Grunff wrote:

Thanks for the input Andy. While I understand your logic, something
still feels instinctively wrong - can't the same calculation be done to
show that the difference between copper walls and 100mm celotex walls is
negligible? What am I missing?

The t/k term! For 100 mm of Celotex you have t = 0.1 m and k = 0.019
W/mK, if memory serves. So t/k = 5.26 m^2K/W and adding the boundary
layer resistances make a total resistance of 5.44 m^2K/W or a U-value of
about 0.18 W/m^2K.

The point is that with the single-glazed window the glass contributes
about 2% of the total thermal resistance whereas the Celotex is
providing nearly 97% in your example. Taking up Nat Phil's point, with
the Celotex wall it doesn't make much difference if the outer boundary
layer is blown away in a gale; with the s-g window it does.

It helps to think in terms of R-values rather than U-values. To use an
electrical analogy, we're talking about resistances in series here, so
to think in terms of their conductances would be a bit silly.

--
Andy