The article claims a high cost of mirrors and steam turbines as the big negative. Without seeing the cost-breakdown of the entire project, it is hard to ascertain if materials are indeed the largest factor or if it is something else like labor/salary or technology/licensing. If it costs $200m to build the tower and buy turbines, $2B/200,000 mirrors = $10,000/mirror. I find it hard to believe that cannot be lowered drastically once the research phase is over.
A reality in California is the high cost of regulatory siting. In this case the desert tortoise issue was costly.
In many other cases, you see multi-year court battles (some in regulatory proceedings rather than traditional legal courts). A lot of the time those pushing the fight are not environmentalists but rather NIMBYs who use the laws/regulations to try to prevent progress. Same with High Speed Rail, of course.
The delay and uncertainty adds a higher-than-expected cost since the investment appeal erodes.
It's interesting that California's pro-environmental laws can sometimes be a hurdle to "green" projects in this sense. Different sides of the environmental issue spectrum.
Environmental concerns about displacing protected wildlife can be mitigated by installing relatively inexpensive things. Ditto for concerns about displacing low-income families. The real problem is that electricity in California is cheap and it is impossible to build a profitable utility-scale solar farm without monetary government incentives or a down-round. The former is winding down and the availability of the latter is constrained by a negative feedback loop.
The expiration of federal, state, and utility incentives, mainly ARRA Section 1603, really screwed up the solar market in most of the US. Uncle Sam used to give developers of renewable projects (solar, wind, geothermal, biomass, hydro, etc) 30% of the total cost of the project in cash within a month or two of completion. Congress refused to renew that program after Solyndra, even though it was probably responsible for building several nuclear generators worth of renewable capacity. Companies can still get tax credits, but these newly-formed single-asset corporations that typically own the farms don't have any tax liability to offset. They can monetize the tax equity by doing some screwed up partnership flip structure, but that is time intensive and costly.
States also used to have incentives which ranged from state tax credits to RECs. These tax credits are even harder to monetize since now you need to find an equity partner who has a multimillion dollar state tax liability in, say, North Carolina. The market for RECs collapsed right after the utilities came close to meeting their renewable portfolio standards.
Most of the current pipeline of utility scale projects in the continental US involves a cramdown of the initial investors and developers. Everyone is looking for "broken" projects where someone else already spent the time and capital to get the project shovel ready. At the smaller end, there are even some contractors who financed the construction loans themselves and are being crammed down by the takeout guys. At the same time, many lenders have discontinued new lending and monetizing tax equity is expensive and difficult.
There are some exceptions such as Hawaii where 90% of their electricity is generated from oil so electricity prices are sky-high (think 5x+ California), but the local utility can't keep up with the requests for permits so projects wait in an ungodly long queue.
Finding and relocating all the tortoises and keeping an eye out for missed ones while the construction was going on is expensive but not relative to the total project cost.
Source: My uncle does this work and was on this project.
That's true, the tortoises cost $22 million, a small percent of the overall cost.
But siting, getting approval, finding a place where few folks will fight you, etc etc make it add up in terms of time and uncertainty and then when you do find a place you can move forward on there's a good chance it'll be remote enough to be extra costly to connect.
However, the large, open, sunny spaces in CA tend to be far from large population centers.
He's now working on the project to replace/upgrade the original transmission lines from Hoover Dam which will pick up this power as well.
I'm in agreement with the other commenters saying energy production should be distributed. However, this'll require a "smart grid" to balance things out - which we don't have - as well as potentially being less efficient.
You can do a pretty nifty heliostat with passive heat collectors and a closed loop containing a fluid which vaporizes at a convenient temperature for shifting the center of gravity. (And dampers so wind buffeting doesn't jiggle the panel too much.)
That said, $30 of motors and $50 of controller board is probably cheaper and more accurate. The passive system is good enough for orienting solar panels, probably not for aiming a mirror.
You can't use passive systems like that for pointing a mirror at a stationary source since they will aim the mirror straight at the sun. The whole point is that the mirrors form a parabola and that you point each mirror slightly different. Passive hookups like you describe would aim all the mirrors in the same direction.
I've done some toying around with this and came up with a system that can run off a limited number of motors.
Have you seen how large the heliostats are? They look over 2 meters on each side, and the surface looks like glass or stainless steel. That's several thousand dollars of light engineering right there. Plus shipping to a desert, plus installation crew.
So you're looking at $100-$1000 of motor. Several hundred dollars for the controller board (how do they know where to point? Are they networked or something?), or thousands if you have to design it yourself. Then you have to cable power to it, all of which has to be up to code. The whole thing also has to be weatherproof in a desert and not blow over in the wind.
(The passive one-axis heliostat does sound interesting for DIY systems, though)
Agreed, turbines are precision mechanical devices and also need regular maintenance. High-pressure steam is aggressive stuff so all the plumbing needs to be built accordingly as well.
Speaking as former Nuclear Sub technician, there isn't much maintenance on the turbines (think transmission in a car -- it's not too often that you have to open it up) but the turbine, high precision machined stainless steel piping and equipment is very expensive. Our steam ran at 1000 psi plus, which we were told was high enough to cut a man in half if a small leak was sending out a stream.
Thanks for the link. Wow: This system used a 100,000 psi (690 MPa) pump to deliver a hypersonic liquid jet that could cut high strength alloys such as PH15-7-MO stainless steel.
Ivanpah is steam, not salt (in salt plants, the storage system is a large component of the cost).
It's not quite as simple as which area is the biggest cost component; in my experience everything adds up. You see that giant structure next to the tower? That's an air-cooled condenser, much more expensive than a wet-cooling tower, but doesn't use any water. Wouldn't surprise me if that one component cost >$50m.
For the record, power towers have been around since at least the 1980s; the technology isn't super novel, but it's all in the cost reductions since then.
Well, I don't know the details, if it works directly with steam then the circuits must withstand a high pressure as well no? (which can justify the higher cost)
But I thought that for thermal solar the most common case would be a molten salt circuit (high temperature low pressure) transferring heat to a steam circuit for better efficiency (and the heat storage capability of the salt)
Ah yes, the technology is not that new (it's heating with the sun after all)
You are correct on both points. But you have to keep in mind that these projects were conceived more than 5 years ago, when everyone was hesitant to build a $2b salt plant. When you combine big money with old engineers, you get very conservative (with respect to technology).
Photovoltaics is more inline with Google's needs because it also offers increased cooling if placed above data centers. Could this be a reason why they favour PV?
1.) Thermal only works on a massive scale. You can't fit 1000 large mirrors, a tower, a huge turbine, and the requisite plumbing on a commercial rooftop.
2.) Even if you could, the city wouldn't let you because the mirrors would distract aviation and anger neighbors.
3.) Technological advances and a capacity glut in PV manufacturing brought the price of panels down so far that it doesn't make economic sense to build thermal anymore.
When you calculate the amount of power the plant is likely to produce over its lifetime, the cost per kilowatt-hour is likely to be much higher than for fossil-fuel power. It’s even likely to be higher than the cost of power from solar panels, thanks to the fast drop in solar-panel prices in recent years. If costs don’t come down — and decreasing the costs of mirrors and steam turbines is hard to do — solar thermal power might prove to be a dead end.
I hate this attitude - the costs of the produced energy is not everything. Who said that energy has to become cheaper and cheaper? If you want to reduce the carbon dioxide emission just introduce additional taxes for fossil fuels or even prohibit burning them or subsidize solar energy. Just accept the increased costs and enjoy your clean air.
> Just accept the increased costs and enjoy your clean air.
That's easy to say, but shouldn't we use existing costs as a base metric for measuring how much increased costs we can bear? Otherwise we would have to accept energy costs that eclipse the cost of everything else--including basic living expenses. Of course, energy cost is tied to basic living expenses, so at some point higher energy costs would mean fewer people can afford to live. Therefore, clean air has to be balanced with the more basic human right to live.
Lack of clean air can also _impair_ the right to live. When coal was the default source of domestic heating in London, toxic smog frequently caused hundreds of deaths and generally caused respiratory diseases.
I hate it as well, but I think it is generally anticipated that any greentech solution is never going to be widely adopted until it achieves price parity.
Which is another way to say that we won't start using "green" energy until we run out of hydrocarbon fuel.
It's a very dangerous attitude -- we need to use what's left of hydrocarbon fuel to bootstrap new energy while we still can, and have the luxury to afford to experiment.
I forget where the quote is from, but as has been said: "The definition of modern agriculture, is that it is a way to transform hydrocarbons into carbohydrates.". Unless we do something soon, running out of oil, will mean running out of food. We need it for tractors, and for modern fertilizers, for irrigation -- and for transport of food.
Oil (and coal) is cheap energy, because it's basically just laying around, waiting for us to burn it. It's grown harder to get -- but it is ridiculous to wait around for other energy to "become cheaper". If we don't make an effort to change the entire infrastructure the world runs on (quite literally) -- what will happen is that oil will become as expensive as "green" energy -- not the other way around.
And that is of course ignoring all the problems hydrocarbons leaves us with, in the form of various forms of pollution. That we'll have to clean up and/or deal with without cheap energy.
I'm not entirely sure I believe this myself, but I'm sure one can easily find studies describing the increased healthcare costs of the entire US population due to fossil fuel burning.
But I think it shouldn't be looked at in that light. The US government claims we, as a first world country, have the responsibility to protect those less fortunate than us. While that line is normally used to justify wars in far off countries (that just happen to be oil-rich), things like this are what I think of. Besides, it's not so much the pollution the US generates that scares me. It's places like China (or the next country that suddenly finds itself in a rampant industrial age) that scare me. Global pollution does not respect political boundaries.
So yeah, this will be more expensive at first. But the research needs to be done before it will get better. Somebody has to take lead so it can be ready by the time more 3rd world countries get "uplifted".
Another thing that people often don't price in to the cost effectiveness of petroleum fuel sources are the many increasing billions that went/go into acquiring said fuel every year…
> Who said that energy has to become cheaper and cheaper?
It needs to if we want to have better living standard in future. Or even more or less equal living standards for all people on Earth. Which I think is a reasonable goal.
If the power is expansive poor people won't use it. Energy is one of the most important productivity multipliers there are. Forbiding developing countries access to reasonably priced energy means artificaly keeping them poor forever.
Does anyone have information on what type of system could be setup using photovoltaics across a similar 3500 acre site? While its obvious not all the acreage is used for mirrors I am curious as to difference is possible output.
Output is expected at 392mw. I found a few acreage references for coal/gas plants but their output many times one of these plants. One solar plant I found is 25mw using about 250 acres.
Just trying to understand the land use efficiency versus other solutions. Fossil fuel plants do not count usage of road/rail in their size calculations nor the size of the source of their fuels. (coal mines/pits/etc)
What's the reflectance of a PV panel? Wondering if the mirrors could be replaced with them, knowing they aren't as good of a reflector as a mirror, but perhaps they reflect enough to run the steam generator, and still produce power themselves. Net win, perhaps?
"Solar cells are made out of N-type and P-type semiconductor material that use the visual light spectrum to generate electricity. Solar radiation with wavelengths of 380 nm to 750 nm (violet to red) strike the material with enough energy to knock electrons from their weak bonds and create an electric current. The unused wavelengths (ultraviolet & infrared) do not have enough energy to dislodge the electrons and are absorbed as heat."
So, I think this idea is pretty feasible if there is some sort of IR and UV filter that could reflect the IR and UV wavelengths at the heat collection tower and let the visible light into the PV cell to have multiple electricity generation sources. I would imagine doing that would more efficiently make use of all the light energy collected, rather than only the wavelengths that can transform the light energy into heat.
I don't know if something like that exists... I know there are UV and IR filters for cameras but I am not sure if by filter they reflect the IR and UV light and let only the visible spectrum through.
That would probably be a bad idea: The main energy input will be by UV. If it is absorbed, the material will heat up.
Physics states that (Reflection coefficient + Absorption coefficient + Transmission coefficient = 1), all wavelength dependent. Now if you want a high number for the reflectivity in IR, you will have a low number for the absorption coefficient.
If you have a low absorption coefficient you will most probably have a low emission coefficient since those two are linked (equal for black bodies, see Kirchhoff's Law/gray body).
So you have a low emission coefficient in the IR, and the thermal emission due to the heating from the absorbed UV will be quite inefficient, and thus there will be more heating, which is bad (edit: the heat will not be transported away and will accumulate).
Of course if you have good materials you might be able to migitate, but at first glance it looks problematic to me. Sorry for inelegant language.
Yes, they do. The question is what happens with the heat from the UV - it will need to be transported away somehow. If we don't have significant heat conduction or convection it will have to be radiated off.
Sunlight does not bring in a lot of IR, and almost all of that is absorbed in the upper layers of the atmosphere. What you have to consider then is ambient IR radiation from the surroundings due to thermal emission. The process is the same as the one we're assessing for the PVs, and the wavelengths are not too different. PVs are normally warmer than the surroundings due to their higher temperature, so less energy will be absorbed that radiated off in the IR.
Hm, maybe I misunderstood how you want to remove the IR? Let's consider some sort of reflectant shell around the PVs but not part of them. On the inside it would receive the heat from the PVs and of course, also reflect great parts of it back, letting it accumulate. We built an insulation. Not good. Maybe you can explain your idea in more detail?
The mechanism I discussed above is of course not the only relevant process, things like heat conduction and heat capacity of the material are also important.
In conclusion: Yes, there would be a net loss of incoming energy, but the mechanims that let energy be transported off might be severely hindered by our alteration. Thus, an accumulation of heat.
There might be ways to make it happen nevertheless, eg with water cooling, but it would require more infrastructure. (But then, I'm not an engineer..)
We clearly need a huge PRISM or a diffraction grating, placed on a tall tower that moves on railroad tracks so that it's always between the sun and the solar cell field. Then we can tune the cells to the right bands and use steam on the low frequency end.
You could also just cool the cells and try to use the waste heat.
Actually in space you could have these gossamer structures more easily as there is more ...space, and the weight penalty is less. I know some satellites use primitive concentration methods coupled with very high efficiency cells to generate a lot of power.
For years there has been talk of building a 1km "solar tower" in my home town.
The basic plan is multiple square kilometers of greenhouses surrounding a 1km chimney. The air inside the greenhouses will heat up and rise up through the tower at high speed. The bottom of the tower will be filled with 32 generators, for a total of 200MW.
> When you calculate the amount of power the plant is likely to produce over its lifetime, the cost per kilowatt-hour is likely to be much higher than for fossil-fuel power.
Solar thermal is cost effective if you take into account the economic problems that carbon emissions cause over their lifetime in the atmosphere. This is why putting a price on carbon emissions is so essential: to expose the true cost of fossil fuels at the time of sale.
So how do you come up with a fair price to capture the carbon externality?
And do you add the externalities to green power like solar? More people have died installing solar equipment than all nuclear accidents. Or what about environmental damage during PV manufacturing? Do we add those to the price of solar at the time of sale?
Indeed, it's difficult to exactly price externalities. There are published estimates of the carbon externality. You do you best; it's far better than the "estimate" we currently use ($0).
When comparing, you add in the externalities to everything. To be specific:
- you don't count the deaths in PV installation as its not an externality. The installers are an involved party; they can (for example) ask for higher wages due to risk. (Also, if you wanted to compare this to nuclear, you need to include deaths building those nuclear plants as well. And possibly mining the uranium.)
- you do count the environmental damage during PV manufacturing (and turbine manufacturing, and coal mining, and oil drilling, etc.)
Perhaps a nice benefit of these plants, presumably, is that your operating costs don't change much due to changes in fuel costs (your fuel is free) and these plants ought to be somewhat easier to site / approve than a nuclear plant.
The other benefit to these plants over nuclear is that the time it takes to get them operational is much shorter. This allows for faster iteration time, and therefore more opportunities to improve the technology and reduce costs.
Any lifetime cost calculations must factor in the fact that the fuel is free. I've seen quick comparisons with combined cycle gas, coal, etc. that neglect this obvious difference.
api, this is a great point and one that is often overlooked with most renewables vs non-renewables cost arguments. However, when you compare $/kwh of the electricity it does take all of this into consideration.
I think of renewables' higher cost as basically paying up front for the fuel instead of over time. At least that's how it should be thought of economically.
You're right, but traditional fossil fuel plants still cost a lot (a comparable amount infact) to build ($2billion for 600MW[1]). Renewables just have a much lower energy output (only work when the sun shines/wind blows, and at a lower capacity) and therefore a much longer payback. But you're right, not having to pay for fuel would off-set some of these negatives.
I wasn't aware we were building these solar towers in the US. How does this compare to the solar updraft ones being built in Australia? It seems like the solar chimney design is more simple, however there are obvious efficiency concerns (kW/acre).
I was going to bring up the same point. The solar updraft systems seem to be more cost effective and generally easier to build. I don’t understand why they’re not being built more. However, they definitely take up more land. And nature advocates claim they are harder on the environment because they essentially bake the top soil, killing everything (which is probably true). But everything has trade offs.
It looks like the solar updraft systems are only cost effective when you have vast amounts of cheap/free land - they are very inefficient in Watts per square meter, getting around 1% efficiency compared to 20-35% for other systems.
Do you mean, concentrating sunlight onto PV? or Having PV cells that are also reflective?
Reflective PV cells/mirrors wouldn't work well as you're taking the energy out one way or another. The more you reflect, the less you can use for PV, and vice-versa.
I've heard of some folks using a Fresnel lens to concentrate light onto a smaller area, where the silicon is. It improves the efficiency significantly, but can also cause heat issues (which can hurt efficiency).
I've tried this. With standard PV panels going over 2:1 is not advisable. The panels will live a lot shorter and if you go higher than 2:1 they will break (wiring will de-solder, the plastic bits & pieces will melt and they sometimes simply stop working). I've tried watercooling as well (with very thin panels) and that works better but the lifespan reduction is still significant, you can see month-to-month performance degradation (as compared to a control that is running on 1 sun).
Does anyone have information on the water-consumption (steam-loss, cleaning the mirrors, cooling the mirrors, and so on)? I would really love to see, if this is a feasible method of producing energy in areas without that much water.
But I really lack the necessary numbers in terms of consumption of water, to built an idea, if this is a viable technology for deserts and other dry and sunny places on this planet.
How can it be more expensive to produce a mirror than a solar panel? This doesn't need astromony grade mirrors... Are steam turbines so expensive compared to inverters?
I guess it's not the mirror, but all of the motors and electronics that keep the mirror pointing in the right direction. If only the Sun didn't move around the sky, this would be very cheap!
The cost does seem rather high, I wonder if (as a DOE funded project) contractors or suppliers were massively overcharging knowing that the money was there to do so.
The biggest flaw I see with this solar project is that it's just another continuation of the centralized model. We should be encouraging homes and businesses to be producers of energy and decentralize our power grid.
$2 billion on a centralized project in the age of increasing decentralization seems like waste of money.
I've heard various arguments for this, none of which seemed that convincing. Centralized power generation seems like one of the great successes of the 20th century. What do you think the advantages of decentralization are?
> Centralized power generation seems like one of the great successes of the 20th century.
From the point of view of large producers and investors, sure. Tesla's idea of decentralized power was viewed as pie-in-the-sky by his benefactor, JP Morgan.
While I'd love to see more decentralization, that's not really what's happening, is it? Urbanization is AFAIK stronger than ever, fewer grow their own food (or even have the ability to do so), institutions are getting more centralized, not less...?
We're even breaking the Internet, giving people asymmetric connections forcing them to depend on centralized services -- so what exactly are we decentralizing?
In another generation or two, people will have the ability to design and print their own circuit boards. Technology itself is becoming decentralized which will have the longterm effect of decentralizing society.
Less than 200 years ago, railroads were the height of technological achievement and the most powerful centralizing force in society, industry and government.
While I feel Solar Thermal (more appropriately CSP http://en.wikipedia.org/wiki/Concentrated_Solar_Power ) is very promising, for some reason many don't. Google invested in this company ( http://www.google.org/rec.html ) but has stopped investing more into CSP, preferring photovoltaics ( http://www.evwind.es/2011/11/24/google-cans-concentrated-sol... ).