On Thu, 22 Jan 2004 17:24:50 +0000, Dean Richard Benson
wrote:

On Thu, 22 Jan 2004 15:28:12 +0000, Andy Hall wrote:
That's exactly what I'm saying. Insulate it to the level of the walls
which is a U value of 0.25 to 0.35 for a place with cavity walls and
insulation. That equates to 100-150mm of glass fibre, not
350mm-600mm.

I think I am seeing your point.

On an older property, the U value of the walls can easily be 2.0.

Yep, my house is 250 years old, so really thick walls, but certainly no
cavity.

This will have a U value better than that which is for a 220mm solid
brick wall with plaster. A 335mm one is at about 1.6 W/m^2.K

For other materials it's necessary to know the construction of the
wall. If there are several components to it then the effect can be
summed mathematically using reciprocals but if there is air inside
that can make a fair difference even before insulation is added.
For example for a cavity wall of two 105mm brick layers with 25mm air
gap, the U value drops from 2.0 to 1.5.
As soon as insulation is put into the gap it falls to around 0.5




There is very little point in reducing the U value for the roof below
0.25 in the context of that.

Is there a way to find out the current heat loss through a wall?

There is a reasonably rigorous way of doing it in the Approved
Document to the Building Regulations for Part L1.

http://www.odpm.gov.uk/stellent/grou...reg_600288.pdf

although the examples work on how much insulation is needed to achieve
a certain U value for typical constructions.

There are also various references to U values on the internet if you
search with Google.

If the wall is very thick, then you will also have the issue of
thermal mass - in other words the amount of heat required to raise the
temperature of the masonry itself. This has an impact on how
quickly the air warms from cold, since to begin with, until steady
state is achieved, you are also having to warm the walls.
This is a separate issue but can influence heating design and
controls.



Thanks

Dean

..andy

To email, substitute .nospam with .gl