Jaques d'Alltrades wrote:
:: The message
:: from "Phil L" contains these words:
::
::: Yes, it's the 'long before it freezes' which needs
::: understanding... once it has expanded enough to freeze solid,
::: excess water has been pushed out through the open tap.
::
:: Not so, or ice would not float. Water expands by a significant
:: amount as it turns to ice.
::

which is why the tap is open...

::: It can only burst the pipe if it has nowhere else to go! - if
::: there is an outlet it will go through this (the tap)....to
::: suggest that the tap is colder than the pipework is false, the
::: copper is less than a millimetre thick while the brass is up to
::: 10 mm, ergo the pipewater will freeze first, pushing the rest of
::: the water towards the outlet.
::
:: Nonsense - while it isn't colder, the tap is bulkier, so can drain
:: heat more efficiently from the copper and the water, whereas the
:: thin copper of the pipe has to give up heat to the water and to
:: the surrounding air.
::
The pipe contains 15 mm of water, encased in a 1 mm copper sheath, the tap
contains 25mm of water encased in a 10mm brass sheath, which is going to
freeze first?

:: Air is not a good conductor, nor does it have a high energy
:: potential.
::
:: However, leaving the tap open will certainly lessen the liklihood
:: of a burst, especially if the freezing is very slow. Ice at or near
:: freezing-point melts readily when under pressure, resulting in a
:: certain amount of it being pushed through the tap as ice, and once
:: unconstricted, the fluid fraction immediately freezes, making it
:: appear as if the ice has been pushed out as a lump.

Like I said, it expands out of the open tap.