Assuming a very large device was constructed which could harvest up to 20% of the mechanical energy of a tornado. It is a mobile platform of some sort, and one or several of these are ready to deploy in a prone region such as America's tornado alley.

Assume the simplest scheme: the device turns the force of the tornado into rotational work, by deploying some construction of large vanes, kites, or other deflecting surface capable of withstanding the forces. On placing itself in the center of the cyclone, it is able to draw off 20% of the rotational energy and store it before the cyclone wanders away. The construction is unknown at this point. It is designed to handle the most common EF-2 cyclones, and can shut down and secure itself against larger ones. For calculation purposes, it can complete the harvest by maintaining itself within the center of the cyclone for two minutes, then the funnel wanders off with 75% of its original energy (the process leaves 5% lost as deformed metal, heat, deafening acoustic energy, and eroded metal).

EF-2 tornado strikes near Des Moines, Iowa, USA

While the energy capture device itself may currently be beyond our technology, the question simply concerns a storage mechanism that can recover as much of this huge rotational work as possible, while being accessible to a mobile platform (the unpredictability of tornadoes make it impossible to pre-stage your energy batteries, or the device, within some large construction). As such, the hard science tag only assumes the input of a known quantity of energy at a known rate from an unknown location or time.

What would be the most efficient means to store captured tornado energy given the high rate of delivery, and inability to predict the location?

My worlds are largely wind-powered so this answer can serve several applications generally.

  • $\begingroup$ (1) A tournado tournadoes over a very large area. The described device is very much smaller than the area over which the tournado tournadoes. It will only have the opportunity to harvest a tiny fraction of the energy of the tournado, because the vast majority of it will never come anywhere near the device. (2) What on earth is "rotational work"? (3) What is numerical meaning of "huge" work? in joules please, or, if one must use olde skoole measures, even in foote powndes-force. $\endgroup$
    – AlexP
    Commented Mar 7, 2022 at 16:25
  • $\begingroup$ @AlexP Rotational work is force that changes the angular velocity of a mass. The science-fiction premise is that your concerns have been resolved by technology. The cyclone is an EF-2 and the device captures 20% of the net energy, this is what "huge" means. There is a possible consequence of baffling and diminishing the funnel but that is beyond this question. Most of the kinetic energy in a tornado is invested in rotational movement, not in moving the cyclone itself. $\endgroup$
    – Vogon Poet
    Commented Mar 7, 2022 at 16:33
  • $\begingroup$ @VogonPoet Just so that nobody else can become grumpy at that, is the "rotational work" you describe torque or "rotational force"? Hard-sciency people like hard-definitions :p. $\endgroup$ Commented Mar 7, 2022 at 16:46
  • 1
    $\begingroup$ @Tortliena No, torque is not work at all. Rotational work results after torque ($\tau= \text{F} \times \text{r}$) has been applied to a mass, and the measure of rotational work (in standard work units of joules) is related to torque by the angle the mass rotates because of it (W$= \text{F} \times \text{s} = \text{F} \times \theta \times \text{r} = \tau \theta$). Here is the calcualtion of rotational work in physics. $\endgroup$
    – Vogon Poet
    Commented Mar 7, 2022 at 17:14
  • $\begingroup$ Capturing a tornado's energy sounds cool, but insanely impractical. The capture device needs to be strong enough to withstand a tornado, but mobile enough to follow tornados. Why not just put lots of windmills everywhere, which may get a boost if a tornado is kind of close? $\endgroup$
    – Atog
    Commented Mar 7, 2022 at 17:26

6 Answers 6



Handwaving the many reasons why storing a tornado in a bottle (metaphorically) would not work, and would be an impractical way to gather energy, if you were going to try, based only on the parameters you’ve provided, I’d recommend flywheels.

  • Are already a rotational form of energy storage
  • Are heavy but can be portable
  • Can charge very rapidly / under high impulse

Once the portable flywheel rig is set up and the tornado has passed, you can transfer the power to longer term storage e.g batteries at your leisure.

  • $\begingroup$ My assumption is that an estimate of the size of such a flywheel would have a large surface producing friction in the air, and a mass riding on remarkable bearings. This isn't disagreeing, but my intuition says the energy would be draining out fairly quickly. It certainly does buy more time than what the tornado grants, however. This is certainly an answer. $\endgroup$
    – Vogon Poet
    Commented Mar 7, 2022 at 16:41
  • 1
    $\begingroup$ @VogonPoet Storing the flywheel in a vacuum chamber and having it ride on magnetic repulsion might help overcome these issues. I'm not an engineer and I'm not sure how practical this technology is, but perhaps it would be worth checking out. $\endgroup$
    – Qami
    Commented Mar 7, 2022 at 16:51
  • 9
    $\begingroup$ @VogonPoet There are commercial flywheel energy storage manufacturers on Earth today. The flywheels are usually contained in vacuum and often operate in multiple sets. $\endgroup$
    – user535733
    Commented Mar 7, 2022 at 16:55
  • 2
    $\begingroup$ @VogonPoet if the device is getting significantly moved around, it certainly isn’t going to be absorbing 20% of the tornado’s power, since the tornado is moving quickly relative to the ground, not itself. I’ve taken for granted that you have some kind of mystical, stabilized platform that is capturing the wind. The flywheel setup (in a vacuum chamber, is are going to be at least as portable as your device) is simply there to provide a power storage solution with a very aggressive charge curve. $\endgroup$
    – Daniel B
    Commented Mar 7, 2022 at 21:42
  • 1
    $\begingroup$ In general, your energy-storage shouldn't need to be placed in harms way. You could feasibly bury the whole system underground and feed energy into the system electrically. A motor to add velocity to the flywheel hooked to a dynamo on a turbine that gets spun by the tornado. Engineering a turbine sufficiently durable and efficient to not be destroyed by a tornado is a much harder challenge. $\endgroup$
    – Ruadhan
    Commented Mar 8, 2022 at 12:09


Use windmills to capture the storm, then pumps to lift something like water from a low place to a high place, as in pump-storage hydroelectric dams.

It is a better way to store normal wind than tornados, but tornados tend to come with high winds.


A general alternative is "weight lifting", you use the energy you harvest to lift many tons of weight to a higher altitude; and then the controlled lowering of this weight, which can start and stop as needed, generates electricity.

A practical implementation is already in use: Excess solar power is used to pump water from a "low" reservoir to a "high" reservoir during the day, and during the night, controlled draining of the high reservoir into the low reservoir, through turbines, generates electricity at night. These are lake-sized reservoirs, obviously, often replenished by rain. In the high reservoir, this is a bit of free energy.

The reservoirs can be built, by digging for the low one, and using the material to build the high one. Or think of something like the Hoover Dam; which operates on nearly the same principle, except we aren't pumping to fill it; a river running downhill fills it for us.

This is a way to turn intermittent energy into energy on demand; although there is about a 15%-20% loss in the energy due to unavoidable inefficiencies in pumping. (That's why they use solar power directly if they can, to avoid that loss, but they are designed to over-generate by a significant factor so they can pump and cover the night needs.)

  • $\begingroup$ I considered this as well, but I don’t think it would be easy to make portable. $\endgroup$
    – Daniel B
    Commented Mar 7, 2022 at 21:43

You have a big problem.

First, we need to know how much energy will need to be stored. According to the NIH, the typical EF2 tornado has 88 TJ worth of energy, so 20% means storing 17.6 TJ.

How much is that? Well, converting 17.6 TJ to TNT equivalent (4.184 GJ per tonne TNT) gives 4.206 KT of TNT roughly 30% of the atomic bomb dropped on Hiroshima. Energy storage systems are often expressed in terms of kWh. 17.6 TJ is 4.89 GWh

So, ignoring the other costs just for the moment, what is the cost of grid-scale storage for 17.6 TJ. According to the National Renewable Energy Laboratory in 2021, and using the low-end of the range for storage costs.

Lithium Ion    352 USD/kWh
Lead Acid      380 USD/kWh
Sodium-sulfur  599 USD/kWh
Pumped Water   150 USD/kWh
Compressed Air  97 USD/kWh
Flywheel      4320 USD/kWh
Thermal         20 USD/kWh

Pumped Water hydro storage is clearly not portable. Thermal and compressed air also also very unlikely to be portable - knocking out the 3 cheapest methods. So, what it the cost for lithium ion storage - 5.89E9 * 352 / 1000 = 2.07E9, or over 2 billion dollars.

Though this is a stretch, it becomes much less feasible when you consider have much time you have to charge your storage system. Attempting to charge that much lithium ion storage in 2 minutes would result in a very large fire. Other storage systems would have similar problems. You need to store energy at the rate of 146.7 GW - 6.5 times that output of the Three Gorges Dam.

Regardless of you storage system, you need to be able to store the power at the rate of 146.7 GW - this will never be true in a portable device that you must guide into the storm path. An F2 system has a remarkable amount of energy.

I would say the nobody is ever going to capture that tornado

  • $\begingroup$ Why would you eliminate thermal or compressed air? A large pneumatic tank can be portable and reinforced, filled via a turbine-driven pump. An ablative friction clutch, or a resistor bank could capture a great deal of heat energy and contain it in a vacuum sealed tank (or maybe even a large aerogel kiln). Perhaps it generates a vat of molten lead. I agree that an electrical intermediary would be challenging (unless you use ultra capacitors which can charge as fast as you can get it there). $\endgroup$
    – Vogon Poet
    Commented Mar 7, 2022 at 20:37
  • $\begingroup$ @VogonPoet, utility-scale compressed-air storage generally uses salt caves or abandoned hard-rock mines, which are not particularly portable. $\endgroup$
    – Mark
    Commented Mar 8, 2022 at 2:29
  • 1
    $\begingroup$ @Mark … I see. ¯_(⊙︿⊙)_/¯ $\endgroup$
    – Vogon Poet
    Commented Mar 8, 2022 at 2:39
  • $\begingroup$ @VogonPoet thermal storage is also not as you expect. At that price it is a substitute for space heating, where heat is stored when a heat source is available and released when it is needed. This can be a result of heat pumps, which have excellent performance when the temperature differences are small. It relies on the thermal storage materials be co-located with the structure being heated or cooled. $\endgroup$ Commented Mar 8, 2022 at 12:50
  • $\begingroup$ @GaryWalker then a transportable item with high heat capacity in container of low heat conductance is needed? At least as long as it takes to get it to the heat exchanger for storage (like the flywheel)? $\endgroup$
    – Vogon Poet
    Commented Mar 8, 2022 at 13:30

You want to harvest static electricity and temporarily store it in super-capacitors.

These can be charged very quickly and will hold their charge for a few days (depending on the capacitor design). It would essentially look like one of those mega trucks. You wheel it in front of the tornado path, charge your capacitors, then drive it off to some central location to offload the energy.

This would be similar to a Van de Graaff generator (https://en.wikipedia.org/wiki/Van_de_Graaff_generator) but the mechanical element is already provided by the tornado.

As the air particles drag though your apparatus, an enormous amount of static will be created which you can then store in super capacitors (https://en.wikipedia.org/wiki/Supercapacitor).

Capacitors are very elastic as well so you can do this over and over with minimal degradation. As a bonus, it should be possible to design a static generator with no moving parts, which will last much longer (you know, if your energy collecting truck doesn't get hit by a shed)

Edit: As noted by @Goodies, the specific energy of capacitors is low in comparison to other storage mediums. The wikipedia page has the higher end at about 100 Wh/kg. Supposing we want to store 1 GWh (for simplicity) we would need 10 million kgs of capacitors.

Considering that the worlds largest truck (https://en.wikipedia.org/wiki/BelAZ_75710) can load a meager 450000 kg, you would need about 20-25 of these bad boys working together to harvest the tornado.

  • $\begingroup$ Hi @TsSkTo, welcome as WB-answerer .. I have to make a remark about the last part: you can release a lot of power (kilowatts) from supercapacitors.. but energy storage (kilowatt hour) in a supercapacitor is less than 5% of a much cheaper lithium ion batteries per unit of weight. Study the table here en.wikipedia.org/wiki/Supercapacitor $\endgroup$
    – Goodies
    Commented Mar 8, 2022 at 15:43
  • $\begingroup$ @Goodies Which is why we typically don't use them in this aspect. There are a few Lipo/capacitor combos out there that bridge the gap. But as far as fast energy storage goes, you really can't beat a capacitor. Unless we're using high energy density fuels. $\endgroup$
    – TsSkTo
    Commented Mar 8, 2022 at 17:06
  • 1
    $\begingroup$ +1 nice edit but don't overestimate my status, please spell my name right :p hehe $\endgroup$
    – Goodies
    Commented Mar 8, 2022 at 17:24

Ignoring like others, the issue that your device isn't ever exposed to more than a small proportion of its highly movable and unpredictable rotational energy, I reason this way for my answer......

@GaryWalker outlines the extreme high rate of charging or storage implied. @Amadeus explains his lifting a large mass against gravity in theory works but in practice wouldn't be much help.

Solution - large volumes of stored/possibly somewhat pressurised air

Building on those, I'd suggest your device converts rotational energy into a pump, and pressurises an immense air container made of some elastomer. Here's why:

  • Lifting an object stores energy but is limited by the mass involved, you want a lot of mass for it. Pressuring air also stores energy very nicely but you don't need to carry round a 100,000 ton block of concrete or steel (or whatever mass it is) for the purpose. Its far more scalable.
  • Its also much better suited to this situation. Air can be pressurised by fans at very high rates, if the fan is adequately powered to rotate fast - just think of a jet airplane's turbofan engine. Now power an array of them with vertical intakes, by rotation from a tornado. (The engines can be stripped right down as you need a gearing element and something the tornado will rotate to turn it, and they never leave ground level so they can move easier or form from individual carriers that pack together and move cohesively or something to track the storm)
  • Airbags can accept ultra fast pressurisation easily if well designed and large. They'd need restraining or the storm would catch them, that's easy enough, exercise for the OP. The elastomer design ensures that they have a tendency to deflate, makes extracting energy easier, although really all that's needed is to roll them up and use the emerging air jet.
  • Simple and fairly foolproof.

It's important to note the purpose of pressurisation in this solution.

We are using pressure to quickly store huge volumes of air very very fast, while the storm rages, not to store it at ultra high pressures or liquefy it. (Those are options, of course, for variant answers).

Air can take immense pressurisation, and we need that for rapid storage. But really you only need a big airbag, because even without high pressure storage, as you roll it back up, you'll create a powerful air jet to reuse, purely because of volume of stored air, even if not especially pressurised.

(Think of the blast of air from an airbed as you sit on it to get the air out when you pack it away)

So the pressure is much more about ability to capture ultra-high air volumes at an ultra-high rate, which may be at some pressure (but doesn't really need to be), than about storing it at high pressure. So the bags don't have to be able to resist huge pressures, except at the point of intake.

Other physical side effects

Also note you'll have thermal effects. Compressing and decompressing does that. We can handwave those, but really, the only high pressure process here is compression, and we design the intake fan system to dispel thermal effects. they'll get hot, and the incoming air will be a bit warm I guess, but not super hot. The outgoing air is released at a much slower rate, so that's much less an issue.

You could have condensation effects too, but that's veering more to hard science/realism, and if they're ignored in a story nobody will notice.

  • $\begingroup$ What did I write that led you to this idea: "your device isn't ever exposed to more than a small proportion of its ...rotational energy"? $\endgroup$
    – Vogon Poet
    Commented Mar 8, 2022 at 13:21

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .