CleanTechnica is the #1 cleantech-focused
website
 in the world.


Clean Power CSIRO CSP with supercritical steam

Published on June 11th, 2014 | by Tina Casey

38

AUS Concentrating Solar Power Breakthrough Could Hit USA Shores

Share on Google+Share on RedditShare on StumbleUponTweet about this on TwitterShare on LinkedInShare on FacebookPin on PinterestDigg thisShare on TumblrBuffer this pageEmail this to someone

June 11th, 2014 by
 
We’ve been on a concentrating solar power/thermal energy storage tear this week, so let’s keep the ball rolling. The Australian science agency CSIRO is working on a super duper “supercritical” steam system for concentrating solar power plants, and in a lucky twist for us over here in the US it turns out that one of CSIRO’s partners in the project is Abengoa Solar.

We stand to gain because the partnership will complement Abengoa’s considerable experience in the field, and the company has just teamed up with our own Energy Department on an advanced new concentrating solar power/thermal energy storage system. The project is aimed specifically at bringing thermal energy storage technology into the competitive energy market.

CSIRO CSP with supercritical steam

CSP with supercritical steam courtesy of CSIRO.

More Concentrating Solar Power With Thermal Energy Storage!

The Energy Department’s advanced CSP/TES project comes under the National Renewable Energy Laboratory. The role of Abengoa Solar will be to manage the systems integration end of things and perform technical-financial analysis, all with the aim of keeping the cost of the technology down.

The cost factor is critical because CSP is typically more expensive than photovoltaic cell technology, leading some solar cell fans to look down upon CSP.

However, the “built-in” thermal energy storage feature of CSP is a key advantage (see yesterday’s post for a brief overview of CSP/TES). It enables CSP plants to keep generating clean electricity long after the sun goes down.

There’s also a ripple effect on other forms of clean energy tech, namely solar cells and wind turbines. By introducing more time-shifting capability, CSP/TES can create more space on the grid for technology that only functions when the sun shines or when the wind blows.

Abengoa is best known for its trough-style concentrating solar power technology, as illustrated by the company’s new Solana CSP plant with a molten salt thermal energy storage feature. Located in Arizona, Solana is billed as the largest parabolic trough CSP plant in the world.

Abengoa is also a global developer of tower-style CSP plants. Last year the company teamed with BrightSource on the Palen concentrating solar power project in California, which last time we checked will sport the world’s largest CSP towers (for the record, BrightSource is also behind the Ivanpah CSP in California).

Super-Duper Supercritical Steam For Concentrating Solar Power

Now, here’s where it gets interesting. Australia’s CSIRO (short for Commonwealth Scientific and Industrial Research Organisation) teamed up with Abengoa and the Australian Renewable Energy Agency to create a CSP breakthrough that consists of a solar powered system for generating super hot, pressurized “supercritical” steam.

The project apparently resulted in the highest temperatures ever achieved globally for steam generated by non-fossil sources.

Currently, solar thermal power plants in commercial operation use subcritical steam, which is hot enough but performs at a lower pressure. Boost the pressure and you boost efficiency, leading to lower costs.

The supercritical steam achievements means that solar energy could have the same rate of performance as coal or gas in power plants. When you combine that with thermal energy storage capability or with other advanced battery technology, the need for building new fossil fuel power plants pretty much evaporates.

Here’s CSIRO Energy Director Dr. Alex Wonhas waxing enthusiastically about the project:

It’s like breaking the sound barrier; this step change proves solar has the potential to compete with the peak performance capabilities of fossil fuel sources.

Instead of relying on burning fossil fuels to produce supercritical steam, this breakthrough demonstrates that the power plants of the future could instead be using the free, zero emission energy of the sun to achieve the same result.

In Australia, supercritical steam is a new thing even for fossil fuel power plants. According to CSIRO, 90 percent of the electricity in Australia is generated by fossil fuel, but only a few of those power plants use supercritical steam.

According to the CSIRO blog, the R&D team achieved a temperature of about 570° Celsius, which is hot enough to start melting an aluminum alloy. The pressure reached 23 megapascals, equivalent to a two-kilometer diver under the ocean’s surface.

 

The secret sauce for the achievement is a fully automated control system for fine-tuning the tracking movements of the heliostats (mirrors). That enables the system to focus the maximum amount of solar energy on the receiver without overheating it.

Oh, The Irony

The team cautions that additional R&D is needed to bring the technology to the marketplace, but in the meantime our friends over at Think Progress have tipped us that Australian Prime Minister Tony Abbott is touring around the US.

Maybe he’s on his way over to the National Renewable Energy Laboratory to see what CSIRO partner Abengoa is up to, but according to Think Progress his first stop was in Canada to meet with Canadian Prime Minister Stephen Harper, apparently to keep the global market for fossil fuels humming right along:

…Abbott “flagged intentions to build a new center-right alliance led by Canada, Britain and Australia along with India and New Zealand,” in an effort to “dismantle global moves to introduce carbon pricing, and undermine a push by U.S. President Barack Obama to push the case for action through forums such as the G20.

If Abbott is successful, that would pretty much cut the legs out from under CSIRO’s project in the competitive global energy marketplace, supercritical steam or no supercritical steam. He’s scheduled to meet with President Obama later this week so stay tuned.

Follow me on Twitter and Google+.

Keep up to date with all the hottest cleantech news by subscribing to our (free) cleantech newsletter, or keep an eye on sector-specific news by getting our (also free) solar energy newsletter, electric vehicle newsletter, or wind energy newsletter.

Print Friendly

Share on Google+Share on RedditShare on StumbleUponTweet about this on TwitterShare on LinkedInShare on FacebookPin on PinterestDigg thisShare on TumblrBuffer this pageEmail this to someone

Tags: , , ,


About the Author

Tina Casey specializes in military and corporate sustainability, advanced technology, emerging materials, biofuels, and water and wastewater issues. Tina’s articles are reposted frequently on Reuters, Scientific American, and many other sites. Views expressed are her own. Follow her on Twitter @TinaMCasey and Google+.



  • Hasan Basri SENAL

    3, July, 2009, in the waste, elimination of the excavation and back with built my business to create renewable energy innovation. Renewable energy also redesigned with a new component Policarbonsilisy my agent; I’ve managed to collect solar energy falling on the earth. able to directly generate electricity. At the same time I had the same amount of cold. Ending them when summed energy in the battery was obtained. Now I need Finance. Enlarge all over the world to send much more. Can you help for the future of our world.
    Bank: Bank of Turkey
    Swift Code: ISBKTRIS
    IBAN No: TR52 0006 4000 0021 0113 3783 91
    Thank you to everyone.
    Sincerely,
    Hasan Basri SENAL.

  • Ronald Brakels

    Get it while it’s hot! Because the CSIRO kept doing its job, producing evidence that climate change is real and having serious effects on the country and developing technologies that threaten the dominance of coal power, the Abbott government has cut the CSIRO budget by around 10% to punish them for acting as though reality is a real thing and not something that will go away if you pretend hard enough that it doesn’t exist. So enjoy this little breakthrough because there may not be too many more.

    • Ronald Brakels

      Sorry, since the CSIRO also gets money from selling diet books and breakfast cereal (don’t ask), the Coalition is actually giving it more of a 20% cut in government funding. So perhaps Australians can look forward to further breakfast cereal innovaton while they watch the oceans die and the koala become extinct.

      • Zer0Sum

        When Australians realise the the budgets cuts are a savage attack on the most needy while the wealthy business elite continue to absorb all of the generated credit that the Government is magically creating out of thin air forcing the aforementioned needy to pay for it all with increased taxes, environmental degradation and climate change they will get rid of Abbot and his corporate cronies.

  • Bob_Wallace

    Just thinking…

    The role for CSP seems to be to store heat and sell power when PV and wind aren’t producing. There might be another role – as deep backup when both wind and sunshine are running short for extended times.

    In the Budischak, et al. study they found it best to use a small amount of NG generation in order to avoid building storage which would be mostly unused.

    https://docs.google.com/file/d/1NrBZJejkUTRYJv5YE__kBFuecdDL2pDTvKLyBjfCPr_8yR7eCTDhLGm8oEPo/edit

    Hook natural gas to these plants and make them the “7 hours a year” deep backup facilities. No need to sink capital into gas turbines/real estate/transmission that are infrequently used. No need to have double staffing, the same people can run the combined site.

    • Ronald Brakels

      And when the weather report shows there will be both low wind and solar output and high demand they can start storing heat straight away so there will be maximum heat stored ready for when it’s needed. This will cut down on the capital costs required for the gas system. And it doesn’t have to use natural gas, biomass could be used, although it is possible that since it is used so infrequently it would be cheaper to use natural gas and then remove the CO2 released from the atmosphere.

      But in Australia we have so much surplus fossil fuel capacity we will just keep some around to deal with rare shortages in supply and it will act like our diesel and kerosene generators do now, only being used for maybe a few hours a year. (We have a 440 MW kerosene plant in North Queensland that may not have been switched on for over a year. Thank you rooftop solar and improved efficiency and no thanks to a stupid planning decision.) And if it looks like there will be a shortage in supply that our fossil fuel backup can’t handle for some reason, we could always start them up before the shortfall in supply strikes and use simple (and cheap) electrical resistance heating to top up thermal storage. This may not be an energy efficient way to do things, but it may be a cheap way to do things if it’s only required on an average of a few hours a year.

      Of course, since we look set to install energy storage in our homes and businesses thanks to the low cost of solar and the ridiculous cost of grid electricity here, we may end up with little or no thermal storage.

      • Zer0Sum

        Perhaps Abbot thinks he can tap the heat coming from the bush fires as a renewable energy source.

        Australia actually doesn’t have that much conventional fossil fuels resources. Most of the mega projects are inherently energy negative or just outright frauds designed to put public money into the pockets of the wealthy elite who managed to get into positions of authority where they could influence key policy decisions in their own personal favour.

        However the abundant fuel that Australia has is Thorium. Combined with CSP and it can be used as a high energy and low cost thermal storage fuel. The Chinese and Indians are already working on CSP with Thorium Salt Storage.

        The same tracking technology that CSIRO has implemented with the supercritical steam engine can be used to heat Thorium too…

        The big problem for Australia is the lack of vision of the Political elite. They have had the opportunity to implement large scale projects with this combination for at least 25 years and have refused to acknowledge the possibility. The main reason is this tech is not dual use so it represents a direct threat to the Nuclear industry who have managed to convince everyone that Plutonium and Uranium are the only viable options.

        • Ulenspiegel

          OMG, why is it so difficult to accept that nuclear power plants or NPP projects die due to simple economic reasons? Wind and PV are cheaper, sorry.

          • Zer0Sum

            The real question should be is it energy positive or not? If you think that we can sustain global growth, feed the masses and do everything we need with just wind and conventional solar alone then you are putting an unnecessary barrier in humanities path to greater abundance.

            The list price of a mega project has very little impact. Generally the more expensive it is the better chance it has of getting approved because that provides more opportunity for graft.

            In this case CSP with Thorium salt storage creates a very efficient and non military grade nuclear process whereby CSP can be run 24 hours a day.

            It might be hard to get your mind around the concept that Thorium can be used as a heat source that is initiated and sustained through CSP but there is no good reason for it NOT to be considered a serious contributor to the energy production cycle.

            This is not LFTR tech. Uranium/plutonium is not required. The concentrated heat from the sun creates enough energy to allow the Thorium to sustain the temperature. However in much the same way as LFTR tech if things start to get too hot just drain the salts away into a cooling chamber and reuse them again later.

            The storage medium is literally dirt cheap, does not create excessive carbon pollution especially if mined with electric shovels, transported with electric trains and processed with electric systems. It is abundant across the world in countries like Australia, India, China, USA. It does not create massive amounts of deathly toxic nuclear waste that cannot be easily disposed off, cannot be used for military weapons.

            It does provide a constant reliable baseload for industrial scale energy supply.

            But there will always be those who are too blinded by their own dogma to see the potential.

          • Bob_Wallace

            The world has limited capital. If we spend more on nuclear for electricity then we have less to spend on other infrastructure such as potable water, irrigation systems, transportation systems for moving produce to market.

            Nuclear, in the US, is 2x to 3x the cost of wind and solar. Even in Germany with its poorer onshore wind and solar resources new nuclear would be more expensive.

            It’s time to let the nuclear dream fade away. We’ve spent 60 years trying to make nuclear affordable.

            Wind and solar are affordable. With the storage technology we have they can be 24/365 sources for less than the cost of nuclear.

            “But there will always be those who are too blinded by their own dogma ….”

          • Mint

            We certainly have not spent 60 years trying to get nuclear more affordable. The last 30 years have been almost stagnant due to public fear, and even the decade before that saw only marginal changes.

            There is a huge economic opportunity for nuclear in industrial heat. No steam turbines are needed for such plants, and use of wind/PV via an electrical heater is way too expensive unless another technological miracle slashes the cost of renewables. To compete with $5/GJ natural gas, renewables powering an electric heater would have to cost 1.8c/kWh, including transmission (since they need so much area and cannot be located nearby). Small modular nuclear has a chance.

            The industrial sector uses almost as much natural gas as the electrical sector US:
            http://www.eia.gov/dnav/ng/ng_cons_sum_dcu_nus_a.htm

            I also imagine that in other countries short on natural gas resources, coal is used a lot for industrial heat.

            So this is certainly an important target for NG usage reduction.

          • Bob_Wallace

            “We certainly have not spent 60 years trying to get nuclear more affordable.”

            Sorry, I can’t accept that claim. Many countries have had nuclear industries and to claim that they we’re looking for ways to reduce the cost of nuclear along with their other work just doesn’t make sense.

            Public fear didn’t kill nuclear. Cost killed nuclear. Nuclear starts had fallen away well before TMI. It had become obvious that the promise of cheap energy from nuclear simply wasn’t happening.

          • Mint

            TMI was not the beginning of fear against nuclear power. Demonstrations against nuclear had been building for over a decade before that:
            http://en.wikipedia.org/wiki/History_of_the_anti-nuclear_movement

            And fear goes even beyond that. Research was essentially limited to evolving the naval designs plus the military incentive for breeders. Why? Because it was the low risk option and a relatively proven tech. Even in France there hasn’t even been an attempt to try Gen4 designs.

            First-of-a-kind attempts aren’t about saving cost. Spain’s PS10 thermal solar has a 20-year PPA at 37c/kWh. Why is it acceptable when PV is a far cheaper way of doing solar? Because it’s about technological advancement and a new way to produce power that potentially has other benefits. Nuclear power never had even remotely close to the diversity of research and pilot projects that renewables get, and I contend it’s about fear, directly from demonstrations and indirectly through policy.

            You know how renewables benefited greatly from ultra-low interest rates? Back in the days of investors wanting 15% ROI, even a low $1500 per nominal kW of wind would amount to 9c/kWh in interest alone. So even today’s wind construction technology would be uncompetitive in the 70′s, 80′s and 90′s, but conditions are different now.

            For similar reasons, nobody felt it was worth the risk to invest in any new nuclear technology when there was something that worked well enough already. Too much cost for too little benefit and too much potential public backlash.

            This statement is just plain myopic:

            It’s time to let the nuclear dream fade away. We’ve spent 60 years trying to make nuclear affordable.

            Ask anyone in the industry, and they’ll tell you the innovation going on in the last 5-10 years has dwarfed what we’ve seen in the previous 30 (or more), and this is without even building anything yet. The internet and modern computing have made it immensely easier for all sorts of bright minds to engage in a previously inaccessible topic.

            BTW, you still haven’t offered any suggestions for economically eliminating natural gas and coal used for industrial heat.

          • Bob_Wallace

            You are making what seems to be an erroneous assumption that fear of nuclear caused nuclear builds to cease in the US.

            The high cost of nuclear energy killed the nuclear industry in the US.

            http://www.vermontlaw.edu/Documents/Cooper%20Report%20on%20Nuclear%20Economics%20FINAL%5B1%5D.pdf

            The high cost of nuclear energy has killed/is killing the nuclear industry around the world (graph below).

            I haven’t paid enough attention to the issue of industrial heat to make a reasoned reply. My guess is that given the high cost of nuclear, along with the siting issues, industry would not turn to nuclear heat unless forced by legislation. A sufficiently high tax on carbon is unlikely in the US for the foreseeable future.

            ” Nuclear power never had even remotely close to the diversity of research and pilot projects that renewables get”

            And that is a totally bizarre claim. Nuclear energy has received massively more support than have renewables. The US has spend far, far more on research reactors than on PV and wind prototypes.

            “Ask anyone in the industry, and they’ll tell you the innovation going on in the last 5-10 years has dwarfed what we’ve seen in the previous 30 (or more), and this is without even building anything yet. The internet and modern computing have made it immensely easier for all sorts of bright minds to engage in a previously inaccessible topic.”

            Perhaps, but all that innovation has not led to competitively priced reactors. When/if it ever does then the nuclear industry might come back to life.

            But with wind now below 5c/kWh and solar to soon be there the task of lowering the cost of nuclear by 2x or more is daunting.

          • Mint

            You’re derailing the discussion.

            Your original claim is that we tried to reduce the cost of nuclear for 60 years and failed, therefore there’s no hope to do so in the future.

            You’re wrong. The only thing we legitimately tried was PWRs (and a few breeders, mostly outside the US). It doesn’t matter how much money you claim was spent on it, because it’s minimally relevant to what’s possible with other nuclear technologies. New power generation technologies are always uncompetitive at first, e.g. solar and wind. You can’t use that as a reason for nobody even trying the non-PWR reactors in being designed today.

            But with wind now below 5c/kWh and solar to soon be there the task of lowering the cost of nuclear by 2x or more is daunting.

            You would need wind at ~3c/kWh to produce heat at the same cost as 10c/kWh nuclear.

            If the only application of nuclear was electricity, then I’d at least half-agree with you that nuclear’s prospects are dim. But high temperature industrial heat (heat pumps don’t apply) strips away the ~30% efficiency of thermal electricity generation, thus making it 3x harder for renewable energy to compete. The processing of oil sands and shale oil in particular are great examples of industrial processes that need lots of heating energy, and often in locations without a connection to the national grid.

          • Bob_Wallace

            “Your original claim is that we tried to reduce the cost of nuclear for 60 years and failed, therefore there’s no hope to do so in the future.

            You’re wrong
            ​…”​

            You are willing to argue that there has been no serious attempt to lower the cost of nuclear energy when the industry has been acutely aware that it was dying due to high costs? That we’ve done no research on alternate reactor designs?

            Sorry, that fish is swimming tits up.

            As I said, I have not paid enough attention to industrial heat needs to form an opinion.

            If you are using this example, then I’m likely to start forming a very negative opinion…

            “The processing of oil sands and shale oil in particular are great examples of industrial processes that need lots of heating energy, and often in locations without a connection to the national grid.”

            The last thing we should do is anything that would aid cooking tar sands.

          • Mint

            You are willing to argue that there has been no serious attempt to lower the cost of nuclear energy

            I didn’t say that. I fact, I directly said that we legitimately tried with PWRs.

            That we’ve done no research on alternate reactor designs?

            I didn’t say no research, but it was minimal. How can you claim otherwise when nothing was even built? Know of any alternative reactors designs being offered a 20-year 36c/kWh PPA?

            The last thing we should do is anything that would aid cooking tar sands.

            They’re going to be extracted whether we like it or not, and I’d much rather have them use clean energy instead of natural gas for it. It’ll take until maybe 2050 before EVs mostly flush out the 200M gas cars on the road in the US, and the developing world is only going to increase demand for oil as traditional wells dry up.

          • Bob_Wallace

            With the cost of current design reactor electricity well over 10 cents per kWh it’s unlikely that there’s a way to cut costs in half and make nuclear competitive. If big business saw a possible route they would be pushing the government to give it a go.

            You’re free to dream away about some miraculous breakthrough design that can compete with <5c wind and solar, but obviously the US companies that invest in energy don't see it coming anytime in the reasonable future. I'm afraid that with nuclear it's "been there, tried that, moved on".

          • Mint

            Why are you continuously putting words in my mouth? Go find a single post in this article’s comment where I said I think it’ll be less than 5c/kWh.

            This is not going to start from “big business”. They operate on a razorblade model: Build the reactor at minimal profit with someone else financing it, and sell fuel for 30-60 years. PWRs use reactor-specific fuel rod construction, while MSRs just need LEU fluorides dumped into the salt (they can even run on spent fuel), and 1/6th of that needed in PWR fuel bundles. They have no incentive in seeing MSRs succeed.

            This is going to come from startups that don’t have to worry about protecting an existing business model or IP, just like it took a startup in Tesla to put a legit effort into building an EV. Such startups are indeed pushing gov’ts (China, Canada, US, France) to give it a go.

            I really don’t understand why you’re adhering so rigidly to this “been there, tried that” conclusion. Wind and solar energy were written off that way 10 years ago by many, many people.

          • Bob_Wallace

            I put no words in your mouth. I pointed out the target price new nuclear has to hit in order to be competitive. The market is not going to pay 11c for something they can get for less than 5c.

            Good luck with your nuclear dreams.

          • Mint

            You said, “You’re free to dream away about some miraculous breakthrough design that can compete with <5c wind and solar"

            I never claimed such a dream or implied it in this discussion. That's putting words in my mouth.

            I can tell you that industry will not pay (5c/kWh+5c/kWh) *(278 kWh/GJ) = $28/GJ for heat via renewables + grid charges when natural gas costs $5/GJ.

            If you can make on-site nuclear heat for $7/GJ (which, if used for electricity generation, would cost way over 5c/kWh, so stop with your disingenuous assertion that I am claiming <5c/kWh for nuclear), they'll consider it.

          • Ulenspiegel

            If you actually check how many green equivalents (5 GW PV = 1 NPP, 3 GW wind = 1 NPP) are already built in reality per year and if you keep in mind that both wind and PV are on exponential growth with a doubling time of 6-8 years you have to admit that NPP can not play a significant role in future for simple technical reasons.

            There is no reason that we won’ t see 100 GW PV per year in 2020 and 200 GW per year around 2027 (40 NPP per year!). For wind the developement is slightly slower.

            A nice graphical representaion of this issue is found here:

            http://www.renewablesinternational.net/comparison-of-nuclear-with-renewable-power-worldwide/150/537/77822/

            Add the issue that available NPPs are more expensive than wind and PV, nuclear power is in danger of becoming completely irrelevant within on generation.

    • Mint

      Anything less than 4 hours of continuous use can be served with batteries at a cheaper construction price per kW than natural gas (7h/yr will all but likely fit into that category), and probably 6 hours in the near future. It’s the longer periods that are tougher to serve without natural gas.

      But you’re looking at the wrong scenario in Budischak’s paper. We’re not going to get to 99.9% in the free market. He did system level optimization, not the marginal price analysis by which all markets evolve. Look at the excess generation for the hydrogen storage case (other cases are similar): 1.18GWa, 29.9GWa, and 54.0GWa for 30%, 90%, and 99.9% respectively. That means going from 30% to 90%, you add 11.7GWa to the grid (incl battery charging) while spilling another 28.7GWa. Going from 90% to 99.9%, you add 0.8GWa to the grid while spilling another 25GWa.

      Even the 30%->90% scenario will never happen in reality. Who is going to build renewables while only getting paid for 30% of the energy they produce?

      Note: “30%” in Budishak’s paper means 30% of hours completely covered with renewables and storage. It’s actually 50-60% of total energy generation. By far the most relevant case is, IMO, the 30% 2030 centralized battery case. It’ll take too long for there to be enough EVs to do GIV.

      Anyway, back on topic: You’re right about using NG generation to avoid excessive storage.

      • Bob_Wallace

        “Even the 30%->90% scenario will never happen in reality. Who is going to build renewables while only getting paid for 30% of the energy they produce?”

        In 2011 the (US) CF for natural gas was 24.2%. In 2012 it was 28.8%.

        Someone built NG capacity that sits unused well over 50% of the time. We’ll build excess wind/solar as long as curtailing is cheaper than storing. With cheap storage the CF numbers for NG would never have been so low.

        • Mint

          Renewables are currently paid for 100% of the electricity they produce, and are just about competitive with NG at 28.8% CF. They would have to come down in price by more than a factor of three to be worthwhile if they only get paid 30% of the time. That’s not gonna happen.

          NG has low average CF because a big portion of its capacity is from peakers (not just daily peakers, but seasonal peakers which have <10% CF), which get big capacity payments. You can't use wind or PV (the two renewable types in Budishak) as peakers, because they're not dispatchable.

          CSP can be used as a peaker, but it needs to get construction cost way, WAY down to compete with NG.

          We’ll build excess wind/solar as long as curtailing is cheaper than storing. With cheap storage the CF numbers for NG would never have been so low.

          Depends on what you mean by “cheap”.

          If you mean insanely cheap, like New Zealand’s high altitude lakes, then yes. They don’t really need peakers at all.

          If you mean moderately cheap, like future $100/kWh batteries, or paying people to use their electric cars for storage, then no. Look at the Budishak results, Table 3:
          -With 0% renewable, they have 72GW capacity and 31.5GWa load. That’s 44% FF CF.
          -For the 30% case with quite a bit of storage, it’s 15.4GWa/61.7GW = 25% FF CF.
          -For the 90% case with even more storage, it’s 2.18GWa/56.9GW = 3.8% CF.

          I fully expect NG CF to go down with batteries, as freak occurences like the polar vortex, heat wave, wind drought, etc become their primary purpose.

  • http://www.michaeljberndtson.com/ Michael Berndtson

    Alright, you made me refresh my steam thermo. Here’s a nice primer on the differenced between supercritical and subcritical boilers. This technology is essentially a rankine cycle, but with the sun as heat source.

    “How are Supercritical Boilers different from Subcritical Boilers?”
    http://www.brighthubengineering.com/power-plants/32896-how-are-supercritical-boilers-different-from-subcritical-boilers/

    And way too much on the rankine cycle from the squids over at MIT.
    http://web.mit.edu/16.unified/www/FALL/thermodynamics/notes/node65.html

    Positives: higher thermodynamic efficiency, higher heat transfer

    Negatives: supercritical steam is hell on materials. coal fired boilers have figured this out for the past 20 years or so.

    My question is this. Where’s the storage? After condensing, steam will have to be made again. It doesn’t seem like there’s a heat storage component here.

    • JamesWimberley

      This is a research plant, not a commercial one. Maybe you can’t get salt hot enough for supercritical. If that’s so, this will be enough to doom the technology: inbuilt storage is the selling point for CSP, the one thing pv can’t offer.
      The Holy Grail is getting hot enough at volume (>1000 deg C) for a Brayton cycle, aka a direct gas turbine. CSIRO were working on this too.

      • Ulenspiegel
        • JamesWimberley

          No mention of a Brayton cycle, but they are getting the temperatures. See also this.

        • http://www.michaeljberndtson.com/ Michael Berndtson

          After finally figuring out how to translate from German – that technology was very interesting. Kind of a fluidized bed design using ceramic beads.

          • Joel

            I think you are confusing two technologies. The ceramic storage is for an air receiver design, not for the steam receiver presented here. This gives some background: http://me.byu.edu/sites/default/files/IversonArticle1.pdf

            My understanding, after working as a PhD student on a project related to a air receiver design for CSP, is that air receivers can achieve much higher temperatures. We had 680 C in our plant (in Jülich), but you can go much higher. Central parts of our receiver had 1000+ C.

          • http://www.michaeljberndtson.com/ Michael Berndtson

            While your dissertation should be an interesting read, I’m afraid to become even further confused. I’m already, apparently, confused by two technologies, why take on a third?

            My question still stands. Does the subject technology (this here blog post here) have heat storage? A simple yes or no would suffice. No, I don’t need to visit another German website.

            (this comment may be more for James) Frankly, there shouldn’t be any process operational difference between pilot testing (development) and full scale, i.e. an essential component (heat storage) is not in place during pilot testing. A research project without all units of operation is called a mechanistic study. I believe the CSIRO work is beyond that.

          • Joel

            I don’t know what – if any – kind of storage CSIRO’s plants use. I’m not an expert on this (my dissertation was on a different topic). But the kind of storage you use is closely connected to what type of receiver you use (air, steam, molten salt, …).

          • Joel

            Anyway, I think the answer is that there is no storage. Not sure if it’s even possible for a steam receiver. And if there were one, it would have been in the diagram… But that’s just my speculation.

          • Ulenspiegel

            Joel,

            the DLR design is a modiefied solid particel receiver (section 4 of your) review. Correct interpretation?

            I had to learn 20 years ago in “Technischer Chemie” that the corrosion cause by steam is the reason for the low temperatures, last reports published by Siemens still mentioned the same in case of their steam turbines and as motivation for ongoinig research.

            OTOH gas turbines work with >1000°C. Is it possible to combine an air receiver with another kind or turbine to get an economic concept?

          • Joel

            Again, I’m not an expert on this, but I think DLR is investigating multiple concepts for energy storage. In Jülich (where DLR is also involved), the storage used to be essentially a regenerative heat exchanger (a.k.a. a regenerator). This year, they also installed a “Sandspeicher”, which I think is the solid particle receiver you refer to. Though I’m not sure how involved DLR is in that – it might be Kraftanlagen München (who built and now operates the plant) and/or FH Aachen doing this. DLR is (or was at least) also involved with Platforma Solar in Spain, with yet more technologies. And that’s just for the solar tower design…

            Different from a (fossil fuel) combined cycle power plant, you don’t extract any power from the air cycle (in an air receiver CSP). The hot air has small pressure gradients and is just used to generate steam (or be stored), I think at around 450 C due to material limits of steel. So efficiencies are lower. But don’t quote me on that.

  • JamesWimberley

    There’s no “lucky twist” about Abengoa’s involvement, After the collapse of renewables policy in Spain since the financial crisis devastated its public finances, this Spanish company has successfully gone global. It dominates the still small market for CSP plants rather as BYD does that for electric buses. Abengoa is cautious in innovation, and was not involved in the ambitious but much delayed high-temperature Ivanpah plant. Its involvement in the Australian supercritical project is a good omen for the technology’s transition into commercial use.

    • Mint

      The most important application of supercritical CSP is retrofitting coal plants. When the sun is shining, they can use this solar steam as opposed to burning coal, and everything downstream stays the same (turbine, condenser, generator, etc).

Back to Top ↑