Solar panels will always require a backup power plant as the Sun isn’t shining 24/7 and storing large amounts of electric energy isn’t trivial.
I‘m really disappointed that this kind of non-sense gets posted on HN over and over again.
I‘m from Germany, we have 50% renewables in our electricity mix and our electricity prices are the highest worldwide.
France has 70% nuclear and their consumer electricity prices are half of the German ones.
Additionally, France emits only 50 grams of CO2 per kWh while Germany emits 400 grams on average per kWh.
I know lots of comments will counter-argue „Yes, but the high taxes in Germany.“ but those aren’t the main price drivers.
The main price drivers are the costs for subsidizing solar and wind (the so-called EEG-Umlage) and the costs for keeping backup power plants available which are allowed to produce only when solar and/or wind won’t deliver as both solar and wind have precedence over other electricity providers in the electricity net (the so-called Netzentgelt).
Solar might be cheapest when you consider the pure production costs per kWh but it’s not the cheapest when you consider the costs for a conplete system to provide electricity for a population.
Even single houses aren’t able to live off just solar while every house can live solely on power from a nuclear power plant which provides electricity 24/7.
And in Germany and Denmark, taxes and levies ARE the main price drivers. The maintenance costs are amortized in the per-kwh cost to consumers, same as with nuclear or gas or coal.
Single houses can't live off just solar energy because they aren't diversified like a nation can be (in terms of storage, routing, diversity of technologies, etc).
Nuclear is nice in theory, but unfortunately they are managed by humans, and humans do stupid things like run them past their decommission date or fail to locate backup generators above flooding level or skimp on maintenance and inspections. And nuclear accidents create no-go zones that last millenia. Only truly "meltdown-proof" reactors would avoid this problem, but R&D into this is terribly slow and seems to be obstructed at every turn by incumbents.
In particular, those prices aren’t stable and are often negative or soar up to 4000 Euros per MWh in Germany.
Please, for the sake of an honest discussion, look at the actual data in Germany.
And, no, there is neither a strong fossile nor nuclear lobby in Germany. The renewables lobby is - by far - the strongest lobby as the overwhelming majority in Germany supports renewables.
Germany has already invested more than 500 billion Euros for the transition to renewables and we are still among the dirtiest electricity producers and we are still building new fossile plants.
100% renewables based on wind and solar does not work and will never work on a large scale.
However, i have studied Gen3 and Gen3+ designs, i think Gen3 design are really safe enough, even if knowledge decline as much as it did on Trantor: anybody with small training could keep one functionning if they follow the simple rule: if something unexpected happen, Don't touch anything.
Gen3+ add a new surface of attack, a digital one, while the main issue in Gen3 is not fixed (for information, the main security issue is the pool). I'm not a fan.
Still, Gen3+ and their core catchers are overengineered and i think are not worth the additional cost in GHG.
Having said that: even a contained meltdown is devastating for the utility. The owner of TMI was gravely wounded by the cost of that accident.
These renewables produce a lot of power that isn't needed during their peaks, which gets sold at low or negative prices on the market, yet is compensated for with the full subsidy.
During their troughs, you need another source of power, usually natural gas, which is expensive, or coal, which raises the baseline production and therefore exacerbates the peak production problem. The result on CO2 emissions is pretty damning.
This is the danger of picking "winning technologies", you are telling the market to waste resources. Instead, you need to incentivize the market for your actual goal and then let it do its thing. Putting a price on CO2 would've incentivized all sorts of technologies that reduce net CO2 emissions.
No problem of this magnitude is ever solved with a single-prong approach. It calls for some far-thinking: Figure out which energy sources will eventually be the cheapest to harvest sustainably, and then take a financial hit now for a bonanza later. Complaining about the price comparison today is just silly, considering that people in the industry only look at out-of-pocket money when making their decisions.
How long do you want to continue kickstarting? Until 2150?
And anyways fossil fuels have received and continue to receive far greater subsidies.
Frances issue is the demand curve isn’t a steady state so Nuclear needs on average more supplemental power than solar to make up the difference. Unless you leave plants mostly idle at extreme cost or have vast energy storage you can’t get to even 50%. Consumer rates are heavily subsidized and their close ties with Europe’s grid mostly hides these issues. However power plant utilization still hovers around 80% which means the country pays significantly more for nuclear than countries with better utilization.
Things would look much worse for France if surrounding countries increased Nuclear generation as the seasonal, night time, and weekend surplus would increase.
It's obviously cheaper to use existing infrastructure than to build new infrastructure.
One day Germany will be able to coast on the green infrastructure it is building today.
Nonetheless, if you take two countries with only coal and one only builds nuclear while the other builds solar and wind I would expect the latter to have far lower electricity prices based on LCOE and variability costs. Even without externalities nuclear is incredibly fucking expensive.
Seems to me that France has built up a lot of debt in terms of old reactors that need to be decommissioned and replaced. As far as I’ve read, their plan is to replace most of those with renewables (targeting 50% renewables)
I feel like I see this all the time - nuclear advocates being extremely nitpicking about cost and challenges with renewables, while ignoring or brushing off those of nuclear power. I’m not against nuclear power, we should put more R&D into it to hedge our bets. But I’ve yet to see a compelling case for a huge bet on nuclear power that’s not ridiculously simplistic.
We all know that energy storage and synthetic fuel production is the missing key to going all in on renewables. But we also know that breakthroughs in these areas are absolutely essential to making transportation CO2 neutral. So going nuclear doesn’t avoid solving that problem. I’d argue that going for renewables is a huge benefit in that it channels more money into energy storage, increasing R&D, making it easier to solve the climate crisis across the board.
We can return to nuclear power in 2-3 decades and hopefully free up some of the land claimed by solar and wind power plants.
Electric production tWh
Nuclear 64.3 39%
Hydro 63.6 39%
Wind 19.9 12%
CHP 8.1 5%
Industry 6.6 4%
Other 1.5 1%
Where CHP and Industry was 75% biofuels. Renewables are about 90 tWh. Production has increased over the last years mainly due to new wind power. Installed solar is still small (0.2% of production) but nearly doubled in installed capacity from 2017 to 2018 (231 MW to 411 MW). With gCO2/kWh about 60. 
Switzerland, the country I live in, also has a ton of hydro (60% ) but the thing about hydro is that it's entirely dependent on geography. Iceland has 100% renewable electricity  but that's due to its geography.
Same goes for all renewables. Wind depends on areas where wind is harvestable and solar depends on daylight hours.
Not saying this to say that you're wrong or your point is irrelevant or anything, just that the energy solutions applicable to one country don't apply to all.
Anyway I'll bite, I work at a startup that does demand side flexibility with electric hot water heaters (and working on other devices as well). We install our controller at customers homes and steer the water heating according to: electricity spot price, greenness of energy and also participate in frequency stabilisation markets. Customer saves by using cheaper electricity and earning money by allowing their heater to be used in the grid stabilization, in effect they get the money that would have gone to the gas peaker plant for firing up during an imbalance event.
A 300L water heater stores about 16kWh of energy as heat and has a heating power of about 3kW. Aggregating thousands of homes together and making sure the water is heated when it's best for the system is one of the ways we're working on the problem of intermittency.
The funny thing is that in our key market (Finland) the homeowners have dimensioned their water heaters to only heat at night when the electricity was usually cheaper. This meant that a family of 4 could use water all day without it getting cold and heat it during the day. Now we can flip this around and store energy when it's best.
Demand side flexibility is key (along with storage and electrification of transport and heating) to enable green renewable generation to come up to the levels we need. Sure, we don't have solutions for the last 1% of the problem but we're far far away from the last 1%.
Intermittency is only a problem when one has a medieval image of self-sufficiency in mind. Solutions are either averaging out peaks and dips temporally (by storing energy e.g. in pumped hydro plants, in batteries, ...) or geographically by moving power where it is needed and from where it is currently being produced. You do not even lose gigantic fractions of the power in the process: high-voltage direct current power lines have losses of around 3%/1000km. China, for example, has power lines that move a dozen gigawatt over a distance of over 3000km. It can be done.
For physical goods, we have ditched self-sufficiency a long time ago. Your government might sustain emergency reserves for grain, food or oil and gas but even that lasts only a few weeks. Here is what Germany has stockpiled for emergencies for example: .
Mutual dependency does not have to be a bad thing either. It keeps both sides from doing stupid things (assuming a balanced level of dependency in both directions).
 (german) https://de.wikipedia.org/wiki/Bundesreserve_Getreide
 (german) https://de.wikipedia.org/wiki/Zivile_Notfallreserve
 (german) https://de.wikipedia.org/wiki/Strategische_%C3%96lreserve
From wikipedia :
```Worldwide growth of photovoltaics is extremely dynamic and varies strongly by country. As of 2020, there are at least 37 countries around the world with a cumulative PV capacity of more than one gigawatt. By the end of 2019, a cumulative amount of 629 GW of solar power was installed throughout the world. By early 2020, the leading country for solar power was China with 208 GW, accounting for one-third of global installed solar capacity. By the end of 2016, cumulative photovoltaic capacity increased by more than 75 gigawatt (GW) and reached at least 303 GW, sufficient to supply approximately 1.8 percent of the world's total electricity consumption.```
So the installed power in 2016 is at least 303 GW. Let's check how much it produce (and btw, solar is not like wind or water and can't easely be disconnected from the grid, at least for large solar farms, so broadly: solar power generated = solar power used).
According by ourworldindata , 328.2 TWh in 2016 and 724 in 2019, so it roughly double every two years. That's the good news.
And the cost have nothing to do with this. The bottlneck is twofold: the installation (peopl trained to instal the solar
panels as well as the land usage, and the other one is the production. That can keep doubling every other year, but one thing won't, and that the mines. The growth decline already started for wind power generation because of the first bottleneck (the demand don't follow the offers), i'm afraid the second one will limit the solar panel usage.
I've read that installed solar was 3% of installed electricity in 2019, so roughly .3% of electricity used. Electricity generation is 1/3 of our GHG emissions, and we need to limit those to 0 by 2050, or at least by 2070. This is not hte way. It will help, but this is not the way.
There is no this way (i.e. no 'silver bullet') to fix the climate crisis. If we want to tackle the problem, there will only be lots of smaller it will helps that - in total - will make a difference then.
To store just a fraction of the needed power would require millions of large expensive batteries, whereas pumped hydro storage requires geological features that are in limited supply.
Once you factor in all these costs, suddenly the "cheapest electricity in history" becomes really expensive.
No country has solved large scale energy storage that would be needed for storing electric energy in the range of TWh.
But if you are factoring in total costs for nuclear, you also need to factor in the security risks and costs (eg terrorist threats to nuclear plants), costs to dispose of nuclear waste, inability to provide housing close to reactors, etc, as well.
Yes because the other 50% is powered by coal.
>not the cheapest when you consider the costs for a conplete system
Same with nuclear and coal, you dont count the potential health problems (coal) and the 10000y storage prices for the waste (nuclear)
In fact, we’re building 16 new gas-fired and 2 new coal-fired plants.
Boop, you're wrong about the first claim. Storing large amounts of electricity isn't trivial, but neither is building a backup powerplant (or really anything at that scale). In my region there are multiple grid level battery storage facilities with batteries from decommissioned electric vehicles and as far as I know they're also financed through EEG-money.
I haven't yet done the research to see how much of that storage we'll need, but I'm crazy enough to claim that of course we can do it. Though, also certainly, that will increase the systemic cost of solar power.
But nothing is free and the transition to renewable energy is inevitable and needs to happen as quickly as possible.
It seems rather convenient that the average electricity consumption is also lower when the sun doesn't shine.
Here's a Wikipedia page on cost of energy per source (in LCOE) that has a table of sources for Germany. According to it, solar is 37-115 €/MWh, while CCGT (gas) is 78-100 €/MWh:
I'd be interested if you can guess the fantasies i have knowing that.
Also, endogenic species are dieing faster than the forest replace them, forests are dryer, this summer we were 50km close to have the Russian weat fields burn more than they did in 2010 (with effects on the world hunger and politics in 2011 well know). This could be mitigated honestly, and i'm pretty sure that this, we can manage. Katrina every year, maybe not as much.
I can't, please elaborate.
The solar panels are expensive, but the sunlight is free. The furnace to burn wood is cheap and can be turned up/down as needed, but of course the wood is expensive, A fine backup.
But this is a significant change - Germany, from what I recall, moved to solar when it was more expensive and less reliable than their existing situation. In future, people moving to solar will get cheaper and less reliable.
That is a very different situation. There is incentive to take the less reliable power now. Uses will be found for it. I'm happy to make changes to my lifestyle if it saves me a bunch of money, and industrial consumers will start to adapt over a decade if the incentives line up.
It won't be the end of reliable power generation, but if it is economically sensible then the options will be explored deeply.
The prospect of continued decline in costs of renewables and storage is real enough to deter investment in nuclear. The market is rendering its verdict.
Yes, it is accurate that solar can't supply more than 30-40% of the energy mix. But we are still far from reaching that level of supply in the developing world and faster the cost reduces, the better it'll be for continued deployment en masse. Battery innovation as well as other tech will continue to improve the amount of solar in the energy mix.
Edit: You may also find this helpful - https://de.energy/blog/solars-future-is-insanely-cheap-2020-...
When I see new solar projects that have tendered absurdly low prices, are the bidders bidding with today’s solar prices or are they betting that solar prices will fall further, allowing them to eventually make a profit? For example Adani solar won a contract to supply energy at INR 3 or $0.05 per kwh. Is that the price of building and operating a solar in 2022 or is it the price that the winning bidder hopes it’ll be in 2024 when they’re building the second and third phases of their contract?
For example, at a cost of 30,000 INR/kW ($405), and a bid price of INR 3, assuming yield of 4.25 kWHr/day per kilowatt, we are looking at an IRR of 10.5% over 20 years. Now add long-term debt to the mix and we are upwards of 12%. Solar doesn't have any moving parts and this assumes operations and maintenance at 15%. If the cost squeezes further by 2024 - the IRR is even more attractive.
Goldman, Walburg, Pension funds, etc are deploying billions in solar for this reason - IRR upwards of 10% over 20-25 years with low risk. That's why we started this business too :)
Since you’re so on well informed about this here’s another question. I notice that battery prices are roughly halving every 3 years. At what point does it become cheaper to attach batteries to Indian solar plants than it is to operate coal powered plants?
A lot of them offer warranties for 20-25 years. Most tier 1 panels are expected to last 25-30 years.
- I notice that battery prices are roughly halving every 3 years. At what point does it become cheaper to attach batteries to Indian solar plants than it is to operate coal powered plants?
At a grid level this is a complex question that I'm not completely qualified to answer because it goes into the larger question of energy mix and energy security. At a factory or building level, we are not that far - we're already offering monthly packages (in terms of cost) which amounts to 10-15c/unit of a blended rate (solar + battery). The min it breaches 8c/kWHr blended rate, this can become more permanent. I'm speculating but some of what I've seen in Li-ion and ESS flow costs we may be two years or less out.
Is anyone using excess solar energy for desalination plus pumped hydro?
No experience w/ solar for desalination or hydro as its outside our focus right now.
solar can't supply more than 30-40%
of the energy mix
Solar + battery is where the future of solar is, but the economics of batteries don't work yet relative to grid electricity price in the developing world. So far we've deployed solar+battery in petrol pumps and a few off-grid locations. In the markets where we operate, on grid, it is economically not an option yet.
So in a world of rock bottom interest rates, incentives, subsidies and "revenue support mechanisms" geared toward green projects, solar is the cheapest alternative? That's not surprising; it's the goal.
The money ecosystem in play now favours renewables more than non-renewables, and that's reason to celebrate.
We're not so fragile that cloudy days will shut down our energy grid.
(Away from the equator, seasons are more significant than cloud or day/night cycle..)
Again, I’m asking out of curiosity. My expertise is a few Reddit threads and some cursory Wikipedia research.
uses real historical weather data, and surprisingly the cheapest energy systems it models for synthesized baseload from renewables + storage is in... Greenland?! I'm not sure I really believe that, but it made me sit up and take notice.
Storage methods need tens of trillions of investment to reach that capacity and power prices would have to increase 10 to 100 fold to cover those investments.
Solar is not a solution for an industrial civilization.
That’s adding x7 over the “cheapest electricity in history” solar and x2.5 for the relatively expensive solar in Europe and the USA, definitely not x100.
And even that’s assuming you have to store all the energy before it gets used. Before I left the U.K., I had two different rates for energy depending on the time of day — cheaper at night because demand was lower.
It still saves you money, because the fuel costs more than the solar power. It also reduces CO2 emissions in the period the sun shines.
I guess your counter argument is “but it doesn’t remove ALL the CO2!!”
But why should it? That is a long term goal. First we remove all CO2 emitted when the sun shines. We still got plenty of work to do there. Only when that has been done all do we need to think about storage. By that time storage solutions will be cheap.
Yes it literally does. My employer has shuttered three out of their fleet of nine gas power stations because they are no longer peak competitive with batteries. And base load designs aren't competitive with solar since they are more expensive and will only be turned on during extreme events, like a snow storm, or an extremely hot night.
Sucks to be us when we need electricity during an extreme event though.
> Storage methods need tens of trillions of investment to reach that capacity and power prices would have to increase 10 to 100 fold to cover those investments.
This sounds like you’re claiming your employer made electricity 10 to 100 times more expensive.
Yes, producing electricity when it's needed is more expensive to producing it when it's easy.
I’d still disagree (PV in the right places can be so cheap that it can already make sense to get through winter by putting down x3 as many panels as you need for summer), but it would be clearer.
I have very much looked at what the prices are doing and the answer is "making my employer filthy rich and making the grid unsustainable over the next decade".
Solar and batteries are eating the profit from base load and peak generators. Yet are unable to provide the sustained power requirements to keep the grid going on a day that gets over 35C.
If you like having power in hospitals at all hours you can't let solar and battery storage invade the grid without regulating it such that the grid survives.
Nationalization of the grid assets and a massive nuclear building spree is the only solution if we want children to grow up in a world where brownouts are the exception.
> Nationalization of the grid assets and a massive nuclear building spree is the only solution
I'm not sure even you know what you want.
Only in America could the two be conflated.
A market does not work for a natural monopoly.
Turns out telling businesses what they can and can't do just means they hire smarter people to do what they want in more convoluted ways.
If you have time, I’d suggest reading the following, and — if you feel like it — rephrasing your argument: https://wiki.lesswrong.com/wiki/Inferential_distance
Unfortunately you tick neither box.
If you're interested my hourly rate for this type of work is $300, you're welcome to commission a report on the country of your choice.
I totally understand the sentiment; thing is, this isn’t a quant forum, it’s mostly tech, and even then more software than hardware — either you should take the downvotes as a hint to simplify, or as a hint to not bother with that topic here.
Again, I really do sympathise with your frustration: no matter how I try, I can’t get non-programmers to understand even the basics of cryptography and why police master-keys/backdoors are a bad idea.
Usage and production are two different things. You can produce more energy than you need and have a stable grid by just throwing money at the problem. In 2016, you had EEX spot prices around 2-3 ct/kWh, below the production cost of any technology. This signals waste and it's also a major reason why renewables rollouts stalled in Europe.
Not just storage but we need to work on energy efficiency as well. We just smoke a huge amount of our generated energy.
Being told that "we have locked in enough climate change already that by 2100 industrial civilization will be impossible anywhere and by 2200 so will agriculture." is not something anyone wants to hear.
If we were serious about climate change we would be spending the next 20 years under martial law building nothing but nuclear powered CO2 scrubbers and producing enough food to keep us alive. Anything else is superfluous.
Or you can vote for Biden and get a vague promise that by 2050 we will be emitting as much CO2 as we were in 2000. And the less said about Trump the better.
The difference in policy options in the US is like the difference between jumping out of a plane and jumping out of a plane tied to an anvil.
It's actually a real thing with the gradual roll out of HVDC between countries. Eventually the gaps will be filled.
(And that’s assuming you get less than 60% losses going from electricity to light, through the atmosphere, back to electricity; and the ground station is free or negligible cost; and nobody worries about orbital death rays either malicious or accidental).
And (/s notwithstanding) people have considered floating PV on reservoirs to avoid evaporative losses.
Floating PV is pretty interesting, has anyone done it on a large enough scale for the economics/efficiency of it to be evaluated?
I don’t know enough about economics or statistics to answer your question, but here is a story about a pilot project: https://www.adb.org/news/features/inside-southeast-asias-fir...
Realistic in the sense of “if the $2 trillion per year currently spent on energy was spent on bootstrapping known tech but in space”, rather a lot more.
Many cheap things have hidden downsides, but not all of them, and some of them have fewer and smaller downsides than you seem to be assuming.
Germany is a poor example as it shut down perfectly fine nuclear power, thus incurring a cost on replacing it. UK has cut emissions more than Germany but had lower prices.
No. Where did you even read that? EDF gave France government 20 Billions in dividends over 10 years, not the opposite. While selling electricity on average 42 euros/MWh. No subsidies, and EDF pay for the whole French network AND the international cables (those going to UK are a net loss). And like USPS have to put money aside to take care of their employees retirement, EDF had to put money aside to take care of retired nuclear plant and waste storage over 300 years.
Also EDF had to absorb the cost of linking solar "farms" to their network themselves (in their report in 2012 they estimated this cost to be close to 1 billion by 2017). On this point this is likely that it did not cost that much, because they managed to be able to refuse individuals trying to connect to their grid.
The EPR was a mistake and made EDF valuation down by 80%, but despite that and no financial help from the French government (except with sales contracts), the dividends are still comming through.
EDF is run by engineers (or at least was) and it shows.
But oops nope we don’t pay that now, we’ll just let future generations handle it.
And if solar isn’t cheaper is you are getting ripped off or you live in a cloudy place or far north.
Did your country factor that into the costs of fossil fuels?