Spanish blackout of 28 April 2025

[IvanV]

It was quite unclear at the time, or in the following period, why the blackout occurred. We now have an official report from Red Eléctrica. They essentially admit that they were chancing it already, and so unable to cope when coming into contact with additional cockups.

The following useful summary should be credited to a former power systems engineer called Simon Gallagher, and came to me second hand:

1. On the day, the system did not have enough large synchronous generators online to control voltage. This was due to some thermal plants being unavailable - the TSO [Transmission Systems Operator] did not replace them with alternative voltage-controlling units. Over 80% of the generation was inverter based.

2. Earlier in the morning, before the blackout, the system experienced multiple oscillations. The TSO took lots of actions to dampen these (including meshing x10 400 kV circuits, switching the DC interconnectors with France to constant power mode, and disconnecting shunt reactors). These actions successfully stopped the oscillations - but, and it is a big but - they also caused voltage to rise and made the system much less flexible in managing it. With over 82% of generation now inverter-based, and few synchronous units available, the system became ‘stiff’ to voltage control. The stage was set.

3. The first generators tripped on overvoltage - even though voltages were still within the grid code limits. This was wrong settings or insufficient ride-through capability. Questions here on compliance.

4. From there, it was all downhill. The system lacked enough voltage control resources - and those still connected did not behave as they should. Generators that should have been absorbing reactive power were not doing so sufficiently, and some were even injecting reactive power. This led to increasing voltage rise, triggering even more generator trips until the system collapsed entirely.

We have previously discovered in Britain that over-sensitive trip-out settings can turn what ought to be small incidents into major blackouts. It was a major factor in that large blackout that occurred in eastern England a little while ago. In that incident, an initial modestly-sized problem caused a large off-shore windfarm to trip out, which made the overall incident too large to handle. Its trip-out settings were too sensitive. It was discovered that over-sensitive trip-out settings are widespread.

[Grumble]

Is there any particular reason that inverters can’t perform the task that synchronous condensers currently do?

[IvanV]

Is there any particular reason that inverters can’t perform the task that synchronous condensers currently do?

There are now inverters called grid-forming inverters, which can provide "synthetic inertia" for the grid. But the power resource behind the inverter needs to have some headroom for them to be able to do this. Because you need to be able to turn it up quickly. So they tend to be attached to batteries rather than to solar and wind resources, because it is the whole point of batteries to have headroom. Whereas it is typically the point of solar and wind to max out its export generation. But getting headroom from such generators does not necessarily mean the electricity is wasted - they could be trickling that headroom power into a battery, or some other optional on-site use. But clearly they would want some compensation for such arrangements, whereas traditional grid inertia comes for free.

Interconnectors is another issue. Synchronised AC interconnectors automatically provide inertia and for free. With a DC interconnector, I suppose again you have the option of a grid-forming inverter. But you'd have to arrange for it to be turned up fast, and that would probably involve dealing with a longer supply chain than local inverter resources.

[Grumble]

Analysis of the report from a US perspective:
https://gridstrategiesllc.com/project/a ... solutions/
Talks about whether or not the US grid could experience the same problems, points out that the US has required renewables to provide voltage regulation for years now whereas in Spain this is forbidden. This rule needs to change in Spain

[IvanV]

Analysis of the report from a US perspective:
https://gridstrategiesllc.com/project/a ... solutions/
Talks about whether or not the US grid could experience the same problems, points out that the US has required renewables to provide voltage regulation for years now whereas in Spain this is forbidden. This rule needs to change in Spain

So, apologies, my previous post was misconceived. I was assuming it was the same as the usual difficult issue of inertia. That is when voltage falls away due to lack of frequency control. But in fact voltage control is also about reactive power, and that was the main issue on this occasion. It is straightforward for inverter-based resources to adjust reactive power without the need for special inverters and headroom.