If We Only Had The Power.

Let’s suppose that we don’t like nuclear power. Fission power is non-renewable; even if you are breeding U-238 into Pu-239 you are only staving off the inevitable time when you run out. Uranium is essentially a fossil fuel, fossilised since a supernova explosion two billion years ago. What are our options with renewables?
The longest-used, and therefore easiest to use, renewables are hydropower and wind-power. Unfortunately, South Africa is a semi-desert country. Therefore, we don’t have many rivers. We are blessed with some fairly high mountains and plateaus, so where we do have rivers they fall fairly far before entering the sea, but very often these falls are steep. Steep falls usually mean that there is little room for reservoirs; you can build a weir to block the river, but it will be expensive to do so and the power will be erratic, depending on how much rain fell last week. (A reservoir smooths the flow out from month to month, or even year to year, depending on size.) There are other problems with dams, but the real problem from a power-generation perspective is that there aren’t enough of them, and also they don’t exist in places where people live and power is consumed.
Wind-power is a whole other story. There’s a lot of excitement about wind-power in the United States at the moment, which should make us a little suspicious (there have been periodic power fads in the Anglo-American world since World War II) but in principle it’s a simple and good idea: build a lot of windmills to catch the wind. There’s a lot of wind in South Africa. Global warming should produce even more wind, so it’s not going to go away.
There are two ways of dealing with wind, which may be summed up as bloody big windmills and nice little windmills. The bloody big windmills, with blades dozens (ultimately, perhaps hundreds) of metres long, are supposed to catch more wind, not only because they are bigger, but because they are lifted above the human activities which affect the wind (houses, trees etc). They also generate megawatts of power. They are ferociously expensive to construct and hideously costly to repair when they break, which they frequently do.
Nice little windmills are much smaller. They catch less wind, because they are down where human activities and even natural vegetation might affect them. They come in two varieties; horizontal-axle, where the whole windmill plus generator is set on top of a tower thirty or fifty metres tall, rotating according to the direction of the wind, and vertical-axle, where the windmill is closer to the ground and the generator is right at the ground, and where, because the windmill is the same from any direction, it doesn’t matter which direction the wind blows from. (Some of them look rather like inverted egg-whisks.) These windmills generate multiple kilowatts of power.
The Creator feels the need to endorse nice little windmills. Gigantism is all very well, as in our mighty new coal power station (if we’re going to shit in our own bed, let’s at least be proud that we’ve produced the biggest turds in the Southern Hemisphere). However, the advantage of the little ones is twofold; they can be manufactured and assembled with relatively low-skill labour and require little special technology, and they can be easily maintained if they break, since all you need is some scaffolding and a conventional long-arm crane. Bloody big windmills require imported components, high technology, highly skilled steeplejacks absorb the resources of the nation to mend if they go wrong.
Where one would need gigantism is in the manufacture of them. Let’s suppose a ten-kilowatt windmill. The Medupi power plant is supposed to generate 4.8 million kilowatts. In round figures we need a million of these little windmills to replace it. That’s a million generators, a million turbines, a million structures. A hundred people could doubtless erect one of these windmills in a week. For a million of them, those hundred people would take 20 000 years. Two million people could do the job in a year. It looks like the end of the unemployment crisis, because manufacturing a million windmills is going to take a fairly large factory. Several fairly large factories, in truth.
How much land would this take up? Spaced in a grid a hundred metres apart, a million windmills would take up ten thousand square kilometres, or about a third the size of Lesotho. It’s not enormous. Ten million windmills — enough for ten Medupis — would take up the area of the Free State. If you build bigger windmills capable of more than ten kilowatts, reduce the area. Notice that “take up” here doesn’t mean that you can’t do anything between the windmills; you can farm crops or graze animals or even have some low-lying housing provided that the locals don’t object to the noise of the windmills.
Now, that sounds wonderful if rather scary, but there’s a small problem: storing the power. The wind doesn’t blow all the time whereas you need electricity all the time. Same problem as the problem with hydropower. You have to build some additional plants to store the power; the most sensible being to build a water-tank at the bottom of a koppie and another at the top of the koppie with a pump and a water-turbine, a pumped-storage system. Since these windmills are going to be placed in hilly, largely uninhabited areas anyhow, this isn’t going to be a technical problem, but it does add to the cost and reduce the efficiency. All this dispersed electricity will have to be then plugged into the national power grid, which further adds to the cost (power pylons all over the place). Incidentally, you could put some of the windmills out at sea, where there are no human activities to obstruct the wind — although the corrosion problems are gigantic and the power-transmission problems immense. George Monbiot likes off-short windmills; good for him.
Well, great. If we spend a hundred billion rand a year over the next decade, we can probably get something like this going. We aren’t going to, of course (there are no plans for anything except importing a few big sexy-looking windmills for propaganda purposes). But is there anything else?
The others are solar thermal, solar electric, wave and ocean thermal.
With solar thermal you have a lot of great big mirrors. It’s an expanded version of the solar water-heater you have on your roof if you have the money. (If the government were smart it would build a vast factory to produce ten million free solar water-heaters for every building in the country, but if wishes were horses, the Creator would drive a coach and six.) The obvious problem here is that the sun moves around. Therefore the focus of the mirrors move. Therefore you either have to move the mirrors, or move the thing they are focussing on (usually a pipe containing some volatile liquid such as alcohol, water or sodium). This in its turn takes power to move it. However, these are design problems; there’s no reason why solar thermal shouldn’t be effective. Eventually you have a boiling liquid which you run through a turbine. Then you have to condense it (there are all manner of ways to capture the heat for usable purposes, rather than just blowing it up a steamy cooling-tower as we usually do) and re-use. Solar thermal can be of any size you fancy, but a hundred megawatts or so seems a plausible upper limit. Sunlight, conveniently, generates around a kilowatt per square metre, so a hundred megawatts would entail 100,000 square metres of mirrors, which doesn’t seem a huge area.
Unfortunately, it isn’t so simple. The wastage of energy is considerable; mirrors don’t reflect all the power, the liquid wouldn’t absorb all the energy, turbines are not totally efficient — most likely you’d need more like a million square metres. A square kilometre, probably scattered across two or three square kilometres of territory. For our national 100,000 megawatts, we need 1000 of these solar thermal stations. That takes up only 3,000 square kilometres, not much bigger than Johannesburg. We could do it, area-wise..
Whereas the wind doesn’t blow all the time, at night the sun doesn’t shine at all. By day, you sometimes have clouds. Therefore, once again, you need power storage systems. That would take up more space, and cost, too. And, of course, the above-cited figures are wildly overoptimistic. However, again it’s something which could be done, which does not require any advanced technology, and which would lead to the end of unemployment as hundreds of thousands of people cemented glass-aluminium mirrors to moving frames, welded pipes or worked in factories producing turbines and condensers.
Solar electric is the big easy. You stick the beautiful blue-black cells on a frame, and when the sun comes up electricity comes out. All problems solved! Just by wearing a hat covered with solar cells you can solar-power your environment! It’s perfect!
Well, not exactly. There are several problems with solar electric. Solar cells, like computer chips, are made of silicon. Silicon is extremely energy-intensive to make; you have to first remove the oxygen from silicon dioxide (aka sand), which is hard because silicon loves to bond with oxygen and then crystallise and extrude the silicon in a billet (which is violently energy-intensive; silicon melts at high temperature) and this has to be done in a vacuum or an inert atmosphere with super-clean graphite tools (because any dirt or impurity in the tools, or any reactive gas around and the hot silicon will bond with the stuff). If the silicon is impure, you have to start again, because only very special impurities in the silicon can be tolerated.
Then you saw the silicon into chips (throwing a lot of it away, but mercifully South Africa has plenty of diamonds for the sawing) and you stick the chips in a vacuum and bombard them with ionised elements which turn their surfaces into something capable of having electrons bounced out of the general electron sea in a metallic crystal, and thus generating electricity. (Incidentally, to get rid of the parts of the surface you don’t like, wash the crystal in sulphuric acid. Yuck!)
Then you have your cells, which you can fit onto a surface preferably facing the sun and wait for it to come up. As with solar thermal, ideally your surface should follow the sun. If it doesn’t, the power is erratic.
Unfortunately, over time — 20 years or so — the crystal structure of the surface of the cell changes with all that photon bombardment and all those bouncing electrons. Gradually the surface ceases to function as a solar cell. At that stage, you throw it away and buy another one. At the moment, this invariably happens before the electricity generated has paid for the cost of making the cell. Ouch. And, what with the pollution caused by creating the silicon crystal and then turning it into a solar cell, it’s not all that environmentally sound, either. The efficiency of such cells is low and you need enormous numbers of them, ideally in areas where the air is clean (polluted air means less power per cell, and urban pollution accelerates the degradation of the cell’s surface, so sticking such cells in cities, where they are needed, is a problem).
Actually things are not quite as bad as this. The Creator is dumping on solar electric purely because it is often seen as The Answer. Things are not so bad, partly because (as almost nobody has pointed out) silicon cells are recyclable. Take the cell back, grind off the degraded surface layer and you have a silicon chip which can return to the vacuum chamber to have its solar-cell capacity restored. This saves the energy required in extracting the silicon, which is most of the expense. Even if you could only do this three or four times, it means that a solar cell would last sixty or eighty years instead of twenty, making it a lot more viable.
At the moment huge amounts of silicon chips are being made. A billion a year, perhaps, for our computers and smart bombs and so on. Two billion square centimetres. But that’s 200 000 square metres, folks. At 1% efficiency, that’s 2 megawatts. A long way too little power to run all those computers. We are going to have to jack up our silicon production, just in this country, until it is thousands of times greater than the world’s current production of silicon chips. Uh-oh; can we do that? Perhaps, but it’s not something easily contemplated. Maybe this high-tech system, though it has a lot of long-term potential, is not ideal for South Africa’s needs.
Wave power seems potentially better. There are essentially two ways of getting wave power: by having something rigid in the water against which a float bobs up and down (much like the wave-power system developed in Britain in the 1970s called Salter’s Ducks), or an even older system of having a vertical pipe, open at the bottom, with a plunger in it which rises up and down as the waves rise and fall, and compresses air as it rises. With a one-way valve you get quite a lot of compressed air which can emerge through a turbine. In both of these cases you store wave energy up and then release it as rotational energy, turning a generator.
Of course, as with wind power, you need a lot of this stuff. The British are fond of wave power because they are an island on a continental shelf, and Denmark, a peninsula on a continental shelf, has similar potential. But South Africa is not an island and our shelf is a lot less shallow than the North Sea and the Irish Sea and the English Channel and the Baltic. In other words, there are fewer shallow places where the wave-power systems can be easily anchored.
Another problem is that anything out to sea is a) highly vulnerable to corrosion from salt water, b) highly vulnerable to being broken by bad weather, and c) rather difficult to repair when corroded or broken. The British have had some bad experiences with their North Sea oil rigs, but those rigs were few in number compared with the scale of a thousand-megawatt wave-power system off the South African coast. It would be basically a city taken out to sea, with all that that implies. Again, you need to store the power; wave energy is more reliable than solar or wind because the waves are there all the time — but they aren’t always the same height.
Ocean thermal has some possibilities, but the Creator wants to discuss that elsewhere.
Fundamentally, the Creator thinks that we need a mix of wind power and solar thermal power for starters — on a very large scale — and, as a subsidiary, start manufacturing a very large amount of solar electric chips and make that grow and grow. Over time, perhaps, solar electric might be the best. Over time, perhaps, wave power might also be worthwhile.
The main thing is that if we start now, we can manage it. It will in the end be little more costly than the plan to provide antiretrovirals to all our AIDS sufferers, and it will be much more practical because whereas buying expensive drugs means giving money to a very small clique, building massive numbers of small power plants linked to a mammoth electrical grid will employ huge numbers of people over a long time. We should get going. This afternoon, if possible.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: