17 Comments
Oct 13Liked by Oliver Johnson

As an engineer I am delighted to see mathematicians finally discovering the correct value of pi.

Expand full comment
author

Absolutely. Plenty good enough for these purposes at least

Expand full comment
Oct 14Liked by Oliver Johnson

Hi Oliver, it shouldn't come as too much of a shock that the Daily Telegraph have misunderstood the proposal.

The plan to build flywheels is part of NESO's stability pathfinder project.

The problem they are designed to solve is not an energy storage problem (although clearly a rotating flywheel does store energy).

The electricity network, as currently congured, relies on their being a reasonable number of large spinning turbines on the system, the synchronous generators. These all spin essentially at the same frquency of 50Hz. If one slows down, or starts to speed uo, there's a restorative force (technically a torque) from the inertia of all the other synchronous generators.

This has two important effects.

First, if generation or demand is suddenly lost, it takes time for the system frequency to change. The more inertia there is, the slower the rate if change. This provides stability from sudden shocks to the system.

Secondly, the effect of the inertia is to synchronise the whole country, so it's all running at 50Hz and in phase. If different geographic regions run at different frequencies or in different phase, it is impossible to transfer electrical power from one region to the next, and the resilience you get from having a synchronised grid is eroded.

On the current system, you get the inertia that performs this for free. It's a side product of the generation of power. But in a system with far fewer synchronous generators (just the nukes and a bit of hydro once the gas units are gone), you need to get that inertia from somewhere else to preserve stability, and that's going to be the job of these flywheels, alonside old gas units that will "work in reverse" with the grid power spinning the old turbine, rather than the turbine generating power. Such technology is usually referred to as Synchronous Compensation.

A description of one project can be found here: https://www.neso.energy/news/latest-boost-stability-pathfinder-construction-flywheel-begins

Expand full comment
author

Hi, I don't think that the Telegraph have necessarily misunderstood the proposal, and I agree that there are other benefits, but I just wanted to do the sums to explain why this wasn't a feasible means of storage at the necessary scale, in case anyone was thinking that. (Maybe they weren't)

Expand full comment

Why only one flywheel? One photovoltaic cell, one wind turbine, one heat pump, one coal mine, one oil/gas field, one nuclear reactor would never be enough either.

Expand full comment
author

Sure, but personally I find it useful to visualise just how much stuff we would need to move fast for this to be a viable solution, and to compare the capabilities of the systems which do exist.

Expand full comment
Oct 13Liked by Oliver Johnson

Very interesting - thanks for continuing to prod away at this issue.

Expand full comment
author

Thanks!

Expand full comment
Oct 13Liked by Oliver Johnson

Will be particularly interested when you get to compressed air (unless I’ve missed your coverage elsewhere).

Expand full comment
author

No, I haven't written anything about that (or hydrogen storage, or others). Don't feel completely qualified to at the moment, to be honest, but I will try to find time to read up about it at some stage

Expand full comment

I’m really not sure I understand the point of treating intermittency as a problem that must be solved by looking at single solutions in turn. The correct approach is a systems approach, and that’s what’s being done. It comprises:

1. Heterogenous supply mix, ie more than just wind and solar. Other renewables include tidal, geothermal, wave, etc. Other low carbon includes nuclear and gas&CCUS

2. Heterogenous placement of wind and solar, ie floating offshore where the wind is more constant, building out SW of the UK to complement the North Sea wind farms, etc

3. Interconnectors, both intracountry and intercountry, eg HVDC to Morocco

4. A mix of short-, medium- and long-duration storage including distributed storage via EVs and fixed on-site batteries, plus pumped, plus a bit of flywheel, plus other chemistries eg flow, plus compressed air, thermal, and even god save us green hydrogen

5. Demand flex

6. Overbuild

There are just not that many dunkelflautes to worry about, and while the last 5% of net zero is orders of magnitude harder than the first 95%, the first 95% gets us massively far forward. We should be so lucky to be in a position to have to worry about the costs of maintaining gas peakers for just the Dunkelflautes because we’ve sorted the build-out for the energy system for the rest of the time.

All a very long winded way of saying: “don’t let the perfect be the enemy of the good”

Expand full comment
author

Thanks for dropping by to tell me that. All I can say is that I personally find it interesting to do these calculations, and the people who've chosen to subscribe to my stuff seem to tolerate it as well, so I'm happy to continue to write for those people.

Expand full comment

I’m not trying to be a dick. I’m saying that I think readers may infer your message is “no technology is good enough to support the move to net zero”. If that’s what you intend, then I disagree with that inference for the reasons stated. And if not, then I think it’s a shame if people do infer that. Or perhaps I’m wrong, and no one will infer this.

But on a broader point: Substack is a fairly public forum. I didn’t stop by to message you, in particular, on your post, but to make a comment in response to your article, aimed at other readers. I realise you might not like my comment, or disagree with it, but isn’t that just the nature of being on Substack, that you’ll get comments from people that you disagree with? I did try to write my comment in a way that didn’t act as any kind of personal attack, but instead was a disagreement with the approach you’d set out in your post, and set out in a bit of detail the reasons why I think we can in fact get close to net zero for electricity generation.

Expand full comment

Whoops, forgot 7. Efficiency, eg insulation etc

Expand full comment

Another way of phrasing this: we used to punch ourselves in the nuts every hour, on the hour, every day of the year. Now we still punch ourselves in the nuts, but it’s more like once a week. We aspire to a world where we don’t punch ourselves in the nuts at all, but it turns out to be quite hard to do, and in the meantime, if we can get to a position where we’re only punching ourselves in the nuts for the two weeks after Xmas, that’s much much better than the nut-punching situation in which we find ourselves today. We should do what we can at the moment and for the next few years, and figure out what we do about the final two weeks of nut-punching along the way.

Expand full comment

As I learnt on my visit to Cruachan pumped storage power station in Scotland last year, they have the current contract to use their turbines (flywheels) to keep the grid at 50Hz. I think similar to what you mentioned at the end of the article. By the way love these sense checks on the challenge of moving to renewables.

Expand full comment

The reason for the people suggest flywheels for grid frequency control is that is effectively what we use now. The rotating intertia in the turbo machinery at thermal powerplants keeps the grid stable during periods of increasing or decreasing demand (generation tends to lag but enough that it would matter).

Though that isn’t enough, on its own, when you loose a big ‘generator’. I believe the largest single ‘generator’ we have right now is the 1.4 GW North Sea Link to Norway. The Danish link will be that size soon and the 2GW IFA link to France is really 2x1 GW. The NSL went down last week and the system responded quite well. The frequency only dropped to 49.56 HZ, as the grid scale battery storage kicked in. This is much better than what happened in 2019, where a similar loss of generating capacity resulted in load shedding (it ended up larger but at 1.4GW down the the frequency went below the lower limit).

Expand full comment