Heat Loss from a Building

[Edward W. Thompson]

What is the relationship between internal temperature and heat loss from a building? In particular if the outside temperature is 0 deg. C and the internal temperature is, say 18 deg. C what percentage increase in heat loss would occur if the internal temperature is raised to say 20 deg.C? Does an increase in the internal temperature translate to the same increase of external skin temperature, that is is the heat transfer linear across the interface?

I appreciate that boundary conditions are important but for the sake of the exercise let us assume the building in in a sheltered location in a city.

Although I don't think it is relevant to the question, assume the construction to be brick cavity walls and insulkated, loft is insulated and windows double glazed.

On a staright linear relationship I suppose the answer is about 10% but I suspect it is not as simple as that, hence the question.

[nightjar]

Complex. It depends upon the u-values of the building materials used and even in buildings of similar construction, the proportion of different materials can change the overall value for a surface. Exposure also affects the loss - a north facing wall in an exposed location loses more heat than a south-facing wall in a sheltered one. As rooms are rarely all at the same temperature, you also have to factor in heat transfer between rooms.

Try this programme:

Bignell

[robert]

For conduction and convection ( when the temperature difference is small and the air flow forced - probably applies in this instance) the rate of loss of heat is proportional to the temperature difference.

So if its 0 deg outside reducing the internal temp from 20 to 18 will reduce rate of heat loss by 10%.

Although not a totally accurate statement of the situation I would suspect it is good enough for this query !

[LSR]

Basic O-level physics. Newton's law of cooling states "the rate of heat loss of a body is proportional to the difference in temperatures between the body and its environment". So the difference is 2/273 or about 0.73% more heat loss. The outside temperature is irrelevent.

[The Natural Philosopher]

yes.

It is precisely that simple.

To a very good approximation.

Obviously hot outside walls induce more convection and therefore a bit more heat loss. As does wind..

[The Natural Philosopher]

Probably the least correct answer anyone has posted on here ever.

[Bob Mannix]

Well the first bit was nearly right, he forgot the air had to be moving and it's the temperature of that air (the "environment") that counts. If he'd actually used this he would have seen that the new heat loss would be 20/18 or 1.11 times the old or 11% more.

[mark]

Something I don't understand here about the linear relationship of heat loss. For example: a heater maintains a room at 20 deg C when it is zero outside. If you put in a second heater I'm pretty sure you wouldn't be able to maintain 40 deg C

mark.

[Andy Burns]

But there you're talking about loss from the heater into *its* environment (the room), as well as loss from the room to its environment (the outside world).

[The Natural Philosopher]

You would.

[The Natural Philosopher]

Oh..in the sense that its not going to do the job if the radiators are only at 35C? :-)

[dennis]

Its a straight line as far as insulation goes. However probably not as far as ventilation goes. So about 10% is a good guess unless its very draughty.

[LSR]

Oops. I thought it seemed a bit low. But there's definitely some calculation I remember where the external temperature surprisingly "cancels out" . Not this one though! Sorry.

[Hugo Nebula]

On Tue, 25 Mar 2008 12:14:44 +0000, a particular chimpanzee, The Natural Philosopher randomly hit the keyboard and produced:

Oh come on! Drivel, reclaim your crown!

[Edward W. Thompson]

Thanks to all for your input.

[robert]

Yes if the room was surrounded on all sides by a zero degree environment of moving air! However in practice the room is bounded on most sides by other rooms at say 15C so the rate of heat loss through those surfaces is going up (40-15)/(20-15) times ie 5x.