News UK power cut: National Grid promises to learn lessons from blackout

[Notebook]

Strange this, only two power sources disconnected from the grid within two minutes of each other. Only 5% of the capacity.
It took over two hours to restore and up to half a million people without power at one point.

Thought our National Grid was a bit more robust than that?

https://www.bbc.co.uk/news/uk-49302996

EDIT: Looks like we aren't getting all the facts yet. National Power spokeschap says the two generators were back online within 15 minutes. They are "a substantial part of the generating capacity", so not 5% then?
I get the impression they were took by surprise by how much of the grid auto-disconnected?

N.

 

[tl8]

Strange this, only two power sources disconnected from the grid within two minutes of each other. Only 5% of the capacity.
It took over two hours to restore and up to half a million people without power at one point.

Thought our National Grid was a bit more robust than that?

https://www.bbc.co.uk/news/uk-49302996

N.

Instantaneous power is very sensitive to sudden changes. 5% is plenty big (actually pretty bug full stop).


Battery tech helps a lot as it can support whole percentage point changes until more power comes on line (15 minutes generally).


Somewhat topical:

2 hours is also pretty fast. Must have meant on major [computer controlled] interconnects tripped and not the smaller manual ones

 

[Urwumpe]

Well, small events can cause chain reactions.
The last big blackout in Wolfsburg was also caused by a tiny bar of metal failing in the wrong spot.

First a small bar of metal failed in the power plant of the factory and four generators went offline.

Then, this drop in power production in the network inside Volkswagen triggered the safety systems of the eastern substation of the factory and it went offline.
Because the substation was feeding electrity into the power network of the city as well at this time (and now started to demand lots of power from the network outside), the substation a few kilometer east of the factory also sensed problems and went offline.

That triggered a first 2 minute long blackout until the two other substations of the city reacted and transported more power from the two still functioning connections to the world outside Wolfsburg into the city, restoring power.

But then the power demand in the city increased rapidly with people waking up, showering and starting cooking coffee. This send the second of three major substations offline and the third one was not capable of supplying power to the whole city.

So, two hours of blackout in the end and that because everything except a small bar of metal actually did its work properly. The substations that went offline did so for preventing damage to the consumers, we would have needed a better connection to the national power lines to prevent the blackout there.

I doubt it is something greater in UK.

See it positive: There will be many new UK citizens in 9 months.

 

[Notebook]

Nicely done documentary, the "Electric Mountain" pops up every now and then as another researcher discovers it!
That is some water-valve, I'll never complain about my plumbing again. The house's not mine of course.

 

[Artlav]

I'm a bit surprised that a powerplant overload results in the whole plant being disconnected rather than parts of the load being dropped. Is it difficult to auto-switch off only a bit of the city or only the factory?

 

[Linguofreak]

I'm a bit surprised that a powerplant overload results in the whole plant being disconnected rather than parts of the load being dropped. Is it difficult to auto-switch off only a bit of the city or only the plant?

The problem is, when you get big changes in supply or load, it can cause frequency, voltage, current, etc. to go out of spec, and it's generally better to have no power than out-of-spec power, so substations that see power going out of spec on any of the grid segments they connect will isolate the offending segment automatically. From what I've heard, the relevant safety systems are generally triggered within a few AC cycles of the fault becoming apparent (so maybe a tenth of a second), before a human operator would even be aware of a problem, and complete whatever action they take within a few seconds (it takes a while to safely open a switch that deals with multi-kilovolt potentials at high current). This can then cause abrupt supply/load changes in other parts of the grid, triggering protective action from other substations.

 

[tl8]

I'm a bit surprised that a powerplant overload results in the whole plant being disconnected rather than parts of the load being dropped. Is it difficult to auto-switch off only a bit of the city or only the factory?

As the grid is well, a grid, if you switch off a whole area, there could be another path and therefore no drop in usage. Better to isolate the power station and let another one take the load. Also, as the station is privately owned, it is easier to isolate your own asset rather than ask someone else to turn off theirs.


Another Tom Scott video on the UK Grid:

 

[Urwumpe]

I'm a bit surprised that a powerplant overload results in the whole plant being disconnected rather than parts of the load being dropped. Is it difficult to auto-switch off only a bit of the city or only the factory?

Well, the part that failed was supposed to do exactly that. It should connect two networks. Also switching high voltage takes a few seconds, especially if its AC. And you can't just switch everything at once, you have to do it in increments or you will cause something else to be thrown offline by transients.

 

[jedidia]

The last big blackout in Wolfsburg was also caused by a tiny bar of metal failing in the wrong spot.

Huh... I hear "tiny bar of metal failing" in connection with electricity, and my first thought is "hit the deck, that bus is going to come through here *really* fast... pieces of it, anyways!".

 

[steph]

Any possibility of it being due to hacking? I hear there've been quite a few large outages worldwide recently. Perhaps someone's testing the capabilities.
Rural western France here, the biggest power outage lasted almost two days. 'natural' causes, some guy with a big tractor managed to clip the high voltage line going into the village. Dude was fine, but the tractor was done. This being in the eve of a national holiday, it took a while to get things going.

That being said, the power network does go haywire from time to time. There are several camping grounds, which get filled during the summer. Basically meaning you've got several thousands more users on what is still a rural old-school network. But, seeing and hearing transformers blow is pretty cool and happens a few times per year, so...

 

[Artlav]

Any possibility of it being due to hacking?

I'd say not likely.
Hacking won't look like a knocked down pole, an exploded transformer or a vaporized spanner.

Hacking would look like a critical connection suddenly disconnected for no reason, causing a cascade of failures.
Or, if it's party time, the grid segmenting, drifting in frequency until the parts are at opposite phases and reconnecting, blowing up a bunch of stuff at once.

 

[Urwumpe]

Not sure, could also have different reasons. Venezuela for example could also be caused by lack of foreign investments and economic decline.

In the UK, the initial "Antispark" was at a gas-fired powerplant, that went offline. It was followed by a big offshore wind park, that also disconnected. The scenario is again very similar to what happened in the UK in 2008.

 

[N_Molson]

Last blackout here was like 2 or 3 years ago for a couple of hours... Usually happens the coldest day of winter when there is really a lot of demand. Else... I don't have to complain really. I enjoy the soft glow of nuclear electricity. :idea:

The real fun happens after every serious storm in rural areas, trees fall on lines, stuff like that. The infrastructure is showing its age. The nuclear power plants too but shh...

 

[Urwumpe]

The nuclear power plants too but shh...

In 50 years, ITER will be ready... just in 50 years... :lol:

 

[N_Molson]

50 years wouldn't be that bad... Controlled nuclear fusion would be pretty much the end of energy issues...

But even if it works, the 40+ nuclear power plants will have to be cleaned. Well, put in barrels and stored (and secured) for several thousand years. Its going to cost a huge lot regardless.

 

[Urwumpe]

50 years wouldn't be that bad... Controlled nuclear fusion would be pretty much the end of energy issues...

But even if it works, the 40+ nuclear power plants will have to be cleaned. Well, put in barrels and stored (and secured) for several thousand years. Its going to cost a huge lot regardless.

And fusion wouldn't be the end of nuclear waste as well. It would just be different waste. Maybe somebody invents a smarter way one day than to hide it "under the carpet". But I don't see one coming soon.

 

[steph]

And fusion wouldn't be the end of nuclear waste as well. It would just be different waste. Maybe somebody invents a smarter way one day than to hide it "under the carpet". But I don't see one coming soon.

I heard thorium reactors (or at least some variant of a subcritical reactor with external neutron source) can be used to 'burn' through some of the waste. I don't know the details though

 

[Artlav]

There are many types of reactors that could burn through the waste. Problem is, most of them are the types which breed fissile material, and no one wants more nukes floating around.

The issue with fusion is the neutron flux it would produce. That tends to turn everything around it into radioactive waste, rather than just the easily removable stuff in the core. It's not going to be any cleaner or much safer.

 

[Urwumpe]

The issue with fusion is the neutron flux it would produce. That tends to turn everything around it into radioactive waste, rather than just the easily removable stuff in the core. It's not going to be any cleaner or much safer.

Exactly, though that is also a matter of the fuel used... Helium-3 would for example be almost neutron free...and perfectly neutron free, if filtering the resulting helium-4 from the plasma would be perfectly fast.


(Since even at too low energy, some helium-3 can fuse with helium-4)

 

[N_Molson]

Helium-3 would for example be almost neutron free...

At the pace things are going maybe we'll be able to extract some from the far side of the Moon in 500 years or so...