Water meter leak test
A water meter that keeps moving after every outlet in the property is closed is measuring water leaving the system, and two readings taken at least two hours apart are enough to put a number on it. On a typical UK domestic meter 0.5 litres per hour is about the smallest movement worth acting on, and 60 litres per hour is a major loss.
Last reviewed 14 August 2026
What this calculator tells you
It converts two meter readings and the time between them into a rate of loss, and then into the volume and the cost that rate produces if it continues.
- Rate of loss, in litres per hour, which is the figure that classifies the result
- Daily volume, in litres per day
- Annual volume, in cubic metres per year, which is the unit UK water is billed in
- Annual cost, in pounds, at your own supplier rate
- A severity band, shown as a position on a scale rather than a single verdict
- What the reading does and does not establish, and what to measure next
The result is not gated. There is no email step, no sign up and no booking form between you and the number.
How to take the measurement
The test is two readings with nothing drawing water in between, and the accuracy of the result depends almost entirely on how completely the property was isolated.
- Close every outlet. Turn off every tap inside and outside the property, and stop anything that refills on its own: a washing machine or dishwasher mid-cycle, an ice maker, a water softener part way through regenerating, a heating system on an automatic filling device, and any toilet still refilling after a flush.
- Read the meter and note the time. Write down every digit, including the ones on a red background. On a UK domestic meter the black digits are whole cubic metres and the red digits are the fractions of one, which is where a small loss shows first. Write the time down with it.
- Leave the property alone. Use no water at all for at least two hours. Overnight gives the steadiest result, because nothing else in the property is drawing water and the reading is not competing with ordinary use.
- Read the meter again and note the time. Take the second reading the same way, all digits, with the time. The two readings and the interval between them are the whole input.
If the reading changed and you are certain nothing was running, that movement is water leaving the system somewhere between the meter and the last closed tap.
If you are not sure which digits to write down, which digits on your meter are cubic metres covers the dial types fitted in the UK.
Using the calculator
Rate of loss
What your result means
The rate of loss in litres per hour is what classifies the result, and the four bands below describe what a rate of that size is consistent with rather than delivering a verdict.
| Rate of loss | Band | What a rate this size is consistent with |
|---|---|---|
| Under 0.5 L/hr | Below the meter specified measuring range | A small real leak, ordinary residual draw somewhere in the property, or the meter working at the edge of its range. This test cannot separate them. |
| 0.5 to 5 L/hr | Small leak | A passing toilet valve, a dripping tap, or a small joint leak. Continuous, and large enough to show on a bill over a year. |
| 5 to 60 L/hr | Significant | A loss worth tracing. At this rate the annual volume is a substantial share of a normal household total. |
| Over 60 L/hr | Major | A large continuous loss, and at this rate water is going somewhere it can cause damage. |
The first band is the one most often described wrongly, including in an earlier version of this page, so it is worth being exact about it.
A domestic meter is specified for accuracy only above its minimum flow rate, Q1. On the Itron Aquadis+ DN15 R160, a volumetric meter of the type widely fitted on UK domestic supplies, that is 15.6 litres per hour. A reading of 0.5 litres per hour is roughly thirty times below it. OIML R 49-1:2013 sets a maximum permissible error of 5 per cent between Q1 and Q2 and 2 per cent above Q2, doubled once the meter is in service, and it defines nothing at all below Q1. So a movement that small is not "within tolerance". It is outside the range where any tolerance applies.
Two different floors are worth keeping apart there, because they answer different questions. The 31 times is how far below the STANDARD's lowest specified flow the reading sits. But the manufacturer publishes more than the standard requires: the same sheet gives an error of ±5 per cent at 3 litres per hour, so between there and Q1 a figure does exist, it is simply not a regulated one. Below 3 neither document says anything, and 0.5 is six times below that. So the reading is real, and its size is unguaranteed by the standard from 15.6 downwards and by anybody at all from 3 downwards.
That cuts both ways, and the second half of it is the more useful one. The same standard does not define a starting flow rate at all, because the flow at which a meter first turns is a manufacturer specification rather than a regulated figure.
For the same DN15 meter the published starting flow is 0.4 litres per hour, and that figure is footnoted to one register type, with others starting higher. So a loss smaller than that can run continuously without moving the meter at all, and the range it can hide in is wider than the number alone suggests. why a still meter does not rule a leak out sets out what an unmoving meter does and does not establish.
If the movement appeared only after you had closed everything you could find, a meter that moves with everything off works through what is usually still drawing. If you want a threshold to judge an overnight reading against, what a normal overnight reading looks like has it.
How the calculation works
The difference between the two readings is a volume, the difference between the two times is a duration, and the rate is the first divided by the second. Everything after that is unit conversion.
- Volume in cubic metres, then × 1,000 for litres
- Rate in litres per hour, the volume in litres divided by the elapsed hours
- Daily volume, the rate × 24
- Annual volume in cubic metres, the daily volume × 365 then ÷ 1,000
- Annual cost, the annual volume × the volumetric rate that applies to it
Three things are deliberately left out of the cost, because including them would overstate it.
- The standing charge, because it is fixed and a leak does not change it.
- The sewerage rate, unless the leaking water actually reaches the sewer. It is a toggle, not a silent addition, and it defaults off for an underground supply pipe and on for an internal leak.
- Any allowance for the fact that wastewater is billed on a stated share of metered volume rather than all of it. That share is set per company, and the calculator uses your own supplier figure rather than assuming the whole volume.
The rates themselves come from your supplier, and what your supplier charges per cubic metre holds them with the date each was published. If you want the arithmetic on its own, converting litres per hour into a yearly cost works a rate through to pounds step by step.
Assumptions and sources
International Organization of Legal Metrology · primary-standard source
- MPE, accuracy class 2, lower zone Q1 to Q2: 5 %
- MPE, accuracy class 2, upper zone Q2 to Q4: 2 %
- MPE, accuracy class 1, lower zone: 3 %
- MPE, accuracy class 1, upper zone: 1 %
- In-service MPE multiplier: 2 ratio
- Q2 / Q1: 1.6 ratio
- Q4 / Q3: 1.25 ratio
Aquadis+ DN15 and DN20 specification sheet
Itron · manufacturer source
- Q1 minimum flow rate, DN15, Q3 2.5, R160: 15.6 l/h
- Q2 transitional flow rate, DN15, Q3 2.5, R160: 25 l/h
- Q3 permanent flow rate, DN15: 2500 l/h
- Typical starting flow rate, DN15: 0.4 l/h
- Typical starting flow rate, DN20: 2 l/h
- Q1 minimum flow rate, DN20, Q3 4.0, R160: 25 l/h
- Flow rate at which accuracy is within +/- 5%, DN15: 3 l/h
- Flow rate at which accuracy is within +/- 2%, DN15: 5 l/h
- Minimum scale interval, DN15 and DN20: 0.02 l
Litres in a cubic metre
Exact by definition
SI definition. The litre is defined as one cubic decimetre, so one cubic metre is exactly 1,000 litres.
Hours in a day
Exact by definition
Definition of the day as 24 hours.
Days used for an annual figure
Editorial convention
Annual figures on this site are a daily rate multiplied by 365. Leap years are not modelled, so a leap year is understated by one day, which is 0.27 per cent.
Test length below which a reading error matters
Editorial convention
A meter test is a difference between two readings, so a misread of one small division is a fixed error spread over the length of the test. The smallest division on a DN15 register is 0.02 litres, so a one division misread over two hours is 0.01 litres an hour, which is about 2 per cent of the 0.5 litres an hour band boundary. Over a twenty minute test the same misread is 0.06 litres an hour, which is more than a tenth of it. The two hours is still a chosen round number rather than a derived one, and it changes nothing in the arithmetic: it only decides when the tool adds a note.
Where the first severity band ends
Editorial convention
A practical threshold set just above the 0.4 litres per hour starting flow quoted for a DN15 meter with the TVM register. It is not a meter accuracy tolerance. Against the STANDARD it sits about 31 times below Q1, the 15.6 litres per hour at which OIML R 49-1 first defines any permitted error. Against THIS MANUFACTURER it sits about 6 times below 3 litres per hour, the lowest flow at which Itron volunteers a typical accuracy that the standard does not require. Neither figure reaches down to 0.5, so nothing quantifies a reading here on either footing. Meters with other registers start higher, so this threshold sits above the most sensitive case rather than above every meter.
Two of the figures above deserve a note beyond the table, because their status is easy to misread.
The severity thresholds of 0.5, 5 and 60 litres per hour are practical bands, not measured boundaries. The lowest of them sits just above the best published starting flow of a DN15 domestic meter, which is what makes it a sensible floor, and meters that start higher put it further inside the invisible range rather than outside it. It is not a meter accuracy tolerance, and the pages that describe it as one have the right number for the wrong reason.
The meter figures are for one widely fitted model at one connection size. Your own meter may differ, and its minimum flow rate and starting flow are printed on it or in its datasheet. Where those figures matter to your reading, use the ones for your meter rather than the ones quoted here.
What to do next
What the reading justifies doing next depends on which band it fell in.
- Repeat the test overnight. A longer interval reduces the effect of reading error and confirms whether the rate is steady or intermittent.
- Split the property from the supply pipe. Close the internal stop tap, then repeat the test. If the meter still moves, the loss is between the meter and the stop tap, which is the underground supply pipe. If it stops, the loss is inside.
- Check the toilets with dye before anything else internal. A passing flush valve is silent, it is the most common internal loss, and it is the one you can confirm yourself in half an hour.
- Take the reading to your supplier if the rate is significant. Leak allowances exist and they are applied to the volumetric charge, so the measured rate and the dates you took it are the evidence for a claim.
If you are weighing a repair against leaving it, what this rate of loss costs over twelve months takes the rate you have just measured and carries it forward. If the rate is above the significant band, how a loss of this size is traced describes the methods used to locate it, and what each one can and cannot find.
A leak allowance can be applied to the volumetric charge, and whether a leak allowance can reduce that charge covers who grants one and what it covers.
Common questions
- Does this test work on a shared supply?
- No. If the meter serves more than one property it cannot be isolated, because another household can draw water at any point during the test, and any movement you record may be theirs. On a shared supply the meter measures the group, not your property.
- How long should the test run?
- At least two hours, and overnight is better. A short test divides a small measured volume by a small number of hours, so a single digit of reading error becomes a large error in the rate. Over 8 hours the same reading error is spread across four times the interval and matters four times less.
- The meter is not moving. Does that prove there is no leak?
- No, and this is the most common wrong conclusion drawn from a meter test. A meter has a starting flow rate, the flow below which it does not turn at all, and the best published figure for a DN15 domestic meter is 0.4 litres per hour, with some starting higher. A loss smaller than that can be real and continuous and still leave the meter completely still. A still meter rules out anything large. It does not rule out a leak.
- Should the sewerage charge be included?
- Only if the leaking water reaches the sewer. Water lost from an underground supply pipe soaks into the ground and never enters the sewerage system, so the sewerage rate should not be applied to it. Water from an internal leak usually does reach the sewer. The toggle defaults accordingly, and what your supplier charges per cubic metre sets both rates.
- Does the standing charge go up because of a leak?
- No. A standing charge is a fixed annual amount and it does not move with the volume you use, so a leak does not change it. Only the volumetric charge, the amount billed per cubic metre, responds to a leak, and that is the only part this calculator applies.
- What if the property is in Scotland or Northern Ireland?
- The volume and the severity are calculated exactly the same way, but the cost is not. Unmetered household charges in Scotland are set by Council Tax band and collected by the council, so the bill does not move when water is lost. Domestic water in Northern Ireland is not charged to households at all. In both cases the calculator reports the loss and states why there is no cost figure rather than showing zero.