The progress we're seeing in solar has made me the most optimistic about a de-carbonized future in years.
I have no doubt that the decision the Chinese government made to heavily subsidize solar manufacturing will make this world a better place.
Already we're seeing that solar energy is more cost effective than all other forms of energy production [1], that the growth of solar has been consistently underestimated by very large players [2], and that solar democratizes energy production more than any other form of energy.
Distibution needs to be improved, but this issue also holds for other non-fossil energy sources. I predict a lot of these problems will be solved through hydrogen generation and storage [3].
> project life for solar and wind both seem to be 20 years to calculate LCOE
Not disputing, but where do you see this?
> Solar will simply churn along for far longer, while wind will have to becreplaced
We don’t have good numbers for the longevity of wind turbines either, though it’s probably under 25 years [1].
> doubts that presented numbers are accurate
Unlikely. The terminal value of that residual at any reasonable cost of capital is, while non-negligible, not going to be significant. (See Slide 6 for how each source’s LCOE reacts to rates.)
> the decision the Chinese government made to heavily subsidize solar manufacturing
This is often claimed, but does anyone have budget numbers or even a decent order of magnitude estimate for how much subsidy was applied here? Or was it actually the free market supplying compounding cost reductions through technological improvements?
> does anyone have budget numbers or even a decent order of magnitude estimate for how much subsidy was applied
Cheap producer credit, covering up to 50% of new-facility costs and feed-in tariffs [1]. At least the latter began getting phased out after costing Beijing over $15bn in 2017 [2].
Haven’t run the precise numbers, but that one-year figure seems to line up with the IRA’s total solar package [3].
Thanks. [1] in particular is a very interesting read. If I understand it correctly, it's all through subsidy of generation rather than directly of manufacture? That is, while there might be subsidy ensuring demand for the product, if you purchase a solar panel from a Chinese producer at any time during this process there wasn't a direct subsidy included in that cost? Just a guarantee over that time of enough demand to keep the factory chugging along?
That is, it's identical to Western-style renewable subsidies?
> it's all through subsidy of generation rather than directly of manufacture
At least at the federal level, this seems to be the case.
> it's identical to Western-style renewable subsidies?
At least early on, most Western subsidies didn’t discriminate based on where the panel was produced. We also don’t have visibility into provincial books, where if how they treat coal is any indication, where the plants are there to buy coal from their coal mines, there is probably cross subsidy.
But given what we know, one could argue they’re structurally similar, at least in respect to what we’re doing now.
There was a report from MIT a few years back that looked into the price reductions of solar between 1978 and 2012, during which time module costs fell by 97%.
They suggest that early on R&D support from government was key, then later market support to help grow the scale of deployment. Since 2001 its been manufacturing scale that has dominated price reductions.
It's claimed in the source economist article and reuters also claims it [1].
It's unlikely that the Chinese government wants to say how much it subsidizes, as this will prompt tariff increases from importing countries.
I wish some of the HN crowd can take a crack at building an affordable hybrid inverter with good software. Fronius/Ecoflow/Anker solutions are too damn expensive.
Currently the cheap hybrid inverter on the market usually comes with crappy monitoring software.
Entertaining idea, the panel with a 50Hz fan blade in front of it, but in practice it's much better to not compromise the beautiful simplicity of the photodiode and just stick an inverter on the back.
Happy to pick up the gauntlet. EE here with strong skills in ucontrollers, comms, and software stack, seeking to partner with other EEs to take on the power electronics part. Anyone interested, please LMK.
ugh tell me about it. I have a Delta inverter.. the interface is rou out of the late 90's... its so bad it passes the password through the url in plain text.
Meanwhile in Germany, the price for Solar power has literally plummeted below zero. There is vast oversupply during peak times and no storage capacity.
On scale, free energy supply is pointless when it occurs during times when nobody can use it. It's all about storage capacities.
The electricity doesn't need to be free. We've had range of interesting business models and regulations taking advantage of off-peak energy for decades.
Anyone looking to take business advantage of solar should be looking at predictably below average prices for most of the day, rather than only when it hits zero.
edit to add:
An example, yesterday on CAISO in California they broke records for solar generated, and battery discharge, but the price never hit 0 and batteries were charging from 6am to 6pm and making good money meeting evening demand:
Negative energy prices are a relatively new phenomenon, and it attracts attention because of its counterintuitive nature (I can get paid to use electricity?!?). Yet, it doesn’t happen that often. If I look at the Dutch market, this mostly happens on sunny+windy Sundays.
I think 2 things are happening that will make arbitrage feasible:
1. As solar production capacity increases, so will opportunities for arbitrage
2. As battery costs keep going down (capacity learning curve), so will the investment to do arbitrage.
At some point the curves for (1) and (2) will cross convincingly enough that investing in storage is just good business, and people will invest. If it’s anything like solar (home batteries), adoption might actually go very quickly because decision making would be decentralized, small investments, and without permits.
Your number 1 relies on producers not changing their behaviour, which is what negative pricingis designed to accomplish. It also relies on people so poorly taking advatange that the arb doesny dissapear.
your 2 will probably happen at some point but its been forecast within the minimum likely range for arbitrage opportunities to cause widespread capital allo.ation to take advantage of the arbitrage that it has stunted investment in your 1.
in short its questionable that arbitrage will sort this any time soon (although it will likely be accounted for in new projects that were happening anyway. That˙makes it something that will move the needle over decades, not years) sorting this soon needs another order of magnitude or two improvement in total cost of storage, ideally in a cheaper and less environmentally harmful chemistry that can also take advantage of the installed battery production capacity, or rethinking to make environmentally friendly demand generation (ie nuclear) cheaper and faster to deploy to provide the power when the solar can't.
Very timely article for Australia, which is just beginning another debate about whether they should go Nuclear (there is currently no Nuclear power in Australia).
Seems Australia has dodged a bullet by NOT having Nuclear, now some are trying to catch that bullet for political points.
It probably would have been nice to have one build now or even better 10 years ago. It's a good base of stable green energy to have during the transition. But planning to build a new one now seems purely political. It doesn't make sense when it takes 10 years to build at best. Every recent plant build in the west has had huge cost and time overruns. But then to make those costs back the plant has to compete with solar/wind/batteries for a good 20 years, STARTING IN 2035-2040. That's never going to work out financially. Solar is already cheaper and you will have to make those investments in the grid either way before the nuclear plant comes online.
It's bad because of Australia's unique situation. They are uniquely well suited for solar while being uniquely unsuited for new nuclear.
Australia has 2x the sunshine of Europe, more spare land for panels, and less seasonal variability.
Australia has no existing nuclear plant experience. No experienced regulator or legal regime. High labor costs and little relevant local labor. And a track record of project cost blowouts and time overruns on large projects.
Australia's small energy needs are also an issue. The marginal cost of new nuclear drops after you build the first few plants but Australia has such small energy needs that it won't reap the fruits of scale benefits.
Australia has large community opposition to nuclear but not to solar. About 35% of the electorate approve of nuclear, with almost all state premiers publicly stating opposition, while 80% approve of solar. This will lead to social licensing risks like what Germany, Japan, Taiwan and California face with planning and legislative delays, and potential early plant closures leading to wasted capex and higher energy costs.
Australia's peak scientific body, the CSIRO, estimates that a mostly decarbonized grid will be 2x more expensive with nuclear than pure renewables with transmission and storage. The above local factors contribute to that conclusion.
In other words, Australia is not China. And even in China, solar is beating nuclear.
FTA: "The next ten-fold increase will be equivalent to multiplying the world’s entire fleet of nuclear reactors by eight in less than the time it typically takes to build just a single one of them."
are there limits to the growth (other than the obvious land limitations which only exists in certain parts and the dark winters for some places)? like certain expensive materials that will get harder to find, or where production is limited and can't easily scale up?
In some senses, everything is finite, but the reason crystalline Si cells have beaten all the other fancier technologies in the market is the raw simplicity. Pure boule silicon, phosphorous and boron dopants in very small quantities, then wiring (silver/copper/aluminium). Then a glass/upvc frame structure.
The low kerf diamond wire saw is also a critical technology for this, but the powdered diamond required can be synthetic.
That isn’t the relevant bit. If decarbonisation is your aim, you build both as quickly as you can.
The fact that “the all-in cost of the electricity they produce promises to be less than half as expensive as the cheapest available today” is what sinks nukes. Burn fossil fuels longer, be judicious in adding gas so batteries have a chance to take hold (we’ve already fucked this up in Europe and America), and accept that while the transition will be dirtier you’ll have a cheaper grid in the end.
That said, just as decarbonisation isn’t the only variable, LCOE isn’t either. Australia will have to maintain a nuclear fleet for military purposes. Taking into account that sunk cost, a civilian fleet’s math might change. (I’m doubtful, but maybe.)
Considering it's lifespan and low external costs (like grid extension and energy storage for renewables) - it is cheap. What makes it expensive is politics.
that's because the cost of nuclear includes every possible cost, while for everything else we only include the immediate costs. for example, oil and gas have only been so cheap for so long because we didn't include the cost of climate adaptation, mass migration, geoengineering, etc.
I don't know what you base this on. Nuclear is heavily subsidized and cannot survive on its own. The price of nuclear waste storage is usually not included and even when money is put aside it's not enough. Germany's plant operators paid 23 billion to wash their hands of the mess and the tax payers will cover the rest (estimates go over 100 billion). Sellafield in the UK will cost over 200 billion to clean up. The French government is perpetually bailing out EDF and doing so as we speak because the EDF can't fund maintenance and building of its own planta. Etc. Etc.
> price of nuclear waste storage is usually not included
We don’t include the cost of disposing of spent panels and turbines either. Nuclear waste’s risks are hyped beyond reason. What kills nuclear is the capital cost of building it.
In the end, it’s fine. We’ll do gas + wind + solar and that will take us through 2050.
i should have said, "the cost of nuclear in any conversation". because the moment you give a cost for nuclear, some very smart person will slam their hand on the table and start talking about the cost of waste disposal (which is pretty small), but they never seem to even try to quantify how much it'll cost to remove carbon from the atmosphere for the coal that's getting burnt instead (e.g. Germany).
Even this is wrong as that number you will give will end up being just a fraction of the actual cost to build it. The cost of waste disposal is not small either, it is easily in the trillions for the waste we have now. E.g. Sellafield is estimated to be in the 250B pound range, and that's just one country. It also doesn't include the reactor decommissioning costs which end up in the billions per reactor range also. And it doesn't include the cleanup of a meltdown. Fukushima cost 200 billion to clean up, a massive effort that in the end mostly failed. People are still measuring high levels of radiation.
It's interesting you mention carbon costs. It's a fair point but also does not end up in nuclear's favor. Germany has been deploying more than a 1GW of solar per month this and last year. This will produce as much as two reactors worth of electricity (actually more but ok). If they were trying to shutdown coal plants with nuclear it would take around 15 years for those two reactors, likely more. So instead of their coal plants running for decades as they build out nuclear, they are sitting idly because solar deployment is quick. This matters too.
If you are talking about eliminating carbon you are currently producing this doesn't matter at all. If you deploy 6GW of solar in a year, you eliminate as much carbon as a 1GW reactor would, only it would take you 15 years of further carbon emissions to get there. You talked about hidden costs of removing carbon, and this is a good example of something that you will need to remove.
And that's the situation we are in. The discussion changes once you saturate the market with solar but even advanced economies are far away from that and the world as whole especially.
Will it though? It seems like the "free market" and political maneuvers might be able to keep the energy costs elevated, similar to what OPEC is currently doing for oil.
It seems to me that monopolising solar will be a lot harder than monopolising oil, because of the substantially lower overhead and resource competition.
Not really. The most important part is already monopolized: transport. Then we have storage, which is not monopolized yet because there's very little of it. My bet is that the transport monopoly will take over storage as well.
> The resources needed to produce solar cells and plant them on solar farms are silicon-rich sand, sunny places and human ingenuity, all three of which are abundant.
The last one reminds me of a quote attributed to Einstein [1]:
> Two things are infinite, the universe and human stupidity, and I am not yet completely sure about the universe.
Already we're seeing that solar energy is more cost effective than all other forms of energy production [1], that the growth of solar has been consistently underestimated by very large players [2], and that solar democratizes energy production more than any other form of energy.
Distibution needs to be improved, but this issue also holds for other non-fossil energy sources. I predict a lot of these problems will be solved through hydrogen generation and storage [3].
[1] https://www.carbonbrief.org/solar-is-now-cheapest-electricit... [2] https://www.linkedin.com/pulse/why-does-almost-everyone-unde... [3] https://www.iea.org/energy-system/low-emission-fuels/hydroge...