Thermal Imaging for Air Leakage Detection (Blower Door)

Thermal imaging for air leakage — finding draughts under depressurisation.

On its own, a thermal camera struggles to tell air leakage from missing insulation — both look cold. But run the building under depressurisation with a blower door, and air leaks come alive: incoming cold air streams over the internal surfaces, painting unmistakable 'fingers' and plumes that move and grow. This combination is the single most powerful air-leakage detection method we have.

Air leakage is invisible and, in still conditions, often thermally quiet — the small amount of air trickling through a gap may barely register. The trick is to amplify it. By sealing a blower door into a doorway and depressurising the building, outdoor air is forced in through every leak at once. As that cold air streams across the internal surfaces around each leak, it chills them in a distinctive, dynamic pattern that the thermal camera captures clearly.

The signature of a leak

Air-leakage patterns look different from static insulation defects, and learning to read them is the key skill:

Air leakage vs insulation defect — thermal signatures

Feature	Air leakage (under depressurisation)	Insulation defect
Shape	Cold 'fingers', streaks, fans spreading from a gap	Cold patch matching the missing area
Behaviour over time	Grows/cools as depressurisation continues	Static — doesn't change
Response to the blower door	Appears or intensifies when fan runs	Largely unchanged by the fan
Location	Junctions, penetrations, skirtings, hatches	Wall fields, between studs, cavity

How a combined survey runs

Establish a sufficient temperature differential (warm inside, cold outside) so incoming air contrasts with the surfaces.
Seal the blower door into an external doorway and depressurise the building to a steady pressure (around 50 Pa).
Sweep each room with the thermal camera, watching for cold fingers developing at junctions, skirtings, service penetrations, loft hatches, window frames and sockets.
Confirm each suspected leak with a smoke pencil at the surface (it should be drawn into the gap), and quantify the significant ones with an anemometer.
Photograph and annotate each leak, building a prioritised register tied to the blower door's whole-building leakage figure.

This is exactly the workflow described in the Airtightness guide's leak-detection article — thermal imaging and the blower door are two halves of the same technique. The blower door tells you how leaky; the thermal camera (plus smoke) tells you where.

Where leaks show up

Skirting and floor-to-wall junctions — cold air drawn up from underfloor voids fans across the lower wall.
Loft hatches and ceiling penetrations — cold fingers spreading down from above.
Service penetrations — pipes, cables, ducts and downlights, each with a localised cold plume.
Window and door frame perimeters — cold lines tracing the frame-to-structure gap.
Electrical sockets and switches on external walls — small cold blooms, especially with dot-and-dab plasterboard.
Intermediate floor-to-wall junctions — where the floor void connects to the external wall.

Why this matters for moisture, not just heat

The complementary external view — pressurisation

Depressurisation, which pulls cold outdoor air in, is the workhorse for internal leak hunting because the incoming air chills the internal surfaces the camera is watching. But the reverse — pressurising the building so warm indoor air is pushed out through the leaks — has its own use. Under pressurisation, an external thermal survey (in cold weather) can catch warm air escaping at the outer face around junctions and penetrations, and a smoke fogger inside makes leaks stream visibly out for photography from outside. Some components, particularly flaps, overlaps and one-way details, leak differently depending on flow direction, so surveying in both directions catches leaks that a single direction would miss. As with the blower door itself, averaging and cross-checking the two directions gives the most complete picture of where the envelope leaks.

Reading the magnitude of a leak

Not every cold finger matters equally, and part of the skill is gauging significance so the sealing budget goes to the right places. The thermal signature gives qualitative clues — a long, fast-developing, deeply cold fan indicates a strong leak driving a lot of air, while a faint, slow cold smudge indicates a minor one — but the camera does not measure airflow. That is where the anemometer comes in: held at a located leak, it measures the actual air velocity, turning 'there is a leak here' into 'this leak passes this much air'. Combined with the leak's position in the stack-effect pressure field (high-level escape and low-level entry leaks matter most), this lets the survey rank leaks by genuine impact rather than by how alarming they look on screen. A dramatic-looking cold streak near the neutral pressure plane may move far less air than a modest one at the wall head.

Common questions from clients.

Should leaks be found with the fan pushing air in or sucking it out?

Both have a place. Depressurisation (sucking air out, so cold air streams in) is best for internal hunting, because the incoming air chills the surfaces the camera watches. Pressurisation (pushing air out) suits external surveys and smoke-fogger demonstrations from outside. Some components leak differently by direction, so testing both ways catches leaks a single direction misses — which is why best practice averages and cross-checks the two.

How do you know which draughts are worth sealing?

By quantifying and ranking them, not by how dramatic they look. The thermal image locates leaks; a smoke pencil confirms genuine airflow; an anemometer measures how much air each significant leak passes; and its position in the stack effect (high-level and low-level leaks matter most) sets its priority. A vivid cold streak near the middle of the building may move far less air than a modest one at the wall head — so we seal by measured impact.

Can thermal imaging find draughts without a blower door?

Sometimes it catches obvious leaks on a windy day, but unreliably — in still conditions many leaks are thermally quiet and a cold patch can't be distinguished from missing insulation. Pairing the camera with a blower door under depressurisation is what makes air leaks reliably visible and distinguishable, because they bloom and move when the fan runs. For serious leak detection, the two go together.

How do you tell air leakage from missing insulation on a thermal image?

By how the cold area behaves. Run the blower door: air leaks appear or intensify as cold fingers and fans that grow over time and respond to the pressure change, while insulation defects stay as static cold patches that ignore the fan. Confirming with a smoke pencil (drawn into a genuine leak) removes any remaining doubt. Watching the response to depressurisation is the key.

Will sealing the leaks you find cause condensation?

Only if you seal without ventilating. Reducing air leakage removes the accidental ventilation a leaky house relied on, so controlled ventilation (extract or MVHR) must be provided as the building tightens — 'build tight, ventilate right'. Done as a pair, sealing leaks makes the home drier and healthier; done in isolation, it can trap moisture. We always design the ventilation alongside the sealing.