Power by convection

In June 2006, an agreement was reached to develop a 50MW solar tower generator, in which the sun heats air, which drives a turbine in a chimney, in New South Wales in Australia. The idea is not based on radically new technology. Chimneys with ‘smokejacks’ in which a fan in the hot air turned a roasting spit were certainly in use in the UK in the eighteenth century and probably before. It is well known that a tall chimney produces a better draught, presumably because there is a greater temperature difference between bottom and top, wind speeds are greater higher up, inducing more air to be drawn out of the top of the chimney by Bernoulli effects, and a long distance up the chimney should help hot air build up speed and kinetic energy. But modelling all these effects and producing an optimally efficient design seems to be almost impossible, at least we have so far been unable to find anyone to take the task on. Alan Williams, a former MP, has since turned to championing such ideas. This conception describes how a large, sealed ground level solar air collector generates warm air which rises. “It flows through a convergent nozzle where enthalpy is converted into kinetic energy, which then drives an air turbine,” he says. “Post-turbine, the air is cooler and denser and falls under gravitation to the base of the solar collector to complete the closed cycle. The cycle is driven by natural convection.” He believes that an overall efficiency of over 50% conversion, from solar energy into electricity, should be achievable. Dr Williams points out that some remarkable working models were built in the nineteenth century that give some indication of how it is possible to use convection to convert very small amounts of low grade energy into mechanical work. He cites two in particular: the Joule Thermoscope, of which there is only a description; and the Convection Mill, which can be seen in the Heat gallery of the Science Museum in South Kensington. The Thermoscope was a glass tube, divided into two compartments with a blackened piece of cardboard dividing the two sides but with a gap at top at bottom. Radiant heat arriving on one side, would heat the air there, causing it to rise and circulate, deflecting a magnetic needle on a strand of silk. The Convection Mill was devised by Alfred R Bennett in 1897. It is also made in the form of a closed glass vessel, but in this case there is a tube in the centre to conduct the warmed air upwards with the return path on the outside of it. This acts on turbine blades both on the inside and the outside of the central tube. The importance of these two devices is that they both respond to moonlight, which is about the lowest grade heat source one could imagine. Dr Williams argues that they defy the Second Law of Thermodynamics, because they still work in spite of the low effective temperatures of their sources. We might argue that the effective temperature of the source of moonlight is the Sun, which has a very high effective temperature. What they show is that the energy of even the tiniest amount of sunlight can still be harnessed in this way and turned into mechanical energy. Dr Williams has come up with a large number of different configurations, but has no idea as to which would be the most cost-effective. The only certainty is that, to be efficient, they would need to be large. A 50kW pilot open loop solar chimney was operated at Manzares in Spain from 1982 to 1989. The chimney was almost 200m high and the collector roof covered an area of 46,000m2. The Australian project, in the hands of a company named EnviroMission, was originally to have built a 200MW demonstrator with a 1,000m high chimney and a 7,000m diameter collector area. EnviroMission Dr Williams web site, Global Warming Solutons Pointers * Engines in which rising hot air turns mechanical turbines have been known for centuries * Nineteenth century development showed that they could be made very efficient, so much so that they could even be powered by moonlight * Solar heating of air used to drive a turbine has been demonstrated, and funding agreed for a 50MW demonstrator in Australia – but the level of funding to improve, optimise and develop the technology to its full potential is lacking