Apple to build battery-based solar energy storage project in California | Hacker News Reader

Apple to build battery-based solar energy storage project in California

reuters.com

154 pointsdrdrey85 comments

Battery storage is a great tool for leveling demand day to day, but I wonder how soon we're going to need to start looking at mechanisms for leveling seasonal demand. Many places need a lot more energy for heating in the winter, right when solar production is lowest.

Synthetic methane seems like a great solution to these problems, and very few people are talking about it. If you used DAC for CO2, or got your CO2 from organic sources, and then combined it with green hydrogen to make methane, it would be a carbon-neutral fuel, and could take advantage of all the existing infrastructure we have for natural gas. Plus, it would also bypass the problem of trying to replace millions of gas furnaces and other appliances in existing buildings. LNG is ten times more energy dense than lithium-ion batteries by volume, and a hundred times more energy dense by mass; generating methane with surplus solar energy in the summer and storing it for later use in the winter seems like it might be doable.

There was recently on HN a discussion about a proposed solution for seasonal heat storage based in aluminum - air batteries.

Doesn't this simply mean you increase baseload during winter? Solar isn't a baseload power option, it's best used in a portfolio of energy sources.

Energy fluctuations are what cost Texas so much the last month, and that could be greatly mitigated with battery storage solutions.

To an extent.

But imagine you were writing a plan for northern europe to become carbon-neutral for electricity, heating and transport.

If you need the most power in the coldest, darkest months, and you manage to meet that demand with renewables that aren't solar, why have solar in the portfolio at all?

Honestly, good-old heat pumps work well for most climates. You heat up the ground underneath your house in the summer, and harvest that heat during the winter. That seems a lot more economical than synthesizing methane and hoping it doesn't leak back out into the atmosphere.

If we can get synthetic methane from atmospheric CO2 to work in an economic fashion, I imagine we could also use it to directly reduce atmospheric carbon (maybe by permanently fixing it into a solid). Problem solved!

Heat pumps are certainly a good option in some cases, but other times they are not practical.

For dense urban areas, there may not be enough open ground available to bury the heat pump coils. Also, in colder climates, many houses and buildings use hot water or steam radiators, and heat pumps can't efficiently heat water hot enough for this type of system. Retrofitting ducts to replace radiators and switching to forced air heating can be prohibitively complex and expensive.

Methane leaks are certainly a valid concern, but if we mandate better monitoring systems and do a better job of repairing leaks when they occur, it seems plausible we could get leaks down to acceptable levels. Methane is indeed a potent greenhouse gas, but it doesn't last nearly as long as CO2 before it breaks down, so if leaks are kept to a minimum I don't know whether they would ultimately be a huge deal in the long term.

> Methane is indeed a potent greenhouse gas, but it doesn't last nearly as long as CO2 before it breaks down

Obviously, methane breaks down to CO2 but there is no need to state this when referencing methane emissions. The Global Warming Potential can be used to relate the magnitude of damage over a given time horizon, for methane over 100 years it has 34 times more global warming potential. Obviously you can take exception to some of the methodologies used to calculate these but in general it's better than leaving the impression that CH4 emissions aren't important.

Interesting, I'd never heard of this before. If we didn't use carbon-neutral CO2 sources would it still be an improvement over status quo?

It definitely is doable, but the efficency of the process probably prevents market adoption, until fosil gas gets more expensive. (cheap fracking gas)

And if you use a methan fuel cell - you can use its surplus heat for heating while generating electricity.

I don't think they are market ready, though. But there are lots of options out there and the best solution is likely to combine what works now.

Btw. I think there was a breakthrough with natrium based batteries. (No lithium needed). They were just not as energy dense as lithium based ones, but that matters not so much for stationary use. So with no ressource limitation, you could produce them in very big quantities..

the efficency of the process probably prevents market adoption

I think this is correct right now, but the biggest cost by far for generating green methane at scale would be the energy costs for electrolysis to get the hydrogen from water. If renewables continue to fall in cost, we may rapidly reach a point where we have a huge surplus of green energy for parts of the year. There are already places in the US where energy prices regularly go negative due to excess wind energy, for example.

Folks have talked a lot in the past about using excess energy from renewables to generate hydrogen gas, but hydrogen alone is a lot harder to manage, is much less energy-dense by volume, is impractical to liquefy for storage and transport, and isn't compatible with existing infrastructure. Methane solves all of these issues, and doesn't take much additional energy to create beyond what you must spend for hydrogen alone.

Sodium ion batteries[1] are another promising one for stationary batteries that pop up from time to time. Sodium is obviously abundant and cheap, and has similar properties to lithium. For grid storage, it’s heavier weight and less energy density isn’t a big issue. However like many alternative battery solutions it it’s the there yet for whatever reason.

1 https://spectrum.ieee.org/energywise/energy/batteries-storag...

Lithium batteries are mostly nickel, despite the name. AFAICT the main benefits are replacing the nickel.

YC backed a company in this space: Prometheus Fuels. The science is sound, and the scaling is economical as you look at renewables for the electricity. Some links:

https://www.prometheusfuels.com/

https://www.sciencemag.org/news/2019/07/former-playwright-ai...

Prometheus Fuels looks cool, but (yeah, standard HN comment) their website is janky on firefox.

Fusion + Syn Fuel is definitely in the playbook to reverse climate change.

In the very long term converting excess generation into methane is almost sure to happen provided that can get accurately carbon priced versus ground extracted gas.

It's not even just seasonal capacity shifting, any situation that weight matters in will favor LNG over electric batteries for the foreseeable future even if batteries get 10x better.

There are many power-to-x chemistries that I think are much easier than methane once you grow out of easily available concentrated CO2. They don't quite reach the energy density of hydrocarbons, but that's not much of an issue for stationary storage in the "seasonal scale battery" use case. Almost everything can serve as an "oxidizable" in reduced form, and if the resulting oxide isn't gaseous you lose the "tank gets lighter" effect that is only really important for aircraft but win a solution to the capture problem for completing the cycle. Iron for example is said to get an energy density roughly comparable with coal. Another x in power-to-x is ammonia, which could be used in an open cycle scenario where DAC would be trivial because so much of our air is nitrogen.

I thought apple was already going to do something with methane and power, from what I recall when they announced the new campus.

I fully expect grid energy storage to be where used BEV batteries spend the end of their usable life before getting recycled.

It'd just be nice if the manufacturers would establish some vendor-neutral standards already, paving the way for this future. I hate the idea of buying a BEV with a proprietary battery of such obvious long-term value beyond my vehicle's needs, especially when it forms such a huge portion of the vehicle's cost. At this point I'm half assuming I'll need to DIY a powerwall out back when the time comes to replace my future BEV's battery.

I watched this last week and was blown away at how useful "used" cells could be in the right hands. Also, imagining what low skill labor looks like a battery offset world.

https://www.youtube.com/watch?v=PenPYwa00CA

My understanding is that used laptop batteries are particularly good for this sort of thing, because what happens is that one cell (out of six or nine) dies and the laptop battery becomes unusable since they are wired in series, and then you have five (or eight) used but relatively healthy cells to reuse.

I am not sure whether the same idea could be applied to a car battery with much more advanced battery management systems, since in those situations dead cells don't cripple the battery, so the other cells keep getting used until they lose efficacy. So I suspect cells scavenged from car batteries will be of lesser quality than those scavenged from laptops.

Not totally sure if I have it right.

I'm imagining for a more advanced BMS system that manages dead cells for you, you don't scavenge cells but instead use the battery pack as a whole unit, just separated from the car.

Think something like a conversion kit that turns your Model 3 battery pack that you've removed from your car into a powerwall.

People on the DIY solar forum already use old Tesla batteries for powerwalls so there’s likely to be kits in the future to make it plug and play.

Not sure about EVs, but replacing single dead cells is common on hybrid vehicles. It's pretty common for them to be series'd up as well.

The power plant operators do not agree, currently. the energy storage margin is thin and li-ion lifetime too short already... if you refer to grid scale. Look at the APS battery 'event' and the hazard to emergency response personnel batteries pose as a grid asset. Some form of home hybrid using recycled batteries may work, but the safety expertise and market payback don't work except at the DIY scale... Maybe it's a biz opportunity.

That sounds right. What utilities want are components that you install and thereafter take little attention. Re-using recycled batteries does not fit that model. They'd be much happier with something that cost more but lasted 20-30 years.

It's not even really a question of attention so much as maintenance. Fixed up front costs are easy to handle as capital that can go into their rate cases, but more so power equipment lifetime is measured in decades and the discharge lifecycles of available battery technology is already not great for an installation expected to go to near full discharge on a regular basis.

If the planned usage is only a 10% semi-daily scenario for stability support that may result in 5-10 years longer lifetimes versus daily discharges for capacity shifting.

They form a huge portion of the vehicle’s cost today, but battery prices have been dropping 21% a year. Assuming the pack in the car lasts 10 years, by then a new pack would cost just 10% of the one put in the car. At that point is it even worth reconditioning, adapting and reusing over recycling?

I didn't mean cost strictly in the monetary sense.

The battery in a BEV accounts for a substantial portion of the resources involved in its manufacture. The battery forms basically half the mass of the entire vehicle, and we're not talking about a sack of unprocessed dirt here. If it has usable life left for energy storage purposes as-is, we should absolutely take advantage of it.

The decision on whether to reuse or recycle should be informed by lifecycle analyses of the costs and externalities of both options. Maintenance isn't free of externalities and it will presumably be much higher in a system made up of older batteries. The opportunity costs shouldn't be entirely dismissed either - lower hanging fruit for removing emissions may be available elsewhere.

For Tesla's the battery pack is about 1/4 of the vehicle's weight. The battery pack weighs around 500Kg and the car is around 2,000Kg, depending on the model. The main problem as EV manufacturers look to shave weight the batteries are going to become more integrated into the car's structure, which is going to make them harder to remove and reuse the cells at the end of the car's life.

Do you have any examples of designs where batteries are literally integrated into vehicle structure?

The iconic Tesla battery-skate design is still modular/replaceable for example (they're all 18650/2170/etc cells).

Individual cells might in theory be modular, but Tesla for example is gluing them to their battery packs which makes them much harder to reuse at the individual cell levels. The battery pack it’s self is becoming a structural element though it can still be removed, but unfortunately it’s proprietary.

On top of this EV batteries are getting very close to lasting the normal lifetime of the vehicle. Car manufacturers are likely to tray and leverage this as a form of built in obsolescence by making replacing the battery unappealing under normal usage patterns.

Here's some pictures of Tesla's plans for it's battery pack:

https://electrek.co/2021/01/19/tesla-structural-battery-pack...

In that article it specifically mentions the cells being "bound" (their words not mine) into the honeycomb pack.

> The main problem as EV manufacturers look to shave weight the batteries are going to become more integrated into the car's structure

For those looking for a source, here's Elon Musk saying exactly that to Sandy Munro.

https://youtu.be/YAtLTLiqNwg?t=1335

I feel like it's unlikely for pack prices to continue to drop 21%/year. Tesla+Panasonic have gotten pretty good at reducing pack costs, and while Tesla's in-house cells could reduce costs significantly, IIRC, a 21%/year reduction in costs for a decade would be significantly less than the materials cost.

> It'd just be nice if the manufacturers would establish some vendor-neutral standards already, paving the way for this future.

I don’t think we’ll ever see this happen. Being able to tweak battery chemistry to suite different vehicles is just too valuable. I imagine at some point different manufacturers will start touting their specific magic battery sauce as being a reason to buy their cars over a competitor. Not to mention the onward march of incrementally better battery chemistries.

But equally the fundamentals of the battery chemistry don’t change that much, and in a recycling scenario where you don’t need to squeeze every last electron into, or out of the battery. You can probably throw a very generic battery management circuit over the top and call it a day.

As for the physical form factor. Again I suspect this will always be different for different cars. But equally, if you can find the positive contact and the negative contact, then the physical form factor doesn’t matter much.

So at the end of day, we probably won’t see much standardisation. But equally it’s probably not that important if you’re recycling. Fundamentally all the batteries will be electrically similar enough it won’t really matter.

None of that prevents them from establishing a standard size and connector type and a standard format for communicating properties of the batteries. NiCd batteries come in all kinds of minor configurations but as long as you know the basic chemistry and voltage you can use a cheap charger to charge them. Even if we hit a period of explosive experimentation with geometry (conical batteries, sheet batteries, string batteries, string batteries wound into comical coils made into slabs), we will eventually have a few specific form factors and that matters more to reusing these than the exact chemical composition.

Imagine if you had to buy Ford branded gasoline because that was slightly more efficient in Ford vehicles and therefore couldn’t be used in Toyotas. That’s effectively what you are advocating.

I think in a recycling scenario you’re gonna be stripping these batteries down to their cells anyway, so their physical format is kinda irrelevant.

As for their properties, you’re almost certainly better of just measuring them, rather than relying on some sort of digital spec sheet. You’ll have no idea what life these batteries have led, and thus no idea how much they’ve drifted from their spec sheet.

The battery chemistry pretty much determines all their important characteristics such as voltage, and charge/discharge pattens. There’s a very limited number of fundamental battery chemistries to pick from, and you can almost certainly ignore the secret source car manufactures are using. You just end up running the batteries slightly less efficiently, but again their at the end of their automotive life, which is the bit when people actually care about peak efficiency.

As for you analogy, it’s a bit more like wheels and tires on a car. There are hundreds of different tire treads, and plenty of different wheel sizes. But if you’re recycling a wheel or tire, you almost certainly don’t care much about either it’s size or tire tread, because it’s trivial to build a recycling process that can handle almost any type of wheel or tire.

> Apple said the project will store 240 megawatt-hours of energy, or enough to power more than 7,000 homes for one day.

By my reckoning this would make it the largest grid connected battery in existence. 50MW larger than the Hornsdale Power Reserve.

Edit: My reckoning was wrong.

There are larger. Moss Landing is about an hour northwest of this install (California Flats in southeastern Monterey County).

> Phase 1 of Moss Landing Energy Storage Facility was connected to the power grid and began operating on 11 December 2020, at the site of Moss Landing Power Plant, a natural gas power station owned by Vistra since it acquired the facility’s previous owner, Dynegy in 2018.

> At 300MW / 1,200MWh, the BESS is considerably larger than the 250MW / 250MWh Gateway Energy Storage project brought online earlier this year by LS Power, also in California. Not only that, but Phase 2 of Vistra’s project will add another 100MW / 400MWh and is scheduled for completion by August this year.

> The site at Moss Landing then offers what Vistra called a “unique opportunity” to expand the project’s size and storage capacity even further: the company claimed that the industrial zone in which it sits offers the potential to support up to 1,500MW / 6,000MWh of energy storage capacity, “should market and economic conditions support it”.

https://www.energy-storage.news/news/at-300mw-1200mwh-the-wo...

Nice! I hadn't heard about Moss Landing. It's only been operating for 4 months.

Looks like this [1] should be it - that's a lot of panels!

1: https://www.google.com/maps/@35.8713857,-120.3390994,7197m/d...

Scary to see a major energy storage facility less than 20' above sea level and right next to the Pacific Ocean. I hope they have flood walls built to withstand a Tsunami.

There is nuclear waste in the cliffs in San Onofre. They reportedly have nowhere to put it, and expect to only be able to store the depleted fuel there for a few decades.

Why isn't it in Yucca mountain?

Harry Reid? I am not old enough to know what happened there.

"In January 2019 Governor Steve Silolak vowed that "not one ounce" of nuclear waste would be allowed at Yucca Mountain, and a May funding bill did not include funding for the site. In May 2019, the Reno Gazette-Journal published a long-form essay cataloging opposition to the Yucca Mountain project. According to a tribal elder, the Western Shoshone view Yucca Mountain as sacred and a nuclear storage facility "will poison everything. It's people's life, our Mother Earth's life, all the living things here, all the creatures; whatever's crawling around, it's their life too." The tribes say they lack funds to discredit federal safety claims, but will be directly affected by a potential disaster."

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

Reading up on it.. they store it 18 inches above high tide in single walled corrosion prone storage that can't deal with ground water moisture that is expected to eventually seep in and they came within 1/4th of an inch of dropping a barrel in a way that would break the air ventilation cooling system and require water cooling and possibly cause a Fukushima like radioactive steam release.. just a few miles from Los Angeles.

Tesla/PG&E are also building 182.5MW/730MWh presumably at the same site? https://www.teslarati.com/tesla-megapack-moss-landing-instal...

That XYZ Energy Corp. of Delaware is building a storage facility, and that Tesla is selling battery packs in the same place, does not really imply two different projects. Could be the same project.

Not really. Look at google maps sat view. These are two different projects.

Bath County Pumped Storage Station("the largest battery in the world") can store 24,000 MWh, which puts it at 100x this battery.

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

This installation is right at the base of my family ranch. I am excited to hear about the development of new ways to generate and store power, but man if these panels don't mess up the natural beauty of the area. I can only assume the batteries will as well.

How does this compare with Tesla's offering, technology wise?

It seems to me that battery-based storage isn't very advanced technology, and is something that is enabled by declining battery prices. I see Tesla fans pointing to it as another one of Tesla's trillion dollar markets, but it never seemed like something Tesla would have an exclusive lock on.

Except for patents nobody ever has an exclusive lock on anything.

What Tesla does have is:

1. A helluva lotta actual battery cell production.

2. Practical experience deploying / analyzing / etc more Joules worth of batteries than probably anyone else.

3. Advanced battery research that is focused on more practical improvements than the pie-in-the-sky research we've seen a lot of over the past few decades ("this battery is flexible and has 10x storage of Li-ion and charges in 1 second") that never makes it out of the lab.

With (3), there is always the risk (to Tesla) that one of these futuristic battery techs actually works out and makes it into production, but until then Tesla's approach (making improvements to things we know work in production / at scale) is probably a wise one.

1 and 2 -> easily duplicated by some Asian country.

3 -> how much improvement are you talking about?

> 1 and 2 -> easily duplicated by some Asian country.

This implies some curious assumption that Asian countries are somehow innately better at this sort of work? I'm not sure that holds up.

Certainly the environments are different, different standards for acceptable pollution, acceptable levels of corruption, wages, safety, government subsidies, etc. These vary as wildly within Asian countries (Japan, Korea, China, ...) as they do between western countries and Asian countries though, and it's really not clear to me which countries environment is superior.

Assuming you're comparing to a high corruption low safety/environment/wages country... Lack of corruption and fraud goes a long way. Wages and safety cost less as things become more automated. I'm not sure how subsidies break down, but I suspect that there are more available for this sort of technology in the western world. Etc.

The point was that battery tech is not very advanced (compared to, say, building a microprocessor), and hence the assumption that an Asian country can easily duplicate it, especially considering the points you make (low wages, different standards, etc.)

- Battery tech can be very advanced. Tesla's manufacturing techniques are far from trivial.

- The most advanced chip fab (TSMC) is Taiwanese (i.e. Asian).

1. Tesla is outpacing Asian manufacturers, not the other way around.

2. Your first point doesn't really apply to my point (2), as that was more about how Tesla is actually getting feedback on their batteries in the field, through servicing their own cars / battery packs and getting constant battery data from the cars themselves and from use of their Supercharger network. This is incredibly useful, others aren't really doing it, and "being Asian" isn't really a silver bullet for this one...

3. https://www.youtube.com/watch?v=HK79ioBW8Mg

Tesla has first mover advantage in that they secured lithium salt flats in Peru back in like 2015. Last year they announced an aggressive pathway to halving the cost of kwh storage due in part because their business depends so much on it.

It appears this is Tesla’s offering

https://www.theverge.com/2021/3/31/22360839/apple-tesla-mega...

I am curious what battery technology they will use. I have recently found out about Ambri's liquid metal battery (https://ambri.com) which got me excited about grid level batteries. If this technology pans out it could be the missing ingredient for making renewables feasible.

Did they ever solve the corrosion issues with those?

Are there any online resources if one is interested in how to write a battery management system? Suppose I have some battery packs around, I want to learn how to properly connect them to a home grid.

I've come across diy powerwalls that have battery management, but haven't looked in 3 years

This is an open source BMS if you want to see how it's done. https://github.com/Teslafly/OpenBMS

Otherwise, there are lots of BMS's you can just buy a hook up to raw cells.

Need a plug n play open source design for 18650 cells kinda like Jackery.

Minimal integration pain & maximum expansion or swapping out bad cells without dismantling the battery pack.

>enough to power more than 7,000 homes

I get that this is used to quantify the energy capacity, but no homes are directly benefiting from this.

At some point you have to wonder what products these are practice for.

Cars?

Smart homes?

A whole City?

The Apple Arcology? Their headquarters is a prototype?

Wouldn’t surprise me, given Musk is hoping to do a Mars colony and Bezos is hoping to do O’Neill cylinders.

I've been waiting for over a decade for IKEA to get in on the action with an IKEArcology!

They already have model apartments with the square footage listed. They just need to establish different SKUs, and then instead of going to the warehouse to get your flat pack, you get a key card for the elevator and go up to your apartment! Though I think if the Swedish Meatballs became my staple diet, and I were exposed to their desert display daily, I'd die an early death.

I've had those books on Space Colonization by TA Heppenheimer, and some of the illustrations inside have a definite IKEA-ish vibe.

I would guess data center cost savings from off peak energy consumption.

Has anyone looked at copyrighting iPower?

Why not a private nuclear power station, Apple

I wonder how much more a project like this costs than in California than in a state like Arizona.

That’s about 2-3 million Teslas of storage.

No, a few thousand. 240 megawatt hours = 240,000 kilowatt hours, a BEV car is around 80 kWh

To be precise, 4800 if you are conservative and use the battery pack capacity of early production Tesla Model 3s (864,000 MJ / 180 MJ).

Thanks! Should have checked, my bad. Buying a few thousand doesn’t seem that newsworthy, Amazon just committed to buying 100,000 EV delivery vehicles...

Yes, because I trust a company that makes the design decision to glue their batteries into their laptops, to make battery facilities.

Reply on news.ycombinator.com