Fly, you fools!

You spin me right round, baby

Question: what do you get if you fix twenty four aircraft carriers to the side of a skyscraper?

As you might remember, I’ve been trying to make sense of the UK Government’s ambition to decarbonise our electric grid by 2030. Because the wind doesn’t always blow and the sun doesn’t always shine, a key part of this needs to be energy storage, to overcome problems of intermittent supply. A few weeks ago I described the limits of pumped storage schemes, by imagining pumping the entirety of Loch Ness up to the top of Ben Nevis. This helped us see that even putting up there a cube of water 500m on each side (about 40 football pitches in area, filled to a height greater than the highest tower in Europe) would only cover about a day’s worth of the UK’s electricity needs, not enough to cover long windless periods in winter.

That wasn’t encouraging. However, it’s not the only game in town. While that sort of scheme looks to convert electrical energy from windmills and solar panels into potential energy, other forms of energy also exist. In particular, if we take something heavy and make it move, then we know from our physics lessons that it will gain kinetic energy, so this can be a way to store a glut of courgettes renewable electricity. Indeed, the Daily Telegraph has suggested something like this might be on the cards:

Ed Miliband reveals plan to prevent net zero blackouts

Giant metal ‘flywheels’ to be installed across Britain to help stabilise electricity grid

so in the interest of fairness I’d like to repeat my kind of previous calculations, to see what flywheel systems can offer us in terms of storage capacity.

As with pumped storage systems, it’s not controversial that we can store energy this way. These kinds of ideas have been used for decades to try to make vehicles more efficient, including public transport systems around the world.

Starting in 2009, some Formula 1 cars were equipped with KERS (Kinetic Energy Recovery Systems) technology, which made perfect sense in a sport where pole position can be decided by thousandths of a second over a minute-long lap. When a car brakes into a corner, instead of converting its kinetic energy into heat dissipating into the atmosphere, we could try to capture some of it to spin up a flywheel in the car. Then, when it accelerates out of the corner, if we can get that energy back from the flywheel in the next few seconds, we can pull away faster than before and go on to win the race.

But can these kinds of system store enough energy to “prevent blackouts” as the Telegraph headline suggests? Well, we can try to repeat the same sorts of calculations that we did before.

By Dronepicr - Own work, CC BY-SA 4.0, https://commons.wikimedia.org/w/index.php?curid=123892891

Lets imagine the biggest flywheel we can. Suppose we mount an axle at the top of the Shard in London, stick a metal disc to it, and spin it around. We can fit a disc of 300 metres radius to this axle without it scraping the ground. We can be ambitious and make our disc out of solid steel a metre thick, and spin it at 100 revolutions per minute. What would that give us?

The volume of the disc would be pi times the radius squared times the thickness. Pi is 3, the radius squared is 100,000, and so the volume is about 300,000 cubic metres.¹ A cubic metre of steel weighs about eight tonnes, so in total this weighs about 2.4 million tonnes, or about the same as 24 Gerald Ford class aircraft carriers.

The maths gets a bit complicated to work out how much energy we can store this way, but thankfully there is an online calculator which I can plug the numbers into. If I’ve done that right, I think we can only store about 2TWh of energy, or a couple of days worth of energy supply, in this gargantuan system.

Again, this isn’t really encouraging. Even if we imagine that we could reinforce the foundations of the Shard to stop it toppling over with the weight, we have an enormous disc spinning fast in the centre of our capital city.

100 revolutions per minute doesn’t sound like a lot, but the circumference of the disc is about 2000 metres, so points on the rim are travelling at about 3000 metres/second, or around nine times the speed of sound. We’re going to have to think about sonic booms, and vibration. There’s going to be wind resistance from the outside of the disc, and friction on the axle. We’re going to have to worry about the integrity of the material, because if it shatters due to these physical pressures then it will fire pointed lumps of metal at high speed into some expensive real estate in the home counties. To manufacture this thing in the first place we’re going to need nearly a months’ worth of Europe’s entire steel production capability. It doesn’t seem like good news.

Realistically then, we’re not going to be able to build flywheel systems on anything like this scale. Wikipedia tells me about a system in California which can store 80MWh of energy, for example. Such flywheels are a few metres in size, can be constructed out of composite materials and spin in a vacuum chamber to reduce air resistance. 80MWh sounds great, but remember that there’s a thousandfold difference between a MWh and a GWh, and a further thousandfold between a GWh and a TWh, so in practice we are talking about one of these systems being able to store about seven seconds worth of the UK’s electricity needs.

In fact, reading the Telegraph article, the use of flywheels in the UK seems more likely to smooth out local fluctuations in supply over periods of a few seconds, to be more like the Formula 1 KERS systems than our giant Shard system. That can be very handy, for example by filling in the gaps between gusts on a wind turbine, but it still doesn’t help with the vital problem of long-term storage, so it’s still not clear to me what the plan is there. Still, I guess we’ve got six years to figure that out still!

Update: I’m still very busy with my new job. There are lots of aspects of it I can’t talk about here, but I can tell you that I had a nice trip to London on Thursday. This was for a steering group meeting and a dinner at the Royal Society with the new Scholars from the Martingale Foundation, an organisation committed to funding MSc and PhD studentships (including in Bristol) to broaden the pool of UK researchers. So please spread the word about the current call for studentships. I also enjoyed Ludwig on BBC iplayer, and managed to run 10k about four minutes faster than I had in a long time, as well as being increasingly grumpy on Twitter.

1. Ok, pi isn’t 3 and so on. If you do it properly you get about 280,000 cubic metres.