Heat Pump Pool Heater Service Standard

Purpose

This procedure guarantees a heat pump complaint is judged against what the unit can actually deliver at today's air conditions, with the air and water sides verified before anybody suggests a refrigerant problem. Two failures pay for it: a unit condemned as "low on gas" when the fault is a starved evaporator coil or a pump speed somebody lowered, and a customer told the unit is fine because it runs when at the measured ambient it delivers a fraction of the table value and the pool never reaches setpoint.

Scope

Covers seasonal and diagnostic service of a residential or light commercial electric heat pump pool or spa heater: isolation, stored-energy discharge, ambient and flow readings, air and water side service, condensate discrimination, output verification and referral.

It does not cover opening the sealed refrigerant circuit, recovering, charging or leak-searching with gauges, which require certification under 40 CFR Part 82 Subpart F; gas-fired heaters, which have their own SOP; or the pump programming that supplies the flow, which the variable speed pump SOP owns.

Roles and responsibilities

Output is an ambient number, not a nameplate number

A heat pump does not make heat, it moves heat out of outdoor air into the pool water, so capacity falls as that air gets colder and drier, and published capacity is stated at a standard rating condition - a specified ambient dry bulb, wet bulb and inlet water temperature. Read the capacity off the manufacturer's table at the ambient, humidity and water temperature you measured, and record the rating condition beside it. A nameplate figure against a 55 F morning compares two different machines.

Delivered heat is gpm times 500 times the rise across the unit in degrees F, where 500 is 8.33 lb per gallon times 60 minutes per hour times water's 1 Btu per lb per degree F at ordinary pool water conditions. Because pool flow is high relative to a heat pump's output, the rise is small: a unit delivering 120,000 Btu per hour at 40 gpm produces 120,000 divided by 40 divided by 500, which is 6.0 F. A tech expecting a gas heater's rise will call a healthy heat pump dead.

Frost on the evaporator at low ambient is expected and the defrost cycle handles it; a frosted coil at 60 F ambient is not, and a steady drip while running is condensate, settled at step 6. Set a shop default and hold it: delivered heat within 10 percent of the table value at the measured condition passes, anything short with the air and water sides verified is a referral.

Procedure

1. Open the disconnect, lock and tag it, and prove the load terminals dead on a meter proved on a known live source immediately before and after. Work practices are 29 CFR 1910.333(b)(2), not 1910.147, which excludes electrical utilization work at its own (a)(1)(ii)(C); 1910.333(b)(2) is written for a qualified person as 1910.332 defines one, so if nobody on site is qualified the visit stops at the disconnect and isolation goes to an electrician. Proving sequence per NFPA 70E-2021, 120.5. Accept when every conductor combination reads zero and the meter re-reads correctly on the known source. Wrong is a dead reading from a dead meter. The fan starts on a control call with no warning and no local switch involved, so a thermostat satisfied is not isolation.

2. Discharge the capacitors per the manufacturer's method and confirm zero volts across their terminals before a hand goes near the control box. Accept when the meter reads zero across each capacitor and stays there. Wrong is assuming a locked-out unit is a de-energized one: a run capacitor holds a charge after the disconnect opens, and that is the hazard this step exists for. Stop and get a second opinion rather than shorting terminals with a screwdriver, which damages the capacitor and throws molten metal.

3. Record the conditions the visit will be judged against before touching anything mechanical. Ambient dry bulb, relative humidity or wet bulb, pool water temperature, setpoint, and the capacity-table entry at exactly those conditions with the rating condition noted. Accept when all five are on the ticket, sourced to the table by model. Wrong is writing down the nameplate figure. Hazard: none here, instrument reading with the unit locked out.

4. Verify the water side and measure actual flow through the unit. Filter differential against its recorded clean baseline, valve positions, the pump's carded speed, and flow at the heat pump. Accept when measured flow is at or above the unit's published minimum. Wrong is a unit starved by a reduced pump speed or a loaded filter, which produces the symptom the customer called about. Stop rule: below the published minimum, do not judge output at all - restore flow per the variable speed pump SOP's speed card or clean the filter, then re-run this step and step 7.

5. Clean the evaporator coil, clear the air path, and check the fan. Wash the coil from the inside out at the dilution and dwell the label states, comb bent fins, clear leaves and clippings from the base, and confirm the clearances the installation instructions require. Accept when light passes through the fin pack across its face, the fan turns freely with no bearing noise, and every stated clearance is met. Wrong is a pressure washer, which lays the fins over and drives water into the control box. Coil cleaner is a hazard on three routes at once - mist to the airway, splash to the eye, contact to skin - so stand upwind, wear sealed eye protection and chemical-resistant gloves, and follow the label rather than a remembered dilution; any label-specified respirator waits for a written program under 29 CFR 1910.134. When it reaches someone anyway: an eye splash is flushed for 15 minutes per the safety data sheet and then seen by a clinician, skin contact is washed under running water with the wetted clothing off, and anyone who breathes the mist and starts coughing leaves for fresh air and is evaluated rather than finishing the coil. Send the SDS with them, because these cleaners run strongly alkaline or strongly acidic and are treated differently.

6. Settle any reported leak before it becomes a warranty claim. A running heat pump condenses water out of the air it cools, and that puddle is the most common false leak call on this equipment. Test the drip for chlorine with a strip or drop test against pool water. Accept when the drip is chlorine free, tracks the run cycle, and leaves the base pan through the intended drain. Wrong is a puddle carrying chlorine at pool concentration: a water-side leak inside the cabinet. Stop rule: chlorinated water in the pan means shut the unit down, isolate the water side and quote the exchanger, because a leaking exchanger in a running cabinet puts pool water on live terminals.

7. Re-fit the panels, restore power, run to steady state, and measure what it is actually delivering. Panels go back on before the unit starts, because line voltage and a large fan blade return at once: nobody stands in the discharge and no hand or tool enters the cabinet while it turns. Take running amps with a clamp on one line conductor at the disconnect, cabinet closed and body clear of the fan path - energized work reserved for step 1's qualified person under 29 CFR 1910.333(b)(2), in rated eye protection and gloves. If the only clamp point is inside the running cabinet, the amps are not taken: record that and judge the visit on delivered heat alone rather than opening a turning unit. Accept when running amps sit at or below nameplate, the rise times measured gpm times 500 lands within 10 percent of the table value from step 3, the fan discharges freely, and the condensate drains away. Wrong is reading rise in the first two minutes, before steady state.

8. Close on a number, and refer rather than guess. Where delivered heat clears the 10 percent band, record it, state the hours to setpoint at that output, and release. Where it falls short with flow verified in step 4 and the coil clean in step 5, refer to a Section 608 certified technician in writing with every reading attached. Wrong is connecting gauges without certification, which is a violation under 40 CFR Part 82 Subpart F and the fastest way to lose a charge nobody had measured. Never open a refrigerant circuit or apply flame near one: liquid refrigerant freezes skin on contact, it displaces oxygen in a pit or an enclosed pump room, and it decomposes on hot surfaces into hydrogen fluoride and other acute airway irritants.

The record this produces

One heat pump block on the work order, filed to the equipment record: the five conditions from step 3 with the table capacity and its rating condition; filter psi against clean baseline, pump speed, measured gpm and published minimum; coil before and after with cleaner and dilution, fin and fan condition, clearances; the chlorine result on any drip and its drain path; running amps against nameplate and where clamped, or recorded as not taken; measured rise, delivered Btu per hour, table value and percentage of table; and the disposition.

The shop lead reads the percentage-of-table line across visits: the same unit at 95 percent this spring and 71 percent next spring is a conversation before it is a breakdown.

One service, filled in

An 18,000 gallon pool with a heat pump rated 120,000 Btu per hour at its standard rating condition. Complaint: will not reach the 84 F setpoint. Step 3: ambient 62 F dry bulb, 55 percent relative humidity, pool water 74 F. The capacity table at those conditions gives 84,000 Btu per hour, and that is the number the visit is judged against, not the 120,000.

Step 4 fails. Measured flow is 22 gpm against a published minimum of 30 gpm. Filter reads 12 psi against a 10 psi clean baseline, so it is not the cause; the pad's speed card says filtration runs at 2,000 rpm and the drive is set to 1,450, changed at the panel by the homeowner. The stop rule holds: no output judgment on a starved unit. Speed restored to the carded 2,000 rpm, flow re-measured at 38 gpm, above the 30 gpm minimum, and steps 4 and 7 re-run.

Step 5: the coil face was matted with grass clippings, washed at the label dilution and dwell, fins combed, and the fence line cut back to the stated clearance. Step 6: the reported drip tested chlorine free and stopped within twenty minutes of shutdown, so it is condensate.

Step 7 passes. Panels back on first, running amps clamped at the disconnect with the cabinet closed: 21.4 A against a 24.0 A nameplate at 240 V. At steady state the rise measures 4.2 F at 38 gpm, so delivered heat is 38 times 500 times 4.2, which is 79,800 Btu per hour. Against the table's 84,000, that is 79,800 divided by 84,000, or 95 percent, inside the 10 percent band.

Step 8: released. The customer's real question is how long the pool takes to come up 10 F. That is 18,000 times 8.33, or 149,940 lb of water, times 10 F, which is 1,499,400 Btu, divided by the measured 79,800 Btu per hour, about 18.8 hours - and that is a floor: it ignores heat lost off the surface while the unit runs, which is why the cover conversation happens at the same visit.

References

• 40 CFR Part 82 Subpart F for the certification required to work on the refrigerant circuit.
• 29 CFR 1910.333(b)(2) for electrical work practices, qualified person at 1910.332; 1910.147 excludes electrical utilization work at (a)(1)(ii)(C). NFPA 70E-2021, 120.5, for live-dead-live proving.
• 29 CFR 1910.134 for the written respiratory program any label-specified respirator depends on.
• The manufacturer's capacity table, rating condition, minimum flow and capacitor discharge method for the model; the cleaner's label and SDS for dilution, dwell, protection and first aid.
• See related: the variable speed pump SOP for the speed card and minimum flow; the gas heater service and combustion check SOP.