If you design an electricity grid to never partition, then you need to keep a substantial headroom of transmission equipment unused to prevent cascading failures leading to a partition in case of failure of just one or two sites.
Instead, an electricity grid should be able to survive any partition like this while still keeping all frequencies within the nominal range. The total cost of such a system is lower in most cases (you need more idling generation capacity or droppable load, but less transmission capacity)
The frequencies were in nominal range. The lowest was around 49.7, the highest about 50.6, both of which are within the allowed range of frequency (47.5 - 52.5 is the absolute emergency range, 49-51 is the nominal range, before you start capping power to or from customers).
Romania had a power blackout for a bit, but everything worked as intended and the romanian grid blacking out didn't pull in the rest of the european grid.
A partition between power grid is acceptable to preserve the largest amount of the grid remaining functional. Marrying the partitions together is bothersome but not something that takes forever (took 1 hour in this case).
While 49.7 is very low, most of your devices will not notice and will be fine. Most industrial equipment is largely already coded to handle frequency shifts in favor of keeping the grid stable.
I understand that some equipment, such as consumer grade solar panels, already disconnects at 49.8, and one early analysis I read[0] said that that in fact exacerbated the problem somewhat on 8 January.
Power producers != Consumers for small devices like consumer grade solar panels, especially if the inverter is too cheap to handle a frequency excursion.
The load shedding mechanisms worked, supportive power generation was automatically activated and there was no widespread blackout.
You are right that it appears like there was insufficient capacity at the north-west/south-east separation point and I guess that is going to be investigated but apart from that everything looks like it worked.
The system survived the partitions, there was no widespread blackout.
Further, the system is not organized by the EU, but the ENTSO-E and contains e.g. countries from North Afrika.
There was no blackout _yet_, but it was very close.
According to the German Bundesnetzagentur (the equivalent of the FTC/EIA), the number of times where they have to intervene with the grid due to grid instability is constantly rising due to Germany shutting down nuclear and coal plants.
> There was no blackout _yet_, but it was very close.
Where do you get that from? None of the sources reported a close blackout, as far as I understood it there was a lot of emergency capacity left. We weren't even in the emergency frequency range, as the other commenter pointed out.
Even the linked article just states that those interventions got more often after shutting down coal+nuclear, but it's not critical, it _only_ costs money to compensate the operators: https://de.wikipedia.org/wiki/Redispatch_(Stromnetz)
It's probably much less money than all the nuclear subsidies.
Nuclear power went from 20GW to 10GW.
Hard coal went from 28GW to 22GW.
Ok, but gas went up from 23GW to 29GW. Brown coal stayed the same.
Renewables went up by by 50GW for PV and 50GW for wind. Consider that wind often hits a 50% capacity factor. That is 25GW in additional power just from wind alone.
Some of those plants may not be running continuously but they are still useful for emergency responses.
It is a choice, spend a fortune on nuclear power that nobody wants in their own backyard, or deal with a less stable grid due to solar and wind energy.
It's perfectly possible to have a stable grid even with solar and wind.
The problem is that neither nuclear nor coal power stations are load following and "base load" has lowered over the years (industry has moved to Asia, devices became more efficient, etc.).
Stand-by power (like natural gas powered generators) hasn't been built up the way the way it should have been. Same goes for smart grid technologies and buffer storage; not to mention the maniacs (especially in South Germany) who basically protest everything - from nuclear power, to wind power, to required infrastructure like north-to-south high-voltage transmission lines.
It's way too oversimplified to reduce the issue to just wind and solar.
Instead, an electricity grid should be able to survive any partition like this while still keeping all frequencies within the nominal range. The total cost of such a system is lower in most cases (you need more idling generation capacity or droppable load, but less transmission capacity)
The EU system here failed to do either here.