Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

Interesting. There's a good video describing the operation here: https://www.youtube.com/watch?v=cOWjwwKSR78

After watching it, I now understand that they aren't flooding the entire borehole, but rather are building a smaller high-pressure chamber at the bottom. The chamber is connected to pipes to a surface reservoir. When air is pumped into the chamber, it displaces water up toward the surface. When generating, the water flows back into the subsurface chambers, forcing out the high pressured air.



A similar thing is done for natural gas storage. A well will be drilled into a rock dome that lies atop a layer of saltwater-saturated sandstone. Gas is then pumped into it, displacing the water, and then can be withdrawn later.

A report from 1961: https://www.ideals.illinois.edu/bitstream/handle/2142/42910/...


What goes into building and burying large scale underground structures like those air caverns?


you don't have to build those caverns as they already exist and we have extracted gas and liquids from them, also the volume should be already known as we know how much we pumped out (unless it's an old site)


Isn't the complexities related to making those caverns air and water tight for long period of time for long term energy storage requirements?


They don't need to be air/water tight for long term, just tight enough with a known leakage factor.


But how cheap is it to determine the leakage factor? I suppose they can just pour in the equivalent water and see if it leaks?


Not much, technologically this is a conventional mine project, only that you mine air volume instead of rock and have to take care of the cavern being reasonably airtight afterwards.


Also you could do it underwater if you live near a coast. Basically an upside down cup with a pipe in it, sunk as deep as possible.


The same company has actually done this[1] with the city of Toronto.

[1] https://www.youtube.com/watch?v=GicQwXbNnv0


The video is from 2015, and says it was a 2 year pilot project. Do you know if there are there any updates on how the project went? Is it still operational? Did they build any more?


Not sure, this is the page on the companies website: https://www.hydrostor.ca/toronto-a-caes-facility/


Is the voice in that video AI generated?


I wonder what is the loss/recovery %age.


I don't have specific numbers, but in other batteries of this type, the length of storage time before discharge affects the efficiency. Heat stored from compression is lost over time.


I was curious about that as well. I think it's best application would be to compensate for large amounts of rising and falling generation from solar grids, that way the cycle is daily. Being quick to come online and meet demand makes it good for maintaining grid stability at a much lower cost as well, like the Hornsdale Power Reserve facility https://hornsdalepowerreserve.com.au/.


The big question is how on earth are they digging out these caverns? That seems like a huge challenge, given that they're so far underground. Everything else seems relatively straightforward


It says repurposing old mining operations. Not sure how feasible that is.


I had a similar idea for repurposing an old mining operation. You take a closed gold, uranium or copper mine which is 1000-2000 meters deep. You create a nuclear rector at the bottom with "fire and forget" design. Only robots can do the maintenance. After 60 years you disable it, pour concrete into the shaft and move to the next location. If it melts, who cares? It's not gonna poison water supply.

Why was this not built before? I have no idea. Maybe robots are essential, we didn't have deep enough mines that were abandoned, energy was cheaper, no push for no emissions, it's not economical, other risks.


In France we’re storing highly radioactive items underground. Here is how dissipation works:

- After a few dozen years, the concrete is expected to breach,

- Radioactive atoms mix up with soil and dissipate both upwards and downwards, mostly thanks to water,

- After 400 years and for thousands of years, they reach the surface, where they should be diluted enough to not be dangerous,

So I guess having badly contained radioactive containers would be much worse.

One thing to remember is that pressure underground is extremely high (stone weighs a lot more than water, and light rock tends to “float” onto denser rock). If a melted reactor were squeezed, it would spread materials into the soil much quicker.


You need a hot and cold reservoir to generate power. Where is the heat going to go? This is why power plants are often built on rivers or near the ocean.


you need to pump water, probably flooding 80% of the mine. Since there are no humans involved 70+ celsius is not a problem.


The heat still has to be managed, otherwise Kyshtym[1] happens. 160 tons of flying concrete is no joke.

https://en.m.wikipedia.org/wiki/Kyshtym_disaster


The footprint of a nuclear power plant is pretty huge. That machinery is generally not optional even if doing a "fire and forget" design. Underground space would come at a premium cost, likely far more than could be saved on reactor design.

Also nuclear power plants obviously generate an absolutely massive amount of heat that needs to be dissipated, a task that would be difficult and expensive underground.


sure, but it doesn't have to be U-235 (thorium maybe?). It's more geothermal power plant with fission booster.


inb4: not a specialist. AFAIR remote control robots were tried when Chernobyl disaster was being cleared up, and the radiation destroyed the electronics. But perhaps the shielding technologies progressed enough to mitigate that.

Having nuclear stuff underground requires extremely precise geological surveys, so that the stuff does not wind up in aquifer.


lots of mining caverns already had pressurized gas that was extracted, so those caverns are pretty tight to pump back some air


Shaft mining is often used to reach underground deposits of iron, coal, etc. I believe they'd use the same process here.


Totally out of my depth here, but does cooling the gas down have the added benefit that it'll be heated by the earth whilst stored?


I'm guessing not. Earth's heat is way lower down. Think of a cave: always cold. I think what's going on is they're taking the heat out and storing it more efficiently than just dumping it down the ground, where (1) it would leak away, representing energy loss, and (2) as it leaks away, the air would lose pressure. (It would leak away b/c it's harder to insulate an entire mine, basically.) So, they pull it out ahead of time and put it back in when they decompress the air (which I think would add pressure back in due to heating a gas and maybe reduce problems caused by super-chilling other plant hardware? Anyway, seems fitting to add back what you took out if you're shooting for a closed system.)


This isn't correct. Deep mines are notoriously hot, because you're digging towards magma. Google says temps increase by 3° C for every 100m.


Do you have a source for that increase? I was not able to find it. Seems very suspect considering that the earths outer crust is approximately 20 to 30 miles thick. Are you suggesting that if we go approximately halfway through the outer crust, say 15 miles, the temperature is going to be 725 degrees warmer than surface temp?


Page 62 of https://archive.ipcc.ch/pdf/supporting-material/proc-renewab...

> The heat is transferred from the interior towards the surface mostly by conduction, and this conductive heat flow makes temperature rise with increasing depth in the crust on average 25-30°C/km


https://www.nationalgeographic.org/encyclopedia/crust/

Not sure about the rate per-mile, but literally it's that hot, yes, but not even 15 miles. Just down 5-7km under the oceans, for example at the Mohorovičić discontinuity the temp ranges from 392 to 752F.


Under the ocean is already much, MUCH closer however to the mantle. Oceanic crust is drastically thinner than the land we walk on.

In the Nat Geo article they quote a mine in South Africa reaching up to 55C (131 F) at the bottom and the mine is 4km deep. At a rate 3 degrees Celsius per 100m it should be 120 degrees Celsius over ambient. Which obviously does not add up.


They reached 160C rock when building an industrial railway in Japan:

https://en.m.wikipedia.org/wiki/Kurobe_Seny%C5%8D_Railway

Cooled down to 40C since the 60s.


Naive question but couldn't our forerunners who otherwise succumbed to cold in harsher climates have exploited this fact to dig subterranean villages and towns?

What's the element I'm missing as to why they didn't?


Pre-industrial holes may collapse long before you reach -100 meters. Even -5 meters is a challenge in certain places.

Also, digging (or rather drilling) in bedrock is hard without motorized equipment and good steel.

Also, supplying fresh air down there is a problem.

Also, preventing the mine from flooding is usually a huge problem.

Mining is hard and a lot of people lost their lives doing that. That said, if your only intent is to get a bit warmer, you may basically try a good cave. Caves tend to have temperatures above freezing for the whole year.


And of course the difficulty with caves is hibernating predator(s) (bear is the one I can think of maybe more?)


And lions:

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

IDK if these lions hibernated, but their bones are definitely found in caves all over Europe. Often with human remains or artifacts nearby.


Difficulty of digging with the tools they had, inability to stabilize the structure to prevent cave-ins, lack of pumps to remove accumulating water, insufficient ventilation and filtration technology/unable to deal with poison gases, inability to light their environment without contributing to the poor air quality, difficulty accessing resources like food and clean water, lack of desire to live in a damp, dusty hole.


Also for example Finland can be at least in north considered harsh climate. But these are also areas that suffered of ice age that scrapped most of the softer rocks away leaving only the tougher stuff like granite. Which makes digging very difficult, specially pre-industrial times.


That it is really hard to dig down 100m without modern digging equipment.


Ever tried to dig down a few 100m in your garden?


More like, ever tried to dig down more than a foot or two? Unless you have really soft ground, even a good steel spade and shovel aren't any guarantee of success. Powered augers are often needed to make holes for fence posts, etc.



Digging 100m down is difficult and dangerous


Have you been in a cave? They're chilly, usually around 50F, year round. IIRC those "deep mines" are a LOT deeper than any compressed air storage system is likely to be.


This comment is an example of why I’ve become skeptical of the downvote button. It sounds like people read it, knew better, and politely pointed out that it was incorrect.

Everyone who might have shared the same misapprehension learned something and the parent wasn’t being an asshole.

Doesn’t that add to the discussion?


Yes, when considered along with the corrective responses, it adds to the discussion. This is why it shouldn't be flagged and removed. On the other hand, it's factually wrong and would be misleading if people were to trust the claims uncritically, which I think justifies a downvote. Keeping it visible but lower on the page and showing it in gray seems like a pretty good compromise. Can you suggest a better way of giving more prominence to correct information? Or do you think that's not the right priority for the site?


Side question - do you need to reach a certain level of membership with HN to see the downvote button? I may be missing something, but seems that's only allowed for certain individuals.


Yes, there's a minimum number of points required to downvote. I think it's currently 500. Here's an unofficial FAQ that give more information about the thresholds: https://github.com/minimaxir/hacker-news-undocumented/blob/m...


Didn't downvote, but I am not a big fan of simply making stuff up on the fly as response. I've met my share of people who rather go this route than put in a 5sec Google search or simply admit they don't know something, and I try to avoid them.


"Looked the temp gradients up, after 50ft, temp of the earth is 50F, and raises one degree F every 60ft." (So, at 300 feet, 54 degrees; at 500, 57.5 degrees.)

(http://www.welshcoalmines.co.uk/forum/read.php?14,46717,4671...)

"...but other parts of the mine was very cold you made sure you had plenty to ware it depended where you worked"

(https://www.quora.com/Are-coal-mines-cold/answer/Keith-Scott...)

"Underground mines maintain a constant temperature, around 55 degrees Fahrenheit."

(https://www.quora.com/Are-coal-mines-cold/answer/Tom-Inghram)

«The temperature on average is in the 50s, but you still sweat an enormous amount when you start laboring.“ - Alan Bates, working in the coal mines of Letcher County, Kentucky.»

(https://www.quora.com/Whats-it-like-working-in-a-coal-mine/a...)

Yes, a 4000m gold mine is hot. I doubt that's where they're storing their compressed air.



> Earth's heat is way lower down

Very incorrect.


what would be the benefit to it being heated by the earth during storage?

all the temperature management here looks to just be counteracting the temperature changes that naturally occur when you compress or expand a gas. i imagine if you compressed it and then pumped hot air into water-filled caverns you'd get some negative effects of thermal shock.


Hot air should have higher pressure, so theoretically you could get more displacement. But your pumps also have to work harder to create a pressure differential. Hard to say which effect would be bigger.


Wouldn't the cooled down gas expand when heated leading to higher pressure, so you need to pump gas down?




Consider applying for YC's Fall 2026 batch! Applications are open till July 27.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: