107 comments

[ 3.0 ms ] story [ 208 ms ] thread
Both TFA and the Bloomberg article it references [1] mention the 100MW power throughput of the battery, which is pretty meaningless without also knowing the energy capacity. For reference, the similar 100MW battery installation in Australia has 129MWh capacity [2].

This project is less remarkable when you read "About 2,100 megawatts of battery storage ... are in advanced stages of connecting to Ercot’s grid." [1].

1. https://www.bloomberg.com/news/features/2021-03-08/tesla-is-... 2. https://arstechnica.com/cars/2017/12/tesla-beats-deadline-sw...

Also for comparison: a single pumped-storage plant will often have storage capacity in the GWh; e.g. the Linth-Limmern pumped storage part in Switzerland can store 33 GWh and provides 1000 MW peak power.
When I overlay the population density map, on top of the elevation map. I think there aren’t many opportunities in Texas for grid scale pumped hydro storage. Although pumped hydro has huge potential it is not homogeneously distributed all over the world according to this map. https://www.hydroreview.com/world-regions/22-million-gwh-of-...
Bill Gates mentioned pressurized pumped-hydro storage in his book, (which was the first I had heard of it), and it sounds quite promising and something Texas is well-suited for:

> GLIDES costs as low as $13/kWh and $346/kWh and roundtrip efficiencies as high as 80% can be achieved using depleted oil/gas reservoirs and high-pressure pipe segments as pressure reservoirs, respectively.

https://www.energy.gov/sites/prod/files/2019/12/f69/02_ORNL_...

> pumped-storage plant

Seeing the title, I jokingly thought of generating hydroelectric power off a dam, and storing excess power by pumping the water back up, Sisyphus like. I'm very surprised this is actually a thing.

Taken to extremes of solar+pumped-storage, we could end up with a situation where the earth spins more quickly in the early morning when the moment of inertia of the planet is lower because the solar panels haven't started moving all that water mass to its higher elevation yet.
There is always a morning somewhere!
This is an excellent point and I can't believe I overlooked it. Obviously there could still be significant lopsidedness since places where you'd be moving mass around might not be evenly distributed, though. No idea why someone downvoted you for this observation. :D
Yes!

It's also only slightly less efficient than batteries.

Elevation translates in to 'head' and that's the single reason that this works well in Switzerland which has something that many other places do not have: very high mountains. With sufficient head the potential energy of a reservoir of modest size can be very impressive, and without you can have a huge reservoir that barely moves the needle.
I wonder if you could do something similar with underground storage in a mine?
Not really enough volume you could only use the lowest levels, and even the largest mines are dwarfed by small lakes. You'd have to put the generators and pumps at the bottom where they'd be easily submerged.

Typically pumped storage uses two lakes, one high and one low. But points for creative thinking :)

> which is pretty meaningless without also knowing the energy capacity

I disagree. You can reasonably assume a battery bank will have a capacity between an hour and a few hours.

Safe assumption, at least for Tesla Megapack based projects. The Megapack has a default inverter that supplies 1/4W of power per Wh of energy. But there are options to buy bigger inverters.
(comment deleted)
Good lord can we please move away from the common misnomer of kW, MW, GW, and TW when we mean kWh, MWh, GWh, and TWh? It makes me angrier than it should, but aside from BTUs (heating/cooling) it's the only one I'm aware of that is so stubborn.
As far as I read the article they are really talking about 100 MW maximum power output; they mention that the Australian version can supply that 100 MW for a bit more than an hour (so store a bit more than 100MWh), so they seem to be aware of the distinction but apparently don’t know (?) the energy storage capacity of the planned battery installation.
To put it in perspective. During the recent cold weather the State of Texas was short of 16GW of power for more than 40 hours making it 640GWh shortfall in amount of energy needed. You would need 6400 batteries 100MWh each to meet such demand. The only thing that a single 100MWh battery in Texas will be good for is stabilizing the grid against short-term load fluctuations -- the job a similar Tesla battery has been successfully doing in Australia.
That is a very important job though and is dramatically cheaper than peaker plants.
‘Undeniable success’: South Australia’s 129MWh Tesla battery

https://www.energy-storage.news/news/undeniable-success-sout...

South Australia’s Tesla ‘Big Battery’ saved consumers $116 million in electricity costs last year

https://www.startupdaily.net/2020/03/south-australias-tesla-...

According to recent figures, South Australian energy prices are 51% higher than Victoria, or 134% higher than Texas.

Throwing poor people under the bus (they're most impacted by high energy prices) so you can claim you're green (while opening brand new gas power plants) is pretty awful.

South Australia is still building new gas plants. Natural gas, which is nearly as dirty as coal.

I was extremely surprised in claim that Natural gas is nearly as dirty as coal as that contradicted my intuition. So I did some Googling and was not able to find supporting evidence. Below is a quote from: https://www.gasvessel.eu/news/natural-gas-vs-coal-impact-on-...

> Natural gas is a fossil fuel, though the global warming emissions from its combustion are much lower than those from coal or oil. Natural gas emits 50 to 60 percent less carbon dioxide (CO2) when combusted in a new, efficient natural gas-power plant compared with emissions from a typical new coal plant.

From the above natural gas power plants do produce a surprising amount of CO2, but is far far better than new coal plants. Legacy coal plants also are a significant air pollution source, so if we get out of coal power plants by building out natural gas seems like a win.

Finally I checked renewables for South Australia and they seems to be doing really great at 53%, compare to U.S. at ~23%: https://indaily.com.au/news/science-and-tech/2021/01/27/sa-r...

List of currently operating power plants and the ones under constructions corroborates renewable energy percentage and actually suggests that South Australia is doing great on renewables: https://en.wikipedia.org/wiki/List_of_power_stations_in_Sout...

>Natural gas emits 50 to 60 percent less carbon dioxide

You're comparing the absolute best natural gas plant to the average coal plant? The source also isn't neutral.

>Finally I checked renewables for South Australia and they seems to be doing really great

At the expense of the poor and minorities, which I guess are acceptable collateral damage in this crusade. [1]

[1] https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6920209/

Are you just generally writing about green energy, or about the Telsa battery (the subject of this discussion)? As the ancestor comment noted, the battery saved money, and saving money doesn't increase costs.
>You're comparing the absolute best natural gas plant to the average coal plant? The source also isn't neutral.

Source claims comparison is for new plants in both directions. Concur it is likely not neutral, do you have a better one?

> At the expense of the poor and minorities, which I guess are acceptable collateral damage in this crusade. [1]

Disproportional impact of higher energy prices is absolutely a factor... What I do not get is how much better South Australia could have done?

Nat Gas Greenhouse emissions are under reported a lot. The trouble is we do not know by how much. As a person who has spent decades in measuring CO2 my personal guess is 2-3 orders of magnitude. I also both coal and oil emissions are underreported a well.

It is not in people's interest to actually know the correct answer than major changes would be required.

South Australia pricing suffers from a failure of deregulation, not renewable power. The gas/wind/solar mix for generation is fine for a state with a very small population and abundant wind, sun and natural gas. SA has very poor deposits of brown coal which were not sustainable and the country lacks any nuclear power expertise. The state privatised their grid and it is now controlled by foreign owners. Many believe the pricing is the result of gold plating. Generation costs have fallen in recent years and wholesale prices are competitive with the other states.
I remember that some facilities were shutdown because their auxiliary power generation units froze. So if batteries would only supply these mission critical apparatus their overall effect can be multiplied.
(comment deleted)
Critical facilities in locations where the grid is shut down? Well in that case a giant central battery can't help those places either, you need on-site batteries.
Batteries are not going to substitute for winterized generation. They are to provide ancillary services such as frequency response (as you mention regarding the Tesla Hornsdale Power Reserve install), previously provided primarily by thermal generators, as well as short term dispatchable discharge (think expensive peaker plants). ERCOT (in Texas) also has a market for generators able to provide black start services (note the grid location of Gambit's install in this article, it sits at the intersection of transmission infrastructure connecting it to thermal generators in the area [1]).

During the rolling blackouts, no attention was paid to avoiding circuits that were powering natural gas compressors and other similar natural gas distribution infra [2] (sidenote: I expect Tesla to do well in battery backing such loads). More distributed battery infra makes the grid more durable, and less at risk for totally black starts [3] [4].

From the city's link to project information [5]:

"Will the battery provide energy to Angelton during a black out or natural disaster?

If charged, the battery can help the local electric system come back online by providing energy to ‘jump start’ electric generators in the. This service is called ‘black start’ capability; the battery is able to help the grid come back online after going ‘black’."

[1] https://openinframap.org/#9.18/29.1828/-95.4064

[2] https://finance.yahoo.com/news/giant-flaw-texas-blackouts-cu... (A Giant Flaw in Texas Blackouts: It Cut Power to Gas Supplies)

[3] https://en.wikipedia.org/wiki/Black_start (A black start is the process of restoring an electric power station or a part of an electric grid to operation without relying on the external electric power transmission network to recover from a total or partial shutdown.)

[4] https://www.texastribune.org/2021/02/18/texas-power-outages-... (Texas was "seconds and minutes" away from catastrophic monthslong blackouts, officials say)

[5] http://angleton.tx.us/DocumentCenter/View/3793/Gambit-Energy...

Smaller gas turbine gensets are far more effective at filling the black start capacity role than batteries are. You don't have to worry that they are charged. They don't really go bad as long as they are stored properly. You can run them on a huge range of fuels. They can be jump started & towed around by light/medium duty trucks if necessary. They produce reliable power for a reasonable duration (days-weeks). With batteries, you don't have much time to screw around before the big turbine needs to be spinning again.
(comment deleted)
I would argue that a water damn would be a better storage of energy than batteries. Although it can be environmentaly problematic depending where you put it, it can hold an enormous amount if energy which can be switched on in seconds.
(comment deleted)
TFA says 100MW, not MWh. So it's not clear if this is the maximum discharge rate or the actual storage capacity.

Side note – I don't think I'll ever stop being annoyed that we use Wh instead of J for things like this, very confusing to people and makes conversions harder.

Thanks, I can't believe I've never realized that Wh and J are measuring the same thing!
Just to be clear 1 Wh is not 1 J.

3600J = 1 Wh.

Yes, a J is technically a Ws (watt-second).

Or more accurately, since Joules are the fundamental unit and Watts are a derived unit, a Watt is a J/s (Joule per second)

Damn. Never made that connection. Have a upboat
It makes a lot of sense to specify energy in Wh in electrical engineering. It's much easier to convert to other useful measurements involving volts, amps, and watts than joules are.
>It makes a lot of sense to specify energy in Wh in electrical engineering. It's much easier to convert to other useful measurements involving volts, amps, and watts than joules are.

Easier how? A watt-hour is just 3600 joules - the number of seconds in an hour.

1 joule = 1 volt * 1 amp * 1 second

1 watt-hour = 1 volt * 1 amp * 3600 seconds

Actually the article does say that the 100MW is served for about an hour that makes it 100MWh. But I do agree that it is a sloppy journaism to confuse power and energy. Basic high school level physics.
That was in reference to their Australian installation, not this new Texas one.
Thanks for the figures. Can this also alter the economics of Texas' grid ? after all a stabler supply probably means stabler prices, the lack of control worsened the recent crisis I believe.
At a certain point you would think it's cheaper to soak up excess energy with something other than lithium-ion, like gravity batteries or flywheels, but apparently not...
Not an expert by anymeans, but it takes a long time for those to start producing energy when required (5 mins+). I believe in Australia, the Tesla battery was able to start providing energy in seconds.
I don't know that this applies at utility scale, but on a home scale they respond within the cycle; that is, they take up the slack before the next 60hz peak or trough.
Actually I think batteries will be the cheapest option for a very long time. They can be produced at a much higher scale than today (I wouldn't be surprised if we see 10x the production in the next five to ten years and consequently prices below $30/kwh) and they can be setup incrementally. Most alternatives are fixed-size all-or-nothing affairs. Oh and the biggest downside of batteries, their low energy density doesn't really matter for stationary storage.
Pumped-water storage is a big thing elsewhere, but afaik Texas doesn't have any big mountains to make this convenient.

(also, these dams have failed before to great misfortune downstream)

Lithium, iron, phosphorus, and oxygen are all pretty abundant and cheap. The economies of scale and better processes need to kick in (even more) to make them cheaper. Say, steel was very expensive until the Bessemer process became widespread.

Gravity batteries are very low-density, and flywheels are prone to dangerous catastrophic failures. They also need mechanical generators which aren't free. I'm suspect that chemical batteries are the future.

It is actually. Pumped energy storage is crazy cheap and efficient, assuming you have land and water.
Batteries are just straight amazing, aren't they? Every other solution is just so complicated. Like pumped storage. You run power to a powerful electric pump, which moves water into a huge reservoir, which produces power through a hydro plant. Or a flywheel. You run electricity to a motor, which spins a huge piece of levitated metal, which then can run a generator.

A battery is an inert block you send electricity directly into and get electricity directly out of. It's quite literally a black box of electrical energy storage.

No other solution is ever going to be quite that simple -- electricity in, electricity out. There's always some other complexity, some other mechanism, some other pump, some shielding, some motor.

Capacitors are even simpler. Even supercapacitors cannot compete with old battery chemistries, but it really doesn't get much simpler than storing electrons in conductive material.
I'm used to thinking of chemistry as a one-way street. The "moving parts" of chemistry are untold gazillions of molecules. That introduces a lot of opportunities for entropy, and makes it very hard to reverse a chemical reaction at anywhere near the efficiency of the forward reaction.

But modern battery chemistry upends that intuition. I'll be honest that I'm not 100% certain how. Whatever it is, it works unbelievably well.

100 MW for 10 msec is 1 MWh

100 MW for 100 h is 10 GWh

It helps to use proper units, doesn't it?

Large-scale battery storage is completely asinine when PES is far superior.

What is PES? I only can get results for Power & Energy Society, which seems tangentially but not meaningfully related
I think it means Pumped Energy Storage
I was wondering if he was suggesting we roll trains up hills with excess power and use regenerative braking or something like this, I hope he will respond
No need for a hill! You can dig a deep hole, and fill it with concrete weights.
What is PES?
Pumped Energy Storage (or rarely: Potential Energy Storage).

Aka pumped hydro.

In Switzerland, and maybe in, say, Appalachia, it's an obvious thing to do.

But if your land lacks serious mountainous features, like much of Texas, it becomes impractical. It's also seriously more capital-intensive: you need to pour like a million tons of concrete.

There are very few areas suitable for pumped energy storage. It requires specific topology and geology.
I like your point about proper scientific units for headlines, and disagree with pumped energy storage. If you aren't gifted with appropriate geology, it really isn't nearly as effective an option, is it? Also, response time is a lot different. There is enough room for both it would seem.
(comment deleted)
(Dons his prophetic hat.)

Even if this battery seems to be gigantic now, in 15 years it will look small, much like Falcon 1 is now just a tiny teeny rocket compared to Starship + Super Heavy.

(dons his prophetic tie clip)

In 110 years time Elons' interstellar Universeship will dwarf both his battery and the Starship.

This battery is obviously not for shock winters since capacity falls in cold weather. When i first saw the title i pictured in my mind a titanic model y rolling through the rural areas of texas...
If only there were some technology available for conditioning the interior airspace of a building so that it stayed at room temperature year round...
powered by....

(oh)

Powered by... the giant battery conveniently kept in the same facility?

No seriously: this isn't a problem. Heated buildings are solved, large ones stay warm for days once power is cut. There's a surprising amount of thermal mass in something that large, building walls are thicker, more simply constructed and easier to insulate than wood houses, and the mass to surface ratio is much smaller than what you're used to in a home. And in any case the waste heat produced by generation at that capacity is more than enough to keep an active facility warm, they most likely need to engineer for net cooling even during the coldest weather.

Tesla batteries can operate in freezing temperatures in heat mode.
Even bigger batteries entered service in California last year.

"At 300MW / 1,200MWh, the world’s largest battery storage system so far is up and running"

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

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.

Right, the difference between these older, larger facilities and Tesla's massive and giant facility is Telsa has a dedicated person with magazine editors on speed dial, while Dynegy doesn't.
There seem to be a lot of these? There are two different ones in Moss Landing, one by Tesla:

> What might be a little confusing is that PG&E itself is also building a similarly named battery storage project in the area - called Moss Landing BESS - at the site of the utility’s Moss Landing substation. PG&E’s project, currently under construction using Tesla Energy battery storage system equipment, will also be among the world’s biggest battery storage projects when completed, at 182.5MW / 730MWh.

I guess it's a good location due to the substation, which is probably there due to the power plant.

(Also, I thought Tesla got rid of their PR department?)

There are quite a few going up right now. There is a 50MW/4-hour facility almost complete near me in Pittsburg, California. PG&E has 750MW new capacity planned through 2023, and the whole state has 2000MW planned for the next 12 months. The state wants 2000MW per year for at least the next ten years.

I don't know why Moss Landing is so intense but basically any place with a peaker, an existing large-scale solar, or any kind of transmission capacity in California is getting batteries now or soon.

I'm having trouble figuring out how the money flows around these batteries. Any insight or links that could be useful?
I dunno, the Tesla Texas installation was secret, working from early last year iirc, and was working under the name Gambit LLC, and if the bloomberg journos hadn't done their own digging, no one would have known (tesla STILL hasn't actually admitted this is their subsidiary, afterall)

also, iirc, Tesla fired their entire PR department, so I guess we can safely say tesla does NOT have anyone with mag editors on speed dial.

The Name Tesla and Elon Musk just sells, I guess.

I wonder if the boring company has any plans to create tunnels that could be used for gravity energy storage.

After all, if they have the tech to continuously bore long tunnels, boring straight down becomes a much easier task as you don't need as much anchorage to provide forward pressure.

You can essentially use the front plate of a TBM, and have it "fall" into the ground with a large weight behind it.

Better idea (maybe): store even more gravitational energy by sending heavy weights into orbit. Just use booster rockets to make it fall through the atmosphere, regenerating vast amounts of energy in the process.
The problem then being: how do you remove the mass you're extracting? Existing horizontal TBM systems make this much easier.
Don't, just price it into the project and leave it down there.
I meant the dirt, not the TBM itself.
Geothermal looks promising, using the technology for drilling wells.
Can we have "100 MW" in the title, either after "giant", or in substitution with "giant"? I think it would be a much better title.

Edit: also, forgot to mention that most people don't seem to know that batteries used for these type of large projects are not in competition with batteries used in cars, in the sense that they are either of a different kind, or recycled/used ones.

(comment deleted)
Tesla recently build a big battery like this in Adelaide (south australia) to much fanfare. A friend who works reasonably high up in electricity distribution engineering tells me that you open the doors on the tesla-emblazoned cabinets and it's simply all siemens (IIRC) gear inside. (Not to say that tesla isn't value-adding by organising the sale!)
Electrical distribution is a solved problem though and has been for almost a century. The notable part of that battery facility isn't the Siemens "gear", it's the Tesla batteries.
Even the battery cells themselves are made by Panasonic, the packaging of a complete product is the value of the PowerPack.
Indeed, the Tesla powerpack is years ahead of the competitors, probably because they engineer only the hard parts themselves and use commercial products for the easy stuff.
Do you have a source for the “years ahead”. Not being sarcastic, I’m genuinely interested.
It's my opinion, mostly, but I've written drivers to control several different models and makes of battery installation, so firsthand experience. The difference is in the control software and reliability: Tesla batteries are way more reliable and have less downtime than anything else I've worked with. It's getting better.
heh interesting, although considering Tesla kinda made the whole thing extremely fast, I'm not surprised they rebranded parts already available from other brands.