Non-Invasive Leak Detection Methods Compared: What Each One Senses and Where It Stops

The five methods in one grid

Read across a row to see one method. Read down a column to compare. The time column is relative to the other methods, because the real time depends on the building and is not promised here.

Which method suits which situation

The first grid compares the methods by what they sense. This one compares them by the situation on site, which is how a choice is narrowed in practice. Fits means a reasonable match, weakened or masked means the signal is reduced, and not applicable means the method cannot be used there. No row shows one method working alone.

Acoustic and tracer gas: two ways of finding the position

Acoustic detection listens. Tracer gas smells, in effect. Sound is quick and needs no preparation but needs pressure and a pipe that carries sound. Gas needs the pipe isolated and filled but does not care about noise.

A published study of plastic water pipe by the Institute for Research in Construction in Canada, reported in June 2000, found that leak noise in plastic pipe is dominated by low frequencies and that correlators were often set too high to hear it. Acoustic work on plastic is therefore possible but depends on settings. Acoustic versus tracer gas sets out how a survey chooses between them.

The usual tracer mixture is 5% hydrogen in nitrogen. One industrial gas supplier describes it as non-toxic and not flammable, and warns it can asphyxiate in confined spaces, so the area is ventilated. See the tracer gas page.

Thermal imaging and moisture meters: two ways of reading the surface

Both read what has arrived at the surface and neither reads the pipe. The Carbon Trust, writing on thermography for traditional buildings, notes that brick and plaster give reliable thermal readings and that metals and glass reflect nearby temperatures, so they do not. It gives a working minimum of 5 degrees Celsius between inside and outside.

The Property Care Association guidance of July 2020 states that moisture meters cannot differentiate between dampness from one source and another. Used together the two methods are still useful: a cool patch on an image is measured with a meter, and a high meter reading is checked against a pressure test. See thermal imaging limits, moisture meter readings explained and pin versus pinless meters. A phone-sized thermal camera behaves differently from a survey camera, as this comparison shows.

What none of them can do

None of the five can see inside a wall. All of them infer from the surface or from a pipe that can be closed off. A leak that comes and goes may be gone when the survey arrives. A leak under a thick slab may be beyond every method on this page. A leak that is not plumbing, such as rain entering through brickwork, may show up as damp with every pipe clear. Which method leads depends on where the pipe is hidden, as choosing a method for a wall, floor or ceiling leak sets out.

When the five methods leave doubt, a camera through an existing access point is the next tool, covered in the endoscope guide. The order the methods are applied in, the price and the point where opening up is unavoidable are on the non-invasive leak detection page. Where the findings need to go to an insurer, the leak detection report for insurers page describes what the written report contains.

Sources and dates

Pressure, sound, gas, temperature and moisture behaviours above are checked against: the Plumbing and Mechanical magazine article on the Institute for Research in Construction plastic pipe study (June 2000); the Carbon Trust article on thermography for traditional buildings; the Property Care Association guidance document on moisture meters (July 2020); and a supplier product page for a 95% nitrogen, 5% hydrogen mixture (Formier 5). All were read in October 2026.