# Why is there no spring based energy storage?

I'm wondering about kinetic energy storage for homes. Imagine a concrete plate resting on hundreds of firmly attached sturdy springs, and a couple of electric winches attached to the top.

To store energy, pull up the plate. To release energy, release the plate. Springs store energy with the square of the displacement.

What part of this won't work?

EDIT: I was wrong about the square of extension behavior, the actual behavior is linear+non-linear over a quite short distance as described here. Therefore the answer is low energy density as explained by the accepted answer.

• There are spring-based energy storage devices. This is how watches that you wind work. A little web searching brings up plenty more information. What are you looking for that you can't find? Commented Jan 10, 2021 at 21:25
• – LShaver
Commented Jan 10, 2021 at 21:41
• A wind-up watch?
– J...
Commented Jan 11, 2021 at 14:41
• Spring-based energy storage is common in toys: jack-in-the-box, snake-in-a-can. Commented Jan 11, 2021 at 15:52
• I've heard of the opposite system - a concrete block on springs, say, a sidewalk paver in a busy city. As people walk on this block, it compresses the springs underneath it and - something-something, kinetic energy is stored for later use. Not sure the exact mechanism or if it was ever implemented. Seems like it'd be a good source of "free" energy in any densely populated city. (Though less so during a quarantine of course) Commented Jan 12, 2021 at 21:48

### Because springs have low energy density

When storing energy, especially in a residential setting, you want to be able to store a lot of energy, or not take up too much space. To store a reasonable amount of energy with a steel spring, you need a large spring (or a lot of small springs). The 2014 paper "Benefits and challenges of mechanical spring systems for energy storage applications" includes this table comparing the mass-based and volume-based energy density of various energy storage systems:

A steel spring is 100 times larger by mass than a battery system, and 50 times larger by volume, for the same amount of energy (using the low end estimates for batteries).

To visualize this, let's compare it to a Tesla Powerwall, which is about the size of a large television (0.13 m3), and stores 13.5 kWh of electricity. For the average U.S. household, this would last about 11 hours. It's energy density is about 100 kWh per m3.

To get a similar amount of energy from a system of springs, it would need to be 45 m3, or about 350 times larger. Assuming a ceiling height of 2.5 m, this would take up about 18 m2 -- about the size of a single car garage. Alternatively, if the spring system took up the same amount of space as the powerwall, it would only store 0.04 kWh, or enough to run a single LED light bulb for about six hours.

The article also explores the possibility of carbon nano-tube (CNT) springs in various arrangements. While they are still less energy dense than batteries, they are superior to steel springs, and their unique structure and properties may make them suited to storing vibrational energy in some niche applications.

• I'm fairly sure a spring with a diameter of 5 meters would be unusable as a spring. Commented Jan 11, 2021 at 14:13
• @Nzall indeed -- it can be a system of springs, rather than just a single spring. In addition, most of the systems explored in the literature are mainsprings (like in a watch), not coil springs.
– LShaver
Commented Jan 11, 2021 at 14:44
• In some sense, lithium-ion batteries and lithium-polymer batteries can be thought of as "spring-based" storage systems. They rely on intercalation of the lithium ions between the layers of some other substrate, which has the effect of pushing the substrate layers apart when the battery is charged. This places the substrate under strain, storing (at least some of) the energy as elastic energy. Commented Jan 11, 2021 at 20:56
• +1. Great answer! It's also important to point out that converting electric energy into the potential energy of a spring system would involve much more losses than storing that same energy into the electro-chemical energy of a battery. The same problem regarding the loss will persist when you use the energy stored in the springs.
– ACat
Commented Jan 11, 2021 at 23:23
• A few years ago I worked through the possibility of energy storage using the structural changes in granitic minerals as they were compressed. I can't remember the detail, but the figures were astoundingly unimpressive. Commented Jan 12, 2021 at 21:10

This is a question that I've heard several times, though it is the first time I've seen it here. The main problem is, I think, efficiency.

Firstly, though, not long ago, when I was a child (OK, quite a few decades ago, but in living memory), energy storage in springs was a common way of powering clocks and watches. Energy stored in weights was used for clocks by my grandfather's generation. My grandfather still had one, and would wind the weights up each day to run the clock for the next 24 hours .. one weight for the time keeping mechanism and one for the chimes.

So much for memory lane.

Suppose you had a 1 tonne weight suspended somewhere in your home, and raised in 5m (assuming your house has 5m vertical extent). The energy stored is weight * gravitational forces * distance = 1000 kg * 10 m s-2 * 5m = 50kJ = 0.013 kWh.

This amount of energy is, unfortunately, not going to last the modern consumer very long. The typical household usage in the UK, for instance, is 8kWh per day, so that 1 tonne weight gives enough energy for less than 3 minutes. Depressing.

• That's why I added the springs, they store energy at the square of the extension
– w00t
Commented Jan 11, 2021 at 5:30
• On the other hand, having your house run on wind-up energy for 3 minutes at a time would pretty darn awesome. Remind me to build one when I'm rich and famous and properly educated in electrical engineering. Commented Jan 11, 2021 at 15:48
• @w00t : OK .. I should have dealt with springs as well LShaver has answered that fully now. Commented Jan 11, 2021 at 18:29
• @user253751: I did invest in a wind up radio a few years ago. For a while it was OK ... 5 minutes winding for an hour of play back. After a couple of years it was 10 minutes winding for 15 minutes play back, at which point it became another piece of waste electronics. Commented Jan 11, 2021 at 18:30

You can easily calculate the maximum strain energy in a material by considering the maximum stress (for metal: yield stress) and stiffness (Young's Modulus). The maximum strain energy in a material is then

E = 0.5 x yield stress ^ 2 / Young's Modulus.

If you take high quality steel with a yield stress of 1000MPa and a typical Young's modulus of 200GPa, you get a measly 0.7kWh/m³ of energy per cubic meter of steel. I would need 10m³ of steel just to power our home on a cloudy day, and apparently an average US residence would need 40m³ of steel. This assumes perfectly stressed steel, whereas a typical spring loaded in bending has quite some 'unused' material.

Note that steel is also quite heavy at 8000kg/m³, so we're talking 320 000 kg of steel stretched to its limit to power a US household for a day.

• Thanks, this is very helpful!
– w00t
Commented Jan 12, 2021 at 13:59

Not only is the energy density of such systems low, as other posts explained - stored energy that will be released as mechanical energy immediately in case the storage system fails has been shown to be extremely dangerous in practice. Even the example of a wind-up clock shows it - if you disassembled a fully wound old school alarm clock the wrong way, there would be a non-negligible chance to get hurt. Pressurized vessels are known for similar tendencies to fail catastrophically... While failure of an energy storage system with the energy released mostly thermally is bad enough, there appear to be more ways to safely contain some unwanted thermal energy than containing some unwanted projectiles.

• The most energy-intense spring that most people encounter are the torsion springs that help reduce the apparent weight of garage doors. Those springs are notoriously dangerous, and the energy they store is not more than the work to open a garage door. Commented Jan 12, 2021 at 8:46
• Mousetraps, Crossbows, Harpoon guns .... while these are intentionally dangerous devices, they "just" have a bit of energy stored in a spring of some description... Commented Jan 12, 2021 at 19:54

### There is - it just doesn't use metal springs

Metal springs have numerous failure modes. In tension, they only remain "springs" for a certain point, after which they simply become wires/rods (which also have a Young's modulus, but in a different way). In compression, they also only remain "springs" until the coils touch, at which point again you have a solid rod. Coiled springs of any type simply don't work well for this.

What does work well is any compressible liquid or gas. As anyone who's bounced a ball knows, compressible gases are a near-perfect spring (and by extension liquids too; they just generally need more force to compress). Gases and liquids can also be compressed over an extremely wide range of pressures, only limited by the pressure/volume at which a phase change (gas<->liquid or liquid<->solid) starts to take place. An "accumulator" pressure vessel is pumped up to pressure for energy storage, and energy release is simply handled by letting the gas/liquid flow out through a turbine.

This is one type of system that has been used for kinetic energy recovery systems in cars. As per Wikipedia on KERS, Bosch and PSA developed a hydraulic KERS system for road cars in 2015. Some places are still looking into the concept.

The problem is that this design has competition from other methods of energy storage which are generally preferable. A pressure-storage system inherently needs two tanks, one for the liquid under pressure and one for the liquid not under pressure, so for starters you've made it larger. And a high-pressure storage tank requires significant effort in construction otherwise you've essentially created a very powerful fragmentation bomb (which of course has been why hydrogen-powered vehicles took so long to come around).

For mechanical energy storage, flywheels generally give higher energy density for smaller applications like cars; and on a larger scale, gravity storage (pumped-hydro) schemes give you scaleability with relatively low cost. Pressure storage can't generally compete with either (although you could think of pumped-hydro as a very large pressure-storage system running at very low pressures).

And that's just mechanical. Electrical storage is also popular, of course. For small applications, KERS systems today generally use supercapacitors because they give very high energy density with the added bonus of being able to control the energy output very accurately; and longer-term electrical storage of course uses batteries for the same reason. There are currently not many options for larger-scale electrical storage, but flow batteries have been discussed conceptually for a while, and the technology is starting to come around. This is especially relevant in the context of renewable energy sources, of course, where there is a definite need to store energy whilst the renewable source is not generating.

TLDR: Yes, you can do it, and people have tried it. And people haven't then taken it further, for reasons.