Scheer strikes back at Khosla…

If you read Vinod Khosla’s critique of German environmentalist Herman Scheer, then you’ll want to read this retort from the man himself. Scheer decided to respond to Khosla by writing a rebuttal for CNET’s In a nutshull, he calls Khosla’s viewpoint naive, contradictory and an example of established forces trying to perpetuate the centralized power infrastructure that has served and made profitable a handful of the world’s biggest power suppliers. Scheer makes some solid points here — it will be interesting to see how — and if — Khosla replies.

Before you read Scheer’s reply, I do want you do read the following comment that Khosla posted on this blog: “I love PV and am invested in PV but don’t believe it can replace 50-100 per cent of coal… we need something that can match the scale and ‘utility requirements’ of coal at the price of coal-based electricity.”

Until cheaper, reliable and large-scale storage is available, this won’t happen with solar PV. I’m surprised that Scheer didn’t address this point in his reply. I think the debate between these two is not solar thermal power versus solar PV, but rather solar PV as a majority of the world’s power versus solar PV as a restricted minority player.

40 thoughts on “Scheer strikes back at Khosla…”

  1. Tyler, I believe that there is no real reason to believe solar and wind can’t be “baseload”. If you don’t mind, I’ll post a relatively long explanation.

    Wind (and solar) isn’t the only generation source that has variance. In fact, all sources do. Most of it (maintenance, refueling, etc) can be scheduled, but not all. Nuclear can be tripped very suddenly – it doesn’t happen all that often, but when it does the plant is offline for more than one day. The size of nuclear plants, and the duration of outages amplifies the impact of the variance, such that a small market like Ireland, for instance, has ruled out nuclear.

    The key is managing the variance, and reducing it to tolerable levels. As discussed, this can be done in many ways, and in much the same way as is done to match nuclear’s flat output with the variation in demand.

    You need:

    Geographical diversity, including expanded long-distance transmission, perhaps with HVDC that has roughly 5% loss per 1000 miles). Additional LDT would make the grid more robust, and reduce the variation of wind by increasing geographic diversity and reducing the ratio of variance to mean production;

    Demand management, similar to the kind of daytime demand charges that moved so much industrial/commercial consumption to the night time, thus creating “baseload”.

    “Baseload” itself is a bit of a misconception. Humans live in the light, and in effect have evolved to use solar energy. “Natural” night time energy use is very low. A large % of what we call “baseload” is Industrial/Commercial demand which has been shifted from daytime to night time by very simple Demand Side Management (DSM): charging higher rates, or “demand charges” for peak daytime usage.

    DSM could be easily expanded. The first, obvious place to start is eliminating flat pricing for residential. Other steps: home electricity monitoring — allowing homeowners, business and factory owners to track their electricity use in real time; dynamic pricing, to reflect variable costs; smart, grid-networked appliances that can modulate their electricity use based on current power availability and pricing; and utility control of those appliances.

    Solar insolation is pretty nicely correlated with demand. It would require very little DSM to shift the A/C demand curve to match solar insolation.

    “Negawatts” in the form of reduced demand as a result of DSM can be very cheap.The most obvious use is with plug-in hybrid-electric vehicles (PHEV’s), such as the Chevy Volt series hybrid that could be charged at night and during peak production periods. PHEV storage will be cost-justified by the vehicle owner, and reduced rates for scheduled charging will be a bonus. As PHEV’s expand they will provide an enormous synergy with variable sources like wind and solar;

    Storage can be very cheap, and storage that is here now or will be very soon includes pumped storage and PHEV’s. The Ludington, MI pumped storage facility has time-shifted nuclear production for 30 years(Pumped storage is very cheap at about .6 cents per kwhr., which is no more than a 10% cost premium for 100% storage) , and PHEV’s are certainly on their way. PHEV’s won’t arrive for several years. On the other hand, neither wind nor solar will reach a level that needs storage until then;

    Backup generation capacity, such as inexpensive gas turbines for the rare extended outage, powered by gasified biomass (which is very efficient for power generation, even though very, very inefficient for liquid fuels). Remember, capacity is very cheap, if you don’t have to use it often. The cost of diesel and natural gas generators is almost entirely in the fuel.

    This is one big reason pumped storage hasn’t been more widely used: until very recently natural gas peak capacity has been dirt cheap, and so relatively large-scale, long-term projects couldn’t be justified. They often had to be paired with other large projects, like nuclear plants.

    So, the upshot of the above is that wind doesn’t have to be 100% reliable, just reliable enough.

    As a system this would be significantly cheaper than coal, once you added in coal’s external costs. Whether it would be cheaper than nuclear depends on how you value nuclear’s external costs, especially the Price-Anderson liability cap, weapons proliferation risk, and opportunity costs for foregone investment in renewables.

    What some have described as irrational NIMBYism and unreasonable regulatory delay is really the political process, lurching about in an effort to put a value on those external costs.

  2. hmmm. I realize that I didn’t address the question of whether solar can be as cheap as coal when you don’t include external costs.

    The answer is that it can, though it’s marginal costs will rise above those of coal’s direct cost at a somewhat lower % of market penetration, perhaps around 35% of the overall market as a wild guess.

    The reasons?

    1st, solar PV is mostly a retail, consumer side technology, and competes with retail pricing. In the US that means that it’s competitive at $.10 per KWH, not $.04-.05.

    2nd, it provides peak power, which is more expensive for Industrial/Commercial (I/C) consumers, and hopefully will become so for residential consumers.

    3rd, most power demand is daytime, especially when you include the I/C demand which DSM has shifted to the night, and which people generally, and erroneously, include as part of “baseload”.

    4th, PV costs are plummeting. Solar costs are now around $.30/kwhr (for retail, rooftop PV). Given that solar competes with retail electric rates, this is actually competitive (meaning a payback of roughly 15 years or less) without subsidies in some places: So Cal and Japan in particular (though subsidies are phasing out in Japan, and growing in So Cal). Solar costs are dropping about 10% per year, which puts it at $.125/kwrh in 10 years, and $.06 in 20 (this is a cost-reduction path which is reasonably well accepted among experts in the area – actually, it may be much faster, with things like Nanosolar happening).

    Demand has really gotten ahead of supply. PV supplies are expanding at about 40% per year, but they can’t keep up with demand, especially in Germany. CA has increased subsidies, and France has raised the price they’ll pay for PV power, but Germany is gradually reducing theirs. Lately supply seems to be catching up with demand, and prices started to fall in mid 2006. PV suppliers can still charge a heckuva markup to ration their product, until supplies catch up in a year or two.

    5th, consumers can buy and install PV with very little cooperation from utilities. If they don’t care about selling back to the utility then they can cover roughly 75% of their consumption and rarely have unneeded production.

  3. Nick… I’m curious about your assertions regarding industrial/commercial as non baseload. Are you claiming that a sizable percentage of night load is from factories, facilities and equipment that are idle during the day and only used at night to take advantage of lower electric rates? Seems to me that many/most industries that run at night do it because of capital equipment cost structures and/or shutdown-startup procedures that preclude intermittent operation.

    Perhaps you could pass along the source of your info.

  4. I’ll state up front that I have a great deal of respect for Vinod Khosla, and having come from the same industry have even had the pleasure of meeting him once or twice. I have followed a lot of what he has written on energy and environment issues. While not agreeing with every detail or conclusion, his insight and analysis is always beneficial to the larger public discussion. Being an avid supporter of CSP, I was especially excited to hear that he may be involved in some breakthrough technology in that field.

    Having said that, it may be no surprise that I would find Mr. Scheer’s arguments less than convincing. I do recognize that there may be some mitigating factors that would make PV attractive even in the non ideal solar climate of Germany. Factoring in things like trade balances, job creation, national energy security and the grid benefits of some level of distributed generation may justify a certain level of government subsidy. I’m certainly a supporter of California (where I live) subsidies for PV. However it seems Mr. Scheer is wanting to throw some sort of anti-corporate, socialist romanticism into the debate. If he wants to use the power of government to limit the profits of energy companies… do so, but don’t confuse that goal with the goal of making our global society sustainable.

  5. I wish people would take the time to read Vinod Khosla’s original blog more closely. He has very few problems with PV in a ‘Rich Nation’ context – subsidies and all. It’s just that he’s a realist in that if we want to make the greatest difference WORLDWIDE, and soon (i.e. in the next 5-20 years) then we should ALSO be developing technologies that have a chance of being adopted by developing nations.

    Unless there’s a huge breakthrough in PV, Bangledesh (for example) will NOT be able to put a PV installation in every village – let alone on every rooftop – for several decades. There is absolutely no way!

    However, when Bangledesh goes to purchase its next power plant in 2012, lets put some serious attention into green alternatives (like Solar Thermal) so that they’re not FORCED to go with another coal plant because that’s all they can afford. Makes sense to me.

  6. Anonymous, I am having trouble understanding your reasoning behind calling Scheer’s policies “socialist romanticism.” In your mentioning his wanting to limit the profits of energy companies, you don’t specify that it is only the conventional energy companies whose profits have stood to be limited by Scheer’s Renewable Sources Act, but that by contrast renewable energy companies are finally starting to thrive. The part that government subsidies, which you also mention, have had to do with this is limited, the incentive created by the feed-in tariff having played a very significant role. I wouldn’t call this policy socialist because energy is not owned and produced centrally by the government as a result, nor commonly among all members of society. I would neither call it romantic because it has worked. Germany has become one of the top two world leaders in renewable energy in a few short years. It has worked so well, in fact, that countries all over the world (Italy, the UK, Spain, India, Australia) are adopting similar policies, having found that every other system they tried (e.g. cap and trade, portfolio standards, doing nothing) failed. Even Al Gore includes similar policy measures in his 10 point plan to save the environment.

    Do you think that the established, more or less centralized, conventional energy corporate structures will make the shift to renewables on their own without policies like Scheer’s? If so, I think it is your misplaced faith in these companies that is romantic.

  7. I read Khosla’s article quite closely. For a long response to it point by point, I invite you to check out my blog at

    As I have shared there, evidence shows that decentralized renewables like PV actually show more promise in developing nations than centralized energy systems. According to the Earth Policy Institute: “As the cost of solar cells has declined, … it is now often cheaper to provide electricity from solar cell installations than from a centralized source…. At the end of 2002, more than 1 million homes in villages in the developing world were getting their electricity from solar cells. If families average six members, then 6 million people are getting their residential electricity from solar cells. ”

    There is a more recent report that in China, 40 million solar roofs (PV and solar heating combined) have already been installed. Cities like Kunming (population of over 3 million) already have solar on every roof. (Sorry link is in German: Solar thermal plants cannot boast the same progress, neither in the developed world nor the developing countries.

    Bangladesh’s population is just under 150 million. If we are to assume that at each home in Bangladesh has at least 3 people, doesn’t this suggest that it could realistically take a lot less than several decades for every house to have a solar rooftop?

    I, and I’m sure anyone with common sense, a heart, and a few basic facts, concur with Khosla that there are still far too many people without electricity in poor countries. But researchers such as the Earth Policy Institute, don’t put the onus on the cost of the technology, as does Khosla. “The principal obstacle to the spread of solar cell installations is not the cost per se, but the lack of small-scale credit programs to finance them. As this credit shortfall is overcome, village purchases of solar cells could climb far above the rate of recent years.”

    Maybe Khosla should consider micro-financing solar as another option for where he puts focuses his admirable micro-loans if he wants to help the world, or more of it if he has already been doing this.

    Also interesting to note, according to The Economist: “Contrary to conventional wisdom, people in poor countries do care about cleaner energy, and are prepared to pay for it….It is a widespread misconception that the poor cannot or will not pay for energy…there is ample evidence that the poor do pay, often heavily, for inefficient, dirty energy—say, from kerosene, candle wax and batteries. Indeed, they often pay more per kilowatt than do middle-class, urban households or wealthy farmers who benefit from heavily subsidized grid electricity. For example, families in Peru’s remote highlands on average spend about $4 a month on candles. For a bit more, they could afford the much higher-quality power offered by village power units: experts say that local entrepreneurs can turn a profit by leasing out a small, 35-watt solar unit, enough to power two bulbs and a radio, for about $80 a year.”

  8. Your supposition that 3rd world countries cannot afford PV is quite wrong. In fact, for many countries (Kenya is a strong example) that lack an electrical grid, photovoltaics are the sole source of electricity for many families. They buy a tiny 8-16 W panel that can suffice for powering various tools like water pumps and radios.

  9. PVs may indeed be very well suited for some of the poorest countries. I’m sure it would be cheaper to provide PVs and some rechargeable lanterns to a village for lighting than it would be to extend a grid from far away. In these applications the amount of energy per person is still very low compared to western standards.

    I think the main issue that Mr. Khosla speaks about is the rapidly developing countries like India and China. In these countries large scale industrial and commercial centers are being built and people are adopting developed world lifestyles as quickly as the economic expansion will allow. New homes, cars, TVs, appliances and air conditioning will all require vast new amounts of energy. It’s in that context where the differential cost PV really matters. If China’s economy has X amount of dollars to develop new energy supplies, it will flow to those technologies that maximize the number of GW they get.

  10. I’m puzzled by Khosla’s skepticism regarding PV in the context of his advocacy for cellulosic ethanol.

    The costs for PV are coming down relentlessly and quickly. Prices for installed systems haven’t fallen lately, due to demand which is rising even faster than supply, but they are almost certain to be down by 50% in 5 years, something which can only be hoped for with cellulosic ethanol.

    Nanosolar makes a credible promise of a 75% cost reduction for PV cells in 2007.

    There are so many different ways in which PV can be manufactured, and so many companies and research teams working on this, that it is impossible to believe that cheap PV will not come, and soon. The only question is, how soon, and what are we doing to accelerate it?

  11. “Are you claiming that a sizable percentage of night load is from factories, facilities and equipment that are idle during the day and only used at night to take advantage of lower electric rates? ”

    Yes. A good example is steel mills, which shut down their very high electric consumption operations during the day, and gear up at night. Data on this is hard to come by, but I’ll see what I can find.

  12. If you’ve read the discussions about PV, you’ve seen that one of the issues is that cell price is far from the only cost that goes into a system. While cell prices may be down by 50%-75% within five years, already the other costs (module creation, installation, etc.) are the majority. At some point you get diminishing returns with reductions in cell cost. Unless of course you know of some way to convince building contractors to work for 50% less in five years. But at least here in California, construction costs are going up about as fast as anything.

    If there is some breakthrough technology which makes PV truly as cheap and easy to install as shingles or siding… that could be a game changer, especially for new housing.

  13. And as a postscript to the preceding post…

    I’m speaking for myself, though I think I’m echoing some of what Mr. Khosla supports. It’s not an issue of whether PV is bad or impractical. Indeed I think the current expansion of PV production is great. It’s got some fundamental advantages such as adding a distributed component that can help relieve grid congestion, working in bright but hazy environments and supplying power in off grid applications (such as truly underdeveloped countries). However, the economics of large scale concentrating solar thermal power are MUCH better and are likely to remain so for quite some time. Plus, with concentrating , storage is economically viable to match round the clock demand… hence it could make up a larger portion of the entire electricity portfolio. Bottom line, for a given amount of government subsidy, there will be a much larger GHG reduction with concentrating solar thermal than with PV. Doesn’t mean we shouldn’t do both. However, it seems the discussion is often skewed, with many more people supporting PV just because it’s the more widespread of the two.

    I would ask all supporters of PV not to sour on it, but rather learn about concentrating solar thermal and become a supporter of it as well.

  14. I am curious about your sources and if you are speaking of specific locales when you support large scale concentrated solar thermal as viably making up a larger portion of the entire electricity portfolio than PV.

    A report jointly issued by Greenpeace and the European Photovoltaic Industry Association states that the that

    “the market segment grid-connected PV rooftop systems…, has the potential to generate an average of 16% of electricity consumption across the OECD (industrialised) countries” by 2040, and the total PV “contribution would equal 21% of the world’s electricity output.

    Greenpeace also issued a report with the European Solar Thermal Power industry, which states that concentrated solar thermal has the potential to supply 5% of worldwide electricity by 2040.

    These figures don’t support your argument.

    As I have stated in other posts, I am also in favor of developing both technologies, but I have not seen any concrete information to convince me that concentrated solar thermal should be favored over PV.

  15. I’m glad you asked.

    I phrased my comments narrowly, and didn’t address installation costs, for lack of space. As you note, they are passing cell costs in importance.

    “If there is some breakthrough technology which makes PV truly as cheap and easy to install as shingles or siding”

    Yes, it’s standardization and building integration, especially for new construction. Currently PV installation is like car manufacturing before the assembly line: hand done, one-off, incredibly inefficient. Worst of all, they’re retrofitted. Installers are small, often without central warehouses, volume purchasing or financing; plans, permits and site inspections must be redone for every installation; wiring must be installed, which can be enormously expensive; and support structures and framing must be added to an existing roof.

    All of this could be eliminated by regulation like the California requirement for solar as a standard option by 2011. Just requiring sufficient wiring conduit for later installation would give big savings. It’s a disgrace that installation is so expensive: it could be cut by 75% easily.

    “If there is some breakthrough technology which makes PV truly as cheap and easy to install as shingles or siding”

    Funny that you should say that: one of the keys is making PV disappear into the shingles or siding.

  16. I had seen both these reports before, but had never really compared the two. You gave me a nice excuse to take the time to look at both more closely. Here are my observations which would lead me to not agree with their projections about 2040. Though I might add that projecting out 35 years is always prone to large degrees of error.

    First… Both these reports are optimistic scenarios (not predictions) that are projections based upon expansion of current government trends of subsidies. Neither report tried to compare the advantages of one technology verse the other to determine which is more cost effective or to predict which would proliferate most if subsidies were on equal footing.

    Second… If you’ll notice section 5 in the second report “Solar Thermal Power Plants in the Mediterranean” you’ll see that in fig 5.3 the year 2040 shows about 1500 TWh/yr with 2025 being about 400 TWh/yr. This compares with other tables which show 95.9 TWh/yr world total for solar thermal in 2025 and imply around 1800 TWh/yr world total in 2040 (the 5% figure). How is the rest of the world producing a negative amount of solar thermal in 2025? Why would prime locations in North America, South America, Australia, China, India and South Africa account for only 20% as much as the Mediterranean in 2040? It’s really hard for me to comment intelligently about their conclusions when they have contradictions like this and very little explanation of how they reached their conclusions.

    Third… The first report estimates PV electricity generation costs (in euro cents per KWh) as 0.15-0.19 in 2010 and 0.13-0.17 in 2015 (these are ranges for high insolation regions to be fair comparison with CSP). The second report estimates the concentrating solar costs as 0.06-0.07 in 2010 and 0.05 in 2015. Two reports with Greenpeace involvement and the conclusion is that concentrating solar will be less than half the cost of PV. So why, if equal subsidies were given per KWh, do you think PV should/would play a much larger role than CSP.

    Fourth… The issue of energy storage is not even addressed in the first report. While I believe PV producing 1.1% of the world demand in 2020 is quite reasonable without addressing the issue of energy storage, having that figure reach 21% would strike most people in the power business as being totally unrealistic unless some way of storing the energy is developed. While there may be some possibilities, the fact that the report doesn’t even touch upon the availability and costs of such storage leads me to be very skeptical of their conclusions. On the other hand, CSP can easily integrate enough thermal storage to spread the power generation throughout the day. It can also cost efficiently integrate gas turbines to address the rare times when multi-day cloud cover would deplete thermal reserves. These items are discussed in the solar thermal report, since it is known, working technology.

    Finally… The first report shows the cost of producing solar PV electricity dropping by at least 50% in each of the cities they list between 2005 and 2020. From you can see that the actual cost of generating via PV has dropped by at best 5% over the period of 2000 to 2007. Also from Solarbuzz you’ll see that the modules make up 50-60% of the installed cost. If the other 40% (labor, overhead, transaction costs) is held constant this means that the price of modules would need to drop by 83% by 2020. Keep in mind that a module also has labor, overhead and material costs that don’t drop as fast as the silicon cells. This is why many believe PV, though a great technology, is being marketed with slightly unrealistic goals. So even the estimate that PV will merely cost twice as much as CSP per kWh may be too generous a view of PV.

    Summary… I hope, and optimistically believe, PV can achieve 1.1% of world demand by 2020. I do not believe 21% in 2040 is economically viable when compared to other solutions and when the issue of energy storage is taken into account. Further, I believe that if the same level of media attention, public support and government subsidies were applied to CSP, it would achieve at least equal status for grid connected applications as PV.

  17. hmmm. Have you read my comments, above? I think you’re not integrating some new information, including:

    Balance of System (BOS) costs like installation can be slashed from their current levels.

    PV is a consumer side technology. Once costs fall to a price competitive with retail prices it will grow as quickly as cell phones. That may make consumer demand look more variable to utilities, but they’ll just have to continue manage the matching of supply and demand as they always have.

    Finally, Solarbuzz is reporting prices, not costs, which has misled a lot of observers into thinking that the historic trend of falling PV costs had somehow stopped. Prices have stayed high due to a shortage of supply (primarily of purified silicon) relative to skyrocketing demand. Costs have continued to fall. As the temporary shortage abates prices will plummet.

  18. I looked more deeply into these studies–thanks for motivating me. And I agree there are many contradictions therein beyond even the ones you cited. So it is difficult to draw clear conclusions.

    One concern I have with solar thermal is that I am not sure how grid installation and maintenance are figured into costs. This is major consideration especially in developing countries where grid installation is limited or non existent. Also of serious concern is that CSP, from what I read, uses more water per kw to operate than any techology out there, even nuclear and coal. This is in large contrast to other renewables like PV and wind, which use very little to no water. Scheer reports that together, nuclear and coal use up 50% of water consumption in the U.S. If the world were to switch to primarily using a technology that is an even bigger water consumer than what we already have, I think we would be in trouble.

  19. I will be thrilled to see the price of PV systems drop. I would love to get one for my own house here in Southern Cal, but I’d need to have prices drop a bit before I could make it work financially for me.

    I fully support both things you mentioned of requiring it as a standard option and having conduits for later addition.

  20. Maintenance I would think is included in the given estimates. The plants in the Majove have many years of operational experience so there is a benchmark from which to extrapolate. Grid issues would obviously vary from location to location. Though HVDC transmission lines are really quite economical if you lump together a lot of generation within one area. The loses can be under 5% per 1000km (this number is from memory). Personally I think the US needs to invest in a really beefed up HVDC transmission grid. Having excess long distance capacity would allow more trade-off of renewable resources… wind from the Midwest and offshore, wave from the Northern coastal region and sun from the Southwest. Less need for storage the more you can shift the power around.

    As for water… you got a very good point there. Is it more than nuclear and coal because the steam doesn’t get as hot? One could use reclaimed water if near large cities such as Vegas or ocean water if coastal, but your right, that is an important limiting factor. Would be nice if Greenpeace really dove in an addressed issues like that rather than just creating the glossy marketing brochure for each technology. I wonder if that mucky stuff called water in the Salton sea might still be useful as coolant. The pollution and mineral levels have unfortunately already killed most life in it, so raising the temp a few degrees probably wouldn’t change much.

  21. as I understand it, the water issue is an asset of CSP, not a liability. When located near a sea shore, salt water is destilled and can be used accordingly. That’s one of the strong points of plans by

  22. Running a risk offending posters here, please allow some reflections.

    As I see it, postings stay within the domain of current decision making paradigms, which keep on failing to meet the challenges we’re facing a species. In my PhD-research on this topic I have found reasons to assume the urgency of the matter requires a system change (techno-, socio-, psycho-, law-, insititutional systems in sync).

    Limiting our scope to one domain (here: the technical) any perspective/opinion/vision will remain ineffective. See also Gladwell’s ‘Tipping point’

    That’s where my take comes from, which you can read here & here.

    I’d love to shift this thread into this direction, since imo only then can we make real progress, rather than merely exchanging ideas (which is fine for starts, but needs -concerted-follow up).

    Emil M

  23. Speaking for me, I’m not offended. Though I’ll have to admit that in various posts on this site and other sites dealing with similar issues I see most of the aspects you mention (techno-, socio-… etc.) being discussed. If you think shifting the thread in some direction would be productive, you might think about just doing so… bringing up some other aspect of the issue and see if anyone else has something to contribute. For the most part there is no censorship.

    If you’re wanting to discuss the idea of radically rearranging the way the world works when you say system change, I’ll have to admit I think addressing climate challenges within the current system would probably be the more timely solution. People behave with a large degree of self interest. Corporations function to make profit. Governments function to keep politicians off the streets (just kidding). But in any case, the struggles and conflicted interests have been going on from the dawn of man and seem highly likely to continue. While there are a few different broad decision making paradigms, such as divine right of kings, the one most of the modern world has settled on is some variant of representative government combined with a capitalist system where other decision making powers flow with money.

  24. And of course a closed cycle air cooled process is an option, though I don’t know how much of an effeciency hit and increase in $/KWh would result. I know air cooled is more of a problem in hot climates, which of course is where you’d be putting these things. I’ll have to dig deeper and see if I can find out what assumptions on cooling were built into the price projections I quoted.

  25. And of course a closed cycle air cooled process is an option, though I don’t know how much of an effeciency hit and increase in $/KWh would result. I know air cooled is more of a problem in hot climates, which of course is where you’d be putting these things. I’ll have to dig deeper and see if I can find out what assumptions on cooling were built into the price projections I quoted.

  26. An intermediate option is evaporative use of water, rather than just cycling it through as a coolant: this uses much less water.

    It’s important to note that there’s a big difference between drawing water and returning it a bit hotter, and consuming it. The very high water demand mentioned above is for coolant, and doesn’t represent consumption.

    While I’m posting, let me ask: are you in PG&E’s territory? If so, are you getting one of the new meters? What’s your current marginal price/kwh for your last kwh?

  27. Hi Glue, indeed I’m in for a revolution. One of the spirit. The heart and mind will automatically/inherently follow suit.

    The threads presumably addressing other aspects of a system change, fail to see the larger -essentially spiritual- picture.

    What I’m aiming at is the perspective of Ken Wilber.

    This is imo where threads should be heading. It has become the backbone of my research, since all other perspectives turned out to be true, but partial. And therefore fundamentally flawed, when looking for ways how to serve the interests of future generations of all species.

    Pace e Bene,


  28. I am not offended :)

    The holistic view of the problem is one I strongly support. Scheer’s tackling the problems holistically is one of the things that makes me an admirer of his That said, at the risk of focusing on details, when you may just mean to bring up a general direction, I have trouble with some of the points you referred us to. Such as cooperating with North Africa/Sahara to give Europe energy. I think Africa should use their resources primarily for their benefit. The hegemony of the Euro-American energy system has been a big driver of Africa staying stuck in poverty, and this has to change.

  29. I always wonder whether the cost of getting and running water through the CSP system is part of the maintenance cost estimates. I have no idea. I tend to guard my skepticism until proven wrong, simply because centralized energy systems tend to leave out external costs when making calculations about costs. 😉

    I agree with you that the US needs to beef up its grid. My German husband looks at the grid here in L.A. and shakes his head in disbelief. :) Not sure about, but interested in, the efficiency loss of 5 %. Am very curious to see info on this, if you or anyone gathers it.

    Of course, our grid system problems are different from a lot of the world’s, where there is no grid at all. This is a bigger, more costly and cumbersome problem for CSP, as far as it becoming the primary global supplier of electricity any time soon. And a primary reason I remain skeptical about CSP’s usefulness in areas other than a few regions in the world, where it seems poised to work very well.

    I’ll deal with water below. A bunch of people have commented on it…

  30. Since CSP can’t really work on most seashores because seashores tend to have coastal cloudiness and fog, a substantial infrastructure would need to be created to transport the water to the CSP plants. This is a cost.

    I am not sure how this cost weighs against the extra cost of storage for PV and CPV (Concentrated PV plants).

    I am also not sure, in a time when many people have water shortages, about spending the infrastructure costs of water desalinization on electricity rather than on water for people to drink and irrigate with. Maybe both can be done, which could be great, but I would think that the latter needs to be the bigger priority, especially for developing countries.

  31. Wow… I’m really enjoying the dialogue on this post. Thanks for all your contributions.

  32. Diane, I agree we should take the whole of humanity into consideration, when thinking about electicity supply.

    Given the logistical, organizational, financial clout the rich countries can muster, I have no doubt the rich can supply all with 100% RES within a decade. What it takes is compassion, translated into vision, political will, courage, boldness.

    Even when there are cost not yet taken into account for CSP > when there’s compassion etc, there’s no final block, just a bump in the road ahead.

    For small communities and communities far away from easily accesable electricity, all kinds of options are available, which are entirely feasable when Spirit has been allowed to unfold

    The 100 book project points in an interesting direction.

    Emil M

  33. Just thought I’d add one extra post on the issue of CSP and water usage.

    The two largest planned facilities in the US, Southern Cal Edison and San Diego Electric, are for an initial 800MWp with possible expansion to 1750MWp (though only about 24% capacity factor). These are both using Sterling Solar Dishes. These are not steam based systems so they require only the small amount of water necessary to infrequently clean the mirrored surfaces. The drawback to dish technology compared to solar tower or solar trough is that storage cannot be economically integrated. Of course Southern California’s daytime usage peak matches solar production pretty well, so until solar makes up a march larger percentage of the total mix, storage will probably not be a major issue.

    Note from the below source that this is requiring no government subsidy. The attractive thing about CSP is that it’s cost is low enough that, unlike PV, it can compete with conventional sources right now.



    Why hasn’t Stirling Energy’s technology made more of a splash in the power business? “Our dilemma has always been how to get costs down,” explains Osborn. The dish assemblies now run $250,000 each. But that’s because most have been handcrafted in sporadic lots of one or two units. Building a group of 40 or so would trim the cost to $150,000 each, Osborn estimates. With real mass production, that could drop by 50%.

    So when SCE said it wanted to buy more renewable energy, Osborn’s outfit proposed the 500 MW project as the means of moving beyond its chicken-or-egg impasse. Producing that much electricity will require 20,000 dishes, built in a steadily increasing flow over several years. “We’re ramping up now,” says Osborn.

    He expects to have 40 dishes in place for a 1 MW facility by the end of next year, followed by 50 MW in 2008. The electricity will be delivered only when the sun is shining, but that’s when the utility’s customers place peak demands on electricity. “Our system is a really good match, providing peak power at times of peak load,” notes Osborn.

    The price per kilowatt-hour (kWh) that SCE will pay is confidential and must be approved by the California Public Utilities Commission. But there’s little doubt that the contract will get a thumbs-up, perhaps as soon as next month. One reason: SCE says the price it negotiated is so attractive — “well below the 11.33 cents per kWh” it now pays for peak power — that it won’t seek any subsidies from the state.

  34. Thanks RhapsodyInGlue. The thing about Sterling, another wonderful technology, is that it doesn’t have the easy, inexpensive storage of the solar thermal CSP’s Khosla is talking about (the steam-based kind). So Sterling might not use water, but it has similar storage issues as, say CPV (Concentrated Photovoltaic plants) or other photovoltaics. Again, I am not knocking any of the technologies. Just cautioning against looking to any one as “the winner.” Diversity and as much decentralization as possible, as Scheer and others suggest, is going to be key to make the radical and fast shift scientists are telling us we have to make to save life on this planet.

  35. Hegemony will only change with a change of heart (make that Spirit). After that -shore line based or otherwise- CSP will deliver abundant water. Making people wanting to move into the areas where CSP residual water is available, instead of fleeing to the rich countries & being at the mercy of people trafficers and the high seas.

    Transporting water has been done for millenia and is a minor challenge, when compared with a change of heart.

    Emil M

  36. I think a possible need for storage gets too much attention.

    First, most demand is during the day, and is correlated with or caused by light levels (especially A/C).

    2nd, Demand Side Management (DSM) appears to me much more useful, and much cheaper. Don’t forget the demand that primitive DSM moved temporarily to the night in search of cheaper power, which would move back if peak power decreased in cost.

    3rd, gas turbines provide a large % of electricity in the South West, including California, and they provide an easy way to match demand with supply.

    4th, there are efficiency losses to thermal storage, which reduce it’s value.

    5th, I have read that there are no installations of significant size using solar thermal storage – that it is more theoretical than real at this point.

    6th, by the time storage is needed, it is highly likely to be available from PHEV’s.

    I would estimate that solar electricity will have provide at least 10%, and probably 20% of the kwh’s in a grid before storage begins to be necessary. Even then, only a relatively small % of incremental power would be stored, in order to reduce supply variance and match supply with demand.

    On the other hand, I agree that this does not diminish the value of CSP, and that we need diverse forms of supply.

    I simply think that decentralized PV is going to grow much faster, due to it’s consumer-side, decentralized nature.

  37. Solar, even CSP, is still considerably more expensive than night time hydro, nuclear and other baseload. If you’re talking about things like aluminum production, it simply wouldn’t be economic to do it with solar electric. In other words, solar is not a way of bringing day time rates down to night time levels.

    In areas where A/C demand is the clear culprit for peaks, such as Southern Cal, I too believe that a fairly significant development of solar can occur before storage becomes an issue, as I stated before. However, there are many areas of the country where this is not necessarily the case. For instance, I was surprised to find that Florida’s absolute peak occurs during the winter, not summer, due to backup electric resistance heating that kicks in when it’s too cold for heat pumps. I ran across that interesting fact in one of the DSM reports that can be found at

    The “Solar Two” tower in the Mojave had molten salt storage tanks to extend electricity generation after the sun went down. Storage has been studied for both towers and troughs and found to be economically feasible. Solar Tres, the new solar tower being built in Spain will have 16 hours worth of heat storage… enabling it to run 24 hours a day during the summer and with an annual capacity factor of 65%. Some of the solar troughs being built in Spain will also have storage. It’s not THEORETICAL, unlike V2G. The reason some plants, such as Nevada Solar One, decided to not use storage is simply due to the low penetration of solar and the A/C peak in that area. Storage can also be added later if it becomes advantages to do so.

    The loses for thermal storage with CSP are not that great. This is because it doesn’t require an extra conversion. Solar Tres, for instance, will heat the molten salt to around 565C. The salt is then either sent to the steam generator directly or stored for later use. The conversion losses are not increased by the storage… unlike any other renewable (wind, wave, PV) storage schemes. The amount of heat lost from the insulated storage tank can be kept to a fairly small amount… significantly less than the other efficiency loss sources in the plant.


    I just ran across another source of info on CSP. At you can find workshop papers discussing many aspects of solar trough facilities including storage and also the issue of water use. I’ve yet to read through them, but plan to soon.

  38. “solar is not a way of bringing day time rates down to night time levels. ”

    Well, it is, just not real soon. I was really addressing the larger question of the maximum market share for solar, and my point was that if solar production began to exceed daytime demand that there was the potential for some night time demand to move back to the day, where it would normally be, depending on relative costs, of course.

    Similarly, morning and evening residential resistance heating peaks are entirely an artifical result of flat pricing (you wake up, and come home from work, and turn up the heat), and could be altered quite easily (with nothing more than electronic scheduling thermostats).

    That’s interesting about heat storage. I guess it makes sense that heat could be stored pretty efficiently. I suppose the main additional cost is the extra capital expense of the storage – I don’t know what the relative size of that expense is.

    I would note that I wasn’t talking about V2G, but about something much simpler, which was simply managing the charging of PHEV’s. This would be an easy way to buffer intermittent power generation by renewables.

Comments are closed.