The data behind New York's increasingly dirty electricity peaks
stevegattuso.me
stevegattuso.me
I totally agree with the analysis and conclusion of this article. The power grid is is a complex legal, economic and engineering system where our usual political narratives cannot produce a valid analogy for our common understanding. Right now, most green power sources (except non run-of-the-mill hydro) have non-controlable and hardly predictable outputs. Wind is much harder to forecast than temperature, and not every market are equiped to deal with the new variability. This create price distortions and other unintended consequences. Pushing against nuclear when there are still coal power plants running is nonsense, and in my opinion proves that twitter beats sound policy in terms of political priorities.
What I do not like in this article is the tendency to regroup all fuels type together and call them 'dirty'. Coal, natural gas and gasoline are not equals, and play different roles in electricity production...
That's why I love HN. Most conversations would have people like him chiming in and giving their views. We should all feel privileged that this forum exists.
- NY is moving away from nuclear (counted under "clean")
- none of the "clean" sources can easily scale on demand. Sun, wind, hydro and nuclear have either fixed or random output
- all peaker plants and less-desirable and so less-fully-maximized sources fall under the dirty category. Sun not providing enough juice? Ramp up nat gas peaker plants
No surprises here. We need batteries, more reliable clean base generation if we want to use less dirty peak generation.
https://en.wikipedia.org/wiki/Glass_battery
https://spectrum.ieee.org/john-goodenough-glass-battery-news...
Quebec hydro is actually developing it to go to grid scale. They have much better performance and none of the problems. In fact the early evidence is that the batteries somehow get better over time.
> The ninth research paper in Braga and Goodenough's "glass battery" work regrettably shows many of the hallmarks of pathological science. ... ad hoc theory, violations of the laws of thermodynamics, basic mistakes, disregard for established knowledge, absent or invalid chemical characterisation and, when all is said and done, devices that don't work the way they're said to.
http://lacey.se/2020/03/13/braga-goodenough-glass-battery-pa...
I wouldn't say it 'sounds too good to be true' when a nobel laureate is the one publishing. The same guy who invented the battery everyone uses today. With work being confirmed by multiple countries and multiple universities.
Flipside, this guy has a bias as he's a direct commercial competitor.
and as a battery expert he's saying things like:
>This is, as best I can work out, how it goes:
So he's not a battery expert? He doesn't understand?
>I am not especially familiar with field-effect transistors, but I will touch on this briefly to try and put it in to some sort of context.
https://en.wikipedia.org/wiki/Field-effect_transistor
A very simple and old tech that's in tons of tech today? I couldnt even tell you how many mosfets ive let the smoke out of.
>The electrolyte is almost certainly not a ferroelectric glass of extraordinary properties, it is a wet mush of different salts
This isn't an accurate representation at all.
Moreover, you need an absolute MASSIVE amount of storage to make intermittent sources like solar and wind equivalent to baseload power.
In winter months, solar in New York will produce less than 5% what the same facility will produce in the summer. Shorter days, lower sun, snow, leaves, clouds... all contribute to this.
One cloudy winter and the entire state would be without power - for weeks. It would be catastrophic.
The size of the water reservoir needed to replace plants like Indian Point plant don't exist on the East coast. You would need to flood absolutely massive areas of land.
Nuclear isn't an option. It's a necessity.
From the numbers I’ve seen, on rooftop systems, output is 2x in the 6 best months versus the 6 worst months:
https://www.lighthousesolarny.com/blog/2017/february/the-sea...
But some of this summer production increase is by design: summer power is worth more on the grid, so you over-design to capture more sun in summer (steeper angles, ignorance of winter shadowing, reduced focus on winter cleaning/maintenance) at the cost of winter production.
If you were building an off-grid system, the summer/winter discrepancy would be smaller. You might even overbuild for winter production at the cost of summer production.
I've generally seen pumped-hydro quoted as 80% efficient.
Such projects could scale to a large enough size to solve the intermittency problem.
This is pretty much a press release, but a bit of a summary: https://www.utilitydive.com/news/ess-sb-energy-softbank-reac...
That said, it's entirely out of my realm of expertise for vetting how good their tech or plans are.
"Approval for 100MW / 400MWh battery storage project at site of New York fossil fuel plant"
https://www.energy-storage.news/approval-for-100mw-400mwh-ba...
The project is expected to reach commercial operation by the beginning of 2023. Enabling the storage of electricity to be used when it is most needed will help increase the amount of variable renewable energy that can be put onto New York’s grid. It will also, as with some recent high profile projects in California, help reduce the state’s reliance on peaker plants; which are only called into action several times a year when electricity demand is at its highest.
What is the carbon emissions per kwh, cost per kwh, etc. I'm not saying that the company building this needs to give those answers.
But if this is part of the solution for climate change, those things will need to be known and reasonable.
The raw materials are lithium, potassium and iron. All are abundant but require significant energy to extract. So if we have a clean grid they can be produced with almost no carbon emissions. So if using "dirty" batteries is necessary to create the clean grid, it's still a massive win.
Everything is possible. But not everything is practical.
Energy density matters. Especially in industrial applications.
For example, electric passenger vehicles are a great solution because the weight of passengers is negligible.
Electric semis hauling freight? Not so much. The payload has to be reduced by a significant amount because the battery is so damn big.
I don't know what the extraction process involves, but I've been to a few mines. Truck, the size of a house moving massive loads. And lots of big stationary machines, which I assume are already electric powdered.
Which is why your question is so impractical. You're asking about kilotonnes of carbon on a project that will save megatonnes annually. If your question forces them to source batteries with a 10% smaller carbon footprint, you'll will likely have done net harm to the environment by delaying the project and forcing us to rely on dirty electrical sources for longer.
Your question is standard FUD technique. On the surface it appears to be a reasonable question. Because it is, at least in other contexts. But in the context of a green energy production plant, it's a distraction.
Yes, we need to make mines greener. So ask your question in the context of mines.
> Electric semis hauling freight? Not so much. The payload has to be reduced by a significant amount because the battery is so damn big.
More FUD. The batteries + motors weigh about one tonne more than the engines and diesel they displace. In response, Europe has increased weight limits for electric semis by 2 tonnes, and the US by 0.9 tons. Even if they didn't, a 1 tonne difference in a 35 tonne load is not especially significant.
So no new invention is needed, but some of this tech needs maturing/getting cheaper.
Extending the grid is also often helpful, albeit still somewhat expensive. I think we need someone to work on making that cheaper.
If you just need energy for heating, you can store the heat in a big insulated pond with an insulated floating lid on. That's cheap, and good enough for seasonal storage. There are several of those ponds in production already, village-sized ones, but they're going to build a town-sized one not far from where I live.
Your largest-scaling options are thermal storage (molten-salt thermal, not to be confused with molten-salt electric batteries), compressed air energy storage (CAES), and pumped-hydro. The last as with hydroelectric dams is limited by available sites and environmental impacts.
Any thermal storage or thermal-process generation (e.g., molten salt thermal storage, synfuel-based generation) will be limited by Carnot efficiencies, with about a 30% energy recovery to thermal input possible. Hydrolysis loses about half of input energy, hence the 15% return on synfuel storage.
Synfuels are another option. Most of these involve creating hydrogen, many (and the ones I tend to favour) will then combine that with carbon and/or oxygen to create hydrocarbon analogues or alcohol. These are very-long-term stable, and have high energy densities. They're valuable for specific uses already (portable power tools, vehicles --- especially off-road or remote, aircraft, and marine shipping). The total net energy recovery is low, on the order of 15--25%, but the storage capacity, the storage durability, the handling characteristics, safety, and extensive extant experience and capital for storage, transport, and utilisation, are all positives.
I've followed the electric battery story reasonably closely for about a decade. It's characterised by big promises and relatively low delivery. LiON is likely the best light-weight battery, for mobile and portable applications, simply based on chemistry. There are only so many light atoms, and the ones lighter than those we're using are exceedingly anti-social. (Notably flourine and chlorine.)
Air-metal, molten-salt, and molten-metal batteries might afford large-scale capabilities, though most research seems to have had limited success. All involve inconvenient behavioural properties of the electorlytes and cells themselves. None are well-suited to mobile applications. Several should be kept some distance from other infrastructure (e.g., residential/commercial zones, etc.).
Energy banking through direct thermal storage (hot water, ground/geothermal heat/cold storage, etc.) are possible, though would require considerable revisions to existing land-use and infrastructure interconnections. Reducing overall energy loads through passive designs minimising heating, cooling, lighting, and other loads, is also probably a factor.
We're headed to a future in which energy economics will be markedly different from those of the past 50, 100, 150 years. It's those economics which have shaped our activities, infrastructure, and land use. I strongly suspect all three to adapt substantially to the new regime. Assuming that the lifestyle we've become accustomed to will continue forward is probably at odds with future realities.
https://en.wikipedia.org/wiki/Hydrogen_production#Electrolys...
Carbon capture is also possible at low energy cost, from seawater, as described in a set of research articles published by the USNRL through the 2010s. The Google X Project "Foghorn" failed to develop this in an economically feasible manner, but my view is that that's more a measure of the mis-pricing of fossil fuels than of the method itself.
The resulting liquid hydrocabons are largely perfect analogues of petrol, kerosene (jet fuel) or deisel, and require no compression or refrigeration.
I’m not sure a reactor designed, built and operated by scientists at the university of Chicago has much to do with Idaho.
Note: as a native Chicagoan and a lover of nuclear energy, I take this stuff overly seriously :)
e.g.: https://www2.census.gov/geo/pdfs/maps-data/maps/reference/us...
I am personally disappointed because it provided good jobs in my community and 1/3 of my school district's revenue.
"Five states have implemented programs to assist nuclear power plants"
https://www.eia.gov/todayinenergy/detail.php?id=41534
It says that some nuclear plants in New York receive subsidies since 2017. I don't know what the New York selection criteria are. Maybe Indian Point just didn't qualify like other plants did, and (regrettably) shut down sooner as a consequence.
I.e. the rich people who have the spare fucks to give about this sort of thing.
If it were in a crap neighborhood in Newark nobody would care.
Because in the US, Twitter determines policy priorities, not science or reason.
This is exactly the problem when it comes to Indian Point. The only way out of New York is over a few bridges and tunnels. It isn't possible to evacuate even a fraction of the millions of people through a few bridges and tunnels if a radioactive plume is released only 36 miles away. There are arguments to be made for nuclear power, but that we should have one 36 miles from the most populous, densely packed city in the country, which also happens to be an island, is not one of them.
0: https://news.gallup.com/poll/153452/americans-favor-nuclear-...
The economics of building or even operating nuclear power plants continues to be unfavourable relative to either wind/solar, or natural gas (for peaking plants). Nuclear has, of course, a smaller carbon footprint than gas, but until that is built into economics (through emissions costs or higher costs for natural gas fuel), that doesn't translate into a financial benefit. Long-term costs of nuclear have been rising whilst those of alternative renewable and low-carbon sources (notably solar and wind) have been falling for well over half a century. If you're hiring^Wbuying based on slope rather than intercept, nuclear is not attractive.
Technically, nuclear power is poorly suited to peak-power loads. It doesn't ramp easily or quickly, and performs most economically when operated at constant power outputs. Gas (and hydroelectric or pumped-hydro storage) by contrast can follow demand-side changes rapidly, in a matter of minutes. For pairing with a variable-input solar-and-wind capability, something other than nuclear would be a better supply-side match. Pumped hydro, compressed-air energy storage (CAES), thermal-electric storage (e.g., molton salt), electric battery, load-banking (e.g., as thermal energy) or demand-side shaping (adjusting heavy loads to maximum generation) would be better fits. For now, natural gas turbine peaking plants fit the bill, though those also need phasing out.
Politically nuclear power is a challenge for numerous reasons, spanning those I'm raising and with others. The economics make for challenging financing and popularity, with very long lead and pay-off times. Cancellations of plants during construction or operation means that potentially-realised benefits are lost with major costs. In many ways, nuclear solves the wrong power problems (though it does solve the right emissions problem).
Nuclear continues to carry risks, and very-long-tailed ones, despite the claims of supporters. Many of those are not technical in nature, but operational, organisational, or reflect global threats outside the purvue of a utility itself. Nuclear plants typically have a paramilitary security presence armed, trained, and authorised to use lethal force. Relative to coal and oil plants, the net safety record is better, but one of the characteristics of nuclear power is the capability for things to go from operating very well to behaving exceeding poorly in a matter of minutes. This has happened repeatedly, across a wide range of designs, despite assertions of safety. Once things go poorly, then tend to remain that way for centuries or millennia. Long-term environmental consequences of fossil fuels notwithstanding, other power options don't have this specific handicap.
Options that are often cheaper than massive batteries and "more reliable clean base generation" by which I assume you mean "build new nuclear".
- don't shut down existing nuclear
- hydro as implicit or explicit storage
- interconnecting regions
- better green energy mix
- overbuilding green production
- geothermal
- demand management aka variable pricing
Hydro scales on demand actually, somewhat slowly, but it does. Pumped Hydro in fact turns hydro power into a battery for perfect response vs demand.
Nuclear can scale in theory: but its too expensive to scale down. The fuel is _basically_ free, so there's no point ramping nuclear power plants down. You spend all the money on safety / construction, very little on the ongoing costs.
The biggest Hydro plant in the US is the Grand Coulee dam in Washington, with a nameplate capacity of 6,809MW. This is pretty impressive, but it's a heck of a lot more than most hydro plants could ever hope to produce. Only one other dam beats out 3,000MW, barely, and most are in the ~2,500 MW range. Overall America has 79GW[1] worth of hydro power capacity.
Now ~2,500MW per dam is a lot, but it's actually kind of low compared to nuclear power. America's newest operating power plant, Watts Bar, has a nameplate capacity of 2332MW, which would put it at number 6 in the hydro plants list. And Watts Bar is fairly small compared to most other nuclear power stations around, since it only has two reactors. 6-7GW nuclear power plants are far from unheard of[2], and most of the older French reactors hang out in the 3GW range.
In places where hydro fits the geography, it's a great choice. Not without its tradeoffs, but miles better than coal or natural gas. But overall nuclear is still going to beat the pants out of it for versatility, and total generation capacity.
0 - To be fair, this can be a benefit in cases where you need to both generate electricity and store water for later use.
1 - Peak production. Yearly energy measured will be higher, roughly 270TWh, because there are a lot of hours in the year. Still a heck of a lot less than the ~800TWh of nuclear power we were making per year last year.
2 - Japan, China, and South Korea run a total of 5 nuclear plants with >6GW capacity. Japan and South Korea each have a plant with >7GW capacity. South Korea, Ukraine, China, and France run a combined seven reactors with a capacity between 5 and 6GW. The French reactors are interesting because they're all pretty old, from the 1980s, and were built and commissioned on a tight schedule, with an average construction time of 7 years.
Now, some people associate hydro with mountains. BC, admittedly, has something like 15-20% of the hydro capacity of all of Canada. Washington is similar.
Still, he biggest generator isn't actually BC, it's Quebec. Quebec has the largest hydro dam in Canada. It also has a greater total hydro capacity, and percentage of power from hydro, than BC. This despite Quebec not being particularly known for its mountains. Why?
Well, geography.[0] The Canadian Shield is pretty good for hydro. The mountains are more of hills, these days, but they were mountains once, and the corresponding rivers and height changes still exist. Also, you know that wonderfully cold Hudson Bay, especially the attached southern bit, James Bay? Yeah. It's cold. No one wants to live there. But water? Water wants to get there, so Quebec built dams, and now generates a bunch of power there.
Ontario, on the other hand, went mostly for nuclear. I'm not sure whether that's due to some inherent unsuitability of the land or rivers on the Ontario side of the Bay, or something else, but Ontario has several nuclear power plants, and relatively few hydro facilities.
[0] Okay, so I dramatized this paragraph a fair bit, but the message stands. Quebec has a very large amount of hydro generating capacity.
While true, they do correlate to demand. The sun is going to be the strongest when A/C demand is greatest, shaving peak demand.
Same with wind: a hot wind in summer will drive A/C demand. Same in winter with winds driving resistive heating or the 1hp load from furnace circulating fans.
This falls apart when your system becomes significantly double digit solar/wind, but smaller contributions should pair well with demand.
Hydro can be the worst because you get the most supply at spring thaw, but this the 3rd or 4th lowest demand season.
Nuclear and thermal are both slow to ramp up and ramp down, and unusable for short term peaks.
Hydro is actually great for that, because you can regulate the power production relatively fast (as long as you have enough water).
Most heating in NY will be carbon-based. For now anyway.
Not necessarily. There are lots of hot summer days when the sky is cloudy.
Also, there can be direct sunlight heating up NYC while there are clouds over the solar plant, which could be miles away.
About the latter, that’s where smart grids come in. Clouds move and it’s incredibly unlikely to be sunny but cloudy a few miles away for an extended period of time.
(Obvs don’t build solar where fog sits regularly).
That does surprise me a lot. Here in Brazil, it's hydro which follows the load. Nuclear has fixed output (always the maximum except when it's offline), non-nuclear thermal is mostly constant (with large changes whenever generators are dispatched on or off), solar and wind do their own thing, and hydro is the one left to fill the gaps between generation and demand. See it for yourself at http://www.ons.org.br/paginas/energia-agora/carga-e-geracao (select "SIN" on the first drop-down for the whole country); selecting "Geração Hidráulica" (hydro) on the second drop-down you can see how well the shape of the generation curve matches the shape of the demand curve at the top, while for instance "Geração Térmica" (non-nuclear thermal) is nearly flat.
Hydro can do load following. Nuclear can too.
https://en.m.wikipedia.org/wiki/Load_following_power_plant#L...
This is the talking point that needs to make the rounds in the media. Shutting down nuclear plants is beyond idiotic. Environmental groups need to do a 180 on nuclear, and politicians supporting this need to lose all "green" support.
The thing I do want to debate is the classification of Hydro as Clean Energy. The meager power generation that hydroelectric dams produce isn't nearly enough to offset the environmental damage necessary to produce a hydroelectric reservoir.
A more standard measure of progress would be carbon intensity (carbon emitted per kWh generated) I have a hunch that it doesn't support this argument as the decision to bucket everything into "clean" or "dirty" is otherwise bizarre.
It's also misses things like greater electrification which might be graded poorly on this, e.g. every car running on gas powered electricity is an improvement but will show up here as a bad thing.
Is it? If an energy source requires natural gas or coal to deal with peaks of increasingly hot days, this doesn't tell the whole story.
If we increase solar or wind to be a primary source, then we need to take into account the carbon cost of storage. Since this doesn't exist, it is an unknown.
Finally, the sun doesn't always shine and the wind doesn't always blow. Carbon emitted per kWh generated assumes it will be operating at 100% over its lifetime. So we will need an over capacity of wind and solar generators. The actual carbon emitted per kWh will be much higher.
We're there now.
https://en.wikipedia.org/wiki/Pumped-storage_hydroelectricit...
Put solar panels on literally every building. Put wind mills where you will. We know we need to do this. No nat gas or coal, we do pump storage. When we can, we do the next gen batteries at massive grid scale.
We dont need to shutdown anything early. Those are all amortized for their time. When we can, we eliminate those things as they come of age.
(Also "dual fuel" basically means gas, but with the ability to burn other stuff if they get into shortages, but its basically a last resort, its 99% gas)
NY appears to have plenty of hydro from that graph. It's not all that hard to close the sluice gates and open them later.
Grid scale batteries are also cost competitive (as of the last 12 months or so), though they wouldn't be as cheap as hydro for dispatchable energy.