One of my most intriguing discoveries as a geeky kid was to learn (from the Solar Energy book I was given as a birthday present) that you can run a refrigerator on heat instead of electricity!  As today is the first time in about a month that we have had sunshine here -- the darkest June on record -- and the temperature and humidity are rising, my cerebrations at this time of summer solstice return to this idea of solar-powered air conditioning.

Solar electricity to power an AC?

It's easy to imagine how you could run a conventional AC with photovoltaic panels generating electricity, but that's a pretty costly proposition.  Consider that a room (window) AC requires 10-15 amperes from a wall outlet.  Multiply those figures by 120V and that's about 1200 to 1800 watts.  According to Solarbuzz.com, photovoltaic panels cost averages $4.75/watt, not including incentives (or so I presume).  So your solar panels to run that room AC would cost:

1200 watts x $4.75/watt =$5700

1800 watts x $4.75/watt =$8550

Yikes! And those figures don't include installation or the necessary inverter electonics to create AC, or any means of storing electricity when the sun gets lower in the sky.  But let's be generous towards solar power and assume you're just adding photovoltaic panels to a system already in place, and doing the work yourself.

What if you had $5700 lying around, how long would the payback take?  Average electricity cost in the US is about 12 cents per kilowatt hour.  Let's assume the AC runs about 8 hours a day for three months of the year, with the thermostat cycling the AC on and off for equal amounts of time (called a 50% "duty cycle").  The cost of buying that electricity is:

1200 watts x 50% duty cycle x 8 hours/day x 1kilowatt/1000 watts x $.12/kilowatt-hour = 57.6 cents / day of operation

If the AC runs 90 days out of the year, the cost is $52 per year.  The payback period for the extra photovoltaic cells?  For the smaller AC (1200 W), that's $5700/($52/yr) or about 110 years!  If you have to pay for installation (generally about equal to the photovoltaic panel cost), the payback takes twice as long.

No cost win there.

Using heat to cool?

Using heat to create cool sounds totally wack at first blush.  You can sort of see, however, how it might work.  It takes energy to run an air conditioner, and heat is a form of energy.  If you have a camper or cabin that's "off the grid," you may be familiar with refrigerators that run by burning propane.  (Alternatively, you might be Amish, some of whom eschew electricity and buy gas-powered fridges, but if so, you're not reading this blog, are you?)  These fridges run on about $.50 of propane a day, and don't chill exceedingly well in general, but enough to keep your excellent farm-made butter from melting.

You can imagine how this principle, whatever it is, could extend to air conditioning as well, using heat from the sun.  With AC, however, you're trying to cool at least an entire room, or about 1000 cubic feet.  A camp fridge is about 2.5 to 12 cubic feet.

The problem is this: propane flames are pretty hot -- 'way more hot than a sun-absorbent collector: a black plastic car seat, for instance, that has been sitting in the sun will not set your shorts on fire the way a propane torch would.  (Do not try this at home.)  The only way to get solar heat hot enough to substitute directly for propane is to make some sort of concentrator, like a magnifying lens or reflector.  A magnifying glass in the sun can set your shorts on fire.  And you can similarly concentrate solar energy with a silvery curved surface, like a satellite TV dish covered in aluminum foil.  (Line the bucket seats of your convertible with aluminum foil at your own risk.)

Using a solar concentrator generally means having to move the device to track the sun; and then you have to put tubing in the focal point where the sun's rays converge.

So... is solar air conditioning impossible? Find out in my next exciting post.