Here's my idea - is there a prayer it will work? I am designing a three season house on an island off Rhode Island. I plan to build with ICF's, SIP's and the floors to be steel bar joists, metal deck with 3" concrete (polished for a finish). The heat will be rediant hydronic in the slab with solar thermal sourcing. For both irrigation and fire protection water I plan to have an underground water vault containing about 10,000 gallons with a footprint of about 400 SF. The "floor" of the vault would consist of only a rubber membrane (so good thermal transfer) The sides and top could be insulated to whatever degree needed. The key is that this island has the highest electric rates in the country - $.62 per kWh!! so I don't want to run a heat pump. Here is the question - could I use the 10,000 gallons as a heat resevoir and as a collector of ground heat through the vault "floor" in order to circulate water of about 45-50 degrees through the slab (possibly via an exchanger) in order to keep the house above freezing during unoccuppied winter months? My thought is that the vault would act like a compressed horizontal ground loop because it would be efficient at capuring all the heat in it's footprint and equallizing it within the water volume. I imagine that horizontal loops buried directly don't do well at capturing the heat between the pipes so I wonder what the difference in overall area required would be for a vault versus loops - all else being equal. My other question would be whether i would have enough temperature differential to effect a heat transfer from the vault temperature water to the house interior? Remember this is in an unoccuppied house seeking to keep the house above the high 30's range of temp. All comments (even those about my sanity) are welcome....

This is what it all boils down - temperature differential.   In my opinion, I don't see this working.  The water in the tank could very easily get down to 40° in the cold of winter which is just not warm enough give much heat to your house.

Good luck

If the house was 40 and the tank was 40 and I just kept circulating would the house "stabilize" at 40? Would this be thermally equivilent to placing 10,000 gallons of "thermal mass" inside the house?

The house, no matter how well insulated, will lose heat to the outside when it really cold. The question is, can your 10,000 gallons of water replace the heat as fast as it escapes to the outside? I just don't see it happening. I guess you can try it out and let us know.

With those construction methods and materials and enough solar thermal sourcing I think you could make it work. South of Cape Cod winters aren't so bad relative to the rest of New England (I grew up near Gloucester, MA and still have family in Bristol, RI).

Solar thermal panels have greatly improved and the combination of a thermally massive and tightly constructed house AND 10k gallons of water should be enough to let the house 'flywheel' through a cold snap.

Get a good load calc and apply worst case rather than typical design day conditions. But since all you care about is keeping the house safely above freezing (40) rather than comfy for habitation (72) the numbers should look much more doable.

Off the top of my head the worst case scenario is the possibility of 3-4 days with little sun and lows near zero. Something like that (with better data than my dim memory / hunch) becomes your design challenge

With electric so high how will you cook, dry clothes, and heat domestic water...propane? If so, proceed with your design and rely on propane to heat some water in the event of an extreme cold snap?

Solar Photovoltaic probably looks pretty good at 62 cents per kWh - can you net meter or not? At that price I'd be looking very hard at going completely off grid.

I haven't completely wrapped my mind around whether to insulate the 10k gallon cistern. My hunch is NOT to leave it open to ground temps but rather to insulate it and let it accumulate solar thermal heat.

What would be the purpose of keeping the house just above freezing? My first guess would be to keep the pipes from freezing, but if it's going to be unoccupied for an extended period of time anyway, it would be far cheaper to turn off all the water. Drain the lines, clear the toliets of standing water and use a heat strips to protect the water source values. It will be far less electric power for heat strips on the values than it will be to circulate 10k of water all winter. 

Even if it is possible, your going to have to have some kind of safe guard to provide emergency heat the the temperature drops below say 35 degrees, if it snows and your solar panels are covered, it could take some time for the snow to melt off of them.     

For a water vault to work well, I think it should be sized correctly for the load, insulated on all sides including the bottom, and pre-charged during the summer with solar thermal.  I would not depend upon the earth to raise the water temperature enough to be effective.

I used a water storage vault under the floor in a solar energy home in South Carolina. I found that Insulating a water vault on the bottom side can be tricky.   I had to install an EDPM liner because the weight of the water moved the concrete floor down enough to cause the storage vault to leak.  To correct for this situation, the insulation should go between the liner and the concrete floor and not under the concrete slab.

I am with Tech on this one, why bother heating it at all, just drain everything and let it freeze.  If you plan to come and go, doing this will require more heat to get it up to temp since it will have additional "cooling" you’re adding, although maybe the starting higher temps would offset this? 

 

And as Alton said, 10,000 gallons weights a lot, something like 8.5 lbs per gallon or 85,000 lbs.  I use our pool as a heat storage tank, it is heated via geothermal (off peak only) and then the pool water is circulated via pex under the living space.  The pool is 16,000 gallons and has 2 inches of insulation under and around the 1 foot thick shell.  So I am holding approx 10 tons of heating per degree F in the pool.

Good points. Thanks to all for taking the time (and no one questioned my sanity) Some thoughts of mine: Why keep it above 40 - I would like the ability to come and go during the winter - the thermal mass of concrete slab flloors may work against me in this case as it may take a whole w/e to warm them up and then i leave. I've had a house on this island (an 1892 farmhouse) for 40 years and letting it freeze is hard on finishes, furniture, canned goods etc. It is nicer to walk into a place that isn't 10 degrees. I'm thinking of this water system is a sort of back up to get through the sunless cold snaps and that the majority of the time the house woudl stay above freezing simply because the average daily temp is above freezer (plus thermal flywheel, passive solar gain etc. I would have some sort of thermal shutters on the north windows and of course no ventilation load. Is there any sort of data on how much recoverable heat is in the ground (say 5' down) on a per square foot basis? I realize that 10,000 gallons = 85,000# = about 4.25 Tons per degree F and that I don't have many degrees to play with - I guess my question is if I "drew off" two degrees or about 8 tons, how would I figure how long it would take to recover that? Again thanks for the comments....

Oh yeah - Alton - I would not be using concrete for this vault. The system is pretty cool - dig a flatbottom hole - lay out a sheet of EPDM membrane (basicly pond liner) - then stack modules on top that look sort of like plastic milk crates (but designed for this purpose) - then wrap the membrane up and over the "crates" , insulate, backfill and done! This system is good for H20 loading so it can support trucks driving over it. It can be configured in any form depending on how you stack the crates (broad and shallow, deep and tall). The bottom can be insulated or not with no real worries about settlement cracks. The only limit would be the depth of the vault and compressive strength of the insulation.

Birdman -

I think your idea will work great.

Here's an additional low cost feature that will help it work even better. It's an idea that is sort of adapted from PAHS:

In summer, circulate the water through the floors, and you could cool the house cheaply to 68F which feels great in the summer. By the end of summer, your tank is 68F, and the ground around the tank will approach 68F instead of its normal 55F. It's like charging a big battery.

If the home doesn't need enough cooling to get the tank all the way to 68F, then a loop to an outdoor fan coil could do the same thing. Turn this fan coil on whenever the outdoor temp is above 75F, and you can charge the thermal battery. Be sure to drain that loop come October.

I've been chewing on a related idea for "air-to-air heat pump geothermal"

The problem I see is the tank size. To really do this you would need a LOT of storage, ball park in the 150,000- 200,000 gallon range, that’s big, like Olympic swimming pool big. Even doing the same thing with the ground doesn't seem to work out as well, the ground just wants to be at the temperature it is and slowly swings back there no matter how much you push it one way or the other. Granted it holds the temps for a while, but so far I have seen our ground temps from the field vary from 45F in winter to 60F in summer, oddly enough this almost exactly matches what the incoming city water is. Even when we ran our heat constant for 2 weeks and got the ground to 30F, in two weeks with normal run times it was back to 40F, and that’s a lot of thermal mass.

Honestly I think the fact that the ground wants to maintain it temperature is what we count on for geothermal to work the way it does. If it did hold heat from summer or cold from winter better it wouldn’t work as well as it does.