Update | Root causes of Iberian blackout
Figure 2. Three major power plant tripping events in south-west Spain. Source: ENTSO-E
without a ramping restriction, or a disconnection of the unit for unknown reasons.
In the same period, there was an increase in net load in the distribution grids of approximately 317 MW, which can be attributed partly to disconnections of small embedded generators < 1 MW (mainly rooftop PV) and partly to an increase in voltage-dependent load as a result of a voltage rise. The exact reasons for some of these events are not known.
From around 12:32:20, a particularly steep rise in local voltage occurred at a substation in the Granada region (corresponding to the black plot shown in Figure 1). Some milliseconds after 12:32:57, a 400/220 kV transformer in the same substation tripped due to the activation of an overvoltage protection on the 220 kV side of the transformer, which connects several generation facilities (photovoltaic, wind and solar-thermal (CSP)) to the transmission grid. Just before the disconnection, the transformer was injecting 355 MW into the grid and absorbing 165 Mvar of reactive energy, and the voltage in the 400 kV part of the grid was 417.9 kV. This event created an additional step change in voltage (also shown in Figure 1).
The next event consisted of two sets of trips, resulting in an additional loss of 727 MW of PV and CSP facilities connected to two 400 kV transmission substations in the area of Badajoz. At the first substation, an evacuation line tripped at 12:33:16.443 due to the activation of the overvoltage protection. The voltage level at the time of this trip is estimated at 432.4 kV via indirect calculations due to the lack of high- resolution voltage data at the connection point. In the second substation, the trip occurred at 12:33:16.820 due to overvoltage protection. After that, several trips between 12:33:17 and 12:33:18.020 occurred, leading to the disconnection of wind and solar generation in Segovia, Huelva, Badajoz, Sevilla and Caceres, amounting to 928 MW. Some of these trips occurred due to overvoltage protection, but the
Expert Panel was not able to establish the cause of most of these trips. The analysis carried out by the Expert Panel indicates that the overvoltage protection settings at some generation units were set below the voltage limits established in accordance with the applicable requirements. There was an increase in voltage beyond 435 kV during this sequence.
Key events are summarised in Figures 2 and 3. Some generation units were consuming reactive power, which reduced the voltage. However, the disconnections of these units without adequate compensation for the loss of reactive power absorption by other resources capable of controlling reactive power led to increased voltages not only in Spain but also in Portugal. Furthermore, the frequency decreased. Between 12:33:18 and 12:33:21, the voltage in the south of Spain increased sharply, and consequently also in Portugal. The overvoltage triggered a cascade of generation losses, causing the frequency of the Spanish and Portuguese power systems to decline.
At 12:33:19, the power systems of Spain and Portugal started losing synchronism with the rest of the European system.
Between 12:33:19 and 12:33:22, the automatic load shedding and system defence plans of Spain and Portugal were activated, but were unable to prevent the collapse of the Iberian power system. At 12:33:20.473, the AC interconnection to Morocco tripped due to underfrequency. At 12:33:21.535, the AC overhead lines between France and Spain were disconnected by protection devices due to a loss of synchronism, preventing the propagation of the perturbation into the CE power system. From that moment on, the Spanish and Portuguese grids operated in island mode. After this AC separation of the Iberian Peninsula, the power imbalance continued to grow, leading to a further decline in frequency. Finally, at 12:33:23.960, the electrical separation of the Iberian system was completed
14 | April 2026 |
www.modernpowersystems.com
by tripping of the HVDC lines that were still transmitting power from Spain to France (due to the constant power mode setting at the time). All system parameters of the Spanish and Portuguese electricity systems collapsed.
Establishing root causes The report presents analysis of the phenomena and actions that preceded the blackout, to understand the causes of the incident. Based on the results of the analysis, the Expert Panel established a root cause tree describing the key factors that ultimately led to the blackout, a full, highly complex, root cause tree and a simplified version (see Figure 4). Main factors considered in the root cause analysis included voltage control, oscillations, flows between TSO and DSO grids in Spain, adequacy of system defence plans and management of voltage-related alarms.
Voltage control
As already noted, voltage control was a very important aspect of the incident. Analysis by the Expert Panel has identified several factors that contributed to the deterioration of system conditions on 28 April 2025, particularly the sudden voltage increase observed from around 12:32:00 onwards. Switching of network components, such as shunt reactors, for voltage control was undertaken manually, which required a certain amount of decision-making and processing time. Some shunt reactors had been previously disconnected due to low voltages during the oscillation episodes up to 12:24.
According to the requirements applicable on the day of the incident, the RES generators followed a fixed power factor for reactive power provision, thereby not contributing effectively to the voltage control of the system. The fixed power factor mode means that generators absorb reactive power (Q) proportionally to active power (P), and do not respond to voltage fluctuations.
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45
