---
title: "Automated flushing systems to prevent Legionella"
source_url: https://legionella.io/articles/automated-flushing-systems-to-prevent-legionella/
canonical_url: https://legionella.io/articles/automated-flushing-systems-to-prevent-legionella/
pillar: "Technology & Remote Monitoring"
summary: "Automated flushing stops low-use outlets stagnating - but only when fitted right. When valves earn their place, and when to remove the dead leg instead."
primary_keyword: "automated flushing systems"
date_published: 2025-06-01
date_reviewed: 2026-06-26
author: "Legionella.io editorial team (REMOTE TECH LTD)"
reviewed_against: "HSE L8 and HSG274 guidance"
region: "United Kingdom"
license: "(c) REMOTE TECH LTD. Quote freely with attribution and a link to source_url."
---

# Automated flushing systems to prevent Legionella

An automated flushing system runs water through an outlet on a set schedule so it never stands still long enough for Legionella to establish. On paper that fixes the most common weakness in any flushing regime: the flush nobody got round to. In practice it only helps if you fit it to the right outlets, set it to do real work, and keep checking that it still runs. Get those wrong and you have bought an expensive way to keep a dead leg alive.

## What does an automated flushing system actually do?

At its simplest it is a solenoid valve and a small controller fitted at or near a low-use outlet, or on the end of a run. The controller opens the valve on a timer, or after the outlet has sat idle for a set period, runs water for a set time or until a target temperature is reached, then closes again. Better units measure the temperature at the outlet and only stop once the cold water is genuinely cold or the hot water genuinely hot; cheaper ones simply run the clock. The job is turnover - swapping the standing water in that branch for fresh water drawn from the moving part of the system. Flushing infrequently used outlets is a long-standing control for stagnation in HSE guidance; automation is just a way of doing it that doesn't depend on someone remembering [1][2].

## Does automated flushing replace our manual flushing duties?

No. It automates the task, not the duty. The risk assessment, the written control scheme and the obligation to show the controls actually work all stay with the duty holder, and none of that transfers to a box on the wall [1][3]. What automation removes is the labour and the "we forgot" failure. What it adds is a new asset that can fail quietly. You are really trading one failure mode - the missed manual flush - for another, a valve that has stopped working while the schedule keeps reporting that it hasn't. The duty to verify doesn't disappear; it moves from "did someone flush?" to "is the kit still doing what we think?"

## Will the HSE accept it as compliant?

HSE guidance doesn't mandate or rule out any particular method. It expects you to control the conditions that let Legionella grow - stagnation and the wrong temperatures - and to keep records that show you did [1][2]. Automated flushing can sit inside your written scheme as the named control for specified outlets. What an inspector cares about isn't the brand of valve; it's whether the assessment specifies it, whether it's set to achieve turnover and the right temperature, and whether the records show it ran and was reviewed. A cupboard of humming valves that nobody has checked the logs on is not control - it's decoration.

## When is automated flushing actually worth installing?

The honest test has three parts: the outlet has to exist, it can't be used often enough to stay fresh on its own, and manual flushing is unreliable or impractical to keep up. Plenty of real settings qualify - a school wing standing empty over the summer, hotel rooms void for weeks at a time, a sports pavilion used twice a week, safety showers and eyewash stations, intermittently occupied care or custodial accommodation, ensuite rooms in a building that is only part-let. But before you automate anything, ask the harder question: does this outlet need to be here at all? Where it is genuinely redundant, removing the dead leg beats flushing it forever - see [on the danger of stagnation](https://legionella.io/articles/neglected-water-systems-the-danger-of-stagnation/). Pulling a redundant branch once ends the risk and the water bill on it; automating it just pays to keep both alive.

## Is any flush good enough, or does it have to reach temperature?

A flush only earns its place if it exchanges the standing water and, where temperature is the control, brings the outlet back to a safe reading - cold run cold, hot run hot. A ten-second trickle that never shifts the tepid slug sitting in a long branch is theatre, not control. Set the duration and frequency through the risk assessment for that specific run, because pipe length, internal volume and how the branch is fed all change the answer; treat any rule of thumb as a prompt to check, not a setting to copy. Where you can, choose units that flush until temperature is achieved and log that temperature, rather than ones that run for a fixed time and hope. If you are building this on top of an existing manual regime, [on implementing a flushing programme](https://legionella.io/articles/how-to-implement-a-flushing-programme-for-legionella-control/) covers the groundwork the automation has to match.

## What tends to go wrong with these systems?

The failure modes are predictable, which is the good news. A solenoid can stick open - a slow flood and a quietly rising water bill - or stick shut, which silently removes the control while the log still reports "scheduled". A controller can lose power or drop its programme. Flushing can keep running for years on an outlet that should have been removed. And the subtle one: false confidence, where a wall of green ticks on a dashboard stops anyone actually walking the building. The defence against all of these is the same - treat the flushing system as a monitored asset with alerts on failure and on out-of-range results, and have a competent person review the exceptions rather than admire the totals.

## How does it prove control to an auditor?

The whole point of automation is the audit trail, but a trail is only worth keeping if it captures the right things: which outlet, when it ran, for how long, the temperature reached if measured, whether that was in range, and what happened when it wasn't. A log that says "flush completed" and nothing else tells you the valve opened - not that the water came good. Wire the alerts into your normal escalation route so a missed or failed flush becomes an action with an owner, exactly like any other out-of-range reading; [on poor temperature control](https://legionella.io/articles/poor-temperature-control-a-recipe-for-legionella/) covers how those escalations should work. That is the difference between data and evidence.

## Doesn't constant flushing waste a lot of water?

Yes - and that is a real cost to weigh, not a reason to dismiss the approach. Every automated flush sends treated water to the drain, so both the running cost and the environmental cost rise with the number of outlets you automate and how hard you flush them. It is another argument for working in order: design out the dead leg first, cut the flush volume to what genuinely achieves turnover second, automate only what is left third. A site that automates every outlet indiscriminately is usually one that never did the survey to find what it could have removed.

## Where do automated valves leave the monthly walk-round?

They shrink it, they don't end it. Once flushing is automated and logged, your people can spend the saved time on the things a sensor can't see - the cold tank quietly warming in a hot riser cupboard, the showerhead furred with scale, the "temporary" hose left coupled to a tap since spring. Automation is at its best when it takes the dull, easily-missed jobs off the rota so the human checks land where judgement is actually needed.

## Start with a survey, not a shopping list

Before you price up a single valve, walk the building with the risk assessment and put every outlet you currently flush by hand into two columns: must keep and could remove. Automate the must-keeps that are genuinely hard to reach or easy to forget, and raise jobs to take out the rest. That one sort protects people better than the most capable controller on the market, and it shrinks the problem - and the bill - before you spend anything.

One caveat worth stating plainly: the durations, frequencies and target temperatures mentioned here are illustrative, not settings to lift. A competent person sets them for your specific pipework, water source and users, and the agreed figures belong in your written scheme, not in a generic schedule copied off a web page.

## Related reading
- [Smart thermometers - using IoT for Legionella control](https://legionella.io/articles/smart-thermometers-using-iot-for-legionella-control/)
- [IoT sensors for water system monitoring: an overview](https://legionella.io/articles/iot-sensors-for-water-system-monitoring-an-overview/)
- [Forming a Water Safety Group: roles and benefits](https://legionella.io/articles/forming-a-water-safety-group-roles-and-benefits/)

## Sources
[1] 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
[2] HSE, "Legionnaires' disease: Technical guidance (HSG274)". https://www.hse.gov.uk/pubns/books/hsg274.htm
[3] HSE, "Legionnaires' disease - what you must do". https://www.hse.gov.uk/legionnaires/what-you-must-do/index.htm