Electricity is expensive. It takes a lot of capital to build the ifnrastructure required to produce and distribute it, a lot of fuel to produce it, and a lot of time and (personal) energy to regulat, bill, and pay for it.
It's also the most useful form of energy we've got; and our modern wourld could not exist without it.
Over the years; there have been a hell of alot of attempts at ways to replace centralized fossil fuel, or nuclear electrical generation; but none of them are really practical on a large scale, because of their general costs and inefficiencies.
The most promising technology for providing off grid power for individuals is fuel cells; but they still require a uitlity connection (either startup electricity for the water cracker, or a natural gas line).
If you want to be off the grid entirely; without relying on fuels which either runout, or require a centralized distribution system; you rbest option is solar.
The problem again though, is cost, and efficiency.
Up 'til today, the general run of photovoltaic cells (solar cells), has been about 6-8% efficient; with some super efficient, super expensive cells going to 12-14 or even 16-18%, and some ourageously expensive (100 times the cost of the 8% efficient cells), miracle, nearly unproducable ones all the way up at 24% (or even 30%, but those are pretty much only used by NASA for sattelites and the like, and they're even MORE expensive than the 24% efficient cells).
What that means, is for every 100 watts of sunlight it recieved (between 125 and 375 watts per sq meter average during daylight hours over a day; up to 1000watt per sq m at noon in high summer), a photovoltaic cell would produce between 8 and 30 watts.
Now, 24% efficiency isn't too bad. That's about the same energy efficiency the best internal combustion engines can do (most are below 15%); but it isn't enough to make converting to solar energy on a large scale very practical.
The breakeven point for solar (presuming a resonable cost per cell, and a reasonable cell lifetime), is generally reckoned at around 40% efficiency.
In practical terms, what's that mean?
Well, a typical household will use about 9000 kilowatt hours per year; and a super efficient house can (in moderate climates), get down to as low as about 1000 kilowatt hour. Without too much work, a typical family in a temperate climate could reduce their consumption down to 3000-5000 kilowatt hours for example. Obviously Alaska and Arizona are the outliers here; and older homes and appliances in general arent very efficient.
Okay so like I was saying, the average power use in the US is 9000 kilowatt hours; and the average power bill according to several net sources, in the US is about $3600 a year... hich seems awfully high to me, because the average cost of electricity is about 6 cents a killowatt hour, with fees and charges and taxes usually about doubling that. That cost may include heating, cooling, cooking, and general electric (so gas, oil, and electric) though.
With an 8% efficient cell, at 125watts per square meter, you get about 10 watts per square meter; for on average 220 days per year; and an average of 10 hours per day. That gives you 22 kilowatt hours per sq meter, per year.
These are worst case numbers by the way, because you don't plan power consumption on anything other than worst case; unless you feel like brownouts.
Anyway, with that low a power output; we're talking about a household needing 410sq meters of cells to meet their power needs; and that's assuming a uniform output, and a uniform demand, with a 100% efficient power storage system; none of which are actually possible.
Now, my house is about 52ft long, and 56 ft wide, in an L shape; so about 25% of that area is part of my driveway. It's a fairly typical house, with about 1800 sq ft of usable space; but I've got a porch, a covered parking area etc... so my total roof sq footage is probably about 2500 sq ft. 1sq ft is 0.0929 sq meters; so my roof has about 232sq meters of surface area.
So, to meet our entire energy needs with solar; we'd need to have a minimum of about twice our roof area.
Oh, and did I mention that it would cost at least $250 per sq. meter (that's including tax credits and environmental rebates etc... currently a 1sq meter panel can run as high as $2400 for the 18% efficient cells, and at a minimum run about $500), with an expected lifespan of about 25 years; and adding in an additional 20% for the wiring and battery cost. Would you like to spend $125,000 on your electric bill over 25 years? That's about $420 a month. Now I dunno about you, but my electric bill is a about $3000 a year.
Even if we best case it to 375 watts, and 300 sunny days a year for an average 10 hours per day (which even in Arizona is a bit unrealistic), and we presume a 3000 kwh per year home (which would be difficult given AC costs here, but with really good insulation could be achievable); we're still looking 90kwh per sq meter, and looking at needing 40sq meters of cells.
Let's be totally unscientific and split the difference. Say we need the whole 232sq meters of my roof; and we manage to get the costs downs to the $250 per sq meter I mentioned; then add about 20% for the associated hardware and storage costs (like huge battery and inverter banks). That's still about $70,000 for my house, and amortized over 25 years, it's just about breakeven; but agian that assumes a uniform demand, 100% efficient storage, and the ability to completely cover my roof with solar cells.
Ok, so let's assume we shell out the bucks for a much more expensive, and much more efficient cell. GE has a cell that they claim will put out 110 watts per sq meter on a 1000 watt day (about 11% efficient); it costs appx $600 per sq meter in bulk, and is warranted for 35 years.
Presuming again, a 100% efficient storage system, an even load, 300 completely sunny days a year for 10 hours a day (all completely pie in the sky); and that average of 375 watts per hour, thats 124kwh per year per meter; so we'd need about 73sq meters of cells. Again, let's split the difference between the minimum and maximum efficiency and say we really need 150sq meters at $800 per meter; and lets assume 20% additional cost for the wiring and battery network; for a total of about $145,000, or about $4,000 per year.
Still, a lot more up front cost than almost anyone would be willing to pay.
Today, the DOE has announced they have a new 40.7% efficient solar cell; at what they call a cost of $3 per watt (assuming a 1000w/sqm solstice noon). That's 407 watts per sq meter, or a cost of $1220 per sq meter.
Ok, so what does that do to our numbers?
Assuming the 375 watt maximum average, over 300 sunny days 10 hours per day, and 40.7% efficiency; to meet our 9000 kwh households electrical needs, we would only need 20sq meters of cells, at $1220 per meter; for a total of $24,000 without the supporting infrastructure.
Even assuming the worst case, of 125 watts per hour average, over 200 sunny days, 10 hours per day; you only need about 88sqm of cells, for about $110,000.
Again, we split the difference, and get about 55sqm, for about $67,000. Adding in 40% (because the cell cost is so much less than other designs) for the cost of the supporting infrastructure, and we still get $95,000. Amortized over 35 years, thats about $2700 per year; or about $225 a month.
That's quite a lot less than I pay for power right now; and I live in one of those areas where you really do get 300 plus watts of sunlight average; and I would only need to cover between 1/5 and 1/3 of my roof.
If we went and got all efficient about things with appliances and insulation, we could cut that down to 3000kwh per year, for between 10, and 30 sqm of cells (plus about $20k for the infrastructure) and between about $35,000 and $60,000; again amortized over 35 years, for $1000 to $1700 a year in power costs; or $85 to $140 a month for power.
And that folks; is when people start ponying up the extra on the front end. 15 or 20 percent more up front cost, for a house that will never have a power bill? Yeah I'd do that.