Industries / Construction and property / Commercial and residential construction / Pool construction
Plumbing and equipment
Plumbing and equipment is the pump, filter, heater, pipes and controls that draw water out of a pool, clean and warm it, and send it back.
7
Milestones achieved, from 1907 to 2025.
1
To watch for: announced, not yet confirmed outside.
2
Open deliverables. Nobody has shipped them.
The road
AchievedWatch forOpen
Achieved
1907
Pools drained to stay clean
Early pools were "fill and draw": filled, used, then emptied and filled again. A 1907 builders' handbook from the Chicago trade journal Domestic Engineering said the water should be filtered as it came in and changed often, daily if possible, and described one pool with 2,250 bathers a month whose water was changed twice a week. Its 1915 edition added that where water was costly and the pool had to be emptied daily or at short intervals, "a re-filtering and sterilizing plant" was needed as well.
Lessonthe cost of fresh water, not only health, pushed pools toward cleaning and reusing their water.
Credit
John K. Allen, with Alfred G. King for the 1915 edition
1946
Recirculation becomes the rule
The Portland Cement Association's 1946 guide to concrete pools sorted them into fill-and-draw, flow-through and recirculating pools. It called fill-and-draw "the old" system, said most health authorities would not allow it in new public pools, and found small private pools to be about the only ones still built that way. A recirculating pool pumps its water through filters and back without stopping; the guide's example used a turnover, the time to pass the whole pool through the filters once, of 8 to 12 hours, and Illinois set a pool's daily bather limit from that turnover time.
Lessonthe guide found refilling cost "considerably more" than recirculating, besides keeping the pool closed while the water was changed.
Credit
Portland Cement Association; Illinois Department of Public Health
1996
Slower pumps tested in yards
A pump's power falls much faster than its flow as it slows: at half speed it moves half the water on about one-eighth of the power, the pump affinity law as a 2012 guide for the US Department of Energy's Building America program explains it. In California, Davis Energy Group and Southern California Edison fitted two-speed motors to the pool pumps of five homes and doubled the filtering hours, so the same volume of water was filtered each day. Energy use fell by 38% to 65%, with no loss of water clarity.
Lessonrunning a pump slowly for longer filters the same water for a fraction of the energy.
Credit
David Springer (Davis Energy Group) and Fred Rohe (Southern California Edison), at the ACEEE Summer Study on Energy Efficiency in Buildings
2007
Pool pumps gain variable speed
A variable-speed pump's motor can run at any speed in a range, so a pool can be filtered slowly through most of the day. In 2007 Ike Hornsby, who ran Pools by Ike in Bakersfield, California, introduced the Dyna-Miser VS, built from a Sta-Rite pump head, a three-phase motor and a Square D variable speed drive; Gary Fernstrom and three colleagues from California's utility efficiency programs call it "the first variable speed pump". Pentair, Sta-Rite and Hayward brought out their own later that year, and Jandy-Zodiac in late 2008. Pentair's IntelliFlo used a permanent magnet motor and could hold a set flow as the filter clogged.
Lessona pool service contractor built the first one from parts already on the market, and the big makers followed within the year.
Credit
Ike Hornsby; Pentair, Sta-Rite, Hayward and Jandy-Zodiac
2021
A federal floor for pumps
Pool filter pumps made in or imported into the US from 19 July 2021 must meet minimum efficiency standards set by the Department of Energy (DOE), measured as a weighted energy factor: thousands of gallons pumped per kilowatt-hour, averaged over set test speeds. For standard-size self-priming filter pumps, DOE wrote in 2023, the standard would "likely require" variable-speed motors. The levels were agreed on 23 June 2016 by a working group of pump makers, electric utilities, efficiency advocates and the California Energy Commission, whose state had set design rules for pool pump motors from 2006 and controls from 2008. DOE estimated savings of 3.8 quads (quadrillion British thermal units) over 30 years of sales, 61% of what the pumps would otherwise use.
Credit
DOE and its dedicated-purpose pool pump working group, which included Hayward, Pentair and Zodiac; DOE found those three held about 90% of US shipments
2022
Australia rates every pump
From 1 October 2022 Australia's Greenhouse and Energy Minimum Standards (GEMS) determination for swimming pool pump-units requires each covered pump to reach a minimum weighted energy factor, in liters pumped per watt-hour of electricity, and sets the rules for its star energy rating label. Pumps are tested to the Australian Standard AS 5102.1:2019; single-speed pumps under 600 W or over 1,700 W, and multi-speed pumps under 600 W or over 3,450 W, are outside it.
Credit
Australian Government, under the Greenhouse and Energy Minimum Standards Act 2012
2025
Replacement motors must vary speed
DOE's 2021 standard covered whole pumps, not motors sold alone, and commenters warned that owners replacing a motor could fit one less efficient than the original. A 2023 DOE rule requires variable-speed control on pool pump motors of 1.15 to 5 total horsepower made from 29 September 2025, and on motors of 0.5 to under 1.15 horsepower from 28 September 2027, whether sold in a pump or on their own. DOE estimated savings of 1.56 quads.
Lessona standard on a finished product can be sidestepped through its replacement parts.
Credit
DOE
Watch for
Heat pump claims COP 30
A heat pump's coefficient of performance (COP) is the heat it delivers divided by the electricity it uses. A datasheet for the Norsup ProX 20, a pool heat pump that uses propane (R290) as its refrigerant, promises "Efficient COP up to 30" and calls it "the quietest & most energy efficient heat pump on the market today", next to a seasonal figure (SCOP) of 8.4 that it says TÜV SÜD certified. The same sheet's table gives a COP of 6.46 to 10.98 with air at 15 °C and water at 25 °C.
Achieved whenan independent lab publishes the test conditions behind the figure of 30, or measures a COP near it in a standard pool heat pump test.
Claimed by
Norsup, in a datasheet on the distributor Bosta's website, last modified 2 November 2026
Open
01
How pool pumps really run
A survey of the speeds and hours that installed variable-speed pool pumps actually run, read from their controllers' schedule screens, and the energy that implies, set beside the figures behind energy rules and utility rebates.
The briefCredit · unclaimed02
Leak or evaporation, tested
An independent accuracy test of the tools that tell an owner whether a falling water level is evaporation or a leak: weather-based evaporation calculators, level sensors that send leak alerts, and the bucket test, in which a bucket of pool water set on a step shows the evaporation alone.
The briefCredit · unclaimed
Open 01 · the brief
How pool pumps really run
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
A survey of the speeds and hours that installed variable-speed pool pumps actually run, read from their controllers' schedule screens, and the energy that implies, set beside the figures behind energy rules and utility rebates.
Why it is still open
A variable-speed pump saves energy only if it is set to run slowly. DOE's estimate of 3.8 quads of savings rests on modeled power and operating hours. The field measurements found are old and small: the 1996 Edison study metered five homes with two-speed pumps, and the load shape that California's utility programs use as their pool baseline comes from PG&E surveys in 2000 and an ADM Associates evaluation in 2002, before variable-speed pumps were sold, by Fernstrom and colleagues' account. A 2012 guide for DOE says real savings "can only be accurately determined after the variable speed pool pump has been installed". A search found no survey of how installed variable-speed pumps are set.
Built with what exists
Phone photos of each pump's schedule screen or app, taken with the owner's consent by the owner or a service technician; an AI model that reads each photo into a table of speeds and hours, such as Claude Fable 5.2 (see the Note "Three working devices"), with a person checking each row; each model's power at each speed from its maker's data sheet; Python for the tally.
First step
Photograph the pump schedules on one service company's route, about 30 pools in a week, and draft the table's columns from them.
Done looks like
300 pools in at least three climates, with hours a day at each speed, estimated kilowatt-hours a day, and the share of pumps that run mostly at high speed, measured in summer and again in winter and published as an open table.
Open 02 · the brief
Leak or evaporation, tested
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
An independent accuracy test of the tools that tell an owner whether a falling water level is evaporation or a leak: weather-based evaporation calculators, level sensors that send leak alerts, and the bucket test, in which a bucket of pool water set on a step shows the evaporation alone.
Why it is still open
Every pool loses water to evaporation, and a leak search costs money, so tools now make the call. A free calculator from a pool leak-detection trainer says it tells "exactly how much water your pool should be losing to evaporation" from local weather data; Kona Labs' LevelSmart sensors send "leak alerts". Neither page cites an accuracy test, and a search found only these tools, other calculators and patents describing level sensors compared with evaporation models, not a published test. The equations behind such calculators can miss by a wide margin: Charles Smith, George Löf and Randall Jones of Colorado State University measured evaporation from quiet pools at 74% of the standard equation's value, and from pools in use at 40% to 70% more than from quiet ones (ASHRAE Transactions, 1998).
Built with what exists
Two identical above-ground stock tanks; a needle valve and a measuring cylinder to drain known amounts as a simulated leak; ultrasonic or pressure level sensors on ESP32 boards; a home weather station; the free calculators and a consumer level sensor; Python.
First step
Fill one tank, log its level and the weather for two weeks with no leak, and compare the calculators' predicted loss with the measured loss.
Done looks like
60 tank-days with leaks of 0, 3 and 6 mm a day, set at random and hidden from the person running the tools, with each tool's missed leaks and false alarms published; then 20 home pools checked against what a leak-detection company finds.