Is it really worth it to dig a hole? I seem to recall some other initiative that was stacking large blocks with some kind of tested COTS construction cranes. The stack-of-blocks way seems to intuitively scale better. You can increase capacity without digging another super deep hole, or ever upgrading your cabling.
There are a bunch of these seemingly questionable ideas out there trying to score funding. A few years ago there was one to run a train up a mountain or create a giant compressed gas reservoir underground.
It may seem strange to use energy density in this context but in many ways those are even worse than batteries due to the massive amount of work they require and poor scalability because of all the physical infrastructure and space.
In fact, this exact scheme looks to have been posted to HN in 2014. From my understanding they are looking to re-purpose holes that have already been dug, for the most part, which is a good idea, but if the planned prototypes are only doing 250 kW nameplate with maybe 100 KWH of stored power, there is almost no way it's going to beat a battery plant you can have more or less just built to order already commercially. And said battery plants aren't even remotely good enough at purpose for meaningful energy storage, they're mostly used for what are considered "ancillary services" in the US power markets, like eating or producing reactive power, voltage support, etc, because the modern grid scale inverters can react to grid conditions faster than grid frequencies (50-60 Hz). More rarely they might be scaled big enough to shave the peak off a demand curve for 5-10 minutes to save transmission capacity or congestion.
Keep in mind all this in happening when LNG is basically free except for the cost of moving it, so you can also put in 1-60 MW combined cycle gas plants in short amounts of time with very well understood technology and proven manufacturers.
That's 110 MWs for 26 hours. Probably not 2.8 GW-hrs of energy storage (they can't sustain 110MWs for all 26 hours), but this plant is probably GW-hr range. I'm pretty sure all of Li-Ion right now is less than 2GW-hrs for the entirety of the USA.
---------
The 290 MW plant in Huntorf Germany (1978) also is successful.
--------
There's a 2nd, 300MW+ plant (that's 300MW power for multiple hours. So near GW-hr scale as well). This is the more recent Apex Bethel Energy Center you were badmouthing earlier. Its not fully built yet, but given the history of CAES, I'm optimistic.
--------
There aren't many CAES plants in the world. But from what I can tell, they are all safe and successful. The one issue is that you need to reheat the air as it leaves the caverns: compressing it underground reduces its temperature. A bit of natural gas is used in this heating process, but not nearly as much as an actual natural-gas plant.
50MWs is somewhat small, but there's something to be said about the simplicity and ease-of-deployment of rail energy storage. Those 50MWs were spec'd out with only 7 trains. It isn't too hard to imagine scaling the system up to support more than 7 trains.
Indeed; I'm not sure why exactly so few of the CAES plants have been built, it's well understood, has minimal safety problems, and doesn't seem unreasonably expensive. If I had to wager an uneducated guess, workable salt mines may not be located anywhere the power pricing makes sense for long term storage offsetting like that.
Yeah, I have to say I'm actually pretty impressed with the "rail up a hill" thing. Sure, 50MW isn't a ton, but it takes up nearly no space, infrastructure is very simple (5 miles of plain electrified track and that's it!), doesn't even need a crazy grade (7%), and most importantly: locomotives are cheap. IIRC, less than a million bucks for a good DC electric one, used. So you can be up and running pretty cheap and simple.
I'd love to see people running more of these setups. Especially in places where DC locomotives are cheap (former CIS countries), this could be even more cost effective.
Brief reminder that the predominate means of energy storage is literally pumping water up a mountain to flood a valley. On the surface, it's equally zany sounding, but of course the massive storage potential there is the big selling point.
Pumped Hydro energy storage is the #1 energy storage in America, and probably the world. A single plant provides 24GW-hrs of energy storage, dozens of more energy storage than all American utility scale Li-Ion batteries combined.
Bath County is the biggest pumped-hydro battery in the USA (and probably the world), but there are dozens of pumped-hydro plants across the country.
------------
Turns out that all the water in a lake can hold a huge amount of potential energy through gravity: just pump it up a mountain.
Thanks for the link and apologies for the skepticism. I didn't know we had systems that basically slosh water from upper to lower during high demand and lower to upper during low demand to even out capacity.
Hydrogen is difficult and consequently expensive to store. Also, the low efficiency is a problem. Pumped storage hydroelectricity is way more efficient.
Then combine with carbon from air and store methane.
Pumped is great when you have the geography for it but most countries don't. And building dams is pretty harsh on the environment too, just in different ways than CO2 emmisions.
Right, I think we will convert any hydrogen produced via electrolysis to methane for better storage. That methane would be completely compatible with the storage/distribution/usage infrastructure which many countries already have in place. I was commenting on the statement, that hydrogen by itself is easily stored.
At small scale, this is correct. At very large scale, hydrogen is incredibly easy to store. You can just fill up and old depleted gas well and store terawatt-hours of energy easily.
Germany has enough gas well storage to survive several months. This infrastructure is currently used for natural gas, but it can be repurposed for hydrogen.
I still think that commoditized battery storage will win in the end, but storage cost is not an argument against hydrogen.
Besides it is not trivial to repurpose natural gas storage for hydrogen - you need an entirely different level of tightness, you would be disappointed about the amount of energy you can store this way, as hydrogen is way less dense than natural gas, so the capacity of the storage would be poor.
For very large scale storage we are not talking about steel pressure vessels that are subject to hydrogen embrittlement, but depleted natural gas reservoirs which are just gas-tight geological formations and have no issues whatsoever with hydrogen embrittlement.
For small and medium scale storage you are right. Which is why hydrogen for cars and even trucks is a bad idea.
And you would get much less hydrogen stored at the same pressure than methane. To store reasonable amounts of hydrogen, you need very high pressures. Gas powered cars use like 60 bars of pressure, hydrogen 700.
Yes, that is correct. But the end to end efficiency of Electricity -> H2 -> Electricity is much better than Electricity -> H2 -> CH4 -> Electricity.
Hydrogen at 0°C and 100 Bar (~10MPa) has a density of 8.3447 kg/m^3. It has an energy density of 120 MJ/kg. So about 33 kWh/kg. So you end up with 278 kWh/m^3.
Now of course you have to multiply this by the H2->Electricity efficiency. Let's be very pessimistic and take 0.6 or 60%, which is what a gas turbine plant can achieve today.
You end up with 166 kWh/m^3 of usable electricity per cubic meter.
One of many german gas storage facilities https://www.nafta-speicher.de/en/company has a volume of 1.8e9 m^3. That translates into 300 Terawatt-Hours of storage capacity for just this one facility.
Right now the German gas infrastructure can only handle a hydrogen content of around 10% or 20% with modest infrastructure upgrades. This is why there are projects where that hydrogen is further processed into methane at which point all of it is compatible with the existing gas infrastructure.
That is a limitation of the pipeline infrastructure, because of hydrogen embrittlement of steel pipes.
It is not a limitation of the large storage facilities, which are just exhausted natural gas wells and work just fine with hydrogen.
The cost of outfitting one such facility with steel pipes that are not subject to hydrogen embrittlement is trivial compared to the cost for the electrolysis units etc.
Note that I am not a big fan of hydrogen at all. But large scale storage is one of its few redeeming qualities.
That was part of the old nuclear age dream back when a fleet of reactors would make power too cheap to meter, although using methane or propane instead may be easier to work with from an engineering perspective. Moving energy around as liquefied gas is also significantly more efficient in most circumstances.
IIRC there was some study on doing something like that in Germany in the 2000s but I'm not finding references to it now. If you don't have to build the dam (since it already exists), pumping water back up into the reservoir is going to be better in the limited locations you can do that, and given the relatively tiny amount of energy storage vs power production I suspect the economies of scale on that won't change without a preceding sea change in global commitment to fossil fuel retirement.
There is also a better transition path to that as we already are building amazingly good gas turbines that could just as well be burning synthesized or biogas on demand from solar or nuclear over-generation. I sort of suspect that may end up being politically poisonous to the people that would normally be pushing for carbon-neutral storage options.
It would be way more than double the solar panels though. Solar generation is generally maxing at around 25% capacity factor, so replacing an existing fossil fuel generator requires 4x or more the nominal capacity in solar plus whatever efficiency loss in storage, so going with a pure solar + storage replacement of a 1500 MW combustion plant could easily require 12,0000 MW of equivalent generation.
> It would be way more than double the solar panels though.
We're comparing energy storage so no matter how much fossil fuel powerplants we replace - it would be x solar panels for energy storage with 80-90% efficiency (for example this gravicity stuff or hydro or batteries) and 2x for methane synthesis.
Germany is already investing into power to gas infrastructure for both hydrogen and methane. What you are talking about is already reality albeit at a small scale.
I think the stacked approach has people concerned about the towers falling. This approach might be safer, but I think the other thing is it may aesthetically pleasing since the working bits are hidden below ground. This increases its viability.
Yeah, and as I mention above, the key thing is height. You can build 10x taller underground than above ground. And that gives you 10x more potential energy for the same amount of weight.
This proposal would likely be a lot more economic if they could reuse existing holes, like abandoned mines. However, mines tend not to dig down vertically and the cost of retrofitting and maintenance would be non-trivial.
The vertical train approach of terrament (linked above) could theoretically work with that. But we'll probably have to wait for the first prototype to see if it can really work with existing infrastructure.
It is worth it to dig a hole! Because building one tall tower is ~exponentially~ (edit: quadratically) better than building multiple shorter ones side by side with the same combined height. You can dig a mile deep which is about 10 times taller than the practical height of a tower. My startup, Terrament, is working on a similar technology, and there is a slide in our deck that illustrates this.
https://docs.google.com/presentation/d/17FI-jrI9RWS3q7Ng44Yh...
Reusing abandoned mine shafts could have an advantage. But there is also ample researching showing that the cost of digging from scratch is still worth it. The US Dept of energy even studied it back in the 1980s when they were evaluating underground pumped hydro energy storage. I wrote a white paper on it that's linked from here.
see here https://www.terramenthq.com/uphs/
It is not exponentially better. Energy stored in a stacked mass scales as n^2 whereas horizontal stacks scales as n. That’s quadratic scaling, not exponential.
Damn, you're right - thanks spot5010. I thought that I could use "exponential" colloquially and it would still be accurate enough (because quadratic scaling is even more esoteric to most people). But thanks for the callout, you're right that saying exponential is technically just wrong. I'll figure out how to re-word.
Take a look at Terrament slides. They are driving a train vertically into the ground.
With increased depth more cars can be added, increasing total weight, so m = Ch and ep =~ Ch*h/2
Yeah, I clearly need to change the wording here as it's just adding confusion.
The key thing here is that adding 10x height gives you 10x more PE per weight without extra digging costs (or tower costs assuming the cost per unit of height is constant)
Here's how to think of it: by digging one deep hole instead of many shallow holes (with the same amount of digging) you get (n^2 / 2) instead of (n/2) Potential energy. The average height of all the modules is half the height of the shaft. With one deep shaft, the modules pass through the same excavated volume so they can all go deeper on average.
There is no stacked mass however. Just a single piece of mass moving up/down that results in linear scaling from height.
Edit: reviewed the Terraent slides they indeed plan to literally drive a mile long train vertically into the ground. In which case the quadratic argument stands.
At the cost of significant added complexity. It comes down to drilling cost vs installation/ maintenance cost.
But also, getting your quadratic gain on a mile deep hole requires a mile of weights to lower into it. That's a lot more above ground infrastructure than just the single shaft!
Haha well, 70 of those slides are appendix slides.
The way our design works, the modular weights are autonomous on a track. So we don't have huge cranes above ground. We just have a track above ground that also runs a mile. That track can be partly or fully buried if desired. It will be enclosed protecting it from weather, and solar can be installed on top to save real-estate.
But yes, these installations are enormous. They might cost around $150M and could provide about 200MW.
You're not making any sense. mgh is mgh. If you lift the blocks above or below the ground makes no difference. But you absolutely save a ton of money by not digging the expensive hole that will fill up with water.
- Keeping water out of a hole is a solved problem.
- as discussed in other comments here, adding more height gives you more PE for the same amount of weight so it's a very valuable investment that pays profits over 20-100 years for your one time investment.
Energy storage by lifting or releasing a weight, in their case, in a well underground. Nothing new; I am wondering what's special about them, compared to tens of other companies trying to offer the same type of energy storage.
AFAIK, the most common is done around hydro dams, by pumping water upstream as a form of energy storage. The infra is already there, but it's not as efficient as systems like Gravitricity. But it costs a negligible amount of money to "activate" energy storage in a pre-existing dam.
I believe the argument for new storage concepts is that renewables will drive a need for large amounts of new energy storage capacity, batteries are expensive, and that not everyone is situated near a suitable geography for pumped hydro.
Interesting concept but digging a big hole sounds expensive and prone to flooding. I can see how this would have better efficiency than pumped-storage hydro but I wonder if you couldn't raise and lower weight near cliffs rather than drilling wells.
If it does work, I would imagine the Boring company would be all over this.
I can see why you'd suspect that. But it turns out that the oil and gas industry has 100 years of precedent demonstrating how to build such shafts into bedrock. Flooding and earthquakes are all real concerns, but we have well-proven solutions.
Maybe not as large as possible, but we're looking at about 10m diameter and about 1 mile deep. Here's an example of existing mining tech that meets this capacity.
https://www.youtube.com/watch?v=z5vAWR7rpco
Isn't this the exact same concept as a dam which pumps water back up when you have excess power? Also how heavy does that weight have to be to be able to store enough energy worth digging a hole that large? Can you even get a weight heavy enough to store a significant amount of energy?
Same concept, but the tl;dr is that no, you can't, unless you're talking hydro from a dam in a good location. It's not just a not-so good idea, it's an orders-of-magnitude-awful idea when you compare it to just about any other form of energy storage per dollar.
Anything involving dams typically also has to involve water release schedules, which at least in some parts of the western US will produce some of the most vicious political battles you can imagine.
You can in fact! Pumped hydro is already the cheapest form of energy storage - about 95% of all energy storage is pumped hydro. We can't build more pumped hydro because it's not feasible to build enough dams. (We've used up the best locations)
The video has some numbers in it about performance. They say dig a shaft of 150-1500m and hang a weight of 500-5000 tonnes. This translates to energy storage of between 204-20400 KWh storage.
Obviously, this seems very technically challenging. 500 tonnes is 64 m^3 of iron. We will see if their engineering is good enough to pull off their claimed 171 US$/MWh.
26 m^3 of tungsten is what, less than a 3x3x3m cube? If they stack 10 cubes at the bottom they get mostly the full storage (-3m at least) for each weight. Seems feasible.
This is a technology where energy density doesn't matter at all because the amount of energy stored per ton is the same. So you want as many tons as cheap as possible. Therefore we either use water or concrete.
Quite a few similar companies are looking at compacted earth mixed with concrete to reuse some dirt and rock that is excavated. See also https://energyvault.com/
I always thought it'd be interesting to do the gravity storage inside the column of a wind turbine. After all you've already built a tall steel tower. Individually it's tiny storage, but in aggregate it's not meaningless, especially as we keep building more towers.
But I assume there's structural issues with suspending a few extra tons on your structure.
I feel like "gravity storage" should take advantage of natural landscapes better.
ARES (rail energy storage) builds a rail-line uphill, for example. Rail cannot handle a very steep slope, but a gentle hill climb will build up potential energy fine.
In the case of vertical-based gravity storage, I'd imagine that lifting blocks to the top of a cliff (or down a valley) would be most efficient.
I mean, Pumped Hydro is gravity storage, and does just that. Pumping water up a mountain and generating energy by dropping it back down. But presumably, we don't want to use water in the Western states (where water is scarce). So Gravity-energy storage WITHOUT water is the goal.
The issue with using natural landscapes is that it limits your height. And adding more height is ~exponentially~ (edit: quadratically) better than just adding more installations with an equivalent total height. (You can dig about a mile deep). You can see more details in my comments below - search for the text "It is worth it to dig a hole!"
I was thinking of combining techniques. Best case scenario in USA is something like Owens Valley: 14,000 at its peak, but 4000 feet at the bottom of the valley.
Leading to ~10,000 feet (or nearly 2-miles) of elevation change. From there, you can dig another mile underground, leading to 1-mile (under ground), or -1000 feet elevation, to a peak elevation of 14,000.
If a tower were built on the top of the mountain: you could gain another 2000 feet or so on top: so maybe 16,000 (a 2000 foot tower on top of the mountain peak) to -1000ft (1-mile deep from the bottom of the 4000-ft elevation valley), for a total differential of 17,000 feet.
Ignoring earthquakes and other issues, of course. :-) Just purely from a hypothetical perspective: working with nature and the natural landscape seems like it'd be better than "just" digging a hole.
EDIT: Repurposing abandoned mine shafts might be worthwhile, depending how deep they are.
Gravitational potential energy is (approximately) linear in height. I say approximately because this assumes constant g (which is a good assumption when h is small compared to the radius of the earth, which it is).
And in fact, a consequence of gravity's 1/r^2 nature is that one is only subject to gravitational acceleration from what is beneath them (shells above cancel out), so mine shafts are less efficient than towers (the effect size is small to the depths we can mine).
So adding more height doesn't help, and if that height is underground it could actually hurt net efficiency.
Efficiency in this context refers potential energy stored per unit height. The field is conservative no matter what you build.
I'm assuming that the quadratic parent refers to comes from the increased potential energy per kg of material multiplied by the ability to store more material due to the additional volume.
If you picture a dense weight like a cannon ball on the end of a string you're right, but if you're digging down n meters, encasing n/2 meters worth of dirt and moving it up and down the free n/2 meters of shaft, the energy storage would indeed be proportional to n^2.
I don't know anything about the field and had the same reaction you did, but considering parent is running a startup in it they're either a lunatic that doesn't know the equivalent of FizzBuzz or there's something we missed on first inspection, and we should charitably assume the latter...
I read the whole pitch deck and it seems like you've thought it through well. I'm curious about the potential for failure; if one of the last links freezes up somehow (i.e. the disengagement you mention elsewhere fails), is there a process for clearing it?
The single shaft vs multiple parallel approach does seem a bit risky in the early days. If there's a 10% failure rate, and you built one shaft, that's a 10% chance of an existential threat to the company. 10 shorter shafts mean one will likely be inoperable.
Of course over the long term worrying about this doesn't make sense. Once you've scaled, 1k large vs 10k small shafts would not matter from this perspective.
I don't think it's fair to call that "scales quadratic". Certainly the energy stored in the total number of blocks is growing quadratic by digging deeper, but it also ignores that costs and effort grow when digging deeper. In effect, adding a block at the bottom comes with substantially higher effort than adding a block right next to it on the same level.
The issue here is that not everyone has that landscape (and as we know, doesn't want to destroy the view). I grew up in the US midwest--no hills for miles on which to build railcars, pump hydro etc.
High voltage power lines already exist. They are part of the landscape, and connect large parts of the country together. Case in point: a solar panel is Missouri could be providing power to Iowa.
Bonus points: if you want to build a big battery in Iowa, you can take advantage of excess power in Iowa or Missouri. Building tiny batteries here and there will only lower your efficiency.
There's a reason why a lot of discussion here is on large 100s of MW proposals: because anything smaller won't really be a big win economically. And it shouldn't be too hard to fill up 100s of MW capacity because of the nature of our large and reliable power grids here in the USA.
That's pretty clever! I would guess that wind turbine towers are mainly loaded in bending, and an extra compression load wouldn't hurt much as long as it doesn't cause buckling.
Yeah, this would be similar to what energyvault.com is doing. I love that energy vault got funding ($110M) and is pushing the industry forward. But IMHO, I ultimately don't think they will be able to compete with underground gravity storage because of the massive 10x advantage you get from digging underground and getting 10x more height (see my other comments below why that is.)
This idea assumes that the interior of the tower is empty space that could be used by the moving weight. Towers have an elevator for maintenance personnel that would reduce the available space. I'm guessing that the towers aren't high enough to store enough energy to justify the cost of the extra equipment.
I doubt that suspending the weight is the main structural issue. It's the high center of mass when the weight is at the top.
Yup, internal space is a very valid issue, given that cheap weights are probably big. I'd worry about oscillations for a high-up, suspended weight too.
Regarding justifying the cost--obviously depends what the cost is. If it's really just the generator, cabling etc, it might work as a short term load balancer. I saw somewhere about GE investing $X millions to smooth turbine output over 5min intervals.
In exchange for nearly doubling the weight of the average 3MW tower, adding enormous complexity, you'd be able to boost power output to 3.02 MW for one hour.
That gets us back to the fundamental problem with gravity storage - it's all mgh. You want to store a lot of energy, you better have a shit-ton of either m, g, or h, and one of those numbers is already tough to change :)
I love Gravitricity. I'm the founder of a similar startup called Terrament. We are also building gravity storage underground, but our patents are extending this idea to use autonomous, modular weights. This enables us to maximize both height and weight, which is the simple recipe for cheap gravitational energy storage.
We are also working on a seed-round of investment. It's an exciting field with plenty of room for competition. And it's so important for fighting climate change! We need to build this asap.
https://www.terramenthq.com/
Terrament is patenting a number of different designs which enable us to maximize both height and weight to nearly 100% of a mine shaft. When you think about the physics of suspending thousands of tons inside a mile deep hole, the devil is in the details. To paraphrase Boromir, One does not simply hang a mile of concrete and steel from a cable ;) There are already some videos on the website that give away some details. We're waiting to secure funding before sharing the rest.
edit: oh sorry, I think you meant gravitricity sorry. I'm not sure what they're patenting. Also I don't think their patents are in the US.
You would need to supply a lot of power just to maintain a vacuum and levitating your flywheel. Not sure this solution would be very effective. Maybe with superconductors, but then cooling is your problem instead.
A basic thing to understand about electricity storage is that there are very different needs for different kinds of storage.
Flywheels from what I understand are for very shortterm storage needs, i.e. balancing out shortterm fluctuations in electricity use vs. generation. But they're unsuitable for any kind of longterm storage, because they loose power over time.
In the long run with a high-solar-high-wind-scenario we'll need some seasonal storage to get us over a couple of weeks in some circumstances. This will need some storage that doesn't loose power over time.
(FWIW I have no idea if these gravitational storage techs will play any role in that, and one can be doubtful about it.)
Apparently not that easily or the half-life would be longer than two hours... You want to try if storing energy in a flywheel while using a lot of energy for the magnetic field can outperform gravity storage?
How do you and Gravitricity deal with ground water? Wouldn't you require constant pumping to keep water out, using additional energy? Or can they be sealed that well to work for decades without maintenance?
But we're talking about energy storage close to consumers, not in remote locations I thought? For deployment in Europe, I don't think it'd be possible to find areas where ground water is that deep (luckily).
And pumped storage still doesn't work as you'd need surface area for new lakes. Also, not sure if you'd want to pump out huge amounts of ground water for pumped storage, lowering the water table for everyone.
My civil engineer partner could speak to this in more detail, but it's a well-understood solvable problem. You seal the shaft wall as you dig past the water table. And some small amount of pumping is used as needed.
1. So the units are actually modular, i.e., not connected. Aha. If a unit jams, it's annoying but not fatal, as all the units above it continue to be usable.
2. Since the weight units are separate, then the force on the gearing (both wall and unit) is constant and thus the wear and tear is manageable.
3. The max energy stored is when all the weights are horizontal, of course. So there needs to be a transmission line built into the shaft, along with a motor/generator per weight unit. Some cleverness needed so the transmitted power is passed to (and received from) the correct units. The sliding contact where power is transmitted would also be a major possible point of failure.
4. Yeah, that's the downside of separate modular units. Each unit now needs to transmit/receive power.
5. It would seem prudent to build two or more shafts adjacent to each other. So if there is a problem in one shaft, you can go down the other and fix it. Increases reliability significantly.
6. If the units and power transmission are waterproof, water in the shaft isn't too much of an issue. You might lose ~20% of energy storage due to the buoyancy of the water, but that's it. Might save energy not having to constantly pump out the shaft.
7. Right. Definitely need a failsafe mechanism so a unit doesn't plunge down (say, when the motor/generator clutch fails) and take out the units at the bottom.
- Regarding jams, failures, etc. These should be extremely rare, but in the event, each unit is designed to "disengage" if needed so it's just dumb weight, and modules adjacent to it will be designed to lift it. Then when the train is above ground, the unit can be swapped out for another one and repaired while the system keeps operating.
- Yup failsafe mechanisms will be built in as well.
- Yeah each unit has its own motor/generator. A power line will enter through the module's axle.
Both solutions present heavy machinery, presumably aimed at big industrial installations.
How about home use? Could it be used small-scale, e.g. as an alternative to Tesla's PowerPack? I've long been curious about distributed energy solutions, from heating (e.g. cogeneration / CHP, heat pumps, geothermal energy, solar+battery for off-grid energy), could this be a low-cost, low-tech, low-maintenance, low-risk alternative to batteries?
At small scales, the setup costs world dwarf those of lithium by an order of magnitude. You can have a powerwall shipped to you and a local electrician install it. This, you’d have to have an engineering crew dig a shaft, unless of course you happen to own an old mine.
With nuclear you still need storage to flatten the peaks and troughs though right?
Also this does not seem very convoluted, it's using an electric engine pretty directly. If you are actually boring straight down it seems like the failure modes are pretty okay.
> With nuclear you still need storage to flatten the peaks and troughs though right?
Why? Just build more. Keep expanding capacity so the base load covers the peaks and thensome. Push the price of electricity down while keeping it carbon-neutral and crush competition. Get everyone 2c/kwh power. Wouldn't that be more fun?
The problem is the speed at which the reactors can be ramped up and down, this is very limited. Also, if you have enough capacity to cover all peaks, most of the time the reactors are not running at full power. That increases the cost of operating the reactors, which is far higher than you claim. Currently, projects building new nuclear reactors are challenged by their costs. Never mention the safety concerns and of course the waste problem.
So print more money, what does it even matter anymore in the land of MMT?
> the waste problem.
Ehh, it's well managed in Canada and France, I don't see why Americans can't handle it. Just keep it away from fault lines, right? I bet some of those old missile silos would be perfect.
Like every sane country, they reuse it - only the final, hard-to-use fractions are stored, which takes a relatively small hall (pretty sure typical US football stadium is bigger).
And those hard-to-use fractions aren't going to be a problem for long, pretty sure they are considerably shorter-time risk than the timeframe for cleanup of WW1 battlefields in France (which currently stands at around 700 years).
Current versions of the traditional LWR design can ramp at about 5% of full power per minute, which is about the same ramping rate as a CCGT plant. With steam bypass it's possible to ramp even faster.
Demand response, frequently promoted as a way to increase penetration of intermittent renewables, can also be used to reduce the cost of a system composed on high capital cost, low marginal cost dispatchable generators like nuclear. Charge the EV's and run heat pumps to warm thermal storages during the night when demand is lower, say.
Yes, demand response will play a huge role in the grids of the future. But when you say "current versions", which current operating nuclear plant achieves this ramp speed?
Also, it means you have to run your reactors regularly at below 100%, so you still can react on additional demand, which increases further the cost of a very expensive technology.
> But when you say "current versions", which current operating nuclear plant achieves this ramp speed?
IIRC that 5% figure I read was wrt EPR and AP1000, presumably older generation LWR's are slower, by how much I'm not sure. France has run older generation PWR's in load-following mode for decades, but I'm not sure which ramp speeds they achieve; fast enough in practice in any case it seems. CANDU reactors in Canada have steam bypass and can apparently ramp at >10%/min.
In any case, my point is that ramp speed is in practice not a technical limitation. Of course you want to run a generator with high capital cost but very low marginal cost at 100% as much as possible, but if you now and then need to ramp (say, if the wholesale price goes negative) you can do it.
> Also, it means you have to run your reactors regularly at below 100%, so you still can react on additional demand, which increases further the cost of a very expensive technology.
To be clear, I'm not advocating a 100% nuclear grid. I'm just pointing out that the "nuclear can't ramp and is thus unsuited for the grid of the future" isn't correct. In particular, I think solar and a moderate amount of storage is very well suited to cover the daily variation in many parts of the world. I also think that dispatchable low-carbon sources (which could be hydro, or nuclear, or something else like geothermal where available, CCS where geological formations for storing CO2 are available, etc.) have a role to play in least-cost deep decarbonized grids. See e.g. https://doi.org/10.1016/j.joule.2018.08.006
From my understanding some of the "delay" in ramping up is related to enrichment level of the nuclear fuel and thus how fast the speed of the reaction can change.
Kind of. When reducing power there's a buildup of a particular Xenon isotope, which is a powerful neutron adsorber. So until that isotope decays away sufficiently you might have problems starting the reactor back up again. Unless you have enough excess reactivity, such as by having fresher fuel loaded.
AFAIU France primarily uses reactors which are earlier in their fuel cycle for load balancing, and ones which are near the end run with a flatter profile.
Didn't know about specific as it's not really my area, but would running a PWR with very high enrichment level (AFAIK some designs use 93% U-235) allow quick spin down and spin up?
> would running a PWR with very high enrichment level (AFAIK some designs use 93% U-235) allow quick spin down and spin up?
Well.. there are a lot of factors in a reactor design affecting the ability to increase or decrease power quickly. Geometry, fuel/moderator ratio, fuel density, burnable poisons (for flattening the reactivity swing over the fuel cycle), amount of control rods etc etc. Fuel enrichment being only one thing, which in turn affects other things as well (e.g. reactors using highly enriched uranium tend to use different fuel designs than low enriched fuels).
But yes, military reactors for navy ships obviously have very different demands on them than civilian power reactors, and are designed accordingly.
And yes, while US navy reactors use 93% enriched fuel, it's not necessary, e.g. French and apparently Chinese submarines use low enriched fuel (7% for French).
A lot of the costs is politics, and second-order effects from lack of production lines making everything a one-off.
Which "funnily" enough makes it so that the more dangerous reactors remain in use for longer, as the new designs aren't built, despite huge safety improvements (even if you go just for late 1970's self-sealing molten lead/bismuth reactors).
Overcapacity is a problem, especially for low-enriched reactors, but that's why it shouldn't be the only answer.
Are you joking? Building nuclear power to cover peaks is insanely uneconomical. More expensive than building a nuclear power plant in a flexible grid.
If you don't know anything about how a grid functions then maybe you should stop shoehorning a single solution to solve every problem when everyone is already aware that a large array of different technologies is necessary for a modern grid.
If you have more production than demand you need to store it somewhere, which is exactly the point of such projects, if only to prevent network desynchronisation that causes brown/blackouts due to power flowing in different way than you want in the grid.
Energy storage has also been planned for nuclear power plants as starter and emergency power.
even if nuclear had no contamination horrors, this is no longer true. Hydrino energy is much more energy dense and will be the ubiquitous portable power source for the 21st century and beyond.
My comment might be read as 'Nothing will EVER compare to nuclear'.
While in fact what I meant was 'CURRENTLY nothing scientifically viable compares to nuclear.'
I do have high hopes about fusion and molten salt thorium is also up there as something I'd like to be further explored.
But currently nothing up and coming seems to compare to nuclear.
I'm not an expert, but reading those criticisms makes me think this is nothing more than a money grab scheme without delivering anything usable for the past 30 years.
If that weight were a fission reactor you wouldn't have to move it up and down. And it would be 100% efficient. And you wouldn't need an external power plant.
The nuclear debate is interesting, with valid talking points on both sides. But even if nuclear were safer and cheaper than wind+solar+storage (which it isn't), it would take decades to build. We might not have that much time.
Climate change is urgent and we need all hands on deck to build as fast and as cheap as possible. May the best designs win asap in this fight!
Some of the known SMR designs are amenable for mass production, if someone would kickstart the line by putting money on the table. Very safe ones at that.
A lot of the cost goes down when the reactor isn't a practically one-off build, and when you can for example use prefabricated components to "assemble" a power plant quickly.
Some designs go even further, and have power blocks that are essentially something you slap on large railcar, including option that instead of refueling you send back the module while the vendor sends you a freshly-fueled one.
If you dig a well with cross-sectional area A and depth D, the amount of energy you can store with compressed air at 10 MPa is 10AD MJ if A and D are expressed in m^2 and m respectively. [1]
If you drop a steel weight down the same well of cross-section A and height H in the same units the energy stored is 0.08HA(D-H) MJ[2]. Again, H is in meters, and AD > A(D-H), and in order to beat compressed air 0.08H > 10, so your steel weight needs to be 10/0.08 = 125 meters long! That's taller than most of the buildings in downtown San Francisco -- and in order to get any use out of this thing, your hole should be at least twice that deep.
Of course, compressed air has its inefficiencies and complexity, but the feasibility of a metal rod even close to that long seems pretty low to me. Compressed-air caverns use as much as 7.5 MPa, but a purpose-built well could potentially go much higher. Plus you don't have to deal with the damn thing vibrating from Coriolis forces and seismicity.
Now, I know what you're saying -- you're saying, if you're so smart, why don't you do it? -- but there are simply too many huge caverns out there to even think about constructing CAES chambers. There are several GW in service today. And even with that huge resource people wonder if batteries won't simply corner the market. Storage is getting here painfully slow, it seems like, but the competition is very fierce.
1: True isothermal decompression cycles are impossible, so of course I'm approximating by using the ideal gas law.
Yeah, compressed air is a very viable and promising solution. It does have efficiency trade-offs as you mention though. And typical compressed air designs still use fossil fuels in the compression process. There are some new designs I've seen that get around using fossil fuels.
One of the most prominent new startups out there exploring advanced compressed air is hydrostor. Note that they also dig underground :)
https://www.hydrostor.ca/technology/
Sorry for repeating myself, but gravity is weak. If you hang a 500t weight in a 150m vertical tunnel, it only holds 500 * 1000 * 9.8 * 150 = 735000000 J, or 735 MJ, that is, 204 kWh.
A Tesla model 3, basic model (MSRP ~$38k) has battery capacity of 50 kWh, so we're talking about four Tesla 3's.
I don't think digging a 150m hole is cheaper than four Teslas - and Teslas come with the rest of the car you can use for driving.
Gravity is weak, but it's cheap. It is indeed surprising, but plenty of research shows that the levelized cost of gravity storage is cheaper than Li-ion batteries - even if you assume that Li-ion will get 4x cheaper in the next 20 years. There are other problems with Li-ion as well that will prevent it from scaling up to the amount of storage that our grid will need. I discuss all this with lots of citations in a white paper here https://github.com/syllable-hq/uphs-feasibility-study
No worries. Here is a good resource that talks about the levelized costs of gravity storage. It's estimated to be much cheaper than Li-ion battery storage on large scales.
The primary benefit is reliability and lifetime. Your lifted rock isn't going to "malfunction" and suddenly weigh less.
I agree that gravity based energy technologies are mostly junk though. There are two important factors: cheap weights and unlimited scalability. Doing something "cute" such as using a mountain side or old mineshaft is reducing the scalability.
Isn't pumped water storage the same thing but much easier to manage? Like think of a 10 ton weight suspended from cables 100m high vs a $300 above ground pool on a 300 foot hill. Isn't pumped water storage going to be hundreds of times cheaper for amount of energy stored?
>"Our patented technology is based on a simple principle: raising and lowering a heavy weight to store and release energy. The Gravitricity system suspends weights of 500 - 5000 tonnes in a deep shaft by a number of cables, each of which is engaged with a winch capable of lifting its share of the weight. Electrical power is then absorbed or generated by raising or lowering the weight."
Idea: Could elevators be retrofitted with something like this?
Then they could use/store energy only while going up, and regenerate some of it going down... and take passengers to various floors while doing that!
"The Regenerative Elevator"!
Stores energy going up, regenerates some of it while going down!
Invented here on Hacker News, by yours truly, 9/8/2020!
(Yes, I know, it's a stupid related idea! <g>. But my other related idea was more stupid, and that one was to fill up a U-Haul truck with trash, put a steel cable on it, find a hill, and use the steel cable (in conjunction with the overweighted U-Haul truck and hill!) to drive a motor/generator/winch assembly that uses electricity going up the hill, and regenerates some of it back, going down the hill... <g>)
On a serious note however (for non-passenger elevator energy storage), I think Gravitricity has a good idea, and I wish them much success with it!
> Idea: Could elevators be retrofitted with something like this?
I think it would not work so well because they have counterweights to make the up-down operation much easier. If you remove the counterweight, it would store energy, but be prohibitively slow when going up and probably frighteningly fast going down.
Regarding the uhaul, this is not too far off from other people mentioning driving a train uphill and letting it fall downhill to generate electricity. Interesting ideas.
I'm not sure if you are joking or not, but regenerative elevators exist, and have existed for over 100 years:
> DC-driven winding-drum elevators—the leading design until the 1930s—use a DC motor in the basement that winds and unwinds the elevator’s steel cable on a steel drum, thus lifting and lowering the car from pulleys atop the elevator shaft. DC drive was the only way to go at the time for a speedy elevator, because only DC could deliver variable-speed operation for smooth starts and stops. The DC motors were also energy efficient, capable of something that has only recently become possible with modern elevator designs: regenerating power when the elevator descends.
Quite a few comments question the cost of digging -- which is understandable. But the context to remember is that these are huge infrastructure investments that will return revenue from that investment over 20-100 years with little maintenance costs. The excavated shaft is like a factory not a product. The product is each cycle of energy generation giving you a profit day after day. The shaft is a one time investment that will yield that product forever.
There's lots of research showing how the costs work out (see my other comments here)
Also, hasn't digging become much cheaper in recent years with new technologies from the oil industry? I'm sure some of that efficiency gain can be used for those shafts (they're obviously bigger than a bore hole for oil).
Pump air into a giant pressure chamber...that's also a super-deep hole with a pulley system...and the weight for the pulley is a flywheel sealed in a vacuum chamber, magnetically levitating to avoid any friction losses. Oh, and the mass for the flywheel? A bunch of batteries.
Not entirely joking. I just saw the big hole and immediately thought it'd be a great place to store pressurized air, and then I continued thinking of other energy storage mechanisms you could cram in there.
This will work, obviously. But I don't see how it can be competitive with batteries.
High energy density lithium batteries are currently being commoditized. They are also increasingly able to handle many thousands of cycles. E.g LiFePo4 cells.
The raw materials for batteries are not actually that expensive or rare, so that leaves the manufacturing.
You might think that making something as complex as a battery can never be as cheap as hanging a weight from a rope. But there are examples of very complex products (solar cells, LCD displays) that became incredibly cheap due to mass manufacturing.
A square meter of solar cells, requiring extremely pure silicon and nanometer scale engineering, is now not much more expensive than a square meter of good roofing shingles.
Why would this be better than a set of railway lines down a hill side into a forest pulling up standard goods carriages full of rocks?
You get a forest (which buffers runaway carriages and does all the other lovely forest things) and energy storage on the hill. All using largely commodity items and you can build it anywhere there's a spare slope.
I don't get how any of these gravitational energy storage startups get past the napkin phase. There is no competition with pumped hydro. Who cares if weights on winches are marginally more efficient when you have many orders of magnitude more mass to move.
This site claims a max weight of 4.5e6 kg. Lake Powell, for example, has a max capacity of 3e13 kg!
So, SEVEN orders of magnitude less mass to move. Sounds competitive.
Pumped hydro is great but unrealistic in many countries. You need hills or mountains with space for new lakes. All obvious places already have them, for the others the ecological impact often is quite significant.
All obvious places may have hydro plants already, but not all hydro plants are reversible.
The country point is interesting, so the market is countries lacking topography suited to hydro plants? How big is that market really?
And considering that pumped hydro is currently the only viable grid scale energy storage mechanism (inb4 but li-on), those countries currently either 1) share a grid with neighboring countries that do have pumped hydro, or 2) forego energy storage altogether and use peaker plants like everyone else. If you do 1 you better trust your neighbors to never go to war with you, otherwise you better have 2.
So the market is: countries lacking topography suited to hydro plants, who currently share energy storage with neighbors who they don't trust, and who for some reason aren't just building peaker plants (which are essentially just normal fossil fuel power plants) and are holding out for a non-extant technology to be developed.
I'm not much of a capitalist myself, but even I can tell that that's a bad investment.
Power to gas is promising. Just generate hydrogen or methane and feed it into existing gas pipelines. Storage infrastructure for gas already exists and has several TWh of capacity.
Assuming that pretty much everyone is planning on eventually phasing out peaker plants for environmental reasons: I'd say that the market is pretty huge if you can develop viable technology to service it.
> All obvious places may have hydro plants already, but not all hydro plants are reversible.
If a hydro site is convertible to reversible, chances are it’s been already. Most sites are not really suitable because regular hydro plants don’t need a lower reservoir or water source and so can be built in locations which don’t have one.
Meanwhile pumped hydro can be built in locations where normal hydro makes no sense because there is no natural downflow or upper reservoir e.g. Taum Sauk.
I may be wrong, but there is a limited capability for doing pumped hydro - you may run out of elevated lakes in a region, and if you want to create new ones then you will affect the environment.
A hole in the ground can be dug up everywhere, e.g. next to each single PV plant.
With seven orders of magnitude of difference, you'd have to dig a whole lotta holes before hydro has to start worrying about competition on the capacity level.
Compared to energy loss over distance, which you don't really have if you dig your hole right next to you, it's more like 6 orders of magnitude, maybe less.
But maybe capacity is only part of the story. What about locality? Distributed energy storage may not be as efficient, but it has a lot of other benefits.
Many of these projects also falsely assume that there are enough abandoned mineshafts or other types of holes to convert into a battery storage system. They have to do this because you would see through the scam immediately. Digging a wide and deep hole is expensive and not economic unless there is something valuable at the bottom of the hole.
There is also the fact that weights are expensive relative to the energy they can store so you have to take them directly from nature (usually via dams or pumping water uphill).
Right, the thing to worry about is power, not energy. Ideally the pumps would be sized to fill/drain lake Powell in 12 hours. Which is ludicrous. The turbines in the Glenn Canyon Dam have a max (output) capacity of 890m^3/s [0]. It would take just over a year to drain the reservoir through them.
Which is fine, as many regions have a strong seasonal availability of renewables. Storing energy for some months will be required to go fully renewable. There are also short term storage requirements, over the course of a day or a week.
You are looking at the wrong metrics. Capacities and efficiencies are interesting from a technical point of view. But the only metric that really matters is $/kwh of storage. And depending on where you are pumped hydro can be stupendously expensive and disruptive to the local environment or require expensive infrastructure to be put in place. Even just having access to water is a non trivial thing in a lot of places (e.g. deserts).
In some places pumped hydro does make sense of course but you have to compare basically on $/kwh what is the best solution taking into account the local environment. There are no silver bullets here. Which is why a lot of places seem to be ending up with lithium ion batteries, because they are there and can be put in place without a lot of fuss or rearranging of the local environment.
This an overly narrow view of the problem. The footprint of pumped hydro is enormous and you can't just put it anywhere where you actually need the power.
The footprint of anything that stores enormous amounts of energy is enormous. (If you think it isn't - batteries, say - it's because you've ignored the socking great mine needed to extract the materials, the factories needed to make them, etc.). And we don't need it to be that near - we have pylons.
This seems to be one of the key indicators of green pseudoengineering - an obsession with avoiding transporting power. Pylons work. Really. If you can get large amounts of RELIABLE renewable power/storage, with low labour and machine costs, from a large amount of cheap land, land far away from people or crops or even trees, you do not have to worry about the rest. Worry about how to get those labour and machine costs down.
Okay, let's just say the footprint of something that stores energy vertically is going to be enormously lower. For what it's worth, this "Graviticity" thing may be utter nonsense, that's besides the point.
Pylons may work, but you need to pay off everyone owning the land those lines go across. In many areas, they can't build pylons at all because of rampant NIMBYism.
The power density for gravity storage such as this is directly proportional to the density of the storage medium.
Water is 1 tonne per cubic meter. Steel is 7.9 tonnes per cubic meter. Lead is 11.34 tonnes per cubic meter - which means that in the same amount of space, you can store vastly more energy if you use these metals instead of water.
The initial cost of set up will also vary depending on the metal you use but lead is cheap.
Cost / density. You quite literally want the cheapest dead weight.
Cost of lead/tonne : USD 1900 (approx)
Cost of Uranium/tonne: At least USD 60000 (approx regular uranium) even if you are able to source it.. and then you still have to build containment for this, get permits etc..
For that price, you could get much higher efficiencies if you went for a cheaper metal - even something like Tungsten which is extremely dense will be much cheaper.
To me it smells like https://www.youtube.com/watch?v=uzV_uzSTCTM and all about patents, marketing and money burning, but I don't know as much as these professors do so maybe I am wrong.
The concept is quite obvious and has been proposed several times in the past. The key question is, what are the costs? Even if the costs for drilling the hole are written off over a very long time span, the wires that hold the weight and the machinery have constant operation costs.
It probably can't compete with pumped storage, but pumped storage isn't viable without at least some hills. This concept could be deployed in many regions, also wouldn't take much surface space, you could deploy it even in densely populated regions. The question is: how does it compete with e.g. batteries?
Does anyone know if something like this has been implemented in some way already? In old mine shafts maybe? The geology has been studied, and some shafts at least there already...
My father has an idea to use sodium (Na) as fuel for fuel cells. This could easily replace batteries by having more or less instant reactions to demand, being much more energy dense and being simple to handle. Also, there is no CO_2 that would need to be captured compared to "bio" fuels. The resulting sodium hydroxide (NaOH) and hydrogen (H_2) can easily be used further e.g. in the chemical industry or recycled, electrolysis of NaOH is well known and also produces hydrogen as a byproduct. The resulting sodium metal can again be used as fuel. It creates a circular economy. Handling sodium securely at scale is also probably even easier than handling gasoline or diesel. Most of these reactions at industrial scale are more than 50 years old but nobody bothered to actually implement it (even though a hint about sodium cells for electricity generation is present in "Twenty Thousand Leagues Under the Seas: A World Tour Underwater" by Jules Verne already) instead of the more complex approaches like e.g. synthetic gasoline.
You can look at the orgpage about these ideas https://orgpad.com/s/energiewende my father also gave a talk last week about it, there is a recording, which will be posted during the next days.
Disclaimer: I work for the small startup OrgPad, which tries to create a tool for easier decomposition of linear ideas/ content into a network of ideas/ content. An ex-Googler, Pavel Klavík PhD. describes the technology (hint Clojure and ClojureScript) and approaches behind OrgPad in a recent talk https://www.youtube.com/watch?v=4UoIfeb31UU
Many years ago I briefly had a thought experiment about converting a gasoline/oxygen combustion engine into a sodium/water combustion engine. Solid fuel mixing concerns aren’t the primary issue. The primary issue is the energy density is just so danged low compared to gasoline. However, compared to electric batteries, sodium is far more energy dense.
Well, there are patents for running diesel engines with sodium, at least I was told so much by my father who researched the problem thoroughly.
The density isn't that low considering using a fuel cell, you can convert the energy a lot more efficiently. Also you don't just burn up the fuel into the air as currently, there is no industrial process to collect the CO_2 and other gases produced from the exhaust.
You could theoretically store energy by winding springs and making use of it by letting them unwind (by spinning a generator or whatever). Toy-car style.
This doesn't involve digging a deep hole in the ground and hope that the earth keeps it level.
A drawback would be the hazards that heavy tension/forces bring with them.
287 comments
[ 2.4 ms ] story [ 246 ms ] threadIt may seem strange to use energy density in this context but in many ways those are even worse than batteries due to the massive amount of work they require and poor scalability because of all the physical infrastructure and space.
In fact, this exact scheme looks to have been posted to HN in 2014. From my understanding they are looking to re-purpose holes that have already been dug, for the most part, which is a good idea, but if the planned prototypes are only doing 250 kW nameplate with maybe 100 KWH of stored power, there is almost no way it's going to beat a battery plant you can have more or less just built to order already commercially. And said battery plants aren't even remotely good enough at purpose for meaningful energy storage, they're mostly used for what are considered "ancillary services" in the US power markets, like eating or producing reactive power, voltage support, etc, because the modern grid scale inverters can react to grid conditions faster than grid frequencies (50-60 Hz). More rarely they might be scaled big enough to shave the peak off a demand curve for 5-10 minutes to save transmission capacity or congestion.
Keep in mind all this in happening when LNG is basically free except for the cost of moving it, so you can also put in 1-60 MW combined cycle gas plants in short amounts of time with very well understood technology and proven manufacturers.
The McIntosh CAES plant began in 1991 and from my understanding, has been widely viewed as successful. https://www.smithsonianmag.com/innovation/salt-power-plant-m...
That's 110 MWs for 26 hours. Probably not 2.8 GW-hrs of energy storage (they can't sustain 110MWs for all 26 hours), but this plant is probably GW-hr range. I'm pretty sure all of Li-Ion right now is less than 2GW-hrs for the entirety of the USA.
---------
The 290 MW plant in Huntorf Germany (1978) also is successful.
--------
There's a 2nd, 300MW+ plant (that's 300MW power for multiple hours. So near GW-hr scale as well). This is the more recent Apex Bethel Energy Center you were badmouthing earlier. Its not fully built yet, but given the history of CAES, I'm optimistic.
--------
There aren't many CAES plants in the world. But from what I can tell, they are all safe and successful. The one issue is that you need to reheat the air as it leaves the caverns: compressing it underground reduces its temperature. A bit of natural gas is used in this heating process, but not nearly as much as an actual natural-gas plant.
------------
The "Rail up a hill" project is 50MWs with 12.5 MWhr storage capacity: https://s3.amazonaws.com/siteninja/multitenant/assets/21126/...
50MWs is somewhat small, but there's something to be said about the simplicity and ease-of-deployment of rail energy storage. Those 50MWs were spec'd out with only 7 trains. It isn't too hard to imagine scaling the system up to support more than 7 trains.
I'd love to see people running more of these setups. Especially in places where DC locomotives are cheap (former CIS countries), this could be even more cost effective.
Pumped Hydro energy storage is the #1 energy storage in America, and probably the world. A single plant provides 24GW-hrs of energy storage, dozens of more energy storage than all American utility scale Li-Ion batteries combined.
Bath County is the biggest pumped-hydro battery in the USA (and probably the world), but there are dozens of pumped-hydro plants across the country.
------------
Turns out that all the water in a lake can hold a huge amount of potential energy through gravity: just pump it up a mountain.
Actual mechanical pumping.
Free as in money, costly in consequences. The whole point of energy storage schemes is to allow us to migrate away from fossil fueled energy.
Efficiency would be low, at about 70% * 60% ~= 40% but capacity would be huge. 1 kg of hydrogen has specific energy of over 140 MW.
We could even capture CO2 as well with that energy and make methan and use normal gas powerplants for retrieving the power :)
Just build double the number of solar panels to account for efficiency loses. Should be cheaper than grid-scale energy storage with 90%+ efficiency.
Pumped is great when you have the geography for it but most countries don't. And building dams is pretty harsh on the environment too, just in different ways than CO2 emmisions.
Germany has enough gas well storage to survive several months. This infrastructure is currently used for natural gas, but it can be repurposed for hydrogen.
I still think that commoditized battery storage will win in the end, but storage cost is not an argument against hydrogen.
For small and medium scale storage you are right. Which is why hydrogen for cars and even trucks is a bad idea.
Hydrogen at 0°C and 100 Bar (~10MPa) has a density of 8.3447 kg/m^3. It has an energy density of 120 MJ/kg. So about 33 kWh/kg. So you end up with 278 kWh/m^3.
Now of course you have to multiply this by the H2->Electricity efficiency. Let's be very pessimistic and take 0.6 or 60%, which is what a gas turbine plant can achieve today.
You end up with 166 kWh/m^3 of usable electricity per cubic meter.
One of many german gas storage facilities https://www.nafta-speicher.de/en/company has a volume of 1.8e9 m^3. That translates into 300 Terawatt-Hours of storage capacity for just this one facility.
It is not a limitation of the large storage facilities, which are just exhausted natural gas wells and work just fine with hydrogen.
The cost of outfitting one such facility with steel pipes that are not subject to hydrogen embrittlement is trivial compared to the cost for the electrolysis units etc.
Note that I am not a big fan of hydrogen at all. But large scale storage is one of its few redeeming qualities.
IIRC there was some study on doing something like that in Germany in the 2000s but I'm not finding references to it now. If you don't have to build the dam (since it already exists), pumping water back up into the reservoir is going to be better in the limited locations you can do that, and given the relatively tiny amount of energy storage vs power production I suspect the economies of scale on that won't change without a preceding sea change in global commitment to fossil fuel retirement.
There is also a better transition path to that as we already are building amazingly good gas turbines that could just as well be burning synthesized or biogas on demand from solar or nuclear over-generation. I sort of suspect that may end up being politically poisonous to the people that would normally be pushing for carbon-neutral storage options.
It would be way more than double the solar panels though. Solar generation is generally maxing at around 25% capacity factor, so replacing an existing fossil fuel generator requires 4x or more the nominal capacity in solar plus whatever efficiency loss in storage, so going with a pure solar + storage replacement of a 1500 MW combustion plant could easily require 12,0000 MW of equivalent generation.
We're comparing energy storage so no matter how much fossil fuel powerplants we replace - it would be x solar panels for energy storage with 80-90% efficiency (for example this gravicity stuff or hydro or batteries) and 2x for methane synthesis.
Reusing abandoned mine shafts could have an advantage. But there is also ample researching showing that the cost of digging from scratch is still worth it. The US Dept of energy even studied it back in the 1980s when they were evaluating underground pumped hydro energy storage. I wrote a white paper on it that's linked from here. see here https://www.terramenthq.com/uphs/
The key thing here is that adding 10x height gives you 10x more PE per weight without extra digging costs (or tower costs assuming the cost per unit of height is constant)
Here's how to think of it: by digging one deep hole instead of many shallow holes (with the same amount of digging) you get (n^2 / 2) instead of (n/2) Potential energy. The average height of all the modules is half the height of the shaft. With one deep shaft, the modules pass through the same excavated volume so they can all go deeper on average.
Edit: reviewed the Terraent slides they indeed plan to literally drive a mile long train vertically into the ground. In which case the quadratic argument stands.
At the cost of significant added complexity. It comes down to drilling cost vs installation/ maintenance cost.
But also, getting your quadratic gain on a mile deep hole requires a mile of weights to lower into it. That's a lot more above ground infrastructure than just the single shaft!
The way our design works, the modular weights are autonomous on a track. So we don't have huge cranes above ground. We just have a track above ground that also runs a mile. That track can be partly or fully buried if desired. It will be enclosed protecting it from weather, and solar can be installed on top to save real-estate.
But yes, these installations are enormous. They might cost around $150M and could provide about 200MW.
- Keeping water out of a hole is a solved problem.
- as discussed in other comments here, adding more height gives you more PE for the same amount of weight so it's a very valuable investment that pays profits over 20-100 years for your one time investment.
https://docs.google.com/presentation/d/17FI-jrI9RWS3q7Ng44Yh...
"We plan to roll out our technology in disused mine shafts worldwide."
AFAIK, the most common is done around hydro dams, by pumping water upstream as a form of energy storage. The infra is already there, but it's not as efficient as systems like Gravitricity. But it costs a negligible amount of money to "activate" energy storage in a pre-existing dam.
If it does work, I would imagine the Boring company would be all over this.
and prone to Earth quakes, I would assume.
The US Gov has studied plenty of research showing that underground pumped hydro is cost effective and obviates the need for dams. (See my white paper here: https://github.com/syllable-hq/uphs-feasibility-study)
So startups like Gravitricity (and Terrament, my startup) are innovating on what is already well-researched territory.
Obviously, this seems very technically challenging. 500 tonnes is 64 m^3 of iron. We will see if their engineering is good enough to pull off their claimed 171 US$/MWh.
https://www.metalary.com/tungsten-price/
2000 tonnes of tungsten is 60 million USD, of iron is 200k USD.
Lead goes for around $2k/ton, so this might be a more feasible compromise.
Civilian sales look at least possible, I heard of at least one sailboat with a DU keel.
But I assume there's structural issues with suspending a few extra tons on your structure.
ARES (rail energy storage) builds a rail-line uphill, for example. Rail cannot handle a very steep slope, but a gentle hill climb will build up potential energy fine.
In the case of vertical-based gravity storage, I'd imagine that lifting blocks to the top of a cliff (or down a valley) would be most efficient.
I mean, Pumped Hydro is gravity storage, and does just that. Pumping water up a mountain and generating energy by dropping it back down. But presumably, we don't want to use water in the Western states (where water is scarce). So Gravity-energy storage WITHOUT water is the goal.
Leading to ~10,000 feet (or nearly 2-miles) of elevation change. From there, you can dig another mile underground, leading to 1-mile (under ground), or -1000 feet elevation, to a peak elevation of 14,000.
If a tower were built on the top of the mountain: you could gain another 2000 feet or so on top: so maybe 16,000 (a 2000 foot tower on top of the mountain peak) to -1000ft (1-mile deep from the bottom of the 4000-ft elevation valley), for a total differential of 17,000 feet.
Ignoring earthquakes and other issues, of course. :-) Just purely from a hypothetical perspective: working with nature and the natural landscape seems like it'd be better than "just" digging a hole.
EDIT: Repurposing abandoned mine shafts might be worthwhile, depending how deep they are.
Gravitational potential energy is (approximately) linear in height. I say approximately because this assumes constant g (which is a good assumption when h is small compared to the radius of the earth, which it is).
And in fact, a consequence of gravity's 1/r^2 nature is that one is only subject to gravitational acceleration from what is beneath them (shells above cancel out), so mine shafts are less efficient than towers (the effect size is small to the depths we can mine).
So adding more height doesn't help, and if that height is underground it could actually hurt net efficiency.
Efficiency in this context refers potential energy stored per unit height. The field is conservative no matter what you build.
If you picture a dense weight like a cannon ball on the end of a string you're right, but if you're digging down n meters, encasing n/2 meters worth of dirt and moving it up and down the free n/2 meters of shaft, the energy storage would indeed be proportional to n^2.
I don't know anything about the field and had the same reaction you did, but considering parent is running a startup in it they're either a lunatic that doesn't know the equivalent of FizzBuzz or there's something we missed on first inspection, and we should charitably assume the latter...
The single shaft vs multiple parallel approach does seem a bit risky in the early days. If there's a 10% failure rate, and you built one shaft, that's a 10% chance of an existential threat to the company. 10 shorter shafts mean one will likely be inoperable.
Of course over the long term worrying about this doesn't make sense. Once you've scaled, 1k large vs 10k small shafts would not matter from this perspective.
Best of luck mate!
There's another thread on this comment page talking about why height is important. Please see this slide in our presentation illustrating it. https://docs.google.com/presentation/d/17FI-jrI9RWS3q7Ng44Yh...
There's hills somewhere in there, and those hills can form energy-storage solutions that can be transmitted across the entire MISO grid.
Bonus points: if you want to build a big battery in Iowa, you can take advantage of excess power in Iowa or Missouri. Building tiny batteries here and there will only lower your efficiency.
There's a reason why a lot of discussion here is on large 100s of MW proposals: because anything smaller won't really be a big win economically. And it shouldn't be too hard to fill up 100s of MW capacity because of the nature of our large and reliable power grids here in the USA.
For example, all the dam failures that have obliterated entire towns.
I'd rather a failure obliterated a bunch of earthworms and voles.
I expect that the voles disagree with me.
I doubt that suspending the weight is the main structural issue. It's the high center of mass when the weight is at the top.
Regarding justifying the cost--obviously depends what the cost is. If it's really just the generator, cabling etc, it might work as a short term load balancer. I saw somewhere about GE investing $X millions to smooth turbine output over 5min intervals.
Nah.
In exchange for nearly doubling the weight of the average 3MW tower, adding enormous complexity, you'd be able to boost power output to 3.02 MW for one hour.
That gets us back to the fundamental problem with gravity storage - it's all mgh. You want to store a lot of energy, you better have a shit-ton of either m, g, or h, and one of those numbers is already tough to change :)
We are also working on a seed-round of investment. It's an exciting field with plenty of room for competition. And it's so important for fighting climate change! We need to build this asap. https://www.terramenthq.com/
I note from the website: "Our patented technology is based on a simple principle ..."
I think it's a great idea but I don't see what there is to patent or what IP they could defend ...
edit: oh sorry, I think you meant gravitricity sorry. I'm not sure what they're patenting. Also I don't think their patents are in the US.
I know Pennsylvania and New York both have 20 MW storage systems that takes up a few acres and are relatively cheap per unit of storage.
A weight you can just hang there
A basic thing to understand about electricity storage is that there are very different needs for different kinds of storage. Flywheels from what I understand are for very shortterm storage needs, i.e. balancing out shortterm fluctuations in electricity use vs. generation. But they're unsuitable for any kind of longterm storage, because they loose power over time.
In the long run with a high-solar-high-wind-scenario we'll need some seasonal storage to get us over a couple of weeks in some circumstances. This will need some storage that doesn't loose power over time.
(FWIW I have no idea if these gravitational storage techs will play any role in that, and one can be doubtful about it.)
Basically flywheel cannot store energy for a long time. It needs to be constantly used to be effective.
Megawatts measure the peak discharge rate, not that amount of energy stored.
For a fixed number of dollars of investment, gravity will store far more joules of energy.
I suspect that if the water table is high enough to cause problems, you'd do pumped hydro storage instead.
A lot of solar power is out in the desert where the water table is hundreds or even thousands of feet below surface.
And pumped storage still doesn't work as you'd need surface area for new lakes. Also, not sure if you'd want to pump out huge amounts of ground water for pumped storage, lowering the water table for everyone.
1. So the units are actually modular, i.e., not connected. Aha. If a unit jams, it's annoying but not fatal, as all the units above it continue to be usable.
2. Since the weight units are separate, then the force on the gearing (both wall and unit) is constant and thus the wear and tear is manageable.
3. The max energy stored is when all the weights are horizontal, of course. So there needs to be a transmission line built into the shaft, along with a motor/generator per weight unit. Some cleverness needed so the transmitted power is passed to (and received from) the correct units. The sliding contact where power is transmitted would also be a major possible point of failure.
4. Yeah, that's the downside of separate modular units. Each unit now needs to transmit/receive power.
5. It would seem prudent to build two or more shafts adjacent to each other. So if there is a problem in one shaft, you can go down the other and fix it. Increases reliability significantly.
6. If the units and power transmission are waterproof, water in the shaft isn't too much of an issue. You might lose ~20% of energy storage due to the buoyancy of the water, but that's it. Might save energy not having to constantly pump out the shaft.
7. Right. Definitely need a failsafe mechanism so a unit doesn't plunge down (say, when the motor/generator clutch fails) and take out the units at the bottom.
Cool idea. Good luck with power transmission.
- Regarding jams, failures, etc. These should be extremely rare, but in the event, each unit is designed to "disengage" if needed so it's just dumb weight, and modules adjacent to it will be designed to lift it. Then when the train is above ground, the unit can be swapped out for another one and repaired while the system keeps operating. - Yup failsafe mechanisms will be built in as well. - Yeah each unit has its own motor/generator. A power line will enter through the module's axle.
How about home use? Could it be used small-scale, e.g. as an alternative to Tesla's PowerPack? I've long been curious about distributed energy solutions, from heating (e.g. cogeneration / CHP, heat pumps, geothermal energy, solar+battery for off-grid energy), could this be a low-cost, low-tech, low-maintenance, low-risk alternative to batteries?
Also this does not seem very convoluted, it's using an electric engine pretty directly. If you are actually boring straight down it seems like the failure modes are pretty okay.
Why? Just build more. Keep expanding capacity so the base load covers the peaks and thensome. Push the price of electricity down while keeping it carbon-neutral and crush competition. Get everyone 2c/kwh power. Wouldn't that be more fun?
> the waste problem.
Ehh, it's well managed in Canada and France, I don't see why Americans can't handle it. Just keep it away from fault lines, right? I bet some of those old missile silos would be perfect.
And please enlighten me, how does France manage their nuclear waste problem?
And those hard-to-use fractions aren't going to be a problem for long, pretty sure they are considerably shorter-time risk than the timeframe for cleanup of WW1 battlefields in France (which currently stands at around 700 years).
Demand response, frequently promoted as a way to increase penetration of intermittent renewables, can also be used to reduce the cost of a system composed on high capital cost, low marginal cost dispatchable generators like nuclear. Charge the EV's and run heat pumps to warm thermal storages during the night when demand is lower, say.
Also, it means you have to run your reactors regularly at below 100%, so you still can react on additional demand, which increases further the cost of a very expensive technology.
IIRC that 5% figure I read was wrt EPR and AP1000, presumably older generation LWR's are slower, by how much I'm not sure. France has run older generation PWR's in load-following mode for decades, but I'm not sure which ramp speeds they achieve; fast enough in practice in any case it seems. CANDU reactors in Canada have steam bypass and can apparently ramp at >10%/min.
In any case, my point is that ramp speed is in practice not a technical limitation. Of course you want to run a generator with high capital cost but very low marginal cost at 100% as much as possible, but if you now and then need to ramp (say, if the wholesale price goes negative) you can do it.
> Also, it means you have to run your reactors regularly at below 100%, so you still can react on additional demand, which increases further the cost of a very expensive technology.
To be clear, I'm not advocating a 100% nuclear grid. I'm just pointing out that the "nuclear can't ramp and is thus unsuited for the grid of the future" isn't correct. In particular, I think solar and a moderate amount of storage is very well suited to cover the daily variation in many parts of the world. I also think that dispatchable low-carbon sources (which could be hydro, or nuclear, or something else like geothermal where available, CCS where geological formations for storing CO2 are available, etc.) have a role to play in least-cost deep decarbonized grids. See e.g. https://doi.org/10.1016/j.joule.2018.08.006
AFAIU France primarily uses reactors which are earlier in their fuel cycle for load balancing, and ones which are near the end run with a flatter profile.
Didn't know about specific as it's not really my area, but would running a PWR with very high enrichment level (AFAIK some designs use 93% U-235) allow quick spin down and spin up?
Well.. there are a lot of factors in a reactor design affecting the ability to increase or decrease power quickly. Geometry, fuel/moderator ratio, fuel density, burnable poisons (for flattening the reactivity swing over the fuel cycle), amount of control rods etc etc. Fuel enrichment being only one thing, which in turn affects other things as well (e.g. reactors using highly enriched uranium tend to use different fuel designs than low enriched fuels).
But yes, military reactors for navy ships obviously have very different demands on them than civilian power reactors, and are designed accordingly.
And yes, while US navy reactors use 93% enriched fuel, it's not necessary, e.g. French and apparently Chinese submarines use low enriched fuel (7% for French).
Which "funnily" enough makes it so that the more dangerous reactors remain in use for longer, as the new designs aren't built, despite huge safety improvements (even if you go just for late 1970's self-sealing molten lead/bismuth reactors).
Overcapacity is a problem, especially for low-enriched reactors, but that's why it shouldn't be the only answer.
If you don't know anything about how a grid functions then maybe you should stop shoehorning a single solution to solve every problem when everyone is already aware that a large array of different technologies is necessary for a modern grid.
Energy storage has also been planned for nuclear power plants as starter and emergency power.
Regardless of your thoughts on Nuclear, this is clearly a much simpler device. Also, I suspect the failure modes in this are pretty boring ...
Most people just don't want to accept the fact that when it comes to energy density, nothing compares to nuclear. Nothing even comes close.
https://dothemath.ucsd.edu/2011/11/pump-up-the-storage/
https://brilliantlightpower.com/news/
I do have high hopes about fusion and molten salt thorium is also up there as something I'd like to be further explored.
But currently nothing up and coming seems to compare to nuclear.
I just did a quick skim of the wikipedia page of that 'hydrino' stuff: https://en.wikipedia.org/wiki/Brilliant_Light_Power#Criticis....
I'm not an expert, but reading those criticisms makes me think this is nothing more than a money grab scheme without delivering anything usable for the past 30 years.
Also, this thread: https://news.ycombinator.com/item?id=24423103
Climate change is urgent and we need all hands on deck to build as fast and as cheap as possible. May the best designs win asap in this fight!
A lot of the cost goes down when the reactor isn't a practically one-off build, and when you can for example use prefabricated components to "assemble" a power plant quickly.
Some designs go even further, and have power blocks that are essentially something you slap on large railcar, including option that instead of refueling you send back the module while the vendor sends you a freshly-fueled one.
By my calculations you would need to raise 40 tons up 3 meters to store 13.5kWh. Definitely not a home storage revolution!
Now, if the entire house was built on a lift....
You would need to raise the 40 tons up 120 meters, or conversely, you would need to raise 1,600 tons up 3 meters.
If you drop a steel weight down the same well of cross-section A and height H in the same units the energy stored is 0.08HA(D-H) MJ[2]. Again, H is in meters, and AD > A(D-H), and in order to beat compressed air 0.08H > 10, so your steel weight needs to be 10/0.08 = 125 meters long! That's taller than most of the buildings in downtown San Francisco -- and in order to get any use out of this thing, your hole should be at least twice that deep.
Of course, compressed air has its inefficiencies and complexity, but the feasibility of a metal rod even close to that long seems pretty low to me. Compressed-air caverns use as much as 7.5 MPa, but a purpose-built well could potentially go much higher. Plus you don't have to deal with the damn thing vibrating from Coriolis forces and seismicity.
Now, I know what you're saying -- you're saying, if you're so smart, why don't you do it? -- but there are simply too many huge caverns out there to even think about constructing CAES chambers. There are several GW in service today. And even with that huge resource people wonder if batteries won't simply corner the market. Storage is getting here painfully slow, it seems like, but the competition is very fierce.
1: True isothermal decompression cycles are impossible, so of course I'm approximating by using the ideal gas law.
2: (8000 kg/m^3)(10 m/s^2)/(mega = 1000000) = 0.08
One of the most prominent new startups out there exploring advanced compressed air is hydrostor. Note that they also dig underground :) https://www.hydrostor.ca/technology/
A Tesla model 3, basic model (MSRP ~$38k) has battery capacity of 50 kWh, so we're talking about four Tesla 3's.
I don't think digging a 150m hole is cheaper than four Teslas - and Teslas come with the rest of the car you can use for driving.
I'm sorry, I'm not trying to be pedantic, but what does "gravity is cheap" even mean?
https://www.storage-lab.com/gravity-based-storage
I agree that gravity based energy technologies are mostly junk though. There are two important factors: cheap weights and unlimited scalability. Doing something "cute" such as using a mountain side or old mineshaft is reducing the scalability.
I didn't do the math but this reeks scam to me
Idea: Could elevators be retrofitted with something like this?
Then they could use/store energy only while going up, and regenerate some of it going down... and take passengers to various floors while doing that!
"The Regenerative Elevator"!
Stores energy going up, regenerates some of it while going down!
Invented here on Hacker News, by yours truly, 9/8/2020!
(Yes, I know, it's a stupid related idea! <g>. But my other related idea was more stupid, and that one was to fill up a U-Haul truck with trash, put a steel cable on it, find a hill, and use the steel cable (in conjunction with the overweighted U-Haul truck and hill!) to drive a motor/generator/winch assembly that uses electricity going up the hill, and regenerates some of it back, going down the hill... <g>)
On a serious note however (for non-passenger elevator energy storage), I think Gravitricity has a good idea, and I wish them much success with it!
I think it would not work so well because they have counterweights to make the up-down operation much easier. If you remove the counterweight, it would store energy, but be prohibitively slow when going up and probably frighteningly fast going down.
Regarding the uhaul, this is not too far off from other people mentioning driving a train uphill and letting it fall downhill to generate electricity. Interesting ideas.
> DC-driven winding-drum elevators—the leading design until the 1930s—use a DC motor in the basement that winds and unwinds the elevator’s steel cable on a steel drum, thus lifting and lowering the car from pulleys atop the elevator shaft. DC drive was the only way to go at the time for a speedy elevator, because only DC could deliver variable-speed operation for smooth starts and stops. The DC motors were also energy efficient, capable of something that has only recently become possible with modern elevator designs: regenerating power when the elevator descends.
https://spectrum.ieee.org/tech-history/dawn-of-electronics/s...
There's lots of research showing how the costs work out (see my other comments here)
Pump air into a giant pressure chamber...that's also a super-deep hole with a pulley system...and the weight for the pulley is a flywheel sealed in a vacuum chamber, magnetically levitating to avoid any friction losses. Oh, and the mass for the flywheel? A bunch of batteries.
High energy density lithium batteries are currently being commoditized. They are also increasingly able to handle many thousands of cycles. E.g LiFePo4 cells.
The raw materials for batteries are not actually that expensive or rare, so that leaves the manufacturing.
You might think that making something as complex as a battery can never be as cheap as hanging a weight from a rope. But there are examples of very complex products (solar cells, LCD displays) that became incredibly cheap due to mass manufacturing.
A square meter of solar cells, requiring extremely pure silicon and nanometer scale engineering, is now not much more expensive than a square meter of good roofing shingles.
Why would this be better than a set of railway lines down a hill side into a forest pulling up standard goods carriages full of rocks?
You get a forest (which buffers runaway carriages and does all the other lovely forest things) and energy storage on the hill. All using largely commodity items and you can build it anywhere there's a spare slope.
[1] https://www.aresnorthamerica.com/
This site claims a max weight of 4.5e6 kg. Lake Powell, for example, has a max capacity of 3e13 kg!
So, SEVEN orders of magnitude less mass to move. Sounds competitive.
The country point is interesting, so the market is countries lacking topography suited to hydro plants? How big is that market really?
And considering that pumped hydro is currently the only viable grid scale energy storage mechanism (inb4 but li-on), those countries currently either 1) share a grid with neighboring countries that do have pumped hydro, or 2) forego energy storage altogether and use peaker plants like everyone else. If you do 1 you better trust your neighbors to never go to war with you, otherwise you better have 2.
So the market is: countries lacking topography suited to hydro plants, who currently share energy storage with neighbors who they don't trust, and who for some reason aren't just building peaker plants (which are essentially just normal fossil fuel power plants) and are holding out for a non-extant technology to be developed.
I'm not much of a capitalist myself, but even I can tell that that's a bad investment.
> All obvious places already have them
That is, we’re already practically at max capacity for hydro.
https://en.wikipedia.org/wiki/Pumped-storage_hydroelectricit...
Hydro plants are built reversible whenever possible. The ones that aren't ("run of river") are that way because it was impractical to impound storage.
If a hydro site is convertible to reversible, chances are it’s been already. Most sites are not really suitable because regular hydro plants don’t need a lower reservoir or water source and so can be built in locations which don’t have one.
Meanwhile pumped hydro can be built in locations where normal hydro makes no sense because there is no natural downflow or upper reservoir e.g. Taum Sauk.
A hole in the ground can be dug up everywhere, e.g. next to each single PV plant.
But maybe capacity is only part of the story. What about locality? Distributed energy storage may not be as efficient, but it has a lot of other benefits.
There is also the fact that weights are expensive relative to the energy they can store so you have to take them directly from nature (usually via dams or pumping water uphill).
[0] https://web.archive.org/web/2016*/http://www.gcrg.org/bqr/6-...
In some places pumped hydro does make sense of course but you have to compare basically on $/kwh what is the best solution taking into account the local environment. There are no silver bullets here. Which is why a lot of places seem to be ending up with lithium ion batteries, because they are there and can be put in place without a lot of fuss or rearranging of the local environment.
This seems to be one of the key indicators of green pseudoengineering - an obsession with avoiding transporting power. Pylons work. Really. If you can get large amounts of RELIABLE renewable power/storage, with low labour and machine costs, from a large amount of cheap land, land far away from people or crops or even trees, you do not have to worry about the rest. Worry about how to get those labour and machine costs down.
Pylons may work, but you need to pay off everyone owning the land those lines go across. In many areas, they can't build pylons at all because of rampant NIMBYism.
Water is 1 tonne per cubic meter. Steel is 7.9 tonnes per cubic meter. Lead is 11.34 tonnes per cubic meter - which means that in the same amount of space, you can store vastly more energy if you use these metals instead of water.
The initial cost of set up will also vary depending on the metal you use but lead is cheap.
Also, water evaporates etc.
Cost of lead/tonne : USD 1900 (approx) Cost of Uranium/tonne: At least USD 60000 (approx regular uranium) even if you are able to source it.. and then you still have to build containment for this, get permits etc..
For that price, you could get much higher efficiencies if you went for a cheaper metal - even something like Tungsten which is extremely dense will be much cheaper.
To me it smells like https://www.youtube.com/watch?v=uzV_uzSTCTM and all about patents, marketing and money burning, but I don't know as much as these professors do so maybe I am wrong.
UPHS has never been fully built to my knowledge, but it's been well studied and quite a few projects have tried to get funding for it.
U.S. DOE research from 1984: https://www.osti.gov/biblio/6517343.pdf
Projects trying to work on this: https://utilitymagazine.com.au/pumped-hydro-storage-the-futu...
Other proposed projects: - https://www.waterpowermagazine.com/features/featureinvestiga...
- http://www.eaglecrestenergy.com/project-description.html https://www.osti.gov/biblio/6517343.pdf
You can look at the orgpage about these ideas https://orgpad.com/s/energiewende my father also gave a talk last week about it, there is a recording, which will be posted during the next days.
Disclaimer: I work for the small startup OrgPad, which tries to create a tool for easier decomposition of linear ideas/ content into a network of ideas/ content. An ex-Googler, Pavel Klavík PhD. describes the technology (hint Clojure and ClojureScript) and approaches behind OrgPad in a recent talk https://www.youtube.com/watch?v=4UoIfeb31UU
This doesn't involve digging a deep hole in the ground and hope that the earth keeps it level.
A drawback would be the hazards that heavy tension/forces bring with them.