At severe risk of a hijack of another thread:

http://www.greenbuildingforum.co.uk/forum114/comments.php?DiscussionID=9243

Posted By: SteamyTeaNot convinced about water walls yet, the SHC is good, the price is good, but have a nagging doubt about something I just can't put my finger on at the moment.

If you do put your finger on it do describe the experience here. I for one would be interested to hear of any likely problems with such set ups.

In the same thread:

Posted By: tonyI have seen people use oil drums full of water -- they are best at storing heat but I would be scares of them leaking

I too like the idea of oil drums but have some doubts about their longevity. Any thoughts on using them with some sort of inhibitor in the water, anyone?

Given that water has a specific heat that's only around 30% greater than dense concrete or stone, is it really worth it?

You're right, Ed, I took the data from the Engineering Toolbox without noting that they had transposed the imperial/SI columns between the table for fluids and the table for solids. On the solids table the imperial units are the left column, on the fluids table the imperial units are the right column. Very annoying!

Here are the links: http://www.engineeringtoolbox.com/specific-heat-solids-d_154.html and http://www.engineeringtoolbox.com/specific-heat-fluids-d_151.html

Right, my views:

You want to get the interior of your house at a stable temperature, say 21°C
One way to do this is to just heat/cool the air as it comes into the house. If it gets too hot or cold, you either change the flow rate or the size of the heater/cooler. Air has a SHC of 1 J.kg^-1.K^-1 (lets not get bogged down with varying humidity).

Another way is to heat or cool some thermal mass. So to keep the air at 21°C, you have to heat the thermal mass to above that temperature, cool it to below that temperature when the air is too hot. The rate of transfer will be affected (or is it effected) by 3 properties of the thermal mass, its mass, the surface area that is in contact with the air to be heated and the thermal properties, this is nicely described by thermal inertia J·m−2·K−1·s−1/2.
Plain old stone (granite) has a VHC of about 217 kJ.m^-1.K^-1, and a conductivity of about 2.8 W.m^-1.K^-1.
So that will be about 780 J·m−2·K−1·s−1/2.
Say the air that needs to be heated needs 10 kWh a day, that will be 3.6 MJ that needs to be transferred to/from the wall.
Heat transfer is most effective when half the energy from the hotter has transferred to the colder. So let us say that there is a 6°C (HDD 15) temperature rise needed to keep the air at 21°, the wall will need to be 16°C above the internal air temperature, so 36°C (quite warm for a wall, but manageable and not so different from UFH).
So to work out the mass of thermal mass needed we need to know the wall area and the lass of air to be heated.
Lets say that there are 3 tonnes of air and the wall is 5 m long and 5 m high, 25 m^2, but transfers energy from both major surfaces (lets ignore edges to keep things simple), so 50 m^2.
The wall has to be able to release at least 3.6 MJ so that the mean temperature of the air stays at 21°C.
The wall starts releasing energy at 36°C and at a rate of 780 J·m−2·K−1·s−1/2.
So for a wall of 50 m^2 that is 39 kJ·K−1·s−1/2
When it gets to half way between 21°C and 36°C it must have delivered enough energy to have supplied the air with 3.6 MJ (or 10 kWh).
So that is 39 kJ·K−1·s−1/2 x 7.5°C or 292.5 kJ.s^-1/2

Now 3 tonnes of air need to be heated and that requires 3.6 MJ per day or 3.6 MJ per 86400 s (3.6 MJ.86400s^-1).
3.6 MJ.86400s^-1 / 292.5 kJ.s^-1/2 =41.67
As I had dropped the M^2 and kg units earlier but have specified that this to heat 3 tonnes of air, I have to put the units back in. So that is near enough 42 tonnes of granite.
Granites has a density of about 2.8 tonnes per cubic metre, so that will be 15 m^3 of granite. 15 m^3 / 25 m^2 = 0.6 m
So your granite wall be 5 m wide, 5 m high and 0.6 m thick to store enough energy to stabilise 3 tonnes of air in a house at 21°C when there is a 6°C temperature difference.

To be honest I am not sure if this is right, though it does seem to make sense and granite walls are often about 2 foot thick down here. I may also have made heavy weather of the calculations, but that was mainly because I was working it out as I typed, it is probably easier to introduce the mass of air and the temperature difference right at the beginning, but that would assume that the equations/values for thermal inertia, volumetric heat capacity, thermal conductivity and density were known.
More than willing to be put right on my maths/thinking though and shown a simpler way to calculate the heat transfers as I have not taken any airchanges into account, I just made the assumption that the air needs heating, not why it needs heating.
But just out of interest to heat 42 tonnes of granite from 15°C to 36°C would take 370 MJ or 100 kWh.

Posted By: SteamyTeaAnother way is to heat or cool some thermal mass....I just made the assumption that the air needs heating, not why it needs heating.

All interesting, but (& I may well be missing the point) isn't it more the idea that provided the only effective thermal interface is with the house interior (i.e. no losses from the mass directly to the outside) the mass simply(!) absorbs & releases heat whenever its temperature is below or above the temp of the house air. The greater the mass, the greater the inertia in the system and so, as tony said in
http://www.greenbuildingforum.co.uk/forum114/comments.php?DiscussionID=9243

Posted By: tonyThe more the merrier and the longer and better it will work so long as it is inside the thermal envelope.

With the properties of the thermal mass affecting how swiftly the mass exchanges heat with the interior at any given temp difference.

Once an initial equilibrium[1] has been reached does the presence of the mass add anything at all to the energy input required by the house?


[1] During which phase the mass will be being 'actively' heated (or in a hot climate, cooled) by whatever is being used to maintain the indoor air temp, of course.

Posted By: SteamyTeaSo the shape of the thermal mass is important as well as the mass.

Yes. I'd intended to mention that too in along with "thermal proprties". Thanks for adding that clarification.

Another way is to think of it as a pendulum, but one where the return swing is inversely proportional to the initial swing.
The quicker you move it to the right, the slower it will return to the left.

Nicely put.

So filling all (the normal thickness, or just a little more) stud walls with thermal mass (dry sand ?) might be better than having one big thick wall of equivalent thermal mass in the middle of the house?

Increasing thermal mass will indeed help reduce temperature changes due to external input but the title of the thread talks about "thermal storage".

And insulated water wall with an air passage would be similar to a fish tank with bubbles blowing though it.
Or a shower, if you think of the water as bubbles.

Here is a bit about thermal inertia in buildings:
http://www.telefonica.net/web2/josepsolebonet/index_archivos/Workshop_Thermal_Inertia.pdf

From my understanding of it (and I think I have made a school boy error in the calculations by not using mass at the begin, though the concept is the same), if you have too much mass, you never get to store the energy quick enough. Seems to me that going under size on the mass is the most effective method, bit like going undersize on an inverter, you can discard the very rare occasions when it is too hot, or very cold, These outliers add very little.
Though that Telefonica Workshop paper does say that it works best in a very well insulated house, which possibly has good airthighness. The test house is in Madrid though, a very different climate regime to ours.
Tony, do you log your house temperature on an hourly basis? Could line it up with some weather data and gauge really what is happening.

Hard to tell what is really happening at any sampling rate over an hour, even an hour is a bit crude, can lead to large error margins.

What should happen when working out how much thermal mass to incorporate into a house is to decide on the maximum temperature swing that is permitted, then work backwards to find out how much mass is needed. This is one advantage of water, it is easy to increase the quantity, and sometimes even easier to reduce it.
Water also behaves differently at room temperature than timber, rock or brick as convection currents can, in effect, increase the surface area that is used to transfer the heat. Though the high VHC means that a lot of energy has to be transferred before a rise/drop in temperature is noticed.
I still think that low mass, well insulated and controlled ventilation/heat recover is the way to go in the UK.

Posted By: SteamyTeaHard to tell what is really happening at any sampling rate over an hour, even an hour is a bit crude, can lead to large error margins.

The Hockerton houses drop from summer temperatures to around 17 °C at the end of February or whenever. I can't imagine you'd need hourly monitoring to follow that.

Are there other factors that could have affected it, such as a large south facing roof (assuming it is room in roof).
And do you think that a large pile of bricks or a 45 Gallon barrel of water would have reduced this.
I live in a low mass place, not ideally orientated to catch the sun, but it is very really too hot, and very really too cold. Even got some data somewhere to show how stable it is.

I think the trouble is that it is very hard to compare identically designed houses, side by side.

At least one interesting spreadsheet on Tony's site. Some reading required.

ETA: done some reading now. To partly answer one of my own questions above, in the very cold weather at the end of 2010 in a period of a few hours over a week (171.5 hours) ending 2010-12-10T17:59 his total electrical consumption was 384 kWh for an average of 2239 W. He seems to have a floor area of 240 m² so that's 9.33 W/m² - less for actual heating as such.

Posted By: Ed Daviesso that's 9.33 W/m²

One of the things that the GBF Energy use thread seems to be showing is that houses with a larger floor area use less energy per m^2. So not always a good gauge.
Still low though.

Given that sunshine entering windows lands on the floor...would that be a better place to put water tanks for the purposes of increasing thermal mass?