A few weeks ago at the end of a post about the myth of stone walls as insulation, I mentioned that high mass materials can be useful when included inside the insulation layer. Here’s why.
Good thermal mass materials are those with a high specific heat (i.e. they require a lot of energy to warm up) and a high density. Examples include dense concrete, masonry, and stone. Because of these properties, you can use thermal mass as part of a cooling strategy in buildings.
The idea is that you let your mass soak up heat gains (from the sun, warm air from outside, occupants, equipment, and lights) throughout the day. In a good design, the mass absorbs your heat and doesn’t give much back, so you feel cool in the space. Then at night when the temperature outside has fallen below the temperature inside, you flush the day’s gains out of the mass by using high ventilation rates. See the (idealised) chart at the right. But there’s a catch.
This will only work well if the mass is exposed because of the way our bodies lose heat by radiation. Everything above 0° Kelvin radiates heat. This is infrared radiation that travels in straight lines like visible light. In typical conditions in temperate countries, radiation accounts for almost half of the heat lost from your body (45%). The rest of your heat loss is by convection (30%) and evaporation (25%).
We lose heat by radiation but we also receive it from our surroundings. Think about sitting around a campfire on a very cold night. If the fire’s big enough, you can sit there comfortably in a T-shirt and shorts, not because the fire is warming the air (all the warm air is going straight up) but because the fire is a fantastic source of radiation. Inside heated buildings, warm surfaces all around us have having this same effect on a lesser scale.
Now think about standing next to a large expanse of glazing on a cold night. Even if the window is completely sealed, you’ll feel cold because you’re radiating to the window and getting little in return.
I’ve worked on a few projects where the architect goes to a lot of trouble to include thermal mass in the scheme and when you visit the site, the contractor has covered it in lightweight plaster or plasterboard. Hidden behind the low mass material, the thermal mass can no longer exchange radiation with occupants. It’s like putting a curtain between you and your campfire. You’ve got to be able to see the mass in order for it to do its job.
You must also have a sufficient swing in temperature throughout the day and use high levels of ventilation at night. If the temperature outside stays warm all night, there’s little opportunity for flushing heat out of the mass. Similarly, if the outside temperature drops but you don’t move enough air over the mass to draw out the gains, the material will still be warm in the morning and won’t do its job the next day.
How much mass is enough? There’s a lot of uncertainty about this. The official line (BS 13790:2004) is that only the first 30mm are useful in a single diurnal cycle. But the original source for this figure isn’t clear. Certainly, beyond 30mm it’s going to be much more effective to increase the area of thermal mass rather than the thickness.
The optimum thickness will also depend on the diurnal temperature swing, levels of ventilation, and the amount of heat gains. But the truth is that there aren’t many sources of information on how much is enough. If you know of any, please post in comments.
Using some thermal modelling software, I’ll try to come up with some answers and post them here. I’d like to see how much cooling thermal mass provides as it increases in thickness, not just over 24 hours but over a week or a month. And particularly relevant to old Mediterranean houses like this one, what happens if you have thick high mass walls with no insulation layer? Results to follow.
Two things worth mentioning:
1. As part of a heating strategy, thermal mass can be a double edged sword. In continuously occupied buildings, mass can help make better use of gains and lead to lower heat consumption. However in buildings that are occupied intermittently, exposed thermal mass tends to lead to higher heat consumption as useful gains are absorbed by the mass and then re-emitted after hours when the occupants have gone home.
2. In lightweight structures you can decouple your mass from the building using thermal labyrinths or earth pipes. There’s a slightly different principle at work since the heat exchange with the mass is by convection and not radiation. Also, the ground is usually used as a heat sink. But these strategies can be effective where exposed thermal mass with night cooling isn’t practical.
carbon limited 
That’s the most succinct definition I’ve seen of thermal mass in a long time.
What software are you using? I’m just being nosy, but would be interested in the findings. Whilst you’re at it, it would be good to look at the effect of a prolonged heatwave. How many hot nights in a row breaks down the mechanism, etc.
Cheers
Thanks Mel. I’m using IES and will post results as soon as I’ve got any. Your point about the heat wave is interesting and I’d like to take a look at that as well. The 30mm figure presumably only applies to a single cycle but what happens in a thicker wall if you have 3 consecutive days of hot weather? Or 7? I’ll let you know what I find out.
OK, I have some real life experience with this problem. I lived in a very cold part of Canada – and decided to use the long history of stone to regulate temperature – as employed by the original settlers of New England.
The idea is to build a mass of stone INSIDE the house, at it’s core, as the fireplace and chimney – instead of at one end of the house. Stone at the exterior is useless in a very cold climate, as it leaks out heat rapidly, and brings in cold.
We built a 50 ton, 8 foot wide monster, which also contained pipes for 2 wood stoves. As a result, as you say, it took a few days to heat up, but once warmed up, would prevent the house from freezing even if left unoccupied (and thus unheated, we had no heat source other than wood) for three or four days. And it radiated heat out to us nicely. Unlike most wood-heated structures, the heat was gentle and very steady.
Now, with climate change, and a change of location, I’m looking at a different problem, for a new house I’m planning. The location is semi-desert, where the days in summer can be very hot (up to 45 degrees), and the nights generally cooler. In winter, the snow lasts about 3 to 6 weeks only, but night can be fairly cold, as there is little cloud cover.
Usually sun during the day.
As there are two seasons, I almost need two different houses. However, my plan is to build a poured concrete shell (for various reasons). Now the question is: should I put tons of insulation on the outside, and leave concrete (or a stone facing) inside, to radiate our own heat (or cooling) back to us; or should all the insulation go INSIDE (leaving the concrete outside as a deterrent to expected bush fires and storms…)
I shall attempt to watch this space for replies, but also, if any could be forwarded by email, that would be great.
Alex.
Alex, I can’t comment on the fire risks but here are some ideas for the rest. It would be best to put insulation on the outside of your mass, which as you say would leave the mass exposed to the inside space.
Provided your house is (more or less) continuously occupied, having exposed mass shouldn’t raise your heating requirement in winter. In summer, it will give you the opportunity to use cooling by night-time ventilation. So give some thought to how you can best move air over your mass. For example, strategically placed tilt and turn windows could give secure vent.
Also consider how you minimise your heat gains in summer. Some ideas are: careful shading particularly on south and west façades, using low energy lighting and appliances, slowing ventilation rates during the day or even using heat recovery ventilation to recover “coolth” (if your house is very air tight).
Regarding the storms, I’d expect rigid insulation and render to be up to the job, though obviously that needs checking with manufacturers. They could also comment on fire protection. Alternatively you could put extra cladding on the outside of the insulation – stone or some other hardy material?
Good luck. I’d be interested to hear how you get on.
Case,
I was meaning to ask you an insulation question, and I suppose this is as good a thread as any.
Basically I want to insulate the attic (really just the space between the trusses) on my house. Third story (or 2nd I think in the UK). Semi arid climate. Any suggestions as to material I should use?
Nick,
If you are to insulate in the attic, between joists/rafters/trusses, I would suggest mineral wool quilt or batts. It is cheap, adn easy to retro-fit as you can push it into all the areas you need to. You can start by laying the insulation parallel to the trusses, and then perpendicular over the bottom cord of the truss to build up the thickness.
However, when you say semi arid climate, I wonder what the issue is? Warm days and cold nights or do you want the insulation to reduce solar gain through the attic space to the upper floor rooms? If it is cold nights, then the above suggestion should sort that out.
One of the issues with insulation, is that the inherent lightness of the material, ie the small pockets of air created within the material reduce the proportion of energy passing through the construction element (whether that is inward or going out) but due to their low mass do not offer any time lag, or decrement to the speed at which the remaining proportion of energy passes through.
This can be problematic where you have a large expanse of say flat roof area with a tin roof, no ventilated cavity and mineral wool insulation above the ceiling finish. Whilst the mineral wool will reduce heat loss, as there is little decrement factor, solar gains on the roof are transmitted to the spaces below relatively quickly.
By the sounds of it, this is not an issue for you as you are insulating the attic space, which (in the UK at least) would be highly ventilated underneath the eaves, assuming is is not used as an occupied space.
Nick
Sunnyboy, thanks for the response. This was exactly the info I was looking for.
Basically, we live in a dry climate that gets quite hot in the day and although never too cold, can drop rapidly to around 0 celcius at night.
My issue has been that I notice, especially on the upper floor, rapid temperature changes. In other words its hot when I wish it were cool, and cold when i wish it were warm.
I am hoping a change in insulation in the attic will help reduce this.
ps.
what would the difference be using mineral wool or batts instead of common fiberglass insulation?
Nick,
there would be no difference, as the products are virtually identical in what they do. The only difference is that mineral wool is made from rock, heated and spun to create the fibres, whilst fibreglass is the same, except from glass, or silica I suppose. not too sure on the exact compositional differences.
I would imagine that addiing the insulation should certainly reduce the temperature swings.
good luck
Nick
You’ve ignored a major problem with high mass houses in summer, although your graph neatly illustrates it. Because mass evens out the diurnal temperature differences, it makes for hotter homes at night which for most people is precisely the time when the would most appreciate a bit of coolth. It’s all very well saying you need to employ a good ventilation strategy but this is independent of the mass of the house – you need a good ventilation strategy in all housing in hot climates, full stop.
Whilst there is a coherent argument that high mass homes will stay usefully warm in winter, the idea that high mass somehow contributes to keeping houses cool in summer is dubious at best.
Mark,
I take your point about mass causing a lag in peak temperature. However, the strategy of night ventilation to flush the heat out of the mass means higher vent rates than in a lightweight structure. In this case “a good ventilation strategy” involves a lot more air changes than in a low mass building.
You say that the idea that high mass contributes to keeping cool in summer is “dubious at best”. I’d say it’s tried and tested. It works well in high tech buildings where you’ve got a controls system to monitor temperature and manage purge vents. It also works well in the vernacular architecture of Mediterranean houses – small windows, thick walls; stay cool by limiting gains during the day and opening up at night. If these houses were timber frame, they would quickly heat up the following day after just a few air changes. Instead, they stay cool.
In my case, it’s not just theory (or something we push on clients). I live in a house in central Italy with half-metre thick masonry walls. We’re pretty good about minimising solar and ventilation gains during the day and opening up the house at night, and even when it’s very hot outside it is invariably cool and comfortable inside. We haven’t had a problem with bedrooms being uncomfortably warm at night as the air coming through open windows quickly brings down the resultant temperature (even if the mass keeps the radiant temp higher for a few hours until it cools).
I’m not sure what is dubious about that, except that half a meter of mass is probably a lot more than necessary to do the job.
As an aside, a colleague of mine has used a combined strategy on a domestic project: she’s put high mass in areas of daytime occupancy and low mass in the bedrooms. In this way, the daytime areas can be purged at night (so they’re ready to absorb more gains the following day) while the bedrooms, once vented with evening air, quickly cool to a comfortable temperature.