Solar will supply almost all growth in U.S. electricity generation through 2025 | Hacker News Reader

Solar will supply almost all growth in U.S. electricity generation through 2025

cleantechnica.com

186 pointskieranmaine222 comments

I found this podcast with Jesse Jenkins to be a really helpful start to understand non fixed energy resources.

https://podcasts.apple.com/us/podcast/what-the-sun-isnt-alwa...

I thought that the most cost effective way to supply energy would require a certain amount of base load. It seems like this is a misunderstanding because of how much cheaper and more available these non fixed resources are than fixed resources. Energy this cheap gives us a huge incentive to change our energy use patterns to match power abundance rather than to take energy demand as a constant.

Yes and:

I appreciated Jenkins' explanation of the different variability time scales. "Base load" is folk understanding and not especially helpful for planning our future green energy grid.

IIRC, seconds, minutes, hours-days (diurnal), and months-years (seasonal). Meaning there's huge opportunity for all sorts of novel energy storage (and demand shifting) systems. Dynamic load balancing might use modestly sized of Li-ion batteries scattered around the grid. And thermal batteries could be used for long-term storage.

The opportunities are insurmountable. -- Yogi Bera

Yes, but please keep things in context. We're transitioning away from baseload, but it's a major endeavor and not going to happen over night. I think the HN crowd is a little overzealous in this area. NERC is reporting the potential for energy shortages in many regions in the United States. This was once virtually unheard of. Solar alone isn't going to fix this in the next few years and many regions still have only tiny amounts of demand that is responsive to price.

The ramp up is inherently slow, we’re still generally seeing cheaper nighttime rates than daytime rates even after years of ‘heavy’ investment in solar.

Solar’s impact on prices is going to remain fairly trivial for the next few years. But start looking 20 years out and a great deal of nighttime load will shift, which dramatically reduces the need for storage even as decades of X0 GW per year adds up.

Aka it’s not just one trend you need to follow but several.

The merit order effect of renewables like solar and wind has already been substantial on wholesale prices with the spot price being $0 or below $0 in significant percentages of the year. This probably hasn't trickled down to retail rates as we're stuck paying for the fixed costs of many of those assets via the PPAs that the utilities sign.

Thus the rapidly expanding introduction of storage, which can take advantage of those periods.

3400GW of solar nameplate is expected to be built from 2023 to 2028 globally, which is maybe 40% of all existing global electrical generation. So I don’t think “20 years” is the right time period to be thinking in.

That's shifting generation, not demand. Most people won't respond immediately, so 20 years isn't a crazy guess for a significant amount of the equipment in peoples' houses to cycle out.

US produced ~4% of its electricity from solar in 2023 and it’s predicted to hit 5.6% in 2024 and 7.0% in 2025. So ~4 years to break 10% of total generation which still doesn’t offset higher daytime demand.

Thus for the next few years we’ll see higher daytime rates than nighttime outside of specific regions and specific times of the year.

Beyond that you need to consider increasing EV adoption, energy storage etc. Cheaper daytime rates is going to happen faster than people realize, but not fast.

Looking to the US to be on the cutting edge of the energy transition is like looking to the US to be on the cutting edge of steam engine development during the early Industrial Revolution. In a global sense, prices are going to change rapidly and huge amounts of industry will adapt to intermittent sources. The US will catch up a decade later, adopting whatever ideas worked overseas.

The US close to average in terms of solar with most countries are on a similar curve +/- a few years. However, I used the US because our electricity demand is fairly static. The US is only up 13% since 2000 unlike Africa’s electricity demand which doubled, or China’s 600% increase over that period.

China is where energy things are happening, both on the generation side and on the industrial consumption side. The US feels like a quiet backwater in comparison.

Only in terms of scale not innovation.

China still gets most of its elected from coal and hydro. They are still building a little nuclear and lots of coal alongside all that wind and solar.

In this case scale is the better part of innovation. The people dominating future PV, wind, and battery industries are going to be the ones implementing the “learning curve” that brings prices down through continuous manufacturing improvement. The folks who innovate the most will be the ones deploying the new terawatts of generation, building factories that operate on intermittent and geographically-dependent energy sources, and learning to operate a massive continent-wide renewable grid. There’s this funny idea that you can keep your technical advantage doing research while another country is actually doing the things, but I don’t think that’s ever been the way the world works.

Cost competitive panels are manufactured inside and outside of China. PV already reached a point where panel costs aren’t the majority of grid scale instillation costs cutting their costs in half just doesn’t change much.

Managing the grid when your dependent on PV not just feeding in a trivial amount you can make up from other sources is the uncharted territory.

> Yes, but please keep things in context. We're transitioning away from baseload, but it's a major endeavor and not going to happen over night. I think the HN crowd is a little overzealous in this area. NERC is reporting the potential for energy shortages in many regions in the United States. This was once virtually unheard of.

What do you mean? Do you recall Enron? Also when I was living in Tennessee for a year in the 90s brown outs occurred at least a couple of times and some of my US friends tell me about interrupted power happening quite regularly still. I'd call that energy shortage.

Enron was a special case when California first started their electricity market and made many mistakes in not having processes to take care of market manipulation. An entity deliberately and unethically putting a generator on outage to drive prices up is not what I'm referring to. I'm saying the reserve margins of baseload generation have shrunk.

Others and myself have been arguing against the baseload myth on HN and other places for ages, but it just persists.

The thing is, while we are still ramping up renewables wind and solar compete with the traditional "base load" large nuclear and coal plants, they are not replacing load following gas turbines until much later. Even Wikipedia says that variable supply like wind and solar are one of the sources to supply base load.

We're already incentivized to do things like running the dishwasher in the evening do to costs, no? Or is the implication that it would be considerably even more cheap?

Depends on where you live -- where I'm at, there's no time of day based savings (unless you're a big industrial user, IIUC).

In my area, summer "peak hours" often make up half my electricity usage. The off-peak hours are the other half. The price at peak times is more than double the flat rate, but only 30% cheaper for off-peak. I'm already reducing my energy use as much as is comfortable, if I want to maintain that I'd be paying more money than the flat rate.

The scheme only saves energy if its not optional.

This is rapidly changing worldwide though: as more variable production comes online, the value of shaping demand grows.

Missouri is moving to hourly rates this year IIRC. EU already is on 60-min day-ahead prices and pushing hard to get to 15minutely prices.

Depends on whether your jurisdiction has time of use pricing.

This is where I'm really intrigued of all-house battery packs like the Tesla Powerwall (it's the only name of one I'm familiar to call out). I like the idea that it charges at night during the lower rates, and during the day the house is powered from the battery. Just haven't seen any longevity reports on them in general vs brochure data.

Tesla warranties the Powerwall for 10 years, 15 years if part of one of their virtual power plants (orchestrated by Autobidder [1]). Look into the 30% federal tax credit for it, as well as state, local, and utility incentives.

If you don't yet have solar and your site is favorable, I strongly recommend considering it if you can have it installed at a reasonable cost.

[1] https://www.tesla.com/support/energy/tesla-software/autobidd...

Solar doesn't work for my property. There are too many trees doing exactly what they were intended to do by shading the roof top. Otherwise, it would be great as the roof alignment is pretty ideal.

Can you submit a residential battery to Autobidder?

If Tesla has a virtual power plant program in your geography.

https://www.tesla.com/support/energy/tesla-virtual-power-pla...

https://news.ycombinator.com/item?id=32500926

One thing to consider is that if that's profitable to do at house scale, then it's probably even more profitable to do at grid scale and you'll likely be out-competed by utility companies.

Edit: Unless you're taking advantage of tax incentives to do it at house scale. Its hard to beat the economics of just being handed free money.

An individual can take action now whereas the grid scale action is molasses like. It's even worse when your state is not in the pocket of but the actual pocket of Big Oil

But an individual can't take action without the grid taking action to provide customers with the different pricing rates.

What? The entire premise was that it is cheaper at night. Thanks for jumping into the middle of the convo though and trying to jerk the wheel.

Settle down. A lot of the U.S. doesn't have time-differentiated rates.

And that has precisely squat to do with my intrigue into looking at a system like this. If it didn't make sense, I wouldn't be interested. So what's your point exactly?

> If it didn't make sense, I wouldn't be interested.

It may or may not make sense depending on your location. Many utilities would prefer to make customer-owned solar a very poor financial choice thru rate design (e.g. very high minimum charges, per-kw-installed solar charges, low daytime rates relative to evening rates, wholesale-rate compensation for exported production, etc). Some succeed.

And much of that does is nowhere near the real cost differences. I might see a difference on/off peak price difference of 0.05 on a $0.40/kwh, but wholesale prices fluctuate from negative prices at peak to much higher rates.

If you claim is that "An individual can take action now" by installing a battery to charge at lower rate, then the question of if they can actually purchase power at that lower rate is fundamental.

If household power purchase price doesnt change with time/generation cost, then there is not ability to act faster than the power company.

> One thing to consider is that if that's profitable to do at house scale, then it's probably even more profitable to do at grid scale and you'll likely be out-competed by utility companies.

Depends. It may be more profitable for utility companies, but they keep the profit of the economies of scale. It is almost never passed on to the customer.

If you DIY your own energy system, you're free to keep all the surplus generated by it through the life of the system.

Only if they let you by giving you a big price discount to charge

Their alternative is to find some way to generate a lot more cheap electricity after sunset. If whatever they're doing can't beat the cost of solar panels on your roof + batteries and a small generator to top them up in exceptional circumstances, people will just stop paying them altogether.

> One thing to consider is that if that's profitable to do at house scale, then it's probably even more profitable to do at grid scale and you'll likely be out-competed by utility companies.

Utility companies have to pay for distribution networks, and generally pass that cost on in the price per kWh. Which you avoid the less you use the grid -- and which then become a higher proportion of the prices the utility has to charge the more people do that.

In a lot of urban locations, it is not legal to have a residence that is not connected to the grid. If you have home generation capabilities and storage that means you take the least amount from the grid provider, you can be technically still connected to the grid without using any power. If this continues, I would not put it past the grid provider to come up with a minimum monthly bill so it is impossible to have a $0 payment. Otherwise, who's going to pay for the administrative costs of having you as a customer?

Large minimum monthly fee + required to buy service seems like the kind of thing that would piss off voters.

Ah, I see where you strayed from reality. Your comment seems to be based on a premise that voters are rational.

The rational thing for voters to do is to mostly ignore what the government does unless it directly impacts them in a way they can understand well enough to rally around and get it changed. Which, it turns out, is mostly what they do.

That just has terrible results most of the time, because it means that people generally ignore or otherwise don't understand what the government is doing, which leads to all kinds of inefficiency and corruption.

But "law requires you to buy this expensive thing you don't want" is simple enough for people to understand and get mad about, and affects enough people to make politicians care.

> In a lot of urban locations, it is not legal to have a residence that is not connected to the grid.

Woah, I didn't know this. In any case, surely it's legal to have grid service to the house and simply not use it.

> I would not put it past the grid provider to come up with a minimum monthly bill

My electric company (a community owned co-op) has done this for many years. You pay $15/mo no matter what. Your power usage is billed on top of that.

Good point. Seems like a problem with the utility company's pricing structure if consumers can avoid like 95% of the costs of grid maintenance without having to become fully independent of it. I wonder if eventually they'll have to split electric bills into a fixed "connection fee" and a variable "usage fee".

> if that's profitable to do at house scale, then it's probably even more profitable to do at grid scale

There is a reason home generation wins. It's because retail prices are 3 times what the generator is paid. (That's true where I live).

So yes, in absolute terms power generated by home solar costs more than a utility generating it. But it's not 3 times more, so if you install home solar and use the power, it's always a big win. It will be the same with batteries.

The traditional forms of generating power, coal, gas, hydro and nuclear only win because home generation costs using those technologies (eg, putting a coal mine and generator in your back year) costs far more than 3 times. You can't put wind, or pumped storage in your back yard either.

But solar and batteries are different animals. They are different in other ways too. It always makes sense to over provision solar. The question is not whether to do it, but by how much. So home generators will always be looking to sell power, and since it's free selling it at 1/3 of what they would pay for the same power looks like money for jam to them. It makes you wonder how many utility generators will be left standing if solar takes over the world.

> it's always a big win

AFTER you finish paying off the financing of the source of home generation. so maybe it'll be a big win for the people in your location after you're gone???

This brings up an interesting consideration. Is it possible that in the age of solar, power grids are simply uneconomical in areas of insufficient density? That people in more rural areas would be better off not having a grid and just installing rooftop solar and batteries instead? My gut tells me that the added reliably of a grid is probably worth paying 3x the price, but its hard to be sure.

Does this reduce your energy usage though? You're still using the same amount of power during the day. So it saves you money, and takes some load off the grid during peak usage. What does that do to reduce emissions from power generation? Without solar, you're still charging off the same infrastructure.

As if someone trying to save money is such a ridiculous thing to do? WTF?

I subscribe to a power company that produces power via wind. So my money is supporting green energy generation. So yeah, it would still be doing plenty. Not really sure why you gotta be so negative about something

Do you even actually save money, accounting for the financing and installation of the solar panels?

A while back there was a thread on the SolarCity subreddit linked from HN where the OP discussed his savings thanks to nice weather, but still figured he had a 20 year payback period on his setup.

That is crazy.

I never even suggested solar. I just want a battery pack that charges when grid prices are cheap, and powers off battery when grid prices are high. Complicating it with solar is beyond the scope of the original intent.

That seems like it would be hard to ever make a financial return.

Say a battery costs $10k and a home uses $100/ month electricity. If you assume electricity is free at night to fully charge the battery and meet energy demand, that’s 100 months (8 years) to break even. Real world costs are going to make it quite a bit worse than that.

As they say, "location location location". In a Texas summer, it is easily 3x that per month for 4 months out of the year for a house. So, that's 12 months in one quarter based on your math. The vast majority of that is just running the A/C. So again, this makes much more sense when you look at the nuances vs just some lame internet posting of using averages. If it makes no sense for you, that's too bad, but to come out with broad brush comments alluding to it not making sense ever is just naive.

I guess you could further insult me or provide your own napkin math.

Sure, let's say the summer months bump the Texas electric bill to $167/month (8$100 + 4$300). The previous model was assuming free electricity. My quick search is revealing that using differential pricing in Texas, the difference from the normal rate (~$.16) and the low rate (~$.08) saves ~$.08/kwh. Assuming you could perfectly offset all usage onto the cheap rate, that would yield a monthly savings of $83.5. Which yields a $10000/$83.5 = 120 month or 10 year payback period.

Then again, we forgot to right size the battery system. This site[0] claims an AC is going to use 16kwh/8 hours. A Tesla Powerwall is 13.5kwh. If you want the battery to fully cover your AC needs alone, you will need at least one additional battery system.

Naturally, we are missing out on maintenance, opportunity costs, grid connection fees, potential rebates, etc, but I think we have indeed landed on a more nuanced model.

[0] https://learnmetrics.com/how-much-electricity-kwh-do-air-con...

On a modern grid, off-peak, cheaper energy coincides with low carbon.

I think the implication is that we'll see a further split around pricing.

One thing that stood out to me is that in the case of thermal loads (perhaps your dishwasher but more realistically a hotwater heater). It will become more worthwhile for that system to know when high and low electricity prices are and store more heat (i.e. hotwater) at low electricity prices and wait until prices fall again to refill it. In this sense, a heater would be operating like a battery.

In a world where energy prices are stable throughout the day, it doesn't make sense to manufacture, for example, a price aware water heater with extra storage capacity. It seems like that calculus is changing with solar and other renewables creating an incentive for variable rates compared to traditional fixed rate alternatives.

Recall your entire house is a thermal battery too!

In fact, all the four major energy loads in a house - Heating/Cooling, Water heating, EV Charging, House Battery - are extremely flexible.

But we're still lacking good standards for reliable control. Today we do stuff like tell your HVAC to stop via vendor X's cloud API. What happens if your router dies after we send a "stop" command in the middle of the winter?

For many of the minisplits I see control code for, they'd just let the house freeze, waiting forever for another command that'll never arrive.

For critical functions like "keep my water pipes from freezing" / "be able to drive to work in the morning", we need much better standards for reliable control and safe failure modes.

French's "fil pilote" to control electrical radiators is quite bullet proof in this regard: No power on command wire = comfort thermostat, full power = "eco" thermostat, and there are other more convoluted signaling for various modes (send only positive voltage, only negative, only a third of the phase) , including simple anti freeze mode. There is an off command as well, but you really can't send it accidently.

If your IoT dies, you'll most likely get to the comfort mode.

Only nit is that in 2023 the default mode would probably be "eco" rather than "comfort"

Tangentially related gripe:

My biggest issue in making this work reliably is that my "smart" thermostat is trying to be too smart.

Our power gets cheap at 10pm. I can't just set my HVAC to turn on at 10 and go full blast until power gets expensive again at 6am. I have to program a set temperature at a time. if I set that at 10pm, my thermostat tries to make sure that it _reaches_ that temp at exactly the time. So it will actually turn on an hour early. So I have to guess how long it will take my HVAC system to reach whatever temperature I have it set to, and then set it for that much time _after_ 10PM. It's inexact and annoying, and means that I am leaving performance on the table.

I have idly searched a few times for a replacement thermostat that would interact locally with home assistant and have had zero luck (partially due to the fact that my system is complicated and seems to have a non-standard wiring setup).

Everything else: hot water heater, EV, laundry, etc. are all trivially easy to make run during the desired periods. But the biggest use of energy I have (yes, larger than charging the car) is a PITA to make it play well with TOU billing.

What’s your “smart” thermostat?

Lennox 7500 series.

Check the manual for the thermostat. Mine has a setting to disable this feature.

Just buy a 240v DPST 24 hr mechanical time switch and wire it into the electric furnace circuit (or furnace fan circuit if you have a gas-fired furnace). Intermatic makes them.

The ‘smart’ thermostat can call for heat all it wants, but nothing will happen if the time switch relay contacts are open.

I’ve seen 50 year old 24 hour mechanical time switches operating just fine, and you won’t need to replace your thermostat.

This one is $122 and comes inside a NEMA 1 enclosure:

https://www.intermatic.com/Product/T104?setcontextlanguageco...

> In fact, all the four major energy loads in a house - Heating/Cooling, Water heating, EV Charging, House Battery - are extremely flexible.

The dominant energy load in existing houses is heating. EV charging is 100% new generation capacity and doesn't take any of the existing load off the power grid. House Battery isn't a load, it's the expensive thing you suffer when the actual loads are inflexible.

Heating is slightly flexible, you can lower your thermostat at night, but it's also colder at night. And most existing heating is fossil fuels. Converting that to electricity is going to require more electricity consumption at night than we have now.

> EV charging is 100% new generation capacity and doesn't take any of the existing load off the power grid.

All the more important to ensure this additional demand is flexible. Shaping it to match renewable production and low-usage hours in the distribution grid means higher capacity factors across the board and less new transmission+distribution need.

> Converting <home heating> to electricity is going to require more electricity

Yes.

> All the more important to ensure this additional demand is flexible.

It's inherently flexible. Shaping it to be that way isn't a particularly hard problem. The issue is that it's entirely new demand -- what you're really replacing is the petroleum burned in existing cars, not any part of the existing power grid. So how do you solve the demand currently served by coal and natural gas and home heating oil, rather than gasoline and diesel?

> Converting <home heating> to electricity is going to require more electricity

I feel like editing out the "at night" part hasn't actually provided a mechanism for supplying that electricity at night when it's cold and you need heat.

This is an interesting thought. There are already some energy meters you can add to monitor all of the circuits in your house. It's a short step from there to have smart breakers or switches to help retrofit old-fashioned heaters, radiant systems, etc as thermal storage. Electric buffer tanks may become a cost effective installation when paired with an on-demand water heater and time-of-use electric pricing.

This is how things are done in NZ already. Most electricity company have contracts with two tarrifs (sometimes even more), one normal one and a second significantly cheaper one where the circuit is controlled by the power company and they can turn off things when there is hight demand. Typically you'd put your water heater, electric car and other things on this circuit.

I the more long term, exchanging resistive water heaters with heat pumps is the way to go (and already done throughout Scandinavia)

My energy provider (BGE) used to offer a program where you could get a rebate in exchange for letting them control your hot water heater and/or thermostat. The HVAC portion is still running, but only for people who signed up prior to 2020. The water heater program has been completely stopped since 2021.

The good thing with both approaches is you don't need the appliance built to support it. For HVAC, the control system is in the thermostat, which is cheap and has a standard interface.

For waterheaters, you can just stick a relay in sequence with the heater.

> We're already incentivized to do things like running the dishwasher in the evening do to costs, no?

Not where I live. Electricity costs the same here no matter what day or time of day it is.

My (Swiss, small) provider just got rid of differentiated pricing "for simplicity sake". On the other hand they continuously increase the solar installations, so I really don't get their point. But I'm not an economist so...

My jurisdiction went all-in on residential smart meters for all to do time of use billing.

Even apartments where hvac, hot water and laundry are central, and the builder/owner chooses the big appliances.

But it turns out people don’t care and didn’t bother with enough load shifting to cover the cost.

“Smart Metering Costs to Date Exceed Projected Costs and Benefits” says it all.

They did get/build a $249m centre to monitor all the data though.

https://www.auditor.on.ca/en/content/annualreports/arreports...

Even after ~15 years, most equipment still doesn’t care about time of use.

Most savings have been due to better regulations/promotions around efficiency. Subsidized LED light bulbs accomplish more than telling people to turn off the incandescent.

But you know, giving useful stuff to people for free is WeLFarE but forcing untested policies on them and making them pay for it is okay.

What are the relative costs?

It depends very much on the weather, but in Denmark the cost can vary from around 1DKK/kWh (€0.14) to 5kr/kWh (€0.67), or a few hours a year even more.

Recent and future prices are shown here: https://andelenergi.dk/kundeservice/aftaler-og-priser/timepr...

I don't know of any studies done here, but anecdotally I do know people who have the app version of that webpage and will schedule their laundry/dishwasher accordingly. (And of course EV owners will configure their car to charge at the cheapest period.)

Personally, the saving isn't worth thinking about for me. I run the dishwasher overnight anyway and sometimes the washing machine, but that's to avoid the noise during the evening rather than any economic or environmental reason.

For the power price, in winter, it’s 8.7 cents/kwh at off peak, 12.2 mid peak and 18.2 peak. (Canadian prices, about CAD$1.35:US$1.00).

There are other charges that don’t vary by time of day or kWh, so those numbers are only good for absolute, not relative comparisons.

https://www.oeb.ca/consumer-information-and-protection/elect...

> But it turns out people don’t care and didn’t bother with enough load shifting to cover the cost.

This doesn't surprise me a great deal. Personally, the price difference would have to be very high in order for it to affect my usage patterns. It would have to exceed the cost of having yet another complication for me to keep track of.

Some of it is mispricing.

We’re getting the kids version of market pricing, but when you give people real market pricing without cross-subsidization, they tend to get confused/angry/bankrupt.

I still think they could do real-time pricing with some kind of cap that gets balanced out amongst all users, but the uptake still may not be there. It averages out for those that treat it as noise and still not a lot of equipment out there to take advantage of it at the personal residential level.

I checked my electric provider and although I said in an earlier comment that they don't do this sort of differential pricing, I was incorrect. They've been doing it for years now -- but it's optional and you have to opt in to it.

They say that the percentage of users who actually do this is somewhere around 10%.

Personally, I value having electric prices being stable very highly. I'd prefer pay higher rates (up to a point) to keep that stability.

if you have the choice, you'd have to run the numbers, but then you create adverse selection issues.

(Technically you can choose a tiered pay X for first kWh and then X+Y above that in my system, but if you're in a building, the building gets to decide which of the two every user is put on)

I had a solar system installed last year. During the day, from around 8am til around 5pm we are "off grid" in the sense that we can generate more than we use.

I've already switched hot water and pool pump to daytime rather than night. The dishwasher now runs in the morning, not after dinner. Washing and drying, Aircon/ heating mostly happened during the day anyway.

Overall our use of grid power has dropped to 20%, and I think we can get it down to 10% (at least I summer time.)

So yes, in the sense that the daytime cost is now a -lot- cheaper (think free) I'm seeing how we are heavily incentivised to change our usage patterns.

I moved to a house with a solar system in October. I now make a point to run the dishwasher and dryer during day time hours. And avoid using the dryer on cloudy days.

It doesn't really matter when I run these appliances. Between fixed rate electricity and net metering, the cost will be the same even if they run at night. But I get weird satisfaction knowing that the sun is providing the power.

And don't forget the ultimate flexible load: electric vehicles.

In 2023 near half of my (relatively well insulated all electric home) energy use was charging my EV at home.

That's in the game plan for 2024, no question. My new home came at the cost of doubling my commute distance. An EV charged by my own rooftop would be better for both the planet and my own bottom line.

Edit: Any chance you're in the US? If so, what EV are you driving?

I'm in the south of France, owning a Tesla Model 3 since 2019 now with 104k km on the odometer (152 Wh/km average). Before that it was a Renault Twingo v1 2006 model.

Jenkins is a worthwhile follow.

https://twitter.com/jessejenkins

https://bsky.app/profile/jessejenkins.bsky.social

> Energy this cheap gives us a huge incentive to change our energy use patterns to match power abundance rather than to take energy demand as a constant.

I have not listened to the podcast, but I know for a fact that’s much harder than it sounds. Parking spaces, highways, factories, offices, parks, beaches, hiking trails, skiing, museums, stadiums, even grocery stores have an extreme variability in demand. Almost all demand comes from human activity, and like all other animals that activity is highly variable and intermittent. Yes, there are a few exceptions: batteries, some household appliances like washers, and perhaps certain factories/refineries that could be run with minimal operations at night. Incentives don’t even work to suppress leisure: people go through massive traffic, no parking and crowds to get to the beach on sunny weekends. Imagine chores or even actual needs: like keeping the heat or AC on. Electricity is survival-critical in many parts of the world.

In short, I think the incentives needed to alter fundamental human behavior are less like cute coupons and more like a gun to your head. In most places, there won’t be enough political lunacy to push that through, and instead the glorious green energy programs will be stalled, delayed and eventually scrapped. I can’t think of a better outcome for fossil fuel companies.

You can replace a significant proportion of the grid with intermittent generation without any trouble. Using solar for the air conditioning load in the summer is perfect. Use whatever you want to charge electric car batteries because they're batteries and most cars are parked >90% of the time. You can can't shift 100% of the load based on time of day through pricing, but you can do a non-trivial amount.

Then you pick the low-hanging fruit and start trying to find a solution to power heat pumps in the winter at night and it's... not as easy. But a grid can be 0% fossil fuels without being 100% solar and wind.

Using solar for the air conditioning load in the summer is perfect.

Ish. They don't completely correspond.

It's close enough that you can compensate by setting a lower indoor temperature during daylight and letting it rise a few degrees around sunset before the outdoor temperature declines.

Whether it be the thermal battery of the house, or something else, some form of storage or buffering is going to be needed. Therefore, not perfect correspondence.

It's perfect correspondence in the amount of power required seasonally. You need more air conditioning in the summer because there is more sunlight, which also yields more solar generation. The times also align closely enough that the thermal buffer of the house is all you need, and since that is available, it's a solved problem. Whereas using solar power to provide heat in winter is exactly the opposite.

> Then you pick the low-hanging fruit and start trying to find a solution to power heat pumps in the winter at night and it's... not as easy.

But it's also not impossible, especially as heat pump efficiency keeps marching up in colder conditions and stationary battery prices keep going down with new chemistries that trade off energy density for cost.

Heck, if batteries are still too expensive, we could even use residential hydrogen fuel cells in that situation, which could easily power a heat pump.

Storage isn't just a bad fit for that because it's expensive, though. What do you do when it runs down, because it's unusually cold or you have an extended period of low generation, or both? Now it's 20 degrees below freezing and you have no heat.

> What do you do when it runs down, because it's unusually cold or you have an extended period of low generation, or both?

I'm not talking about going off grid, just energy price arbitrage. You still have the option of pulling power from the grid to heat your house, although potentially at a higher price.

With storage, you could also coordinate grid loads during the night so that at any given time the total load on the grid doesn't exceed the grid's generation capacity.

> I'm not talking about going off grid, just energy price arbitrage. You still have the option of pulling power from the grid to heat your house, although potentially at a higher price.

It's not a different problem for the grid. If the grid is generating nearly all of its power from solar and wind and it's calm and cold and night, where does the grid get the power to run your heat pumps? Storage is not only expensive, it has finite capacity. On the coldest week of the decade when everyone's heating system is using 250% more power than usual, and has been for days, the storage runs down and then what?

> On the coldest week of the decade when everyone's heating system is using 250% more power than usual, and has been for days, the storage runs down and then what?

Since we're talking about the grid of the future, there are several possible options:

- Hydrogen electrolyzed/stored (pressurized or as NH3) during the summer can be used to add supply to the grid.

- Pumped hydro increases its output during the cold snap.

- Power imported from regions that aren't experiencing the same cold snap with no renewable generation.

> If the grid is generating nearly all of its power from solar and wind

Since you said "nearly", there is some room for geothermal, synfuels, and maybe SMRs (if they can ever be built economically).

The point is that we won't have a lack of carbon-free generation sources or storage options in the future.

> Hydrogen electrolyzed/stored (pressurized or as NH3) during the summer can be used to add supply to the grid.

> Pumped hydro increases its output during the cold snap.

These are essentially other types of storage, and you're trying to go for something where you trade the cost of storage against the cost of the conversion equipment. (It can't be cheaper for both or that's what you'd use all the time.)

The nice thing about batteries is that you can draw power from them as fast as you'd usually ever want to, to the point of discharging the entire battery within 30-60 minutes. A 100kWh battery could give you 200kW for half an hour. And of course then two 100kWh batteries could give you 400kW for half an hour.

Whereas if you have a 100kW generator and 100kWh of NH3, you can generate 100kW. If you have a 100kW generator and 200kWh of NH3, you can generate at 100kW for two hours, but you can't generate 200kW, much less 400kW.

Which kind of leads to the problem here. If you're going to have a 50GW average shortfall for a week, you need 8.4TWh of storage. Ouch. Okay, so that's why we're using the ones with the cheapest storage, but that's still not cheap. And then you need 50,000,000 kW of generation capacity for that type of storage, on top of the storage.

But you're only going to use either of them for one week every two years.

That seems impractical.

> Power imported from regions that aren't experiencing the same cold snap with no renewable generation.

It's not obvious how this one is supposed to work either. Say it's -15 in the Northeast, that's trouble, we'll have to import power from Texas. Where it's also winter, and therefore 25 degrees with significant heating demand, and generation there isn't as low this week but it's below average. Texas can handle that, they charge a higher rate at night, people shift demand enough or lower their thermostats and they can break even. But they have none to spare.

Okay then, the Midwest? Nope, too close to the Northeast, it's -15 there too. The South then. Finally, a region with spare capacity... but it's only 10% of what you need and the Midwest wants it too.

And the general problem with this one is that people are going to optimize it out. If you have cross-continent transmission lines, they'll be used whenever they're useful, which will lower the price of electricity. And then people will build less generation capacity, because it's less needed and less profitable. And then the whole grid has less slack in it and you can't rely on the slack existing in situations like this. See also "just in time" supply chains in an emergency situation.

> Since you said "nearly", there is some room for geothermal, synfuels, and maybe SMRs (if they can ever be built economically).

But this is kind of the point. You don't build geothermal or nuclear plants for periodic use in an unusual supply crunch. If they're enough of the grid to address this, they're that much of the grid all the time.

> You can replace a significant proportion of the grid with intermittent generation without any trouble.

Yes, in some locations. That’s a no brainer and where it’s economical it’s already happening. But unpredictable supply particularly needs on-demand energy. Nuclear can’t respond quickly, so if you don’t have hydro you’re likely left with gas.

> Use whatever you want to charge electric car batteries because they're batteries and most cars are parked >90% of the time.

Agreed. We have some way to go with universal infrastructure support for that though.

> You can can't shift 100% of the load based on time of day through pricing, but you can do a non-trivial amount.

Possibly. The major sources today are heating, AC, fridges. Cooking will come up there if you remove gas. In either case, industrial and commercial are ~60% of use and how do we move that significantly out of peak hours?

For short term unpredictable demand it is likely that batteries will outcompete gas in the next couple of years.

> But unpredictable supply particularly needs on-demand energy. Nuclear can’t respond quickly, so if you don’t have hydro you’re likely left with gas.

Nuclear is baseload, but baseload offsets intermittency.

If you're using solar you've got no generation at night. Suppose you do all the load shifting to the day that you reasonably can and you still need 600 GWh of power for the 12 hours between sunset and sunrise. Well, add 40GW of nuclear and now you only need 120 GWh. You've cut back the amount of gas or batteries you need by quite a lot, but you still have some, which is enough to respond to short-term demand changes. And the nuclear plant generates during the day too, which isn't as valuable but is still not zero and contributes to paying its costs.

> We have some way to go with universal infrastructure support for that though.

At some point there was a fear that this was chicken and egg: No one buys electric cars because there are no chargers so no one builds chargers.

There are enough chargers now that it's viable for a significant proportion of people to buy an electric car. At that point people will just build more chargers in proportion to how many cars need them and there isn't really anything to worry about. And the power can be generated from renewables without having to worry about intermittency because you're charging a battery which is typically stationary for most of the day.

> The major sources today are heating, AC, fridges. Cooking will come up there if you remove gas.

AC is fine, aligns well with solar. Heating is the hard one, but heating is approximately baseload. The best candidate for that seems to be nuclear -- which can also potentially provide cogeneration for buildings in reasonable distance from the plant.

> In either case, industrial and commercial are ~60% of use and how do we move that significantly out of peak hours?

Businesses disproportionately operate 9-5, which is daylight hours. Many industrial operations operate 24H because the cost of paying people a premium to work second and third shift is less than the cost of building three separate factories that only operate 9-5. Now they'll each have to decide if it's also worth paying the cost of energy storage to operate at night.

I’ve always said this! Make daytime energy 100x cheaper and the market will respond.

Appliances will start building in batteries, behaviors will change , and who knows what else the market will come up with.

That's just externalizing costs though. Instead of building a large battery at the power station, you're building lots of little batteries everywhere else.

Even demand management can be an externalized cost. If a factory shuts down half the time, then you need twice as many factories to produce the same stuff.

> If a factory shuts down half the time, then you need twice as many factories to produce the same stuff.

No. It means you might need twice as much of whatever the bottleneck is, assuming that's a real time manufacturing resource which it might well not be. You definitely don't need twice as many whole factories. And if that bottlebeck isn't energy intensive that's not the problem at all, you can just re-arrange the energy intensive parts and you won't even need more factory.

I'm not sure I would call it "externalizing costs" if the only change was making daytime energy significantly cheaper (and not increasing off-peak energy prices). You could simply not add any batteries and your costs would still go down. You would presumably only decide to add batteries if the cost of doing so was less than the money you'd save by shifting energy usage to cheaper times.

How is daytime energy supposed to get 100x cheaper and still allow solar, which only generates during the daytime, to recover its costs? Unless you mean for the nighttime price to increase by a huge factor, but then how would that out-compete existing technologies for generation at night?

I think he means solar panels becoming so cheap that the total cost of solar is 1/100 of nighttime energy sources.

Even if the panels were free, you couldn't cover the labor to install them for that.

or the land. A lot of people seem to forget that...

You get a larger market share of the total energy consumption/payments to be during the day. E.g. because solar is cheaper, I shift my usage so that all of my energy costs are during the day, and if a majority of people do that, there will be enough money to keep the businesses afloat.

"Lose money on every kWh but make it up on volume" doesn't actually work.

The most electricity should cost at night is approximately what it does now, because you can get that from nuclear or hydro, and why wouldn't you unless you have an alternative that costs less than that?

The lowest sustainable daytime price has to pay for the panels, the labor to install them, the land, maintenance, loan interest etc. And a significant part of the consumer price for all sources is the distribution grid. The price difference is not going to be a factor of 100, and you'd be lucky to get a factor of 2 and still be able to pay back your costs, no matter how cheap the panels themselves get.

> The price difference is not going to be a factor of 100, and you'd be lucky to get a factor of 2 and still be able to pay back your costs, no matter how cheap the panels themselves get.

Perhaps not x100, but likely more than you think too, as european domestic PV installation costs were significantly lower than US installation costs last time I compared them.

I'm not sure what installation costs are now, but here in Germany several local supermarkets are selling PV systems you can install yourself on your apartment balcony for €199:

https://www.archyde.com/the-lidl-parkside-solar-panel-review...

I've seen other things like this with batteries for a similar price, though I can't remember the brand or exact specs.

Four quick solutions, there may be more!

(1) Transmission: When it is night somewhere, it is daytime somewhere else. All you need is transmission lines. Shipping electricity is so much easier than shipping fossil fuels, 40% of all shipping[1]: https://qz.com/2113243/forty-percent-of-all-shipping-cargo-c.... With vertical panels we can capture power early mornings and evenings, reducing the duck curve, but also for transmitting power long distances. There are long undersea cables being deployed, Denmark Wind to England recently, Morocco to Europe, Singapore to Australia. These can be bidirectional.

(2) CSPs with molten salt storage: In addition to solar panels, we could also leverage CSPs and provide 24/7 power. Hopefully, CSPs follow the same learning rate as panels: https://insideclimatenews.org/news/16012018/csp-concentrated...

(3) EVs: EVs can absorb power whenever it is overabundant. Power prices go negative 200 million times a year, EVs can charge at that time. Eventually, we'll have 100 - 200m cars, this is a gigantic, distributed and resilience battery. Cars are parked 22 - 23 hours/day, their primary purpose is energy storage, we get to take a ride on them for some time during the day.

(4) Wind: Wind is not daytime only. We can overprovision wind also, which complements with solar. "Offshore wind farms generate electricity from wind blowing across the sea. They are considered more efficient than onshore wind farms, thanks to the higher speed of winds, greater consistency and lack of physical interference that the land or human-made objects can present.": https://insideclimatenews.org/news/16012018/csp-concentrated...

We switched from circuit switched networks (the dominant network since 1850 telegraphs till early 2000, about 150 years), to packet switched networks. Packet switched networks transformed telecom infrastructure which led to all the developments we see today and the cost reduction to zero, which was unimaginable to anyone until early 2010s. Telegraph/phone used to cost a lot! We can imagine similar trajectory with electricity, the legacy grid being the equivalent of circuit switched network, which will be replaced with a new grid, with EVs taking the role of energy "switches" (store and forward switches enabled packed switched networked, store energy and supply the grid back with a few hours latency is the role of EVs)

We need solutions, likely a mix of them, because they all come with challenges:

1) Transmission takes time and $$$ to build (especially to get the land). Maintenance ain't cheap either. Longer transmission == greater losses.

2) CSP requires a specialized site and equipment, with limited generation/storage and greater losses

3) EVs as batteries to back up the grid externalizes battery wear to car owners. To compensate owners would probably cost more than just building dedicated battery back up on the grid (which would be much more reliable). This idea has never made sense to me.

4) Wind is a great resource, but also variable. There will be times when the wind doesn't blow at night.

HVDC transmission can help here as well.

Who has ever said that solar/wind or any other form of power is going to be 100 times cheaper than what we have now?

I think your parent is not comparing to right now, but is comparing times of abundance with times of scarcity.

AND how we manage our personal energy - for example, adding batteries to arbitrage the cheap daytime energy during the night

I do this today with residential batteries on a time-of-use rate plan that is highest between 3PM and midnight (with an average $0.30/kWh price difference between peak and low rates).

I end up arbitraging about 5kWh/day (1825kWh/yr ) of usage between those rates, which works out to about ~$500/year in utility bill savings. Compared to the cost of the battery system, that's not much and the payback period is well beyond the life of the battery.

In some locations (alas not mine), you can sell your battery's charge back to the grid during, providing grid ancillary services, for an even better ROI.

There is still, however, some resistance against opening the gates to leaf-node distributed energy resources from traditional utilities and generators.

The EIA (and the IEA) have historically underpredicted solar deploys, to a comical level. They need 2-4 years of spot on predictions before they can regain credibility on their projections, IMHO.

So I would put these projections as a bare minimum floor on how much solar will get installed. The true number is likely quite a bit higher.

> would put these projections as a bare minimum floor on how much solar will get installed

Is there evidence they haven't fixed this bias? The EIA and IEA are in general credible forecasters. Assuming past errors will persist doesn't sound like a good strategy unless you have methodological clues that they haven't made adjustments.

This isn't like trying to beat the stock market. There's no intrinsic reason why their predictions should be any good, and in fact they've been comically bad, coming in too low every year for 21 consecutive years:

https://www.pv-magazine.com/2018/10/09/iea-low-balls-solar-g...

(this is from 2018, but the same trend has continued up through 2023)

> no intrinsic reason why their predictions should be any good

This is where the "generaly credible" bit comes in. They're good forecasters in other domains. Part of being a good forecaster is adjusting as you make errors. So we'd need a hypothesis for why they wouldn't be making those adjustments.

As a trivial example, Hoekstra's chart shows the WEO serially predicting false plateaus between 2006 and 2018. Those plateaus are no longer the case for the IEA's solar predictions [1].

Again, not rejecting that this may be undershooting reality. But if your only evidence is past performance, that's likely oversimplistic.

[1] https://www.iea.org/reports/renewable-energy-market-update-j...

The link you're referring to only goes to next year, whereas the infamous pessimistic graphs tried to extrapolate for around a decade. I suppose capitulation would count as an adjustment.

> link you're referring to only goes to next year, whereas the infamous pessimistic graphs tried to extrapolate for around a decade

This CleanTechnica article references the short-term energy outlook. The context is near-term forecasts.

> capitulation would count as an adjustment

What else would it be?

If they are specifically bad in an area, with a very strong reputation, why should we assume that specific reputation would change to match the general reputation?

Also, I'm not familiar with any projections that are generally good. Collecting existing data? Sure they are good at that. But projection is an entirely different skill set and I have not ever heard somebody cite great projections from EIA.

Their predictions are comically bad, year after year they massively under appreciate the growth of solar. Just look at this graphic: https://pbs.twimg.com/media/DsX2rpPW0AIVORG.jpg?name=orig

You don't want methodological clues. You just want to monitor their predictions and see if they get better.

Well, we could try something with skin in the game. We could test our strategies. $1k to a charity of our choice 1:1? I'll say their best predictions today for capacity 2 years from now will be under the capacity 2 years from now.

If you're in SF, I'll send someone to make a deal with you. It's not on Kalshi or we could have some fun.

> > would put these projections as a bare minimum floor on how much solar will get installed

> Is there evidence they haven't fixed this bias? The EIA and IEA are in general credible forecasters. Assuming past errors will persist doesn't sound like a good strategy unless you have methodological clues that they haven't made adjustments.

I would contest the credible forecaster label when it comes to solar. By now we require evidence that they fixed their biases. If somebody gets something so badly wrong for over 20 years why should we assume they suddenly fixed their models. All historical evidence points to the contrary.

No, 2050 isn't some sort of tipping point[0].

[0]: https://www.liberalpatriot.com/p/to-avoid-the-worst-impacts-...

This link has no relevance to my comment or to this thread.

I’d hardly qualify a sub stack with a policy agenda as a source here.

Excellent.

Witness the power of the cost learning curve (Wright's Law).

Per Jenny Chase, we also need new wind power generation capacity to keep pace. If solar gets too far ahead of wind on the cost learning curve, it'll (further) spoil the financing of wind. Our glorious green energy future requires both wind and solar, in roughly equal measure.

This Volts episode is quite good:

Checking in on solar power - A conversation with Bloomberg NEF's Jenny Chase. [2023-11-29]

"longtime solar industry analyst Jenny Chase, author of Solar Power Finance Without the Jargon, catches us up on the current state of the global solar industry and looks to where it’s going."

https://www.volts.wtf/p/checking-in-on-solar-power

This was fascinating! Thanks for sharing!

We really shouldn't have waited this long to start transitioning to clean energy. If 2050 really is the point of no return (somewhere around then, from what I've heard), we'll have to replace over a century of infrastructure in about 20 years. And scaling is the hardest part. (we've already done some scaling, but renewables only account for ~20% of US energy, so we have to 5x that) Then, there's less developed countries, which can't fund subsidies and have unstable governments making it less likely for big infrastructure projects to get started anyways.

Not to mention oil-based economies. Many of them are just sticking their heads in the sand pretending oil will hold up their economy forever. This will be disastrous for them, and I'm sure many will lobby to stop clean energy (which will only make things worse for the planet). Only Saudi Arabia is really trying to diversify. As I said, we really should have started sooner so we could get the politics out of the way and we could just be focusing on deployment at scale today.

Information takes time to disseminate to people. I think the transition happened "as fast as it could" given that people tend to oppose change, have a hard time understand scientific results and science takes a long time anyhow.

If anything I would say "we should educate more people". Because that is the only thing that might prevent the next "we shouldn't have waited..." (for any of the problems that we could solve, but people just don't think/understand them - like antibiotic microbial resistance and other topics)

Solar and batteries are on an expodential path.

The cost is going down so fast it won't make sense to have anything else.

I think that by 2035 most of the transition will be done even if we don't have major new developpement, but if you look at what AI is finding in new materials I think that the next few years are going to be very weird.

Look into Tony Seby, he predicted the current path 10y ago.

I love having our energy grid being reliant on a geopolitical rival. I'm sure this won't end in disaster

>China makes 80% of global solar panels

>controls 95% of overall supply chain for solar

https://www.downtoearth.org.in/news/energy/china-to-dominate...

We only have ourselves to blame (mostly due to being in the pockets of Big Oil/Coal).

This is what the a big part of the Inflation Reduction Act was meant to address. You can thank Biden and Manchin for taking action to address your concerns.

I dislike when people criticize something without being inquisitive or acknowledging the elephant in the room.

What is your alternative? Continue heating the planet through burning oil/NG/coal? Develop solar panels domestically and wait to flip to solar (missing out on benefits in the meantime)? Nuclear?

If China stops selling us solar panels, the ones we already have will keep producing power and we'll be just fine.

I feel like you're making an analogy here to 1973 when Saudi Arabia cut off the supply of oil or 2022 when Russia cut off the supply of fossil gas, but the analogy doesn't work.

The real question is will growth of solar push electricity prices down enough that fossil plants get decommissioned early because they are no longer sufficiently profitable?

Almost certainly, assuming the growth of solar occurs as stated. The real tipping point is the fuel-only operating cost of a given plant, which is a function of the plant's efficiency and the price of its fuel.

Remember that many plants have futures contracts setting the price for both fuel and electricity sometimes 10+ years in advance.

Theoretically futures contracts should never affect wether an activity is financially worthwhile or not, but realistically a profitable contract can keep a plant online even though it otherwise makes no sense to do so.

Don't forget the CO2 price.

It depends on what you mean by fossil fuels. With solar and wind you absolutely need gas peaker plants to come online quickly and provide power during downtime so that’s not going away anytime soon. Batteries may offer some replacement but we’re not there yet. Pumped hydro is probably the best bet if governments can work out the issues with it and start getting it right.

They've already majorly affected what sort of plants get built. For example, most of the new build plants are natural gas based and designed to be started quickly to supplement the current production of power.

How hard would it be to include one graph that plots the relative contributions of solar, wind, gas, coal, etc to total energy usage, rather than growth in energy usage?

If the growth over the next 2 years is 10% of current total, having it mostly be renewable is nice, but not very meaningful. It is reassuring that coal growth is negative, but what is the total? Negative 2% is very different from negative 20%.

For the US, the eia has all of this info in a pretty accessible format.

For electricity - https://www.eia.gov/electricity/data/browser/

For energy - https://www.eia.gov/totalenergy/data/browser/?tbl=T01.03#/?f...

I don't get why solar is so popular.

In more northern regions, you get much less light in the winter. You effectively can only generate around one tenth of the energy you can in the summer - this is from experience talking to others who monitor their solar installations.

But, you need more energy in the winter. People are in more, using electricity for heat cos it's colder. I'm pointing out that there is a supply/demand reversal. Sure there are sweet spots in spring and autumn, but when you really need energy, you have less.

You can mitigate some minor fluctuations with batteries.... Ie you can smoothen things a bit, but you can't really take the summer's excess, to use in the winter.

It just seems like a fundamental mismatch, that leaves you exposed, when you need the most amount of support.

>But, you need more energy in the winter. People are in more, using electricity for heat cos it's colder. I'm pointing out that there is a supply/demand reversal. Sure there are sweet spots in spring and autumn, but when you really need energy, you have less.

Northern areas tend to use fossil fuels for heating at higher rates, and less heat pumps.

Let me introduce you to Scandinavia... https://www.euronews.com/green/2023/10/30/do-heat-pumps-work...

The answer is a bit simple. Only a small amount of Earth's population lives at north European latitudes (https://engaging-data.com/population-latitude-longitude/). Solar is quite effective for the majority of the population, so its adoption rate will have the biggest impact on further climate change.

I get that it's a great solution for much of the world - and that is fantastic.

But they were asking about its effectiveness in northern climates during winter (which I also live in). It sounds like your answer is: "it doesn't work there". If so, that's fine, we can focus on other approaches in such climates.

What seems like an emerging shortsighted opinion (on HN at least) is: "it's so cheap, just use more". That doesn't really work when: - you're in a place that gets a small fraction of solar radiation during the winter - it's often overcast - the requirement for electricity for heat pumps goes up drastically during this time - the nighttime requirement is even higher, as temps can get to -20 C or lower

I can't imagine the system you'd have to build to satisfy such demand with solar and battery (or similar) storage in the worst case, but I don't imagine it's "cheap".

There are solutions that are carbon-free; wind, enhanced geothermal (if fracking was legal in Europe), and nuclear. Solar electricity imports out of Spain or Morocco seems more feasible that solar situated in northern Europe.

The OP was saying:

"I don't get why solar is so popular" based on the fact it doesn't work as well in extreme Northern regions where only a tiny fraction of the world's population lives.

That's sort of like saying I don't get why boats/ships are so popular because they can't drive on land.

I literally started with "I get that it's a great solution for much of the world - and that is fantastic." ... so I think I'm on board with what you're saying. I do agree that the OP sounded a little too sour on it though.

And yes, I think we can all agree that Solar works well where .. it works well.

The issue is many folks are pronouncing it as some kind of a panacea. The reality is that there's a lot of places where it's marginal (and needs effective dispatchable fallbacks) and others where it's just not suitable. Luckily there's comparatively few people that live in those places, so it won't make much of a difference on the environmental bottom line. But for folks like me, these distinctions are important.

And "just add more solar" actually isn't a great idea, as it reduces the ROI on each panel, and there's a minimum amount of dispatchable capacity you need to keep online anyway. Sounds like it really does come down to storage at that level, most of which is still progressing to full productization.

Regions that get down to -20 C in a normal winter have a rounding error of the total human population in them, what with humans not functioning too well in such conditions either.

There are three ways you can have winter power with PV when you're close to but south of the arctic circle:

1. more cells, which is basically fine because of the very low population density and the cheap cells.

2. storage, which you have to be careful about regardless, because (no matter how you actually store it) a month's worth of energy supply in stored form is an industrial accident waiting to happen, even if you're storing it as a petrochemical that we already have standards for.

3. transmission, which is basically fine for everyone except Russia and North Korea, because everyone else far from the equator has decent trade relations with someone closer, sometimes (US, China) it's another place inside the same country. And yes, although it would take a while to build, the losses from HVDC are low enough that this makes sense on paper.

> I can't imagine the system you'd have to build to satisfy such demand with solar and battery (or similar) storage in the worst case, but I don't imagine it's "cheap".

Correct, but this is why so few people live in those parts of the world in the first place: getting stuff there, and staying warm there in winter, is and always has been harder.

In my native Netherlands, in the summertime there's solar, in winter there's is wind.

I have a dynamic energy contract, I pay spot prices + taxes and markup. What solar energy I sell in the summer, I can buy in the winter. With our heat pump, I need to generate about twice as much as I did before we had the heat pump, to match it in money (summer is cheaper than winter at the moment). Watt for watt is not a useful measurement IMO, since I can't store Watts, but I can store euros.

This just assumes someone else fixes the storage/overprovision problem, otherwise energy will get cheaper in the summer and more expensive in the winter until storing euros does not work.

That’s what markets are for: someone else will fix the problem because there’s money in it. It’s generally something you can rely on, that someone will try to get their hands on stored euros.

This is literally called magical thinking - there is money in breaking the law of gravity, so what?

Addressing storage | provisioning issues is not breaking the laws of thermodynamics.

We're talking aviation industry as a reasonable equivilance here, not magical anti-grav machines.

Reddit is really the venue for eponymous comments.

Actually they are exactly the same thing - seasonal storage with efficiency over 50% is basically magical infinite energy battery.

Just like anti-gravity machine, neither breaks the laws of physics, and neither has any known method for an engineer to implement.

Currently, you either need preposterous quantity of batteries, or you are stuck creating synthetic/green fuels at efficiencies of ~20% or less.

Solar PV systems has reached grid parity in many parts of the world (it's cheap), at a national level renewable energy means energy security and solar is usually used with others power sources like wind and/or hydro.

People closer to the equator tend to use more energy in the summer than in the winter, since they need more cooling than heating.

I think solar is becoming popular because it is so much cheaper where it works.

One point of disagreement is that you are saying "you need more energy in the winter". Right now, for example, tomato paste is made in a facility that runs year round. The main cost in converting tomatoes into tomato paste is in the thermal cost of energy. Having abundant cheap energy at one time of year will shift the way such industrial processes work so that we use more energy when supply is the highest rather than using the most energy when demand is the highest.

>Having abundant cheap energy at one time of year will shift the way such industrial processes work so that we use more energy when supply is the highest rather than using the most energy when demand is the highest.

That would require leaving a lot of capital idle, while is itself wasteful from both an economic and environmental perspective.

Solar is super cheap, making it a very effective complement to the matrix even though it's not enough by itself, for reasons you mentioned.

Every kWh from solar is one kWh less from fossil fuels. It seems inevitable that electricity becomes free when there's sun, because of solar.

In northern (and southern) regions the less light you get in winter, the more you get in summer. It is a zero-sum game.

You can't take the power you generate in the summer and use it to heat your house in the winter though.

One thing I've thought about, how hard would it be to set up a lot of solar panels in a region with 99% good weather (like Arizona) and then supply some of the energy to areas with less favorable weather?

Though that would be putting all your eggs in one basket.

It is possible but traditional power grid lines have quite some losses over longer distances. An expensive HVDC connection would be more efficient.

Furthermore Arizona could be too hot and dusty to be ideal. Solar panels work best when they are unobstructed and not too hot.

Most homes in the Northern US outside use propane, natural gas, heating oil or wood to heat their homes in the winter, not electricity.

Electricity demand in summer in the US is far higher than winter due to air conditioning.

Moreover, most of the utility-scale solar generation capacity isn't in the Northern US.

> I don't get why solar is so popular.

> In more northern regions, you get much less light in the winter. You effectively can only generate around one tenth of the energy you can in the summer - this is from experience talking to others who monitor their solar installations.

When we looked at installing solar on our house in Sweden the calculations showed that in winter we'd produce about 20% of the power IIRC. I just had a look for a source and I could only found this [1] which might be more like 10%. That said Sweden is quite extreme, all of the US save Alaska get much more sunlight in winter.

[1] https://adelsfors.se/2022/12/29/solar-vs-wind-power-in-swede...

I've done some math on residential solar in Ireland, on east-west facing pitched roofs, and it looks like a 4-6KW system would be break even in 5-7 years, given the current prices (44c/KWh) and the current subsidies (~2k for a ~8k install cost all in).

It's a little complicated because of time of use, and it's unclear if net metering or feed-in rates will continue.

While Ireland has a good chunk of renewables, mainly wind, our electricity rates are still driven by gas import prices, so they're 2x what they were a couple years ago.

Hydrogen could be viable way to do seasonal storage, it's generation reasonably efficient and can be stored in huge quantities. It's also a bit of a trojan horse for traditional energy companies

Just so we have some numbers:

$0.41 per watt utility-scale solar panel cost in 2020 ($0.17-$0.30 by 2030): https://www.energy.gov/eere/solar/articles/2030-solar-cost-t...

Cost projections (note the 2019 upswing where market forces and installation costs exceeded solar panel cost): https://www.solar.com/learn/solar-panel-cost/

$480/kWh utility-scale battery cost (falling by 32% to $326/kwH by 2030): https://www.energy-storage.news/li-ion-bess-costs-could-fall...

Solar energy will likely be stored in low-cost sodium ion residential and high-temperature sodium sulphur utility-scale batteries. This is probably a solved problem in the lab but will take 5-10 years to roll out at scale.

To put this in perspective, Bonneville Power in the northwest supplies some of the cheapest electricity in the US due to its installed hydroelectric base. But it is no longer able to compete with renewables, as solar and wind prices have fallen below its $35/MWh wholesale price:

https://www.bpa.gov/energy-and-services/rate-and-tariff-proc...

http://www.bluefish.org/notobase.htm

https://en.wikipedia.org/wiki/Levelized_cost_of_electricity

Now the cost-benefit of dams is being examined due to their environmental impact on fish runs (like salmon) and the starvation of wildlife like killer whales near estuaries.

Meaning that since solar is scalable, it has made nuclear the most expensive form of electricity today in terms of levelized cost. That doesn't mean that newer thorium and pebble bed reactors aren't worth exploring for high-power use cases, but that there's no way to build, fuel and decommission a nuclear reactor more cheaply than operating an existing dam which is already more expensive than solar.

The smart investment today is in energy storage and AC/DC transformers. Photovoltaics are second to that because they are a saturated market, but there could be opportunities in thinner and disposable panels, especially for electric vehicles.

I'm suspicious of those numbers. I personally can build a home scale battery storage system for around 280£/kWh. As in, I can go and buy the parts to do that and put it together. I can't believe they can't already do it cheaper at grid scale. That said, it might well be it's as easy to buy at small scale as at grid scale currently.

We are in a weird space right now where individuals can get super cheap batteries and inverters, but what's getting installed for grid scale is somewhat more expensive and meant to be cycled fully daily 5000-8000 times. Residential is much less likely to see discharges as deep and frequent as the large grid installs.

Welcome to the future. https://media.sciencephoto.com/c0/47/01/29/c0470129-800px-wm...

1) That ain't even bad 2) There are better ways to integrate solar into communities

> 1) That ain't even bad

It's not exactly a step towards a 15-minute city, either

Right, in a 15m city this would be on rooftops or integrated in the agriculture.

Being that we've not taken any steps toward the 15 minute city in the US, solar or not, I don't think that's much of a measuring stick.

No, but a 15-minute city would offer no opportunities for this to happen, except on hard town borders. Any plot of land this big in a dense city would be much more lucratively used for houses.

Better yet, there are better ways of integrating solar with agriculture.

solar is 10x cheaper than nuclear generation, but storage is 100x more expensive than nuclear.

Need a massive change in our understanding of storage costs to go 100% renewable.

Citations? I find this difficult to understand.

Do you happen to have a source for this comparison? I'm curious to know what the economics of grid scale electricity storage are.

Cost figures are very hard to come by as:

* Financing and returns are spread over 60 years for nuclear, which makes it hugely dependent on the interest rate. Which we don't know in advance, and also is at risk of irrational public perception (Leading in democracies to early closure of perfectly fine plants. That affects ROI hugely).

* Nuclear too profits from scale: you have to maintain an army of competent engineers. But once done, you can copy-paste plants cheaply. And raw fuel is less than 10 of the operating cost.

* LCOEs (financially Levelized Costs) are all over the place and highly dependent on plugging holes in production

* Chinese solar panels are state-sponsored to an unspecified amount

* Chinese solar panels are produced using coal electricity, and with lots fossil heat in the process. Models for pricing it in a CO2-free (or levelized with nuclear) are inexistent.

* Even calculating the EROEI (Energy Return Over Energy Invested) is difficult as you need to define an envelope. For example: do you need to account for producing gasoline from solar energy to power the mining truck necessary for your materials. How about powering it with batteries? Which you would need more mining.

* Battery storage needs are of humongous size. What's marginally fine now will probably be humongously expensive at scale. If done on lithium, we need about 180 years of mining for 28 days storage. And that's without accounting for other uses of lithium.

* Having cheap or expensive energy from one source may help you get more of another.

TLDR: At scale, it's a system with 150 variables dependent on each other, including geopolitical game theory hidden variables; and of course the climate crisis affecting the whole thing. But maybe we can trust physics: nuclear forces >> electric forces.

Other recent threads have mentioned gridscale battery storage on the order of 8-10c/KWh.

Lithium based is on the order of $100/kWhr: https://thundersaidenergy.com/downloads/battery-storage-cost....

FWIW, that's a different measure -- the 8-10c/kwh is for amortized cost over the lifetime of energy storage, vs the capital cost to build it.

It doesn't have to be batteries, hydrogen gas can be generated from solar at 80% efficiency, and storing hydrogen is way cheaper than batteries. Certainly way cheaper than nuclear.

Round-trip efficiency of hydrogen storage is at best around 40% though. But with abundant and cheap renewables, this might not be much of an issue.

Then you have to find a way to store it.

> storage is 100x more expensive than nuclear.

Please don't make things up out of thin air. Storage is cheaper than nuclear.

Look at the cost of storing 12 hours of energy for a 2 GW solar install (assuming long winter nights) or cost per kWhr of non gravity storage (since mountains aren't available everywhere).

Take these numbers from a smaller install on HN a few days ago, $390/kWh:

https://news.ycombinator.com/item?id=38942951

And you get roughly $9B for 24GWh. Rhis is roughly half the cost of Vogtle most recently completed $18B reactor, which is only 1.2GW and does baseload, so it provides less valuable constant power rather than dispatchable power that can match supply to demand. And that's before further price reductions for the batteries from having far lower power:energy ratio (save a ton on inverters), bulk discounts, and the thigh cost of doing any construction in Hawaii versus other places.

What about the solar to charge it and provide daytime power? Again, using high cost estimates of $1/W, and a capacity factor of 0.25, you'd need 8GW to supply an average of 2GW continuous power, which is $8B. And again, that's using a high installed cost, and ignoring that a major cost of install, the inverters, would be shared with the batteries and could be dual utilized. Solar panel costs are only 1/3 of the total installed cost of utility solar in the US, and batter packs are similarly only able 1/3 of the total balance of system.

So using excessivelt high, conservative estimates, a solar+batter install that could sustain 2GW average, but which is also hugely flexible and can adapt to peaks and troughs in demand, is going to be less than $17B, probably closer to $12B or $10B realistically. That nuclear reactor for an inflexible 1.2GW is $18B.

And then let's look at project completion risk and timelines. The nuclear reactor will take 10 years, and has a >25% of failing to complete construction, leaving billions stranded. The solar plus battery install will be completed in 2-5 years, with a >95% of project success.

That's today, but every year solar and storage costs fall. Nuclear might eke out a 5% decrease in cost when 10+ reactors have been built, but it's not going to catch up to storage and solar.

Gravity based storage won't be able to compete, according to any of the startups' preliminary numbers I have seen published. New hydro storage requires construction, which is super expensive for advanced economies, and is unlikely to be able to even match the costs of hydro built 75 years ago, and will likely be be far more expensive.

In any case, the idea that storage is 100x more expensive than nuclear is just not close to reality.

https://www.nrel.gov/docs/fy23osti/85332.pdf

Which energy storage is $1/kWh? I thought battery was ~100$

I can't match up your numbers with anything in that 20 page PDF or with my comment... could you clarify your question?

Sadly, wind has much better capacity factor, and works during nights and winter.

Why sad?

Getting the wrong ratio of installed green power generation means less CO2 reduction per dollar invested.

Markets are imperfect - even more so when government subsidies play a major role. Wind will play a growing role.

According to the charts, battery storage seems to be growing roughly in parity.

Battery storage does not help with seasonal differences.

That's why it's important to have a have both solar and wind, because in this case wind generates more power in the winter.

This forecast has wind up as well, just not as much as solar.

There's room for both. Solar works when the wind is calm. It's probably good to be diversified.

They actually anticorrelate, significantly reducing the amount of storage required when paired.

Sadly wind doesn't blow as hard during the night anyways

Wind is my favorite generation source and we need much more of it, but I can say that my rooftop solar in USDA Zone 5 (colder than most of the US) produced a net-surplus of energy for my all-electric house in December. So it's pretty good. Saying that solar "doesn't work" in winter is some mix of lazy, factually wrong, and unsubstantive. I've also got some standalone panels out in the yard right now hooked up to an LFP battery that I haul out there most days. Solar is awesome and I'm thinking about the best ways to expand.

Wind has a less aggressive experience curve than solar, so it's probably headed for secondary status in a renewable world. Places like India are probably going to get by with very little wind.

This is bad news for high latitude energy-intensive industries.

This makes me sad, as it means the industrialization of non-urban lands that will get covered with panels. Croplands. Wild lands. If subsidies were only provided for panels that cover existing urban hardscapes, it would be much better.

We currently have 40 million acres of corn dedicated to ethanol production. All our energy needs can be met by 15 million acres of solar panels. The same set of facts that make you sad make me happy.

But can I cut a maze out of solar panels?

Parking, flat roof, unprofitable argriculture land, there is plenty of space.

Reply on news.ycombinator.com