As most readers will know, the Iberian Peninsula transmission grid covering Spain and Portugal suffered a severe failure shortly after 12.30 CEST on Monday April 28th. Nearly 24 hours later power supplies have been restored to most customers in the peninsula, but full recovery to the
Thank you Gordon. A few days ago there was an article in the Portuguese quality news "paper" Publico quoting a source within the Spanish system operator Red Eléctrica de Espanha (REE) warning a few months ago of the risks of blackouts. More recently a Spanish minister declared the electricity network "safe" and that there would not be major outages.
I am eagerly awaiting a similar political reassurance here, just as we were told Covid wouldn't be that big a deal in Britain as the Italian health system was capsizing.
Politico Europe quoted Spanish prime minister Pedro Sanchez as stating "Freak disappearance of electricity triggered power cut". With the likes of him in charge, what does anyone have to fear?
The picture I am slowly assembling, with your help, is that in attempting to operate electricity grids with a large penetration of wind and solar, we are now off the map.
As you say, the exact cause of the breakdown is probably unimportant relative to the need for the system to be adequately resilient to any sudden shock. As it happens, I may have stumbled across the actual cause via a chance link to a post from SpaceWeatherNews. In a 90-minute YouTube, the reporter gives evidence that the Earth’s weakening magnetic field essentially failed, collapsing for a brief second and giving Iberia a blast of solar energy comparable to a massive flare, even though the Sun was mostly quiet: https://www.youtube.com/live/TYuPQ2NbkwA.
I know next to nothing about what the reporter was talking about but he was animated that this was a big thing and was very obvious to space weather catastrophists. We know that at 12:33pm the national solar supply suddenly fell by about half which triggered the cascade, see https://notalotofpeopleknowthat.wordpress.com/wp-content/uploads/2025/04/image-69.png. Could the flare have caused a spike (or even an overload slump) in solar power supply and could that have triggered the shutdown? Could it happen again at any time?
I have no idea or expertise in judging whether such an event occurred. I think that the crucial observation is that a frequency drop of 0.15 Hz should not, in any reasonably managed system, result in a cascading loss of grid supply. Systems are - or should be - designed to cope with extreme weather events and other random factors.
Thank you Gordon, I was looking forward to your insights on this.
Couple of follow-up questions:
1. As you mentioned nuclear may not be most suitable to respond to sudden changes in demand, but is there a case to be made for higher proportion of nuclear in the energy mix to improve the baseline grid stability? Since nuclear are notoriously expensive to build, my guess would be that the economics on this would still show that the cost of building them will be greater than the savings from improved grid stability... I would be also interested to hear your opinion on the smaller nuclear plants that are being researched and developed - does that look to you like a promising technology?
2. What are the key features of the UK electricity grid allowing it to operate within larger frequency range? Is it larger proportion of gas/less proportion of solar? How much of it is the inverters themselves or different frequency response services?
3. Given the costs of the outages such as this one, I can imagine there will be increased pressure politically to avoid them even if that means abandoning/delaying the Net Zero commitments - Europeans may say they really care about the environment, but I doubt they care about it more than they care about having electricity at their disposal. Therefore, is it generally better to think about restarting phased out gas plants or building new ones? Have the efficiency gains in modern CCGTs been sufficient to justify that sort of investment?
Third, about the lessons learned. I am not sure how Europe will respond. If US experience is anything to go by, the response will be to claim "special circumstances", etc - i.e. nothing to learn. The resistance to drawing core lessons from what happened in Texas in 2021 was extremely strong with everyone blaming everyone else. And, of course, there is rarely a smoking gun as system collapses depend on a complex combination of factors.
The obvious solution to assure more resilient grids would be to maintain a minimum level of generation (as a proportion of total load) from high inertia sources. That would primarily be CCGTs (or coal in Germany and a few other countries). However, that would require a redesign of market systems and dispatch arrangements. This, I think, is the central issue. Policymakers who know nothing in detail about electricity systems think that you can dump large amounts of zero marginal cost but intermittent generation into market systems which evolved under completely different circumstances. That is never going to work in the long run and, in Britain, is being made worse by redesigning arrangements to suit intermittent generators while completely ignoring resilience and backup.
I have little doubt that it will be necessary to build a lot of both CCGTs and GTs across Europe but I see no willingness to accept that payment systems and market incentives will have to change radically if that is to happen.
Second, about the UK grid. Its resilience over the last two decades has been very much a by-product of reliance on CCGTs. Gas turbines are of limited use in this context because they are not running most of them but they are essential for black starts - i.e. repowering and synchronizing the grid after a collapse.
There are also an increasing number of small reciprocating engines spread around the grid that are designed to provide frequency response services, but these would not have helped with a repetition of what happened in Spain because the instability was so sudden and unexpected. Most contracted frequency response units are designed to handle a more gradual decline in frequency due to a small number of generators going offline or reducing output (eg a sudden reduction in wind or solar output).
Grid collapses are very unusual. Variations in frequency of the type I refer to in my article are more common and it is understandable that system operators focus on ensuring that those can be managed without severe impacts on the network. I don't think anyone had really focused on handling a sequence of events similar to what happened in Spain. They will now!
Dusan - I will write separate comments on your questions. First, about nuclear. One would never build nuclear plants primarily on grounds of grid stability. I don't know what happened in Spain but synchronizing nuclear plants to a grid that has collapsed and has to be restarted is likely to have been a nightmare. I am not sure whether anyone knows if this would be easier or harder with SMRs. They have been designed with a different goal in mind - easy and efficient load-following so that their output can regularly be increased or reduced by up to 50%. They are basically thermal generators so they would provide inertia but I suspect that they have to operate within a narrow frequency range.
More generally the nuclear industry is its own worst enemy - or perhaps that applies to its regulators who mostly appear to hate nuclear generation. I think that large nuclear plants are mostly a dead-end except when built by China or Korea, which are able to control costs and keep construction times down to ~8 years. The prospects for SMRs are much better *provided* (a) they don't over-promise on costs, and (b) construction periods are no longer than 5 years. What most people don't appreciate is that the overwhelming advantage of CCGTs is that they can be built in 3 years, which not only keeps down costs but allows much greater flexibility in project planning and decision-making.
Further to the lesser-scale UK blackout you mention in 2019, it mirrored the Iberian blackout in that (i) the then UK politicians denied any suggestions that their precious renewables lacked resilience and (ii) just before the outage the cheerleaders for renewables had boasted of setting an all-time high for wind power supply.
“The recent power cuts which affected a million people and caused hours of disruption have done us a favour by highlighting how our efforts to decarbonise the UK economy are leading us into a hopeless energy cul-de-sac. The establishment is doing its utmost to gloss over the incident which was almost certainly due to system fragility caused by having too much non-synchronous wind power connected to the national grid. This problem also applies to solar power and interconnector imports. Shortly before the incident the climate change cheerleaders at National Grid boasted in a tweet that wind power was close to setting an all-time record.
The BEIS response on national grid fragility glosses over the severe doubts on the lack of inertia and poor grid resilience evidenced by the recent power cuts caused by already having too much non-synchronous generation (wind, solar and international interconnectors) connected to the grid. Engineering experts have been warning politicians for years that these inappropriate technologies are unsustainable. Recently an experienced power systems engineer warned that there should be a 30% upper limit on non-synchronous supply yet National Grid had about 50% connected on the day of the power cuts.”
In dealing with such matters politicians and civil servants are sock puppets. They voice words that are written by the PR people at NG (now NESO) whose expertise is equally low. The general problem is the complete lack of accountability. If it was clear that both the Chairman and Chief Executive of NG/NESO would be fired in such an event, there would be a great deal more effort devoted to ensuring such outages either didn't happen or were rapidly corrected. Further. senior staff would be more likely to go on record that the system was vulnerable because of decisions made by policymakers.
Sometimes, strict accountability is a double-edged sword but it is much better than an absence of any kind of accountability.
Miliband is aware of the grid inertia issue as he has talked about installing giant flywheels throughout the grid to emulate the spinning reserve of our now-unwanted synchronous plant. I’d been wondering how these would work and your final paragraph is helpful in suggesting that they are most likely Heath Robinson contraptions, if they ever come about. According to NESO Stability Services Network they won’t be deployed before 2029.
Thank you Gordon. A few days ago there was an article in the Portuguese quality news "paper" Publico quoting a source within the Spanish system operator Red Eléctrica de Espanha (REE) warning a few months ago of the risks of blackouts. More recently a Spanish minister declared the electricity network "safe" and that there would not be major outages.
I am eagerly awaiting a similar political reassurance here, just as we were told Covid wouldn't be that big a deal in Britain as the Italian health system was capsizing.
Politico Europe quoted Spanish prime minister Pedro Sanchez as stating "Freak disappearance of electricity triggered power cut". With the likes of him in charge, what does anyone have to fear?
The picture I am slowly assembling, with your help, is that in attempting to operate electricity grids with a large penetration of wind and solar, we are now off the map.
As you say, the exact cause of the breakdown is probably unimportant relative to the need for the system to be adequately resilient to any sudden shock. As it happens, I may have stumbled across the actual cause via a chance link to a post from SpaceWeatherNews. In a 90-minute YouTube, the reporter gives evidence that the Earth’s weakening magnetic field essentially failed, collapsing for a brief second and giving Iberia a blast of solar energy comparable to a massive flare, even though the Sun was mostly quiet: https://www.youtube.com/live/TYuPQ2NbkwA.
I know next to nothing about what the reporter was talking about but he was animated that this was a big thing and was very obvious to space weather catastrophists. We know that at 12:33pm the national solar supply suddenly fell by about half which triggered the cascade, see https://notalotofpeopleknowthat.wordpress.com/wp-content/uploads/2025/04/image-69.png. Could the flare have caused a spike (or even an overload slump) in solar power supply and could that have triggered the shutdown? Could it happen again at any time?
Solar flares and coronal mass ejections can cause disruption. One such event was that of March 1989 which caused big disruption in Quebec: https://en.wikipedia.org/wiki/March_1989_geomagnetic_storm
However, I didn't see any extraordinary solar activity on 28 April.
I have no idea or expertise in judging whether such an event occurred. I think that the crucial observation is that a frequency drop of 0.15 Hz should not, in any reasonably managed system, result in a cascading loss of grid supply. Systems are - or should be - designed to cope with extreme weather events and other random factors.
Thank you Gordon, I was looking forward to your insights on this.
Couple of follow-up questions:
1. As you mentioned nuclear may not be most suitable to respond to sudden changes in demand, but is there a case to be made for higher proportion of nuclear in the energy mix to improve the baseline grid stability? Since nuclear are notoriously expensive to build, my guess would be that the economics on this would still show that the cost of building them will be greater than the savings from improved grid stability... I would be also interested to hear your opinion on the smaller nuclear plants that are being researched and developed - does that look to you like a promising technology?
2. What are the key features of the UK electricity grid allowing it to operate within larger frequency range? Is it larger proportion of gas/less proportion of solar? How much of it is the inverters themselves or different frequency response services?
3. Given the costs of the outages such as this one, I can imagine there will be increased pressure politically to avoid them even if that means abandoning/delaying the Net Zero commitments - Europeans may say they really care about the environment, but I doubt they care about it more than they care about having electricity at their disposal. Therefore, is it generally better to think about restarting phased out gas plants or building new ones? Have the efficiency gains in modern CCGTs been sufficient to justify that sort of investment?
Third, about the lessons learned. I am not sure how Europe will respond. If US experience is anything to go by, the response will be to claim "special circumstances", etc - i.e. nothing to learn. The resistance to drawing core lessons from what happened in Texas in 2021 was extremely strong with everyone blaming everyone else. And, of course, there is rarely a smoking gun as system collapses depend on a complex combination of factors.
The obvious solution to assure more resilient grids would be to maintain a minimum level of generation (as a proportion of total load) from high inertia sources. That would primarily be CCGTs (or coal in Germany and a few other countries). However, that would require a redesign of market systems and dispatch arrangements. This, I think, is the central issue. Policymakers who know nothing in detail about electricity systems think that you can dump large amounts of zero marginal cost but intermittent generation into market systems which evolved under completely different circumstances. That is never going to work in the long run and, in Britain, is being made worse by redesigning arrangements to suit intermittent generators while completely ignoring resilience and backup.
I have little doubt that it will be necessary to build a lot of both CCGTs and GTs across Europe but I see no willingness to accept that payment systems and market incentives will have to change radically if that is to happen.
Second, about the UK grid. Its resilience over the last two decades has been very much a by-product of reliance on CCGTs. Gas turbines are of limited use in this context because they are not running most of them but they are essential for black starts - i.e. repowering and synchronizing the grid after a collapse.
There are also an increasing number of small reciprocating engines spread around the grid that are designed to provide frequency response services, but these would not have helped with a repetition of what happened in Spain because the instability was so sudden and unexpected. Most contracted frequency response units are designed to handle a more gradual decline in frequency due to a small number of generators going offline or reducing output (eg a sudden reduction in wind or solar output).
Grid collapses are very unusual. Variations in frequency of the type I refer to in my article are more common and it is understandable that system operators focus on ensuring that those can be managed without severe impacts on the network. I don't think anyone had really focused on handling a sequence of events similar to what happened in Spain. They will now!
Dusan - I will write separate comments on your questions. First, about nuclear. One would never build nuclear plants primarily on grounds of grid stability. I don't know what happened in Spain but synchronizing nuclear plants to a grid that has collapsed and has to be restarted is likely to have been a nightmare. I am not sure whether anyone knows if this would be easier or harder with SMRs. They have been designed with a different goal in mind - easy and efficient load-following so that their output can regularly be increased or reduced by up to 50%. They are basically thermal generators so they would provide inertia but I suspect that they have to operate within a narrow frequency range.
More generally the nuclear industry is its own worst enemy - or perhaps that applies to its regulators who mostly appear to hate nuclear generation. I think that large nuclear plants are mostly a dead-end except when built by China or Korea, which are able to control costs and keep construction times down to ~8 years. The prospects for SMRs are much better *provided* (a) they don't over-promise on costs, and (b) construction periods are no longer than 5 years. What most people don't appreciate is that the overwhelming advantage of CCGTs is that they can be built in 3 years, which not only keeps down costs but allows much greater flexibility in project planning and decision-making.
Further to the lesser-scale UK blackout you mention in 2019, it mirrored the Iberian blackout in that (i) the then UK politicians denied any suggestions that their precious renewables lacked resilience and (ii) just before the outage the cheerleaders for renewables had boasted of setting an all-time high for wind power supply.
I captured details of the incident and referred to the importance of grid inertia in an online email to then Business Secretary Anthea Leadsom: https://edmhdotme.wpcomstaging.com/the-case-against-net-zero-co2-emissions-2/
“The recent power cuts which affected a million people and caused hours of disruption have done us a favour by highlighting how our efforts to decarbonise the UK economy are leading us into a hopeless energy cul-de-sac. The establishment is doing its utmost to gloss over the incident which was almost certainly due to system fragility caused by having too much non-synchronous wind power connected to the national grid. This problem also applies to solar power and interconnector imports. Shortly before the incident the climate change cheerleaders at National Grid boasted in a tweet that wind power was close to setting an all-time record.
The BEIS response on national grid fragility glosses over the severe doubts on the lack of inertia and poor grid resilience evidenced by the recent power cuts caused by already having too much non-synchronous generation (wind, solar and international interconnectors) connected to the grid. Engineering experts have been warning politicians for years that these inappropriate technologies are unsustainable. Recently an experienced power systems engineer warned that there should be a 30% upper limit on non-synchronous supply yet National Grid had about 50% connected on the day of the power cuts.”
In dealing with such matters politicians and civil servants are sock puppets. They voice words that are written by the PR people at NG (now NESO) whose expertise is equally low. The general problem is the complete lack of accountability. If it was clear that both the Chairman and Chief Executive of NG/NESO would be fired in such an event, there would be a great deal more effort devoted to ensuring such outages either didn't happen or were rapidly corrected. Further. senior staff would be more likely to go on record that the system was vulnerable because of decisions made by policymakers.
Sometimes, strict accountability is a double-edged sword but it is much better than an absence of any kind of accountability.
Miliband is aware of the grid inertia issue as he has talked about installing giant flywheels throughout the grid to emulate the spinning reserve of our now-unwanted synchronous plant. I’d been wondering how these would work and your final paragraph is helpful in suggesting that they are most likely Heath Robinson contraptions, if they ever come about. According to NESO Stability Services Network they won’t be deployed before 2029.