Bringing a building's water back after it's been empty

A building that stands empty does not put its water system on pause. The water sits still and drifts toward room temperature, and stillness plus warmth is the condition legionella wants. Whether an empty building becomes a problem when people come back is mostly decided in the two weeks before they do, not on the morning the doors reopen.

So the honest answer to "what do we do to reopen safely" is: very little on the day, and a fair amount in the fortnight before. Review the assessment, flush the system through, prove the temperatures came back into control, deal with the tanks that stood idle, and disinfect or sample only where the assessment tells you to. Roughly in that order.

The physics, in one paragraph

Legionella multiplies between 20°C and 45°C. A running system keeps water out of that band or moves it through quickly: cold stays cold because it is drawn and replaced, hot stays hot because it circulates. Stop using the building and both defences fail. Cold water sitting in pipework climbs toward room temperature, hot water sitting in a cylinder falls, and everything converges on the warm, still middle where the bacteria breed. Summer makes it worse, because the incoming mains and the ambient building are both warmer. A fortnight empty in August is not the same risk as a fortnight in January, and the assessment should say so.

Start with the risk assessment, not the flush

The instinct is to send someone to run taps. Do the paperwork first. A change of use or occupancy is itself a trigger to review the legionella risk assessment. The fixed two-year review was removed from ACOP L8 back in 2013, and reviews are meant to be driven by change rather than the calendar. A building going from empty to occupied, or changing what it is used for, is exactly the kind of change that should prompt a look at whether the assessment still describes reality.

We wrote about the wider myth that assessments run on a fixed clock in the legionella risk assessment review myth. For a reopening the practical point is narrow: the assessment tells you which outlets matter and which tanks feed what, and where a sample is worth taking at all. Flushing before you have read it is just exercise.

Flush it through before people arrive

Flushing a stood-down system means running every outlet, methodically, so stale water is drawn out and replaced across the whole network rather than at a convenient few taps. Work outward from the incoming supply toward the furthest outlets, so you are not pulling old water back through pipes you have already cleared. Every outlet counts, including the ones nobody thinks about, like the cleaner's sink or the outside tap that has not run since spring.

An outlet that has gone seven days or more without use counts as little-used, and after a void every outlet is little-used by definition. HSG274 already expects weekly flushing of little-used outlets in normal running, one line in the full routine cadence. A reopening is that same task done once, hard, across the entire building.

Prove the temperatures, don't assume them

Flushing moves water. On its own it does not prove the system has returned to safe control. Once the building is running again, check the sentinel temperatures and confirm the control values actually hold:

• Stored hot water: 60°C or above.
• Distributed hot water: 50°C or above within a minute at the outlet.
• Cold water: 20°C or below within two minutes of running the tap.

A cold outlet that will not drop to 20°C after a fortnight of warm stagnation is telling you something: either the flush did not finish the job, or there is a warming problem the empty spell merely exposed. Either way, that reading is the start of an investigation, not a box you tick and move past.

The tanks that stood idle

Cold water storage tanks are where a void shows worst. A tank that sat warm and undisturbed for weeks can develop biofilm and sediment, and a quick temperature at the nearest tap will not tell you what is happening under the lid. Open it and look before you sign the building off. Check the lid is seated and the insect screen intact, and look for debris or scale that was not there at the last inspection. If the assessment or the tank's condition calls for a clean or a disinfection, do it before occupancy, not after the first complaint.

Disinfect and sample only where the assessment says to

Two reflexes to resist. The first is to disinfect everything as a ritual. Disinfection has a place after a stand-down, but where and how much is a question for the risk assessment and the system's condition, not a fixed routine, so do it where the assessment calls for it and to the extent it calls for it. There are no dose numbers in this post on purpose. The right dose is a decision for the assessor and the system in front of them, not a blog.

The second reflex is to sample the water for legionella as proof of safety. Sampling has a role, but HSE is clear that it is appropriate in specific circumstances rather than as a routine tick. A negative sample on the day tells you about that tap, at that minute, and it can lull you into skipping the flushing and temperature discipline that actually keeps the system safe. Sample where the assessment says sampling earns its place, and rely on control the rest of the time.

When there is no reopening date: flush the void

Not every empty building is heading for a reopening you can plan around. Some sit half-occupied, some empty between tenants for months. For those, the alternative to a big reopening exercise is to stop the water stagnating in the first place, with regular flushing through the void period, typically weekly, across the whole system.

The catch is that "someone flushes weekly" is easy to write into a plan and easy to let slip. A flush that happened is only worth anything if you can show it happened, on the right date, at the right outlets. We went into why that evidence is the hard part in flushing little-used outlets and proving it. An undocumented flushing regime and no flushing regime look identical to an assessor, and to a court.

Schools do this every holiday

Schools live this problem on a timetable. Legionella risk in schools rises over the holidays precisely because the building empties and the water stops moving, and the summer break stacks the longest void onto the warmest months. Every half term is a miniature version, and the return-to-term week is the recurring edition of everything above: flush the building through before the children come back, then prove the temperatures held.

The advantage schools have is that the dates are known a year out. The holiday is on the calendar, so the flushing during it and the checks after it can be on the calendar too. A void you can see coming is a void you can plan the evidence for.

Where this leaves the software

Whether it is a void you are flushing weekly or a reopening you are working through, the same two things decide whether it holds up later: that the rounds actually happen, and that each one leaves a record at the outlet level. That is what the BlueWave scheduler is for. The weekly void flush and the reopening checks get booked as recurring jobs against the real outlets and tanks on the site, so a missed week shows as overdue work on a screen instead of a gap someone finds in an audit. The engineer's readings and flush confirmations land against each asset with a timestamp, which is exactly the per-outlet evidence a reopening needs and a void demands.

The test for a reopening is the one you would want applied if you were the incoming occupier. Two years from now, could you show which outlets were flushed, on which date, and that the temperatures were back in control before anyone turned the taps on. If the honest answer is "we did it, but I couldn't prove it", you did half the job.