the romance of iconoclastic ideas

An online person named Kris de Decker writes an interesting website called “Low Tech Magazine“, where he offers a unique read on technologies old and new. He recently linked this blog in an article on pedal powered equipment. But the latest article, on solar powered factories, er, ‘crystallized’ (more on that later) a theme that’s I’ve been digesting mentally for quite a while.

I am a lifelong enthusiast in renewable energy technology, and I’m currently working with friends on our own project in this field, so I hope nothing here leaves the impression that I am sour on the potential of renewable energy to make a difference. But this article set me off a bit, both in terms of technical accuracy and this more subtle complaint.

The basic idea of his article is that while there’s a lot of focus on renewable energy as a source of electricity, well over 50% of our energy needs go to heating, a need which (he says) is poorly met by electricity. Which has a grain of truth to it, but in his enthusiasm he goes un-necessarily off the rails. Consider:

Although it is perfectly possible to convert electricity into heat, as in electric heaters or electric cookers, it is very inefficient to do so.

This almost qualifies as a howler.

One of the basic facts about turning electricity into heat is that it can be nearly 100% efficient, which is more or less unheard of otherwise – a simple piece of wire such as in your toaster turns basically all of the electrical energy delivered to it into thermal energy – where else would it go? Digging down a level, the wires in the walls and the wires that make up the electric grid are not completely without electrical resistance, and something on the order of 6% of all the electricity that gets put into the grid is lost (as heat) in the wires along the way, by exactly the same mechanism as in the toaster. Those wires are engineered to be fatter than the wires in the toaster for the amount of current carried, precisely to limit the losses to an economically acceptable level. But a 6% loss for a critical function is not at all crazy in this business.

What Decker is trying to get at is that producing electricity from heat, and then turning that electricity back into heat at the end source, is not very efficient when considered as a whole. The Carnot efficiency puts a theoretical ceiling on the efficiency with which a flow of heat can be tapped to produce mechanical work, and the efficiency is a function of the ratio between the hot and cold side temperatures of whatever machine you’re using – basically the temperature of the burning coal or what -have-you, on one hand, and the river or whatever you’re dumping the waste heat into on the other. Not too long ago, one third of the energy was a reasonable estimate (1/3 turned to electricity, i.e. 2/3 wasted); that number has crept up, but even the most efficient commercial heat engine systems are on the order of 50% efficient (ordinary car engines are much worse), and pushing that number toward 60% has involved some truly amazing engineering to get the hot side temperature up, as illustrated in But the statement, which is repeated a few times, is just plain wrong – consider a modern wind or hydro plant, either of which is capable of extracting in excess of 80% of the theoretically extractable energy of the source; then you take the 6% transmission hit, but you still end up with a respectable overall efficiency. So there’s two basic problems – the efficiency of electric-to-heat is actually very high, and the claimed larger inefficiency applies to thermal sources, not mechanical sources.

It is true that concentrating solar power (CSP)-to-electricity schemes are subject to this Carnot conversion efficiency limit, but on the other hand, achievable heat engine conversion efficiencies are well above those of commercial PV solar, and CSP is amenable to thermal storage, so if your end goal is to produce electricity, it’s not an unreasonable idea, hence the several commercial enterprises that are trying to revive it. In the end, whether CSP or PV wins for solar electricity will depend on how the engineering plays out in the economics – who can demonstrate lower initial/operating costs, who can convince conservative banks of the safety of the investment, etc.

Maybe what bothers me more, though, than the carelessness of the blanket statement about conversion efficiency is how it highlights the way that we all re-jigger reality to conform to our ideals. Previous posts suggest that Decker relishes a curmudgeonly dislike for electricity and various other aspects of modern technology, and he has let that predisposition cloud what’s basically an interesting and not-far-off-base article. Direct conversion of solar energy to heat, especially low grade heat, is a very real, practical technology, and as such is in use all over the world, including in places that don’t have nearly as much luxury as we do to indulge in warm-fuzzy environmentalism. I built a 4×8′ collector mostly out of scrap when I was in high school, and despite relatively crude design (I attached flanges of copper flashing to pieces of scrap copper pipe, using scrap lead flashing as solder – any cognitive defects exhibited here may be thus explained), it worked great, providing most of our hot water in the non-freezing months. (It was a simple design that circulated water from the bottom of the tank directly through the collector whenever a CdS cell indicated sun – no freeze protection.) But then we skip blithely to the notion that silicon chips, solar cells, and other high tech products could be produced directly using concentrated thermal power.

I know a modest amount about silicon processing from previous employment, and I believe this is bonkers. One thing about processing silicon for sensitive applications is that it is an extremely sensitive process. A basic step for making wafers (both for microchips and for PV) is an extremely sensitive process that demands precise temperature control over long periods of time, so the silicon cools slowly and large crystals form. Despite the true fact that sourcing heat directly would be significantly cheaper, the purity and stability demands require in practice that these processes be run by electricity. Temperatures must be held within a percent or so at over 2500F for hours on end, or the whole thing goes to pot. Purity wrt common metals must be held at ppm levels, so processes go on under inert atmosphere, or at vacuum (big pumps that run on electricity). Here’s an example of a company that produces very complex (electric) furnaces for the common Czochrolsky process. You can’t just put a CZ furnace at the focus of a big heliostat array and make boules when the sun shines – the first cloud would end the game. In theory maybe you could figure a way to get some direct solar heat involved, but in practice, these furnaces are a _source_ of low-grade heat – even a small plant needs a big cooling tower to dispose of the waste heat that’s drawn off to keep the metal shells of the furnaces from melting. (Some innumerate folks latch onto the fact that making PV takes appreciable energy to condemn the whole renewables project; various analyses of modern products put the energy payback in the single digit years, well under the manufacturer’s warranty.)

He also talks about a simple open source solar furnace (made partially of aluminum) that’s available for 7 grand and can attain the temperatures needed to heat aluminum, and extrapolates to a time when people will be able to produce their own solar furnaces with onsite solar power. Again, bonkers – I have friends in the CPV business who know just how hard it is to make a decent mirror that will last any length of time in the weather, and before you even get to that you have the significant problem of making a high quality, optically smooth thin sheet of metal out of a hunk of slaggy crap you’ve melted in your backyard. (Thanks to Holly, I’ve actually produced slaggy hunks in a backyard…)

The ability to latch onto an interesting idea (perhaps with a largeish grain of truth embedded in it) and go off the rails due to motivated reasoning is something we’re all subject to, but I think it’s important to watch out for it and resist it. For personal aesthetic reasons I might think I yearn for the Jeffersonian ideal of a society largely formed of of sturdy and resourceful agrarian smallholders, but odds are that given the drawbacks I would hastily repent if magically given an average role in such a scene, and in any case I can’t convince myself that it will come to pass. Then there are the self-replicating machine folks, who are captivated by an interesting idea which will have basically zero practical import in my lifetime (I think I have enough background in 3D printing to plausibly make that prediction), and the singulatarians, who think that in a few decades, once we’ve got enough processing power, humanity will transcend biology. Similar the notions of some well-intentioned friends and family who thought that the financial crisis would finally rouse the downtrodden underclass and bring about a more equal society. But the poster child for this sort of thing is James Kunstler, whose naked desire for the downfall of modern suburban life (and a return to red brick and white clapboards) leads him to unseemly cheerleading of every political crisis and market crash, and creation of fictional future worlds inexplicably indistinguishable from the 1800s.

There is an important distinction between what we would like to happen for our own quirky personal reasons, and what we think actually might transpire. There is enough madness (real or studied) in our politics these days, and enough self-interested hucksters in the renewable energy field, that a dose of gravity and skepticism is in order, in ours and every field. Plus, I actually want to know what’s actually going to happen – both from basic intellectual curiosity, and for practical (e.g. planning) purposes. So, let’s resolve to check facts, question assumptions, and make a habit of asking, ‘what is the influence in my train of thought of what I wish would happen, compared to the vast array of forces that will actually cause the future?’ That might give us a reasonable basis for how to actually change something.


One Response to “the romance of iconoclastic ideas”

  1. Gus Says:

    Thanks for this posting, Ben. I recently read the Low Tech article on Solar Furnaces and came across your comment and reference to further criticism – and was curious enough to follow it up. I take your point made above that the demands of hi-tech manufacturing rise far beyond heat that a solar furnace may ever be able to produce consistently. However, I still found De Decker’s article helpful in giving me a place to start in envisioning a future in which hi-tech manufacturing is possible using only cleantech energy sources. It occurs to me that this very issue may be the organising principle around which future communities might be organised before the end of this century.

    My expertise is in commerce and psychology, not the physical sciences, so I am handicapped in this effort. But I would like to know how we might be able to perform hi-tech manufacturing in the future without harmful emissions. Could a solar furnace contribute to the process? Would we have to burn biomass or methane (or something else) to provide the consistent level of heat necessary to the job?

    I’d be very interested to read any further thoughts you might have on this subject.

    Cheers, Gus Griffin

Leave a Reply

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

You are commenting using your 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: