I was trying to get an upper limit on the amount of solar power that could be generated if all rooftops were covered with 15% efficient solar panels.
Can we count captured heat directed toward water heaters and such? This might turn out being more valuable then photovoltaics alone.
Well, I was trying to figure out the upper limit on rooftop electrical power.
In that case, I have no idea.
Just kidding. I guess the first step would be to try to calculate all the rooftop space in, you didn’t specify what region, but I’ll assume the continental United States. A very high percentage right off the bat can be eliminated because the amount of sunlight they get isn’t enough to pay for the energy used to produce and install the solar panels in the first place.
Personally I think a superconducting grid is your better bet, so you can set up oodles of solar panels in places where the payoff is highest and transmit this energy without loss to the cities.
If you must have a completely ballparkian figure, take the southern half of the U.S. and compute how much solar energy falls on it, take a guesstimated fraction, and then 15% of that. It’ll probably turn out to be some massive number that could indeed satisfy all American energy needs if it could be distributed properly.
Mods, I meant to put this in GQ. Sorry.
As I said, I was looking for an upper limit. One problem is that I don’t know how much rooftop area there is.
If only we could pave roads with solar cells. 
Guesstimate the fraction of people in the US that live in houses (suburbs and rural, city folks dont count). Figure the fraction that live in the south or southwest. Divide by 3 to get the number of non city houses. Assume the roof has about 1000square feet. Assume about a factor sunlight of 5 hours a day. Take your efficiency. Go from there.
That doesn’t factor in non-residential roofs.
Double it then.
I live in a two-story 2,400 square foot house in So Cal. I am putting in a system right now, and it will use from 1/2- 2/3 of my optimal roof area (the stuff facing the right direction). To use more would require shifting some exhaust vents, running more cable, etc.
Here are the raw numbers from my proposal to help you:
Proposed System Size
4.510 kW DC(STC)
3.891 kW AC(CEC)
Estimated Annual Production1 6,675 kWh
SolarCity Power System $28,811
Rebates Assigned to SolarCity
EPBB ($1.90 / AC_WATT) -$7,382
Out-of-Pocket Cost to Customer $21,429
Incentives Paid to Customer: 30% Federal Tax Credit (On Out-of-Pocket Cost) -$6,429
Net System Cost to Customer. Includes Sales Tax.
Excludes permit fees. 2
$15,001
I am getting about 1TW. Does that sound about right?
While we can’t do that, is there any way to put pressure sensitive plates in the road to use some of the energy from cars passing over it?
If you did that, you’d rob the energy from the cars. There are no free lunches.
As for the rooftops, it’s impossible to get a precise answer. But it’s rather easy to do some ballpark estimates.
Here’s some Census Data for you to use.
From the cite, we find that there are about 127 million ‘housing units’ in the U.S., of which 26% are in multi-unit structures.
That leaves us with about 96 million homes. Let’s call it 100 million for fun, since we’re just WAGing anyway. Let’s also guesstimate that each home has an average of 1,000 sq ft of rooftop. That gives us 100 billion square feet of domestic rooftop in the U.S.
A quick look on the internet shows solar panels available with output in the range of 10-15 Watts per square foot. Let’s use 15. So, solar panels covering every rooftop of every home in the country might generate 1.5 TW of power in full sunlight. Let’s assume you get 10 full sunlight equivalent hours per day. That’s 15 TWh per day, or about 5,475 TWh per year.
From these WAGs, we can start reducing downwards. Most roofs are peaked, and will not receive full sunlight all day (half will be in shadow). The figure for solar flux is the world average. The U.S. is far north of the equator. Reduce the amount accordingly. Not all the roof will be able to be covered - you’ll see space to walk around, to run pipes and cables, etc. And you could go on. But let’s work with this ‘best case’ number and see what that could theoretically get us:
In 2005, the U.S. consumed 29000 TWh of power. So the number above equals about 18% of U.S. power consumption.
Now, adding in all the reductions above, you’d be very, very lucky to get even a quarter of that. Peak daylight is probably more like 5 hours per day. Because of the northern latitudes of American homes, annual solar flux is probably around half of what it is at the equator. Then you add in the cloudy days. Next, you have the problem of storing all the energy, and the problem of many of the homes not being near where the majority of power consumption needs to be.
So, my WAG is that if you put solar cells on every rooftop in America, you’d be lucky to replace 5% of the U.S.'s power consumption, and probably even lucky to get to 1-2%.
Who says we can’t do that?
That is very interesting.
I’d love to see some research on it in terms of how well it functions as a road surface in terms of control of vehicles, and how long lastign it would be with even large trucks rolling over these panels.
Wrong. Laughably wrong.
It may be that in your country, the only buildings with a roof are single family homes, but in the US, we have roofs on businesses, housing duplexes, townhouses, storage facilities… in fact, damn near every building we make down here has a roof.
They say 21 years before needing to be refurbished. Although they don’t directly address weight capacity on their website, they do state
Well, la-dee-daaaaa…
The idea behind the question, I think is “let’s put solar on every roof! That way we eliminate global warming, etc. etc. etc.”
The reason solar cells aren’t on every roof is that for one thing they are very expensive. It’s not lack of space that limits their use.
The other reason is that solar is intermittent. What do you do when it’s cloudy or at night? Coal, nuclear, etc. can be made to give electricity on demand 24 hours a day. Usually baseload demand is predictable but sunlight on any given day is not. That necessitates backup power. Currently solar is used so sparingly that backup power isn’t an issue. But the point is that solar can’t supply a huge portion of electricity without backup power…and there has to be facilities to provide this. Those backup generating plants are expensive. Much of the cost of coal electricity is not the coal, it’s the physical plant.
Solar because it’s not reliable (you don’t know how sunny it will be on any day) and can’t be ramped up as natural gas or coal can.
Without subsidies, solar cells probably wouldn’t exist at all except in niche applications. I know, I know, coal and nuclear are subsidized too, you just don’t see it as easily.