A common lament of those analysts wishing to get to grips with the real-world performance of solar thermal power plants has been, well… an absence of data. Trainer noted, in ‘Solar Thermal Questions‘:

It would be great to get some actual data on their year round performance. I have found it fiendishly difficult to get such data out of anyone; they seem not to want to make it public, and this makes evaluation of claims very difficult.

During the Equinox Energy 2030 summit, Jay Apt noted some issues with utility-scale PV farm performance, as illustrated in the figure below (from this paper):

Note that this is from a solar PV farm in the Arizona desert — one of the best locations in the US for this type of facility. The associated commentary said:

Observed rapid and deep fluctuations at time scales of 10 seconds to several minutes may indicate that a component of the intermittency is due to low, scattered clouds with significant opacity. We observe a number of examples of output power rising above nameplate capacity before and after deep drops in power. This may be due to focusing of sunlight around the edges of low clouds. If PV becomes economically attractive enough to be deployed at large scale, intermittency is likely to be matched with dispatchable power, storage, and / or demand response

The implied ramp rates to compensate for these types of fluctuations will be challenging. Indeed, some form of large-scale battery energy storage seems vital to maintain quality of the electricity output.

That is PV. Now, at last (and thanks to a tip from regular BNC commenter John Bennetts), I have some data on solar thermal performance. It comes from the final report of the Colorado Integrated Solar Project, which you can download here (25-page PDF).

First though, some details on the facility:

The world’s first hybrid solar/coal power plant has been built near Palisade in Colorado. Xcel Energy and Abengoa Solar are partnering on the demonstration project which uses solar parabolic trough technology to supplement the use of coal. Initially, it’s expected to reduce the emissions generated by the Cameo Station’s Unit 2 plant by three to five percent, but it’s thought that this could increase to up to ten percent.

The system focuses solar energy on mineral oil, which is then passed through a heat exchanger where it’s used to preheat the water used by the coal-powered part of the 49MW plant.

You can also go to its National Renewable Energy Laboratory page for further technical specifications on the plant. In short, the expected generation was 49 MWh per year for the 6 acre parabolic trough facility, with a 2 MW turbine capacity. The NREL page says:

 A parabolic trough solar field provides thermal energy to produce supplemental steam for power generation at Xcel Energy’s Cameo Station’s Unit 2 (approximately 2 MWe equivalent) in order to decrease the overall consumption of coal, reduce emissions from the plant, improve plant efficiency, and test the commercial viability of concentrating solar integration.

How much savings of coal (or actually, CO2-e)?

…should the demonstration be successful, the implications are not to be sniffed at. A ten percent cut in coal consumption in coal fired power stations in areas with the appropriate weather and solar intensity would be a great boost to efforts to limit carbon dioxide emissions globally.

According to the performance review document, the fuel savings would come in three ways:

A reduction in fuel and emissions was expected from three operational changes brought on by the solar heat addition. The three operational changes that contribute to fuel and emissions reduction are: a reduction of high pressure steam extraction, increased available steam for generation, and supplemental heating of feedwater.

Sounds good. So what was the actual result? The system starts up once the Direct Normal Insolation (DNI) reaches 200 W/m2 and can keep operating for a while at DNI below this value (p6). The maximum temperature achieved is ~300C. The results of the integration were positive (p7),  with minimal impact on the coal plant operation.

The total coal savings for the project were 238 tonnes of coal fuel (p10), or a total emissions saving of 528 tonnes of CO2-e, for a facility that cost $4.5 million. Here are the details of the actual performance data:

Obviously not great, but how does this compare to their predictions? The following table of pre-operation forecasts tells the tale:

So, the expected coal savings, converting to tonnes, was 11,017 t of coal and 2,442 t of CO2-e. That is, the actual performance was 2.2% of the predicted performance in terms of fuel savings, and 22% of expected in terms of CO2-e reduction. The report tries to put a brave face on the results (p13):


Overall the performance related to coal and emissions savings were not as good as Abengoa predicted or what Public Service expected. There is reason to believe that these results are attributable to the small scope of the project as a demonstration project. However, the integration into the feedwater cycle of an existing fossil facility was successful. The project was not designed to maximize efficiency or performance. For example, to minimize costs, less insulation was used than what would have been installed for a 20 year design. Mirror washings were less frequent than would typically be performed to reduce O&M expense. As previously mentioned, Abengoa took this opportunity to test a new collector frame design; however, the results were that the redesigned system did not provide the anticipated solar energy collection efficiency instead the efficiency actually decreased.

Future Deployment

At this time, Public Service believes that it would be best to take a “wait and see” approach before deciding on further deployments of solar integration with fossil fuel feedwater systems. Though the Company believes it achieved a successful integration of the solar heating into the feedwater system, the situation regarding costs and efficiencies is fluid, making it difficult to make any definitive recommendations regarding future deployments at this time. Based on our costs for Cameo, the Company would conclude that the cost on an equivalent MWH basis is much higher than wind or PV solar.

However, as discussed below in more detail, technological changes have occurred and it is likely that costs will come down. The Company believes that the best approach at this time is to continue to monitor developments relating to CSP technology, as well as other renewable technologies, before deciding on any future deployment of CSP technology.

The report concludes (p14) with some hopeful statements about future costs decreasing and performance increases. I hope so too (but if wishes were horses, beggars would ride…).

There are similar serious proposals for Australia. such as at Liddell power station, where AREVA have said they will integrate a 38 MW solar bolier feed water  pre-heating to (putatively) save fuel, and thereby CO2-e. After the experience in Colorado, they may be having second thoughts, unless the subsidy is sufficiently high.

These results also highlight an inherent danger with the ZCA2020 plan — until we have good performance data from their proposed CSP solar tower facilities, we risk building our nation’s electricity future on a very expensive house of cards. As the old saying in science goes, ‘Data is King’. Let’s have more of it (real-world data), before we make any irreversible investment decisions, and before we arbitrarily rule out known reliable zero-carbon options like nuclear (for which abundant performance data exists, hour-by-hour, day-by-day, for commercial operations over decades).