I just measured the thickness of the walls in the main building, where I live. They are over 2ft thick (27in). I just don't believe any significant amount of heat would pass through these walls.
I'm not at all convinced of this U-value business. At least in an old building, I believe most of the heat is lost through draughts (badly sealed doors or windows, as well as ventilating bricks, etc) and through the ceiling to the attic.
Contrary to the opinion of experts, I don't believe a significant amount escapes through the walls. In fact I would go so far as to say that I suspect a lot of the so-called experts do not have a strong grasp of physics.
Eg I had new casement windows installed in one small room, and it transformed the heat-keeping properties in that room. I would guess the heat-loss was reduced by 75%.
No, they are plastered on brick or granite (probably a mixture of the two). The cottage was restored to a very high standard about 30 years ago by a trawler-owner for his mistress (who left him the day he went bankrupt).
well think again. thats 24 inches, or .61 meters, so at a K value of
2.7, thats around a U value of 4.4. Not QUITE as bad as a single glazsed window but close. Only *ten* times more than current guidelines for walls ;-)
well so it is, but once they are sealed, it becomes the dominant loss.
what you believe does not change the facts.
Sorry. irespective of what you are losing via the draughts and teh loft, you WILL be losing aroounund 150W per square meter of wall on a icy cold day.
No, its you that doesn't have a grasp of physics.
The experts are spot on, its the double glazing salesmen that are the liars.
Draughts can ruin the way a house is insulated juts as much as leaving a fridge door open ruins the cold inside. But once they are sealed, in your case, with the 4" of padding up top, its the walls that will be the next worsts thing.
Your 24" of granite is as bad as a 9" solid brick wall for heat loss. Worse if its not plastered.
I wouldn't disagree.
Draughts are the first thing to tackle.
Nut after the lofty is dealt with, walls come next.
What you probably do not appreciate, is that the difference between a draughty old stone house and a well insulated one draughtproof, is about
30:1 on heat loss.
So you have halved it with better windows. Big Deal, now do the next 90%...
par for the course. The plaster alone is probably doing nearly as much for the insulation as the granite is.
I've got figures for plasterboard - k value is 0.16,so 15mm of that has a U value of 10...roughly..that will take the U value of your stone walls down from about 4.4 to 3.3 roughly.
Look its up to you whether you believe me or not, but the way to approach this is first of all to draughtproof, secndly to do the loft, but its pretty decent with 4" of rickowwl already. Its not wrthe pending more in frankly.
But almost any insulant, even if its 3mm cork tile will really make a difference on the walls. I would be looking at insulation backed plasterboard and a replaster.
Forget the floors unless they are suspended timber. A bit of decent carpet and underlay (thick stuff) is a lot cheaper than chipping out solid concrete and laying in insulant.
And forget double glazing, unless you need to replace the whole window anyway - use thick curtains instead.
I fear you might be talking complete claptrap here. I can't be bothered to do the calculations here but I'm sure 2 feet of granite are a damn sight better than a 9 inch wall. Maris
I'm afraid that is just contrary to my experience, as is a lot of the expertise I read in pamphlets on energy-saving.
For example, there is actually an internal granite wall in this building that is over 2ft thick. (It was an external wall, but the building was extended almost as soon as it was built.) If a significant amount of heat escaped through this wall, it would presumably come out the other side. But the room the other side of the wall does not seem to me to gain any significant amount of heat this way.
Sorry, I'm a scientist, I don't give great credence to assertion. I have never seen any report of the result of any experiments on heat loss carried out in an old house. If you can provide a reference to experimental evidence I should be happy to read it.
That is simply contrary to elementary arithmetic, and to common sense. The room I am in now is over 3m high, and there are two external wall 4.5m long. According to your calculation it would be losing over 4kW of heat through the external walls, which is absurd.
The temperature outside is about 4 degrees C and inside it is 22 degrees C. The room is warmed by two 5ft radiators.
There is a big difference, but the ratio is nothing like that, maybe 3:1. I know, as this building was completely derelict when we moved in
30 years ago. It was quite possible to keep the living room at a reasonable temperature although it took a lot of fuel. But not 30 times as much.
They may not have been lying, but they may not have based their figures on real-world experiment. They may well have extrapolated from a test of the conductivity of a piece of granite in the laboratory.
As far as I can tell from a quick google, there is a vast difference in the themal conductivity of different kinds of granite.
What on earth does that mean? According to Newton's Law, the heat transfer should be proportional to the difference in temperature on the two sides of the wall. The difference at this moment is just under 10 degrees Centigrade. This means the heat transfer should be about 40% of the 150W/sq metre that you quote. This comes to about 1kW. In other words, the heat transfer according to your calculation should be equivalent to having a 1kW electric fire in the colder room. That is manifestly not the case. I believe the heat transfer is negligible, maybe of the order of 100-150W for the whole wall.
I suspect I know a bit more about physical laws than you. In any case, this is not a question of authority. If someone makes a scientific or pseudo-scientific assertion, I like to be able to verify what is said. I do not consider, "I am expert, therefore I must be right" to be sufficient.
I very much doubt if the "heating engineers" you quote did any kind of experiment in the real world. Like you, they probably looked in a table giving the thermal conductivity of samples of different materials.
As I said, the difference between different kinds of granite is so large as to make a statement giving the "thermal conductivity of granite" almost useless.
eg even the difference in conductivity of a single sample in different directions differed by a factor of 2.
These figures, like much of what you say, are contrary to common sense. I wonder if you realize the significance of a factor of 10, ie 1000%?
So the obvious answer is that you will have to measure the temperature each side of *your* granite wall in different circumstances. Of course, measurement of one point will not be enough, many measurement points will be needed. Preferably measure in the middle of the granite as well just in case it is not homogeneous across its thickness. Anything based on any other granite wall in the universe could differ markedly from yours.
The U-value is a simplified calculation and it has worked accurately for many decades. I did lots of such calculations when I were a lad; thermal conductivity, thickness, and factors for heat transfer for inside and outside surfaces, I believe. It's a simple calculation, go and google.
CIBSE moved away from simple U-values about 20 years ago, the revised calculation method involved factors for radiant and convective proportions of the heating. You'd notice this most when heating from cold; the air temperature could be 21, so you should be comfortable. Because the walls are cold you are losing a lot of radiant heat to the cold walls and so you feel it is colder than the thermometer (air temperature) tells you it is. Unless you're a building services consultant and have a suitable computer programme, U-values are good enough for radiator size calulations.
And though you may not want to believe it, dense stuff conducts heat better than stuff with voids in it. Granite would be about the most thermally conductive building material you're likely to encounter, more conductive than concrete or brick, I'd think.
There's not much point in expecting the heat transfer to exactly correspond to the inside & outside temperatures. It only does when it has achieved a steady state. The heat takes some time (hours) to travel through the granite and much heat goes into raising the temperature of the stone. The outside temperature is always changing, so it is never a steady state.
Infiltration or ventilation plays a part and there's a calculation for that as well (1/3 x N x Vol x dT, memory says, but I wouldn't trust it nowadays).
This doesn't seem to be relevant to the paragraph you appear to be responding to.
As to your points, I very much doubt if the temperature of the cold wall varies greatly from one point to another. Why should it? I took the temperature of the room, which was 11 degrees Centigrade. I am sure that is the temperature of all points of the wall, within +/- 1 degree.
As to drilling into the granite wall, which is what you seem to suggest, I do not feel any urgent need to do this. I'm simply stating that I do not believe there is a heat flow of 1kW through the wall, as the figure offered suggested.
To take another illustration, this property is rather large -
1000 square metres - and at a quick calculation the external walls have an area of about 200 square metres. If the heat loss was really 150W/m^2 this would amount to 30kW. If that were so, my heating bill - which is quite large - would be astronomical.
As you seem to agree at the end of your post, I am sure the thermal conductivity of different granite walls differs by a large factor, but I doubt if 2ft granite walls anywhere on earth conduct 150W/m^2.
But did you ever compare the result of your calculations with the real world? That is the issue, not the difficulty or otherwise of multiplying numbers.
I didn't express any view on that, one way or the other.
But people are unlikely to construct a building with concrete walls 27in thick.
Sorry, I should have said the floor area is 300 square meters. However, the external wall area is about 200 square meters, so the point I was making is not affected.
well you obviously haven't seen military buildings then ;-)
check out a mediaeval castle for a start.
Let's just say that the simple engineering solution is to separate the structural from the insulation properties of a structure.
I.e. a modern house has as much material in it as it needs to to nt bend buckle subside or collapse, then it has as much insulation in it to reduce conductive heatlosses to somewhat below the losses it will have when it gets the airchanges it needs to avoid suffocation the inhabitants.
So the regulations
1/. make sure it wont fall down
2/. seal it up totally
3/. add enough insulation to it so that its heat loss is less than..
4/. ..what it will be when they punch necessary airholes in it.
Thats as good as it gets. Unless you go for heat recovery ventilation.
27" of granite is no better than 9" of brock, taking average values for both.
If you want real insulation from concrete, abut 3 meters thick starts to get near the sort of building regs values IIRC.
Yes. Heating, ventilating and air conditioning systems designed around the U-value and heat loss/gain calculations were installed in the buildings. They worked with monotonous regularity. Proof, pudding, etc.. In fact, the only one I recall that didn't work as intended ( I hadn't done the design but was trying to find the problem) had a suspended, uninsulated concrete floor, solid brick walls and very high ceilings. The heating installed was gas-fired, top-outlet, convector heaters which, according to the U-value calcs, should have been adequate. And it was nice and warm, about 8' above floor level. They'd cocked up the heat loss calcs though. I submitted my figures but never heard the outcome.
I came across some implausible construction methods, the U-values allow you to estimate the heat input required for almost anything.
As far as I can see, what you are saying is that you made U-value calculations, and the buildings were satisfactorily warm.
To my mind, this is far from scientific proof. For example, the warmth might have been almost entirely due to ceiling insulation.
I would have thought it would be possible to measure actual heat-flow through windows, walls, etc. But I have never seen any real-world results along these lines. I would be very interested to see such data, particularly on older buildings like the one I am interested in.
As far as I can see, it is relatively easy to estimate heat transfer through internal walls, and the results seem to me to show that the transfer is a fraction - maybe 25% - of that given by U-value computations.
In my case, I am interested to know what form of insulation in a cottage would give best value for money. U-value calculations seem to show that insulating the walls would give the best return, while I am pretty sure insulating the attic will have much greater effect.
The message from Timothy Murphy contains these words:
And no one is likely to construct an ordinary dwelling with solid walls
2' thick either.
2' thick stone walls are common in older properties in many parts of Britain and I would think just as common in parts of Ireland but they are not solid. The normal method of construction is a double skinned wall with the centre packed with with rubble and just occasional 'through' stones to tie the wall together.
Because of the slightly tapered section of the stone used and a a comparative lack of 'thrus' such walls often bulge with age and can end up considerably wider than they should be in some areas. Numerous layers of plaster can also add to the thickness even if they haven't bulged.
If the walls of my hovel are typical the facing stones may almost touch in places but the lime mortar rarely extends as much as 6" from the faces so there are plenty of air pockets in the central section.
I would say that is a fairly accurate description of the walls in the building I live in (an old orphanage). Which confirms my suspicion that computing heat loss through these walls from U-values is likely to be wildly inaccurate.
HomeOwnersHub website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.