On Mar 26, 1:57 am, phil...@doramail.com wrote:
Here is another somewhat similar system from David Mears:
It uses gravel under the slab with water to store heat. There is a
liner to prevent the water from leaking into the surrounding ground.
The house has a long and successful track record.
I did ask David about the house, and what he would do differently in a
new version. He thought (without running the numbers) that with
better insulation and sealing that just the slab might suffice for
thermal storage without the extra complication of the lined gravel
chamber under the slab.
Some of the references on storing heat in dirt under a slab put a lot
of emphasis on making sure that there is no ground water movement
through the earth, because even very slow moving water will transfer
heat away quickly. It seems like some form of liner to keep any water
you add to increase the thermal storage would be wise?
On Mar 26, 4:57 am, phil...@doramail.com wrote:
This idea of storing heat under a building seems to hold a lot of
interest and promise. I'm trying to work out a design for a post and
beam, straw bale workshop with underfloor heat storage that would
(hopefully) provide most of the heat to the building.
As the various posters have alluded, there are pros and cons with the
various choices of heat storage medium.
Sand: The biggest advantages are that it supports the floor, cannot
leak out, and it is easy to work with. Disadvantages are that it
only holds about 1/3 the heat as an equivalent amount of water, and it
can be moderately to very expensive to bring in truckloads of the
stuff. Also, piping must be embedded in the sand to distribute the
heat from the collectors.
If I remember correctly, the heat flow characteristics of sand are
what seemed to generate the most rancorous debate. The moderate
thermal conductivity and moderately high specific heat make for a time
delay effect, where, in a properly designed system, the stored heat
doesn't start making the floor warm until the heat is needed. It
seems that this final issue is the least well characterized and
tested. One would hate to have a 90 degree floor in August! Alan's
system at Daycreek seems to work well for him, but I haven't heard if
warm floors are a problem in late summer, early Autumn.
Water: Advantages are that it holds about 3x heat compared to sand.
Also, a tank of water makes system design much easier. Just pump the
cool water from the bottom, through the collectors, and let the heated
water stratify on the top. This would be a natural application for a
drain-back system. No antifreeze or heat exchangers required, thus
helping to trim back system costs. The one-time cost of filling the
tank would be moderate in most locations.
The disadvantages of hot water storage are that a structural floor
must be installed over the tank, and the water can leak out. A water
tank under a floor will give a warm floor all of the time, as the hot
water will naturally stratify to the top.
Gravel / rocks or mud present design characteristics not unlike sand
and water. The idea of wet sand is interesting. Wet sand has a
higher specific heat, yet still retains the supporting characteristics
of dry sand. However, I would expect that heat in wet sand will tend
to rise faster than in dry sand due to the boyancy of hot water. This
may cause problems with warm floors in the summer. As for mud, I don't
see any advantage to using mud over plain water, as mud cannot(?) be
used to support the weight of a floor, and mud has a lower specific
heat than water. Someone please correct me if I'm missing something
As has been widely recognized, groundwater migration, if not carefully
controlled, can literally wash away a large amount of the stored heat,
regardless of the storage medium. It must be controlled in order for
any ground based storage system to work.
My idea is to insulate the bottom of the floor with 2" of rigid foam
board (or wood, if one is adverse to foam), and put PEX in the
concrete, with a small circulator pump moving heat from the storage
medium into the floor only when required. This would nicely get
around the unwanted warm floor problem, especially with water based
I like the idea of a water tank, but it needs to be inexpensive, and
reliable. What about filling the tank with some material that would
support the weight of a simple concrete floor and building structure,
yet be mostly empty, to hold the water... and satisfy local building
codes? What might such a material be? I would like to reduce the
amount of concrete to a minimum due to it's cost, and high embodied
energy. Could some of the technology being used with pressure treated
wood foundations be used to make a tank? Would it last?
My hope it that a well insulated building with good passive design
will not require a vast heat store, and make such a system feasible.
Here in north-central Indiana, we get very little sunshine from Mid
December to Mid February, so stored heat must be relied on to get
through the worst of winter.
So much for my ramblings.