The Last Metre of Hot-Water Pipe: How Branch Length Turns Waiting into Waste
A hot-water system may satisfy pressure, temperature and capacity calculations while still making a user run the tap before useful heat arrives. The cause is often hidden in the last branch between the heater, riser or recirculation loop and the fixture. Its contained volume must be displaced, while the pipe wall and surrounding environment absorb part of the arriving heat. This makes the final metres of routing an architectural question: equipment position, wet-core planning, ceiling coordination and fixture layout determine a daily experience long before the tap is selected.
What the original study examined
Roger W. Moss and Robert E. Critoph studied domestic hot-water delay and the performance of a recirculating alternative. Measurements in three existing properties recorded delays of up to 46 seconds after a hot tap was opened. The researchers then developed a transient plug-flow advection model for water and pipe temperature, validated it against a single-pipe test, and extended it to paired supply-and-return arrangements.
The study did not simply treat the pipe as a cold cylinder that empties once. It modelled heat exchange among the moving water, the pipe and the surrounding environment. That distinction matters because the first water leaving a branch may be warmer than the standing water yet still below a usable threshold. The authors’ model was designed to optimise return-pipe diameter, flow and insulation rather than to declare one universal recirculation layout.
Measured result and modelled result are different
- Measured: the three properties showed hot-water arrival delays reaching 46 seconds.
- Modelled: for a 15-metre pipe, the proposed recirculation arrangement was expected to require about 30 W less electrical power than a local thermal store at the delivery point.
- Modelled system behaviour: calculated pumping power was below 1 mW under the analysed conditions, and the authors considered a thermosyphon arrangement possible in some installations.
These figures belong to the configurations and assumptions analysed by Moss and Critoph. They should not be copied into a project specification without checking pipe sizes, temperature set points, insulation, draw patterns and climate.
The pipe volume is only the starting point
The geometric volume of a branch is:
V = π × D² × L ÷ 4
where D is internal diameter and L is branch length. Every additional metre therefore adds a predictable standing volume, but the water discharged before reaching a usable temperature can be larger. Experimental work synthesised in the ASHRAE Handbook reports that actual discarded flow commonly ranges from about 1.0 to 2.0 times pipe volume after cooling. For preliminary calculations, Carl C. Hiller proposed a ratio of 1.25 to 1.75, while warning that long, uninsulated branches at low flow can perform much worse.
| Internal diameter | Standing volume per metre | Standing volume in an 8 m branch |
|---|---|---|
| 10 mm | 0.079 L | 0.63 L |
| 12 mm | 0.113 L | 0.90 L |
| 15 mm | 0.177 L | 1.41 L |
| 20 mm | 0.314 L | 2.51 L |
These are geometric calculations, not measurements from the 2022 study. As an illustration, an 8-metre branch with a 15 mm internal diameter contains about 1.41 litres. Applying Hiller’s preliminary 1.25–1.75 range gives roughly 1.77–2.47 litres discharged before usable hot water arrives. The estimate excludes heater ignition delay, valve behaviour and any cold-water crossover.
Architectural application: design the branch, not only the plant
The evidence supports a sequence of project decisions rather than a universal product choice:
- group fixtures with genuine hot-water demand around short, legible distribution routes;
- locate heaters, risers or recirculation take-offs using contained volume and predicted wait time, not plan distance alone;
- avoid oversizing remote branches as a precaution when the larger diameter increases standing volume;
- compare a compact trunk-and-branch layout, demand-controlled recirculation and point-of-use heating using the same draw profile;
- coordinate access to insulation, balancing devices and controls so performance can be maintained after handover.
Recirculation is not automatically the answer. A continuously warm loop can reduce waiting while increasing distribution heat loss. Timers, demand control or a thermosyphon may improve the balance, but each option needs its own energy, hygiene, control and maintenance assessment.
Limits of the evidence
The field measurement involved only three properties, so the observed 46-second maximum is evidence of the problem, not a population estimate. The model was validated for a single pipe before being extended to recirculating arrangements; full-building behaviour still depends on simultaneous draws, installation quality and control settings. The study focused on older housing and heat-pump conversion conditions in the United Kingdom, which limits direct transfer to other climates and building types.
Hiller’s waste ratios are also design aids, not immutable constants. Pipe material, internal roughness, fittings, starting temperature, insulation, flow rate and the chosen “usable” temperature all change the result. Low-flow fixtures reduce discharge rate but may increase the time needed for hot water to arrive; water and time do not necessarily improve in the same proportion.
How to verify the design
- Measure developed pipe length and internal diameter to every critical fixture.
- Calculate contained volume and state the assumed usable temperature.
- Predict wait time and discarded volume for the specified fixture flow rate.
- During commissioning, record inlet and outlet temperature and flow at one-second intervals after a defined cool-down period.
- Compare the measured result with the design threshold, then adjust controls, balancing or routing where practicable.
A pass/fail requirement can be simple: maximum seconds to a declared temperature at a declared flow, after a declared idle period. Without those three conditions, “instant hot water” is not a testable promise.
✦ ArchUp Editorial Insight
Coordination pressure, rather than user preference, often produces this waste. When plant rooms and risers are fixed early but kitchens and bathrooms keep moving, the hot-water branch absorbs the change: an extra turn, a longer ceiling route, or a wider pipe selected as hydraulic insurance. Procurement and programme then reward successful pressure testing, not the seconds between opening a tap and receiving usable water. The decision is rational within each package, yet its recurring cost moves to the occupant, who pays in water, energy and time after the design team has left. The same transfer appears when an air-conditioner drain is allowed to terminate above a pedestrian route: a small service endpoint becomes an everyday spatial condition because no single party owns the interface. A project response would therefore be institutional, not cosmetic: record contained volume and predicted wait at each remote fixture in the design review, assign a maximum performance threshold, and repeat the measurement during commissioning. This is an ArchUp hypothesis for practice, not a tested contractual model. Its value is that it makes a hidden coordination choice auditable before finishes conceal the branch and responsibility becomes diffuse.
Prepared by ArchUp Research Lab. The editorial image is awaiting preparation by the art direction team.
References
Roger W. Moss and Robert E. Critoph, “Optimisation of a Recirculating Domestic Hot Water System to Minimise Wait Time and Heat Loss,” Energy and Buildings, 2022.
Carl C. Hiller, “Hot-Water Distribution System Piping Time, Water, and Energy Waste: Phase I Test Results,” ASHRAE Transactions, 2005.
James D. Lutz, Peter J. Biermayer and Derek A. King, “Pilot Phase of a Field Study to Determine Waste of Water and Energy in Residential Hot-Water Distribution Systems,” Lawrence Berkeley National Laboratory Report, 2011.
