---
title: "Dead legs and blind ends: how to find, assess and remove them"
source_url: https://legionella.io/articles/dead-legs-and-blind-ends-how-to-find-assess-and-remove-them/
canonical_url: https://legionella.io/articles/dead-legs-and-blind-ends-how-to-find-assess-and-remove-them/
pillar: "Legionella Risk Assessment"
summary: "A dead leg is stagnant pipework feeding a tap that barely runs. How to spot blind ends on a survey, measure them in pipe diameters, and decide what to cut out."
primary_keyword: "dead leg Legionella"
date_published: 2026-02-09
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."
---

# Dead legs and blind ends: how to find, assess and remove them

A dead leg is a length of pipe that water rarely moves through, feeding an outlet that barely gets used. A blind end is worse: pipework that goes nowhere at all, capped off and forgotten, with water sitting in it permanently. Both are stagnation features, and both are physical things you can put a tape measure against.

That is the useful way to think about them. Not as an abstract risk, but as bits of metal and plastic on a wall that you can find, measure and cut out. The bacteria do not care what the pipe was for. They care that the water is still and warm.

## The core principle: still water at the wrong temperature

Legionella multiplies in water that sits between roughly 20°C and 45°C, and stagnation lets biofilm establish where flow would otherwise scour it away [1]. A dead leg gives the bacteria both conditions at once. Cold water in a rarely-used spur drifts up past 20°C; hot water in a stranded length cools below the temperature that would control growth. Either way you have a small reservoir feeding the rest of the system.

The principle that follows is simple. Every metre of pipe should earn its place by carrying water that moves. Where it does not, you either restore the flow or you remove the pipe. HSE's technical guidance treats the elimination of dead legs and redundant pipework as a core control, not an optional tidy-up [1].

How short is short enough? HSG274 works in pipe diameters rather than a fixed length, and a long-standing rule of thumb is that a leg off a circulating or regularly-used main should not exceed about five times the internal diameter of the branch [1]. A 22mm branch therefore wants to be under roughly 110mm to its terminal fitting. Treat that as a design aim to test against, confirmed by your own risk assessment, not a pass/fail line you can apply blind.

## Reading the system: a survey walk-through

Picture standing in front of a plant room wall. The diagram you are building in your head has three things on it.

First, the live path: the incoming cold main, the storage cistern or the calorifier, and the distribution running out to the sentinel and other outlets where water genuinely flows every day. Draw that as the spine.

Second, the branches: every tee off that spine. At each tee, follow the smaller pipe with your eye and your hand. Where does it terminate? If it ends at a tap someone uses daily, it is live. If it ends at a capped stub, a disconnected appliance, or a tap nobody has turned in months, mark it. The give-aways are a pipe that is cold along its whole length when the rest of the run is warm, a fitting furred with old jointing compound, or a valve seized open to nothing.

Third, the orphans: pipework that loops away and comes back, or that was once part of a ring and is now a cul-de-sac because a section was isolated. These are the hardest to spot because from the front they look like part of the live system. Trace the run end to end before you trust it.

The honest way to record this is feature by feature: location, what it serves, internal diameter, length to the terminal point, and a temperature reading taken after the live system has settled. That last reading is the tell. A branch that reads close to ambient when the cold should be below 20°C, or tepid when the hot main is at distribution temperature, is a dead leg whatever the drawings claim [2].

## Four situations you will actually meet

The decommissioned shower. A wet room was stripped out, the shower removed, but the supply was capped at the wall rather than cut back to the main. You now have a blind end holding a few hundred millilitres of permanently still water, often warm because it runs alongside the hot. The fix is not a cap closer to the tee. It is removing the branch back to the live pipe. As covered in [Neglected water systems: the danger of stagnation](https://legionella.io/articles/neglected-water-systems-the-danger-of-stagnation/), a capped stub is the classic neglected feature that survives three refurbishments because nobody owns it.

The oversized tee. A low-use cleaner's tap is fed by a 22mm branch off the main when 15mm would have carried it. The leg is short but fat, so the stagnant volume is larger than the length suggests. Volume, not just length, is what matters. Cutting the branch back and dropping the diameter both help.

The stranded riser. A vertical run that used to serve an upper floor now feeds nothing because that floor was isolated. From the basement it looks live. Trace it up and you find a closed valve and a dead column of water. This wants isolating at source and draining, then physical removal when access allows.

The water heater spur. A point-of-use instantaneous heater was swapped for a different unit and the old supply spur left in place. Short, but feeding nothing and rarely flushed. These accumulate one refit at a time. Good system design avoids them in the first place, which is the theme of [Avoiding stagnation: design tips for consistent water temperatures](https://legionella.io/articles/avoiding-stagnation-design-tips-for-consistent-water-temperatures/).

## The trade-offs in remediation

Removal is the right answer and not always the available one. Cutting back to the main needs the system drained or frozen, an isolation that may take a wing offline, and a competent plumber rather than a flush of the rota. Where you cannot remove immediately, the interim is to manage the feature: flush it on a defined frequency so the still volume is regularly displaced, and put it on the register as outstanding remedial work, not as solved [2].

In my view, flushing a dead leg is a holding measure, never a destination. Every flushed stub is labour you pay for weekly for the life of the building, plus a residual risk between flushes. The pragmatic call is to schedule physical removal of the worst features at the next time the relevant supply is down for other work, and to stop treating the flush list as if it were a fix.

One caveat worth stating plainly. This is general guidance to help you recognise and quantify stagnation features; the actual control measures, frequencies and what counts as acceptable on your site are decisions for a competent person working from a current, site-specific Legionella risk assessment under the ACoP L8 framework [3]. A diameter rule of thumb does not replace that judgement.

## What to do today

Take your asset register or schematic and pick one wet area you refurbished in the last five years. Walk it, trace every branch to its end, and write down any capped stub or cold-when-it-should-be-warm spur you find. That single honest list of features, with location, diameter and length, is the thing a flushing rota never gives you. If your records still live on paper or a spreadsheet that nobody updates after a refit, this is exactly the kind of feature-level data a digital logbook keeps visible, so a dead leg created during one project does not quietly survive the next three. Related background sits in [Identifying Legionella hazards in your water system](https://legionella.io/articles/identifying-legionella-hazards-in-your-water-system/).

## FAQ

### Is a dead leg the same as a blind end?

Not quite. A dead leg is pipework that still feeds a used outlet but rarely sees flow, so water stagnates between uses. A blind end is capped pipework that serves nothing, so the water never moves at all. A blind end is generally the higher concern because there is no flushing through it.

### How long can a dead leg be before it matters?

HSG274 frames the limit in pipe diameters rather than a fixed length, with a common rule of thumb that a leg should not exceed about five times the branch's internal diameter [1]. The shorter and narrower, the smaller the stagnant volume. Your risk assessment confirms what is acceptable for the specific outlet and use pattern.

### Can I just flush a dead leg instead of removing it?

Flushing displaces the still water and is a legitimate interim control, but it is labour you repeat forever and it leaves risk between flushes. Physical removal back to the live main is the durable answer, so flushing is best treated as a holding measure until removal can be scheduled [2].

## Related reading

- [Neglected water systems: the danger of stagnation](https://legionella.io/articles/neglected-water-systems-the-danger-of-stagnation/)
- [Avoiding stagnation: design tips for consistent water temperatures](https://legionella.io/articles/avoiding-stagnation-design-tips-for-consistent-water-temperatures/)
- [Identifying Legionella hazards in your water system](https://legionella.io/articles/identifying-legionella-hazards-in-your-water-system/)

## Sources

[1] HSE, "Legionnaires' disease: Technical guidance (HSG274)". https://www.hse.gov.uk/pubns/books/hsg274.htm
[2] HSE, "Hot and cold water systems". https://www.hse.gov.uk/legionnaires/hot-and-cold.htm
[3] HSE, "Legionnaires' disease. The control of legionella bacteria in water systems - ACoP and guidance (L8)". https://www.hse.gov.uk/pubns/books/l8.htm