By Jason Deign in Barcelona

Cooling is a headache for CSP. The traditional way to cool a plant’s steam turbines is with water, but that is a precious commodity in many of the desert environments where CSP is being introduced. In some cases water may be more valuable than the power a plant can produce.

Which is why project developers are increasingly turning to dry cooling systems that use fans to keep the turbine temperature down.

Such systems come with their own set of problems, though: namely, that they add significantly to the capital cost of a project and they guzzle a fair amount of power, equivalent to up to around 5% of the total output of the plant. This can dent the profitability of the whole venture.

There could be another way, however. At the 2010 SolarPACES conference, a trio of Paris-based investigators put forward a study on whether tall chimneys could provide a passive dry cooling system for CSP plants.

Denis Bonnelle, of the Centre Énergétique et Procédés de l’École des Mines de Paris, Frederic Siros, research engineer for EDF, and Cedric Philibert, senior analyst at the International Energy Agency, proposed employing convection-driven Heller cooling towers in place of fans.

Such systems are already employed in other types of power plants, such as the 1,400MW combined-cycle gas turbine Bursa power station in Turkey, which boasts 135m-tall chimneys.

And CSP itself has a long history of fascination with chimneys, dating back to early solar updraft tower pilots. However, as with the solar updraft concept, one of the main challenges associated with Heller tower-based dry cooling appears to be one of size.

According to Bonnelle et al’s analysis, chimneys are highly subject to economies of scale, and are not worth building in place of fans for any project delivering less than 300MWe. Beyond that point, they begin to make a lot of economic sense.

Conservative hypothesis

“Even with conservative hypotheses, chimneys 500m high and 60m in diameter, costing USD$28 million, could be profitable for CSP outputs from 300MWe onwards,” says the study.

“For a 1,000MWe field, a 660m chimney could generate net savings of 55% of its $71 million cost.”

That is great news. Except in 2010 nobody was building plants anything like that big. It is perhaps not surprising, then, that the paper failed to spark a rash of projects with chimney dry cooling. But times are changing.

Next year the Ivanpah Solar Electric Generating System plant is due to begin delivering up to 370MW to the Californian grid, making it the first project to enter the feasibility range for chimney dry cooling (in the event, it will use fans).

And this year work is scheduled to start on the first 160MWs of a solar project in Ouarzazate, Morocco, which could ultimately deliver 500MW.

The project will combine CSP with PV and at this stage the exact split between them is uncertain, although solar thermal is likely to be favoured.

There are a couple of reasons why the creation of large projects in emerging markets such as North Africa is particularly interesting from a tower-cooling perspective.

The first is that water is really scarce in many of these locations, so water-based cooling is much less of an option than in richer, established markets such as Spain and the US, unless CSP is used partly for desalination. The second is that labour is much cheaper.

Financial viability

This could change the 300MW-lower-size-limit for the financial viability of Heller towers since, says Bonnelle, his model was created using Australian labour costs. “In India it could be three times cheaper,” he states.

Improved economics could allow developers to try out towers on smaller plants, which would help provide some real data to prove the concept and might also assist with the solution to a number of other potential issues, such as where to put the cooling chimney or chimneys.

Being large structures, the chimneys would cast a wide shadow and so would presumably need to be located away from the prevailing angle of incoming sunlight, although of course the exact sites would also depend on the distance from the steam turbine.

Another option would be to rotate the solar field around the tower using tracking systems, a concept currently being investigated at the Renewable Energy Advanced Technology Centre (Tecnológico Avanzado de Energías Renovables or CTAER in Spanish) in Andalusia, Spain.

In this instance, a cooling chimney and a power tower could conceivably be combined in single structure at the centre of the solar field, although Bonnelle accepts this might represent too much innovation for most developers to handle within a single project.

Nevertheless, Jesus Fernández-Reche, of the Spanish Centre for Energy, Environmental and Technology Investigation’s Almeria Solar Platform, believes the cooling chimney concept deserves further study.

“It is certainly possible to build chimneys this high,” he says. “But right now the idea is still in nappies.”

To respond to this article, please write to Jason Deign

Or write to the editor Jennifer Muirhead: jmuirhead@csptoday.com