A building water system is designed around expected demand. The type and number of outlets connected tell the designer how much water will pass through the system, and the pipework, storage and plant are sized to suit.

That expectation carries a requirement. To stay in the condition it was designed for, a system needs a minimum throughput of water. In practice that means regular turnover through the main distribution, and HSG274 Part 2 (Table 2.1) points to weekly flushing of infrequently used outlets, or as the risk assessment indicates. Where normal use does not deliver that movement, flushing has to replace it.

ACoP L8 (paragraph 59) lists avoiding water stagnation, which may encourage the growth of biofilm, among the measures for controlling the risk. When water does not move regularly, stagnation follows, and stagnant water is among the most significant factors in allowing Legionella to proliferate, particularly where temperatures sit in the 20 to 45 degrees Celsius range described in ACoP L8.

This is the second article in our MOD series on troubleshooting the causes of persistent Legionella problems. The framework we use is Mechanical failure, Occupant behaviour and Design. We suggest that troubleshooting by category, rather than treating symptoms, is the quickest route from a persistent positive result to the actual cause.

In the first article we covered Mechanical. A passing TMV, an unbalanced return loop, a dead leg at a critical point: parts of the system not performing as designed, usually fixable once correctly identified.

Here we look at a different situation. The system is sound. There is nothing wrong with the pipework or the plant. What has changed is how the building is used. Occupant behaviour has two parts, and they feed each other.

Part 1: Demand Lower Than the Design Expects

Water use in a building often falls well below what the system was designed for. The usual reasons are:

• Over-design, where the system was sized for a peak that rarely arrives.
• Change of use, where floors or wings now serve a different purpose.
• Low occupancy, from vacant areas, seasonal use or hybrid working.
  
  

None of these is a fault in the pipework. The system is sound, but it is being used differently from how it was designed to be used. Flow drops, water ages, and temperatures can drift into the range where Legionella multiplies.

ACoP L8 (paragraph 47) lists changes to the water system or its use, and to the use of the building, as reasons to review the risk assessment. Low occupancy is one of the changes that is most often missed.

There is also a mixed message at play. For years occupants have been told to turn taps off and save water. It is a commendable habit, but an outlet that is rarely or never used is bad for water quality. A tap left alone is not a saving. It is a stagnation point.

Part 2: Why Can Flushing Fail Even When It Is Signed Off?

Where demand is low, flushing is used to simulate the demand the system was designed for. The principle is simple, and that simplicity is part of the problem.

Turning on a tap is considered so easy that the task is often handed to whoever is on the ground: cleaners, care staff, maintenance teams. It is added to an already busy schedule as a bolt-on, often done reactively, and understandably given low priority against their main duties. Two things tend to follow.

Flushing is misunderstood. The aim is to run the outlet long enough to bring fresh water from the distribution pipework to the point of use. Running it for less leaves stagnant water in place. Running it for longer or more often than needed simply wastes water.

HSG274 Part 2 (Table 2.1) advises flushing until the temperature at the outlet stabilises and is comparable to the supply water. The time this takes depends on the length and layout of the pipework serving each outlet, which is why one fixed flushing duration will not suit every building. Without that explanation, some people over-flush and others under-flush.

Records stop being evidence of what happened in the water system. In most buildings, proof that flushing took place is a signature. Temperature checks work in a similar way: a reading is written down against a day and time, but it cannot show what happened between checks. After twelve months of the same routine, results become predictable, and a record can begin to reflect what is expected rather than what was measured.

We describe the condition rather than the person. A process that depends on one unverifiable signature will tend to produce this outcome, whoever is carrying it out.

What This Means for a Risk Assessment

ACoP L8 and HSG274 Part 2 both expect infrequently used outlets to be flushed and the work recorded. The records show that the task was logged as done. They cannot show that the water in the system responded.

A useful question for any building is therefore not "is there a flushing regime?" but "how do we know it is working?"

If the answer rests on signatures and spot readings, there may be a gap between the regime on paper and the behaviour of the water.

Closing the Gap

Three practical steps help.

Compare the design demand with the real demand. Vacant areas, changed use and low occupancy should be reflected in the risk assessment and the flushing regime.

Train the people who flush, and explain why. Fresh water from distribution to outlet is the aim, no more and no less.

Use data rather than signatures where it matters. Legionella Watch provides continuous monitoring of water temperatures and, where appropriately configured, outlet usage. This helps identify whether outlets are being used and how the water system responds, rather than relying solely on manual records.

What to Take From This

Water systems are sized for a demand that many buildings no longer generate. Over-design, change of use and low occupancy can all reduce water turnover without any fault in the pipework.

Flushing stands in for the demand the design expected. Whether it works depends on the people carrying it out understanding the aim, and on the task being given sufficient priority.

A signature shows a task was logged as completed. It doesn't necessarily confirm that the water responded as intended. Continuous monitoring can provide additional evidence of what's happening within the system.

Do You Know if Your Flushing Regime Is Working?

If your building's occupancy or water usage has changed, your existing flushing regime may no longer reflect how the system is actually being used.

LCS can help you understand what's happening within your water system, identify areas at risk of stagnation and assess where monitoring could improve your current approach.

Discuss your water system with LCS

Next in the MOD series: Design: how pipework, system sizing and changing water demand can contribute to persistent Legionella problems.

Keith Elwell is Managing Director of Legionella Consultancy Services Ltd. LCS provides Legionella risk assessments, written control schemes and IoT remote monitoring across commercial, healthcare and industrial premises in the UK.