Most active commenters
  • kragen(5)
  • epistasis(3)

←back to thread

Understanding Solar Energy

(www.construction-physics.com)
261 points chmaynard | 16 comments | | HN request time: 1.292s | source | bottom
Show context
bryanlarsen ◴[] No.43423941[source]
Great article. Unfortunately his California duck curve graph only shows 2023. A graph including 2024 shows how batteries are dramatically flattening the duck curve:

https://cdn-ilcjnih.nitrocdn.com/BVTDJPZTUnfCKRkDQJDEvQcUwtA...

https://reneweconomy.com.au/battery-storage-is-dramatically-...

replies(3): >>43424435 #>>43425755 #>>43426846 #
1. epistasis ◴[] No.43424435[source]
And similarly the battery prices are very outdated. I don't blame the author for using those estimates, I frequently do too just because getting access to current data usually requires paying money.

But making decisions on that data without understanding that current prices and near-term prices will be about half of that price will lead to bad decisions. And when thinking 5-10 years out, not taking the full exponential drop in battery and solar prices is beyond foolish.

replies(2): >>43425520 #>>43433066 #
2. r00fus ◴[] No.43425520[source]
Actual battery prices may be dropping but cost to install batteries to your solar installation in CA have not dropped - in fact they've gone up.

Not sure why this is the case.

replies(1): >>43425678 #
3. epistasis ◴[] No.43425678[source]
This is by design in the regulatory infrastructure, from local permitting offices all the way up to CPUC and rate structures.

We pay about $3/W for solar installation in the US, but Australia pays about $1/W.

For batteries, there's still a supply crunch and the only people getting really good prices are those people who buy in huge bulk or are willing to take a risk on a lesser known manufacturer. If you want well-proven brands the prices can still be very high for small purchases, and a solar installer is not going to want to take a risk with a new supplier.

These systems are not super complex, most technical people could figure them out fairly easily, and in fact off-grid disconnected systems are really easy to do. It's the grid tie that will kill you or first responders to your house, we have made the process of setting the whole thing up very expensive because nobody on the regulatory side has an incentive to make it straightforward and cheap. And since NEM3 killed solar in California, all the installers are barely scraping by and need to rely on very high margins on few projects.

replies(2): >>43429499 #>>43430746 #
4. PaulDavisThe1st ◴[] No.43429499{3}[source]
> This is by design in the regulatory infrastructure

I don't see how this can be true. I installed my own ground mount array, and the costs directly attributable to regulatory infrastructure were about US$35 (for the permit). It would have been no higher if I had added batteries. The material costs were completely comparable with AU, CAN and UK pricing.

Perhaps you're arguing that the certification and licensing regulations for paid installers drives the installation cost up (i.e. that labor costs for US solar installs are too expensive) ?

replies(1): >>43429708 #
5. epistasis ◴[] No.43429708{4}[source]
> and the costs directly attributable to regulatory infrastructure were about US$35 (for the permit)

That may be true if your time is free, but for a company, they must deal with a permitting scheme for every county and city that they do business in. Additionally, unpredictable changes to rate structures will drastically change the demand for solar in areas year to year, and so the solar installers that survive are the ones who are well attuned to that change, and pounce on new markets that are suddenly opened up by new rate structures that make solar easy to finance or pay off quickly. That means that about $1/W of the $3/W that installers charge actually goes to customer acquisition costs.

Most areas do not have super onerous labor requirements for solar installers, and generally the contractor licensing part is quite reasonable. But perhaps insurance like workers comp and disability is a lot higher in the US than in Australia.

I'm surprised that US tariffs have not resulted in higher materials costs than in the other anglophone countries!

replies(1): >>43429835 #
6. PaulDavisThe1st ◴[] No.43429835{5}[source]
I installed my system 5 years ago, when no particularly unusual tariff structure was in place.

Your reply seems to indicate that "regulatory infrastructure" is not responsible for the bulk of the cost, but rather traditional concerns of for-profit business, in this case, the business of solar PV installation.

7. stephen_g ◴[] No.43430746{3}[source]
The same price difference exists with things like heat pumps - I've seen people in the US talking about installing single-head Japanese brand mini-split systems for US$5000 to even over US$10,000, when I can get the same sized units fully installed for AU$1600 to AU$2500 (about US$1000 to US$1600).

Just makes no sense why it should be that different. The units seem to cost similar prices in Europe to what we pay here in Australia so why is it so much more in North America? I assume part of it is that they are not quite as common but it still boggles the mind.

replies(2): >>43434166 #>>43436530 #
8. pyrale ◴[] No.43433066[source]
> And when thinking 5-10 years out, not taking the full exponential drop in battery and solar prices is beyond foolish.

The curve on solar is gradually getting flatter, though. Lazard's last LCOE report even saw it increase, partly because of inflation.

replies(1): >>43434136 #
9. kragen ◴[] No.43434136[source]
PV panels have dropped in cost in nominal euros by 21% over the last year, which is roughly the long-term trend since solar became profitable without subsidies around 02014: https://www.solarserver.de/photovoltaik-preis-pv-modul-preis...

Possibly you are only looking at prices inside the US, where anti-renewable-energy regulations drive the cost of solar energy through the roof.

replies(2): >>43434917 #>>43436151 #
10. kragen ◴[] No.43434166{4}[source]
Fossil fuel companies choose the president and legislature in the US—not exclusively, but they do evidently at least have veto power, and the current president campaigned on an anti-renewables platform which he has delivered on. Bipartisan tariff policy in the US is basically completely locking Chinese solar panels and EVs out of the market, doubling the costs of those products to US consumers and sabotaging any chance of future competitiveness in heavy industry. I don't know if the same thing is happening with Japanese air conditioning systems.
11. hnaccount_rng ◴[] No.43434917{3}[source]
Which does not really matter anymore though. In almost all installations the panels are already negligible for the total cost. This is especially true for rooftops and small installations
replies(1): >>43435036 #
12. kragen ◴[] No.43435036{4}[source]
People often point to how electric motors revolutionized industrial productivity, but not until about 30 years after their introduction. Because that's when factories were redesigned around the flow of products instead of the flow of line shaft power, using small electric motors at every workstation instead of just powering the line shaft with a big one. You might have pointed out at the time that electric motors had a negligible cost compared to the factories they were installed in, but from that you should have concluded that huge changes were in the works, not that further reductions in motor costs were unimportant.

Today the module cost is far from negligible (the article shows SEIA data showing that, even in the US, modules are a third of the cost of recent utility-scale solar) and it's only small because the other parts of the installation are badly lagging behind. If you need to heat or cool your house or train your neural networks, you really just need the energy those panels can provide, and somewhere to store it. Other balance-of-system costs like microinverters, racking, most wiring, transmission, design, civil engineering, land, installation labor, and regulatory approval are only useful as means to that end; they are not strictly necessary to receive the benefit.

If avoiding those forms of waste means you can get energy for a negligible cost, more and more people will find ways to do it.

How can you avoid them?

Well, you can avoid the cost of inverters by using low-voltage dc power, as off-grid enthusiasts, RV retirees, and Google data centers have been doing for decades. You can avoid racking by laying the panels on the ground, as the article mentions, or hanging them on an exterior wall of a house or an existing fence. These also avoid civil engineering and land and labor costs, and also falling off your roof. You can't avoid wiring but you can reduce its cost by using higher voltages (even low-voltage dc can use 48 volts instead of 12) and mounting the panels close to the point of use. You avoid transmission (and distribution) costs by siting the panels onsite instead of in a faraway solar farm. You avoid design costs by buying an off-the-shelf modular power system instead of paying someone to design a custom one. You avoid regulatory approval most obviously by breaking the law, probably more feasible in a slum apartment or an RV than in a utility-scale power plant, or by avoiding doing regulated things like connecting to the electrical grid or running 120VAC or 240VAC wiring.

This clearly points to a near future of ridiculously abundant energy, at what we would have previously considered a negligible cost.

replies(1): >>43436623 #
13. kragen ◴[] No.43436151{3}[source]
To expand on this 21% yearly rate, my notes say that in July 02014 the price on the Solarserver page was €0.55/Wp, presumably for the "low cost" category, which is now (February) at €0.070/Wp. That drop by a factor of 7.86 over 10 years and 8 months (128 months) works out to 1.62% per month (√√√√√√√7.86 ≈ 1.0162) which is coincidentally 21.3% yearly growth in watts per euro (a 17.6% cost reduction per year).

This is staggering, even at its current level. €0.070/Wp at a nominal 15% capacity factor is €0.46/W; at a 5% interest rate, assuming no aging, that's €0.74 per gigajoule, or, in the quaint non-SI units more commonly used for trading energy, €0.0027/kWh†, €0.029 per liter of diesel, 10¢ per gallon of gasoline, or US$4.60 per barrel of oil. And it's pure, undiluted exergy; you incur no Carnot losses to use it to drive motors or train neural networks.

The current WTI oil price is US$68.20 per barrel of oil: https://markets.businessinsider.com/commodities/oil-price?ty.... That makes solar energy fourteen times cheaper than oil, or more than thirty times cheaper if you're using it for transport or electricity.

The US's current policy of imposing prohibitive import tariffs on solar panels is similar to the Arab oil embargo of 01973, but self-imposed, attempting to prolong the energy crisis that began at that time.

______

† Not €0.27/kWh or even €0.027/kWh. €0.0027/kWh. 0.28¢/kWh.

14. dalyons ◴[] No.43436530{4}[source]
Drives me bananas. Aussie in the US - I’m very reluctantly going to spend 4-5x what I would pay in au for a heat pump. I have no explanation for why it can possibly cost so much more, it’s infuriating. You can DIY for a lot less but it’s a big complicated project
15. gpm ◴[] No.43436623{5}[source]
> hanging them on an exterior wall of a house or an existing fence

You can avoid racking by installing them as the fence when you install a new fence.

I mean you don't literally, but the installation cost is a cost you were going to pay anyways.

replies(1): >>43436822 #
16. kragen ◴[] No.43436822{6}[source]
Maybe so. It might depend on the winds in your area.