Just a thought;

I've got one of those model sterling engines that you can run on a spirit flame.

Maybe something like this (but bigger) chould use the heat of the furnace to drive the fan? So you don't lose heating if the power goes down.

As it turns out, somebody in Canada already makes these things. See the mealtimes website.

I originally found it on a Dutch website, slimspul (which translates as "smart stuff"). The text is in my native Dutch, but it has some interesting pictures.

It is a non-trivial amount of money, to be sure. And it would be worthwhile to check if there are cheaper ones available. But if I were living in a remote area like you are, I'd say;

• heat driven fan: $300
• heating in winter even when the power goes down: priceless

Or make a system with natural circulation, like the Roman hypocaust. Even better, a natural circulation system with water as a heat transfer medium, because water is much more efficient than air at transferring heat. My sister lives in a 1920s house that originally had such a system. Maybe its modern-day equivalent like a boiler stove would work for you.

Another alternative you could look at is a thermoelectric generator to drive the fan, like the ones from tegproducts. That might even also allow you to charge batteries (laptop?).

The problem with solar panels is that they work worst when you need them most; in the winter.

If a solar panel is rated at e.g. 30 Watts, that is usually under an illumination of 1000 W/m2!

• In the summer,
• on a clear day,
• with the panel pointing directly at the sun,

you'll get around 950 W/m2 tops.

If any of these conditions is not met, power drops rather sharply. Consider e.g the geometry (projected surface) if the panels are at 45 degrees to the sun... And realize that the average solar radiation in the US is 200 W/m2! (perpendicular to the sun's rays)

(As an aside, I have some practical experience with this stuff. For my graduation project from the Polytechnic back in 1993, I built a single-person solar powerd "car" (without batteries). With six 0,5 m2 solar panels on top, it reached around 30 kph on a sunny day. It had three low-resistance bicycle wheels. I did a lot of testing wich solar panels and their power, and was not particularly impressed.)

Current high-efficiency solar cells like the BP Solar SX_170B will give a maximum of 170 W (under the aforementioned conditions), for a panel of 1.59 x 0.79 meters. That is around 135 W max. per m2. On average you will be lucky to get a fifth of that. Can you work a heater fan with 25 W? This panel will set you back a cool grand! And as for any solar panel, you need some electronics in the form of a maximum power point tracker and possibly a converter.

IMO a $300 heat-driven fan starts to look better and better.

(note that I'm using SI units here. I'm not used to thinking in horsepower and BTUs and what have you)

The power rating for a solar panel is usually the Watt-peak rating. This is at a solar irradiatiation under good circumstances, like a fine sunny day, and with the panel pointing directly at the sun, about 1000 W/m2. The panel you mentioned is about 1/3 of a m2. So under optimal conditions it would get 330 W of solar radiation. If it produces 15 Wp of electricity, that would mean an efficiency of 4.5%. That sounds a bit low for an amorphous silicon cell. It should be about 9%. But there could be several explanations for that. The page is unfortunately remarkably free of specific data, though.

Assuming an average irradiation of 200 W/m2 for the US (found that somewhere on the 'net), and assuming a linear relation between irradiation and power output (which is probably optimistic), a 15 W(peak) panel would give an average of 3 W. Let's assume you get 8 hours of sunshine a day on average. One of those panels then produces 24 Wh (383600/1000 = 86.4 kJ) per day, or 8.76 kWh (31.5 MJ) per year. Now get the amount of kWh you used off last years electricity bill, and you know how many panels you need to live off the grid.

Living off the grid

If you use an average electrical power of 236 W, you'll use 2070 kWh of electric energy per year (how do I know? those were my figures for last year). So I'd need 2070/8.76 = 240-odd panels.

And then there are the batteries you'd need to store the energy for when it's dark or clouded. Suppose you want enough stored energy for five days. Using the same 236 W figure, you then need to store 235243600*5/1e6 = 102 MJ of energy. Look at the batteries article on Wikipedia, and look for energy density for rechargable batteries. A standard lead/acid (car) battery delivers 0.14 MJ/kg. for 102 MJ, you'd need 102/0.14 = 714 kg (1570 pounds) of those! A NiMH does better at 0.36 MJ/kg, but you'd still need 278 kg (600 pounds) of them.

Yet another factor if power. Your energy storage needs to be able to be able to supply the machine that uses the greatest amount of power. E.g. a microwave will use about 1-1.5 kW. An electric furnace much more, I suspect. And there are limits on how fast you can discharge a battery, depending on the type; too fast a discharge can damage them.

Just powering the blower

If the fan motor is, say 50 W, and you run it for ten yours per day, that's 0.5 kWh or 1800 kJ per day. You'ld still need 21 solar panels to generate that, and 120 pounds of batteries to store 5 days worth.

Closing remarks

Have a look at the solar cell article on wikipedia. It gives a good insight into the working of solar cells, and provides a lot of interesting links.

As you can see, there are a lot of factors you need to take into account when dimensioning a solar system. And the initial investment is not the end of the costs. Batteries will also need replacing every few years. The panels themselves should last a decade or two, although their efficiency will drop slightly over the years. But contacts and wires can erode, and electronics (mpp-tracker, charger, inverter) may need to be replaced. A dry storage space for batteries would be nice. etc.

My holidays are over, I'm back to $WORK now, so I won't be reading these comments every day. If you have questions, you can e-mail me at rsmith@xs4all.nl.

The above mirrors my impressions when I tried to find a portable panel that could power a small laptop. There seems to be much use of confusing numbers to make it sound like a panel might work when it's way underspec. When it does work, you might get a few hours per day if you're lucky.

You have an almost ready-made energy storage system though. Your water tank is a great way to store gleaned energy for later, just need to convert that water pressure back to electricity. And if the one tank isn't enough, add more. Or even, eventually, small ponds higher up could do. To water the garden too, of course.

Your creek doesn't have much elevation and is far enough from the house that power transmission could be a problem, but at least it's strong and steady and doesn't dry, so I think it could run some small hydro too.

The turbine here generates 2 amps @ 12v, which is exactly enough to run a small laptop: http://www.otherpower.com/otherpower_hydro.html

It seems my numbers were a bit off.

On the fast facts page of the solarbuzz website, I found the following info. A solar system of 1 kW peak power will deliver;

• 1600 kWh in a sunny climate
• 750 kWh in a cloudy climate

So my flat (definitely cloudy climate) would need an installation with 3 kW peak power to be self-sufficient in electricity. That is 200 of those $99 modules. That's $20000 just for the panels. Again according to solarbuzz, the total cost will be about twice that.

You know, I was actually thinking about water in exactly this way a few days ago! I don't know much about hydraulic ram pumps, but I was pondering whether I could use one of them to pump water up to the top of the hill using the power of the creek, then harness the water falling back down off our plateau. More things to research!

You're probably right that hydro is better than solar for us. Reading back over those reviews on that solar panel page, it sounds like nearly all of the folks are just using their system on the weekends or less often, so the panel probably has time to charge the battery banks all week while they're away.

There is this show on the National Geographic channel called "planet mechanics". It deals with a pair of guys thinking up and implementing aleternative energy solutions for people. If you haven't seen it, I think you'd like that show.

In one of the first episodes called "lake district dilemma" they have to find power for a seriously remote youth hostel in a valley in the lake district in England. They find a stream uphill, and use firehoses to pipe the water down and drive a water turbine. When they open the tap for a test, the two of them can barely hold on to the hose!

But to get significant power with a turbine, you need either a significant height difference, a lot of water flow, and preferably both.

Why not go for something simple like an undershot water wheel? It does not require a huge height difference. See e.g the collse watermolen (that is Dutch for "water mill at Coll", link is in Dutch as well, but has pictures) near Eindhoven in the Netherlands. This mill has been milling grain and oil in its current form since 1681, after the previous mill from at least 1337 burned down. I'd call that quite mature technology.

Take a waterwheel, let it drive some gears to pull up the rotation speed, and tack on a generator.