Efficiency
A thermal camera shows surface temperature and nothing else
Infrared imaging is genuinely useful for finding where a building is losing heat, and it is misread constantly because the picture looks more objective than it is.
By Manish Trivedi3 min read
The image is a map of radiation, not of insulation
A thermal camera measures the infrared radiation arriving at its sensor from each point in the scene and converts that into an apparent temperature. What it produces is a picture of how warm surfaces appear, and everything useful that comes out of it is inference layered on top of that single measurement.
It cannot see through a wall. It cannot measure the thickness of insulation, detect a cavity, or distinguish a damp patch from a cold one without additional evidence. What it does exceptionally well is show variation across a surface, and variation is what points to a problem.
Read as a comparison, it is a strong tool. Read as a measurement, it misleads.
Emissivity is the correction nobody applies
How much infrared a surface emits at a given temperature depends on the material. Plaster, brick and painted timber emit efficiently and behave predictably. Bare metal, foil, glossy paint and glass do not, and a camera pointed at them reports a temperature that can be badly wrong in either direction.
Shiny surfaces are worse than merely inaccurate, because a low-emitting surface is a good reflector of infrared. A metal radiator valve, a foil-backed panel or a window can show you a reflection of something warm elsewhere in the room, including the person holding the camera.
That’s why an experienced surveyor moves around rather than photographing from one position. A genuine cold spot stays put as the viewpoint changes. A reflection moves.
Without a temperature difference there is nothing to see
The technique depends entirely on there being a substantial difference between inside and outside, because that is what drives heat through the fabric and produces the surface pattern being photographed. A survey conducted in mild weather shows a uniform image and proves nothing at all.
Direct sunlight ruins it in the other direction. A wall that has been in the sun retains that heat unevenly for hours, and the resulting image describes the sun rather than the building. Serious external surveys are done before dawn in cold conditions for both reasons.
Rain and wind interfere as well, since a wet surface is cooled by evaporation and a windy one by convection, both unevenly. The conditions are as much a part of the method as the camera is.
What it is genuinely good at finding
Missing or slumped insulation in a loft shows up clearly from inside as a pattern that follows the joists or the gaps between batts, and it is the single most productive use of the tool in an ordinary house. Air leakage paths show as cold plumes around openings, particularly with the building slightly depressurised.
Thermal bridges at junctions, lintels and around openings appear as continuous cold lines, which is exactly the geometry that distinguishes them from an isolated patch. Underfloor heating pipe runs are visible, and so is a blocked or air-locked radiator, which shows an uneven pattern across its face.
What it is poor at is quantifying anything. It will tell you a surface is four degrees colder than the wall around it, and it will not tell you what that is costing.
Damp, mould and the interpretation problem
Wet material is cooler than dry material because evaporation removes heat, so damp shows on a thermal image as a cold region. So does missing insulation, so does a thermal bridge, and so does a draught. The camera cannot distinguish between them, and every one of them has a different remedy.
This is where the tool stops being self-explanatory and starts requiring somebody who knows buildings. A moisture meter, an understanding of the construction and a look at where the water might be coming from are what turn a cold patch into a diagnosis.
A thermal image on its own is evidence of a question rather than an answer to one.
Buying one, or having somebody bring one
Inexpensive cameras and phone attachments have made the technique widely available, and for finding loft insulation gaps in your own house they are perfectly adequate. The resolution matters less than the conditions you use them in, and a cheap camera used at dawn in January beats an expensive one used in April.
For anything consequential — a dispute about workmanship, a damp problem, a decision about spending significant money on a wall — a survey by someone trained in the technique is worth the cost, because the interpretation is the hard part and the interpretation is what you are buying.
Either way, treat the images as a starting point. The camera’s real strength is telling you where to look next.
Common questions
When is the best time to do a thermal survey?
On a cold, dry, still night or in the hour before dawn, with the heating having run for several hours so that there is a strong and steady difference between inside and outside. Avoid direct sun on the surfaces being surveyed for at least several hours beforehand, and avoid rain and wind, all of which produce patterns that have nothing to do with the fabric.
Can a thermal camera tell me if my cavity wall is insulated?
Sometimes, by showing a pattern of variation across the wall that suggests slumped or missing fill, and it is not conclusive. A uniform image can mean uniform insulation or uniform absence of it. The reliable methods are a borescope inspection through a small hole or records from whoever installed the fill, and the camera is best used to decide where to look.
Why does my window look freezing on a thermal image?
Partly because glass genuinely is colder than the wall around it, and partly because glass reflects infrared and has a different emissivity from the surrounding surfaces, so the reported figure is unreliable. Windows are one of the surfaces where thermal imaging is least trustworthy, and comparing one window against another in the same house is more informative than the absolute number.
Deputy editor, Power Your Roof
Manish has been reporting on solar basics, batteries, bills & tariffs since long before it was fashionable and would rather show the working than assert the conclusion.
