts and the glaze under it still reflects light evenly at a low angle. Wrong is a hazed or scratched patch: stop there and change methods rather than continuing along the band. Hazard is the dust route if the surface dries out, so re-wet before every pass, and a wet coping edge underfoot, so work from inside the pool or kneeling on the deck, not straddling the edge.
Where the stone alone will not lift it, apply an acid gel with a stated dwell and rinse on the clock. Brush the gel onto a section no longer than you can rinse in one pass, hold it to the manufacturer's stated dwell, and flood-rinse immediately. Accept when the deposit releases and the grout joints are still flush and hard under a fingernail. Wrong is grout that turns sandy or recedes below the tile face: stop immediately, rinse the whole band, and hand the job back to the service manager, because grout removal is a re-grout, not a cleaning. The label directions are legally binding on dwell and dilution, so follow the label on the tin rather than a remembered ratio.
Where acid does not touch it, stop and quote blasting rather than escalating strength. A deposit that failed the drop test at step 1 and does not move under gel is silicate, and the only removal that works is a media blast, glass bead or sodium bicarbonate, with the water down and containment set. Accept by writing the identification and the referral on the ticket. Wrong is stepping the acid toward full strength on a customer's tile: it dulls glaze, cuts grout, and still leaves the silicate. Blasting liberates respirable dust from the substrate and needs hearing and eye protection plus a respirator under 29 CFR 1910.134, so it is a scoped job, not a truck task.
Restore the level and the pump, then close the chemistry loop or it comes back. Refill to the middle of the skimmer mouth, remove the lockout only when no one is in the pool and the outlet covers are confirmed in place with every fastener, and restart from beside the strainer pot rather than over it. Accept when the pump primes, returns run clear, and a fresh test shows pH back inside the operating band after the acid work with calcium hardness and total alkalinity recorded. Wrong is a pool restarted at a pH the acid dropped into the fives, which etches the plaster you just cleaned around. Where the saturation trend caused the scale, route it to the partial drain and dilution decision rather than scheduling this visit again next year.
Why the substrate gates the method
The instinct is to pick the method from the deposit, and it is backwards. The deposit tells you what will dissolve it; the substrate tells you what you are allowed to do to the surface underneath. A glazed porcelain tile with hard grout takes a wet stone and a short gel contact and looks new. The same deposit on a glass mosaic band is untouchable by both, because the glass scratches and the joints in a glass installation are usually more delicate than the tile.
Natural stone is the trap most shops learn once. Polished travertine and limestone coping are calcium carbonate, the same family as the scale sitting on them, so an acid that removes the deposit removes the polish with it and leaves a dull patch no rinse restores. That is a refinishing bill on a coping nobody asked to have touched. When a tech cannot identify the tile, the answer is a photograph to the service manager, not a test on a visible section.
The record this produces
The scale record, attached to the property, because a shop cannot tell whether treatment worked without knowing what the band looked like and what the water was doing when it formed.
Fields: photographs of the band before and after at the same spot and angle; the drop test result stated as effervesced or did not; tile and grout type with the substrate verdict; water level lowered from and to, in inches, plus the pump lockout time on and off; method used per section, with the gel product name and the dwell taken from its label; the test patch result; grout condition after; refill level; and the post-work pH, total alkalinity, calcium hardness and water temperature.
The service manager reads the drop test result to know whether the next quote is a cleaning or a blasting subcontract. The route technician reads the closing chemistry to see whether the water is still depositing. The office keeps the before-and-after photographs, because glaze condition is the argument this job generates.
Worked pass: a mixed band on a plaster pool with hard water
A 16 by 32 ft plaster pool, 512 sq ft of surface, roughly 1/8 in of white deposit across the tile band, heavier at the spillway end. Customer wants it gone before a party.
Step 1: acid drop on the light film at the shallow end effervesces within about 2 seconds, so that film is carbonate. A second drop on the heavier deposit at the spillway end shows no reaction at 30 seconds, so that section is silicate. The band is mixed, which is the common real case and the reason the test is run at more than one point.
Step 2: tile is glazed 6 by 6 porcelain, grout joints flush and hard under a fingernail. Substrate gate passes for both stone and gel.
Step 3: pump disconnect opened and locked at 09:05, impeller confirmed stopped. Water dropped 8 in below the lowest scaled tile with a submersible discharging to the sanitary cleanout. At 512 sq ft and 0.623 gal per sq ft per inch, one inch is about 319 gal, so 8 in is 8 times 319, about 2,552 gal removed.
Step 4: floating chlorinator pulled out and set on the deck away from the acid, band hosed clear, goggles, face shield, gloves, boots, hose running.
Step 5: test patch behind the ladder with a wet stone. The carbonate film lifts and the glaze under it still reflects evenly at a low angle, so the shallow-end run proceeds. About two thirds of the band cleans up this way.
Step 6: gel applied to the spillway-end section at the label dwell, rinsed on the clock. The deposit does not release. Grout is checked and is still flush and hard, so no grout stop rule fires.
Step 7 FAILS as designed, and it is the step this visit exists to reach. The spillway-end deposit failed the drop test at step 1 and did not move under gel, so it is silicate. The acid is NOT stepped up. The identification and referral go on the ticket, the service manager quotes a bead-blast subcontract with the water down, and the customer is told before the party rather than after a second failed visit.
Step 8: refilled to mid-skimmer, outlet covers confirmed with all fasteners, lockout removed 13:40, pump restarted from beside the pot and primed. Post-work test reads pH 7.0 after the gel work, so soda ash brings it to 7.5 before the tech leaves. Calcium hardness reads 620 ppm, which is the reason the band formed at all, so the ticket also opens a dilution recommendation rather than treating this as a finished job.
References
- The safety data sheet and the printed label directions for the specific acid gel or tile cleaner on the truck; label dwell time and dilution are binding, and a remembered ratio is not.
- 29 CFR 1910.134 for the written respiratory protection program covering any respirator a product label or a blasting scope requires, including medical evaluation and fit testing.
- 29 CFR 1910.147 for lockout of the circulation pump while the water level is down, where the controlled hazard is pump motion and outlet suction.
- See related: the PoolService calcium hardness and scale management reference for the saturation chemistry, the muriatic acid handling and dilution standard, and the partial drain and dilution decision standard.