The thermometer reads the same in both. A back-street office and a 600-bed hospital chase identical figures - hot water hot, cold water cold - yet almost everything about reaching and proving those figures changes with the size of the building. Miss that, and you either bury a small site under monitoring it never needed, or run a large one on a handful of checks that quietly skip whole wings.
So the real question is not which temperatures to hold. Those are settled. It is how much system sits between your heat source and the tap, how much of it one round of readings can honestly represent, and which kind of failure you should be watching for. Those three things scale hard.
The targets are the same; the system isn’t
UK practice points everyone at the same benchmark. HSE’s hot and cold water guidance describes keeping cold water below 20°C where practicable, storing hot water at around 60°C, and distributing it so it reaches roughly 50°C at the outlet within a minute - 55°C in healthcare premises [1]. L8 and HSG274 sit behind that as the core references for how the control scheme is built and verified [2][3]. A corner shop and a teaching hospital both work to those numbers.
What differs is the distance the water travels and the number of places it can stall on the way. That is where small and large buildings stop resembling each other.
What changes as the building scales
In a small building - a single shop, a small office, a two-storey HMO - the hot side is often a combi or one cylinder, the cold usually comes straight off the mains, and the longest pipe run is measured in metres. Heat does not have far to go, so an outlet that runs warm normally runs warm because of the source, not the journey. The characteristic risk here is not complexity. It is a single forgotten outlet: the cleaner’s sink nobody uses, the shower in a room let twice a year, the tap on an outbuilding at the end of a long dead leg.
A large building is a different animal. Several calorifiers feed circulating hot water loops with return legs that have to come back warm enough to count; cold water sits in storage and travels through long risers, roof spaces and ceiling voids where it can drift up towards room temperature. Heat bleeds out across every metre of poorly lagged pipe, which is why insulation quietly does so much of the temperature-control work on a large estate. Balancing the loops so the far leg still returns at temperature is a real engineering job, not a thermostat setting.
Scale also changes what a reading means. In a small building, the two representative “sentinel” outlets - nearest and furthest on the system - can genuinely speak for the whole thing [3]. In a large building, one loop’s sentinels say nothing about the loop two risers over. You need sentinels on every circuit plus a rotating sample of other outlets across the year, or your evidence has holes in it shaped exactly like the parts you never visit.
Energy measures pull on the same thread. Dropping storage or circulation temperatures to save fuel changes the safety margin, and the bigger the system the more outlets that one decision touches - which is worth weighing carefully against the Legionella risk that lower temperatures invite.
Large vs small, side by side
| What changes with size | Small building | Large building |
|---|---|---|
| Typical system | Combi or single cylinder; mains-fed cold; short runs | Multiple calorifiers; circulating hot loops with returns; stored cold; long risers |
| Where temperature is lost | Rarely in transit - runs are short, so the source sets the outlet | Across long, sometimes unbalanced distribution; cold warming in voids and roof spaces |
| What “representative” means | Two sentinel outlets can stand in for the system | Sentinels on every loop plus a rotating subset; one round covers little |
| The failure you should expect | Complacency - “too small to be a problem” | A blind spot - a wing, riser or TMV group nobody checks |
| What good evidence looks like | A short monthly record, completed every time | Auditable coverage proving every loop and a planned outlet sample over the year |
Where building size stops being the answer
Footprint is a clue, not a verdict. None of the above sets your actual numbers, your outlet list, or your monitoring frequency - those come from a competent risk assessment of the specific system, not from floor count. A “small” site with a spa pool, a rarely-used annexe, or a cold supply stored next to warm plant can carry more risk than a large, busy, well-circulated block where the water never stops moving. Size tells you where to look first. It does not tell you what you will find.
Matching the monitoring to the building
If you look after a small building, resist the pull to copy a hospital’s schedule - and resist the opposite pull to do nothing because the place is small. Fix your two sentinel outlets, set a realistic monthly check, and write down the one or two low-use outlets that need flushing between uses. That is usually proportionate, and a clean, consistent record beats an elaborate one nobody keeps up.
If you look after a large building, start with coverage rather than frequency. Map every hot loop and its return, name the sentinel outlet on each, and build a rota that pulls a different slice of ordinary outlets and TMVs into the sample each month, so that across the year the whole estate is accounted for. Then check that the readings actually trigger something: a far leg returning cool should generate an action with a name against it, not just a figure in a column.
Either way, the next move is the same. Pull out your current temperature records and put a blunt question to them - if a wing or a circuit had been running tepid for three months, would these records have caught it? If you cannot answer yes, you have found the gap to close this week, whatever the size of the building.
FAQ
Are the hot and cold water temperature targets different for large and small buildings?
No. The benchmark figures HSE describes - cold below 20°C, hot stored near 60°C and delivered at around 50°C, with 55°C in healthcare - apply regardless of size [1]. What scales is how many points you must check to show you are meeting them, and how easily heat is lost on the way to the outlet.
How many outlets should I monitor in a large multi-loop building?
There is no flat number. Each circulating loop has its own representative sentinel outlets, and HSG274 expects monitoring to be set by the risk assessment for that system, typically with a rotating sample of further outlets across the year [3]. The test is whether your records, taken together, leave no circuit unrepresented.
Is a small building ever genuinely low-risk enough to skip temperature checks?
Rarely. Even a small system needs its conditions kept right and proven, and a single low-use shower or a stored cold tank can change the picture. The duty to assess and control applies whatever the size; the risk assessment decides how light-touch the monitoring can reasonably be [2].
Sources
[1] HSE, “Hot and cold water systems”. https://www.hse.gov.uk/legionnaires/hot-and-cold.htm [2] HSE, “Legionnaires’ disease. The control of legionella bacteria in water systems - Approved Code of Practice and guidance (L8)”. https://www.hse.gov.uk/pubns/books/l8.htm [3] HSE, “Legionnaires’ disease: Technical guidance (HSG274)”. https://www.hse.gov.uk/pubns/books/hsg274.htm