SolarReserve still falling short at flagship solar tower project

Print Friendly, PDF & Email

SolarReserve’s flagship solar tower and storage project is still falling short of design targets. Does it deserve public money?

share
Print Friendly, PDF & Email

On January 18, RenewEconomy reported that SolarReserve’s $650 million, 150MW concentrating solar thermal plant to be located near Port Augusta has been granted development approval.

We question whether the technology is sufficiently proven to justify building this project. If SolarReserve wants to use its own money, or that of a private sector bank, well good on them. Public money – the company is expected to get a $100 million grant from the federal government – is another thing.

SolarReserve’s flagship project is Crescent Dunes, a US$1 billion, 110MW first-of-its-kind project located, perhaps appropriately, near Las Vegas. Ground was broken in 2011 on the project.

The project has 10 hours molten salt storage but the relevant statistic is its design capacity factor of 51 per cent.

Not enough evidence that Crescent Dunes works

Production from the facility was first reported in October, 2015, and out of 26 months since then the facility has been completely offline for 9 months. On only 3 months has it met the annualized design goal.

Figure 1 Crescent Dunes operating history. Source EIA, Wikipedia

SolarReserve management stated that the reason the facility was closed down for 8 consecutive months in 2016/2017 was a leak in the molten salt tank.

Leaving aside any questions of how that could happen, or why it should take 8 months to repair, we note that having started up again in July 2017 the facility did not reach design goal and as of November 2017, on the last data available, it appears to have virtually shut again.

It did not meet its design goal in any month after being restarted.

Leaving aside any questions of project economics we would simply say there is not enough positive evidence to suggest the technology works to the extent advertised.

As far as I am concerned I don’t really care what the company’s explanation of the problems are. If you want to put $100s of millions into a new technology you need enough evidence it works.

The only satisfactory evidence it works is production history of consistently meeting the design goal.

I’d note that there are at least three pumped hydro projects under development in South Australia, one of which Goats Hill, owned by Altura/Delta is only 12 km from Port Augusta with 8 hours storage and 220MW of power.

As I understand it Goats Hill is expected to receive development approval by mid 2018.

Given the evidence available from Crescent Dunes, my personal opinion is that Port Augusta should be put on hold until Cresent Dunes has shown it can operate at something like design capacity for at least  12 consecutive months.

David Leitch is principal of ITK. He was formerly a Utility Analyst for leading investment banks over the past 30 years. The views expressed are his own. Please note our new section, Energy Markets, which will include analysis from Leitch on the energy markets and broader energy issues. And also note our live generation widget, and the APVI solar contribution.

Print Friendly, PDF & Email

25 Comments
  1. GlennM 9 months ago

    Harsh, but hard to disagree with..

  2. Tony Hall 9 months ago

    Sometimes you have to take risks to advance the technology. Failure is not necessary a bad thing, as it closes the loop to improve.

    • Cray 9 months ago

      You can fail in your experiments on your own dollar, not mine.

  3. mick 9 months ago

    possibly value in drought proofing hydro?

  4. Collin 9 months ago

    Finally an article that tells how poor Solar Reserve’s technology really is working! Well done! $2 billion US investment ($1.5b in gov. loans) for a pitifully small amount of inconsistent power generation. In my opinion, their Ponzi scheme will collapse when politicians quit supporting it.

    • Barri Mundee 9 months ago

      It may well be a dud. If so never mind as battery and pumped hydro is proven.

  5. David McKay 9 months ago

    I understand they are in the process of raising the project finance. Arguably the most difficult process of any project. By the time this closes we may have 12 months of operational data? That graph will need to take a significant upward tick.
    Not sure anyone has a solid O&M cost model for circulating all that hot molten salt. There is always risk with any emerging technology & storage is the “holy grail” for renewable energy. Things are moving rapidly & the time to prove you deserve your place in the market is short.

    • Cray 9 months ago

      I have seen statements that Solar Reserve has 63 permanent staff at their Nevada solar plant. At 75,000 each that is almost $5 million per year in operations salaries. Probably double that in total costs. $10m likely per year.

      • Gary Rowbottom 9 months ago

        Not a bad thing if you come off the dole queue to be one of that 75. It would be great to know the full details of issues, but I can understand commercial confidentiality will hold sway. I can say that something like the leak in the molten salt tank could very easily happen, there are millions of components that have to be assembled correctly in a thermal power plant. Some are in areas where problems can shut the whole show down. The molten salt tanks are one of them in CSP plant. I don’t understand the length of time it was off line for, but if there is a time to rectify developmental issues it is in early life. No idea what the current issues may be, I hope they are not serious. The Port Augusta plant should benefit enormously from the learnings from Crescent Dunes. Are there production stats from Gemasolar plant in Spain available? That is only 20 MW but 15 hours storage, and I think the longest operating tower with storage plant – I think it was put on line in 2011.

      • Peter 7 months ago

        The basic problem is that there is so much hot salt (combination of sodium and potassium nitrate) that it takes about 3 months to either heat it up or presumably cool it down. So you are talking about a 6 months gap in generation every time you want to make a repair to a hot salt tank.

        You wouldn’t expect this to happen very often, but it will happen very occasionally. One circumvention would be to provide a third tank and set of piping so you can take any one tank out of service. It’s not a huge expense (it’s only a tank, albeit one capable of storing hot salt), though the Crescent Dunes layout may not have space for it, once built. A third tank (hot; cooler (but still molten); spare) may have been considered and rejected in the design because it wasn’t calculated that it would fail often enough to be worth duplicating.

        All other components, e.g. receiver pipework on top of the tower, you can just shunt all the salt to the tanks, and not generate while you fix the problem in a few days or weeks, but the tanks are different because you have to put all the molten salt somewhere!

  6. George Michaelson 9 months ago

    Geothermal has had similar roadblocks. Odd because we know it works. It works in NZ and it works in Iceland. But, when people tried to do it here, they ran out of money.

    I don’t think solar therman is a lemon in concept, I think we’re not being told about some specifics in context. Its an open question for me if its a lemon in practice.

    • Biff 9 months ago

      The geothermal resources in Iceland and NZ are much easier to access.

  7. BushAxe 9 months ago

    SIMEC ZEN is the fallback I’d say seeing that they have the Govt power contract now (I’m guessing they will make up any shortfalls from the CST with wind/solar PV).

  8. Ray Miller 9 months ago

    Interesting, one needs to go into any project with an open mind. It does point out the risks associated with the various strategies and technology leanings.
    The current strategy which is working, has small well designed and proven units like a PV module, Lithium battery cell, inverter unit and make millions of them, drives the price down, the risk is low and they can be distributed to get the highest value.
    So to try to complete with the failed large centralised model with all the risks, costs etc with the “hope” of lower costs when it clearly is at just the beginning of the technology cycle. We really would need major assurances and proof that the AU version has every chance of meeting a contracted reliability figure.
    Where you have high energy, large machines, complexity in a dynamic environment the risks are elevated, and if anything does go wrong it is even a bigger (expensive) problem.
    I’m not saying the project should not go ahead but like with SpaceX and the new Falcon Heavy rocket you use the best designers, rigorous testing and retesting to maximize your possibly of success, even if you have a glitch the data and leanings inform the next one.

    • Cray 9 months ago

      Solar works great in small applications until the sun goes down or behind a cloud. Never build critical infrastructure dependent on fickle weather!!

      • Ray Miller 9 months ago

        You’ve got the totally wrong idea, the whole planet has always worked on our solar and variable conditions. All our ecosystems have built on variability, it has been the height of human ignorance and arrogance which has tried to do the opposite.
        The problem with the Solar reserve project is not the scope of what it is planning to do but with the implementation of the technology, over the decades various solar thermal projects have been built and many learning’s, so the expectation is that in 2018, money invested into such a project should have a high expectation that the performance and reliability are high.
        As with the Tesla big battery, it was fully installed in a matter of months and has performed beyond expectations, which has resulted in more business for Tesla. So David is pointing out Solar Reserve Mark I has not performed, so why should they get our business? Or why should we shoulder the risk until they can prove they have sorted out the bugs? Or it may be more prudent to use proven (and reducing in cost every year) wind and solar PV to power a pumped hydro system/s?

        I’d be more worried about those grossly inefficient, highly polluting, fickle, poorly maintained, aging coal plants built last millennia.

        • Collin 9 months ago

          With advances in technology, you expect higher quality solutions for your investments. Quality in energy projects = consistency and lower cost. CSP is inefficient as is pumped hydro and other multi process systems. The reason coal plants are underappreciated is because there is a misunderstanding of the value of consistency over production. If you can’t reliably produce power 24/7/365, you shouldn’t be allowed to have any role in grid power.

          The Tesla “big battery” has an hour’s worth of juice on a small grid. Hardly a “huge success” unless you are dealing with poor quality power solutions.

    • David McKay 9 months ago

      CSP is a proven technology. The SEGS plants (300mW Parabolic Troughs) operated in California for some 30 years. No energy storage. There are other successful CSP projects globally that operate. This is pretty standard thermal power plant 101. Make saturated or super heated steam, run turbine, generate electricity, condense, repeat. Storage is the new & make or break technology. SR built a trial plant & ran this to prove the tech. There are at least 3 or 4 other Australian CSP/storage techs looking to enter the market, so the race may be on. Failure is a factor when you are pushing the development boundaries, however, failure for SR would be catastrophic, I believe.

      • Ray Miller 9 months ago

        Yes CSP is proven and Australia has been testing the waters over 50 years.
        I agree, failure of SR would be catastrophic in so many levels, which is why a high level of scrutiny over every aspect of the design, every mirror checked, every weld and all the materials used.
        I’ve personally had many failures of equipment, so called fit for purpose meant for Solar tracking which only lasts a season or two, so I know how difficult it is to have mechanical and electrical equipment in the elements with everything against you. But it can be done with the right planing, expertise and vision.

  9. Andrew Scott 9 months ago

    David,
    you are correct in having reservations about this proposal.
    It is not just the technology used by this proponent that is problematic.
    The concept as a whole – as applied to electricity generation – is flawed.

    We should think about the physics and engineering principles that are in play.

    There is the intrinsic inefficiency of boiling water to drive turbines (Carnot cycle etc) attached to spinning generators to produce electricity.

    There is the large sunshine concentrating system that is inherently complicated, expensive and potentially unreliable .

    There is the system for capturing the heat in some high temperature transfer medium at the focal point and moving it to storage remote from the hot spot – another technically demanding challenge for safe and reliable performance.

    Further complexity arises with the addition of Pumps and Heat exchangers to recover heat from the storage(s) and transfer it to another heat exchange location for boiling water for the turbines. And so the complexity goes on.

    How can this concept not be too complicated, too inefficient, too unreliable and too expensive to be viable for electricity supply?

    Solar thermal collection and storage in its simplest form can be highly valued for providing hot water for domestic, commercial and some industrial processes.

    Perhaps a simple variant for solar thermal collection storage and recovery can also be developed to supply super-heated air in very large quantities for some industrial processes. (eg hot blast air to an iron-making blast furnace.)
    It would be better to direct some of the Commonwealth funds to researching such options for development of solar thermal systems for direct thermal use, rather than squander it on the current proposal for electricity generation.

    • Gary Rowbottom 9 months ago

      Making electricity has long been a very complex undertaking, requiring the coordinated choreography of millions of components to enable something to happen when you turn on your Espresso machine etc. Sure it is complicated, and sure that means there is plenty that can go wrong. But thousands of thermal power stations round the world for over a hundred years say that it can be done and by and large works pretty well, and getting better. Those “complicating” factors of all those components that go whizz, whirr, hum and spin also mean a decent amount of jobs. I like that. We have to solve the emissions problem. The contracted price for the Aurora plants output is pretty competitive, especially remembering it is essentially fixed for 20 years. Of course CSP is not the only answer to the future decarbonised energy mix, but it may very well be one of them. Give it a chance.

  10. Cameron Pidgeon 9 months ago

    CSP with molten salt storage has been operating successfully in Spain since 2009 so I don’t know how it can be said that the technology is unproven. Scaling from the MW Spanish range to the GW US range may have been better done with multiple MW scale reflector-field-and-tower units instead of one big GW unit. The molten salt can be heated electrically when sunny, windy days elsewhere are producing excess. There is also the prospect for hybrid PV/Reflector panels with a PV layer in the panels collecting shorter wavelengths the reflector doesn’t direct to the tower, if the problem of getting layers of differing and chemically complicated layers to stick together at the tiny thickness needed can be solved. Its way to soon to dismiss this technology. I would say that that much funding shouldn’t be given without gov getting IP rights, government research (CSIRO?) partnerships etc.

  11. Dennis Abbott.. 9 months ago

    If those trailblazers whom installed expensive PV in the 1970’s, had not supported the PV Industry, would I have less toxic (no Arsenic dopants ) cheaper, lighter, more efficient PV on my roof today?
    No electrical generation method is perfect. CST in Australia has so much potential; grid electricity, Industrial uses, Ali and Steel smelting, desalination, soil and water remediation.
    In my eyes CST should be supported at Gov, Private sector and Public levels.

    • Tom 9 months ago

      There’s no arsenic in PV.
      Dopants are phosphorus and boron.

      • Dennis Abbott.. 9 months ago

        Sorry Tom my wording was not that clear.
        Early PV cells were expensive, heavy, did not make much electricity, contained toxic chemicals; arsenic etc..
        Today, the PV cells on my roof are more advanced and Boron doped.
        The point I am trying to make is we supported early PV and now its a winner. We need to replace coal and gas, CST with storage may be part of our future energy mix, CST has my support.

Comments are closed.