Solar Power at Night using Concentrated Solar Power CSP

[@EngineeringwithRosie]

To try everything Brilliant has to offer-free-for a full 30 days, visit brilliant.org/EngineeringwithRosie/. The first 200 of you will get 20% off Brilliant’s annual premium subscription.

[@boxmenswear1053]

Rosie you’re a superstar. The more intelligent, articulate and likeable people like you explain the transition to a greener way of living the quicker we’ll get there with EVERYONE on board. The change is exciting, it’s full of opportunities and your way is so beneficial. Thank you

[@EngineeringwithRosie]

🤩

[@makeitwork583]

You really are awesome, one of the only KZbinrs I share with other people. Your vid’s are so insightful! Keep being awesome! Be well.

[@Deontjie]

Have you heard about the hundreds of Spanish investors lost billions on these kind of power stations? Have you heard what happens when there are a few rare cool days in a row and the salt hardens and clogs all the pipes?

[@srenjensen3817]

@@Deontjie If the project was launched before the marked was ready, then the investment can be lost. If there is ANY risk of cold days, one would need a backup system to avoid cloging of the pipes. There is always a risk element in investments. 9 out of 10 investments could be a loss. I thought that CSP was an obsolete tech, but Rosie just pointet out, that there could be a marked in the future.

[@sparkysho-ze7nm]

Very well said tytyty

[@ericgardiner7715]

South Africa has a couple of CSP plants coming soon. Will be interesting to see how it takes on, seems to be a good move by the Government.

[@marinamorrow8234]

I'm so glad to see your video on CSP. I love the historical research that you did. I can definitely add some of this to my CSP talk in my alternate sources of energy portion of my climate change lessons for my physics classes. Great job. Thanks

[@EngineeringwithRosie]

Oh that's great to hear, thanks!

[@christo930]

CSP is a dead end. It's too expensive. It's too maintenance heavy. You have to wash these things nearly every day. The upfront costs are enormous. You can do the same thing with conventional panels and use the electricity to heat up a salt. It's less efficient, but it's cheaper. All of these mirrors need to track the sun. It does not work at all if you are off even by a little bit. It doesn't work with clouds either. Who knows what evil partial cloudy conditions cause. Having the same pipe be hundreds of degrees difference in temperature a few feet apart is not good for the pipe.

[@massatube]

Marina make sure you don't teach the kids global warming activists stuff. There's no climate emergency.

[@johnbash-on-ger]

Be sure to mention an electric heater can be added to basically turn the storage tanks also into multifunctional cheap grid batteries.

[@johnbash-on-ger]

@@massatube Would describing fossil fuel caused global warming as a slowmo extinction-level threat be more your thing?

[@Jawst]

I made a death Ray in early 2000s 😂 used an old sky satellite dish and a mirror cut into squares! Managed to melt aluminium and cook some meat!

[@DADADRTR]

I only burnt wood while tinkering at ADFA, but you've now inspired me to graduate to barbecue level. 😂

[@stephenbrickwood1602]

yes you are right. Let me say that we can import as many vehicles from overseas as we need, overseas supply is unaffected by our small market and its demand. For the Australian market any demand could be met. I suggest that V2G be a priority as it puts the maximum amount of batteries into the grid and every building is a connection to the grid. Not rapid charging connections but daily steady top up or grid feed in. Rapid charging will be at the corner shops as a side business and on the main roads. Basically every EV is fully charged or ready with its next day's power needs. The old days of going to fill up with 100kg of inflammable petroleum when the tank is empty will be seen as like we look at the horse and cart days. The big business will be with the grid owners and the auto manufacturers. Every building will feed in to the grid and the 20million Australian EV batteries will absorb the excess and stabilise the grid 20million buildings = 660gWh daily 20million EV 100kwh batteries = 2,000 gWh DISPATCHABLE, dispatchable storage. Fossil fueled generation is 400gWh avg Peaking at 600gWh if you are lucky. Grid capacity expansion is the killer cost for central generation Renewables or nuclear. And nuclear will tell you that is the killer cost of new Snowy 2 and Renewables. But 5fold increase in electricity generation and the transmission is the killer of centralised any generation.

[@jared_du_jour]

Vehicle to grid is not a viable utility-scale energy storage mechanism. The biggest problem is that the vast majority of car usage is during the day, meaning cars are not on their chargers when renewable energy generation is most abundant. The other big reasons aren't technical; they're political and legal, like the administrative cost of renting battery storage from retail consumers, legal liability in the case of damage, compensation for the degradation of the vehicle's battery, managing customer's relationships and experience etc. Utilities would MUCH rather write one check to buy their own energy storage they own/operate than try to pay thousands of individual people to always have their cars plugged in at the right time.

[@stephenbrickwood1602]

@Jared S horse meat was cheap and available after the T model Ford rolled of the production line. Horse and cart thinking is dangerous. If you start in the wrong direction, you may never get there. 300million vehicles in the USA. 20million vehicles in Australia. If all EV, then the energy storage capacity is 3 times the daily fossil fueled generated electricity. This is a stupendously large amount of dispatchable electricity that the power generators do not want to compete against. The grid owners will love to manage this cheap daily supply. Most vehicles drive building to building. All buildings are connected to the national grid. Most vehicles are parked 23hrs a day. Ezi pezi for a selfparking EV to connect to the grid like a home robotic vacuum cleaner can do. Everything else is simple. The grid knows you. The grid will know you own the EV. The grid knows how much feed in your roof top solar power system supplies. This is baby simple, do the maths. The money will be in the grid ownership. The auto manufacturers know this double benefit. The battery developers know that cheaper and bigger is on the way. The grid builders know how massive the construction costs are to fattening the grid capacity and that it is a dead end. I am trying to help you. I am a Construction Engineer and have worked in power supply, and I know the numbers.

[@adon8672]

With the gradual rise of sodium and flow batteries, I think storing energy from solar PV will only get cheaper. CSP will continue to struggle to beat PV in terms of cost.

[@Alastair510]

Some 40 years ago I was shown a CSP power plant in some remote West Australian town. It worked, kinda. the guide explained that they had to spend a lot more time on maintenance than expected. Mostly cleaning the mirrors.

[@thewordofgog]

And before the Bond villain there was a Thunderbirds episode where there was a mirror supplying energy for a village but the mirror tower toppled over (earthquake maybe?) so the sun's rays were concentrated on the village. Of course they didn't allow for the sun moving but hey it was a kids show in the 60s!

[@wimvanuytven7858]

Great video as always! I think there might be another potential advantage to CSP: we never have to worry about curtailment of PV and wind anymore! In theory, surplus renewable electricity (if all the batteries are already full) could be dumped in resistors to further heat the thermal storage of the CSP plants. The round-trip efficiency is low, about 40%, but that's actually still higher than the round-trip of hydrogen, and much much cheaper.

[@iansmith6728]

That's all well and good in theory, however the problem with that is importing from the grid incurs network transmission charges, typically in the ballpark of $45-$50/MWh. That might not sound like much, but when you're dealing in a margin's game, that is a pretty steep price to pay before you even consider the cost of the electricity and all of the mandatory LGC charges. Before you know it you are breaching the $80/MWh mark, just to heat salt. The major problem is that the round trip efficiency is pretty low. Even the best plants would struggle to get anything more than 40% efficiency, so you're effectively paying $200/MWh for that electricity and the transmission company is making profit on both the inward and outward electricity flow. You'd need prices to swing a lot to recover that expense. You are far better applying storage at the point of generation than using the grid as a conduit. There are many commercial and near-commercial technologies that are more suited to that type of storage/demand management.

[@CaedenV]

The issue of overproduction isn't just a problem of shunting power to a sink, it is also an issue of shunting power through a large enough medium to get to that sink. The nice thing about batteries or gravity blocks is that you can house them on-site to have a local power sink to shunt excess power to. When you need to shunt it through the grid to a centralized sink, then you have rate charges to pay for the upgrades needed by the grid to handle it. In the case of using a CSP's moltan salt heater to dump heat into, then it means that they need to build in extra capacity to store that heat. The problem though is that if the salts cool down too much, then they solidify which is a whole other issue to contend with. So then you need to build in over-capacity at your CSP plant, while also not having so much over capacity that on a really cloudy rainy day you don't need a ton of coal power to keep your salts warm. As with all things renewable... its always complicated. Worth doing, but complicated.

[@Sindrijo]

​@@CaedenV The problem of solidifying salts is probably a non issue as you can design tiered storage with heat exchangers that can be mechanically uncoupled when the extra thermal mass would be a liability.

[@Kangaroo_Caught]

Quite a few years ago, I visited White Cliffs in western NSW and visited the local thermal solar plant. I'm not sure where my photos are, but Wikipedia lists it as the world's first commercial solar plant.

[@andrewgrubb9268]

Engineering with Rose Coloured Glasses. Suggest you have a close look at the failures of Crescent Dunes and Ivanpah CSP plants or the large plant in Spain which was closed down as soon as government subsidies were ceased. It's not just a case of installing a a plant BUT successfully operating such a plant in a cost effective way to maybe even given even a modest return on capital. Doesn't appear to be happening yet with many disappointed expectant consumers. Let's not worry about the impact of CSP on birds especially raptors. This series is more of would, coulda, shoulda, maybe, possibly etc etc.

[@JohnMillerDesign]

Thanks Rosie! Electric cars will be very common in a few years. At around maybe 60kwh storage each, that's big power.

[@nickfosterxx]

Wow, hats off to you for a fantastic concise and well structured script, makes it all so easy to retain. Will check out those references too, thanks so much.

[@chalermako]

Great channel. A lot of your footage is from Noor II & Noor III. After working directly at the site, I have to say the parabolic seems more workable. The solar tower seems like a great idea, until commissioning all those mirrors to move correectly. This is something that can be overcome with quality design and control. One thing that remains problematic, that is the amount of water required to clean the mirrors. The footprint for a 150MW turbine is very big, which in the dessert is not so bad until a day of wind and dust puts the mirrors out of order. A literal fleet of diesel water trucks is used to overcome this problem. Pumped hydro is much more proven at scale.

[@dizzyfpv4903]

CSIRO Newcastle CSP project actually holds the world record for the highest steam temperature generation from CSP

[@patrick247two]

Thank you. It would be nice if we could dispense with the steam part of the process.

[@briankuhl9314]

Fantastic video, was wondering what happen to all those CSP projects I saw in the past, glad to hear all that positive reporting was just blowing steam 😉

[@1locust1]

There are also thermal storage bricks for heat storage and iron/air batteries.

[@ecoideazventures6417]

I think Rosie is 'almost' painting a rosie picture about CSP but anyway, i am gonna believe it :)

[@jokerlecture]

Minus CSP u cant go fully renewable. Solar +onshore wind+offshore wind +psp +convwntial hydro you can reach only upto 65 to 70% For rest u need CSP or nuclear.

[@phelanwolf6747]

CSP should be part of the future city planning, instead of leaving most of the rooftops unused it could be used with the tower concept to concentrate heat there and use it in various possible ways: 1. Energy storage 2. Electrity generation 3. Heating 4. Hydrogen generation (Hydrosol) This way a more decentralized way of providing energy can be achieved on a community level.

[@rubidot]

3 years to build a CSP plant is shorter than I thought it would be.

[@jfolz]

Would be interesting to know where CSP is possible/financially viable geographically. I.e., whether other parts of Europe will build their own plants or lots of HVDC transmission lines from the Iberian peninsula. The latter would certainly be a better alternative to buying oil and gas from "questionable" sources.

[@Kangenpower7]

In America, Ivanpah is somewhat a "Ideal" location to install a CSP. It is near I-15 and the California / Nevada state border, about 2 miles to the northwest of I 15, so you can see it from the freeway, but it will not blind the drivers. It looks like a very bright LED light at the top of a tall tower from the freeway. The problem is not the 4,000 acres of glass, or the amount of maintenance that the mirrors might require to keep them clean enough and pointed in the right direction. The $2,2 Billion cost is what matters, and the 400 MW of power output. So that works out to about $5.50 per rated watt, or 5 times what a normal PV solar system costs. And you must have steam engineers on site while the 800 + PSI steam generator is running. You could install a 2,200 MW solar PV system for about the same cost. And it might be smaller footprint too!

[@vernonbrechin4207]

Tanks for your summary of the various forms of CSP and the comparisons with other power sources. You could have provided some examples of the problems CSP has encountered that were not originally anticipated. This has been especially true for the tower mounted receiver type units. You could have included the story of the large Crescent Dunes Solar Energy Project that has sat idle for many years due to numerous factors.

[@bashful228]

and how Solar Reserve went bust because of technology failures... breaking many hearts in the rapid decarbonisation transition world.

[@alanw356]

Or you could do those things yourself. This video gave me precisely the information I was looking for.

[@christo930]

8:48 All of this is possible with PV. You just use the electricity to heat the salt. Though PV is less efficient than CSP, the electric heaters are near 100% efficient (there would be slight losses in the line). You eliminate most of the maintenance as well. While both benefit from being clean, CSP is much more susceptible to dirty mirrors. CSP doesn't create any power without a clear sky. Any mirror not directly looking at a clear view of the sun will have its power drop to near zero.

[@nicolasblume1046]

I watch a lot of engineering stuff on KZbin and I really wonder, why your channel was never recommended to me until now. Anyways, great I've found you!

[@701983]

AFAIK, dish systems are rather solar stirling dishes, with a stirling engine driving the generator, instead of steam engines/turbines. However, they don't provide heat energy storage, electricity production follows direct irradiation, similar to much cheaper photovoltaics. And photovoltaics can make use of diffused light (cloudy sky) too.

[@stopscammingman]

I wouldn't have guessed that the vid would come with a cool history lesson!

[@zaphodbeeblebrox1130]

awesome !! pumped hydro used to be my favorite, but better thermal storage seems to be coming up.

[@jasonbroom7147]

Very good video, as always. Your content is well thought out and very clearly presented, with excellent production value. I think your enthusiasm for CSP is misplaced, though. The only reason the plant in California is still operating, and I suspect the same is true of those in Spain, is that they were provisioned largely through government funding, and neither grid is robust enough to withstand the loss of generation they provide. CSP has proven to be too costly, in terms of maintenance. This technology will continue to be a fringe solution and will be obviated entirely when hydrogen sequestration becomes more economically viable. The boring of geothermal wells using plasma lasers, something I believe you covered a while back, should offer all of the same advantages, with a much smaller footprint and far better LCOE.

[@bashful228]

sorry but I was a huge fan of solar-CST and one of the main proponents went bust because of technological failures, it's not a "lack of need for the technology" that deployment of power-tower CST. The projects in the Middle East were all parabolic mirror projects. They had a contract at Port Augusta signed with the SA Government and never broke soil on that. Perhaps the molten silicon designs of 1414 (who now have a lease on the land that Solar Reserve were going to use at Port Augusta). Prof. Lovegrove is a legend, I'm sure he'll have a lot to say!

[@waszyrowski]

Great video Rosie, TY.

[@EngineeringwithRosie]

Thanks!

[@durwoodmaccool890]

Good video. I had largely written off CSP mainly due to it's higher LCOE, but as you pointed out the storage angle may make it much more competitive. It would be interesting to see how the cost compares with long range transmission.

[@johnsamsungs7570]

Industry needs heat more than electricity so it is ideal for industrial processes. Having heat available after dark is needed as industry runs all day not just when the sun is shining. So, CSP can do both, electricity plus heat around the clock. The tower CSP can provide the very high temperatures that the channel/trough can't.

[@b9eda9ad]

Only problem I see with CSP is that batteries for grid storage are becoming cheaper and better so will it be cost effective against classical solar combined with Na-Ion or some flow batteries in 3 years time ?

[@adrianthoroughgood1191]

This is the gamble. By the time CSP is really ready for mainstream will cheap new battery tech undermine it? I think it comes down to CSP being able to have a very large amount of storage for very low cost. The power of the input array and the turbine system ate the main costs. So you have low power high capacity. It's very suitable for backup power generation to cover a few days of no wind, but batteries to store enough to cover that would cost an absolute fortune, even with cheaper tech than today.

[@manuelcilia391]

dear Rossie, have a look at Raygen which is an Australian company and has built 4mw pilot plant in Victoria. It is cross between PV and solar thermal.

[@stephenbrickwood1602]

Now you are talking Rosie. The Australian electricity grid is massive and dispersed like Renewables energy..

[@tdevi1]

SO thats what it is!! In the US, between the drive from Los angeles and Las Vegas is the Ivanpah Solar Power Facility!!! (I always thought it looked like heat death ray! haha)

[@matthewbaynham6286]

You're comparing the storage of this to lithium batteries, but there are other types of batteries like the flow batteries. So how does the storage in these CSP compare to other types of batteries like flow batteries for a couple of days of storage?

[@vincenguyen2922]

Have you heard of solar chimney power plant? I always thought this was a lower tech solution but not sure if anybody is pursuing the tech. I also read a while back they thought of using the waste heat from a coal plant for it.

[@mikenewman4078]

I was hoping to find more details about the CSP installation at Port Augusta. To the best of my knowledge it desalinates water for a large scale hydroponic greenhouse. I've passed the site a few times when crossing the country. On one trip it was during a sandstorm which stretched at least to Ceduna. The CSP plant looked quite surreal through the gloom. CSP is a difficult technology, molten salt is a temperamental material with nasty expansion characteristics. Severe damage to the hardware during the melt / freeze phase change cycle can be a real problem. Cold source in desert areas is its own issue. When I visited Solar One Nevada parabolic trough installation in 2014 the generation plant utilised air cooling and masses of fans. Deep space as a cold source seems to have potential and there is some interesting citizen science on KZbin regarding Barium Sulphate nano particles as the heat rejection medium. Wind, PV, storage are all strong established competitors, interesting times ahead. CSIRO were working in silicon metal as a heat storage medium.

[@nitelite78]

I know you are saying there isn't that many CSP stations at the moment but I thought there was far less and thought it was just a struggling technology. Think that's just because I saw a few people post links to articles saying that. So it's quite exciting to hear there's quite a few commercial plants and it looks like there's going to me loads more in the future.

[@tomschmidt381]

Interesting overview of CSP plants. I had thought CSP had turned out to be uncompetitive however you raise a good point about storage being cheap using CSPs.

[@JohnSmith-pc3gc]

The focus of solar light on a material is limited by the melting point of the material. The higher the temperature, the higher the efficiency of the heat engine can be. If the light is focused on a plasma, there is no temperature limit. A gas turbine spins the air in a compressor. If there is a high voltage arc going through the spinning compressed air, it would make a disk of plasma. A plasma tends to absorb all manner of radiation. If the concentrated solar is focused onto a plasma disk, it would heat up the plasma. If some of the light goes through the disk, another disk or a number of disks might absorb most of the light.

[@matteoricci9129]

While on a work trip I stopped by gemasolar in Spain it was a research station with all different experiments, why direct mirror on the side of a solar panels are discarded, and why polar mount tracker are discarded, gemasolar had those experiment and two CSP

[@bru512]

I have watched CSP for decades and have built small home sized systems. Other benefits not pointed out is that it requires highly available, inexpensive materials, and no rare minerals. Other disadvantages not pointed out includes burn risks to people and animals around the concentrators. It's not clear how CSP has significantly changed other than costs have been dropping due to scale. They are still complex systems. The last decade has seen both simple PV and Li-ion battery technology costs drop dramatically, essentially wiping out most of CSP advantages. Will continue to watch this technology evolve, but I just don't see it ever beating Wind, Solar PV and Batteries. Incredibly simple, safe, and low maintenance solutions. That said, I would love to be pleasantly surprised. Let the markets decide!

[@Nikoo033]

Even as storage, CSP will soon be dead in water as well, as long-term renewable energy storage will soon rely on “freeze-thaw molten salt battery for seasonal storage”. Look it up. So solar PV generation coupled with this type of storage will win over CSP in the end.

[@stopscammingman]

Stored for days? That's cool!

[@Ice_Karma]

13:04 Hey, 99% reliability means just over three and a half days downtime a year, 99.9% means a bit under 9 hours, 99.99% is a bit under an hour, 99.999% is five and a bit minutes, 99.9999% is 31 and a half seconds. Small percentages matter. 😸

[@unicornadrian1358]

Aren’t most of the CSP projects in mothballs due to high running costs and far more competitive alternatives?

[@fiegenfiegen]

In Spain, during most of daylight, CSP already accounts for 5-7% of our power demand. Add PV solar and wind, and we regularly have over 55% of our power demand covered daily. This can be achieved by most countries and remember: the less dependent we are on oil, the fewer millions go to absolutist regimes, dictatorships, and terrorists.

[@philipandrew1626]

Water usage in the steam turbines is a potential issue in dry and desert areas. I recall some years ago CSIRO were looking at experimenting with a jet engine (without fuel) at the power tower receiver to attempt to create a way to generate electricity without the high water usage. Never heard how these experiments panned out however.

[@jaganathanaratnasingam4635]

It might be possible to use captured CO2 to drive the turbines. I believe, I have read in the past that there has been some work on turbines that run on CO2 with the prospect of using them in geothermal plants.

[@rmar127]

Now here is a thought, some people are suggesting that electricity companies may soon resort to paying businesses to run their most electrical demanding process during the middle of the the day to soak up the excess solar power in the grid. CSP plants could take advantage of this by installing larger molten salt tanks than would otherwise be used for the size of their plant and use excess solar from the grid to heat the salts directly. Thereby having lower levelized cost of energy

[@BillMSmith]

Thank you Rosie. Your research into the history of CSP is wonderful, it adds a great deal of perspective. As you point out, future grid reliability will require multiple types of storage. CSP does offer additional flexibility because with a thermal storage/turbine combination you can move that stored energy to other storage (battery, pumped hydro etc,) and though efficiency is lost to some extent it adds an additional option that grid operators should see as a real positive. BTW, I shared one of your recent videos to FB with the comment that you were my favorite Australian engineer. Just to let you know, that still holds. (To be honest N=1, but we could add a few and it would hold. 😛)

[@EngineeringwithRosie]

There is more than one engineer in Australia, so I hope if you meet another one I will still maintain your number one spot!

[@AI-antics]

I loved this video! I would be quite curious why sterling engines never took off for this type of grid scale use. I'm sure there is a reason there are many clever people who have thought of it first but I would love to know why it never took the steam turbines job. Thanks Rosie, love the videos as always!

[@thamiordragonheart8682]

while Stirling engines are more efficient, they're also much bigger and need much more maintenance than gas or steam turbines. I think the fuel cost vs capital and maintenance cost just doesn't work out in favor of sterling engines for power generation.

[@Kangenpower7]

Ivanpah cost $2,200,000,000 for a 400 MW CSP tower system, with 4,000 acres of mirrors. And you must have a staff person on duty each hour that the steam plant is running. So that is about $5.5 per rated watt, while a conventional PV solar system will only cost about $1 per rated watt. So if you have $2.2 Billion to spend, would you recommend a 2,200 MW solar PV system, with almost no need for maintenance, or install another 400 MW Ivanpah system where you must hire staff to monitor the steam turbine and boiler system. At least 20 employees who will want to earn about $45,000 - $75,000 a year.

[@konradcomrade4845]

Sweden developed a submarine with a Stirling engine because it can work so silently. Military technology seems always to come first, before civilian!

[@zaurenstoates7306]

Sure all generators are "almost always available" but some are more so than others solar thermal has about a 20% capacity factor (amount of time it's out puting it's full name plate capacity) vs 25% PV and 30% wind on average. But nuclear sits up top with an average of 92% capacity factor. So that means that even with thermal storage you would need ~5x the name plate capacity to match the power made by a nuclear power plant. This also doesn't take into account the winter months generating around 1/3 less solar energy on average.

[@solexxx8588]

I can't see this being cheaper than solar with batteries since you add the complexity of boiling water to run turbines. If you are in a place with 300+ days of sunshine it might make sense. A supercritical gas turbine to run the generator would be a better fit but I haven't seen them available at any useful scale yet. It's an interesting technology with lots of moving parts to maintain that can fail. Thanks for the video. To stop burning carbon fossil fuels the answer is to use every viable technology including nuclear as quickly as possible. This tech works, so we should use it in the locations where it can be efficient. Thanks for the explanation. Cheers!

[@andrewclark3236]

There is a lot of mechanical equipment in the cooling systems for batteries that will constantly need replacing, as well as the slow degradation of the battery materials themselves and the electronics in the inverters.

[@industrialsolar5978]

Thank you for featuring our installation footage of Fresnel Collectors in your great video, Rosie. Well-researched and informative! Though we realize our projects to provide heat, steam, and cooling for industrial companies (so they are smaller in size compared to some of these enormous power plants), every step away from fossil fuels counts!

[@johnbash-on-ger]

☝👍

[@guyreavley5195]

Rosie, we have a patented CSP technology that increases CSP effciency considerably .. Would love to share with you given your enthusiasm. Guy

[@KevinTurner-aka-keturn]

There's a Concentrated Solar Power project here in the States that I had high hopes for: The Crescent Dunes Solar Energy Project. I think it was supposed to be something of a flagship for the molten salt energy storage. It probably even shows up in some of the footage you have of central tower installations. But wow, that place has been plagued with operational problems. Offline for eight months due to a leak in a salt tank. It was expected to be a 40 MWh installation, and it has *never* hit that target, even in its best months at the peak of the season. Most of the year it couldn't even hit half that. Economic failure, bankruptcy, it was offline for over two years and I thought they'd given up for good -- looking it up now I'm surprised to find it's back in operation again! So I'm sure he's heard it before, but that'd be my question for Keith: What has the industry learned from Crescent Dunes, and how will future installations avoid its failures?

[@adrianthoroughgood1191]

Yes this is the biggest thing that needs to be talked about. How can you avoid the molten salt problems. I heard from someone who worked at that place and he said the molten salt was a nightmare that they just hadn't expected. This is also an issue that people never address when enthusing about molten salt nuclear reactors where it could be an even bigger concern!

[@Kangenpower7]

A engineer from McDonald Douglass in Huntington Beach California was at my 7th grade class, to show how the Barstow solar system worked. (I think his daughter was in my class). This was a 100 MW system, also using molten salt at about 850F. The problems include needing staff there to monitor the boiler when it is running. How to warm the salt before you can pump it to the tower to get it hot, so a lot of electric heaters to warm the steel pipes that have salt inside. That reduces the output MW of the system, when you are using a MW of power to heat the salt each morning. The biggest problem with Ivanpah is the $2,200,000,000 cost! For a 400 MW system. You could install a 2,200 MW solar PV system, instead and that will not require monitoring or cleaning the mirrors.

[@fredbarnes196]

Arizona has the Solana project. It takes 3 sq miles of solar to power 70,000 homes. That works out to about 5300 sq miles of desert to handle all us households, or 20% of Arizona. Interesting

[@markjmaxwell9819]

We have the technology at the moment to produce 100% renewable reliable power 24/7 with a relatively small physical footprint. I have designed two power plants that can do the job extremely well with designs that have never been used before. As they used to say on the X FILES "The truth is out there" 😎👍

[@CaedenV]

The issue with CSP is water use, which seems super unintuitive. But almost every time Ivanpah has been shut down was because of water issues, needing to expand their water storage, and cooling towers multiple times before it could run at full rated capacity. There is also an issue with air traffic (including birds) being blinded and/or baked by the bright tower or misalligned mirrors (which seems like they could mitigate by adding a type of hat on top of the tower to help manage some stray light?). It isn't to say that CSP is impossible... just a lot more expensive than standard solar PV +batteries. Between molten salts corroding pipes, and water issues, and having to turn a generator, CSP is very much a bandaide fix to make traditional turbine-turning power work with a renewable heat source. It can be done, but when you can just create electricity directly cheaper... why go through the inefficiencies of legacy systems? Also, there has been some fun stuff with SCP PV. It has never quite hit main stream, but this has the potential to hit 40-60% efficiency as PV directly, without any efficiency concerns of converting to a steam turbine. This is still a research project because colling it, and replacing the short lived cell that gets burned alive are still very real issues, but there is some promise there. But perovskite coated traditional cells look like the real direction of the future with potential to have cheap panels that can hit nearly 30% efficiency, which is remarkable!

[@davidmurray2829]

Very interesting and compelling points. 👍👍👍

[@delusionsalyer]

Are you going to talk about The company heliogen? It solved the problem of high cost of mirrors! The problem now for csp is not the cost of the mirrors but the tower because no one solved the high cost of building the tower, energy storage and maintain it.

[@mikeklein4949]

What an incredible summary. Thank you Rosie.

[@BertVerschuren]

Brilliant video Rosie! My reference now for helping it explain to anyone.

[@williambreen1001]

Unlike PV and batteries(at least modern Li-ion), the likes of CSP+thermal storage and wind can actually be built and maintained at home or the cottage industry scale. Which may be something to bear in mind if you're genuinely seeking independence with an off grid system. The PV and batteries set-up may be a lot sleeker and convenient, ..but at the end of the day you're still just as beholden(albeit with a longer notice period than if you were dependent on the grid) to mega-industries for your energy.

[@5th_decile]

Concerning 4:30 and the rationale behind the Fresnel trough, I came on a youtube video about the launching of the experimental Évora Molten Salt Platform where they managed to use molten salts in a parabolic trough system (and they of course boast that this achievement would enable a higher operating temperature as you already indicate as one of the advantages in your video). My impression of the CSP landscape is that the parabolic trough is maturing faster than the tower system (e.g. for equal land areas, the generated power seems to be higher and the project cost lower). The rationale for not yet giving up on the tower seems to come from a market of high temperature industrial heat (companies like heliogen try to make towers which heat > 1000°C).

[@rmar127]

Forward planning is definitely needed. Unfortunately here in Australia, the LNP is still firmly in the pockets of the fossil fuel industry and the ALP is too gutless to enact the legislative changes to make it happen.

[@GEMINDIGO]

Another fantastically informative and interesting video from Rosie!! Thanks for the education!!

[@Ikbeneengeit]

Thanks for the serious coverage of this interesting topic

[@stephensimpson5283]

Fascinating. Love your videos.

[@Richardincancale]

Very interesting video - and it raises so many questions so glad there’s a follow-up livestream planned! Questions: What is the comparative power generation per square metre (or hectare…) for CSP vs PV? What is the cost/MWh of CSP storage vs Lithium or other batteries (like Redox flow), can the latter even scale? How long term is CSP storage - my guess is after a week the losses become uneconomic so probably more suitable for diurnal cycling? Finally - a key advantage of PV is that it is completely incremental, from a panel on top of a shed with an old battery up to grid scale and use of V2G storage, whereas CSP is big project stuff - beloved of utilities etc.

[@EngineeringwithRosie]

I've added these to the list, I'm sure we'll get to at least a couple of them! p.s. I have a hybrid flow battery project tour video coming up soon.

[@mintakan003]

@@EngineeringwithRosie In the US, I'm skeptical of running long transmission lines from the CSP plant to elsewhere. There are permitting and NIMBY issues. If they close down coal plants, the lines are already permitted. It would be easier to replace these with renewables (wind or PV), or a storage site. I'm not sure how well CSP would scale in most geographical situations.

[@johnbash-on-ger]

It's the same as with other types of power plants.@@mintakan003

[@ThalassTKynn]

I'm a bit of a stirling engine fan, so I always thought the dish arrangement was best. And I suppose there could be situations where that would make the most sense. But Australia should absolutely be constructing large CSP farms with storage!

[@randomguy7253]

Could a coal power plant be converted to CSP? I'm thinking of reusing the turbine and generator, which probably are a big chuink of the cost of either type of power plant.

[@SocialDownclimber]

It could but CSP should be sited for the largest number of clear sky days per year, and it would be a pretty big coincidence if a coal plant was already in the same spot.

[@adrianthoroughgood1191]

​@@SocialDownclimber also there needs to be a large amount of space around for the mirrors, if you wanted to come close to fully powering the turbines. These power plants ar3 usually built near cities that need the electricity. I think nuclear reactors are better suited to reusing old plants. There is a company that has developed a system for doing exactly this.

[@michalbike9305]

en.wikipedia.org/wiki/Martin_Next_Generation_Solar_Energy_Center

[@adrianthoroughgood1191]

@@michalbike9305 while every little helps, that example is pretty little. The CSP is providing a small boost to the gas plant, not replacing gas as the power source. At peak solar output it only provides 6% of the energy to run the turbines in that power unit and can't operate the turbine by itself at all. It takes 500 acres to do even that. Preheating the turbine to allow for faster startup is useful. We need peaker plants to fill in renewable gaps and they are usually less efficient than long duration gas plants, and this helps with that.

[@lylestavast7652]

Maybe a related topic or future show - systems that can recover rejected heat - Rankin or Organic Rankin systems - there's a great breakout of energy flow usage in the US done by Lawrence Livermore National Labs which shows that roughly 65% of all energy ends up as rejected heat. Of 97.3 Quads, 65.4 were rejected heat, and 31.8 usable services. Some is unavoidable, but surely recovery of some of it would provide some useful reductions of other energy. And so much is made of new generation; how about improved efficiencies of end use offsetting new demand volumes too... we see a lot about heat pumps, ground loops, deep wells and things like using CO2 as a refrigerant - industrial process heat pumps etc. And related to this particular topic of your EXCELLENT video, one of the knocks on a CSP plant in CA just over the NV border - constant complaining that they have to use NG to bring up to temp overnight and it never hit its projected/planned performance targets... so, they have to use NG to raise the heat for generation - um...., that's how it would have been generated anyway otherwise and using the stored thermal the rest of the time is decreasing CO2 emissions all those aggregated hours...

[@evil17]

Great vid & info on the pro’s & con’s of CSP Rosie, but I am not sold on it, I think CSP has too many problems, maintenance issues, & costs, etc, for it to be competitive with other technologies for storage & energy creation. While CSP is a very interesting technology, it still has a lot of issues & I think I would like to see more efforts go into other storage & Geothermal technology solutions, I’m sure Australia has huge potential for Geothermal power creation. Cheers

[@BenMitro]

Some pretty powerful arguments for CSP...I buy it!

[@DADADRTR]

Coupling a solar tracking dish to pump a hot media through a concrete slab could significantly reduce a Canberra winter heating bill, if not nullify it altogether.

[@user-um9sl1kj6u]

For long-term energy storage that doesn’t dissipate, why not use it to split water as hydrogen? In addition to sodium. You could heat a transfer fluid or salt as your primary, then use next generation flywheels as a Super Efficient Capacitor, and the Generate Hydrogen for long term storage. All of those are less toxic than lithium I thought about doing that with offshore installations that have solar wind and wave power

[@bernadmanny]

I'm really interested in it's potential use in industrial heat which I've yet to hear much about to my puzzlement.

[@basildaoust2821]

I'm glad I live where I live because we get 97% of our power from renewable hydro, it isn't 100% but it feels like we are doing OK.

[@GarretKrampe]

what size natural gas plant would emit as heat 4.48 x 10^11 W per year given 60% emissivity efficiency and state the natural gas consumption and Co2 emissions use only metric units in the calculations To determine the size of a natural gas plant that would emit 4.48 x 10^11 W per year, given a 60% emissivity efficiency, we can use the following formula: Q = εσAT^4 where Q is the rate of thermal energy emitted per unit area, ε is the emissivity, σ is the Stefan-Boltzmann constant (5.67 x 10^-8 W/m^2K^4), A is the surface area, and T is the absolute temperature. Rearranging the formula to solve for A, we get: A = Q / (εσT^4) First, we need to convert the thermal power output to units of energy per year: 4.48 x 10^11 W x 3.1536 x 10^7 s/year = 1.413 x 10^19 J/year We also need to determine the temperature of the natural gas plant required to emit this amount of thermal energy. We can estimate this temperature by assuming that the plant emits heat at a temperature of 50°C (323 K), which is a common assumption for the temperature of industrial process heat. Next, we can calculate the surface area required for the natural gas plant using the formula above: A = (1.413 x 10^19 J/year) / (0.6 x 5.67 x 10^-8 W/m^2K^4 x (323 K)^4) A = 6.61 x 10^7 m^2 We can now calculate the natural gas consumption required to produce this amount of thermal energy. We can assume a heating value for natural gas of 50 MJ/m^3. Natural gas consumption = energy output / (heating value x efficiency) Natural gas consumption = (1.413 x 10^19 J/year) / (50 x 10^6 J/m^3 x 0.6) Natural gas consumption = 4.71 x 10^8 m^3/year Finally, we can calculate the CO2 emissions resulting from this natural gas consumption using the emission factor of 0.2 metric tons of CO2 per MWh of electricity generated: CO2 emissions = natural gas consumption x heating value x CO2 emission factor CO2 emissions = (4.71 x 10^8 m^3/year) x (50 x 10^6 J/m^3) x (0.2 metric tons CO2/MWh) / (3.6 x 10^9 J/MWh) CO2 emissions = 523,888 metric tons CO2/year Therefore, a natural gas plant with a surface area of 6.61 x 10^7 m^2, consuming 4.71 x 10^8 m^3/year of natural gas, and emitting 4.48 x 10^11 W per year at a temperature of 50°C would generate approximately 523,888 metric tons of CO2 per year

[@adam-g7crq]

Hi Rosie, what are your thoughts on liquid air batteries.

[@rickharold7884]

Awesome need more of these and related solutions. Love it

[@peterfosmark7052]

It could be interesting to hear about the potential in combining PW with the heat storage part in CSP as a battery alternative.

[@CausticLemons7]

What about a single massive parabolic mirror pointed at a tower that is angled such as to not cast a shadow on the dish? Could also be used as a heat ray.

[@GauchoJonny]

And for all dunkelflaute that is coming, we use Biogas powerplants that are easy to store biomethan and hydrogen. Hydrogen is still tricky to store, but we could be using production of hydrogen as a demand load to stabilise the grid. And maybe put some of hydrogen in biogas production to increase methan production if the storiges of hydrogen is full. There are so many synergies to use as a solution to the "sometimes" or "not always". Even taking accound of waist heat of Industries.

[@grahamcook9289]

Solar PV is better than CSP in that it is scalable from domestic to grid. CSP only makes sense at grid level. The drawback of solar PV with curtailment at times of over production is now dealt with by battery storage at the domestic level and a myriad of storage technologies at the grid level with pumped hydro, compressed air and liquid air to mention but 3. There are vast areas of untaped solar PV installations that do not require green field or desert sites, such domestic roof, commercial roof, industrial roof and carparks. I think the carparks is a no-brainer and in the hotter European medetteranian countries you use more and more carpark space, especially in public facilities such as hospitals and supermarkets now being covered in solar PV panels on a frame-work. It generates electricity for local use and export to grid revenue as well as provide shelter/shade for vehicles.

[@GarretKrampe]

Let's have another shot at making a comment and see if it gets deleted. PV solar is ~ 20% efficient meaning that 80% is going into atmospheric heating. If you capture this heat and use it or radiate it back into space through the Atmospheric IR window .. no issues. To date no one in the world is doing that , so solar PV remains a greenhouse heater if you compare a grass paddock at 32 C in noon sun vs a PV array at 53C. We are talking about TJ's of heating over the course of a day for a 1 KM^2 area. 200MW of power vs 835 MW of atmospheric heating ! CSP troughs on the other hand, using around 400 C temps to heat SCCO2 and turbines can leverage ground loops , IR windows, regeneration, and long term (few days) storage as part of it's very design.

[@kayakMike1000]

How do you measure efficiency? Do the financial efficiency to understand how quickly it will be adopted.

[@muzzarobbo]

you do an amazing job explaining these sorts of things. thankyou!!!

[@GarretKrampe]

how much power per year would be produced by a solar pv plant of 20% efficiency in Perth Australia given the area of 1 km^2 The amount of power produced by a solar PV plant of 20% efficiency in Perth, Australia, given an area of 1 km^2 can be calculated as follows: First, we need to convert the area from km^2 to m^2: 1 km^2 = 1,000,000 m^2 Next, we need to take into account the solar irradiance in Perth. The average solar irradiance in Perth is about 1,850 kWh/m^2 per year. Multiplying the area by the solar irradiance gives us the total amount of energy that falls on the solar PV plant in one year: 1,000,000 m^2 x 1,850 kWh/m^2 = 1,850,000,000 kWh/year To calculate the amount of electricity that can be generated by the solar PV plant, we need to multiply the total energy by the efficiency of the solar panels (20%): 1,850,000,000 kWh/year x 0.2 = 370,000,000 kWh/year So a solar PV plant with an area of 1 km^2 and an efficiency of 20% in Perth, Australia, could produce approximately 370,000,000 kWh of electricity per year.

[@wayne8113]

Thanks Rosie 👍

[@Hzwo1990]

I like csp since I heard about it like 2006. Then PV got cheap and efficient. Would be interesting to look at to what degree we really have to replace all these night power. Here in Europe, power is still cheaper at night, which doesn't make any sense already, let alone in a renewable future. If we make night power more expensive than day power, maybe even with flexible pricing refliecting the actual cost of production (which means we dont have to "make" the prices but they more or less set themselves like the prices of shampoo), then we need much less night power than now. I'm guessing like 10% ish, also considering other night power sources like electric cars. So...yes I think what Rosie is saying in this video makes sense, but how much do we really need? If people stop using night power because its more expensive than day, then CSP might even lose money...

[@robertchen7058]

if the LCOE of CSP with storage is lower or the same as combine gas, I think that would be it. CSP shouldn't be competing with solar PV, it should be competing with other on demand energy generation. We can have solar PV for the day and CSP with molten salt energy storage for the night, and that would be that.

[@rogerlafrance6355]

Some people have build low temp systems, storing hot water in arge insulated tanks, to heat their Earthship in the winter, often with great results. Even if you have to use fuels say 20% of the time, that's still a big savings. Also, in some places, Coal or Gas plants could be converted to multi energy, steam is steam.

[@iansevs549]

I love CSP

[@vincewhite5087]

Rooftop solar & battery is huge, but I want lots of all solutions. China has successfully placed CPs in cloudy mountain areas. Lots of distributed power required too. Hydro, tidal, efficiencies etc.