The monthly sentinel round is the single most repeated check in most water-safety schemes, and it lives or dies on technique. Run the tap too briefly, hold the probe in the wrong place, or record the wrong outlet, and the figure that lands in the logbook is worse than useless: it reads as compliance while telling you nothing.
This is the method, not the theory. If you want the background on why these particular taps were chosen and what “sentinel” means, that sits in the separate explainer on sentinel outlets. Here you get the task: what to carry, what to do at each outlet, what a pass and a fail look like, and what to do the moment a reading is out.
What you are doing and why it matters
A sentinel outlet is the tap nearest to and furthest from each hot and cold supply, the points where, if temperature control is slipping anywhere on that leg. It shows up first. Checking them monthly is the front-line test that hot water is leaving hot and cold water is staying cold, the two conditions that keep the system outside the range where Legionella multiplies [1][2].
You are not diagnosing the system. You are taking honest readings, recording them against a clear pass/fail, and raising anything out of range so a competent person can act.
Before you start: competence and kit
This is a skilled task, not a clipboard exercise. You need to know which outlets are sentinels for your building, how your thermometer is used and that it is in calibration, and the limits your scheme works to. If any of that is unclear, stop and ask, a confidently recorded wrong number is the worst outcome of the day.
Carry: the outlet schedule or schematic that lists your sentinels, a calibrated digital thermometer with a surface probe (and an inserted/immersion probe if your scheme uses one), antibacterial wipes, the logbook or monitoring app, and gloves. Treat hot outlets with respect, water leaving a sentinel hot tap can scald exposed skin, so direct the flow to the back of the basin and keep hands clear of the stream.
The round, step by step
Work the building in a logical order so nothing is skipped, and do hot and cold separately at each location. The sequence below is the implementation pattern for one outlet; repeat it at every sentinel, then close out the calorifier and the round.
1. Identify the outlet from the schedule, not from memory. Match the asset reference on the schematic to the physical tap before you touch it. Recording a reading against the wrong outlet is the error that quietly corrupts a whole trend. Done when: the outlet ID on your record matches the one on the schedule and the physical label.
2. Take the hot tap to a stabilised temperature and note the time to reach it. Open the hot outlet fully and let it run. Guidance commonly expects a sentinel hot outlet to reach roughly 50C within about a minute of running, with the actual target set by your scheme [2]. Place the surface probe on the spout or hold the flow against the probe until the reading stops climbing. Done when: the temperature has stabilised and you have recorded both the figure and roughly how long it took to get there.
3. Take the cold tap after a short run and check it is staying cold. Run the cold outlet and let it draw through. The usual expectation is that cold water stays below around 20C after running for about two minutes, again with your scheme as the authority on the figure [2]. Record the stabilised cold reading. Done when: the cold temperature has settled and is logged with the run time.
4. Use the right probe technique for your scheme. A surface (clamp or contact) probe on the spout or pipe is the common non-invasive method; some schemes specify an inserted probe into the flow. Whichever your scheme uses, be consistent month to month, switching technique makes the trend meaningless. Done when: the same measurement method has been used as last month, wiped clean between outlets.
5. Record against the pass/fail there and then. Write the reading immediately, with date, outlet, hot/cold, run time and your initials. Mark it against the scheme’s acceptance limit while you are still at the tap, not back at the desk from memory. Done when: the entry is complete and unambiguous, with any out-of-range value clearly flagged.
6. Check the calorifier flow and return as part of the round. Where your scheme includes it, record the hot water leaving the calorifier (flow) and coming back (return). Stored and circulating hot water is commonly expected around the 60C flow / 50C return mark, but defer to your scheme and the system’s design figures [3]. Done when: flow and return are logged and any drop across the circuit is visible in the record.
7. Close out and escalate exceptions before you leave. Total up which outlets passed and which did not. Any failure, hot too cool, cold too warm, or a tap that never reaches temperature, gets reported to the responsible person the same day, not held until the sheet is filed. Done when: the round is signed off and every exception has been raised through the agreed route.
Where the round goes wrong
The commonest failure is the rushed cold check: recording the first cold reading instead of running the tap long enough to see whether the standing water in the pipe was masking a warm supply. A cold tap that reads cold for the first few seconds and then climbs is the warning sign the round exists to catch, so the run time matters as much as the temperature.
The second is technique drift: a surface probe held loosely, or moved off the spout before the reading settles, will under-read a hot outlet and over-flag a fail. Let it stabilise.
The third is the silent fail. An out-of-range reading written down but not reported is a failure of the whole scheme, not just the round. The number means nothing until someone competent decides what to do with it. Note too that a single low monthly reading is a flag to investigate, not proof the system is unsafe, that judgement belongs to the risk assessment.
How to verify the round was done properly
A good round is auditable on its own. Every sentinel on the schedule has an entry, hot and cold are separate, run times are noted, the calorifier figures are present where required, and every out-of-range value carries a follow-up reference. If your logbook shows a clean run of identical “pass” temperatures with no variation at all, that is a tell worth questioning, real systems wobble, and an unvarying record often means readings copied rather than taken.
FAQ
How long do I run a cold tap before recording the temperature?
Long enough to clear the water that has been standing in the local pipework and see the temperature of the supply actually feeding the outlet, guidance commonly cites running for around two minutes for a sentinel cold check, with the figure confirmed by your scheme. The point is to catch a cold tap that starts cold and warms up as it draws through, so record the stabilised reading, not the first one.
Do I use a surface probe or an inserted probe?
Use whatever your written scheme specifies, and use it the same way every month. A contact/surface probe on the spout or pipe is the usual non-invasive choice and is fine for routine sentinel monitoring; some schemes call for an inserted probe in the flow for certain checks. Consistency matters more than the choice itself, because the value of the round is the trend over time.
What do I do if a sentinel outlet fails its temperature?
Record it clearly as out of range, with the figure and the run time, and report it to the responsible person the same day through your agreed escalation route. Do not adjust thermostatic settings, bypass a TMV, or boost the calorifier yourself, those are decisions for a competent person acting on the risk assessment. Your job is an accurate reading and a prompt flag, not a fix.
How do I know which outlets are the sentinels?
They are defined for your building in the risk assessment and listed on the monitoring schedule or schematic, typically the nearest and furthest hot and cold outlets on each leg or circuit. Do not pick them yourself or substitute a more convenient tap. If the schedule and the building no longer match because outlets have been added or removed, that is a finding to raise, because the sentinels may need reviewing.
Do this before your next round
Pull your monitoring schedule and walk it against the building once, dry, before the next due date. Confirm every listed sentinel still exists and is reachable, that your thermometer is in calibration, and that you know the exact limits your scheme works to. The round only protects anyone if the readings are taken right and the fails get raised, and that starts with knowing your outlets before you are stood at the tap with the water running.
Sources
- HSE, ACoP L8 (2013), “Review of control measures: monitoring and routine inspection”, p.18. https://www.hse.gov.uk/pubns/books/l8.htm
- HSE, HSG274 Part 2 (2024), “Operation and inspection of hot and cold water systems”, p.70. https://www.hse.gov.uk/pubns/books/hsg274.htm
- HSE, “Hot and cold water systems”. https://www.hse.gov.uk/legionnaires/hot-and-cold.htm