Swimming pool bucket test
A bucket of pool water standing on a pool step loses water to evaporation at the same rate the pool does, so the difference between the two drops over 24 hours is the leak. That is why this test needs no evaporation figure, no weather data and no assumptions: the bucket measures the evaporation for you, in your pool, on the day you tested.
Last reviewed 14 August 2026
What this calculator tells you
It converts two millimetre measurements and a surface area into a daily volume, and tells you whether there is a leak at all.
- The net drop in mm, which is the pool drop minus the bucket drop
- A daily volume in litres per day, and an annual volume in cubic metres
- An annual cost at your own supplier rate
- Or, where the net drop is zero or less, a plain statement that this is evaporation and not a leak
That last outcome is a different answer rather than a small number. A pool that is not leaking gets told so, not handed a figure near zero that looks like a tiny leak.
How to take the measurement
A bucket, a pen and 24 hours. The whole method rests on one idea: put a control in the same conditions as the thing you are measuring, and the conditions cancel out.
- Stand a bucket on a pool step. Put a bucket on the top step of the pool, or on a submerged ledge, so that it is standing in the water rather than beside it. It needs to be at the same temperature and in the same weather as the pool.
- Fill it with pool water. Fill the bucket from the pool until the water inside it is level with the water outside it. Matching the two levels is what makes the bucket behave like the pool.
- Mark both levels. Mark the water level inside the bucket and the water level outside it, on the bucket itself, with a waterproof pen or a strip of tape. Two marks on one object, so nothing has to be measured twice from a different reference.
- Leave it for 24 hours. Turn the pump on or off for the whole period and note which, then leave the pool alone. Do not swim, do not backwash, do not top up. Anything that adds or removes water invalidates the test.
- Measure both drops. After 24 hours, measure how far each level has fallen, in millimetres, from its mark. The bucket drop is evaporation. The pool drop is evaporation plus any leak. The difference between them is the leak.
Mark both levels on the bucket itself rather than measuring the pool from a tile or a step. One object with two marks removes every question about whether the two measurements share a reference.
Using the calculator
What your result means
The net drop is the answer, and there are two quite different things it can be.
- Zero or less: the pool lost no more than the bucket, so there is no leak to find. Both were losing water to evaporation and that is what pools do.
- Greater than zero: the pool lost more than evaporation alone accounts for, and the excess is the leak.
Running the test twice, once with the pump on and once with it off, turns the size of the loss into a location for it. This is the single most useful thing a pool owner gets from the test and it costs one extra day.
| Pattern | What it points to | Why |
|---|---|---|
| Loses faster with the pump running | The pressure side of the circulation. | Under pressure the return pipework and its fittings leak harder. With the pump off they leak less or not at all. |
| Loses faster with the pump off | The suction side. | The less common pattern. Running the pump draws air in through the fault rather than pushing water out, so the loss shows up when the system is standing still. |
| Loses at much the same rate either way | The shell, the liner, or a fixed fitting. | A structural loss does not care whether the pump is running, because it is not connected to the circulation. |
| No net loss in either test | Nothing to trace. | Both results were evaporation. |
If the loss stops at a particular level and then holds, that is a further clue about where it is. whether the loss is in the liner or the structure covers what a self-stopping loss means.
How the calculation works
Three steps, and the first one is where the evaporation disappears.
- Net drop = pool drop − bucket drop, both in millimetres
- Volume lost = surface area in m² × net drop in mm, which gives litres directly, because one millimetre of depth over one square metre is one litre
- Daily volume = that volume × 24 ÷ the test length in hours
The conversion in the second step is the useful coincidence of the metric system: a millimetre of depth over a square metre is exactly a litre, so no conversion factor is needed and no constant enters the calculation.
Nothing in this calculation involves an evaporation figure, a weather assumption or a climate dataset. The bucket has already accounted for all of it, empirically, in your pool.
The surface area is the one input you have to supply, and it is the one most often guessed. work out your pool surface area in square metres calculates it from dimensions.
Assumptions and sources
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.
Millimetres in a metre
Exact by definition
SI prefix definition. Milli is a factor of one thousandth.
Hours in a day
Exact by definition
Definition of the day as 24 hours.
Difference that counts as pump dependent
Illustrative assumption, editable above
Illustrative assumption. Two bucket tests run on different days differ anyway, because wind, air temperature and sunshine differ. This is the difference between the two daily losses, as a fraction of the larger, below which the two tests are treated as the same result rather than as evidence of a pressure-side leak.
This calculator has no evaporation constant in it, and that is worth stating rather than leaving to be noticed. The bucket is the measurement, so nothing about the weather, the season, the humidity or the location enters the arithmetic.
This site does publish monthly UK evaporation baselines, and they are used as a sanity reference on how much evaporation is normal for a UK pool and nowhere else. They are derived rather than measured, one of their inputs has no source at all behind it, and a separate one decides most of the answer. None of that touches this page.
The one assumption in the method itself is that the bucket evaporates at the same rate as the pool. It is in the same water, in the same air, at the same temperature, so it is a good assumption, and it is the assumption every version of this test has rested on for as long as the test has existed.
What to do next
Four things, and the second is the one most people skip.
- If the net drop was zero or less, stop. There is no leak. Check again in a month if the water level still bothers you, because a slow leak may not show over a single day.
- If there was a net loss, run the test again with the pump in the other state. One day of work turns a quantity into a location.
- Note whether the level stops falling at a particular height. A loss that stops at the skimmer, at a light, or at a step is telling you where it is.
- Rule out the obvious before anything else: a backwash valve passing to waste, a leaking pipe fitting in the plant room, or an auto-fill device that has been quietly topping the pool up and masking the whole thing.
That last one matters more than it sounds. An auto-fill hides a leak completely, because the level never drops, and the loss shows up only on a water bill. confirm the rate of loss at the meter will see it even when the pool looks fine.
Common questions
- Why a bucket rather than an evaporation table?
- Because the bucket is in the same pool in the same weather on the same day, and a table is an average of somewhere else. Evaporation depends on air temperature, water temperature, humidity and wind, all of which vary hour to hour. The bucket experiences exactly what the pool experiences, so subtracting one from the other removes the whole question rather than estimating it.
- The bucket dropped as much as the pool. What does that mean?
- That the pool is not leaking. A net drop of zero or less is evaporation only, and this site says that plainly rather than reporting a very small leak. Losing 2 to 3 mm a day in summer is normal for an uncovered UK pool.
- Should the pump be running or off?
- Both, on separate days, and that is the most useful thing this test can tell you. If the pool loses more with the pump running, the loss is on the pressure side of the circulation. If it loses more with the pump off, it points at the suction side. If it loses the same either way, it is structural.
- How long should the test run?
- Twenty four hours, so that it covers a full day and night cycle. A shorter test can catch an unusually still or unusually windy few hours, and evaporation varies far more within a day than the leak does.
- How accurate is it?
- As accurate as your two marks. A millimetre of error over 24 hours on a 32 m² pool is 32 litres a day, so a careful mark matters more than a long test. Use one object with two marks rather than measuring from two different references.
- Can I do this with the cover on?
- You can, and it changes what you are measuring rather than invalidating it. A covered pool evaporates far less, so both drops shrink and the leak becomes a larger share of a smaller number, which is easier to see. Just keep the cover arrangement identical for the whole test.