HOW LOW CAN WE GO?

When we first started thinking about designing a new house some eight years ago, I said I wanted a passive solar design and a house that was so well insulated that we could heat it with a candle.

Years of study have taught me that while a passive solar greenhouse may be possible in Wisconsin, the nights and cloudy days would be too chilly for living spaces.  But we have come up with a plan that we hope will offset 50%  to perhaps 70% of our heating needs through a combination  of passive solar design and the addition of solar hot water panels, along with heat storage in concrete and flagstone floors and an interior rock wall to collect and store the heat that we generate  And of course, some good insulation.

For the periods of cold, cloudy weather, we will have backup heat from a small wood-burning stove on the main floor and propane to add heat when necessary to the radiant floor heat system.

SAND BED HEAT STORAGE?

We all know how hot sand can get in the sun. (photo credit: http://www.flickr.com/photos/stevephillips/242665348/ )

We were planning to supplement that heat storage with a sand bed under the house that could collect heat starting in late summer and release it slowly during the winter.  We got the idea from Mark Morgan  a couple of years ago, when we were working at a straw bale workshop he led, and it’s been part of our general plan ever since.  It seemed like an intrinsically good idea to store up heat ahead of time and use it like a battery to provide it later on when needed.  We planned to rest the roughly 900 square feet of our house that will sit on the ground on a two foot deep bed of sand, with a good layer of insulation at its base.

As with a lot of good ideas, the devil is in the details, and now that we have talked with five  area solar heating  installers, we are going to skip the sand bed.

WHY NOT?

Bob Ramlow is a well-known proponent of sand beds.  He has written a book, Solar Water Heating, with Benjamin Nusz. He also wrote an article for the magazine Solar Today, Nov-Dec 2007, Warm, Radiant Comfort in the Sand.  In that article he suggests putting a 2-foot deep bed of sand under your building.  He says, “In my Wisconsin climate the solar energy system begins to heat the sand bed in early to mid-August….It takes about a month to get the sand bed saturated with heat, and then the temperature inside the building is regulated by judicious opening and closing of windows.

Anyone who considers this should be aware of is that if this is the only form of  in-floor heat you will be able to use for that part of your house,  This is a good news/bad news scenario.  Sand beds are slow to collect their heat charge and give it off slowly and as the winter progresses, it will gradually loose its heat.  Then there is a cool mass under your floor.

Some people attempt to get around this with 2 loops of tubing.  One right under the floor to provide quicker heat, and one further down.  This is not a perfect solution because even the upper loop will radiate its heat both up and down.  It’s not like air convection where  heat rises and cold air falls.  Yes there is air in sand, but the idea is that you are trying to store heat by radiating it into the sand particles, so the depth of the bed will slow down the rate at which your room will warm up even from the upper loop.

A sand bed is exceptionally unresponsive.  You can’t crank up the thermostat and get more heat later today.  There will be a lag time of many days before the heat soaked up on a sunny day is available for use in the house.  It might take most of a week – or longer.  After all, it took a month or more to fully heat the sand bed during the time of year when days are long and sun is strong.

Pex Tubing - how that hot water moves through sand and concrete. (photo credit http://www.flickr.com/photos/moosicorn/4546866153/ PEX TUBING )

As we have talked to people who are aware of sand bed installations, most note  that there will be a string of days in the winter – maybe more than one string – that are cloudy and cold during which the heat in the sandbed will be almost fully discharged.  In the middle of the winter, you will not have enough sun to recharge it.  You are left either spending your heating energy reheating the sand bed or living with perpetually cold floors.  It seems like the deeper the bed – the worse this problem can be.  The cases we heard of where people thought two feet = good: four feet = better now have an even less responsive system.

If you want to try to keep a deep sand bed warm with solar panels, you will need a LOT of them.  Everything has its environmental cost, and building and transporting more panels than needed is part of the big picture.  You can’t use solar for the coldest, darkest part of the winter without a lot of waste in the system the rest of the year.

Want the nightmare scenario?  A hydronic heating installer friend told us of an uninsulated four foot deep sand bed under a house he knew of.  The radiative losses downward into the cool earth are immense, and the heat recharging possiblilites are virtually non-existent in the middle of winter – all done with the best of intentions.

So here’s  our proposed plan.  (The details are still being worked out with the solar installer.)

We will have in-floor heating, and we expect that the upstairs concrete slab and the lower level flagstones set in shallow sand will provide heat storage for a few days.  We will supplement with a wood stove for the main floor and a small propane burner that will heat the hot water tank when the sun can’t.

We have been told by several experienced installers that the only way to have any control over the temperature in the house is to keep the heat in the hot water tank and deploy it to a relatively  thin (2” – 5”), well insulated floor as needed.   Again, insulation is a key ingredient.  We’d like to put 4” of Styrofoam under the ground-level floors if we can afford to.  At the very least, we’ll put in 2” of insulation.

We think a well insulated, and not-too-deep concrete or flagstone-set-in-sand storage floor is the best solution.  We’ll update you on the actual solar panel array specifics with a later post, perhaps in January.