Clocks in Europe running six minutes slow because of a power-grid dispute


#1

Originally published at: https://boingboing.net/2018/03/08/clocks-in-europe-running-six-m.html


#2

The problem, affecting some two dozen countries from Spain to Turkey, originates from a political disagreement between Kosovo and Serbia, ENTSO-E said.

LOL - Name one other location that has fucked with the rest of Europe more.


#3

Lead (most voltage) station on the power grid sets the frequency and line voltage…
rest of the power stations follow suit. Lead station carries all the line resistance in the grid,
so they don’t make as much ‘money’ as the ‘follower’ stations.


#4

This got me thinking - are clocks set like that in the US? Or do they just use a simple quartz clock?


#5

Back when everything was powered by linear power supplies, generating timing based off the power line was practically free–you just looked at the signal coming off the secondary of the transformer and watched it go positive and negative. It would do that 100 times a second (120 in 60Hz countries).

In the last 70 years or so, the power grids around the world had a very precisely controlled frequency, so it was a great source of timing accuracy when other methods cost more.

Nowdays, yeah, it’s not used as much, but it’s still more accurate than quartz crystals–especially the 32KiHz ones used in cheap electronics. So, this method is still in use. As switching power supplies replace linear ones, we’ll see less and less of this as there’s no free way to get a safely isolated version of the power waveform. It’s still cheaper to use a linear power supply for little thing like ovens, microwaves, crock pots, coffee machines… So, until switchers get cheap enough to replace them, we’ll keep seeing this used.

Edited to add @Mister44, yes, even in the USA this method is still employed.


#6

Not a big deal, synch off your phone. I generally run my appliance clocks fast anyway, so if I really need the actual time I check my cellphone


#7

But why does a shortfall in power cause the clock frequency to change? I have read a half dozen articles on this and haven’t seen an answrer. Seems likely the answer is that someone with a bad clock somehow got to dominate the system but this is just speculation on my part.

It is sort of fascinating watching all these reputable news outlets just wave their hands at this and parrot each other assuming of course a change in the power flow would change the clock rate.


#8

Output power greater than input power will cause the frequency to drop. The excess power comes from the rotational momentum in the generators.

https://www.forbes.com/sites/quora/2013/10/07/what-is-the-holding-capacity-of-the-us-power-grid/#348e5d9e7d03

Serbia / Kosovo basically robbed some spin off of everyone else’s generators. The NPR article points out that there are gigawatts of power that have to be put back into the grid to restore the frequency back to 50 Hz.


#9

What’s not clear to me is why an error of a few thousands of a Hz is that significant. The grid fluctuates much more than that many times per day, and there is a process in place to adjust drift as well:


#10

I’m pretty sure that France has the most consistent European fuckery at least. If we allow the Frankish kingdoms we’ve got a solid 1000 year streak.


#11

It doesn’t change clock frequency. The clocks are designed to use the frequency of the incoming power as a timing mechanism. If you expect a certain number of cycles per second then you can use that as your clock.


#12

For providing so much history, that article is surprisingly light on engineering.

To answer more directly: everything that holds a charge exhibits some incidental capacitance. Everything that carries current has some resistance, and some incidental inductance. Resistance causes electrical energy to be transfer to heat and lost. Capacitance requires you to spend some energy to build up an electric field. Inductance wastes some energy to build up a magnetic field.

There’s nothing to be done about losses to resistance, but you can do something about losses to capacitance and inductance. For any given inductance and capacitance, you can find exactly one frequency where their contributions to an AC circuit cancel out, and there’s no loss. This is called power matching, and it means maximum power delivery. Obviously, the trick is to choose a standard frequency and design everything to have inductance and capacitance that resonate at that frequency. In the U.S., that frequency for the grid was chosen to be 60 Hz. In Europe, it’s 50 Hz.

When you produce electric power, you did it by translating mechanical power into electric power with a generator. In other words, generating power usually means spinning something connected to a generator, and keeping the grid at, say, 60 Hz means making it spin at a multiple of 60 turns per second, then transforming down predictably. To get it to spin at exactly the right rate, you have to do a good job matching the power going out to the load it’s supplying. Too little demand, too much power conversion, turbine spins too fast, frequency increases. Too much demand, too little power conversion, turbine spins too slowly, frequency decreases.

Yes, power generation and distribution is that complicated.


#13

Probably because it’s a persistent error.


#14

I concede I haven’t made a list, nor do I know all of European history, but coincidentally I have have listened to several history series and Serbia was the cause of problem in several times through out history - or at least was where the problems were, if not the cause. Even back in Roman times.


#15

We should probably all be glad that they’ve managed to keep things at merely killing time. It usually doesn’t go that well.


#16

Right in time keeping if your second is a few thousandths off. You’re loosing a few thousanths 60 times a minute. So after a certain number of minutes. You’re down a whole second. And so forth. Basically all these clocks have been running .004 seconds slower than the actual time. Cumulatively. For weeks. So now they’ve fallen behind by several minutes.

Same thing happens with mechanical watches, especially cheap knock off ones. The spring that handles the actual timing is subtly off just a bit. This second a little short, this second a little long. In the better ones its roughly balanced and it takes a while before the time drifts off. With your $25 Railex? Bigger difference in one direction so couple days and you’ll loose a minute or two.


#17

“so tell me again granpa, how the last great war started because of Servia, Croatia, and some microwave ovens?”


#18

That is why I always travel with a battery powered alarm clock. AC powered clocks run inaccurately in lots of places, as I have discovered. It seems like a design flaw to me, as it relies on an external input to maintain time. my $2.00 Ikea clock was designed without that issue.


#19

Yes, France for the win in the pure net fuckery, but to paraphrase Dan Carlin on Serbia, they punch above their weight in the fuckery dept. The chaos they’ve created for centuries is far disproportionate to their size.


#20

The European grid has a procedure for shifting the frequency up to 0.01 Hz (more than this error) on a daily basis.
I wonder if the issue is more economic than technical at this point. It should be no problem for the other operators to make up this frequency error, but perhaps there’s no guarantee they’d be paid for the extra power input.