What we got wrong sizing our first 200-key hotel — and the rule we use now

In 2019 we sized a 200-key hotel on daily volume. The arithmetic was clean: 200 rooms, 1.6 occupants a room, 45 litres a person at 60 °C. Twelve tonnes of hot water a day. We selected plant to make that in sixteen hours and felt good about it.

The hotel ran out of hot water on the fourth floor at 7:40am, every morning, for a week.

Daily volume tells you almost nothing

Nobody consumes hot water evenly. In a business hotel, roughly 45% of the day’s demand leaves the tank between 6:30 and 8:30am. Add the kitchen starting breakfast prep and the laundry taking its first fill, and the morning peak is three to four times the daily average rate.

A heat pump does not care much about peak. It has a fixed output and it recovers steadily. Storage is what absorbs the peak. We had sized the plant correctly and the storage badly.

Size the heat pump on daily energy. Size the tank on the worst two hours. They are different problems and they have different answers.

The rule we use now

We model the peak two-hour draw-off and hold that volume in usable storage, at delivery temperature, before the heat pump contributes anything. In practice that lands between 55% and 70% of the two-hour peak in buffer volume, because the plant does keep producing during the peak — but we never count on more than 60% of its rated output at the start of the peak, when the tank is stratified and return temperatures are still high.

For that 200-key hotel the answer was not a bigger heat pump. It was 12,000 litres of storage instead of 5,000, and a stratification-preserving return arrangement so the top of the tank stayed at 60 °C rather than mixing down to 48 °C by 7am.

Three things we now check on every hotel

First, the laundry schedule. A single 40 kg washer taking a hot fill can pull 400 litres in six minutes, and laundry managers start when it suits them, not when it suits your plant.

Second, the recirculation loop. A badly balanced loop bleeds heat continuously and shows up as a plant that never quite reaches setpoint overnight. It is almost always a balancing problem, not a capacity problem.

Third, the buffer connections. Bottom-fed cold, top-drawn hot, with the heat-pump return entering low. It sounds obvious. We have opened tanks on four sites where it was plumbed the other way round and the operator had been blaming the compressor for two years.

What it costs to get wrong

Under-sized storage does not cause a dramatic failure. It causes lukewarm showers at 7:40am, guest complaints the front desk cannot explain, and an operator who eventually switches the old diesel boiler back on “just for mornings” — which quietly erases the entire business case for the retrofit.

That is the real risk in a heat-pump project. Not that the technology underperforms, but that one sizing shortcut hands the site a reason to distrust it.