Given the difficulty of producing energy for a region when the wind isn't blowing, why is it not possible to have half of each wind farm producing energy directly, and the other half working to lift a weight, that then descends and produces energy when the wind stops.

Is this a valid solution to produce all of a region's energy via wind?

  • 5
    $\begingroup$ google gravity battery $\endgroup$
    – jsotola
    Commented Nov 11, 2023 at 3:29
  • 1
    $\begingroup$ Canada was one of the first to produce hydrogen from excess energy from windfarms. Check out Dinorwig - same concept could be applied… some easy research. $\endgroup$
    – Solar Mike
    Commented Nov 11, 2023 at 7:10
  • 1
    $\begingroup$ Related: physics.stackexchange.com/questions/305563/… $\endgroup$
    – jpa
    Commented Nov 11, 2023 at 15:21
  • 8
    $\begingroup$ That's what pumped hydro is, the "weight" is water. $\endgroup$ Commented Nov 11, 2023 at 19:34
  • 3
    $\begingroup$ No. All the good gravity storage sites have been identified decades ago, and there isn't nearly enough of them. Other systems to store energy, and systems that reduce the need to store energy, are required. Also, the good storage sites tend to be a very long way from good wind/solar generation sites, and also a very long way from demand centers, although that last item could change. DOE thinks there are 1800 potential hydro storage sites in the Alaskan rain belt. But in New Jersey, not so much. $\endgroup$
    – Phil Sweet
    Commented Nov 12, 2023 at 14:44

7 Answers 7


The storage part of the RES systems is probably the biggest challenge today. There are several ways to store energy. By far the most used way today is Pumped storage hydroelecticity systems.

Essentially what they do is they pump water when there is a surplus of energy (and water) and then use that stored water for electricity production.

enter image description here Figure: example of a storage system (source: https://www.mdpi.com/1996-1073/16/11/4516)

Compared to other methods (electrochemical, spinning inertia etc) they offer significant advantages in:

  • Energy storage scaling: in the order of hundreds of GWh (one of biggest advantages since other technologies are in the order of MWhs)
  • power capacity: in the order of GW. (one of biggest advantages)
  • depending on the size, they can discharge for over a day.
  • Grid stability and load balancing: They have response time in the order of seconds, which means that they can go from rest to 100% production in a very short time (other technologies might also compete with that).
  • Good integration with renewables.
  • They can couple with irrigation systems (if the planning is correct).

The biggest problem is that there are limited locations that they can be built (so its a finite resource) and that they require careful planning and have high initial cost.

Additionally, lately there has been a renewed interest. The following image shows the cumulative installed/operational capacity worldwide. As you can see there are several planned pumped hydrostorage systems planned for the next ten years.

enter image description here

  • $\begingroup$ Thank you. Would you say then that 100% of energy production for say the USA could indeed come from wind, so long as one builds enough of these hydro plants? $\endgroup$
    – apg
    Commented Nov 11, 2023 at 14:08
  • 1
    $\begingroup$ @apg Let's do some math. US electric consumption for 2022: 4.05T kWh. That's 11B kWh/day. Typical load factor for wind is 30%, thus 7.7B kWh * 3.6E+6 Kwh/J = 2.8E+16 J. 2.8E+15 kilogram-meters. Lake Mead (the largest hydro dam in the US I'm aware of) could hold 3.2E+13 kg of water. You'll need to lift that whole lake 87.5 meters every day. To where?? And considering periods of calm you'll need far, far more than that. $\endgroup$ Commented Nov 12, 2023 at 17:18
  • $\begingroup$ Can the individual wind farms not just have a "wind up" mechanism, not put all the energy storage into a large lake (which as you say is impractical). Then you build twice as many farms, and half have winding, and half unwinding. Like lots of small hydro plants localized in the pillars of the wind turbines. I suppose a modern battery basically for each farm. $\endgroup$
    – apg
    Commented Nov 12, 2023 at 18:53
  • $\begingroup$ @apg theoretically yes. As I mentioned the problem is that the locations are finite. That is the biggest challenge. In Japan they've created a small project that pumps seawater but to my knowledge this has not been replicated. $\endgroup$
    – NMech
    Commented Nov 13, 2023 at 6:35
  • $\begingroup$ However if your question was whether wind has the capacity, yes it does many fold. $\endgroup$
    – NMech
    Commented Nov 13, 2023 at 6:36

You've described energy storage, or physical batteries. You wouldn't need half of each wind farm dedicated to this; you would simply store the excess, beyond what is used at a given time. It's not weights that are best to use, but water pumps that raise the level of a reservoir and then use the water to drive hydroelectric generators when it's used.

The main problem with trying to store either wind or solar is that our electrical grid is not set up for it yet. Well over 99% of energy produced by the grid is used as soon as it is produced, and the production rate is increased or decreased with demand. Creating widespread energy storage is possible, but you are talking about a trillion dollar undertaking. And the economic motivation is not there, because our current system of using what we make real-time is working fine for everyone (except the environment – and that would be fine too if we switched to nuclear).

  • 1
    $\begingroup$ Water is the only thing where such storage is remotely economic and it requires substantial cooperation from geology to construct the high and low lakes required. Grid-scale power storage requires a truly massive amount of weight to be moved--and I think the largest installations are still just quick-response standbys to give them time to spin up something bigger but slower. $\endgroup$ Commented Nov 12, 2023 at 6:05
  • $\begingroup$ Pumped reservoir storage is indeed a good form of storage, however in most developed nations, there is very little capacity for increasing capacity. $\endgroup$ Commented Nov 13, 2023 at 15:16
  • $\begingroup$ @JackAidley, pumped storage is actually vastly under-utilized compared to installed hydropower. For example, the Columbia River is a continuous chain of reservoirs from Priest Rapids Dam nearly all the way to the headwaters, but only has one pumped-storage facility (Banks Lake, off to the side). Replace all those turbines with pump-generators, and you'll have a spectacular amount of storage available. $\endgroup$
    – Mark
    Commented Nov 14, 2023 at 1:39

I suppose what you had in mind with your question was something like, every single turbine winches up a heavy weight to its tower while surplus energy is available, then lowers it again to recuperate the energy when needed. This is of course in principle possible, but some back-of-the-envelope numbers quickly show that it's not really practical: going by a typical wind turbine as Wikipedia describes it, we have a power of 1.5 MW at a tower height of 80 meters. Assuming we need to store energy for at least, say, 4 hours, that would be

E = 1.5 MW × 4 h = 21.6 GJ.

If this is supposed to be stored by lifting weight up the 80 m tower, we'd need a mass of

m = Eg × 80 m = 27 500 t,

WolframAlpha helpfully remarks that this is about half the mass of RMS Titanic, and almost 100 times as much as the 300 tons the turbine's nacelle itself weighs. Accordingly, the tower would need to be much sturdier and more expensive. Not something you want to have to build, especially not in a remote location where wind turbines tend to be placed. This is why most existing gravity storage uses existing topography instead.

The only real advantage to keeping the storage at the turbine itself is that only one generator is needed, whereas coupling turbines to pumped hydro requires a generator in the turbine powering a motor+pump in the hydro plant. Plus electricity floating a longer distance along power lines. But both of this isn't actually such a big deal: large electric motors/generators are highly efficient and though they are certainly not cheap, it's relatively harmless compared to the price of the turbines and bearings. Note also that having a winch in the turbine nacelle would complicate the mechanical aspects a lot further, you'd need extra gearboxes and clutches and bearings which might well end up being both more expensive and inefficient than a separate pumping motor in a hydropower plant.

So all of this looks a bit bleak. Nevertheless, there is actually one technology conceptually similar to the one you're proposing that could prove practically useful: the Integral Compression Wind Turbine, an idea spearheaded by Seamus Garvey. It works by using pistons in the turbine's blades to compress air. This compressed air can then be stored and used for electricity generation at a later time. This keeps the turbine simple (indeed simpler than a standard electricity-providing one).

The intended application is offshore windpower, because there it's relatively easy to store large amounts of compressed air, namely in a bladder anchored to the seafloor, using the fact that the pressure there is much higher than atmospheric.

I suppose what also could just about work is a turbine on top of a mountain with a (direct-mechanical) winch that's used to drive a funicular with heavy weights up the side of the mountain. But building that funicular won't be cheap either, and again it would likely be more economical to build it in a separate location and drive with grid power.
  • $\begingroup$ On a more grokable scale, lets say I want to reheat a mug of coffee in a microwave oven (700 W for 143 seconds) using a 80 m gravity fall. 80 m/143 s is 0.56 m/sec. (700 kgm^2/s^3)/(9.8 m/s^2)/(0.56 m/s)=127.5 kg per cup of reheated coffee. $\endgroup$
    – Phil Sweet
    Commented Nov 13, 2023 at 16:31

You're describing a Gravity Battery. While the other answers have pointed out that water is the most common lifted weight, EnergyVault is a Swiss company that's basically lifting concrete blocks. The total energy efficiency is similar to water, but as you imagine, once you've lifted a weight to the "top", you have to put it somewhere, un-rig the crane, and then grab the next weight which is a big engineering challenge. There is a similar project which uses a mine-shaft to avoid this challenge, but it's limited in its capacity because it only has the one weight.


As other answers have pointed out, the key to a system that utilizes renewable energy sources with an energy storage backup envisioned by the question is the ability to store large amounts of energy that are also readily accessible. Pumped hydro is the classical approach which has been in use for more than a century [1].

However, it is often difficult to find suitable sites for pumped hydro facilities, as they require suitable sources of water as well as easily exploitable differences in elevation. In addition, proposed pumped hydro projects are not infrequently challenged on environmental grounds. For these reasons, and with the advancement of technology, large-scale energy storage using electrochemical batteries is currently being pioneered [2], in particular in California which has few sites suitable for pumped hydro storage.

Battery storage systems in California are typically designed with a capacity that allows them to store four hours worth of energy at their rated power, so these are short-term storage solutions. As of summer of 2023:

Representing the largest concentration of lithium-ion battery storage on any grid in the world, the growing storage capacity – we reached 5,600 MW as of July 1 – is critical in decarbonizing the bulk power system and to our ability to keep the power flowing as California transitions to a carbon-free system.

At present, the largest single battery-based storage facility in California is the Moss Landing Energy Storage Facility, which is located on the site of a former natural gas powered power plant. This has the obvious advantage that much of the needed electrical infrastructure was already in place. The battery storage was installed in three phases, with the project completed by the summer of 2023:

An additional 350MW output and 1,400MWh energy capacity has been added to the plant, bringing it to a total 750MW/3,000MWh. This comes after the 300MW/1,200MWh Phase I was completed in 2020, followed by the addition of another 100MW/400MWh in Phase II the following year.

For comparison, the largest pumped hydro power plant in the United States, Bath County Pumped Storage Station, provides peak power of 3,000MW and 24,000 MWh of energy storage, so four times the power and eight times the storage capacity of the Moss Landing Energy Storage Facility.

Because of geography California utilizes relatively little wind energy but plenty of solar energy. The batteries charge up during the day when there is often a surplus of electricity produced from solar energy and discharge in the evening when the sun starts to set and usage often peaks, especially during the hot summer months.

The dashboard of the California Independent System Operator (CAISO) provides handy graphs that show how that works in practice. The battery storage facilities discharge in a staggered fashion depending on electricity market conditions, collectively covering about six hours. From casual observation, the daily peak discharge rate during the second half of 2023 seems to be around 3,000MW to 4,500MW, representing around 10% of California's electricity demand at that time.

Apart from occasional mishaps in the form of fires, large battery storage plants for short-term storage appear to be a solved engineering problem. My understanding is that the biggest remaining issue is the relatively high cost of battery storage, which exceeds the generation cost of electricity from solar and wind by factors. For example, for 2027 EIA projections are for a levelized cost of electricity (LCOE) for solar of \$36/MWh and onshore wind of \$38/MWh, but levelized cost of storage (LCOS) for batteries of \$125/MWh.

[1] The (to my knowledge) very first pumped hydro power station in the world was a 2.5MW facility in Switzerland that became operational in 1909, then called "Zentrale C" of the Schaffhausen power plant complex, but now usually referred to as Engeweiher power station. See: "Die Erweiterungsbauten des Elektrizitätswerks der Stadt Schaffhausen", Elektrische Kraftbetriebe und Bahnen, Vol. 8, No. 21, June 24, 1910, p. 414

[2] According to the U.S. Energy Information Administration: "Before 2020, the largest U.S. battery storage project was 40 MW."


Energy storage is a tricky challenge.

Batteries are one, but the cost of even city scale single day's storage is gargantuan.

Others have proposed using renewable energy to generate hydrogen or in-situ methane fuel (from carbon dioxide and water) or some kind of methyl fuel. However making 1 kilogram of hydrogen takes 50 kWh of power alone. Also requires water, in dry regions, is inapplicable. On the other hand, if hydrogen is produced from shore and brought inland to a location by fuel tankers or pipeline and used in gas turbine (combustion) or fuel cell, water as byproduct can be used as a source of water.

A flywheel storage is what you might propose,

enter image description here

But efficiency is low, as half energy lost.


Not a full answer, but an addition to the existing answers. I see that most of the other users restricted their options for energy storage to the systems that are more publicised, not the ones that are really possible. The fact is that there are a lot of possibilities to store energy, the gravity storage that you proposed, that is not pumped storage can store little energy.

If you have a district heating system reachable by an electric connection heat storage has a good efficiency/cost ratio. It is limited to the availability of the sites like it happens for pumped storage, but it can contribute.

In a place with cold temperatures where there is no district heating encouraging hotels, offices and other big buildings to put a sand tank to store heat for hot water and heating would add up some storage that is cheap and easy to implement.

Flexible production by industrial plants that require a lot of energy. Plants that make silica or magnesia or fertilizers or refine metals using electrolysis and so on. There are a lot of production processes that require a lot of energy and are designed to work at a constant rate. A variable production rate would be as efficient as energy storage.

Heat storage in molten salt tanks to generate electricity or to provide heat to chemical plants would be another option.

Cooling storage as tanks of compressed air in the basements of big hotels, offices and other public building. It would be used to power the air conditioning when needed.

Cooling storage as tanks of liquid nitrogen in food processing plants.

  • $\begingroup$ Saltwater desalination is a very flexible, energy price sensitive consumer. Although enabling addition consumers is probably not central to the op's question, potable water is a critical commodity in much of the world. $\endgroup$
    – Phil Sweet
    Commented Nov 13, 2023 at 16:45
  • $\begingroup$ @PhilSweet Good point. Thinking about all the overlooked details the list could grow large. $\endgroup$
    – FluidCode
    Commented Nov 13, 2023 at 17:54

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.