Why the Frame Material Decides Where the Insulation Goes

The records give a test anyone can apply on site: a magnet, or a look at any exposed stud. Where the frame is steel, the wall will normally show a continuous insulating layer outboard of the sheathing, because a section runs from the warm face to the cold face through whatever is packed between the studs. It will also show a head detail at every floor left deliberately free rather than packed tight.

A timber-framed wall needs neither of those things. It can carry an outboard layer and often does, but it is not compelled to in the same way, and its head detail is a junction rather than a designed movement allowance.

In both walls, insulation between the studs is interrupted by every stud, plate, lintel and noggin, so what it achieves depends on how much of the wall is actually frame. In the steel wall that interruption is not a reduction but a short circuit: the sheathing and the room-side surface opposite each section stay nearer to outside temperature than the average of the wall suggests.

Material added outboard of the frame does something different again. It runs past all of the framing and it raises the temperature of the sheathing, and in the steel wall it changes the temperature of the sections themselves. That is what decides whether the frame can become a condensing surface at all, which is why the records describe the split between the two positions as the decision rather than the total depth.

In a timber wall the outer membrane and the inner sealed layer are described as a pair doing opposite jobs: one open to vapour and closed to liquid water, the other closed to air and more resistant to vapour. Leave the inner layer unsealed and warm humid air moves bodily into the insulation, depositing far more moisture than diffusion alone would.

In the steel wall the same pair exists, but where the internal control layer sits decides what the steel is exposed to. Placed on the warm face and properly closed, it keeps humid internal air away from the sections. Run with services passing straight through it and it is perforated at every fitting. And because the sections are metal with the sheathing pressed against them, moisture that does arrive meets a surface which neither absorbs nor buffers it, so it stays as liquid exactly where it forms.

On a timber wall the sheathing board stiffens the frame and gives the membrane a continuous background, and because it sits at the cold side of the between-stud insulation, its tolerance of occasional wetting matters as much as its strength. On a steel wall the same board sits pressed against metal, so its behaviour when damp and its compatibility with steel fixings both matter more.

Durability follows the same pattern. A light steel frame is thin, so the loss of protective coating at a cut, a drilled hole or a dissimilar-metal contact matters more than it would on heavier steel, and the conditions around the sections and the metals allowed to touch them are design decisions rather than site ones.

Where a light steel frame wall is built as infill between the floors of a primary structure, the head has to let the floor above move without transferring load into the studs, which means a gap whose dimension changes. At the same time the air line, the weather layer and the insulation all have to remain continuous across that gap.

The record states the trap plainly: any solution that closes the joint solidly in order to keep those layers continuous has reintroduced the load path the detail existed to prevent. That is a consequence of the frame being an infill element, and it does not arise in the same form on a loadbearing timber wall.

On a steel-framed wall, every cladding bracket fixes through the outboard insulation into a stud. It punctures the weather layer, crosses the thermal layer in metal and delivers a concentrated load into a thin section, all at one point. Their number and arrangement follow from the facade and its wind loading, which is why changing the cladding material late is never only a change of appearance.

The junction record for a facade bracket through insulation makes the same observation from the other side: as walls become better insulated, these point crossings become a proportionally larger share of the heat loss, and the number of brackets follows from the loads rather than being a free variable.

A solid engineered timber panel is not a framed wall with thicker studs. There are no studs, no between-frame insulation and no separate sheathing board, because one element is doing structure, sheathing and often the air barrier together. With no cavity, the whole thermal layer goes outboard, which keeps the thick hygroscopic panel on the warm side of the thermal line and lets it dry inwards.

That arrangement removes the between-frame interruption entirely and concentrates the question onto whatever crosses the continuous insulation to hold the cladding. It is a third answer to the same question rather than a variant of either framed wall.