Solar power the world! Wait...

September 27, 2014 12:17 PM

My father is talking to me about how a tiny patch of the Sahara could supply all the power needed by the whole world using photovoltaics. But how would it get from Africa to, for example, New York? Wouldn't undersea cable be inefficient? He tells me "superconductors" which as I understand them are not exactly a viable thing. Is solar going to be more of a per-home deal where I power my house with my personal solar plant? I don't know anything about this so please talk to me like I'm a doofus.
posted by Sternmeyer to Technology (15 answers total)

Best answer: If you'd like a good rundown of desert vs. urban solar power generation, I recommend the articles by Chris Clarke on KCET. He's an advocate against desert solar installation for various reasons (including habitat and ecological preservation) and for urban/rooftop solar, so there's a definite bias in the reporting.
posted by carsonb at 12:28 PM on September 27, 2014 [1 favorite]

He's talking science fiction, not science fact. There's always a loss when you transmit electricity by wire.

As of 1980, the longest cost-effective distance for direct-current transmission was determined to be 7,000 km (4,300 mi). For alternating current it was 4,000 km (2,500 mi), though all transmission lines in use today are substantially shorter than this.
posted by Cool Papa Bell at 12:34 PM on September 27, 2014

Here's a 574MW superconductor that's not science fiction, but then again, not in practical deployment, either.
posted by ambrosen at 12:56 PM on September 27, 2014

This seems obvious, but you did say please talk to me like I'm a doofus:

Your father might be wrong. Just because he thinks this is possible does not mean this is a viable idea. Fathers are not automatically always right about everything.

Yes, even if he's a professor or engineer. Really.

You will believe this because I'm from the internet.
posted by yohko at 1:55 PM on September 27, 2014 [2 favorites]

Here's an overview of the project which your dad's most likely referring to. It's pretty ambitious, to say the least. Not much mention of the U.S., either.
posted by ambrosen at 2:23 PM on September 27, 2014

There is a fair amount of discussion in Europe about the economic inefficiency of building solar anywhere other than in southern Europe and wind anywhere other than in western Europe, but people keep building it in other places so that their national subsidies stay within their borders rather than go elsewhere even if it meant they paid less per unit of renewable energy generated.

There is also a fair amount if discussion about putting solar in northern Africa and wheeling it into Europe, on the grounds that the reduced deployment costs and higher output per unit installed capacity would balance the increase in transmission costs (large scale installations within Europe cost about one third less than rooftop installations per unit of energy generated, I suspect that this would be a bit better in northern Africa due to labour costs and improved performance). There is also some talk of having a really high voltage line connecting north Africa all the way into Europe to facilitate just this sort of project. Still some way off though.
posted by biffa at 2:32 PM on September 27, 2014

To concur with ambrosen, this is in no way going to connect transmission from Africa or Europe to the US. But why would you, there's desert in the US or I dare say similar cost advantages with links to Mexico.
posted by biffa at 2:34 PM on September 27, 2014

There is an energy loss from pushing electrons through a wire. This often happens in the form of heat. Sometimes it's useful, like the heating element in your kettle. Other times, it's not, like when your extension cable overheats because you've plugged too many high power items into it.

There is a LOT of solar energy hitting deserts around the equator, and relatively few trees to shade the solar collectors. Getting that energy from the point of collection to somewhere else where it can be used is the difficult bit. The further you go from the collection point, the more losses you incur. Also, there's the laws of physics to contend with. One of them states that there is an energy loss every time you convert energy from one form to another.

I haven't run the math, but I doubt that a tiny solar array in Africa can power the world. It could conceivably power a portion of Africa, as electrical power usage in Africa is so low. With the right technology, it's theoretically possible that an African solar array could power some homes in the US. We're a long way off from that right at the moment, though.

It's far more efficient to stick some solar panels on your roof. Using less wire means less energy loss and is also cheaper. If you don't live on the equator, there's the aspect of getting less energy from the available sunlight, but you'll likely wind up with more energy from a 100w panel on your roof that you will from a 100w panel in Africa.
posted by Solomon at 2:44 PM on September 27, 2014 [1 favorite]

What happens when the Sun has set in Africa?
Then it is not powering the world, so pedantically, no.

If one were to divert the entire planetary production of aluminum for 2 years, a pair of planet girdling cables, each with a cross sectional area of 0.83 square meters could be constructed.

Using a bipolar 800 kilovolt DC transmission scheme, the current average power requirements of the United States could be transmitted from one side of the planet to the other, with transmission losses running about 50%.

So no again: inefficient and difficult both.
posted by the Real Dan at 2:56 PM on September 27, 2014 [1 favorite]

Best answer: And to add one more source, here's a list of the longest transmission lines in the world. At 7.2GW, each of those lines can supply as much power as Ireland (population 4.6 million) or Nigeria (population 170 million) can generate, and at 2,000 km, they'd go from the Sahara to the UK.
posted by ambrosen at 2:59 PM on September 27, 2014 [1 favorite]

It sounds like your father has taken an interesting thought experiment too far.

In broad strokes though, a surprisingly small land area covered by existing PV panels could deliver a significant portion of the US (or world) energy budget. In round numbers 100x100 miles of panels in the southwestern US might be enough to do the job in the US.

Of course, the thing is, you would never want to do that. You'd want to site capacity closer to demand to reduce transmission losses and infrastructure upgrade costs (including crazy stuff like massive superconducting transoceanic cables), and you wouldn't want to shade a huge patch of wilderness, not when you have lots of rooftops, and abandoned industrial land, etc. You'd also want to spread it around to avoid exposure to disasters and routine weather.

The other thing is that, while desert areas do get more sun through an average year than other areas, the difference in annual insolation between someplace like Tucson and somewhere like Seattle is only 2x. Which isn't insignificant, but its not enough to render PV uneconomical outside of desert areas.
posted by Good Brain at 4:40 PM on September 27, 2014 [2 favorites]

The closest point in the US to North Africa is 5000 KM away (source)

A 130 MW HVDC cable has an 800 mm^2 cross section (source)

Underwater HVDC only requires one conductor, not two, since seawater is conductive.

The US consumes 4 billion kwh/year of electricity (source). That's 500 gigawatts, or 3846 of the 130 MW cables.

Copper weighs 1600 kg per m^3.

3846 * 5000 km * 800 mm^2 * (1600 kg/(m^3)) = 25 billion kg of copper.

The world produces 15.1 megatons = 151 billion kg of copper per year (source), so it would be physically possible to buy that copper (especially spread out over many decades).

Copper costs $6700/ton (source), so this project would cost $170 billion in copper alone. Actually that's surprisingly little, the USA has completed many projects of that scale (such as minor wars, fighter jet purchases, corporate subsidies). That's around 20% of US spending on electricity for a single year.

So yes, I would conclude that if the USA had a really, really compelling reason to import all its electricity from Africa, it would be possible.

More realistically, imagine a future where northern Europe is already connected to the Mediterranean via a European grid. It may prove practical to build a much shorter cable and import small amounts (a few hundred MW) of electricity to the US to even out dips in the US grid, and vice versa. In a world where most of our energy is from renewable sources, that could really help reduce usage of temporary fossil-fuel-based generating capacity to fill in the dips. Of course it will be 50-100 years before we get to that point.
posted by miyabo at 6:41 PM on September 27, 2014

Miyabo, that's 15.1 billion kilograms of copper produced per year, not 151 billion kilograms.

Spank your slide rule :-)
posted by the Real Dan at 7:35 PM on September 27, 2014 [1 favorite]

When I saw that infographic come around on Facebook, one thing that struck me was that it only looked like a small area because we were looking at a map of the whole African continent. Actually making a farm of solar arrays large enough to be indicated by anything larger than a dot would be an incredible undertaking, and this area was distinctly larger than a dot.
I didn't follow up on the details, so I'm surprised to hear that it's not just a thought-experiment, "it would only take this tiny percentage land area omg wow!" but actually a proposal.
posted by aimedwander at 5:15 AM on September 28, 2014

Coming back to this, I see people pointing out his this *might* work. But what really separates science fiction from science fact is basic efficiency, economics and politics.

He's talking about an enormously expensive, complicated project, spanning international boundaries, with inefficient technology.

And this will take place in a world where a gallon of gas, available for purchase on every street corner, costs about three bucks before taxes and delivers 114,000 BTUs. And that price is already falling due to increased efficiencies in local solar, local wind, and better extraction tech.

Your father's idea will never work as he's describing. Never ever.
posted by Cool Papa Bell at 10:42 AM on September 28, 2014