A review of concentrated solar power (CSP) in Spain

Euan Mearn’s recent Red Eléctrica de España (REE) post drew my attention to the fact that REE has now begun to show grid data for solar PV and concentrated solar power (CSP) generation separately instead of lumping them together. In this post I use the REE data to review the performance of Spain’s CSP plants and to check among other things whether the claim that they are capable of providing baseload generation, as this 2011 Forbes article claimed, holds up in the light of operating experience. Spain is a good case study because the lion’s share of world CSP capacity (2.3 of 3.4GW in 2013) is installed there.

Gemasolar 20MW CSP plant, Spain. “The molten salt storage tank permits independent electrical generation for up to 15 hours without any solar feed”. Image credit Torresol.

Types of CSP plant

First a brief review of the different types of CSP plant. The Gemasolar plant shown above is a solar tower CSP plant, where dual-axis tracking mirrors reflect solar energy onto a central tower that contains a working fluid, usually molten salt. The salt is heated to temperatures of 300˚C or more and passed through heat exchangers that generate steam to drive turbines. Solar towers are not the most common type of CSP plant but they are easily the most photogenic.

The most common type of CSP plant is the parabolic trough. Here a linear single-axis tracking mirror array concentrates solar energy onto a tube containing a working fluid – again usually molten salt – located along the focal point of the mirrors, and the molten salt is again used to generate steam to drive turbines. The Andasol in plant in Spain is an example:

Detail of single-axis tracking array, Andasol parabolic trough plant

Other types of CSP plant include dish stirling, which uses stand-alone parabolic reflectors to concentrate light onto a receiver and delivers the heated working fluid to a Stirling engine and fresnel reflectors, made of many thin, flat mirror strips that concentrate sunlight onto tubes through which working fluid is pumped. More details are available here.

These different CSP technologies have their advantages and disadvantages but they all have a common feature that distinguishes them from solar PV plants. They convert solar energy into heat, and heat can be stored for re-use. In short, they have a built-in storage battery that allows them to generate electricity when the sun doesn’t shine.

They also have another less-well-publicized common feature. They need an auxiliary source of heat to jump-start them in the morning, and the heat is commonly provided by natural gas. Spain in fact allows CSP plants to generate up to 15% of their electricity from natural gas. A list of Spanish CSP plants is here.

REE June grid data

Now to the REE grid data. REE only started segregating PV and CSP generation in April of this year so a year-long evaluation isn’t possible. Here I review the data for just two months – June 2015, which I have assumed was a typical summer month in Spain, and November up to the time of writing (November 26) which is as close to a winter month as I can get.

Figure 1 is a stacked bar chart showing PV and CSP generation for June, a month during which the sun shone brightly except for a cloudy period before mid-month. CSP generation tracks PV generation but has a somewhat different shape and tends to lag PV by about an hour, presumably because of thermal inertia. The PV capacity factor over this period based on 4.7GW of installed capacity was 25% and the CSP capacity factor based on 2.3GW of installed capacity was 45%. More about capacity factors later.

Figure 1: PV and CSP generation  for June 2015, Spain

The CSP plot shows nighttime generation of power from stored heat. However, so does the PV plot – over 600MW of it on June 29 and 30. There is obviously a problem here but I have assumed that it affects only REE’s PV data and that the CSP data are correct.

Figure 2 shows CSP generation segregated into daytime generation and “nighttime generation from storage”. Two points should be noted:

  1. 1. The plot sums the generation from over 40 CSP plants with built-in storage capacities ranging from zero or near-zero up to 15 hours (although the storage capacity in MWh is never specified).
  2. I define “nighttime generation from storage” as the CSP power generated between the time the sun went down, as defined by the cessation of PV output, and minimum daily CSP generation, which usually occurs shortly after sunrise.

Calculated using this definition only 19% of Spain’s June CSP generation came from storage and the amount of energy stored was never sufficient to deliver stable power through the night. Power delivery always began to fall off before it picked up again in the morning:

Figure 2: CSP generation  for June 2015 segregated into “daytime” and “nighttime from storage”

In addition, no attempt was made to match CSP generation to demand (Figure 3). Daytime generation tops out around 2.2GW, most probably because the 2.3GW of CSP turbines were “maxed-out” at this level, but the working fluid continued to accumulate heat that was released through the turbines later. The blue areas can in fact be visualized as having been scraped off the top of the orange areas:

Figure 3: CSP generation  for June 18th through 22, 2015 segregated into “daytime” and “nighttime from storage” and compared with demand. Note that the left and right Y scales are not the same.

REE November grid data

Figure 4 plots PV versus CSP generation for November. The PV capacity factor has fallen to 14% from 25% in June but the CSP capacity factor has fallen to 14% from 45%. Clearly mucho sol is needed for CSP plants to work efficiently. (Note that PV plants generate electricity in the absence of direct sunlight while CSP plants do not):

Figure 4: PV and CSP generation  for November 1 through November 26, 2015

Figure 5 plots daytime CSP generation versus nighttime generation from storage for November. The percentage of total generation from storage has decreased to 10% from 19% in June, which when combined with the decrease in capacity factor reveals that Spain’s CSP plants delivered six times less stored power in November than they did in June. On the first three days of the month they delivered no stored power at all:

Figure 5:  CSP generation  for November 1 through 26, 2015 segregated into “daytime” and “nighttime from storage”

And again no effort was made to match generation to demand, as shown in Figure 6. (The double peak in CSP generation is particularly intriguing. I have no backup evidence but speculate that the first peak may be when they turned the gas off):

Figure 6: CSP generation  for November 6 through 9, 2015 segregated into “daytime” and “nighttime from storage” and compared with demand. Note that the left and right Y scales are not the same.


As things stand Spain’s CSP plants are nowhere close to providing baseload generation. But then, they have no incentive to provide it. As I understand it they get paid the same kWh rate regardless of when the power is delivered, which gives them an incentive to produce power as quickly as possible after the sun gets up to minimize working fluid heat loss. If they were paid a large enough premium for nighttime delivery the generation plots would be flatter, but how much flatter is impossible to say.

As to whether the plots could ever be smoothed to the point where Spain’s CSP plants provide baseload generation at roughly the same level year-round, however, the answer is an unqualified “no”. Even if the plants had been able to smooth out diurnal variations so that the same generation level was maintained through the day we would still be left with large day-to-day variations in June and even larger seasonal variations between June and November, as shown in Figure 7:

Figure 7:  Average daily CSP generation, June and November, 2015  

Smoothing out the June variations so that power could have been delivered at the monthly average of 1,030MW at all times during the month would have required a large amount of storage, and delivering power from June through November at the average June-to-November level of 670MW a prohibitive amount. Figure 8 summarizes the power storage and release requirements that would have been needed to maintain a constant 670MW of baseload generation during June and November. Approximately 260 GWh of storage would have been needed to cover the shortfalls in November alone. Installing this much storage to support less than a gigawatt of baseload generation is clearly not a viable option:

Figure 8: Power to and from storage needed to maintain constant 670MW baseload generation, June and November, 2015

A final question is whether the higher capacity factors of CSP plants means that they are more efficient than PV plants, as is sometimes claimed:

Torresol says that the plant will provide electricity for about 20 hours each day on average, with numerous days in the summer seeing 24-hours of supply. How does that compare with a similar-sized PV plant? The 21.2 MW Photovoltaic Solarpark Calaveron in Spain generates about 40 GWh a year. This smaller 19.9 MW power tower plant will generate about 110 GWh per year.

CSP plants have higher capacity factors than PV plants simply because capacity factors are calculated based on the capacity of the turbines, not the capacity of the mirror array. When the capacity factor is calculated based on the capacity of the mirror array they are actually less efficient than PV plants. Andasol 1 is an example (data from NREL):

Turbine capacity: 50 MW
Generation at 100% CF: 438,000 MWh/year
Actual generation: 158,000 MWh/year
Capacity factor: 36.1%

Mirror area: 510,120 square meters
Solar resource: 2,136 kWh/sq m/year
Mirror array capacity: 124.4 MW
Generation at 100% CF: 1,090,000 MWh/year
Actual generation: 158,000 MWh/year
Capacity factor: 14.5%

If Andasol’s dual-axis-tracking mirrors were dual-axis-tracking PV arrays we would expect a capacity factor of over 20%. By this yardstick the Andasol plant is roughly a third less efficient than a PV plant of the same size.

ADDENDUM: December 2, 2015

Willem Post suggested in comments that it would be helpful to include the graph below in this post. It plot Spain’s CSP generation and demand for November 2015 on the same scale and illustrates how little CSP presently contributes to Spain’s generation mix:

This entry was posted in Energy and tagged , , , , . Bookmark the permalink.

65 Responses to A review of concentrated solar power (CSP) in Spain

  1. Peter Lang says:

    Thank you for this post. I’ve been looking for the actual output performance of CST plants for some years, and this is the first time I’ve seen it presented. So well done.

    Regarding this important point at the end of your post:

    CSP plants have higher capacity factors than PV plants simply because capacity factors are calculated based on the capacity of the turbines, not the capacity of the mirror array. When the capacity factor is calculated based on the capacity of the mirror array they are actually less efficient than PV plants.

    This is an example of why, arguably, the most important criteria for valid comparisons between technologies is the cost of electricity including properly attributable share of grid costs.

    • Peter: Thank you

      According to IRENA CSP is the most expensive source of renewable energy in terms of both LCOE and capital cost:

      • Peter Lang says:

        Thank you Roger. Good comparison charts. And we should add to the LCOE of the generator technologies around $36/MWh at 10% penetration and $56/MWh at 30% penetration for the grid costs attributable to solar (using the OECD/NEA estimate for solar PV for the six OECD countries studied; six country averages summarised here: http://www.energyinachangingclimate.info/Counting%20the%20hidden%20costs%20of%20energy.pdf ). The additional grid costs attributable to solar PV are higher than for coal by about 40x ar 10% penetration and about 60x at 30% penetration.

      • Günter Weber says:


        with respect to photovoltaic the data is outdated (or is because of the Euro-Dollar exchange rate?). Larger PV (above a few 100 kW capacity) plants in Europe are currently build with feed-in tariffs around 100 Euro per MWh.

        And the data for offshore wind is probably totally wrong with respect to China and India (4 times cheaper than in OECD?).

  2. Davis Swan says:

    This is a very useful and thought-provoking article. Thanks for going to the work of producing it. It also highlights one of the basic problems with LCOE and that is that not all production is of equal value. The ability to produce during peak demand times should be reflected somehow in the value of CSP as compared to PV. For example, it would be easy to imagine Oahu reaching the point where mid-day PV production on many days exceeds total demand (almost there now) in which case any incremental production is literally of no value. When you take into account that some thermal production needs to continue to keep the boilers hot this mid-day limit could be quite low. Although some jurisdictions have implemented time-of-use retail pricing it is very crude and doesn’t reflect real-time market demand. Wholesale prices sometimes do reflect market demand and can vary by orders of magnitude in a few minutes if there is a lot of wind in the mix. I think we have to move to a “available supply” pricing scheme which would reflect a much higher value for CSP with storage over PV. More complicated but ultimately more accurate and would help justify what I believe is the best way forward between 35 degrees latitude North and South.

  3. Willem Post says:


    Thank you for this revealing article about the Spain CSP boondoggle, that has cost Spain a fortune and provides a minimal quantity of unreliable energy to the grid. That 15% natural gas for start-up is likely imported at high prices.

    How economical ARE these CSP plants? Are all costs considered and included?

    Instead of different scales, it would have been even better to show demand and CSP to the same scale, to emphasize it smallness in the overall picture.

    Jacobson and Delucchi, of “100% RE by 2050” fame, will be quite upset to see this, because it blows out of the water their scheme of CSP with at least 10 hours of storage for continuous operation, in the US southwest.

    The CSP capacity would need to be connected to a US-wide HVDC overlay grid stretching up to northern Maine to provide energy 24/7/365, i.e., AT ANY TIME local wind and solar generation and storage would be inadequate to serve demand anywhere in the US.

    The HVDC overlay grid would be required to minimize transmission losses. Based on 4 such proposed lines, mostly overhead, from New England to Canada, the capital cost would be about $7.5 million/mile. At least 10 thousand miles would be required. The Jacobson report ignored/does not mention this $750 billion capital cost, and its associated O&M and replacement/refurbishment costs.

    Here is the URL of their US study:

    Here is a URL of FAQs regarding the US study:

    After my third email to Jacobson, he finally answered and attached this URL with spreadsheets:
    See fifth line down.

    He and his sidekick were invited to present their US study and Worldwide study in Paris. This goes to show what those serious, well-meaning, people like to hear as RE gospel. Soon, they will be mandating having us pay for Heaven on Earth.

    I have emailed them your article, which may give them indigestion, and will let you know their reaction, if any.

    • Instead of different scales, it would have been even better to show demand and CSP to the same scale, to emphasize it smallness in the overall picture.

      And that’s most of the world’s installed CSP capacity you’re looking at there.

      • Willem Post says:

        Just as I thought. Please insert it into your article.

        As Jacky Gleason would say: “And away we go”.

        I sent an email to Jacobson and Delucchi, to which Delucchi responded, and to which I responded. If you like copies of these emails, I’ll be glad to send them to you.

        As a result of your article, I added this section to NOTE No. 4 of my article:

        Example of CSP in Spain: In light of the Spain experience (see URL), CSP w/storage, after 15-20 years of development, appears to be nowhere near ready for prime time. It required great cost to implement, $million/MW, (which Spain could not afford) and provides just a little of unreliable, expensive energy!!

        As peaking, filling-in and balancing is of major importance, and, according to emails with Jacobson and Delucchi, CSP would not have to perform “all gap filling, any time, anywhere”, then how much of that gap, and at what times, is CSP w/storage providing energy, and how much of that gap, and by what means, and at what times, are OTHER systems providing energy?

        Jacobson and Delucchi claim to have made a detailed study of this aspect. However, details of the systems, including the extent of the HVDC system, and capital costs, are insufficiently mentioned in the Report.

  4. Willem, I make that $75B, not $750B. Either way, it’s way out of reach.

    Re that 15% figure for gas as a fraction of Spanish CSP: I guess that is 15% of energy sent out, not 15% of nameplate rating.

  5. stewgreen says:

    Roger it is interesting what you say re capacity factor.
    That PV people state capacity as the power of the panels.
    whereas CSP people state capacity of the turbine.
    #1 No one here cares what capacity is. It is just a PR trick that renewables mafia and media* use to deceive. (Like their trick of quoting power in ‘pixie’ homes, as if that is where most electricity is used.)
    eg “this is 1GW wind park, that power X number of ‘pixie’ homes.”
    What we care about is the average power delivered over a year. eg that 1GW solar/wind farm is like adding a 0.3GW Gas ENERGY PARK.

    If they want to state a derated capacity just like some UK wind operations do to stay within subsidy limits that’s their business.
    ie They might say it’s not 1GW with a Cap factor of 25%, but a 0.5GW with a 50% cap factor.
    The result is the same, it delivers 0.25GW on average.

    * Harrabin in his BBC article hyping Noor1 began by unfavourably comparing Morocco’s 42% Renewable Capacity target with a 30% UK Renewable Delivery target. But the informed outside observer is thinking Harrabin tell us what the Moroccan average delivery is going to be. ie it doesn’t matter that Noor1 is 160MW it matters what it actually outputs.

    Another difference with CSP is that apart from the intermittency, it inherently must have great transmission losses cos it’s not situated close to it’s users. Like nuclear it might be hundreds of Km from them.

    A simple question ?
    For CSP can’t you just couple the solar warming along with controllable gas heating into the same boiler to produce a baseload ? At least during daytime in the summer ?

  6. Günter Weber says:

    Remark 1: The efficiency of a solar power plant is absolutely irrelevant. It would only come into play when the space covered by a plant would be a limiting factor. If it was only cheap enough, you could perfectly use a solar plant with 0.00001 efficiency.

    Remark 2: Figure 6 shows that CSP output roughly coincides with peak demand (as long as the sun is shining). If one aims for using CSP as one of the main pillars of electricity production (not only 2 GW, but maybe >10 GW), it would make sense to optimize their output to cover the daily peak demand. For sure it makes no sense to ask for significant output in the early morning when demand is lowest. Remember that you need to have backup anyways.

    Remark 3: The earlier start of PV output compared to CSP output is a nice feature when looking at the demand curve.

    • robertok06 says:

      “If it was only cheap enough, you could perfectly use a solar plant with 0.00001 efficiency.”

      Not even remotely close to truth!… you may want try and read the excellent paper by de Castro et al, published on Energy Policy a couple of years ago, where the authors dispel the MYTH about the availability of areas to install PV on the planet… a free copy of that paper is here:


      “Although some uncertainties can not be avoid, our estimations for the global potential
      of solar electrical power are 1,75-4,5 TWe, which implies a hard techno-ecological of
      solar power potential, much lesser than other assessments.”

      Note that the calculations are done with PV efficiencies of the order of 15%… so go figure what the conclusion would be with your 0.00001 value.

      “He/she who refuses to do the math is doomed to talk nonsense – J McCarthy”

      • gweberbv says:


        do we really need to discuss if my number of ‘0.00001’ was an auxesis or not?

        Concerning the paper you are linking to:
        1 TWe translates to 10 TWp of installed PV (using a pessimistic capacity factor of 10 – and assuming that I did unerstand this unit correctly).
        In 2011 word electricity production was something like 20000 TWh. If we assume a constant production (of course a simplification), this amounts to slightly more than 2 TW of running capacity.
        -> 10 TWp of PV is likely to supply peak demand (on a sunny day) for most of the world. Without having a realistic storage solution, it would not make sense to install much more PV anyways.

      • Günter Weber says:


        by the way:

        The paper also cites an upper limit for wind energy around 1 TWe. (I had a look in the original publication and the number is correctly cited.)

        By the end of this year, we may have something like 425 GW of wind power installed worldwide. With a pessimistic capacity factor of 0.2 this means we have already come close to 10% of the stated limit.

        So, maybe one should take such estimated with a grain of salt.

  7. Joe Public says:

    An instructive video of Top Gear’s James May’s visit to a Spanish CSP is here:


    The observation that even a thin layer of dust reduces output by ~10% is noteworthy.

    • Euan Mearns says:

      I imagine dust may also defract light – not good when trying to concentrate it. And to remind those who have not read the previous post on “The wind in Spain…”, when its cloudy, CSP simply switches off.

  8. robertok06 says:

    “CSP plants have higher capacity factors than PV plants simply because capacity factors are calculated based on the capacity of the turbines, not the capacity of the mirror array.”

    The correct figure of merit, IMHO, should be the average W/m2, i.e. the energy delivered in one year divided by 8766 h/y divided by the area covered by the installation (everything included, even non-mirro and non-panel surfaces).

    For both technologies, it is a few W/m2… and this puts the nail in the coffin of their viability as alternate sources to FF, run-of-river hydro, and nuclear.

    • Graeme No.3 says:

      Low energy density has always been the problem with wind and solar because the capital costs become so much higher. Despite what most greenies think these have to be paid for.
      Tidal and wave power plans are also intermittent but have much higher energy densities, so much so at times that the plant gets mangled.

  9. Olav says:

    This have potential to reduce CSP cost and eliminate the need for natural gas use.
    This storage does not depend on CSP to be useful.

    • gweberbv says:


      what is so interesting about heat-resistant concrete? This is nothing new.

      • stewgreen says:

        @gweberbv It’s NOT “heat-resistant concrete”
        Heatcrete is a “concrete-based storage medium” an alternative to molten salt that some CSP’s use.
        ..looks like you didn’t read the article

      • Olav says:

        Heatcrete expected to last indefinitely without cracking, has high heat capacity and high heat conductivity. If located close to a heat user its round trip efficiency is very high well above pumped hydro storage. If only electricity is produced from this storage then the carnot loss is higher than PHS. CSP is only the initial setting it is used for. Its main usage will be almost any stationary setting where storage is required.
        Especially where heat is in excess being stored and the usage is also heat. Actually we in Europe uses more energy for heat than for electricity.
        It can be the pumped hydro storage for the “flat land guy” making intermittent energy sources much more attractive.
        Just Google “NEST Thermal storage”

        • Günter Weber says:


          ‘heatcrete’ is just a new name for something that exists since decades. -> http://www.everything-about-concrete.com/heat-resistant-concrete.html
          Facing temperatures of a few hundred degree Celsius is not a unique feature of CSP facilities.

          Replacing a molton salt tank with a ‘NEST’ might be slightly cheaper (concrete is cheap, of course, but what about the hundres of kilometers of pipes inside the storage device?). However, as long as you still take molton salt as the heat transport fluid, you will also have to keep the whole device above 200 degree Celsius.

          When comparing such a device to pumped hydro storage: Thinking about converting electricity to heat and then back is more or less madness (unless you can make use of the huge heat losses).

          • Olav says:

            A NEST module consists of many 25 cm diameter Heatcreate sylinders stacked. It has a steel pipe in the middle for fluid circulation when concrete is poured. Replacing steel bars.
            There may be many kilometers of pipe if many modules are stacked together. But connections are limited to top and bottom. Circulating fluid is not molten salt so temperature range is what the circulating fluid can take..0..400C is possible and higher temp is a possibility.
            The heat loss inside insulated module is minimal over 24h. Losses occur when converting to electricity, therefore having a heat user at hand is preferable.
            But if surplus (low value) electricity is at hand then why not store it as heat.
            Just Google “NEST Thermal storage” is my recommendation

          • Günter Weber says:


            there is no magic bullet for high-temperature heat storage. Many concepts are possible. -> http://energystorage-journal.com/upload/uppic/201212109457304.pdf
            For relatively low-temperature storage there is nothing as cheap as water.

            Even Denmark and Germany have zero electricity prices only for a few hours per year. It just does not make sense to build an infrastructure to make use of this ‘excess energy’. Not yet. And probably not in 10 years from now.
            If you desperately looking for some sort of storage, then simply command all heat pumps in country to aim for a slightly higher (room) temperature. Once a building is heated up, with proper insulation it will last for a day or two (saving heat demand during that time).

          • Olav says:

            This is new.. Present literature on concrete may not be up to date..
            looks like a “M bullet” for me. Viable storage is needed.
            Pilot plant in Abu Dhabi has been through its test now ind indications is that it got results above what is said in below document.


        • Jonathan Madden says:

          One might consider pressurised water for heat storage. Its specific heat remains more or less constant with increasing temperature, so a heavily reinforced vessel with, say, Zirconium heat exchange rods carrying superheated steam, could be raised to 300C to 400C, thereby doubling or tripling thermal storage capacity while retaining the advantages of high specific heat and a liquid medium.

  10. A C Osborn says:

    Euan, Off Topic, but you may like to comment on this article on Germany’s Energiewende over at Judith Curry’s Climate etc as you have done so much work on the subject.

    You might like to answer TonyB’s comment on UK tidal generation.

  11. Mark Miller says:


    The owners of the Ivanhoe CSP facility in CA get paid different kWh rates depending on when power is delivered to the grid.

    Time of Delivery (TOD) factors were built into long term PPA’s as denoted in Appendix B : http://docs.cpuc.ca.gov/WORD_PDF/FINAL_RESOLUTION/154753.PDF Over the years the “Market Price Referents, Appendix A, have come down. The long term contracts signed via the 2008 Resolution had 20 year contracts for projects brought on line in 2011 via a Market Price Referent of $.1173 kWh. I understand, but do not have any hard data that contract prices are now closer to $.05 kWh for eligible RE. In the past (2008 to 2011 resolutions) the TOD factors changed to value (ie pay more) for kWh delivered at “super peak times” .

  12. My ten cents’ worth on heatcrete storage. Everyone is losing sight of the problem, which is the enormous amount of storage needed to do any good. I calculate in the post that 260GWh of storage would be needed to smooth Spain’s intermittent CSP output into 670MW of baseload generation – equal to only about 2% of demand – in November alone. How much heatcrete do we need to store 260GWh? Assuming an energy content of 0.75 kWh/m3 at 340C and 100% heat conversion efficiency we need approximately 800 million tonnes, or if you like a square kilometer of heatcrete standing 350m high. And it would have to be kept warm for months.

    • Günter Weber says:


      would it not make much more sense to optimze CSP as peak load plants for covering the high demand from late morning to the evening?

      • Peter Lang says:

        No! The capital cost is enormous, Their is no indication they can be economic. Ivanpah is about $19/W (average) last time I looked, but that was based on their projected energy generation per year. The actual generation so far has been about half their projections, so you can double the above figure. (The $19/W average is from memory; I could be confusing Ivan[ah and Gemasolar, but the point is stili valid even if the figures are out a bit – i.e. CSP is hugely expensive)

    • Olav says:

      Something missing here Energy content is 0.75 KWh for each m3 with temperature difference of 1K. It will then be 800/500 or 1.6 million tons by temperature range of 100..600 C

  13. Peter Lang says:

    Gemasolar, $19/W and $31/W average

    I estimated the cost per W and per W average delivered based on the published figures for the total project cost, generating capacity, and expected average annual output (see sources below). The inputs and intermediate calculations results are (costs in 2010 AUD):

    Commissioned date May-11
    Capacity (nominal) 19.9 MW
    Energy storage 15 h
    Energy pa 110,000 MWh/a
    Capacity Factor 63%
    Capital cost (M) € 230
    Currency €
    Base cost year 2009
    Exchange rate 0.6
    Escalate costs by (% pa) 3% 3%
    Capital Cost A$395 A$ mil
    Cost per kW A$19,841 A$/kW
    Cost per average kW A$31,443 A$/kW average


  14. Peter Lang says:

    Ivanpah, California

    A presentation on the Ivanpah CSP project in the USA, currently under construction.

    Nameplate capacity is 370 MW peak electric.

    1,000,000 MWh/year. This means an average power production is 114 MW electric.
    Capacity factor is 0.31 or 31%.

    US $2,200 million

    the project comes in at more than $19/Watt average power delivered.

    This is 3x the cost of some recent nuclear powerplant builds that most environmentalists have accused of being prohibitively expensive.

    The heliostats used in the project weigh in at 30,000 tonnes. That’s 262 tons of heliostats per MW electric average. That’s just for the heliostats, not even the foundations, not to mention the tower and power block.

    The powerplant area that had to be bulldozed over is much larger than a nuclear reactor 20x the average (real) capacity (twin unit AP1000).

    • Günter Weber says:

      Very interesting to have a closer look at these numbers. And to put this in comparison to the recent French PV project, that I linked above.

      300 MW of PV in French you can get for 360 million bugs (let’s put Euro and Dollar on par). It will generate 0.3 TWh per year.
      The 360 MW CSP plant in US costs 2200 million bugs and delievers 1 TWh.

      For the running costs, I have no numbers but I think it is reasonable to assume that they are significantly lower for the PV plant than for the CSP plant.

      -> CSP is absolutely dead. (Take into account that the French PV plant located at the same place as the CSP plant would produce a lot more than 0.3 TWh/year.) If you really want to have the storage/delayed production option of CSP, you could probably get it cheaper now with a PV->Heat->Electricity scheme (of course, still much to expensive to be realistic option).

      • Peter Lang says:

        For comparison, a 300 MW nuclear plant would produce 2.4 TWh, i.e. 8 times as much as the PV plant.

        Importantly, Ivanpah costs $19/W average, whereas the nuclear plant costs about $4.2/W average (based on the price of the 5600 MW plant currently being built in UAE by a Korean consortium. Therefore, the CSP plant is 4.5 times more expensive than the nuclear plant per average power delivered.

        You didn’t state the cost of the French plant in meaningful numbers, so I can’t compare that. But remember that PV plants need to include the cost of the back up generators and grid costs.

        Also recall that adding PV to the French electricity system will increase not decrease the emissions intensity of France’s electricity system: http://judithcurry.com/2015/11/29/deep-de-carbonisation-of-electricity-grids/

        • Peter Lang says:

          The French PV plant reportedly cost $382m for 300 MW; i.e. $1.27/W. At 15% capacity factor the cost of average power delivered is $8.5/W average – i.e twice the cost of average power from the UAE nuclear plant.

          However, that’s not a fair comparison because the PV plant is not dispatchable. To make them comparable we need to add the cost of the backup. Let’s assume gas at $1/W. The corrected Average cost figures are:

          The French PV plant plus gas backup cost $382m = $300m for 300 MW dispatchable capacity; i.e. $2.27/W. At 15% capacity factor the cost of average power delivered is $15.2/W average – i.e 3.6x the cost of average power from the UAE nuclear plant.

          Renewables are not even close to being viable, without huge incentives.

          • Günter Weber says:


            I fully agree that relying on intermittent renewables production is not the cheapest way to generate electricity. (Maybe like buying organic food is not the cheapest way to get your 2000 calories per day.)

            But you are grossly exaggerating the costs in your example. In most industrialized countries the backup is already in place. There is no need for building the complete fleet of power plants, transmission lines, etc. from scratch. So, the incremental costs of integrating PV, CSP and wind is much lower.

            On the other hand you take the (estimated) costs of new nuclear power plants in a probably very nuclear-friendly regulatory environment (Arabic Emirates – why not take North Korea?) as granted. How costs develop in a more nuclear-critical environment, we can see in UK, Finnland and France (ok, maybe the EPR is just a shitty reactor design?).

            In general, I want to make one important point: For an industrialized economy, it plays no big role if electricity is generated for 50 or 100 bugs per MWh on average. What is really important is that the electricity supply is as free of any risks (also political ones!) as possible. In that respect renewables a preferable (if not prohibitive expensive).

          • Peter Lang says:

            But you are grossly exaggerating the costs in your example. In most industrialized countries the backup is already in place. There is no need for building the complete fleet of power plants, transmission lines, etc. from scratch. So, the incremental costs of integrating PV, CSP and wind is much lower.

            That’s disingenuous for at least two reasons:
            First – your comment refers to low penetration levels. You can add intermittent renewables at low penetration with small cost but as penetration increases the costs become huge, actually prohibitive. Try for yourself projecting the additional costs of the grid to 50% and 80% intermittent renewables using the OECD/NEA report summarised here: http://www.energyinachangingclimate.info/Counting%20the%20hidden%20costs%20of%20energy.pdf

            Second – Assets last for a period then have to be replaced. You have to compare on the basis of the LCOE using the full life time cost of the technologies. Comparisons have to be done on a rational, properly comparable basis.

            Solar and wind power are not sustainable http://bravenewclimate.com/2014/08/22/catch-22-of-energy-storage/ , cannot do the job and would be prohibitively expensive at the levels of penetration Germany is advocating.

          • Willem Post says:

            Another item overlooked regarding variable wind and solar energy is the future lack of rotational system inertia.

            Fossil fuel, hydro and nuclear plants typically have large turbines, synchronous to the power system, which provide plentiful system inertia for free whenever operating. However, increasing wind and solar energy percentages on the grid, which do not provide system inertia, may result in a power system that is lacking in system inertia, i.e., it becomes unstable. There may be a practical limit to wind and solar energy on the grid. With interconnections to other grids that still have plentiful system inertia, their inertia may be “borrowed”, when one’s own system is lacking in system inertia.

            Example of German and Danish RE Exports: During windy periods, mostly at night when demand is low, Germany and Denmark, to avoid/minimize grid instabilities, due to too little rotational inertia, and/or too little on-line, flexible, ramping capacity, need to export their excess wind energy to the Netherlands, Norway, etc., at near-zero, wholesale prices.

            The value of that energy, less transmission losses, to the Netherlands is the avoided cost of the gas, about 2.5 eurocent/kWh, the value to Norway is near zero, or negative, as Norway, 98% hydro, i.e., plentiful rotational inertia, merely reduces the water flow through some of its hydro turbines. The German and Danish grid connections to foreign grids need to be a mix of HVDC and HVAC lines, because HVDC lines would not transmit the stabilizing function of rotational inertia.

            This excess ENERGIEWENDE, offshore/onshore wind energy has a subsidized average cost of 15 – 20 c/kWh. As Germany and Denmark further build out their wind turbine capacities, these export energy quantities would increase, as they have during at least the past 5 years. RE aficionados crowing about Germany’s grid being stable, even with high variable wind energy percentages on the grid, likely have no idea how that stability was accomplished.

          • gweberbv says:


            also renewable producers can actively support grid stability. You just have to throw a very little amount of money to the problem if it becomes crucial.


            If I remeber correctly, the wind generators used in the El Hierro project already have this capability.

          • gweberbv says:


            I agree that the goal of 80% (or even 100%) renewable electricity production will not be reached. At around 60% most industrial countries will get stuck. The technical challenges to integrate intermittent energy producers into the system will set a certain limit.

            I also agree that there are more cost efficient ways of savong CO2/fossil fuels. Just think about insulation of buildings (at least for countries north of the Alpes). Or a properly working cap-and-trade scheme.

            However, I strongly disagree that a comparison of the ‘true costs’ of PV and wind compared to FF plants (or nuclear) can be done. On the one hand you have no idea about the FF prices in 10 years from now, not even to speak of 20 or 30 years. Last 10 years saw a factor 3 variation in the oil price. You can plug in any number. On the other hand we have no reliable estimate of the average lifetime of renewable plants yet. The wind generators and solar cells produced 20 years ago have nothing to do with what is installed today. And for nuclear, if western countries fail to ramp up again NPP projects, we might end up in a situation where only very few suppliers in the world have steady experience in building such plants. I doubt that the bigger western countries would like to rely on Russia or China.

          • Hinckley is supposedly going to cost 24.5 billion pounds for 2.7 GW of capacity. With an exchange rate of 1.5 and a capacity factor of 0.9, this works out as $15 per average Watt.

            Personally, I think carbon costs should be zero, near term there are benefits, and the long term should be discounted heavily.

            The climate policy I want is focused on sensible technology development and therefore enabling options.

            Hinckley is not about sensible technology development. 40 years ago nuclear power plants got built very rapidly, much more cheaply. If we ever needed or wanted to repeat that feat, we would need to forget any lessons learnt in the construction of Hinckley, not build on them.

            Solar thermal plants have a lot of sensible technology development in them, on the other hand, which will make construction of future solar thermal plants cheaper. And in addition, it is a proving count for innovative storage and heat to X conversion technologies, which in my opinion should prove more useful for a potential nuclear future than the little to be learnt from Hinckley (because most of that learning would be related to stupid overengineering of safety features).

        • Willem Post says:


          Here are some real-world numbers of JUST THE FUEL COST of back-up.

          The Wheatley study of the Irish grid shows: Wind energy CO2 reduction effectiveness = (CO2 intensity, metric ton/MWh, with wind)/(CO2 intensity with no wind) = (0.279, @ 17% wind)/(0.53, @ no wind) = 0.526, based on ¼-hour, operating data of each generator on the Irish grid, as collected by SEMO.

          If 17% wind energy, ideal world wind energy promoters typically claim a 17% reduction in CO2, i.e., 83% is left over.

          If 17% wind energy, real world performance data of the Irish grid shows a 0.526 x 17% = 8.94% reduction, i.e., 91.06% is left over.

          What applied to the Irish grid would apply to the New England grid as well, unless the balancing is done with hydro, a la Denmark.

          Europe is facing the same problem, but it is stuck with mostly gas turbine balancing, as it does not have nearly enough hydro capacity for balancing.

          Fuel and CO2 Reductions Less Than Claimed: If we assume, at zero wind energy, the gas turbines produce 100 kWh of electricity requiring 100 x 3413/0.5 = 682,600 Btu of gas (at an average efficiency of 0.50), then 682600 x 117/1000000 = 79.864 lb CO2 are emitted.

          According to wind proponents, at 17% wind energy, 83 kWh is produced requiring 83 x 3413/0.50 = 566,558 Btu of gas, which emits 566558 x 117/1000000 = 66.287 lb CO2, for an ideal world emission reduction of 13.577 lb CO2.

          In the real world, the CO2 reduction is 13.577 x 0.526 = 7.144 lb CO2, for a remaining emission of 79.864 – 7.144 = 72.723 lb CO2, which would be emitted by 621,560 Btu of gas; 621560 x (117/1000000) = 72.723 lb CO2.

          To produce 83 kWh with 621,560 Btu of gas, the turbine efficiency would need to be 83 x 3413/621560 = 0.4558, for a turbine efficiency reduction of 100 x (1 – 0.4558/0.50) = 8.85%.

          Below is a summary:

          Ideal World…………………………..Btu…………CO2, lb…….Turbine Efficiency
          No Wind gas generation………..682,600………79.864……………0.5000
          17% Wind gas generation……..566,558……….66.287…………..0.5000

          Real World
          17% Wind gas generation……..621,560……….72.723…………..0.4558

          Actually, Ireland’s turbines produce much more than 100 kWh in a year, but whatever they produce is at a reduced efficiency, courtesy of integrating variable wind energy.

          For example, in 2013, natural gas was 2098 ktoe/4382 ktoe = 48% of the energy for electricity generation; see SEIA report. This likely included 2098 – 2098/1.0855 = 171 ktoe for balancing wind energy, which had a CO2 emission of about 171 x 39653 million x 117/million = 791.4 million lb. This was at least 791.4 million lb of CO2 emission reduction that did not take place, because of less efficient operation of the balancing gas turbines.

          The cost of the gas, at $10/million Btu, was about 171 x 39653 million x $10/million = $67.6 million; it is likely there were other costs, such as increased wear and tear. This was at least $67.6 million of gas cost reduction that did not take place, because of less efficient operation of the balancing gas turbines.

          In 2013, the fuel cost of wind energy balancing was 5,872,100,000 kWh of wind energy/$67.6 million = 1.152 c/kWh, which would become greater as more wind turbine systems are added.

          It must be a real downer for the Irish people, after making the investments to build out wind turbine systems and despoiling the visuals of much of their country, to find out the reductions of CO2 emissions and of imported gas costs, at 17% wind energy, are about 52.6% of what was promised*, and, as more wind turbine systems are added, that percentage would decrease even more!!

          *Not included are the embedded CO2 emissions for build-outs of flexible generation adequacy, grid system adequacy, and storage system adequacy to accommodate the variable wind (and solar) energy, plus all or part of their O&M CO2 emissions during their operating lives; in case of storage adequacy, all of O&M CO2 emissions, because high wind and solar energy percentages on the grid could not exist without storage adequacy.

          NOTE: Gas turbine plant efficiencies are less at part load outputs. If gas turbines plants have to perform peaking, filling-in and balancing, due to variable, intermittent wind and solar energy on the grid, they generally operate at varying and lower outputs and with more start/stops. Such operation is less efficient than at steady and higher outputs and with fewer start/stops, just as with a car. Operation is unstable below 40%, hence the practical limit is about 50%, which limits the ramping range from 50% to 100%. Here is an example:
          Simple Cycle………………….100%……….38%……………..40%………….26%
          Combined Cycle……………..100%……….55%……………..40%………….47%


          Australia’s Power System: The Wheatley report states, with 4.5% wind energy on the grid, CO2 reductions were about 3.5%, which means the effectiveness was about 3.5/4.5 = 78% in 2014. The Wheatley report states, if wind energy were 9%, it would be about 70%. By extrapolation, if wind energy were 13.5%, it would be about 62%, and at 18%, it would be about 54%, i.e., the more wind energy, the less its effectiveness reducing CO2 emissions and fuel consumption. This would be similar to the effectiveness of 52.6% at 17% wind energy of Ireland’s power system. The laws of physics apply to Ireland, Australia, etc.


          • gweberbv says:


            in the report for Australia, about half of the 20% efficiency loss (at 4.5% penetration) is attributed to higher transmission losses. Probably because the wind generators ar far away from the users of electricity. I doubt that you will find nearly identical numbers for every country/electricity system.

            However, the relatively poor efficency in ireland is remarkable. One should really investigate this issue.

            In Germany hard coal is used for ramping while gas, nuclear and lignite are nearly constant. So, maybe the wind efficiency is a little bit higher here. However, penetration is still lower than in Ireland. It was about 15% in the first half of 2015.

          • Peter Lang says:


            Your just making uninformed, comments.

            “However, the relatively poor efficiency in ireland is remarkable. One should really investigate this issue.”

            First point: it’s not “efficiency” it’s “CO2 abatement effectiveness”. And, yes, you should certainly do some study of the subject if you want to have your comments taken seriously. Clearly you know nothing about it at the moment.

          • Peter Lang says:


            “At around 60% most industrial countries will get stuck.”

            Renewables are not economic in most electricity grids in industrial economies at even moderate penetration levels. Instead of making many unsupported and mostly irrelevant assertions, why don’t you compare the cost of electricity and CO2 abatement cost of an electricity system with mostly nuclear versus mostly non-hydro renewable energy using methodology that gives properly comparable figures. This is an example of how to do it: see summary in Figure 6 (also see Figures 5 and 7) http://citeseerx.ist.psu.edu/viewdoc/download?doi=

          • gweberbv says:


            maybe it is impossible for you to accept this but in an environment where very few NPP are being build and the public oppinion is extremely sceptical about this technology, it makes absolutely no sense to dream about things that will never happen.

            Even if EDF/Areva were offering Germany 20 EPRs for free and would also take over the O&M for the next 100 years, they would find the door being shut. But in reality EPR needs something like 10000 bugs per kWh and takes 10 years or more for costruction. And better hope to have no Fukushima-like event in between. Because if that happens, public oppinion might force you to close the site and write-off that 10 billion plant.
            Be aware: the idustrial and high-technology heart-land of Germany (Baden-Württemberg, nearly 200 billion euros of exports per year) is run by a green governor!

            So, from the German perspective the costs of NPP are simply irrelevant. Like the costs of Kobe beef are for Hindu people.
            Nevertheless it is funny to see these NPP projects getting more costly and delayed each year.

            The question of the Energiewende is not, if it the cheapest option. Only to what extend it will be possible with respect to the laws of physics.

          • gweberbv says:

            Should read: 10000 bug per kW installed capacity.

      • robertok06 says:

        “300 MW of PV in French you can get for 360 million bugs (let’s put Euro and Dollar on par).
        –> It will generate 0.3 TWh per year.” <–

        Unfortunately (for my electricity bill) it will generate them only during daytime, and practically nothing during 4 long months, Nov to Feb.

        I suggest that at least here on this blog we DO learn something from the discussions and blogs posted, and keep in mind that the TWh generated by PV (or wind) ARE NOT equivalent AT ALL to TWh generated by dispatchable power stations, especially nuclear in France.

  15. I have always looked for sound and professional data on CSP performance.
    And I never found anything. So I am very glad about the article
    “A review of concentrated solar power (CSP) in Spain”.
    The people who spent billions of tax money on CSP should be kept responsible
    for making the performance of those plants publicly accessible.

    Please allow me to add some good news: simple and cheap CSP does
    exist, its called “Linear Mirror” http://www.isomorph-production.it.
    Even better: Linear Mirror systems can transform by means of
    solar heat cheap biomass (straw, hay, waste) to “solar carbon”.

    Until the failure of Desertec, Solar Millenium, Siemens Solar and
    Abingoa it seemed that solar technology is the enemy of fossil
    energy. It was said, that nowadays cars must be substituted by
    battery cars, for instance, or carbon power plants by PV.
    But that was untrue, solar technology goes very well with
    fossil technologies:
    Solar carbon can be transformed to gas or to liquid fuels,
    similar to what is practiced in South Africa with fossil carbon.
    Or it can be used directly in traditional carbon driven power plants
    instead of fossil carbon.
    As a result, going solar can be done without giving up the existing
    industries and technologies. Only difference: solar carbon is CO2 neutral.
    All relevant articles can be found again at http://www.isomorph-production.it.

    The seeming incompatability of traditional and solar industry was just
    a stupid business model of the big solar guys (the ones now bancrupt),
    who wanted to have it all. We can do better.

    with best regards
    Hans Grassmann

  16. Jason Deign says:

    I can’t claim to have any technical experience on this topic, although I thought it was very odd of Andrews to use Spanish data in a review of CSP+storage when, to my knowledge, only a couple of plants (Andasol and Gemasolar) actually have molten salt storage.

    The rest are all an earlier generation of plants that lack storage and are about as good at providing baseload power as PV, albeit more expensive (although it should be pointed out that when they were commissioned PV wasn’t that cheap either, so there was a better rationale for developing both technologies in tandem).

    The other thing to bear in mind (and again, I don’t claim to be a technical expert) is that the solar radiation levels in Chile, South Africa and MENA might make CSP plants a lot more efficient there than they are in Spain.

    Currently the best comparison of CSP and PV that I’m aware of was in the latest Chilean energy tender, where PV came in at $64 per megawatt-hour and CSP was $97 per megawatt-hour–around 50% more, sure, but not prohibitively so (http://www.greentechmedia.com/articles/read/tender-outcome-adds-700-mw-to-chilean-solar-outlook).

    Also, the Chilean tender PV didn’t include storage; a 2015 Fraunhofer study claims CSP with molten salt is cheaper than PV with batteries (in Chile, I presume; see slide 11 in http://www.fraunhofer.cl/content/dam/chile/en/documents/6-Fluri-Fraunhofer-Chile-150527.pdf?utm_source=Energy+Storage+Report&utm_campaign=9d894c902d-ESR_2_10_1210_2_2012&utm_medium=email&utm_term=0_bd57f7e9aa-9d894c902d-34284857).

    The real cost issues for CSP, as far as I can see, are that a) you have to build a big plant for it to be efficient, b) that costs a lot of money, which has to all be invested upfront before you get to see any return and c) there aren’t that many CSP plants out there, so it looks like a risky proposition to sink a lot of money into.

    Put another way, I get the impression that CSP is a bit like nuclear–you can argue what the generation costs will be like but few investors are going to take the risk at a time when energy models are shifting to distributed generation and people like Sonnenbatterie and LichtBlick are building virtual power plants out of assets paid for by consumers.

    • Euan Mearns says:

      Jason, Roger is away this week. He may answer later. But the reason he chose Spain was because the data were available at Red Electrica, and I had just been downloading it. My view on the storage element is that storage to cover the diurnal load and demand cycle is future fantasy. Solar is currently useful to fill daily demand peaks and should be used immediately to do so without adding cost and inefficiencies of storage.

  17. Graeme No.3 says:

    the ‘cure’ for the intermittency of renewables is always storage. When there is an excess of generation it is sold cheaply, indeed at times customers are paid to take it. Storage will ‘solve’ that by saving that ‘free power’. Somehow the cost of batteries (or other methods) is NEVER costed into the budget. Nor for some reason is it ever explained that storage uses electricity, it doesn’t magically generate it. (Nor is it the done thing to mention that at 75% efficiency storage pushes the cost up by a third – 4 units in, 3 units out).

    Yes, the low energy density of renewable sources causes the initial capital cost to be much higher, and as it has to be paid for then the cost of renewables will be higher than conventional sources. There is also the cost of maintenance, which can cost as much as 20% of revenues for wind farms in Germany, and the cost of shorter lifetimes with renewables. PV degenerates steadily causing output to drop with time, so up goes the cost (same or higher expense over lower units of output.) Wind farms seems to need major work after about 9 years.

  18. Pingback: Notizie dalla fine del mondo: Molise, Darth Trump, bail-in, crisi di liquidità in Medio Oriente, esternalità positive tra Porno e Fallout 4, ISIS, cultura aziendale in VW e altre amenità della settimana | Liberali da Strapazzo

Comments are closed.