Faucet and Supply Line Replacement

Purpose

The faucet is rarely what fails after this visit. What fails is the stop the tech turned and the supply line the tech reused, and both failures arrive after everyone has gone home, under a cabinet, onto a particleboard floor that swells before anyone notices.

The other thing this call gets wrong is treating the cabinet as plumbing. It is plumbing plus a disposal, a dishwasher and a receptacle, with the tech's head and shoulders inside it and water about to be released.

This procedure guarantees electrical isolation before anyone reaches in, that every stop operated on the visit leaves it either replaced or proved to hold, that no supply line is ever reused, and that the delivered hot temperature is measured and recorded rather than discovered by the customer.

Scope

Covers kitchen, bar, laundry and lavatory faucets in dwellings and small commercial restrooms and break rooms: faucet replacement, cartridge and aerator service, sprayers and escutcheons, supply line replacement, and replacement of angle and straight stops on existing stub-outs.

Does not cover valve bodies inside a wall, tub and shower valves, water heater work or temperature setting, pressure-reducing valves, soldering inside a finished cabinet, disposal replacement, or any stub-out that has to be cut back or opened into the wall, all of which go to a plumber.

Roles and responsibilities

Procedure

1. Isolate the electricity in the cabinet before your head goes in. Acceptance: the disposal and dishwasher circuits opened at the panel, the cabinet receptacle proved dead on a known live source before and after the test per NFPA 70E-2021, 120.5, with work practices at 29 CFR 1910.333(b)(2), and the breaker held open by a lockout device or, where the panel will not take one, taped with a note and the customer told. What wrong looks like: unplugging the disposal and calling the receptacle safe, which leaves a live device an inch from where a line is about to be opened. Stop rule: if nobody on site is qualified as 29 CFR 1910.332 and 1910.399 define one, nothing gets proved dead; the disposal is unplugged and the receptacle is treated as live all visit. Hazard: this cabinet is where water and energised parts meet, so isolation comes first and the cabinet floor is dried before anything is re-energised at step 7.

2. Find the main, identify both stops, and measure the delivered hot temperature. Acceptance: main shutoff located and confirmed operable, each stop identified as compression, threaded or integral, stub-out material named, and the hot temperature at the faucet measured after a two minute run and written on the ticket. What wrong looks like: guessing a compression stop on a threaded stub-out and finding out with the water on. Stop rule: a main that cannot be found or will not operate ends the visit before any joint is opened. Hazard: water delivered near 140 F scalds badly in a couple of seconds, so run the hot side off and let the line cool before cracking the first joint; if a scald happens, cool running water goes on it immediately and anything blistering is a medical call.

3. Prove each stop holds, then replace or document it. Acceptance: stop closed, faucet opened, flow stopping completely within about 30 seconds and no drip at the spout after 2 minutes with the faucet still open. Shop standing rule: any multi-turn stop that had to be forced, or that is older than the fixture it serves, is replaced on this visit rather than reused. What wrong looks like: a stop that slows to a pencil stream and is accepted as closed. Stop rule: a stop that will not hold means closing the main and replacing it at step 4; a main that will not hold ends the visit and the customer is told a plumber comes first. Hazard: forcing a seized stop shears the stem and releases the line, so hold the valve body with a second wrench and stop at first resistance.

4. Replace the stop with the type the stub-out actually takes. Acceptance: stub-out material confirmed, stop type matched to it, a new ferrule on any compression joint, and the nut made up to the turns past hand tight the stop manufacturer's instructions state rather than to feel. What wrong looks like: a new ferrule crushed onto an old one that was never removed, which weeps at low flow and reads as a faucet problem. Stop rule: a threaded stub-out whose nipple turns in the wall, or any corroded galvanized stub, goes to a plumber, because turning that nipple opens the wall. Hazard: no torch work happens inside a finished cabinet on this procedure, because an open flame against the cabinet back, its insulation and a plastic drain is a fire with the tech lying under it; joints are compression or press, or the work goes to a plumber.

5. Remove the old faucet and clean the deck down to bare surface. Acceptance: mounting nuts off, old sealant and gasket residue removed, the deck inspected under the escutcheon footprint and any crack, soft spot or delamination photographed. What wrong looks like: a new escutcheon set over a hairline crack in a cultured-marble deck, which becomes the shop's crack the day it opens. Stop rule: a cracked or soft deck stops the install until the customer is told and it goes on the ticket, because the new faucet's nut is what finishes that crack. Hazard: overhead work under a sink with old sealant, grit and standing trap water above the face, so eye protection goes on before reaching in and the trap is drained into a pan.

6. Set the faucet with the manufacturer's sealing method and fit new supply lines. Acceptance: the gasket or sealant the instructions specify rather than silicone by habit, the spout square to the sink and centerd in its hole, mounting hardware tightened per the instructions, and new supply lines long enough that each runs in a gentle sweep with no kink and no bend tighter than the minimum radius printed on the packaging. What wrong looks like: a line stretched straight between stop and inlet, which loads the nut every time the faucet is used. Stop rule: no supply line is ever reused, and a line that has to be pulled to reach is the wrong length. Hazard: cut braided stainless ends stand up like needles, so handle a used line by the fittings and bag it rather than laying it in the cabinet.

7. Restore water, flush before the aerator goes back, then restore power and prove the protective device. Acceptance: each stop opened slowly with a hand on the stem, the aerator left out while hot and cold each run 60 seconds to clear debris, no weep at either packing nut after 60 seconds, and a paper towel dry at every joint at 5 minutes and again at 15. Then the cabinet floor is dried, the breakers are reclosed with the tech standing to the hinge side of the panel rather than square in front, disposal and dishwasher are confirmed running, and the receptacle's GFCI protection is proved on its own test button and reset. What wrong looks like: the aerator fitted before flushing, which packs the screen with stop debris and reads as low pressure. Stop rule: a weep, a wet towel, or a GFCI that will not trip and reset fails the step. Hazard: energising over a wet cabinet is the mechanism here, which is why the drying comes before the breaker.

8. Run the fixture out, check for cross-flow, and hand over on the temperature you measured. Acceptance: both handles through full travel, hot delivered on hot with cold closed and cold on cold with hot closed, no temperature drift at either extreme, the cabinet floor dry after a 10 minute run with a towel under each joint, and the delivered hot temperature re-measured and written beside the step 2 figure. What wrong looks like: a mixed spout running warm on full cold, which is a cross-connection through the new cartridge and gets worse. Stop rule: any drip, any cross-flow, or a delivered temperature higher than step 2 recorded is not handed back. Hazard: the customer runs this tap next, so if the measured delivery sits at a scald temperature, say so out loud and route the setting to the water heater rather than adjusting a heater you were not called for.

The record this produces

What was isolated at the panel and how it was proved, main shutoff location, stub-out material, each stop's hold test result and whether it was replaced, the make-up turns used on any compression nut with the sheet they came from, supply line length, deck condition with photos of any crack, the flush duration, the 5 and 15 minute dry checks, the GFCI test result, the cross-flow check, and the delivered hot temperature before and after.

The office reads the stop line when the customer calls about a wet cabinet, because a stop documented as replaced and dry at 15 minutes is a different conversation from one documented as reused.

Worked pass: kitchen faucet swap, 1987 house, copper stub-outs

Complaint: kitchen faucet dripping and the sprayer will not shut off. Disposal and dishwasher both in the cabinet.

Step 1: both circuits opened at the panel, the cabinet receptacle proved dead on a known live source before and after, breaker held with a lockout device. Step 2: main located at the front basement wall and confirmed operable. Both stops multi-turn compression on 1/2 in nominal copper stub-outs. Hot temperature at the faucet after a two minute run measured 138 F, written on the ticket.

Step 3 FAILED on the hot side. The cold stop closed fully and showed no drip at the spout after 2 minutes with the faucet open. The hot stop slowed to a pencil stream and was still running at 30 seconds. Under the stop rule the tech closed the main rather than working against a live line, and both stops went into scope, not just the failed one, because the second was the same age and had to be forced.

The office was called before any joint was opened. Two quarter-turn stops added about 45 minutes to a booking of an hour and a half, authorized by the customer on that call.

Step 4: old ferrules cut off rather than crushed under new ones, new quarter-turn compression stops fitted, nuts made up to one and a quarter turns past hand tight as read off the stop manufacturer's instruction sheet in the box. No torch came out of the truck.

Step 5: deck cleaned to bare stainless, no cracking under the old escutcheon footprint. Step 6: stub-out to inlet measured 16 in, so 20 in braided lines were fitted, each running in a sweep with no bend tighter than the minimum radius on the packaging. The faucet's own gasket was used rather than silicone.

Step 7: stops opened slowly with a hand on each stem. Aerator out, hot and cold each flushed 60 seconds, and grit did come out of the hot side. No weep at either packing at 60 seconds; towels dry at both stops and both inlets at 5 and 15 minutes. Cabinet dried, breakers reclosed from the hinge side, disposal and dishwasher both ran, GFCI tripped and reset on its own button.

Step 8: full travel both handles, no cross-flow at either extreme, cabinet dry after a 10 minute run. Delivered hot re-measured at 138 F, the same as step 2, which is expected because nothing on this visit touched the heater. The customer was told the figure sits above 120 F and that the setting is a water heater question.

References

• The faucet, stop and supply line manufacturers' installation instructions for sealing method, make-up turns past hand tight, and minimum bend radius on the products in the box
• NFPA 70E-2021, 120.5 for the live-dead-live proving sequence, with work practices at 29 CFR 1910.333(b)(2) and the qualified-person definitions at 29 CFR 1910.332 and 1910.399
• The plumbing code your jurisdiction has adopted, in its current edition, for stop and connection requirements on the stub-out material you find
• See related: Faucet Replacement Basic; Which Joint to Check First on a Leaking Supply Line; Toilet Repair and Replacement Standard