This series of blogs is about electricity storage and how it works with renewable energy to create net-zero electricity grids.

Last week I talked about using weights and gravity to store electricity. The most popular method is pumped hydro storage. Pump water uphill into a reservoir when you have spare electricity and let it run down through a turbine when you want the electricity back. This accounts for about 95% of global electricity storage, a very useful 9,000 GWh. There are other methods in development that use the idea of lifting weights in different ways.

Witt Energy electricity generator using turbulent flow in water.

Witt Energy Electricity Generator using Turbulent Flow in Water

Other mechanical devices can also be used for storage. I am not talking about clever little units like the Witt Energy generator. This can convert motion in any of six axes into useful power and has potential applications in getting electricity out of turbulent water flows – think sensors and navigation buoys. No, I mean systems that can be plugged into an electricity grid and provide useful storage on any timescale from seconds to days and months.

The two technologies that have produced the biggest interest are compressed air storage and flywheels.

Spring is in the air

A useful place to store energy is a spring. Not the coiled metal spring you find in a mechanical alarm clock, but in compressed gas.

Diagram of a compressed air energy storage system

Compressed Air Energy Storage System (Storelectric)

The idea is simple. Use electricity to power compressors to force air into a reservoir at high pressure. Once there, it can be a long-term energy reserve. To recover the electricity, you allow the high-pressure gas to escape through an air turbine driving a generator. Compressed air does not store a huge amount of energy per unit volume, so these Compressed Air Energy Storage systems (CAES) use a large underground reservoir, such as a salt cavern or an exhausted oil or gas well.

So far, so simple. However, there is a problem. If you have ever pumped up a bicycle tyre, you know that when you compress air it gets hot. When you release the pressure, the reverse happens, and the air cools. In a CAES system, the temperature swings can be extreme, so you must cool the gas as it is compressed and warm it as it expands. These simple CAES systems lose a lot of energy on the round-trip and are only about 30% – 40% efficient. Even worse, the simplest way to heat the gas as it expands is to use fossil fuels. Contributing to the problem of CO2 emissions you are trying to solve.

There is a way around this. Save the waste heat generated by compression and use it to warm the gas when it expands. These Advanced Adiabatic Compressed Air Energy Storage systems (AA-CAES) should be able to achieve an 80% round-trip efficiency. The problem is how and where to store the heat until you need it? Storing heat without losses is a bigger problem than storing gas at pressure.

So, where do we stand today? We have over forty years of practical commercial experience in CAES. The Huntorf plant in Germany started in 1978. It stores compressed air in two caverns 600 m below the surface carved out of a salt deposit using hot water. The total volume is 310,000 m3, and when pressurised to 100 bar can deliver 321 MW of electricity for 2 hours.

Alabama is home to the other commercial plant. The McIntosh plant started in 1991, and also uses a cavern water-mined in a salt deposit. It has 110 MW output and a storage capacity of 2.86 GWh.

Both these sites use conventional CAES with fossil fuel heating of the expanding gas. Despite successful operation for many years, we have not seen a rush of other CAES projects. We need to see a solution to the heat storage problem of AA-CAES. That would give round-trip efficiencies close to pumped hydro and battery storage. Many demonstrators and prototypes are proposed, but they always seem to get delayed or shelved, or the promoters lose interest.

The UK company Storelectric has suggested using hydrogen to heat the expanding gas. Keeping the simplicity of conventional CAES but without the CO2 emissions. Unfortunately, without ready supplies of green hydrogen, this won’t deliver. And if you have the electricity to make the green hydrogen, why use it to heat a CAES plant?

This is a shame because CAES would be very attractive, with higher efficiencies and no emissions:

  • We have lots of suitable underground chambers or places where we can create them. Finding a site will be easier than for pumped hydro
  • Lower civil engineering costs than pumped hydro
  • A higher power density in W/l than pumped hydro
  • Cost per MWh not too far from pumped hydro, and a similar start up and discharge time

Work continues, and many variants of CAES are in research and development. But none have cracked the efficiency and scale challenges. As a result, the story of compressed air energy storage is of promise, not yet fulfilled.

Spinning tops

Flywheels are the other mechanical storage technology deployed at grid scale. Again, a straightforward idea. We know that once we get a wheel spinning it takes energy to slow it down, because a spinning mass has momentum. So, if we use an electric motor to spin up a disk, we can use electromagnetic braking to slow the disk down and get electricity back. Exactly like the regenerative braking on a battery electric vehicle.

Just like CAES, the idea is simple, but challenging to apply in practice. The energy stored in a flywheel is linearly proportional to the mass, but increases with the square of the radius and the square of the angular speed (rpm). To store a lot of energy, you need a large radius (with the mass concentrated in the rim), fast rotation, or both.

This is a hard engineering challenge. First, you need low-loss bearings to stop the flywheel slowing up. Air resistance is a problem, so you may want to run the flywheel in a vacuum. And worst of all, if you spin a flywheel fast, centrifugal force tries to tear it apart. So, you can’t make a high-speed rotor out of steel; you need a much stronger composite material.

Although a practical flywheel system cannot store as much energy as other methods, it has some real advantages. It has a high energy density, high efficiency (92% round-trip), and a response time as low as milliseconds. That makes it ideal for grid regulation duties.

Cutaway image of a Beacon Power high speed flywheel storage unit showing casing, vacuum chamber, flywheel, hub and bearings, magnetic suspension and motor.

Beacon Power 100kW, 25kWh Flywheel Storage Module

Beacon Power is one company that makes commercial systems for grid regulation. The standard module provides 100 kW peak power and 25 kWh storage with a rotor operating at a peak speed of 16,000 rpm. Magnetic levitation bearings are used to reduce losses and the rotor spins in a vacuum. It is a modular system, and they can link multiple rotor units to provide more storage. Beacon Power has two commercial plants, one in Hazle, Pennsylvania, and the other in Stephentown, New York. Both use 200 flywheel units to give a maximum power of 20 MW and 5 MWh capacity. Each flywheel unit is placed in a covered pit to contain debris, should there be a catastrophic failure.

Beacon is not the only company in this space. Amber Kinetics, ABB, and Hitachi have all deployed flywheel storage to support grid regulation and integration of renewables. And not all flywheels run at the blistering speeds used by Beacon. S4 Energy’s Kinext system uses a 2.6 m diameter steel flywheel weighing 5,000 kg and spinning at a much slower rate of 1,800 rpm.

Since flywheel storage is a simple idea, the research and development pipeline is about improving scope and useability. Increasing the energy density, strengthening the rotors, reducing losses, using superconductors and improving control systems. Flywheel storage will never hold large amounts of energy over long periods, so there is interest in hybrid systems. These combine flywheels for fast response, deep discharge, and rapid cycle times with other energy storage systems.

Almelo hybrid storage system in the Netherlands. Showing 4 Kinext flywheel units and the battery farm

S4 Energy Hybrid Battery-Flywheel Storage System at Almelo,Netherlands

Just outside Amsterdam in the Netherlands is a 9 MW hybrid storage system that combines six Kinext flywheels with a large battery. It provides grid stability and makes more efficient use of a local wind farm. By buffering the output from the turbines, you don’t need to curtail the wind farm on windy days, and you get more electricity for your investment.

Providing grid inertia to prevent fluctuations in frequency is another application of flywheels. Traditionally, the large generators at big power plants provided the inertia. Enormous pieces of spinning metal that could absorb and flatten out changes in grid voltage and frequency. Renewable resources can’t do that, so a solution is a synchronous condenser, a large electric motor driving a heavy flywheel at relatively low speeds. The condenser constantly absorbs and releases power to keep the frequency steady.

Compressed air energy storage is good for storing large amounts of energy for a long time with moderately fast release. Flywheels are better for smaller amounts of storage but with quick delivery. Both can contribute to a decarbonised grid.

Next up – Thermal storage.

Previous parts of the energy storage story:

Electricity Storage Options 3 – Springs and Tops
Tagged on:         

2 thoughts on “Electricity Storage Options 3 – Springs and Tops

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.