I work in the power generation industry. The large scale introduction of Battery Storage is going to be a game changer for renewables since it will allow overproduction during the day to be stored for the evening peak. I think the current fad of peaking plants is somewhat overblown, as the average large scale battery storage system will not face the issues with starting reliability that can be present on large gas turbine power plants, which not only affect the grid but can be very expensive as a failed starts consumes large amounts of fuel to no effect. In the end, this will drive carbon-producing power capabilities to largely only run during the night when solar is out of service.
Thing I harp on is the logistics of battery installations are fantastic in about every way.
If you have a couple of brown field acres of land next to an existing substation you can just buy and install containerized batteries. And you don't need specialized contractors to handle the job either. Pour concrete pads, forklift operators to take batteries off the truck and put them on the pad. And then standard HV techs to hook them up.
And the permitting and environmental review is nil. Go ahead try and get a pumped storage system permitted somewhere.
A fun one. Three gorges dam. You could replace the whole thing with solar and batteries for the cost it took to build it. And the area covered by solar panels would be the same as the lake behind the dam. Except you can put the solar panels on some ecologically and economically low value land where ever.
This is cool. The sheer amount of mater required to build hydro storage should not be underestimated.
Another win is that you can put something below the solar panels, even grow shade-preferring plants below them. This is being done in many places. Otherwise it could be a storage area, a light industrial facility, a shopping mall, even a sports field. And for all the battery storage you can afford, of course.
I think the answer for some area's is sheep. They can graze under the panels and don't chew on stuff like goats.
There is a whole interest in agrivoltaics as it's become apparent that the land under the solar panels remains productive. In the western US there is a lot of completely unproductive land to put solar panels on. But in the east and midwest being able to continue to use the land for agriculture is a win.
Also car parks. Shields the cars from weather, and provides power right where you want it for charging. Can even aid guiding rainwater runoff to avoid contamination with petroleum products. And reduces the amount of heat absorbed by giant swaths of black asphalt.
I’m not an expert on this topic, but I think it comes down to cost and scalability. You have to construct a new project with custom specs for the exact site you’re on, and the permitting for a large environmental change is another drag. For large scale batteries, they all are a somewhat complex power electronic wrapper around mass produced battery cell cans or pouches that can be dropped anywhere. The cost declines of batteries are undeniable and are not stopping anytime soon.
No grid operator going to mess with scrap batteries. Even for home player new cells are so cheap that it's not worth messing around with used ones (esp when considering liability and insurance). Might work in third world countries tho who have even more appetite for reliable power.
Grid battery deployment is faster than EV market penetration in the USA. This is not something that can siphon off a fraction of EV batteries. It is a huge market in its own right.
What would be considered ruined Leaf EV batteries that have significant range loss (20% to 50% degradation, for a car that went around 60 miles with a full battery) are being used to cover peak load right now in California and have been used to do so for years.
But that isn't a mainstream case. It's notable for being weird. The 1000x larger Moss Landing battery facility just uses new, off-the shelf BESS units from LG.
Interesting but it appears to only take account the topographic feasibility of the project.
The first location I looked at (link is not working :(), it relies on building two dams on rivers (which is something we tend to do less not more) and would flood some houses ...
And there is supposedly a price estimate, but it's just an A, B, C, D or E ranking could not find some costs to make comparison with battery storage for example.
Batteries can also be installed basically instantly, with storage being added incrementally. Unlike a years long project of constructing a hydro battery before it enters operation.
There is zero chance that a mega-project like pumped hydro can be undertaken in the US in the current regulatory environment. NEPA review allows even clearly green projects like solar and wind to get bogged down in environmental review lawsuits. A project with major environmental impact like destroying an entire valley ecosystem to build a dam would be DoA.
It's not just nepa. Citizens really don't like pumped hydro which negatively impacts it's ability to be deployed (you need local permits as well as federal).
Bear lake Idaho has been talking about putting in pumped storage since the 2000s. On paper, it's pretty much the perfect geography for it. However, various concerns from the impact on local fishing to the impacts on the lakebed have ground that process to a halt. It's a 20+ year project that's gone nowhere.
That's why I'm largely negative on pumped hydro. On paper it seems nice, in practice it's almost impossible to get off the ground. Batteries, on the other hand, take almost no effort to install.
Yep, agreed. We probably need new chemistries to get the seasonal storage potential of pumped hydro. But I’m quite bullish on batteries in general, it seems while folks have been gnashing teeth on how expensive they are, a substantial proportion of renewables installs in the US include 4-6H of battery capacity, which at least substantially helps with the duck curve.
> We probably need new chemistries to get the seasonal storage potential of pumped hydro.
Yup. It might be possible to overbuild solar/wind to solve the problem (just need enough juice to offset the night). Won't be possible with current chemistries to have enough power to cover cloudy days or snow-covered solar panels.
> it seems while folks have been gnashing teeth on how expensive they are
This to me seems like a fixed perception based on outdated information. Batteries were quite expensive and required loads of rare earth/horrible chemicals to produce... 30 years ago when NiCd was fairly prominent. Now they are dirt cheap and due to get even cheaper with sodium ion coming to market. We are already manufacturing TWh of batteries yearly. Enough that we could theoretically have a full day backup everywhere in 10 to 20 years without any growth in manufacturing capacity.
Do the math on how much energy you can store-- the volumes required for meaningful amounts of energy are just mind boggling. So you don't just need elevation, you need thousand of acres of valley or crater that you can flood and people tend to get a little testy about geoengineering at those scales these days.
> I think the current fad of peaking plants is somewhat overblown
Do you mean peak powerplants will become obsolete? As far as I know, in many places peak powerplants are hydroelectric, which in the future, aided by local batteries, will allow filling the reservoir lakes to higher limits and covering greater peaks - eg. malfunctions - which in turn will make grids more stable.
The big unknown, as far as I understand, is whether or not we have enough rare minerals to cover enough TWh of the daily peaks of energy-demands so we can have a flat daily curve. I'm sure in 10-15 years time the "renewables" will look a bit different than what we expect them today.
You are right: there are a lot of proposals at the moment trying to replace lithium. I'm hopeful, but let's wait until they are widely adopted before claiming victory. Let's not forget the carbon and environmental cost of such alternative batteries are not known because they are not mass built and mass deployed.
I think there is a fairly big difference between these examples. solar panels need pretty complicated semiconductors. Batteries on the other hand can be made by sticking 2 random metals in a jar with some salt water. since grid storage only cares about price per capacity rather than energy density, it seems unlikely for expensive materials to win out.
When I read people talking about replacing lithium batteries I think of the decades of reading about material X replacing silicon in IC's and solar cells.
Won’t it also be useful for power transportation as well? Since batteries on the usage side of a bottleneck can peak shave and trough fill to levellize the load?
There's a future issue looming: at some point, a lot of the natural gas peaker plants build for 2021 capacity will be idle almost all of the time. (Based on how often my local peaker plant seems to run, that may actually already be the case).
Keeping peaker plants around and having them ready year round for just a few days use is going to feel expensive; but decommissioning the plants too early has its own risks.
You can't directly extrapolate from April to November. April is generally the optimal month in California for renewable generation (days are long enough, rainy season is drawing to a close, not too cold or hot, and summer fog from the oceans hasn't started), so the graph is definitely showing the best case. Nonetheless, with more battery installs, we can almost certainly bank the majority of power needs all year round in California.
Here are some useful graphs showing how electricty demand varies by region, month and time of day [1].
As expected, California peaks in July because there's more demand for cooling than heating. But what's a little surprising is that July is peak demand for every region.
There are many reasons why solar is so attractive as a power source, including:
1. It's the only form of power with direct electricty generation. There's no heating water to turn a turbine;
2. Solar basically has no moving parts. You can point panels towards the Sun to increase efficiency but the panel itself and transmission has no moving parts;
3. Solar panels continue to massively reduce in price and increase in efficiency and it's not clear where this ends;
4. Obviously solar generates its pwower when the Sun is shining. This conincides with when most power is usedk. Adding solar to the power generation mix reduces the peak load required to generate from other sources;
5. Solar installations can be really small, including on individual houses, reducing the transmission capacity required for last-mile and regional power transport.
Batteries are just one method to store excess power for variable (renewable) energy sources. Another is carbon sequestration to create fuel directly, called Carbon Capture and Storage ("CCS") [2]. This isn't economic yet. But it may have applications in, say, cold climates where battery performance noticeably degrades.
More to the point, we want them to use electricity for heat instead. Heat pumps are quite efficient and natural gas emits CO2. But contrary to air conditioning load, heating load is inverse to sunlight, so we need a way to supply that electricity at night.
Heat pumps are so efficient that you can take the gas you were going to burn in homes to heat them, burn it in generators (losing about 50%), send it across long wires (losing a few percent more) and still come out ahead.
If you only need to do that at night, you're another 2x ahead.
If you then remember that wind, hydro, nuclear and (the main subject of this discussion) batteries exist, further reducing the times you need to use this displaced gas, it's an overwhelming win.
Examples like this is why electrification becomes more important than decarbonizing the last few percent of grid electricity, if the goal is to reduce carbon emisions and/or money.
> Heat pumps are so efficient that you can take the gas you were going to burn in homes to heat them, burn it in generators (losing about 50%), send it across long wires (losing a few percent more) and still come out ahead.
Coming out ahead by a few percent still means you need to displace the large majority of the current heating load with some form of electrical generation that can reliably supply a large amount of power at night.
> If you only need to do that at night, you're another 2x ahead.
You're not, because the significant majority of heating load is at night, because it's colder at night. Also, the night is longer than the day in the coldest parts of winter, and that's the time you need to spec the grid to be able to handle.
That doesn't mean we shouldn't electrify heating, but it means we're probably better off generating that electricity with nuclear, which is reliable and generates power at all times, than trying to somehow use solar to supply nighttime power in winter.
> 1. It's the only form of power with direct electricty generation. There's no heating water to turn a turbine;
> 2. Solar basically has no moving parts. You can point panels towards the Sun to increase efficiency but the panel itself and transmission has no moving parts;
This is kind of it's own problem, the fact that both solar and wind don't have moving parts that are directly connected to the grid and rotate at grid frequency necessitates a lot of other equipment to bring frequency stability back.
Some old power plants are being converted into synchronous condensers - the turbines replaced with giant flywheels - for this reason.
The big wattage numbers are great - but it is unclear how many watt-hours can be supplied - how many hours can California's grid run with batteries?
It appears that the batteries are 4 hour batteries. Two questions: is there excess solar capacity in the winter to charge the batteries? How do the batteries perform in the winter - total storage capacity and discharge rate?
It is misleading to compare a battery array that lasts for 4 hours to a nuclear powered electric plant that supplies energy at a constant 24/7.
"Developers plan to add 6,813 MW of battery projects in the California Independent System Operator's (CAISO) domain this year, dominated by four-hour lithium-ion systems, roughly double their additions in 2023, according to an analysis of S&P Global Market Intelligence data."
Because of the regulatory/economic regime the state has established, all of these plants are 4h plants, or 4Wh/W if you prefer. That's the number they need to hit to get the top-tier payout from the grid operator.
> It is misleading to compare a battery array that lasts for 4 hours to a nuclear powered electric plant that supplies energy at a constant 24/7.
It is misleading to compare the two:
* One has no risk of a nuclear meltdown in a very earthquake prone area (California).
* One does not have 60+ year long successful history of lobbying governments and establishing regulatory capture (note I wrote 'successful')
* one does have issue where they do not reprocess spent fuel adding billions of risk of nuclear fuel leaking into the environment; or if they do recycle, risk creating material for nuclear weapons.
* one would struggle with it's strong baseline, yet inflexible power generation without batteries to handle afternoon peaks
Keeping hearing random arguments for nuclear fission power (why not tidal?) and yet the regulatory capture and earthquake issue is still present in the US. We moved away from nuclear power after Japan's meltdown for a reason. With recent Supreme Court decisions and fracking causing earthquakes in Nebraska it has only gotten worse. I do not trust the industry with nuclear power. California spent a lot of money to establish a working grid that can be used for sustainable power until we get useful fusion (or that big supervolcano hits). Nuclear power is more needed in Texas right now, with it's broken grid, than California.
Arizona is a great case where agrovoltaics could really work. Solar reduces the sun load to something reasonable and minimizes evaporation, plants cool the air under the panels. Human / nature infrastructure symbiosis.
This is an interesting data point. Grid scale batteries are a pretty big change in how grids are managed, things like "anticipating load" and starting up "peaker plants" to catch that load are mitigated by the ability to wait and see while the batteries fill in the gap.
Batteries charge for free when renewables push spot prices to zero or negative and discharge when most needed and getting paid to do so; they comin’ for fossil gas generation that pays to burn (fuel) and spin (per hour O&M).
They might even be paid for consuming power because that provides stability to the grid [0] in such a situation. Too little load causes the frequency on the grid to rise which is only acceptable within narrow limits.
Absolutely, there is no doubt that are stealing grid service revenue from spinning thermal, and they also provide black start services (jump starting larger generators when they go out through an isolated transmission path).
I was under the impression that “grid forming” as opposed to “grid following” inverters were not a solved problem. Can you cite any battery plant with documented tests of energizing transmission or distribution systems?
Gambit Energy, owned by Tesla in Texas. Megapack supports grid forming in production (Hawaii and South Australia installations as well). “Virtual Machine Mode.”
It is solved, which shouldn’t be controversial considering Tesla’s power control maturity.
It's weird that you're talking about fission when the thing the batteries are really replacing is natural gas peaker plants.
Peak shaving (i.e. the thing storage is best at) is essentially the opposite of baseload. It gets you over the hump where the demand peak is after sunset but you don't want to have a 100% nuclear grid overbuilt to meet the peak demand that occurs when solar generation is zero, and you also want to stop using natural gas.
You still need something to generate power when the load is back to baseline and the peak shaving batteries are dry but it's 8 more hours until sunrise. Especially if we're going to electrify heating loads that currently use fossil fuels, which are higher at night and higher during the part of the year when solar generation is lowest.
"Baseload" clearly exists. There is a minimum load on the grid that it never falls below, which can be satisfied by constant-output generation methods that aren't as well suited to varying output, but also don't vary output and so can't don't create a shortfall when low output periods in variable sources last for longer than the usual amount of time or coincide with high demand.
> The sum of a bunch of intermittent sources is sufficient.
That only works if the intermittent sources are independent. In winter, solar output is going to be lower during the day than in summer, zero at night, you will have more hours of night, and heating load is both higher in winter and higher at night. There aren't going to be panels on one solar farm where this is the case and others that produce more, it will be true for the entire hemisphere for the entire season.
You're then left relying on wind if you don't want to use nuclear, but wind costs more than solar, and now you're lacking independence again. Weather is regional. There will be days when it's still and cold across a thousand square miles. Even batteries that could hypothetically take the load for a night are not going to have anything left by the end of a week of that, so you either need baseload generation methods or some currently unproven/uneconomical ultra high capacity long-term storage system or the week when that happens the power will go out and people will freeze.
As far as I can tell, anti-nuclear people either have a very shallow understanding of energy and can’t really have a conversation with you using well sourced numbers or they can but know the numbers show that nuclear is the only realistic choice and either way they just respond without numbers in very short and easy to repeat (and incorrect) arguments like “storage” and “bigger grids”.
Anti-nuclear environmentalists’ first choice would be to let people freeze to death. That being politically untenable their second choice is to continue using fossil fuels as the baseload and offset it as much as possible with renewables. They think nuclear is “an excuse” to continue the “business as usual” of affordable energy. They want energy to be unavailable and expensive. They oppose human progress and economic development as itself evil, even if the environmental impact is zero. Making energy expensive is their goal, not something they reluctantly accept. All the counter productive positions and bad arguments make perfect sense when you realize that they simply have a totally different goal from you.
> we found that the total annualised cost (including capital, operation, maintenance and fuel where relevant) of the least-cost renewable energy system is $7-10 billion per year higher than that of the “efficient” fossil scenario. For comparison, the subsidies to the production and use of all fossil fuels in Australia are at least $10 billion per year. So, if governments shifted the fossil subsidies to renewable electricity, we could easily pay for the latter’s additional costs.
They are claiming that a 100% renewable system would be CHEAPER than a fossil fuel system. If that doesn’t stink like some grade A bullshit to you I have nothing more to add.
That's an Australian academic in 2013 making that claim, and fair play to him, the official Australian cost estimates in 2024 say:
> ‘Firming costs’ is a term often used to describe the investments needed to make variable renewables a reliable source of electricity for our power system. In the GenCost report our preferred term is ‘integration costs’.
> Integration costs include investments in storage, peaking generation, transmission and system security devices such as synchronous condensers. Modelling determines the most cost-effective combination of these investments.
> ... renewables were still found to have the lowest cost range of any new build technology.
> For more detail go to the GenCost 2023-24 report section 5.2.1 Framework for calculating variable renewable integration costs on page 65.
Lol because you’re citing it. A study is not a magic spell you cast to immediately win arguments. Every study makes assumptions and you have to be prepared to discuss and defend them if you cite one. Another basic question, how much energy storage does their model require, because its probably way less that would actually be required. For comparison, this study of Germany found 56 TWh of storage to be optimal assuming a hypothetical and very cheap hydrogen storage system.
https://iopscience.iop.org/article/10.1088/1748-9326/ac4dc8#...
Or you burn a small amount of gas when these rare occurrences happen to coincide.
I suggest we call the periods that solar/wind/hydro/batteries can cover "baseload", as apparantly if you do so, you can pretend that the other energy being generated at non-baseload times does not matter, and simply declare victory with a job half done.
> Or you burn a small amount of gas when these rare occurrences happen to coincide.
But this presents a different problem: Instead of having to maintain several hundred GW of natural gas plants that you use on a daily basis, you now have to amortize the same cost even though they're only being used five days a year. At which point the incremental cost of building the same capacity in nuclear, which generates power 100% of the time and allows you to avoid not only the natural gas plants but that amount of solar and storage, is looking pretty good.
Why? Why does this new innovation in the california grid have to be all about dunking on texas?
(As a californian) It's especially annoying coming from anyone californian, the state of Enron and PGE wildfires. Like we are so caught up in the vanity of comparison that we quickly rush to forgive our past sins while condemning the mistakes of others for forever and ever.
How about as technologists, we just focus on innovating (which is not a zero sum game) and we don't compare ourselves to people who aren't doing as well.
Because, despite having vast amounts of renewables, Texas is a key political contributor to our lack of a cohesive national strategy for ditching fossil fuels. Its national politicians regularly insist that climate change is a hoax.
If a few decades of trying to use rationality have failed, I'm all for calling attention to their own local failures. If "dunking" on them is in bad faith, it is only because good faith arguments have failed so consistently.
How so? Would the Federal regulators have done a better job of making CenterPoint invest the right amount of money into tree trimming and placing lines underground?
California is under Federal regulation and has significant problems where PG&E hasn't invested enough in reliability.
If California just breezed through every natural disaster with no power outages and lower than Texas energy prices it would truly be a no brainer.
But Texas has retail power rates in the $0.10-$0.20/kwh range. From what I understand $0.20 would be an absolute bargain in California and that $0.30-$0.50 per kwh is more what folks out there pay.
It's not so compelling in that case. Of course I could have missed something. Please do let me know if I have.
Hitting a company with fines for not doing maintenance required by regulation, when the fines cost more than compliance, works well. Given sensible regulations, of course.
Look at airlines: rather strict and reasonably good safety regulations made flying the safest form of transportation, safer than driving and maybe even walking. Nevertheless, we don't see exorbitant prices either; if anything, the cost of air travel is surprisingly low, often competitive with rail, even in EU.
Of course, unreasonable regulations are a bane, see the leaded fuel situation in general aviation, or the whole nuclear industry.
Then you should also know the way they run their energy network is similar to our euro free energy markets. Many many people in Europe pay spot price, the kind that people on HN freaked out when the pricing went crazy is pretty normal in EU.
Texas has added a massive amount of batteries in a year or two. Worth noting, however, that as a fraction of their peak load California has about 3x as many batteries as Texas, because Texas uses far more energy and has much less rooftop solar. So while they are in the 2nd position it seems like a fairly distant 2nd.
I dislike right-wing silliness as much as anyone, but this whole hate against Texas is just weird. We had Montana winter weather for a week and our grid went down. I had a business partner in California during a period a few years back and we had to reschedule meetings around when he would have electricity. We should just focus on doing better instead of fighting.
The hate as you call it is not weird. It’s a reasonable response to a state that consistently enacts laws and regulations that hurt the people. You get what you vote for and people in other states have to put up with idiotic Texas politicians who vote against hurricane Sandy aid while later requesting help. Texas deserves far more “hate” than it gets.
Perhaps it would be more helpful to try to understand what underlying mechanisms have created situations in which apparently self destructive tendencies are expressed in large societies (not just Texas but in general). What can we do to improve on this?
Not a Texas-hater exactly, but there’s something specific here wrt culture and identity. Texans routinely describe Texas as exceptional, as different, as unique.
You may be right that there are generic problems at work, but when the protagonist insists that they aren’t like others, it’s not a surprise that people don’t think that way.
Not all silliness is equal or equally stupid. The Texas legislature routinely calls for secession when a Democrat wins the Presidency and votes to stymie federal aid to other states while demanding federal aid for themselves.
I don’t see the world in stark terms except where appropriate. You do believe cases exist where such views are warranted right? I happen to think that Texas qualifies.
California demands just as much federal aid, and perhaps more - they budgeted assuming that COVID-era federal funding would last forever, until they were recently forced to cut after having to face their absurd assumptions. San Francisco's city government also attempted to boycott working with any contractor in 30 states based on their political views, until it was forced to relent due to not being able to find contractors at an acceptable price.
California gets much less in federal spending than they pay in federal taxes. California is much larger population wise than Texas and as such it isn’t surprising that they need more federal aid. California also doesn’t vote against federal aid to states in disaster situations.
Socal and the general southwest has been setting records[1]. Some highlights:
> Palmdale and Lancaster on Tuesday experienced a record sixth straight day of temperatures at or above 110 degrees, surpassing the prior record of three days for both Antelope Valley cities. But wait, there’s more — officials say the streak is expected to continue the rest of the week, with highs forecast over 110 until Friday.
> Las Vegas on Sunday smashed its record high temperature by three degrees, hitting 120 for the first time since record-keeping began in 1937.
> Las Vegas is also expected to break its all-time record for consecutive days at or above 110 degrees, on its way there with six days in a row Tuesday. It’s forecast to remain just as hot through next week, which would beat the prior 10-day streak.
> Palm Springs on Friday hit 124 degrees, its highest temperature in recorded history.
It was 121º in Palm Springs and 100º in Riverside last Monday. I don't know what the population-weighted temperature was that day but that's pretty hot?
Meanwhile the San Diego heatwave is a scorching two degrees above average at 78 degrees[1]. Now if the Santa Anas start blowing during one of these inland heatwaves and the grid still holds up that will be great.
[1] If this sounds like sarcasm, try living in coastal San Diego for a few years. You really acclimate to the perfect weather and 78 really does feel awfully hot. To say nothing of those freezes where it dips into the 60s. I’ve seen bar patios turn the heaters on then.