Thermal bridging in Australian buildings.

What thermal bridging is

A thermal bridge is a more conductive path that lets heat bypass the insulation in a building element — a steel stud through an insulated wall, a concrete slab edge, a balcony that runs straight through to the floor inside, or the metal frame around a window. Heat takes the path of least resistance, so it concentrates through the bridge instead of spreading evenly through the insulated field around it.

The result is twofold: more heat lost through that path than the insulated R-value suggests, and — more importantly — a cold spot on the inside surface directly behind the bridge. In a Australian context, the bridges that matter most are metal: steel wall framing, and the aluminium that frames almost every window and door.

Why it matters: it's about the cold line, not the spreadsheet

A thermal bridge isn't a problem because of the heat it loses on an energy report. It's a problem because of where it gets cold. The bridge is the coldest continuous surface in the room, and cold air holds less moisture before it saturates. So the air touching the bridge is the first place the surface humidity climbs, the first place mould takes hold — well before the air ever reaches its dew point — and, if it gets cold enough, the first place condensation forms.

Under the NCC's condensation-management provisions — expanded again in NCC 2025 — condensation and mould are no longer a callback you argue about after handover; they're something the design has to manage up front. So the cold line isn't just an energy number quietly getting worse. It's a health, amenity and compliance question, and it sits exactly where the bridges are. (More on how it's measured: fRsi, the Mould Index and TIRS.)

What the NCC & NatHERS actually require

Australia started accounting for thermal bridging under NCC 2022 — but narrowly, and only where the framing is steel.

• Steel framing is counted. NatHERS accounts for thermal bridging in repeating steel frame elements — the studs and joists in floors, walls and ceilings — calculated to NZS 4214:2006, with default values applied where the framing detail isn't given.
• Timber framing is exempt. Thermal bridging isn't applied to timber-framed construction at all.
• Steel frames need a thermal break. Where lightweight cladding (and lining) fix directly to the same steel member, a thermal break of at least R0.2 is required between the frame and the cladding — for external walls and metal roofs alike. Steel-framed floors are addressed separately, through continuous insulation (around R0.5) rather than a thin break.

That's a real and sensible start — a steel stud shorts straight through the insulation, so it earned a number. But two things are worth understanding clearly: how little the prescribed "fix" actually does, and what the whole framework still leaves out.

The thermal break — and its limits

The mandated break is a minimum of R0.2. Invert that and it's a U-value of 5 W/m²K — roughly the insulating value of a single pane of glass. (For scale: a single-glazed aluminium window is about Uw 6.9; a single-glazed timber or uPVC window about 5.5.) The prescribed cure for the steel thermal bridge is, in insulating terms, about one sheet of glass.

It's also a modest intervention physically. R0.2 is a through-thickness R-value, not a coverage requirement — a thin strip over the faces of the framing members satisfies it, isolating the fixing line while the steel's web and sides stay fully conductive. And where the cladding is fixed indirectly — on battens over a cavity — no thermal break is required at all, on the assumption that the cavity interrupts the metal path. A steel batten screwed back to the steel frame doesn't interrupt it — it's metal-to-metal contact, bridging straight through a gap the 1D calculation quietly treats as air.

The code's prescribed fix for the thermal bridge it does count is, in insulating terms, a single pane of glass — and a batten can make it optional.

The bridge nobody counts: the window frame

Here's the larger gap. The framework above is about the steel in the walls, floors and ceilings. The aluminium that frames the windows — the most conductive material in the whole wall — isn't counted as a thermal bridge at all.

A typical house carries around 25m² of windows. If those windows are aluminium, that's roughly 80 lineal metres of frame wrapped around the openings — about three lineal metres of frame per square metre of glass. Architectural aluminium conducts heat around three times faster than the steel stud the code does count, and more than a thousand times faster than the timber it exempts. And unlike a stud buried in the wall, the frame runs fully exposed from outside to inside — the metal path, external face to internal face, with nothing covering it. Yet none of it is treated as a bridge.

Instead, the frame is folded into the whole-window U-value (Uw) — an area-weighted average of glass and frame. Glass dominates the area, so the frame's conductivity is smoothed into a respectable-looking number rather than reported as the cold line it is. The full argument, with the numbers, is here: the thermal bridge nobody counts. And it's the same averaging that lets a good U-value still grow mould at the sill.

The break that isn't: conduction vs convection

There's one more reason thermal breaks promise more than they deliver — and it's physics, not compliance. A thermal break is a conduction control. It drops a low-conductivity material into the solid heat path to slow conduction through the material. It does nothing about convection — heat carried by moving air.

Modern façade and window detailing deliberately introduces moving air. A ventilated façade, or a pressure-equalised and drained window cavity, opens an air path to outside so the assembly can shed water and dry. But that same external air then washes surfaces inboard of the thermal break — reaching the internal skin and carrying the outdoor temperature with it. The cold arrives by air movement, not through the solid, so it walks straight past the break. A "thermally broken" frame with its break out in a ventilated cavity can still run cold on the inside face. It's a conduction answer to a convection problem.

How thermal bridging is actually assessed

This is the deepest limitation of the prescriptive approach, and it's a question of dimensions. A thermal bridge is a two- and three-dimensional heat-flow problem — heat spreads sideways, funnels through the bridge and concentrates at a surface. But an R-value is a one-dimensional quantity: resistance straight through a layer. The R0.2 break and the NZS 4214 framing correction are simplified, essentially one-dimensional calculations. They give you a number, but they can't resolve how the heat actually spreads around the bridge, or how cold the surface gets at the single worst point.

Resolving that needs 2D/3D numerical modelling — ISO 10211 for construction details, ISO 15099 for windows (the method behind LBNL THERM). That's what we do at AFRC Lab: model the real section, in two dimensions, to ISO 15099 — and every section is cross-checked against THERM, more than 1,000 times over in a public register, at about 99% agreement. From that model you get the thing the 1D number can't give you: the actual surface temperature on the coldest line, and whether it stays above the dew point. That's an fRsi condensation screen and, taken further, the AIRAH Mould Index via TIRS.

None of this is a criticism of the code. Counting steel framing was the right move, and the window U-value is a perfectly good comparison number. The point is narrower and more useful: a one-dimensional rule is a screening tool, not the real heat flow. If you need to know whether a window or junction performs — and stays dry — in the wall it's installed in, you model the coldest line, not the smoothed average.

The thermal bridging series

This guide is the hub for a series working through thermal bridging in Australian windows and façades, one piece at a time.

Related reading: a good U-value can still grow mould at the sill · condensation is now a compliance failure · performance isn't fixed until it's in the wall · fRsi, the Mould Index and TIRS.

Common questions

Working estimates and indicative values. The ~80-lineal-metre and ~25m² figures are representative estimates for a typical house (about three lineal metres of frame per square metre of glazing), not published statistics. Conductivity comparisons use the ISO 10077-2 standard frame value for aluminium (λ ≈ 160 W/m·K; architectural alloys run ~160–205), mild steel (≈ 50 W/m·K) and softwood framing timber (radiata pine ≈ 0.10 W/m·K, per NCC Spec J1.2). The NatHERS/NCC framing treatment applies to repeating steel framing in floors, walls and ceilings (NZS 4214:2006), timber exempt; the R0.2 thermal break applies where cladding fixes directly to steel. Project-specific thermal simulation to ISO 15099 / NFRC 100; modelled results are indicative to the systems and details provided, and final NCC compliance is determined by your assessor or certifier. AFRC Lab (Advanced Fenestration Research & Certification) is an independent company; it is not the AFRC and is not affiliated with WERS. TIRS™ is a trademark of AFRC Lab Pty Ltd (patent pending).