Intumescent coatings on thin sections

Why it seems obvious

With hot-rolled steel structures it is established practice to paint the structure with an intumescent coating. In a fire the coating foams up and forms an insulating layer around the steel. It seems logical to do the same with a light gauge steel frame.

In practice this is not a viable route, and for more than one reason.

The physics works against it

The required coating thickness is set by the section factor: the thinner the steel relative to its heated surface, the thicker the coating would have to be. For thin-walled profiles this ratio is extremely unfavourable — the thickest layer would be needed exactly where there is the least steel.

There is no testing behind it

This is the decisive argument. Coating manufacturers’ test tables are valid for a given range of sections. Extrapolating beyond that is not permitted — international certification bodies expressly forbid it, and for good reason: the behaviour of the foaming layer does not scale linearly downwards.

In practice, below about 4 mm wall thickness almost no manufacturer can be found who will certify their product for such a section. A typical LGS profile is a fraction of that.

Where there is no certification, there is nothing to rely on: the coating can be applied, but nothing verifies fire compliance.

And the cost

If someone did push it through, the huge surface area makes the cost of the coating unrealistic. A light gauge steel frame has many members, each with a large surface — this is not an item a residential building can carry.

So what remains

Lining as part of a system

The usual route: the section is protected by a build-up that was tested as a whole. Fire protective boards.

Fire resistance verified by calculation

Fire design of the structure, carried out for the project and signed. Fire resistance design.

The point

The problem with intumescent paint is not that it is a bad product — it is that for this range of sections there is no testing behind it. And extrapolation without testing is an assumption, not verification.