Comparing Solar Technologies to Ausra's Kimberlina Solar Thermal Plant

Earlier this year, on October 27th, Ausra commissioned their first solar thermal pilot plant, a 5.0 megawatt facility located in Kimberlina, just north of Bakersfield, California. It is the first solar thermal power facility to be commissioned in California in over 20 years – significantly, the approximately 350 megawatts of solar thermal power installed back in the 1980′s are all still operating.

There are three basic types of utility scale solar thermal power, all of which have strengths and weaknesses and all of which currently compete to become the most cost effective version. Bright Source Energy has been working on an improved “power tower” design, where a field of two-axis tracking mirrors (which can be installed on single poles placed on unimproved ground) reflect sunlight onto a single boiler. The advantages of this design are less site preparation, no transfer fluid plumbing required in the solar field, and a much hotter boiler which can make condensation and reuse of the steam on the backside of the turbine more cost-effective. Bright Source commissioned a 1.5 megawatt pilot plant in Israel earlier in 2008, and has projects in the pipeline in California.

Acciona Energy’s 64 megawatt plant, located near the Hoover Dam in Nevada and commissioned earlier this year, uses the “parabolic trough” design, where the solar field consists of single-axis trough-shaped parabolic tracking mirrors that turn from east to west each day, reflecting sunlight onto a tube containing heat transfer fluid that runs lengthwise at the focal point above each trough. This heated fluid is collected from the solar field and fed to a central steam turbine.

Ausra’s design, unlike the well-established power tower and parabolic trough designs, combines attributes from each of them, and adds a few unique features. Similar to the parabolic trough design, Ausra runs heat transfer fluid into the solar field into linear heat collection tubes located at the focal point of the reflectors. But the reflectors are only slightly concave, and the heat collection tubes are located further above the reflectors. This innovation allows one heat collection tube to sit at the focal point of several mirrors, allowing higher temperatures and greatly reducing the amount of plumbing required to go into the field.

Two lines from Ausra’s Kimberlina solar field – note how several mirrors
share one collector tube, which is positioned well above the collectors.
(Photo: Ausra)

One of the surest ways to determine which design is most efficient is to compare prices. Another interesting factor is the amount of land required, since concern over land consumption is a common criticism of utility scale solar power. It is probably too early to determine Bright Source’s costs or land consumption per megawatt, since their pilot plant is relatively small, but data is in on Acciona and Ausra.

Acciona’s Nevada Solar One plant cost $260 million and produces 64 megawatts at full output, a cost of $4.1 million per megawatt. Ausra’s Kimberlina plant cost $15 million and produces 5.0 megawatts, a cost of $3.0 million per megawatt. Ausra’s planned Carrizo plant, intended to produce 177 megawatts at a cost of $500 million, is estimated to come in at $2.8 million per megawatt. For comparison, Optisolar’s utility scale Sarnia solar field in Ontario, Canada, using thin-film photovoltaics, cost $300 million and produces 40 megawatts at full output, at a cost of 7.5 million per megawatt. Of course, when calculating cost per kilowatt hour, you have to take into account the “capacity factor,” the full-output-equivalent hours per year divided by the total hours in a year. By this measurement, Acciona’s desert plant has a capacity factor of 23%. Ausra’s estimated capacity factor, in the only slightly less hot and sunny southern San Joaquin Valley is between 18% and 22%. One can only imagine the capacity factor – not disclosed – for Optisolar’s plant located in Canada is significantly less than this.

In terms of land consumption, the comparisons are also interesting. Ausra’s Kimberlina pilot plant, at 10 acres and 5.0 megawatts, generates 0.5 megawatts per acre, or 320 megawatts per square mile. Their Carrizo plant will consume about 550 acres, generating 177 megawatts, which means at scale their space efficiency estimate drops to 206 megawatts per square mile. But that is actually very good by comparison.

Acciona’s Nevada Solar One plant consumes 400 acres and generates 64 megawatts – equating to only 102 megawatts per square mile. And Optisolar’s thin-film solar field consumes 902 acres to generate 40 megawatts, a paltry 27 megawatts per square mile.

When you consider megawatt-hours per year per square mile, the capacity factor comes into play. It is unlikely you will get more than about 1,200 full-sun-equivalent hours per year in Ontario, Canada, which equates to a capacity factor for Optisolar’s Sarnia field of 14%, which in-turn equates to 34,000 megawatt-hours per square mile per year. By contrast, Acciona’s Nevada Solar One, with 2,000 full-sun-equivalent hours per year can generate 204,000 megawatt-hours per year per square mile, and Ausra’s Kimberlina plant, at a slightly lower 1,800 full-sun-equivalent hours per year, but a much higher output of 320 megawatts per square mile in full sun, can generate an impressive 576,000 megawatt-hours per square mile per year. Ausra’s planned Carrizo plant, at full scale, projects a somewhat lower 206 megawatts per square mile in full sun, but that still equates to 370,000 megawatt-hours per square mile per year.

Without going into cost per kilowatt-hour derivations in this analysis, it is probably fair to say utility scale solar thermal plants such as Ausra is developing have a chance to compete with conventional energy on a level playing field – given they are already coming in at $3.0 million per megawatt and have capacity factors exceeding 20%. A thin-film energy plant at scale, at $7.5 million per megawatt, and a capacity factor that is likely lower than 15%, is going to have a much tougher challenge to commercially compete with conventional energy. And the fact that Ausra’s Carrizo solar field, megawatt-hour vs. megawatt-hour, will consume literally ten times less land than Optisolar’s Sarnia solar field, should not be lost on anyone considering desirable options for utility scale solar development. With solar energy, where you build it – and what technology you employ – dramatically impacts the costs and the land consumption necessary.

In a recent email received from Ausra’s Chief Development officer, Rob Morgan, he states “Ausra’s core technology is the most land use efficient solar technology in operation today.” The facts would seem to bear this out, when comparing parabolic trough and thin film technology to Ausra’s novel hybrid design. It will be interesting to see how Bright Source’s power tower technology compares, when that data becomes available.

Some Related Posts: Optisolar’s Thin Film, Bright Source’s Power Tower, Photovoltaic vs. Thermal, Utility Scale Photovoltaics, Acciona’s Nevada Solar One, Ausra’s Solar Thermal Power

18 Responses to “Comparing Solar Technologies to Ausra's Kimberlina Solar Thermal Plant”
  1. Pat McGuinness says:

    Re Ausra:your article talks of ‘heat transfer fluid’, but I think you’ll find that no such fluid is deployed.

    Water in the elevated tubes is flashed directly to steam and dispatched immediately to the turbines, thus skipping the ‘heat transfer’ stage necessary with more traditional trough-based systems.

  2. Ed Ring says:

    Pat: You’re right, one of Ausra’s greatest innovations is they are using a single loop, where the water collects heat from the solar field and becomes steam to drive the turbines. I should have made that clear in the report.

  3. Al Fin says:

    A large solar thermal plant in Ontario? Whose bright idea was that? Nuclear power makes far more sense in the frozen north country than solar. Consider that in the winter months, high latitude solar collectors get very little sunlight for several predictable reasons. The triumph of ideology over survival instinct may be the death of us all–especially with a carbon hysteric in the Oval Office.

  4. fred says:

    Al Fin,
    Obviously you didn’t read the article closely…the solar facility in Ontario is not a solar thermal project: “For comparison, Optisolar’s utility scale Sarnia solar field in Ontario, Canada, using thin-film photovoltaics,”. Your observations about insolation in that part of Canada are the reasons why thin film was selected.
    CSP must be part of the US’s generation portfolio…along with wind, tidal, geothermal, natural gas and yes, coal too. Going back to the exclusive use of fossil fuels is simply stupid.

  5. Cyril R. says:

    What I use for levelised cost is:

    (1/capacity factor)*($/Watt-peak total investment)+(variable cost in cents/kWh) = levelised cost in cents/kWh

    So a 3 dollar per Watt plant with 0.20 capacity factor gets 15 cents/kWh. Operational costs are typically low for large solar thermal electric plants, and can be even lower for large PV plants, so are unlikely to add more than a cent to this estimate.

  6. antoine says:

    Optisolar: $300 million for 40 MW = $7.5 million/MW, not 7500!
    Best regards.
    Antoine – fixed; it was a typo… Thank you! Editor

  7. Rob Moss says:

    I am not sure your data is right. If you are getting 0.5 megawatts per acre then this must refer to megawatts of steam, or megawatts of electricty but only at the time of peak irradience; but not megawatts of electricity; period. It is possible that the figure from Ausra is for steam but the journalist has interpreted it to mean electical power. one megawatt implies one million watts power 24/7 – unless this is clarified.

    Also PLEASE can you people stop giving data in acres, bushels, etc. Or do you only want to speak to ‘”island America”and not the whole World. I am British, from a farm, and know that even British farming magazines have every measure of land or weight in international units. If non-international units are used (mostly orignating from the Saxon era) then the Ha, or metric tonne units will be given alongside. We still love our oldfashioned weights and measures – but not in journals. I think the USA us the only developed country that still does this. The fact that USDA reports grain yeilds in bushels per acre, is mind boggling nonesense. I thank God that I was trained as a scientist and always wonder, slack jawed with amazement, how US scientists put up with this and continue to popagate this practice themselves.

  8. Ed Ring says:

    Rob – Kimberlina’s energy production of 0.5 megawatts per acre is at full output. We could have made that more explicit, the term “full output” is used later in the report. We tend to use megawatt-hours (or gigawatt-years) or simply the term “constant megawatts) to report output over time. During peak irradience the sun delivers about 100 watts per square foot, and at 44,000 square feet per acre that equates to 4.4 megawatts per acre. For Kimberlina to produce 0.5 megawatts per acre implies an efficiency of 11%, which is quite reasonable for electricity output from the steam turbine, not thermal input from the solar field.

    As for your comment regarding the metric system, to simply say that we agree with you would be a huge understatement. But we have a large percentage of readers in the USA, and to go exclusively metric would be difficult. We try to provide data in both units but we have no policy – perhaps we should. The imperial system of weights and measures is probably one of the most archaic and ridiculous legacies of the past imaginable, and undoubtedly condemning Americans to its use makes providing useful information and insights on energy, water, and other resource & environment issues all the more challenging.

  9. Scott R. says:

    Long ago I adopted a personal policy of using metric measurement wherever possible. My collegues here in the USA are generally fine with this. Basically everyone knows how long a centimeter (for instance) is, and conversion is rarely necessary.
    This reminds me of a funny measurement unit-related story (bet you thought there were none of those ;>). Once on my friend Joe’s moving day from his 2nd floor apartment he shouted up at me from his car to toss down a stick he needed to brace a load. “How long does it need to be?”, I asked. Quckly he measured and shouted back up “Two cubits and a hand!”. I smiled, measured a suitable stick, and we went back to work :) .

  10. Golden Sun says:

    Thank you for writing this article. I have been following this project (albiet remote observation) from Santa Clara CA. I have been doing solar demonstrations (PV) with my solar trailer & DJ service. I’m excited about the CSP method for baseload purposes. It seems like a no brainer and a better tea kettle. I still like PV for distributed purposes. Your report has given me a key metric that I was missing from this project. I realise this project was more of a pilot demonstration, but is illustrative of the potential that CSP deployments represent (particularly near Bakersfield). I look forward to your report on Ausra’s Carrizo project. You might want to contrast this project with the nearby Diablo Canyon operation. Using your metric, I believe we can deliver 333MW of CSP for the same amount that the nuclear industry wants just to underwrite a construction loan Wall Street wants before a shovel turns dirt. Those are just the front-end costs. Nevermind operational and back-end costs (Waste disposal) Each nuclear plant in the U.S. has already received about a billion per plant in corporate welfare on plants that were built decades ago. It’s fair to assume that those same expenditures would be considerably higher in today’s dollars. Thanks again.

  11. Cyril R. says:

    The Ausra website clearly states that the steam output of the Kimberlina plant is 25 MWth while the electrical output is 5 MWe.

    That gives 20% steam to electrical efficiency.

    A big plant gets 30+ % steam to electrical easily; steam turbines, including the saturated steam turbines used in this CLFR technology, are most efficient at large scale.

    This puts the cost of a peaking plant down to 2000 USD/kWe (from 3000/kWe at small scale Kimberlina level).

    Big steam turbines, as well as cooling systems, feed pumps etc etc are also a lower cost per kWe installed, this probably reduces the cost to around 1500 USD/kWe or something. That’s nice for a peaking plant. It will really get interesting if they get their large scale storage system ready for prime time.

    These back on the envelope calcs are just to show what really big scale alone can do for cost effectiveness. But even at 3000/kWe that’s reasonable cost power for California.

  12. Cyril R. says:

    As for capacity factor, this is not fixed. Thinfilm facilities are typically non-tracker since they are lower efficiency (larger area = more expensive tracking, relatively).

    The best locations can get over 35% capacity factor for dual axis tracking. Single axis tracking can also get very high (over 30% in the best locations). Thinfilms are decent at converting diffused radiation (cloudy conditions). As mentioned before on this site, this gives a nice benefit, particularly for cloudy but reasonably sunny locations, such as Florida.

  13. R. W. Beck, Inc. is a premier national engineering-based management consulting firm that helps our clients in the Energy and Water & Waste Resource market sectors make better business decisions and accomplish their objectives. We do this by guiding our clients through the volatile nature of energy markets, and confronting the impacts of aging infrastructures, and solving economic and technical challenges.

    With more than 20 offices around the world we are always looking for qualified economists, engineers, and consultants, to fill positions to help our clients in the energy, utility, financial, water resources, and solid waste industries.

    Our differentiator is the quality of our people and the way we do business – with absolute commitment to objectivity and building trust. We blend our technical expertise with proven business acumen to deliver innovative solutions to our clients. And we have been doing it since 1942.

    We offer our employees an attractive total compensation package. Competitive, market-based wages may be augmented by additional profit sharing and spot bonuses, in addition to our comprehensive benefits solutions, tailored to fit every employees needs.
    We are often told that we are the engineering and consulting industry’s best kept secret – come check us out at
    Solar/Renewable Engineer– Boston, MA or Denver, CO
    Working in a team setting, responsibilities include managing projects and providing support to an existing consulting practice within renewable power.
    Responsibilities include: writing independent/owner engineering reviews and reports, managing projects and providing due diligence on various energy technologies including the use of concentrating solar power for electricity production. Other duties may include participating in reviewing operations and maintenance of facilities and witness performance testing of solar and more traditional types of energy facilities.
    Position Qualifications:
     B.S. in Engineering, preference mechanical engineering
     5 years experience in the engineering, construction monitoring, performance analysis/testing or operations and maintenance of solar/thermal or other renewable and/or traditional electric generation facilities
     Experience successful track record maintaining client relations and working with clients.
     Excellent oral and written communication skills.
     Ability to work independently or as part of a team.
     Ability to be a successful manager of a team using proven project management skills.
     Ability to successfully meet deadlines under pressure. Ability to successfully meet deadlines under pressure.
     Clients will include utility companies, private developers, legal and financial firms, commercial and industrial facility owners, and energy service companies
     The job may require as much as 25 percent travel
    R. W. Beck advances the business of infrastructure. As technically based business consultants, we are trusted advisors to thought leaders in the energy, water, and solid waste industries. We serve the financial community, municipal and private utilities, public agencies, government, and industry. Unlike traditional engineering firms, we fuse business and financial acumen with technical expertise to offer innovative planning, financial, and engineering solutions – providing insight with impact on every project with every client. Our engineers, economists, analysts, and business consultants are passionate about delivering solutions that foster our clients’ success and improve the communities where we all live and work

    Andrew Bright
    Sr. Corporate Recruiter
    RW Beck, Inc.
    1001 Fourth Avenue, Suite 2500
    Seattle, WA 98154-1004
    PH: 206.695.4754
    Cell: 360-393-0071

  14. atul patel says:

    which is lowcost energy either parabolaic or photovolataic
    is it possible to established 1mw solar thermal energy plant
    pls. reply me by mail only bcoz we r interested to produce 1mw electricity by solar
    which project is convenince for us and could u help us or need your suggation

  1. [...] made progress with “Ausra’s Kimberlina Solar Thermal Plant” by Ed Ring at EcoWorld but the credit crunch has slowed the development of planned multi-hundred [...]

  2. [...] made progress with “Ausra’s Kimberlina Solar Thermal Plant” by Ed Ring at EcoWorld but the credit crunch has slowed the development of planned multi-hundred [...]

  3. [...] made progress with “Ausra’s Kimberlina Solar Thermal Plant” by Ed Ring at EcoWorld but the credit crunch has slowed the development of planned multi-hundred [...]

  4. [...] made progress with “Ausra’s Kimberlina Solar Thermal Plant” by Ed Ring at EcoWorld but the credit crunch has slowed the development of planned multi-hundred [...]

Leave a Reply

You must be logged in to post a comment.