21
$\begingroup$

Sabotaging the electric network is kind of a well established scheme: find a node of the network, put some explosive on the pole and take it down (or the like). The network is gone for a while.

Now imagine that this organization wants to avoid using explosives and plans to do the following: a large set of conjureds agrees on switching on a high power electric device in their houses (let's say an electric oven or a water boiler, drawing at least 3 kW) at the very same time, to the split second.

Is this a realistic plan to take down the network on (at least) an urbanized area of about 500 square km?

(background info: when I was at the University during a lecture on electricity generation the professor told us that in the '80es a comedian, during a TV show, proposed to complain against the government by switching off the TV when he said to do so. The professor claimed that if he had done it, it would have caused a nationwide blackout.)

|improve this question
$\endgroup$
  • 3
    $\begingroup$ @Rycochet - it wasn't the kettles that was the problem. It was everyone going to the loo, flushing it, and causing the pumping stations to all switch on. $\endgroup$ – Martin Bonner supports Monica Feb 21 '17 at 12:40
  • 3
    $\begingroup$ @MartinBonner - Just found the right name for this - "TV pickup" - en.wikipedia.org/wiki/TV_pickup (including references) :-) $\endgroup$ – Rycochet Feb 21 '17 at 13:03
  • 1
    $\begingroup$ Will your rubes be willing to all install a special appliance in their homes - say, a tampered-with washing machine that's capable of much greater electrical pull than it needs? At the right time, they all activate this particular capacity. Or better yet, activate it remotely (yay Internet of Things!). $\endgroup$ – SPavel Feb 21 '17 at 13:33
  • 5
    $\begingroup$ It's quite plausible that this would take down some local parts of the grid. The company I work for has a single device which consumes about 35MW of power when it's running (it usually runs for an hour or two at a time). We have a special "red telephone" to the nearest power station to warn them when we are planning to use it, to avoid this problem - and they can refuse permission to run it if they can't handle the situation. That's only equivalent to a few thousand kettles. $\endgroup$ – alephzero Feb 21 '17 at 15:43
  • 6
    $\begingroup$ This seems related to my fear of a cyberattack on IoT devices ("the 's' in IoT stands for security"): if you can break into millions of thermostats and simultaneously add or shed the load of millions of AC units, that could maybe cause problems. $\endgroup$ – Nick T Feb 21 '17 at 20:59
37
$\begingroup$

Yes and No

Would you believe this occurs at a predictable time every day in the UK?

During the break at the end of popular soap operas a significant percentage of the population of the UK puts the kettle on. Total load spike on the system can reach 3GW, equivalent to approximately 1.8kW load per household for around 1.75 million kettles at almost exactly the same moment1.

This is a known factor and you can set your watch by the power surges relating to it. Power stations are on standby, more power is made available in France to cover the spike.

However this is a known effect at a known time and the system is geared up to cope with it. Dropping a 3GW load onto the national grid without preparation would likely knock large sections out.

1BBC video source probably not available outside the UK

|improve this answer
$\endgroup$
  • 4
    $\begingroup$ @MichaelKjörling, the overall 3GW figure is a quote from a load balancing engineer at the national grid in the related BBC video. bbc.co.uk/britainfromabove/stories/people/teatimebritain.shtml $\endgroup$ – Separatrix Feb 21 '17 at 12:11
  • 9
    $\begingroup$ This is the reason why the "earth-hour" thing, where everyone is supposed to switch off for an hour, actually has very little effect. Because the power companies need time to roll back and ramp back up in time for the ensueing surge.. it simply is not worth it. Plus, during sudden demand periods they bring online the oil fired power stations since they are a LOT faster to come on line... QED.. EO is a waste of time. $\endgroup$ – Trevor_G Feb 21 '17 at 12:25
  • 5
    $\begingroup$ @MichaelKjörling : Typical european kettles are 3kW each! (That's the big advantage of 230V.) $\endgroup$ – Martin Bonner supports Monica Feb 21 '17 at 12:43
  • 2
    $\begingroup$ @Darael : My source is Amazon.co.uk. The first few hits for "kettle" all said "3kW" or "3000W". Interestingly, the second time I looked it up, the first two hits said 2200W. There's no real risk in drawing the full rated 13A for a kettle - you only leave it on for a few minutes at a time. $\endgroup$ – Martin Bonner supports Monica Feb 21 '17 at 13:05
  • 4
    $\begingroup$ Um, actually, the nominal voltage changed, but if you get out an AVO and stick it in a UK socket, it will still register 240V. (They just fudged the errors that equipment was required to work with so that UK and Continental voltages were both acceptable.) $\endgroup$ – Martin Bonner supports Monica Feb 21 '17 at 14:41
9
$\begingroup$

It shouldn't work.

  • First, there are circuit breakers in flats and houses. These, they could disable, but
  • then, you have breakers outside flats, usually one per building section, like staircase. If you try to draw more than you are allowed, it'll break your circuit. But what if it isn't fast enough?
  • You will have a breakers and monitoring tools at the transformer station, on the stage that transform medium voltage to low one you use in your flat. Often this is the first step where you can encounter active protection and management devices.

And so on. Power grid further away from end user is actively managed and monitored. If these countermeasures fail, you will have a

Cascading power failure

Looking at the examples, for problems like this you need a lightning storm, human error in setting up the protection system, software bug etc.

On the other hand, in India this happened in 2012, just the way you want it. What was the exact conditions no one knows, but it appears as unusually large power consumption in unusually short time. Of course, power grid there was not a state of the art, so you would need to go faster, stronger and overall larger scale to get this done in more modern system. Can't tell you exactly, because simulations say it shouldn't be possible, and we don't have many real life examples to extrapolate from.

|improve this answer
$\endgroup$
  • 3
    $\begingroup$ Also happened around 2000, when Imperial College London was testing its tera(?)watt laser and knocked out half of London. I can't find a citation anywhere, so I didn't post it as an answer, but they were connected directly to the transmission grid without the protections normal distribution grids have $\endgroup$ – nzaman Feb 21 '17 at 7:48
  • $\begingroup$ @nzaman but OP don't want direct approach, I think. Anyway, I'll include it... If and only if I'll find decent article about it. $\endgroup$ – Mołot Feb 21 '17 at 7:51
  • 12
    $\begingroup$ The circuit breakers in buildings are irrelevant though: the proposed 3kW is right at standard 13A fuse capacity in 230V regions. Mains circuit breakers are (here in the UK) typically 20-30A for a single floor worth of sockets, with a 30A breaker for an electric oven/hob. If I wished to participate in an attack on the electric system, the limiting factor for me (and most ordinary people) would be the 60A master breaker, which is nearly 14kW. I have no idea what fuses/breakers are like in e.g. the USA, but it must allow substantial power flow. $\endgroup$ – frodoskywalker Feb 21 '17 at 12:17
  • $\begingroup$ @frodoskywalker only irrelevant if they want to only use one / few per building. When it's a lot, these still gets tripped. $\endgroup$ – Mołot Feb 21 '17 at 12:26
  • 1
    $\begingroup$ Around here in germany with a bit of effort it should be no problem for the average house to produce a load of 40kW and with somewhat bigger installations you get 60-70kW easily, you just need the proper devices which are not that hard to get hold of. Also note that breakers will not instantly pop when the current goes over the limit (so when your breaker pops when you plug in a 3kW kettle, there is something wrong) but when it is over that for a time. From a 16A breaker I can draw 20A for several seconds, maybe minutes just fine. $\endgroup$ – PlasmaHH Feb 21 '17 at 15:16
6
$\begingroup$

Cascade failures are no joke: https://en.wikipedia.org/wiki/Northeast_blackout_of_2003

So how could we engineer one? How big is a 500 sq km area? How many people are involved in this? That's about a third of London or two Birminghams. Let's say you have 2m people. UK peak consumption is very roughly 1kW per person, so your city uses 2GW or half of Drax.

You want to disrupt this.

How much power-delta is required? Let's say 10%. That would require 67,000 people switching on 3kW devices. That seems like quite a lot of people to me, easily enough to engage in more direct action.

Edit: note that you don't necessarily have to switch all of them on. If you can switch off enough load simultaneously enough ("load shedding"), this will cause the local voltage to exceed limits for a short time and trip out the substations - causing more load shedding.

If you've got really good control over the on/off timing, you could try oscillating. Exploit the LC resonance in the largest power line leading to the city?

|improve this answer
$\endgroup$
5
$\begingroup$

500 square km is a small city... or at least, my city is smaller than that. So, I searched news archives for instances of blackout because of overload of the local power station...

I found one such incident: it happened during a test because of which the power plant was running at a fourth of its capacity. While the test was running, a connection to another power station failed, leaving the local power station with all the demand. The system did shut down automatically. Power was restored 20 minutes afterwards.

So, the normal demand was too much for a fourth of the capacity of this single power station when there was no connection to other power stations for backup.

The conditions for this failure are very unlikely... it is expected that the power plant will have enough capacity to supply any foreseeable peak of demand, and second even if it can't handle the capacity it should be able to pull power from the rest of the national network.

So, even if you manage to multiply the demand of the houses and buildings on the city (which would then make the local circuit breakers fail instead of the whole urban area), so that you can cause a peak power demand beyond its capacity... the power plant should be able to rely on the rest of the national network (which you didn't sabotage).

|improve this answer
$\endgroup$
4
$\begingroup$

Maybe, but there would be easier ways on such small scale

As Mołot mentions, cascading power failures are a real threat. However those usually concern much larger areas. Here is an example from Europe, November 4, 2006.

The Day Wind Power Nearly Blew Out Europe.

So if your saboteur times things right, preferably when conditions are at their extremes in terms of load and when a big change is not expected, then yes... if they have hacked every "smart" home and for instance suddenly turn off the main power switch in the houses, then flick them on again, then they can perhaps instigate a failure.

But 500 square kilometers is a square about 22 km on the side, or a circle 25 km across. That is not a very large area. And it would probably be quite easy to just shed the troublesome grid sections restore power to others fairly quickly, no more than 15 minutes or so.

However...

Since the area is quite small, it would be easy for the saboteurs to just map the incoming power lines. Unless the area in question has local power production, attacking the power lines will flick the switch on the entire area. This can be done by bringing down the towers — even though you said "no explosives" there are other ways — or by propelling a chain or similar over them to cause a massive short-circuit.

|improve this answer
$\endgroup$
  • $\begingroup$ I think the idea was to not do anything obviously illegal. $\endgroup$ – Paŭlo Ebermann Feb 21 '17 at 22:14
  • 2
    $\begingroup$ @PaŭloEbermann Sabotaging critical infrastructure for the sake of causing such disruptions is very illegal. "Yeah but we were only flicking switches" is not an excuse. Intent and effect is all you need here... and - ta-dah!! - you just committed sabotage. $\endgroup$ – MichaelK Feb 21 '17 at 22:35
  • 1
    $\begingroup$ @MichaelK It may be inconvenient to prosecute people for making tea. It also is probably much easier to talk people into flipping switches than smashing things. $\endgroup$ – user25818 Feb 21 '17 at 23:58
  • 2
    $\begingroup$ @notstoreboughtdirt Does not matter: it is still sabotage and you are the instigator. Not only that but talking thousands of people into it, you have created thousands of witnesses. And since far from all like the idea of having some fanboi of Anonymous playing vigilante in their area, knocking out their power this in turn means that the police will probably have advance warning of your ploy, as will the power utility. As a general rule: your probability of keeping a secret is inverse to the square of the number of people aware of it. $\endgroup$ – MichaelK Feb 22 '17 at 8:41
3
$\begingroup$

Not only is this possible, it already happens. (two different links there). I am going to take it that you don't live on the US eastern grid where rolling brownouts/blackouts happen during the summer.

|improve this answer
$\endgroup$
0
$\begingroup$

3kW per household is not much, I do not think it will cause a large area blackout. It might cause some on the edge transformators to stop functioning. For reference 3kW is a vacuum cleaner and a hair dryer running at the same time (maybe plus a couple of lights). Unless you are already running couple of other devices eating 3kW, you are out of luck. If you want to overload the system, you should care about turning all that the house breaker could handle, not necessarily at the same time. If the grid is well connected and the event is local, the system might stay up or you could take down quite large sections of it, not just the immediate area.

|improve this answer
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.