Solar power is now cheaper than nuclear | Hacker News Reader

Solar power is now cheaper than nuclear

ncwarn.org

35 pointsbrkumar53 comments

No, it's not. The position paper (from an organization whose whole mission is to push solar and get rid of nuclear) simply makes up a bunch of number. http://v.cx/2010/07/lying-about-solar

Also, this: http://atomicinsights.blogspot.com/2010/07/gullible-reportin...

  Pro-nuclear advocate

as the first post is pro-solar, this one is just the opposite. both posts are nonobjective, there's still quite a lot of religion involved in the discussion.

The top post is claiming something that is not true. The responses are pointing out that the claim is bullshit, but they are not making claims of their own.

Suppose I were to claim that Sasvari owes me one trillion dollars, based on no reliable evidence whatsoever, and Sasvari were to deny it, pointing out the lack of evidence. Are both sides in this scenario equally "religious" in their stances?

Here in Germany we add half of this years world's photovoltaic solar panel production. Much of that is installed by small consumers, houses, farms, factories. This gets us away from the huge expensive centralized energy production. Panel by panel.

A new nuclear power plant has not been built in he last twenty years and is not planned.

Thus we can concentrate on bringing the price of renewable energy down.

The new reactors that are under construction in Finland and France are both much more expensive than planned and are coming years late. France also has to invest several hundred million Euros per reactor to keep them running. They haven't done much for those - a full fifty reactors. Here in Germany we need to spend some ten more more billion Euros on closing old East German reactors and for cleaning up an experimental storage site for nuclear waste. That experiment has failed. Spectacularly. Billions and billions that could have been invested in clean and renewable energy.

I really wish we had the same policy here in the UK. Here it's extremely rare to see a solar panel anywhere, and it seems highly likely that new nuclear (fission) reactors will be built in the near future.

What, to capitalise on the eight days per year of British sunshine?

There's a British company that makes evacuated heat-pipe solar hot water elements. They work very well in England in the winter.

http://www.thermomax.com/

If there's enough sunshine to grow grass and trees, there's enough to make energy.

>>The new reactors [..] are both much more expensive than planned and are coming years late.

There hasn't been many nuclear plants built for quite a while. Teething problems are to be expected.

That said, there seems to be strong arguments for solar power's cost going so low that it takes over in a decade or two. The world should probably be thankful that Germany waste lots of money to hasten that date.

But solar energy won't solve the energy generation in Finland or the rest of Scandinavia (hint -- think sun hours/angle in winter :-(.)

But maybe Polywell or General Fusion works and solves all problems...?

For far northern countries, I'd put my money on advanced fission designs and perhaps Thorium.

Or they can just install their panels in another country and pay someone a bit for the use of that land.

We tend to be wary of such solutions after being screwed by pretty much everyone in the last big conflict. Sure, things are different now in many ways, but ...

In that case - Uranium from seawater!

Hmm. Except that 'huge expensive centralized energy production' produces electricity much more cheaply than the solar panels.

If this wasn't the case, the German government wouldn't need the huge expensive system of subsidies for PV.

By all means spend lots of money to encourage production of more expensive electricity, but don't pretend it's cheaper.

This is an issue that's taken care of by the advancement of technology in the end. Carr describes well the economic sense for centralized power production in his book "The Big Switch," but centralized power production using legacy technology doesn't lend itself nearly as well to cost/benefit ratios as for instance units of computing power which have gotten substantially cheaper over time, so mainframe computing fell by the wayside. PV technology is more akin to computer technology than legacy power technology, so local power production using PV will end up being cheaper, provided your local utility doesn't step in and squash PV for their own benefit. Means becomes ends.

If PV technology is indeed akin to computer technology (how, exactly?) then perhaps subsidies aren't needed, just as they weren't needed in the arena of computing. I am just not convinced that the government funnelling huge amounts of money towards buying technology that isn't economical results in the technology becoming economical any faster.

No subsidies in the development of computing? That's a joke?

Wait, who do you think has paid for the research. Ever heard of ARPA or DARPA? The whole computer technology came out of military applications paid by the tax payer. Ever heard of the origins of what we know as Internet, the ARPAnet?

Google for 'secret history of silicon valley' and view the movie.

That is irrelevant to the issue, which is that of government subsidies on the purchase of computing technology, for end-users. No doubt NASA have done some solar panel research - that would be more analogous to what you mention.

The only example I can vaguely remember was the funding in France for everyone to get some sort of modem-connected device. I don't believe this advanced computing hardware at all, and didn't place France in the lead of the computing technology race.

You're talking about Minitel: http://en.wikipedia.org/wiki/Minitel

It's still in use and generating revenues 28 years after introduction.

Can someone post numbers on the amount of subsidies that are given by the German government? The Belgian system of paying out for each 1000 kW produced is gradually being phased out; For those installed in 2008 and 2009 you got 450 euros, this year 350 and over the next few years this will decrease to 150 I think. Plus you can deduct a mortgage used to buy solar panels from your income. I modeled this fairly extensively a few years ago and it turn out that yields of 7% on the investment are expected. My parents are making more than that because of lots of sunshine over the last few years and higher energy costs so more savings. It's an awesome investment, and unparalleled in its risk class.

I've done some consulting in the area but don't have any hard numbers at hand right now. But basically, it's lots of money. If the subsidies are taken away then the market collapses. Germany isn't the sunniest country in the world yet was (is?) the largest market. Spain had a recent boom, which was followed by a bust when the subsidies were removed.

France has been playing catchup in this area and they were proposing subsidies that would reduce each year to better match reductions the cost basis of PV. They were predicting ~20 years for solar to reach grid parity.

A more interesting debate is whether governments should be interfering in free market forces to push some technologies over others.

Oh sorry what I meant was more how much an individual can make on these subsidies. I know the aggregate is significant and not in relation to the immediate effects. I think it's also, on a rational level, quite clear that the redistributive effects of the subsidies are of dubious nature and politically motivated; every tax payer contributing to elevated savings yields of those with 20k to spare or the creditworthiness to borrow as much ranges from silly when one is in a forgiving mood to outright immoral when one is not.

Maybe I was diverting the conversation, I was just pointing out that those who installed PV panels in Belgium got a very nice yield on their investment. I haven't run the numbers with the latest payouts and panel costs (PV panel cost has come down significantly already over the last few years) so I don't know the state of the art. I was interested in comparing subsidy schemes with other countries.

(as an extra data point, the Netherlands provides only a fixed subsidy for PV panels, and there's a fairly low ceiling to the total that can be paid out a year; based on a first come, first serve system. The period to recuperate the investment is about 20 years, which is quite long; in Belgium one can earn back the investment in about 6 years and everything after that is pure profit).

I apologize in advance for another short comment, but your reply leads to another interesting point: should entities (individual, companies) be using payback period as a criteria for investments? Everything I learned in b-school pointed us towards using NPV as the best metric, but many entities still use IRR or payback period. I'll try to follow-up tomorrow with some numeric examples so you and everyone else can see the implications of various methods. But the stuff I saw from calculations in France (the most generous market for building-integrated photovoltaic installations) was on the order of 10 years.

Your 6-year payback seems a bit optimistic but I last looked into this stuff 8 months ago and things may have changed.

No, payback period is a silly metric, and my 6 years do account for opportunity cost :)

Actually I tried to make a product out of this. The website is still up on http://www.solar-profit.info/. It's in Dutch but the screenshots speak for themselves. I never sold a single copy, for one because my market is about 500 customers big, and second because I didn't manage to explain the concept of time value of money to my leads (installation companies for PV panels).

So yeah that's the answer to your comment. Payback period everybody understands, and that's the only thing it has going for it, but unfortunately it's such a big advantage that you still need to use it if you want to convince some markets.

Energy production was never a free market and never will be.

The nuclear industry is best run without competition. See France. The whole energy industry there is like soviet-style centralized economy. This leads to higher costs and at the same time to low safety. France for example brings a lot of nuclear waste to siberia where it is stored just under the sun, outside. You would never get real costs out of the nuclear complex in France. It all cleverly hidden. Yet they can't even produce enough energy for their country and deliver it. Last winter the power lines in France we almost collapsing. Heating is done with electricity and the winter was long and cold. The whole structure of energy production is very centralized (and not mesh like). At the same time 20% nuclear reactors were off the grid and Electricity had to be imported.

Even for a conventional energy production it is important to have competition. But the big energy producers are getting monopolies for their regions. In France it is even worse: there is EDF and not much else.

Next, the nuclear reactors are not using their heat in any useful way in France or Germany. They are mostly used to heat up the rivers. Which at the same time limits how many reactors one can put at a river. They need lots of water for cooling.

Let me add the hidden cost of mining. From the Areva wikipedia page:

--- Areva has been criticised both in Niger and the west for a neocolonial relationship with the former French colony, having obtained monopoly rights to mine uranium for nearly forty years, while Niger remains (as of 2006) the least developed country in the world.[16] There has been specualtion that Areva was the primary vehicle (and motivation) for French government involvement in the Nigerien government's poor history of democracy and human rights since the late 1960s. Areva argues that it is a good steward of the environment and a partner in development with the Nigerien people. The company stresses that most workers it directly employs in Niger are locals, and that the revenue generated by these mines is the single largest foreign exchange source for Niger.[17] When the population of Niger was exposed to a serious famine in 2005, Areva donated 130,000 euros in June 2005 to the food crisis coordination group of Niger, and 120,000 euros in July in the form of two planes loaded with food and organized by Bernard Kouchner's Réussir NGO. Nigerien critics point out that this aid amounted to 0.05% of Areva's annual profits of 428 million euros.[18]. In January 2008, a Swiss human rights group awarded Areva's Niger operations their "2008 Public Eye Global Award" for the "worst company in the world", citing charges of radioactive pollution, poor working conditions, and the deaths of a number of workers.[19] The Tuareg based rebel group fighting in the Arlit area has echoed these charges. ---

People in the developing world suffer from injustices, but it's silly to argue that this is a hidden cost of Nuclear. If the reserves were in a developed country they would be mined as cheaply and more safely.

And don't forget 'the hidden cost of PV' - massive pollution in China, bad working conditions, damage to workers' health.

"People in the developing world suffer from injustices, but it's silly to argue that this is a hidden cost of Nuclear."

Right, just as Superfund cleanup and cancer treatment is free to the US taxpayer.

"If the reserves were in a developed country they would be mined as cheaply and more safely."

What prevents Areva from doing that in Niger?

I've not followed what's happening in France (there is, shall we say, less transparency in that country), but the Finnish reactor is having problems due to lax supervision of (sub)contractors: foundational concrete was not done correctly, forgings have to be recast, etc.

These aren't nuclear power issues per se but general competence of construction (supervision) issues. I gather the French site has some of these sorts of problems as well.

Getting onto strictly nuclear power issues, these are the first two attempts to build the new European Pressurized Reactor: http://en.wikipedia.org/wiki/European_Pressurized_Reactor. It's not at all surprising if the first examples of a new design end up being more expensive and slower to build (which of course also means more expensive): there's new learning curves, lots of new regulatory work required, likely redesigns required, etc.

Germany's experience with Soviet design East German plants would not seem to be applicable outside of the former Iron Curtain (although we can thank God none of them were of the RMBK design). Your failed "experimental storage site for nuclear waste" is just that, yours. You are of course welcome to decide that nuclear power is a failure and e.g. depend on natural gas from the former Soviet Union (or just living in the cold and dark), but most of the rest of the world doesn't seem to be following your path.

As for your desire for "clean and renewable energy", none that I'm aware of that can be done on a large scale (I'm excluding geo-thermal) are suitable for either base line or peaking power unless you throw batteries into the mix. Which aren't particularly clean, low maintenance or long lived. (There is one trick you can do which is to pump water uphill into a reservoir but the Greens generally veto that sort of thing, we've moved from NIMBY (Not In My Back Yard) to BANANA (Build Absolutely Nothing Anywhere Near Anything)).

>Here in Germany we add half of this years world's photovoltaic solar panel production. Much of that is installed by small consumers, houses, farms, factories. This gets us away from the huge expensive centralized energy production. Panel by panel.

Your solar capacity only exists because of extreme government intervention. It's far more expensive than what you call "expensive" centralized electricity.

The most recent feed-in tariffs are 31.9 - 43.0 Euro cents per kWh (2009 installations), fixed for 20 years. Older installations are given subsidies: the original rates were 45.7 - 54.7 Euro cents/kWh (2004 installations).

http://en.wikipedia.org/wiki/Financial_incentives_for_photov...

http://www.solarbuzz.com/FastFactsGermany.htm

For comparison, wholesale spot prices for utilities in Germany averaged 3.9 Euro cents/kWh for baseload in 2009 (through 3Q), and 5.0 Euro cents/kWh for peaking power. Residential retail rates (as of 2007 -- slightly dated) are about 14 Euro cents/kWh before taxes.

http://rwecom.online-report.eu/2009/ir/3/reviewofoperations/...

http://www.cbs.nl/en-GB/menu/themas/industrie-energie/public...

>A new nuclear power plant has not been built in he last twenty years and is not planned.

Your government banned them.

>The new reactors that are under construction in Finland and France are both much more expensive than planned and are coming years late.

Yes, they've had enormous cost overruns and are years late. And German solar power is an order of magnitude worse.

We're talking about Olkiluoto #3 (Finland) and Flamanville #3 (France), the two new reactors being built in Europe. Both EPRs of 1,600 MW(electric) capacity (net). Olkiluoto had an inital cost estimate of €3 billion, and is overrun by 50% and counting (so €4.5 billion so far). Flamanville' overrun (to date) is smaller, from €3.3 B to €4.0 billion (20%). (These are the most recent numbers I can find; correct me if you find better ones.)

http://www.nytimes.com/2009/05/29/business/energy-environmen...

http://www.bloomberg.com/apps/news?pid=newsarchive&sid=a...

Crucially, nuclear reactors actually achieve their nameplate capacity -- they run at full load almost all the time. For example, look at these reactors' older siblings: Flamanville #2 has a lifetime load factor of 70.53%, and Olkiluoto #2, 93.55% (IAEA data). And this is typical. A 1,600 MWe EPR can conservatively be expected to average 1,100 MWe or more.

http://www.iaea.org/cgi-bin/db.page.pl/pris.ophis.htm?countr...

To date, all of German solar panels (i) add up to just one of these EPRs at best, (ii) have earmarked at the minimum €70 billion in subsidies between them, over their next 20 years. To be fair the present value of that €70 billion is a bit less (it's future money). But it's eye-opening.

According to solar industry sources, there are some 11,500 MW of solar PV capacity in Germany (9.8 GW through 2009, and another 1.7 GW through May 2010).

http://www.pv-tech.org/news/_a/germanys_solar_capa567_by_3gw...

The actual amount of electricity is only a tenth of that. If you look at PV power plants in Germany, their load factor is only around 11-13%.

http://en.wikipedia.org/wiki/Solar_power_in_Germany#Largest_...

Which you could independently predict by looking at a solar insolation map; for instance 1,000 kWh/year from a 1 kW capacity panel is a load factor of 11.4%.

http://en.wikipedia.org/wiki/File:EU-Glob_opta_presentation....

So 11,500 MW of German PV capacity should generate only 11 billion kWh/year, or 1,250 MW agerage. All of it adds up to about one EPR. (To anyone actually reading this, I recommend Google Calculator for these kinds of calculations, type-checked with physical units like 'kWh').

And a lower bound for the committed subsidy is, using the lowest possible rate of 31.9 Euro cents/kWh, €70 billion:

http://www.google.com/search?hl=en&q=%2811%2C500+MW+*+0....

(The "PV tech" article gives a higher figure of €100 billion; I'm not sure what their assumptions are.)

That's easy - you just impose expensive environmental regulations on top of building nuclear reactors and, suddenly, solar power becomes cheaper. With a couple iterations, even hamsters in cages will surpass nuclear.

How green is solar panel production, BTW?

> How green is solar panel production, BTW?

The term "green" is probably too vague to answer that question. However, I once read (a long time ago, so no link) an article on some research that reported that EV solar was net-energy negative, i.e., if you add up all the energy that went into producing EV solar panels and subtracted the energy output over its expected lifetime it was negative. This is not true of hot-water type solar just EV solar panels. Of course the conclusion rested on numerous assumptions but they used conservative assumptions and argued that EV is not an energy source and would need some wild breakthroughs in reducing the energy costs to manufacture or a huge jump in energy output efficiency neither of which, as far as I know, has happened.

Edit: I got that reversed, Energy Output - Energy to make < 0

Times change, 50,000Hz Computers used to be top of the line and cost thousands of dollars, but try selling such things today at those prices.

The value of electricity produced over a modern solar cell’s lifetime > cost of that solar cell today so clearly they take less energy to produce than it takes to make. However, ROI is more than just getting more money back than something costs the payback time is important.

PS: Granted, Ultra high efficiency solar cells built for the space program still cost a lot. However, they are producing energy in space so the incentives change.

dmfdmf's point is that it takes more energy to make a solar cell than it will ever produce during its lifetime.

I am not sure his numbers are current (he made the disclaimer), but it has nothing to do with energy pricing.

My worry is not whether photo-voltaic is net-positive, but whether the production actually reduces pollution. For what I know, they are very close cousins of computer chips and those factories were not particularly "green".

Energy is a complex topic, so calculating the actual energy costs to produce something is hard because different forms of energy have different value. EX: heat pumps don't generate energy even though they can output more BTU's than an electric heater.

However, electricity cost to generate other types of energy works as a shortcut to a lot of this stuff. So generally if something costs X$ to manufacture it takes less energy than X$ worth of electricity.

As to being green; PV is not exactly a single technology with a single type of environmental impact. Assuming you recycle the cells, PV can have a very environmentally friendly footprint, but the factory must also take care to avoid a wide range of pollution.

PS: It’s hard to talk about energy without using enthalpy, entropy, and other thermodynamic concepts. If my point is not clear I can try a more detailed explanation.

In Australia, we have one of the cheapest electricity prices in the world. Wholesale(national pool price) is about USD 65 per MWh + Network charges.

But Solar PV electricity is 1/3 cheaper than the Retail price of electricity WITHOUT ANY SUBSIDIES.

The current Peak Retail price (Midday - 8.00pm)is circa USD 0.29 kWh (Aust$0.32kWh) and the 'real' unsubsidised cost of solar PV is about Aust $0.21 kWh.

With the current subsidies the price is NEGATIVE $Aust 0.50 kWh (USD -0.45)

Yes, we get paid to produce through grants and a Feed in Tariff

If Australia is on the ball with infrastructure and government policy, your little private continent might end up a bastion of energy stability in the post peak-oil world! You have your own deserts with lots of potential for massive Solar Thermal. Add some high-tech distribution, and you're in a very enviable position.

They sell solar panels in Menard's now, freakin' Menards (A mid-western version of Lowe's/Home Depot). I think they cost 300+ for 60 watts. I for one would love to put a bunch of solar panels up on my house, just for the cool factor.

Not to mention, centralized power is vulnerable. Terrorists, grid failure, etc. Something so vital should be so centralized. I'm going to dub the coming DIY power generation "Cloud Power", remember you heard it here first.

What the south needs now are cheap 12 volt AC units. Install some solar PV, and have regulation circuitry that turns it on when the panels are producing enough juice. This would still save a ton on summer AC power costs, but it would be (relatively) cheap. (No batteries! You'd have to have a hefty capacitor for startup surge power, though.)

Here in the UK, you can also harm the environment and emit carbon by buying a windmill from the DIY store and sticking it on your roof. People are mostly doing it for the 'cool factor' since they tend not to generate much power at all.

I'm going to dub the coming DIY power generation "Cloud Power", remember you heard it here first.

You've got that the wrong way round. I already get my power from 'the cloud' rather than generating on premise :)

As has been pointed out this study is filled with a lot of stats and little methodology. They assume that solar tech is getting cheaper and cheaper $/KwH (quite valid), however they seemingly think that nuclear is frozen and the tech costs are rising.

I think it could be argued that the lobbies subsidizing both sides are causing each trend. Westinghouse wants subsidies so more plants are sold, however that also increases the tendency to over-build. A friend of mine in their nuke dept. said this very true. Compare western to foreign nuke prices.

Also, the green movement has every interest to encourage increasingly strict nuclear regs that make it difficult to make new ones while they support solar subsidies.

All of this leads to nuclear power going up in price while the tech remains underutilized. This report shows these price trends, but does little to elucidate the underlying causation in them.

Is solar still coming down in price? What kind of bothers me is even if the panels were free it would still probably cost $10,000+ to have someone install them on my house, thus I'll probably never be able to do it :-(

10k for just the labor? Nah. It takes 2 persons 1 day to install PV panels on a 'regular' house, including the carpentering, electrical installation etc.. That's 1000$ or so in semi-skilled labor.

I think a new roof costs like 20K. So I figured covering my roof with something would cost at least half of that. Is it really that fast?

Panels are put on the roof, you can't use them as a roof (actually there are panels nowadays that can do that, but I'm talking about the 'mainstream' products). Assuming you have a good roof, what they do is open it up in a few places to anchor supports to your roof structure (the rafters) and then put the panel on those supports. The amount of work that the opening up and closing the seams takes will depend on your roofing; some types are easier to work with than others, but still that wouldn't make more than a day's work of difference.

A typical installation has only about 10 or so panels. Once the supports are on there, you hoist the panels up, screw them to the supports and that's it.

In addition (I'm on a roll now ;) ) a complete new roof including materials will cost let's say 20k. The actual work of that is, I'd estimate, only about 5k or so, the rest is materials But then you get a complete new roof with new joists and the whole shebang, which is quite a bit of work.

(of course it's also going to make a difference if you cover a roof in Michigan or in Texas, or in NYC... )

What bothers you (in the present) is that if solar panels were free (they aren't), it would probably cost (a number pulled out of thin air with no research, therefore deliberately chosen to be high enough to annoy yourself) to install them on your house, which you predict negatively that you wont be able to afford ever.

Are you always this cheerful and optimistic? Have you considered ... being happier?

Wow, when you put it like that it sounds ridiculous :-)

I'll try to have more optimism for the solar industry .. maybe in general too.

This is a bunch of anti-nuke nonsense written by non-experts in the energy field.

Base load.

Also W/m^2

Reply on news.ycombinator.com