back to article Europe is living in the past (by nearly six minutes) thanks to Serbia and Kosovo

Electric timekeepers in Europe have been losing minutes due to power frequency deviations arising from a dispute between Serbia and Kosovo. The snappily titled European Network of Transmission System Operators for Electricity (ENTSO-E) spotted a deviation in power from the control area called Serbia, Macedonia, Montenegro (SMM …

  1. Voland's right hand Silver badge

    Mains powered clock

    Is there such a thing at all nowdays? It is cheaper to throw in a quartz osciallator or a real time clock than to measure grid frequency.

    Otherwise it is not new. When I was a kid the country we were in at the time was playing some really weird export/import shenanigans so the frequency was regularly off to the point where people simply stopped paying any attention to a classic institutional wall clock.

    1. Phil O'Sophical Silver badge

      Re: Mains powered clock

      Is there such a thing at all nowdays? It is cheaper to throw in a quartz osciallator or a real time clock than to measure grid frequency.

      That was my thought when I first heard about this, but I suppose there are still some mechanical clocks with synchronous motors around.

    2. Robert Helpmann??
      Childcatcher

      Re: Mains powered clock

      The clock thing is just a symptom. The real issue is that of less electricity being put into the system than was claimed and is not being provided.

      1. bombastic bob Silver badge
        Devil

        Re: Mains powered clock

        "The real issue is that of less electricity being put into the system than was claimed and is not being provided."

        Yes, and that means power was STOLEN.

        One of the following probably took place:

        a) power generating company(ies) reported more power going onto the grid than was really provided;

        b) someone(s)'s tapping power from the system and NOT paying for it

        c) equipment seriously malfunctioning (probably not likely)

        In the case of 'a', they're being ripped off by one or more power generating companies. SOMEONE is getting the money for "providing" more than was actually on the grid, so I'd guess "follow the money".

        In the case of 'b', some marijuana grow houses (or other gross power thieves) tapped directly into the distribution lines somewhere, in a manner that's not being sensed well enough to indicate a problem, and are powering up some large equipment [like growhouse lighting] on a rather massive scale with stolen kilowatts. So an audit of the grid seems to be in order [it was detected down to the region already, so I'd guess "more of the same" to locate it].

        1. John Brown (no body) Silver badge
          Happy

          Re: Mains powered clock

          "Yes, and that means power was STOLEN."

          Must have been one hell of a ganja farm to by-pass that much past the meter!

    3. Warm Braw

      Re: Mains powered clock

      It has always been the case in the past that you get much better long-term stability from mains-frequency-linked clocks than from a quartz oscillator. My heating timer automatically adjusts for daylight savings time and, power cuts aside, should never need to be altered. You'll need to adjust your quartz clock periodically as it drifts away from the right time. This is a real and rare anomaly. Perhaps radio would be better still, but you're adding cost at that point...

      1. Mark 85

        Re: Mains powered clock

        This is a real and rare anomaly. Perhaps radio would be better still, but you're adding cost at that point...

        Here in the States, the standard for a long time has WWV Time Signal. The catch is the circuitry in the target device (say an alarm clock) adds cost. But it you need "exact time" it's worth the cost.

        1. Alan Brown Silver badge

          Re: Mains powered clock

          "WWV Time Signal"

          Yes, there are 4 such standard transmitters in the EU.

          But even in the USA, the long term stability of the 60Hz mains is higher than any crystal oscillator, whilst being cheaper than WWV

          1. Mage Silver badge

            Re: Mains powered clock

            Irish reception of the UK or German Radio time signal is very erratic in the Mid-West and West. One Alarm clock here has no manual setting and can take 1 to 10 days to recover the time. Idiot design.

            A wrist watch uses your body heat to make the crystal more stable. A quartz crystal wall / desk / alarm clock is a victim of the local temperature. Much less accurate than worn decent metal back quartz watch and very much less accurate than synchronous motor or mains counting electronic clock.

            1. Manolo
              Alien

              Re: Mains powered clock

              When I lived in the Caribbean, the time on my weather station would mysteriously go back six hours overnight occasionally. I then remembered it has a DCF receiver and apparently picked up the signal from over 7.000 kilometres away. Officially the range of the DCF transmitter (it's in Frankfurt I think?) is stated as around 2.500 km. This happened after I moved to the Atlantic side of the island, on the Caribbean side a volcano blocked the signal :-)

        2. Anonymous Coward
          Anonymous Coward

          Re: Mains powered clock

          "Here in the States, the standard for a long time has WWV Time Signal."

          In the UK it is MSF on 60KHz - and in Europe there is DCF on 75KHz.

          As MSF has moved from Rugby to Cumbria - then people in the south of England might find DCF is better reception. Some allowances may have to be made for European time differences.

          Surprisingly there seem to be very few MSF receiver modules available in UK/Europe. The only apparent UK source, "Galleon Systems", is only economic if you order in larger quantities.

          There is a Canadian company, " Universal Solder", who sell one at a reasonable price including shipping at about GBP12. You get the receiver board connected to a (longer) 100mm ferrite rod aerial.

          The strong plastic shipping tube can also be used in service. That means the thin enamelled aerial wires do not get strained if you seal the tube end - with the board inside - after connecting the wires for your preferred controller. I use Arduino Nano or Uno.

          The Kris Adcock variant of the Jarkman MSFDecoder library works ok on my Arduino. Initially reception seemed very poor - but changing the tolerance of #define PULSEMARGIN from 30 to 50 made it very reliable.

          The MSF decode is used to synchronise a DS3231 txco Real Time Clock. It only synchronises when two consecutive minute decodes are sensible. The MSF transmission error protection is too simple to trust one sequence on its own. One of the house MSF clocks was once well out of sync for a day.

          My Arduino implementation also tests the DS3231 module's batttery back-up every few days for low or no voltage. Had to be jury rigged with a small croc clip - as no one seems to have thought to bring out the battery + on a connection pin. The RTC itself is also monitored for poor time-keeping in between synchronisations.

          Three LEDs give various indications of problems. Yellow and Green flashing in sync every two seconds mean all is well. Red indicates a probable battery problem.

          1. Fruit and Nutcase Silver badge

            Re: Mains powered clock

            @AC

            "Electronics" The Maplin Magazine published a "Rugby Clock" project. I think I still have a copy...

            Found this which refers to that project, and also mentions salvaging a MSF receiver from suitable radio controlled clocks, as the Galleon Electronics route is somewhat costly...

            http://www.customelectronics.co.uk/rugby.htm

            also take a look here...

            https://www.burningimage.net/clock/2007/10/23/sensitive-60khz-receiver/

            1. Anonymous Coward
              Anonymous Coward

              Re: Mains powered clock

              "and also mentions salvaging a MSF receiver from suitable radio controlled clocks,"

              My big Maplin MSF digital clocks have the receiver module separated from the main motherboard. It looks similar to the "Universal Solder" one with a long ferrite rod aerial. The clock runs off one AA battery - but I suspect the receiver module is a standard one capable of working up to 5 volts. The clock itself keeps very good time so synchronisation is rarely noticed.

              I have a small MSF clock that soon runs slow during the day. Unfortunately it has the MSF function integrated into a compact smd components board. Not a source of anything apart from the ferrite aerial which is of the much shorter variety - as are the Galleon ones judging by the picture.

          2. Anonymous Coward
            Anonymous Coward

            Re: Mains powered clock

            For anyone who wants to test the DS3231 RTC battery automatically - this is my empirical RTC power down/up sequence. Will probably work with other low power I2C RTC modules too. My I2C chain also happily supports a 16 port expander.

            The DS3231 RTC is powered from a digital port as it only needs a few milliamps.

            The RTC CR2032 battery + probe is connected via a reed relay to an analogue port - which has a 1M resistor to ground to drain any residual voltage.

            The Arduino Nano/Uno I2C SDA and SCL digital shadow ports A4, A5 are initially declared as INPUT.

            The I2C interface can lock up the processor if this druidic ritual is not followed.

            //

            pinMode(A5, OUTPUT);

            pinMode(A4, OUTPUT);

            // force SDA and SCL low via their digital shadow ports on UNO and NANO

            digitalWrite(A5, LOW);

            digitalWrite(A4, LOW);

            delay(2000);

            //

            digitalWrite(RTCPowerPin, LOW);

            Serial.println(F("Removing RTC power"));

            delay(5000);

            //

            digitalWrite(RTCBatteryTestRelayPin,HIGH);

            delay(100);

            BatteryVoltage = ((float)(analogRead(RTCBatteryTestVoltagePin)*5))/1024;

            digitalWrite(RTCBatteryTestRelayPin,LOW);

            //

            digitalWrite(RTCPowerPin, HIGH);

            //

            delay(2000);

            pinMode(A5, INPUT);

            pinMode(A4, INPUT);

            //

            delay(2000);

            // I2C needs reinitialising else hangs

            //!! a commonly recommended TWCR value of 0 causes crashes

            TWCR=0x94; // stop

            //

            // Initialize the rtc object

            rtc.begin();

        3. Anonymous Coward
          Anonymous Coward

          Re: Mains powered clock

          > Here in the States, the standard for a long time has WWV Time Signal

          Used to be. Nowadays (like, for the last 30 years) you would be using GPS¹ (full name: Navigation Signal Timing And Ranging Global Positioning System, or NAVSTAR GPS).

          ¹ Non-American alternatives also available.

          1. Mage Silver badge

            Re: GPS

            An accident waiting to happen. To save cost the DAB, DTT and Mobile systems all use GPS for accurate timing. It's vulnerable. It can be terrestrially spoofed and one decent flare will take out most of the satellites. Not an "if", but "when".

        4. Tom Paine

          Re: Mains powered clock

          For 99% of applications you don't need exact time. In real life a few minutes either way rarely matters. For applications that matter, there's NTP, the radio signal,..

          1. John Brown (no body) Silver badge

            Re: Mains powered clock

            "In real life a few minutes either way rarely matters."

            That depends on where you parked your car and what the "no, we really don't have productivity targets" traffic wardens are up to this week.

      2. Doctor Syntax Silver badge

        Re: Mains powered clock

        "t has always been the case in the past that you get much better long-term stability from mains-frequency-linked clocks than from a quartz oscillator."

        Only if you don't have countries playing silly buggers with the grid.

        "Perhaps radio would be better still, but you're adding cost at that point."

        Just check your clocks & watches against the computer these days. That gets synced to some nice atomic clocks somewhere.

        1. Mage Silver badge

          Re: PC gets synced to some nice atomic clocks

          Though some places have the Time client disabled or Time service port blocked for security.

        2. handleoclast

          Re: Mains powered clock

          Just check your clocks & watches against the computer these days. That gets synced to some nice atomic clocks somewhere.

          Last time I checked (a few years ago, and I can't be arsed wading through that crap again to see if it's changed), Microsoft trumpeted their embraced and extended version of NTP (after ignoring NTP for 25 years first). After making many "improvements" they guaranteed your computer would show the time correct to ±2 minutes. I could have misremembered that, and it could be that timestamps would have that slop but the displayed clock would do better, but it's still shit. Maybe they've improved things since I looked.

          On Linux, using the default NTP servers specified by the installation, I'd be surprised if the time it showed was less accurate than ±50 ms. If I were to configure ntpd to use a few stratum-1 servers (that would be naughty, so I don't do it) I could do better than that.

          For those without sane implementations of NTP, every so often Aldi sell a small MSF clock for around a tenner. Or you could install an NTP client app on your phone. Or even tell your phone to get its time from the MNO (not as good, but probably better than Microsoft's "improved" NTP - I just checked and it's about 2s slow).

          1. Anonymous Coward
            Anonymous Coward

            Re: Mains powered clock

            "After making many "improvements" they guaranteed your computer would show the time correct to ±2 minutes."

            Just checked my W7 against my MSF clock. W7 is 23 seconds fast. The MSF clock is always spot on with the FM radio "pips" on the hour.

            1. Anonymous Coward
              Anonymous Coward

              Re: Mains powered clock

              "Just checked my W7 against my MSF clock. W7 is 23 seconds fast. "

              and this morning it is only 2 seconds fast - about an hour after powering the PC back on.

              1. handleoclast

                Re: Mains powered clock

                "Just checked my W7 against my MSF clock. W7 is 23 seconds fast. "

                and this morning it is only 2 seconds fast - about an hour after powering the PC back on.

                I think (for small values of "think") that the next time you reboot it will have a different error. But less than 2 minutes either way. It's something about the granularity of timekeeping on NT-derived versions of Windows (I expect somebody who actually knows about the innards of the beast will be along to correct me).

                It seems to me to be rather important for servers (at least) to have the correct time. So that intrusion/hack attempts can be correlated with ISP connection records. It's no good knowing the IP address an attack came from if your clock could be out by two minutes with respect to the ISP's and you get spurious connection details. That's why, whenever I take over admin of a *nix server, one of the first things I do is make sure ntpd is correctly configured and running.

                It's also important to have the correct time (or at least co-ordinated time) across any setup using Kerberos (such as Microsoft Active Directory). Oh, I vaguely recall rhere's a rather large slop in the specs for timing in all that. "We need to get the right time to avoid replay attacks, but we want to allow a margin for network congestion/lag and clock drift." Yeah, right.

                Microsoft's tech page boasting about how they "improved" NTP adds insult to injury.

                1. Naselus

                  Re: Mains powered clock

                  "It's also important to have the correct time (or at least co-ordinated time) across any setup using Kerberos (such as Microsoft Active Directory). Oh, I vaguely recall rhere's a rather large slop in the specs for timing in all that."

                  Kerberos is fine as long as time difference is under 5 minutes.

          2. david 12 Silver badge

            "after ignoring NTP for 25 years"

            NTP dates to c 1985. If, as you claim to believe, MS ignored it for 25 years, that would put the first MS implementation at c 2010.

            The rest of your post is of similar value.

            1. handleoclast

              Re: "after ignoring NTP for 25 years"

              @david 12

              NTP dates to c 1985.

              It existed, under the name history..

              Some of us look deeper than Wikipedia for our facts.

              If, as you claim to believe, MS ignored it for 25 years, that would put the first MS implementation at c 2010.

              I can't remember the exact date when I no longer had to install 3rd-party NTP solutions on Windoze because Microsoft had finally included their own implementation, but that date sounds about right. It was somewhere between 6 and 12 years ago. At some point in that timeframe I walked into a room full of computers running up-to-date versions of Windoze and the clocks were all over the place. I showed the person in charge how to fix it by installing a 3rd-party solution because at that point in time there was no Microsoft implementation of NTP.

              And that first Microsoft implementation was pretty crap because it used SNTP (intended for querying a serious timeserver on your local network) to query Microsoft-run timeservers across the wider Internet. Because it was SNTP it did almost none of the sophisticated analysis required to compensate for WAN latency/congestion/etc. It was as shitty as I expected, but it was sufficiently adequate that it was no longer worth spending time installing something better.

              Yep, 25 years is about right, because I recall David Mills announcing NTP had been in operation for 25 years around the same time that Microsoft finally got around to implementing NTP on Windoze. Just because you find it hard to believe that Microsoft ignored NTP for 25 years does not mean that they didn't ignore it for 25 years. You should not let your personal credulity get in the way of actual facts.

              The rest of your post is of similar value.

              That appears to be the pot calling an iceberg black.

              1. Red Bren
                Joke

                Re: "after ignoring NTP for 25 years"

                @ david 12

                Perhaps he's using the MS implementation of NTP to set his calendar?

              2. david 12 Silver badge

                Re: "after ignoring NTP for 25 years"

                NTP was implemented in Win2K. It wasn't a very good implementation, but it's not "ignoring" NTP for 25 years -- because the 1975 date you suggest wasn't NTP, wasn't called NTP, wasn't anything like NTP as described in:

                A New Fault-tolerant Algorithm for Clock Synchronization”. Information and Computing, vol.77, no.1, pp.1-36, 1988.

                And where poster said "2 minutes" in the garbage post, he would have meant "2 seconds" if he had known what he was talking about.

                (Recent implementations of MS NTP are aparently quite good, with milli-second syncronisation. I haven't tried..)

    4. tom dial Silver badge

      Re: Mains powered clock

      Why would it be cheaper to use a crystal oscillator - and a counter - than a counter alone? The ICs probably cost the same, and the extra crystal cost is positive.

      And, as another poster observed, power line frequency generally is more stable over time than an autonomous quartz crystal.

      1. Anonymous Coward
        Anonymous Coward

        Re: Mains powered clock

        Back in the 1960s we had a tour round our UK local electricity control centre. They explained that the grid had to let the frequency change when compensating for load. They then had to make a deliberate change in the opposite direction when possible - to bring any synchronous clocks back into line.

        They had a big such clock on the wall. It was apparently their reference for boosting generators ready for the kettles going on at the end of any popular TV programmes.

        1. bombastic bob Silver badge
          Devil

          Re: Mains powered clock

          "They explained that the grid had to let the frequency change when compensating for load"

          Exactly. The frequency (generator shaft speed, essentially) is regulated based on load. There's a "curve slope" that's used, essentially like a speed control, to make the generators turn a bit slower as load increases. Doing it this way is a lot more stable, and handles transient load conditions better.

          So, at high loads, frequency is just a bit lower than at no load. And this has several interesting effects: First of all, if you run generators in parallel, they tend to load balance very nicely. Second, if you get a power transient, it's a very smooth transition with very little over/undershoot. And every generator on the grid would essentially be able to maintain this load balance correctly. You could, for example, cause one generator plant to take up more load by INCREASING its frequency setting, and leaving the others alone (this is what an Independent Systems Operator would be responsible for managing). Doing this could keep the generation and the load close to one another. If Kosovo is using more power, then generators close to Kosovo should be providing it. That kind of thing. And you can keep nuclear power plants running at 100% power as well [they don't do so well with transients, inherent design thing]. Anyway, it's really very cooperative like that.

        2. Dom 3

          Re: Mains powered clock

          "reference for boosting generators"

          These days they just, errr, watch the telly.

          https://www.youtube.com/watch?v=slDAvewWfrA

    5. Jason Bloomberg Silver badge
      Trollface

      Re: Mains powered clock

      Using mains frequency is much more accurate long term than even a good crystal oscillator.

      It also allows the frequency to be slowed down during the working day then sped up overnight so the state can have people working for longer than they are being paid for.

    6. Gordon 11

      Re: Mains powered clock

      Is there such a thing at all nowdays? It is cheaper to throw in a quartz osciallator or a real time clock than to measure grid frequency.
      I have one on my bedside table. I've had one (not the same one) there for the last 45 years.

      Unlike a quartz clock it never needs to be reset (unless there's a power failure).

      1. Anonymous Coward
        Anonymous Coward

        Re: Mains powered clock

        Unlike a quartz clock it never needs to be reset (unless there's a power failure).

        Unlike a quartz clock it always needs to be reset when there's a power failure.

        TFTFY

    7. Mage Silver badge

      Re: Mains powered clock

      Old clocks. I still have a lovely 1950s mantel clock.

      Most mechanical plug-in time-switches.

      A lot of central heating and immersion clocks are still mechanical as that stuff lasts for ever unlike the privacy parasite IoT stuff.

      Very many NEW cheap clock radios and LED alarm clocks actually count the 50Hz. It's built into the ancient design LED clock/display single chips they still use. Some older ones have a PP3 with an RC oscillator. These seem to last forever.

      1. Anonymous Coward
        Anonymous Coward

        Re: Mains powered clock

        "A lot of central heating and immersion clocks are still mechanical as that stuff lasts for ever"

        My mechanical timer built in to the oven housing wore out its gearing after about 10 years.

        I'm still trying to get my 1920/30s Garrard repaired. The last repair job was apparently botched and the mainspring broke for a second time.

        I keep my Omega Seamaster Cosmic wristwatch for dress occasions. The GBP200 maintenance service bills were becoming a burden. In 45 years it has never kept good time. The Isle of Man maker of very expensive custom mechanical watches (GBP 100K) is on record as saying that if you want reliable accuracy then you can't beat a cheap quartz watch.

      2. Anonymous Coward
        Anonymous Coward

        Re: Mains powered clock

        "Some older ones have a PP3 with an RC oscillator. "

        My central heating timer has a PP3 apparently as a back up. The clock runs slow and needs a correction every week. When there is a power failure the PP3 invariably proves to be dead.

        1. John Brown (no body) Silver badge

          Re: Mains powered clock

          "My central heating timer has a PP3 apparently as a back up. The clock runs slow and needs a correction every week."

          Sounds like it's broken if it's so far out each and every week that you notice. I think mine would have to be losing or gaining fifteen minutes per week for it to be noticeable and even the cheapest nastiest timers are rarely that poor.

    8. bombastic bob Silver badge
      Devil

      Re: Mains powered clock

      "Is there such a thing at all nowdays? It is cheaper to throw in a quartz osciallator or a real time clock than to measure grid frequency."

      not necessarily. old-school analog clocks probably still have synchronous motors in them. And I remember constructing a digital clock (well, back in the 70's) that used line frequency. Since you had to have a transformer [back then] it made sense to just use it. Seriously, though, it's not that hard to get the incoming AC frequency, through a very cheap capacitor, and then measure it with a single input pin on a dedicated clock IC. Crystals probably cost $1 more than that (and typical crystal oscillators need 2 capacitors, one on each end of the crystal). And that assumes that you don't want a backup battery for when power goes out, so the clock just flashes 12 after a power outage. This would be, a VERY cheap clock, yeah.

      Anyway, my $.10 worth.

      1. This post has been deleted by its author

      2. Anonymous Coward
        Anonymous Coward

        Re: Mains powered clock

        "[...] it's not that hard to get the incoming AC frequency, through a very cheap capacitor, [...]"

        My gut feeling is that the household supply would have enough glitches from internal load changes to cause false transitions - unless you had a tuned filter or phase locked loop.

    9. Tom Paine

      Re: Mains powered clock

      It's a reasonable question, and what I was wondering myself; I assume the savage downvoting means "No, and you're an unspeakable barbarian for even asking the question".

      Allow me to donate some surplus karma to the cause of asking a question. Who uses clocks that calibrate off the main frequency, and if it's so prone to inaccuracy, why? I've a ten quid plastic click on the kitchen wall that syncs off the radio time signal (or does it?)

  2. Anonymous Coward
    Anonymous Coward

    Curious, so do these clocks run fast in the UK?

    1. Peter2 Silver badge

      UK grid frequency is currently 49.952Hz ( you can check yourself - http://gridwatch.templar.co.uk/)

      The National Grid is (IIRC) legally required to be within 49.5 and 50.5hz. I personally wouldn't rely on the grid being that accurate.

      Shouldn't they have it on a UPS generating a pure sinewave at precisely 50Hz if it's THAT important that their supply be completely, perfectly accurate? I'd hate to know what that lot would be saying if somebody on the same circuit started arc welding or similar...

      1. Anonymous Coward
        Anonymous Coward

        "Shouldn't they have it on a UPS generating a pure sinewave at precisely 50Hz"

        The pumped storage is effectively a very high capacity UPS. Without it the frequency would deviate during rapid load changes.

      2. Anonymous Coward
        Anonymous Coward

        For some reason I assumed the UK was 60Hz. Probably confusing TV signals or something.

        1. Phil Endecott

          > For some reason I assumed the UK was 60Hz.

          You probably also think we use inches.

          1. Anonymous Coward
            Anonymous Coward

            @Phil Endecott

            I suppose I did put my foot in it.

          2. quxinot

            >> For some reason I assumed the UK was 60Hz.

            >You probably also think we use inches.

            And have speed limits set in MPH. Go figure, eh?

            1. Yet Another Anonymous coward Silver badge

              And buy fuel in liters but quote consumption in MPG

          3. bombastic bob Silver badge
            Devil

            "You probably also think we use inches."

            you don't? (how about miles, pounds, gallons, ...)

            yeah only us US'ians still measure based on a King's anatomy. I wouldn't mind using all metric if the damned bolts on my car were all metric... and I needed a 5/16" wrench the other day to loosen the battery terminals so I could replace the battery. Yeah, it's just unavoidable.

            And every damned recipe in every recipe book uses freaking tablespoons and cups. What the hell?

            1. Anonymous Coward
              Anonymous Coward

              "gallons, [...] cups"

              While they are standard sizes in their own countries (USA, UK, Australia) - they do have different absolute capacities. Tablespoons are not apparently standardised if you buy one in a shop in the UK.

              In the UK you may still find recipes using ounces and pints - it is more usual to find grammes and millilitres. All food packaging and content analyses use grammes and millilitres. Can't remember the last time a modern recipe used "cups".

              Not sure if the USA and UK re-aligned their standard imperial yards. The discrepancy was noticed during WW2 when manufactured component tolerances were mismatched.

              1. Anonymous Coward
                Anonymous Coward

                Here in France, my local hardware store sells shelving in 20, 30 and 40cm widths, all 1.83m long. Metric, of a sort...

              2. hmv

                The "international yard" was standardised in October 1958 as 0.9144 meters. So when you're using yards, you're using the metric system (just a funny multiplier).

          4. King Jack

            You probably also think we use inches.

            We use everything here. Miles for distance, Pints for beer in the pub but in the supermarket its litres or parts there of. Cars still use miles per gallon but you buy fuel by the litre. If the PTB had balls all this could have been sorted 50 years ago. But we seem to be happy sitting between the two systems. Petrol in litres so that the prices can rise steeply. 2p on a litre is much higher than 2p on the gallon. We've been conditioned to accept it without thinking too hard.

            1. Phil O'Sophical Silver badge

              Re: You probably also think we use inches.

              If the PTB had balls all this could have been sorted 50 years ago. But we seem to be happy sitting between the two systems.

              It's a question of convenience. Peope use metric measurements for calculations etc. because it's easier. For things like distances, though, no-one actually cares about the actual number. Tell someone "it's 12 miles away" and that translates mentally into "about 15 minutes in the car" or "too far to walk, I'll take the bike", etc. Tell them "20km" and they have no mental reference for it, so it's meaningless. Same for fuel consumption, what matters is "how much to fill the tank" and "more or less economical than my last car". Litres/gallons/pints are irrelevant as such.

        2. Martin an gof Silver badge

          Probably confusing TV signals or something.

          Always annoys me that normal monitors refuse to do 50Hz to match TV. Particularly so when editing video from y'know, video cameras, which is 50Hz. Upscaling that to 60Hz just results in horrid jerkiness.

          M.

          1. bombastic bob Silver badge
            Devil

            I'm pretty sure that HDTV standards allow for different frame rates, like 24fps [movies], 25fps (EU standard I guess), 30fps (US standard I guess). Over here in the USA we're doing HDTV pretty much everyplace now. The 'pulldown' and 'telecine' algorithms are well known and supported by common open source tools like ffmpeg and mencoder, last I checked. And at least one HDTV I've seen had Linux on it (GPL declaration in the printed stuff). I think even VLC can do the conversion. So there ya go.

            I don't think it really matters any more what the actual frame rate is for the video, other than the perception of the viewer. I happen to like 25fps since it makes the files a bit smaller.

            But the reason "back in the day" was to synchronize the vertical sweep with the line frequency in order to avoid the kinds of problems that a CHEAP CRAPPY POWER SUPPLY would have, a line frequency pattern that would vertically sweep across the screen. If you synchronized it with the line frequency, it was less annoying. Fortunately electronics have improved since the 1950's, and that's no longer a problem.

            1. Martin an gof Silver badge

              I'm pretty sure that HDTV standards allow for different frame rates ... I don't think it really matters any more what the actual frame rate is for the video, other than the perception of the viewer

              Yes, HDTV does, but computer monitors usually do not. It's all to do with certain assumptions and "traditions". Look up the difference between CEA and DMT modes. Most TVs will happily display either (CEA tend to be "TV" modes, DMT tend to be "Monitor"), while many computer monitors will only display DMT modes. I had quite a few problems with this in the early days (2012) using a Raspberry Pi as a video player.

              In 50Hz countries we have a legacy of 25/50Hz content. On a TV, no problem, the TV just synchs at whatever is required. Computer monitors will almost invariably be on 60Hz these days, and it's not a nice, simple up-scaling to get from 50Hz to 60Hz (though arguably going the other way is worse).

              Another problem is computers attached to TVs. They will query for supported resolutions and then choose the "best". In this context, "best" means "more", so they will generally choose the highest resolution at the highest frame rate possible, often 60Hz which may not be appropriate for a media player. A TV can report its "preferred" resolution, but an attached device doesn't have to honour it.

              The other problem comes when editing. It's difficult to "audition" video properly if the source file is 50Hz and the monitor is running at 60Hz. Yes, there are algorithms and modern processors certainly have the grunt to make them work well enough, but just like the algorithms for de-interlacing, nothing is quite perfect.

              And all it would take would be a minor software tweak to allow monitors to report 50Hz modes, and perhaps a locale preference that defaults to 50Hz in certain parts of the world.

              M.

          2. Natalie Gritpants

            Your monitor probably does support 50Hz unless it is very old. Probably you graphics card is the fault or the video editing software.

            1. Martin an gof Silver badge

              Your monitor probably does support 50Hz unless it is very old. Probably you graphics card is the fault or the video editing software.

              There is an age thing - but I suspect it's more related to the supported interfaces. I've never come across a monitor which reports anything other than 60Hz on a VGA connector or a DVI connector, but monitors with HDMI ports tend to support 50Hz on the HDMI, though probably not 24/25/30. The oldest monitor I'm using regularly at the moment is perhaps seven or eight years old. My old Philips 17" CRT would synch at anything from about 45Hz to 120Hz, so long as the other timings added up. I had an even older Philips - Acorn-badged - that could even do 50Hz interlaced. It had a SCART socket :-)

              M.

      3. Martin an gof Silver badge

        The National Grid is (IIRC) legally required to be within 49.5 and 50.5hz. I personally wouldn't rely on the grid being that accurate.

        The 1% requirement is correct, and I believe that they're pretty accurate on that.

        The reason our grid hasn't been affected with Europe's is that our connections to Europe are all d.c. and our grid is entirely separate, control-wise.

        M.

        1. Peter2 Silver badge

          The 1% requirement is correct, and I believe that they're pretty accurate on that.

          Oh, totally accurate. The UK grid is rarely more than 0.1Hz out.

          My point is simply that given that it could be as low as 49.5 or as high as 50.5 legally, if your complaining about the average being 49.996Hz because it's *that* important to you then you should be taking corrective action youself instead of relying on the grid, which is not required to provide that level of accuracy.

          1. Anonymous Coward
            Anonymous Coward

            The grid IS required to maintain that level of long - term accuracy.

            It may go up and down, but the long term average is 50Hz +-0.

      4. Anonymous Coward
        Anonymous Coward

        The specific regulation regarding accumulated errors is in the Grid Code, section BC3.4.3:

        "NGET [National Grid Electricity Transmission plc] will endeavour (in so far as it is able) to control electric clock time to within plus or minus 10 seconds by specifying changes to Target Frequency, by accepting bids and offers in the Balancing Mechanism. Errors greater than plus or minus 10 seconds may be temporarily accepted at NGET's reasonable discretion."

  3. wolfetone Silver badge

    Time Is Terribly Slow Until Pub

    You're telling me.

    I've been sat at my desk since 11am looking at the box of beers I've been given, waiting for 5:30 to come for me to get bladdered.

    1. Anonymous Coward
      Anonymous Coward

      Re: Time Is Terribly Slow Until Pub

      It's 6pm here and I'm still stuck on a conf call with Americans who don't know it's already pub time.

      1. m0rt

        Re: Time Is Terribly Slow Until Pub

        This is why you say 'Bar' time.

        Translation issue. :)

  4. Anonymous Coward
    Anonymous Coward

    Some might call that "taking back control"

    Not really, as we've always had it (in this case).

    1. Lars Silver badge
      Happy

      Re: Some might call that "taking back control"

      As Nick Clegg said it, the Brexit referendum was about "tacking back control of our borders" but instead a boarder that did not exist has been created.

      https://www.youtube.com/watch?v=JIsmatwzVRY

      1. Yet Another Anonymous coward Silver badge

        Re: Some might call that "taking back control"

        Obviously it will be necessary to build a wall around Europe to keep them in their place

        1. Stoneshop
          Boffin

          Re: Some might call that "taking back control"

          Obviously it will be necessary to build a wall around Europe to keep them in their place

          ITYM "wall around the Continent". The UK, and Ireland, will be part of Europe until some serious tectonic action occurs across the North Sea.

          "Fog in Channel, Continent isolated"

      2. Anonymous Coward
        Anonymous Coward

        Re: Some might call that "taking back control"

        "[...] but instead a boarder that did not exist has been created."

        As Jacob Rees-Mogg would likely say as a solution to the Irish border conundrum - "Stand by to repel boarders!"

  5. techmind

    Not quite as it seems

    I saw this story elsewhere earlier in the week, and gave it some thought then.

    Normally in a synchronous power grid the instantaneous frequency deviation indicates the mismatch between demand and supply (it runs fast if over-supply, or slow if under-supply). There's also a general agreement to make the daily cycles add up over 24 hours to keep clocks in step.

    It seems to me that allowing the cycles to run short over a period of days and weeks is a conscious (nod-nod wink-wink) decision of the operators of the power grid in order to make a *protest* (to raise it up the political agenda) rather than any direct technical consequence of the allegedly "stolen" power.

    Presumably the electrical grid companies themselves have very limited legal options to pursue in other jurisdictions, so need to get political (governmental) wheels in motion to sort things out.

    Power games...

    1. ashdav

      Re: Not quite as it seems

      Power games.

      I get it.

      Interesting website of UK power usage http://www.gridwatch.templar.co.uk/

  6. JassMan

    For those who wonder...

    and are too busy to look up the tech details: The reason the UK hasn't been affected is that the power under the channel is sent via DC. We have control of our own 50Hz when the DC is made back into AC. All the rest of Europe is currently connected by a mixture of AC & DC interconnects but where you are running both, the frequency of the DC/AC interconnect has to match the AC/AC otherwise you get some very warm wires.

    https://en.wikipedia.org/wiki/HVDC_Cross-Channel

    1. Anonymous Coward
      Anonymous Coward

      Re: For those who wonder...

      Thank you, that's enlightening.

    2. Mark 85

      Re: For those who wonder...

      DC is used? I'm surprised. Seems the very early power grids (in the States at least) were DC but AC was just more efficient with less line loss.

      1. Adam 52 Silver badge

        Re: For those who wonder...

        DC is more efficient. If you follow the HVDC reference from the link above it explains why.

        1. This post has been deleted by its author

          1. Phil O'Sophical Silver badge

            Re: For those who wonder...

            If then Britain is alone with DC then, well, how surprising, not.

            How can you be that unaware of the situation outside your borders, yet still feel competent to comment on another country's political choices?

            Local and medium distance is AC, for long haul DC is more efficient. That's pretty consistent worldwide.

          2. Stoneshop
            FAIL

            Re: For those who wonder...

            Too lazy to prove it now, but it would not surprise me if it happened all over Europe.

            Nope. Domestic DC was uncommon, though it has been in use in some places. In the 1960's there were still locations that had DC as well as AC, for instance the Danish island of Fanø.

            If then Britain is alone with DC then, well, how surprising, not.

            Only the HV interconnects to France are.

            Of course there was this fight about efficiency regarding the electric chair in the USA,

            It wasn't about efficiency, it was about Edison (DC) versus Tesla (AC), with Edison 'proving' that AC was more dangerous by demonstrating electrocutions using AC.

            1. bombastic bob Silver badge
              Devil

              Re: For those who wonder...

              "It wasn't about efficiency, it was about Edison (DC) versus Tesla (AC), with Edison 'proving' that AC was more dangerous by demonstrating electrocutions using AC."

              They certainly weren't more 'dangerous', as high voltage DC and high voltage AC will still kill you. But the AC version would be a LOT more entertaining... [doing the 50/60 cycle jitters in the chair].

              Edison was just wrong about AC and we all know it. But he was a SORE LOSER. And he HATED Tesla because of it. They used to get along, until AC vs DC. Tesla was right. Tesla had the ability to think 'dynamically' and do simulations in his head. That's how he came up with the idea of a rotating magnetic field, and alternating current, as I understand it. He pictured it in his mind. Those of us who can ALSO do that kind of thing understand it completely, I'd guess. And I'd also guess that Edison couldn't really do that, though his 'static' imagination was still pretty brilliant.

      2. Anonymous Coward
        Anonymous Coward

        Re: For those who wonder...

        No AC and DC are equally efficient. The trick is to use high voltage to minimise transmission loss. This is was easy with AC as all you need is a transformer, which is why AC won out. I recall the DC power link to France used thyristors. Technically much harder but it means UK and French grids don't need to be in sync.

        1. Neil Barnes Silver badge

          Re: For those who wonder...

          DC and AC both suffer the same resistive losses, but DC doesn't suffer from capacitive losses n transmission.

          1. John Sager

            Re: For those who wonder...

            That's why they are going to use DC for the big windfarms in the North Sea. The routes are so long underwater that AC would lose too much reactive power in the capacitance to the seawater.

        2. Anonymous Coward
          Anonymous Coward

          Re: For those who wonder...

          "No AC and DC are equally efficient." - apparently not: https://en.wikipedia.org/wiki/High-voltage_direct_current#Comparison_of_HVDC_and_AC

          Yes indeed, voltage conversion for AC is much easier and cheaper and reliable than for high power high voltage DC, but HVDC is markedly more efficient for undersea cables: losses using AC underwater cables are enormous.

        3. Alan Brown Silver badge

          Re: For those who wonder...

          "AC and DC are equally efficient"

          Not for HV transmission. Skin losses start factoring into long cable runs, DC uses the entire conductor and for very long AC runs the transmission line starts acting as a radiating antenna to ground (reactive losses)

          On the other side of the equation if you get a breakdown anywhere along the line, AC arcs are generally self-extinguishing, whereas a DC line might need a shutdown to quench.

          HVDC are also more immune to solar flares, but you can mitigate that on AC links at cost of using 2 transformers instead of one.

          When you run lines underwater the AC problems are multiplied in spades and they also become a problem to some extent in tunnels, but the transmission ones under London aren't long enough to matter.

        4. Mage Silver badge
          Boffin

          Re: DC vs AC

          Once upon a time the AC voltage could easily be stepped up or down nearly lossless (if it's a really big transformer). The only way to do it with DC was a dynamotor (rotary converter), basically a motor and dynamo on the same shaft. Very lossy.

          Why?

          Because high voltage means lower current and thus the wires can carry maybe 1000x more power on a grid. Power distribution is 5,000 V to 400,000V depending on the length of cable and power to be carried. USA houses have to have 220V for wired in high power appliances (Allows about 3 kW) as the 110V wiring might only allow 1 kW to 1.5 kW, I'm not sure. Much less than Europe.

          However the transmission wiring has capacitance and inductance. In theory perfect capacitors are lossless. Also very long wires act as an aerial even at 50Hz or 60Hz so some energy is radiated. DC thus is lower loss. Today the efficiency of electronics based DC-DC converters is very high (I suspect very big SMPSU). The high voltage DC is a big advantage on an undersea or underground cable and increasingly being used for grid distribution.

          Japan has 50Hz and 60Hz, but is 110V.

          TV was originally based on the local mains for the field rate so hum bars would be stationary, it's a pity it wasn't 48Hz (2x film), the 1/2 field frame rate of 25 fps or 30fps is a clever early x2 compression scheme. We are much less able to see detail in movement, so the idea was to transmit odd and even lines in alternate frames. Thus static pictures were about 480 visible (of 525) in USA, or about 378 of 405 lines or from 1948 in Europe/Russia about 576 lines out of 625 transmitted. The extra lines are time to allow the scanning electronics in camera and screen to reset to the top of the frame. The 1080 lines HD is actually from 1125 lines Japanese Analogue HD.

          The lock to local mains wasn't needed from mid 1950s. It's totally sad that the entry level of HD wasn't 48 fps without interlace. The analogue interlace technique makes it HARDER to do digital! Possibly also more than 1080 lines and no 720 mode as that's not much improvement in Europe (rather than 480 USA/Japan). Having 1200 minimum (twice SVGA) would have been better. Digital and HD resolutions and frame rates are a mess, 24p, 25i, 30i, 50p, 48p 60p frame rates, both square and anamorphic pixels. The HD Ready lie, TVs with no HD Tuner and only rescaling 1920 x 1080 to native pixels.

          1. Anonymous Coward
            Anonymous Coward

            Re: DC vs AC

            "The only way to do it with DC was a dynamotor (rotary converter), basically a motor and dynamo on the same shaft. "

            For lower power step up requirements - like a valve radio receiver from 12/24v - there was the "vibrator" voltage converter. Basically a buzzer interrupting the current flow through a transformer.

      3. bombastic bob Silver badge
        Devil

        Re: For those who wonder...

        "AC was just more efficient with less line loss."

        this mostly has to do with distribution locally, not over distance. the problem with DC is that you can't EASILY convert a high voltage (like 100kv) into a line voltage for home use (let's say 220 or 110). So the generators had to make DC at 'what you get at the wall socket', meaning very high current and lots of loss.

        However, if you send DC over distance at 100kv (let's say), then all you need is a motor/generator setup on each end. In fact, an MG can go BOTH directions. A DC line under the channel could then be a bi-directional power connection. It's probably why they do it "that way". Also a DC setup could have surge capacitors and batteries on it to help absorb transients etc. etc.. Sort of like submarine power, which still has backup generators and batteries very similar to old WW2 diesel subs. Hey, it works!

        The motor/generator [for those who do not know] is what it says on the tin. It's a motor on one side, and a generator on the other side. The mode of operation is essentially based on how the regulators are set up to work. A generator on the DC side will regulate voltage. A motor on the DC side will regulate speed. Similarly, if the AC side is operating as a generator, its frequency regulates the DC side. If it's operating as a motor, the voltage and current on the DC side determines the loading. Both the voltage and speed regulation modes would have appropriate settings curves, so that load is shared and regulated properly, and you adjust the regulator's curve point to maintain 50Hz (or whatever), or 'n' volts on the DC bus.

        Also in this way you can parallel the MG sets, so that you can take one or more off line for maintenance [yeah those DC brushes will need to be changed out periodically, and the contacts cleaned, etc.] but with proper care and maintenance, an MG set will spin for DECADES without major problems, and operate very efficiently. They really do work very very well, and are probably the most reliable way to transfer very large power levels between AC and DC, while simultaneously allowing bi-directional flow.

        Additionally, you could use a motor/generator for standard conversion - a 60Hz unit on one side, a 50Hz unit on the other, with an appropriate number of poles on each so that they spin the same speed. Often they're used for 400Hz systems in that way.

        Another advantage of using DC to transfer power is that you separate the line frequency controls on either end of the distribution. I assume this is what they wanted to do here. And when you're putting power lines under water, maybe the inductive losses are much higher with AC than with DC. I'd guess that's it at face value, without researching even.

        1. SImon Hobson Bronze badge

          Re: For those who wonder...

          Another advantage of using DC to transfer power is that you separate the line frequency controls on either end of the distribution.

          Indeed, that is also a very important feature. AIUI over in North America they have the grid split into several sections connected only by DC links - thus avoiding having to try and run the whole continent as one huge frequency control zone. The bigger the area, the harder it is to control and keep stable.

    3. Killing Time

      Re: For those who wonder...

      If it were purely down to cost, it would be cheaper to maintain the grid solely as AC however the DC aspect of the interconnects are there specifically to segregate the 'frequency control zone'. The UK's distribution grid has now effectively become part of the European grid ( there are other DC interconnects with Ireland and Norway for example).

      Frequency deviation within the zones are managed by dispatching generation units of adequate power output in a specific 'frequency control mode' which sacrifices their overall power export capability to the grid in favour of driving the grid frequency in a stable manner.

      Another reason why large conventionally fuelled power plants will be needed along with renewables.

      1. Alan Brown Silver badge

        Re: For those who wonder...

        "Another reason why large conventionally fuelled power plants will be needed along with renewables."

        Apart from the factor that renewables can't supply the entire required load anyway?

        1. Anonymous Coward
          Anonymous Coward

          Re: For those who wonder...

          "Apart from the factor that renewables can't supply the entire required load anyway?"

          Renewables tend to be able to come on line very, very quickly to meet surges in demand. Thermal systems take time to build up a head of steam.

          The Dinorwig hydroelectric power station uses off peak electricity to pump water up to its reservoir - then releases it to drive the generators*** to combat grid surges. It does 0 to 1800MW in 16 seconds - online duration up to 6 hours.

          *** IIRC the pumps and generators are the same turbines? El Reg did an article a while back?

          1. Fruit and Nutcase Silver badge

            Re: For those who wonder...

            @AC(!)

            El Reg did an article a while back?

            https://www.theregister.co.uk/2016/05/16/geeks_guide_electric_mountain/

          2. SImon Hobson Bronze badge

            Re: For those who wonder...

            Renewables tend to be able to come on line very, very quickly to meet surges in demand. Thermal systems take time to build up a head of steam.

            I think the reference was to "mass" renewables like wind and solar. Pump storage is really a big battery, not a source of power in itself.

            The issues with sources like wind and solar is that due to politicians swallowing the green lobby lies hook line and sinker (not to mention the big trough with plenty of snouts guzzling from it), renewables get first pick of the load - so they are all generally running at whatever they can produce at the time. I nearly wrote "flat out", but windmills almost never reach their flat out capability !

            Because of this, there are no renewables sitting there waiting to be called upon to manage short term variations in demand.

            Instead they have highly variable output, thus adding massively to the control needed to keep the system in balance - with other sources (typically open cycle gas turbine) having to ramp up/down or stop/start completely in order to compensate for the variation in supply from the renewables AS WELL AS changes in demand.

            There are also generators (mostly diesel) sat around doing nothing but waiting to keep the lights on when a drop in supply (no wind on those nice crisp winter days, and no solar outside the few short hours of daylight) corresponds with a peak in supply (such as on those same crisp winter days ! Search for articles on diesel generator farms - the green lobby won't tell you about the cost of paying standby subsidies to them as an externalised cost of their so called "cheap" and "green" electricity.

            1. bombastic bob Silver badge
              Devil

              Re: For those who wonder...

              "The issues with sources like wind and solar is that due to politicians swallowing the green lobby lies"

              I'm with you on that last part (swallowing the green lobby lies), but I think we should still use wind and solar generators, because other forms of energy are still finite, and I doubt the EU (or post-brexit UK) wants to prop up the price of oil by using even more of it for electricity. If you want energy independence, you use more of what you have, and less of what you must import. So there ya go. Besides, most of the trouble in the Middle East is ultimately caused by too much world money going into the hands of people who tend to support things _like_ terrorism, and we don't need Middle East politics affecting the world economy so much, now do we?

              So even if the 'renewables' policy was driven by 'human caused global warming' hysteria, it's still a reasonable outcome to have plenty of solar, wind, hydro, geothermal, and whatever other 'renewable' someone can think up and implement at a reasonable cost to the consumer.

              Then, on non-windy winter nights, you can burn the midnight oil for power [so to speak]. But the rest of the time you can get it from Mr. Sun instead.

            2. Deckard_C

              Re: For those who wonder...

              Nuclear gets first pick as they need near 100% output all the time to have a chance of the investment being paid for.

              Coal need to be running constantly but can vary the output, only generally shutdown for there summer maintenance.

              The gas power stations are the one which handle most of the large changes in load, for the UK anyway..

              Hydro and pumped hydro handle the quick changes in load

              Wind is interesting as when there is plenty of power wind turbines are sometimes stopped, because they are the easiest to stop I don't know if that's techincal reasons or contractually. With a power grid being able to stop or reduce supplying power can be as important as providing that power on demand.

          3. Dodgy Geezer Silver badge

            Re: For those who wonder...

            ...Renewables tend to be able to come on line very, very quickly to meet surges in demand...

            Er... no. Pumped storage can be brought online from 'totally stopped' very quickly - but if you want 'very, very quickly' you want 'spinning reserve' - which is typically gas turbine, though it could in theory be anything.

            'Renewables' are most famous for being 'non-dispatchable' - which means that you can't guarantee that they will come on-line at all. Pumped storage, for instance, can only come on-line if there is enough water pumped up the mountain - and that will only last for an hour or so...

    4. itzman

      Re: For those who wonder...

      So much misunderstanding and limiting understanding

      AC is great because its easy to change voltages using just an old lump of iron and some copper windings.

      It's also great because its easy to generate - being what naturally comes out of a magnet rotating inside a bunch of coils - or vice versa.

      It's also great because it shows that the dynamos are slowing down under load, thus slowing down all synchronous motors attached to the grid and lowering the load.

      Its bad for very big grids. Or undersea cables. Because very big grids have enough propagation delay to make a lot of out of phase current circulate - as do undersea links , on account of the massive capacitance they exhibit between conductors and ground.

      This leads to greater losses due to cable resistance.

      The UK grid is stabilised to have a long term average of EXACTLY 50Hz, though it will fluctuate under load.

      It is not connected directly to the European Grid. All links are DC with frequencies at either end being independently variable.

      The move away from turbines & transformers towards electronic inverters is not altogether welcome. Spinning mass accounts for the best short-term storage on the grid there is. load fluctuations are dealt with my the rotational inertia.

      Adding massive amounts of highly variable - and over short time scales, as windmills trip in and out and clouds cover solar panels - renewable energy places a BIG strain on grid management. Germany has already had to relax its frequency control standards to accommodate Energiewiende and many factories that depend on 50hz stability have been forced to essentially install what amounts to massive crystal controlled UPS'es. Increasing electricity consumption still further.

      Meanwhile the rapid deployment of power stabilised switch mode PSUs has meant that lowering frequency and/or voltage does not lower power drawn, thus depriving grids of a simple and natural way to keep supply and demand balanced.

      We are fortunate that we have always run a separate grid even before Brexit.

      Like most of the EU, the European grid is too big to manage and suffers too much political interference to be efficient.

      1. Anonymous Coward
        Anonymous Coward

        Re: For those who wonder...

        "AC is great because its easy to change voltages using just an old lump of iron and some copper windings."

        If your device can stand the heat dissipation then it is cheaper to use a large resistor to drop the voltage. For many years TVs had that system - a transformer was just too bulky and expensive. Earlier radios did too - when the UK pre-grid could be either DC or AC from local power stations. There were valves - notably the "U" types - that had a standard current rating for their heaters so they could be wired in series with a resistor taking up the slack.

        One big disadvantage - apart from wasting power as heat - was that the chassis was connected to one side of the mains. With only two wires in the mains lead it was not unusual to feel a tingle from a wooden case when the connections were reversed and the chassis was live.

      2. Fruit and Nutcase Silver badge

        Re: For those who wonder...

        @itzman

        thus depriving grids of a simple and natural way to keep supply and demand balanced.

        At some point, they'll be rolling out the ability to throttle the usage of the domestic customers via the Smart Meters (whenever that farce manages to get its house in order), and thus be able to control the demand on the grid.

        1. JulieM Silver badge

          Re: For those who wonder...

          I have got my own Solar power system, which exports anything I don't use locally into the Grid; and I've been considering installing a system with some big batteries and a smart charger (probably Raspberry Pi-based); which will continuously read the telemetry data from my Inverter to determine how much power is available for charging the batteries, and set the charging current accordingly so as only to use Solar energy, and no energy drawn from the Grid, for charging the batteries.

          If I install a web server on the Raspberry Pi charge controller, then with a few custom CGI scripts I should be able to see with a mobile phone or laptop just how much energy I'm producing from the Sun, and even ask for some watts temporarily so I can power an appliance such as a 1kW travel kettle (for a not-so-quick brew) without drawing unnecessarily from the grid.

          Then at night, I'll switch over my lighting circuits to another inverter, this one not synchronised to the grid, powered from the batteries charged during the day, and then back to the grid if the batteries become depleted before morning. And obviously I'll be able to run other appliances from the secondary inverter, subject to power availability (my greediest appliance is my washing machine, which peaks out at 2500W while heating the water; so I might just get away with a 3kW inverter.)

          Even if I'm not totally self-sufficient, I should be able to minimise the amount of energy I'm drawing from the grid. The Solar system does need the grid connection, though, because it has to have something to lock against -- and it stops producing AC power altogether if the grid connection is lost. This is for the protection of any people working on the line -- they won''t be expecting there to be any power coming out of the end that goes to consumers -- and also because you can't be sure there isn't any more load downstream of the break, that it would be trying to export into.

          1. Stoneshop

            Re: For those who wonder...

            I have got my own Solar power system, which exports anything I don't use locally into the Grid; and I've been considering installing a system with some big batteries and a smart charger (probably Raspberry Pi-based); which will continuously read the telemetry data from my Inverter to determine how much power is available for charging the batteries, and set the charging current accordingly so as only to use Solar energy, and no energy drawn from the Grid, for charging the batteries.

            I'm contemplating something similar, except that initially I'll be going to install an electrically heated boiler (or two), powered if there's excess power from my panels. If that happens to become depleted of thermally enhanced H2O, then a valve will switch the demand to the hot water supply tied to the gas-powered central heating system. Once that's done I may start looking into electrical storage, although an electric vehicle might well cover that, in a way.

            For your setup you're basically building a big-ass UPS powering (part of) your house, going into bypass mode if its batteries are down or load demand exceeds its capacity, and being charged off solar.

            The Solar system does need the grid connection, though, because it has to have something to lock against -- and it stops producing AC power altogether if the grid connection is lost. This is for the protection of any people working on the line

            The manufacturer of the inverter I have fitted (SMA) offers a setup that turns a standard mains-tied inverter into one capable of 'island' operation (they also have 'island'-mode inverters, but those are meant for fully mains-independent operation), one component of which is a relay fitted in your incoming mains feed. I expect that one's utility co. will want such an installation to bear all relevant certification marks and then some, not just someone's Arduino-controlled flimflam.

            1. JulieM Silver badge

              Re: For those who wonder...

              I haven't the generation capacity for full self-sufficiency (best I can ), but I figured I might as well make the fullest use possible of my homebrew juice. The intention is to have 5 kWh of storage capacity (the 3kVA UPS I'm looking to use as an inverter needs 48 volts, and a bank of 110Ah traction batteries would hold this much).

              I'm going to use a Raspberry Pi as opposed to an Arduino, just so I can do my complicated PID control business logic stuff in a girly, high-level (assembly is for blokes), interpretated language. And for a CGI-enabled web server, so the mobile phone need only be running a web browser.

              In the (highly unlikely) event of loss of grid supply, I should be able to run the solar system in "island" mode to charge the batteries, as long as I isolate it totally from the grid and give it a local voltage and frequency source to lock against.

              1. Anonymous Coward
                Anonymous Coward

                Re: For those who wonder...

                @JulieM "I'm going to use a Raspberry Pi"

                Most Arduino programs are in "C" ..... and this will help https://playground.arduino.cc/Code/PIDLibrary you will find web server libraries too.

                As (a bloke) who many years ago wrote assembly code for commercial PID programmers I would strongly recommend that you also implement a "watch dog timer" to make sure that when (not if) your controller crashes it recovers gracefully without burning your house down.

            2. Killing Time

              Re: For those who wonder...

              @Stoneshop

              'I'm contemplating something similar, except that initially I'll be going to install an electrically heated boiler (or two), powered if there's excess power from my panels.'

              Your boilers would be pretty pointless without a means to store the hot water so assuming you have a hot water tank or a 'heatstore' the solution exists. There is no point in reinventing the wheel, checkout the 'Solar iBoost' or similar devices.

              It monitors any power export automatically and routes the excess power to your immersion heater. If you are on a UK export tariff you are being paid for the export and its actually ending up as free hot water for you.

              1. Stoneshop

                Re: For those who wonder...

                Your boilers would be pretty pointless without a means to store the hot water so assuming you have a hot water tank or a 'heatstore' the solution exists. There is no point in reinventing the wheel, checkout the 'Solar iBoost' or similar devices.

                In my part of the world, a hot-water boiler is a tank holding several tens up to a couple hundred liters of H2O, with a heating element (electrical, heat exchanger coupled into central heating, or independently fuelled), so my setup will not lack the storage capacity you think it lacks.

                And buying ready-made kit: where's the fun in that? I also note that the Solar iBoost in particular (and I expect equivalent devices as well) duplicates stuff I already have in place, and the clamp wouldn't work with the cabling as present at the point where it's meant to be installed anyway.

  7. Herby
    Joke

    Power frequency off??

    Looks like it is more than a few cycles off. The correct frequency for power is 60 Hz. So, they are a bunch (10 Hz) off according to my calculations. They should be spinning faster (3600 RPM).

    Get with the program!

    1. drgeoff

      Re: Power frequency off??

      There are more places inthe world than the USA. All of Europe, half of Japan and many other places use 50 Hz.

      And to correct another poster, Japan is 100 volts not 110.

  8. Jay Lenovo
    Pint

    What's the frequency Kenneth?

    The Balkans, like native Nikola Tesla, are both amazing and prone to acts of wackiness. No wonder the frequency is a bit off.

  9. Fruit and Nutcase Silver badge
    Flame

    240/230/220V

    Some might call that "taking back control".

    The EU Single Market membership robbed us of 10V. Do we get it back after Brexit?

    Our kettles can take longer to boil (depending on the local substation)

    1. This post has been deleted by its author

    2. Alan Brown Silver badge

      Re: 240/230/220V

      My UK line voltage is 252V - kettles boil quickly, but light bulbs don't last long.

    3. Mike Shepherd
      Meh

      Re: 240/230/220V

      The EU Single Market membership robbed us of 10V. Do we get it back after Brexit?

      No country needed to change voltage. Compatibility was achieved by widening the tolerance band.

      1. Fruit and Nutcase Silver badge

        Re: 240/230/220V

        The Electricity Safety, Quality and Continuity Regulations 2002

        http://www.legislation.gov.uk/uksi/2002/2665/regulation/27/made

        "(2) ...and the voltage declared in respect of a low voltage supply shall be 230 volts between the phase and neutral conductors at the supply terminals.

        (3) ...the permitted variations are—

        (b)in the case of a low voltage supply, a variation not exceeding 10 per cent above or 6 per cent below the declared voltage at the declared frequency;"

        ...which gives us 230V +10%/-6%

        Yes, tolerances changed to accommodate the supply ranges across the EU, including some areas of the UK which still had nominal 250V, and the more usual 240V all fall in range.

        As substation equipment/supply transformers gets replaced over a lifetime, the new equipment will supply a nominal 230V, as per EU harmonisation rules,

    4. Mage Silver badge

      Re: 240/230/220V

      It didn't. They changed the % limits for Mainland Europe, GB, NI and Ireland. It's mostly 240V in UK, 220V in Germany and 230V in Ireland, like it always was. Equipment marked CE has to work in any EU electrical supply, though it MUST have the local standard of plug, if it plugs in.

      The most criminal is Amazon supplying 5V chargers with a CE mark. Yet they only have the USA two blades, illegal in all of Europe to retail. The actual electronics like most SMPSU do seem to work 90V to 245V.

      Older 220V gear with a transformer and no taps (Such as German Market radios) may need a dropper resistor in the UK.

      1. Fruit and Nutcase Silver badge

        Re: 240/230/220V

        The regulation was not for the end user equipment, but the electricity supply equipment.

        By harmonising the supply in the EC to 230V, you make it possible for manufacturers of the supply equipment to produce equipment (substations/transformers) that can be used in any of the countries of the single market, thereby driving down the cost as it means the manufacturers do not have to make equipment of different specification by country.

        The small change in supply specification meant "most" end user equipment "could" tolerate the change. And also meant the electricity supply companies did not need to replace their equipment outside of normal life/serviceability.

        Your local substation may be supplying 240V (or 220V) today. If that substation develops a fault and the electricity company were to replace the transformer with new ones tomorrow, then, the new supply voltage will be 230V, as when they went to purchase it, the market only offered 230V (and not 240 or 220V) transformers

        1. SImon Hobson Bronze badge

          Re: 240/230/220V

          Your local substation may be supplying 240V (or 220V) today. If that substation develops a fault and the electricity company were to replace the transformer with new ones tomorrow, then, the new supply voltage will be 230V, as when they went to purchase it, the market only offered 230V (and not 240 or 220V) transformers

          Wrong on several counts.

          Firstly, these are not things that they mass produce and put on the shelf, they tend to be made to order and different areas have different specs anyway. In any case, the difference in winding between 230 and 240V is not major.

          And after that, at the substations they have devices called tap changers, the transformers actually have multiple connections at one end of the winding so you can add or remove a few turns to alter the output voltage. On your local substation (11kV down to 240/415V in the UK) these are manual, on the larger substations (132kV to 33kV and 33kV to 11kV) they are remotely controlled - allowing the control centre to adjust the taps according to power flows.

          One of the problems caused by the increasing amount of embedded generation (specifically solar on house roofs) is that the power flows can change significantly, raising the supply voltage at the consumers' terminals. In many cases they've had to change tap setting to reduce the voltage generally - but they don't like doing this because ...

          It's in the interest of the DNO (Distribution Network Operator) to keep the system voltages as high as they can within the limits imposed on them. The higher the voltage, the lower the current for a given amount of power delivered - and hence the I^2R losses are lower. Eg, roughly speaking, if you drop the voltage by 5%, the current required goes up by 5%, and the resistive losses increase by about 10%. As well as increasing the losses as a proportion of the power delivered to customers (and hence charged for), cables are limited by current rather than power - so if the cables are already approaching the limit, then reducing voltage would mean also having to reduce the power delivered, which can only be done by moving some load to a different supply (typically expensive) or upgrading the cable (almost always expensive). So keeping the supply voltage as high as possible means they can put off potential upgrades as demand grows.

          1. Stoneshop

            Re: 240/230/220V

            Firstly, these are not things that they mass produce and put on the shelf, they tend to be made to order and different areas have different specs anyway. In any case, the difference in winding between 230 and 240V is not major.

            The really big ones are not mass produced[1], but the transformers used in local substations are used in numbers that warrant small-series production. And looking at the storage yard of one of the regional network operators that happens to be nearby, there are a few transformers in there. The largest ones you could just about load on a flatbed truck without then needing 'Transport Exceptionnel' escort, to give you an indication of their size.

            [1] I occasionally go past the Smit Transformatoren factory, and they usually have four to six of those, almost always all different, standing outside being readied for shipment. Impressive stuff.

            1. Killing Time

              Re: 240/230/220V

              'The really big ones are not mass produced[1], but the transformers used in local substations are used in numbers that warrant small-series production.'

              They are all made to order above a few tens of KW output, in part due to the fluctuations in copper price. Local substation distribution units are generally about 0.75MW and above. If your regional DNO holds multiple spares then that is their choice so that they can react swiftly to a unit failure rather than wait the 12-16 weeks lead time.

      2. Stoneshop

        Re: 240/230/220V

        It didn't. They changed the % limits for Mainland Europe, GB, NI and Ireland. It's mostly 240V in UK, 220V in Germany

        Nope. Germany, as well as the rest of continental Europe, is at 230V now.

        Equipment marked CE has to work in any EU electrical supply, though it MUST have the local standard of plug, if it plugs in.

        That's why non-grounded (doubly insulated) equipment is fitted with Euro plugs: two-prong plugs that fit the various sockets in use.

    5. Insert sadsack pun here

      Re: 240/230/220V

      Kosovo would love to be in the EU. The UK wants to get out. We could just do a swap...

      1. Yet Another Anonymous coward Silver badge

        Re: 240/230/220V

        Kosovo would love to be in the EU. The UK wants to get out. We could just do a swap...

        Or you could just rename it United Kosovo, turn up at the meetings and hope nobody noticed?

    6. drgeoff

      Re: 240/230/220V

      Another urban myth. There was no change to the actual voltage. http://coffeetime.wikidot.com/uk-eu-mains-voltage-harmonisation

  10. Paul Hovnanian Silver badge

    No problem

    We just need 1.21 Gigawatts to get back to the future.

    1. Fruit and Nutcase Silver badge
      Joke

      Re: No problem

      We just need 1.21 Gigawatts to get back to the future.

      I'll just pop down to Maplin and get a flux capacitor. Oh, wait!

  11. sanmigueelbeer Silver badge

    NTP anyone?

    1. David Shaw

      'free' German NTP?

      I bought a bunch of Meinberg hardware references, for work

      I also use them for home amongst a few other different tools for time-sync using - for example weak signal propagation reporter (WSPR or the WSJT-X) Ham Radio digi-modes, from https://physics.princeton.edu/pulsar/k1jt/

      you could try Meinbergs's NTP software for Windows

      https://www.meinbergglobal.com/english/sw/ntp.htm#ntp_stable

      they used to have a special offer, write them a mail saying why you'd like to become an NTP node, and they might send you all the hardware!

  12. JulieM Silver badge
    Pint

    Maybe the Young Earth Creationists have it right

    Maybe the Young Earth Creationists have it right. We've been comparing the mains frequency against atomic clocks, but radioactive decay rates can't be constant if the Earth is only 6000 years old .....

    Too many of these --> ?

    1. bombastic bob Silver badge
      Angel

      Re: Maybe the Young Earth Creationists have it right

      "radioactive decay rates can't be constant if the Earth is only 6000 years old"

      I don't think that even the 6000 year old Earth crowd believes THAT. Instead the theory is that the earth was created with isotopes "in that state", including light beams from distant stars and stuff like that.

      just playing "angel's advocate" on this one. I'm familiar with creationist arguments and some of them have merit. Claiming the earth is only 6,000 years old, however, is completely clueless. [you can still reconcile biblical creation against evolution, if you don't take it word-for-word literally, but those who do take it literally are subjecting themselves to rules they can't possibly defend. Still, creationism is an interesting study, at least to me].

      but yeah, it's convenient to 'hand wave' the science when your god can just create it "that way"

      1. Anonymous Coward
        Anonymous Coward

        Re: Maybe the Young Earth Creationists have it right

        "[...] when your god can just create it "that way""

        But the god that created "your god" gave it the ability to do that.

        Or I should say [the god who created the god]*** who created "your god" gave it the ability to do that.

        ***[the god who created the god] is an infinite recursion. It may be said that there is no recursion element and "your god" has always existed without itself being created by a superior being. In which case Occam's Razor proposes it is more probable that the effect of the universe has always** existed without any such a creator.

        **whatever "always" means when time itself is an artefact of the effect.

  13. M7S
    Coat

    This situation with the clocks being wrong just winds me up and

    ...I get really ticked off about it

  14. Anonymous Coward
    Anonymous Coward

    8008L355

    So TITSUP they flew off and now 8008L355

  15. This post has been deleted by its author

  16. Chairman of the Bored

    US grid to float high in freq:

    Regulatory bods to let grid drift generally higher in freq to improve stability and reliability, conditioned on the assumption nobody cares about grid-based time:

    https://spectrum.ieee.org/tech-talk/energy/the-smarter-grid/power-system-experiment-in-us-means-clocks-will-speed-up

    During peak demand the US grid would always run a bit slow, by fractions of a percent, and then run high at night to compensate. My high school had these massive IBM clocks in the hallways, looming darkly from the ceiling. One outside every classroom and all running running in inhuman lockstep. Awful "bang" every minute when the hands would move. Imagining you were in some sort of penal or mental facility was not difficult. I think they were on grid freq and could swear during some classes that freq was actually tens of percent low.

    1. Anonymous Coward
      Anonymous Coward

      Re: US grid to float high in freq:

      "My high school had these massive IBM clocks in the hallways,[...]"

      After restarting from a power cut it was not unusual for a classroom clock to go backwards. Presumably a design economy by omitting the spring mechanism that automatically corrected a "wrong way" start by the synchronous motor. IIRC there was a button to manually force a change in direction - for which the school caretaker needed to use a stepladder. He had to go round resetting the time anyway.

    2. Anonymous Coward
      Anonymous Coward

      Re: US grid to float high in freq:

      "Awful "bang" every minute when the hands would move. Imagining you were in some sort of penal or mental facility was not difficult."

      I would suspect that there was a master clock mechanism somewhere - with a wired connection to the slave clocks.

      Every minute it would send a shaped pulse to advance the slave clocks by an escape mechanism - hence the bang.

      When a slave clock's hands reached the hour point then it would ignore any further "minute" pulses. The master clock would then send a differently shaped "hour" pulse to get the slaves past that point.

      At one minute to the hour the master clock would actually send several "minute" pulses in quick succession. These pulses would compensate for a slave clock having missed a few previous pulses - and get it to the hour waiting point. The "hour" pulse could also be repeated several times to release any slave that had missed the first one.

      If a slave clock interpreted noise as a "minute" pulse it would merely get to the hour stop a little earlier - and then wait. Noise that could be interpreted as an "hour" pulse would be ignored except at that critical point. That case might put the slave clock a minute fast during the next hour only. It would resync at the next hour wait point.

      Daylight saving can be achieved quite simply. To advance the time the master clock sends a rapid series of "minute" pulses to get the slaves to the next hour. To retard the time it can merely omit the relevant "hour" pulse - keeping the slaves stationary for an hour. Not a problem if no one is likely to be wanting to know the time in the middle of the night.

      MSF movements for analogue clocks do that - I watched one at 2am to satisfy my curiosity the first time the clocks went back.

      The most novel master clock mechanism is possibly an early 20th century one. It used a very accurate pendulum mechanism - but avoided the problem of friction from a mechanical signal transfer for the slaves. The pendulum shaft had a carefully shaped hole in it. A light shone through the hole onto a photoelectric cell. The electrical output was a sine wave.

      1. Anonymous Coward
        Anonymous Coward

        Re: US grid to float high in freq:

        "Noise that could be interpreted as an "hour" pulse would be ignored except at that critical point. That case might put the slave clock a minute fast during the next hour only."

        Oops. Erratum. The slave clock would initially be fast at one minute past the hour. It would ignore the genuine "hour" pulse - so would be back in sync after the first "minute" pulse taking them all to two minutes past the hour

        1. Chairman of the Bored

          Re: US grid to float high in freq:

          @AC, all posts - thanks for the insight; genuinely learned something about the time beasts that used to rule my life.

  17. Dejan

    And the Serbian electrical grid company, while being responsible for the electrical grid in that part of the region of Europe in general, points the finger at Kosovo privider KOSTT for the "missing" electricity..

    You can read their response to the ENTSO-E report here:

    http://www.ems.rs/news_more.php?id=17818

  18. Daniel B.
    Coat

    I know what happened

    It's Hackerman, he hacked too much time!

    Mine's the one with the Kung Fury logo.

  19. chivo243 Silver badge
    Coat

    Across the channel->NL

    Both my oven and microwave clocks have lost time, each losing a different amount of time as well... and ofcourse it's a Dutch Oven...

  20. Voidstorm
    Go

    While not so relevant to the exact topic...

    http://www.usno.navy.mil/USNO/time/master-clock/systems-of-time

    Is worth a gander if you want to get reeeealy picky about timing precision. Enjoy ;)

  21. Anonymous Coward
    Anonymous Coward

    If you want good time keeping in a mechanical watch...

    Then Rolex is your only real choice. They keep staggeringly good time.

  22. Anonymous Coward
    Anonymous Coward

    Still getting worse?

    According to the swissgrid.ch page, they correct a disparity by aiming for 49.990 Hz or 50.010 Hz. So if everything were under control they ought to be catching up by about 15 seconds a day. However, since this article was published on Friday they seem to have lost another second. Perhaps they will soon be out by more than six minutes.

    1. Anonymous Coward
      Anonymous Coward

      Re: Still getting worse?

      They've done it: at 14:56 UTC I read "Current grid time deviation -360.177 s". Congratulations.

      1. Hotears

        Re: Still getting worse?

        And another second. And a half.

        http://n1.taur.dk/eurogrid.png

        I began monitoring the power line awfully early sunday morning; It looked reasonably under control for the first 24 hours, now it is entertaining again. If you are the type who enjoys waching phase drift.

        This is plotted every 20 minutes, so jumping on the refresh button more often than that won't help.

        Gigantic data file available on request.

  23. Anonymous Coward
    Anonymous Coward

    Modified mains frequency

    Wonder what the effect of setting the frequency high for n weeks by a few microHz would be?

    would this cause things like motors to run slightly faster? also relevant could sensitive equipment be messed up.

    I noticed that my LED bulbs keep flickering at 1am, also lost a ceiling light for no apparent reason a week back. Possibly excess dissipation in the heater as it was working perfectly up until it failed catastrophically.

    Mine's the one with the ebay surplus atomic clock and Wifi sender, on the hour every hour using nanoceramic antennas on every clock in the house to make damn sure they are accurate to +/- 90ms :-)

POST COMMENT House rules

Not a member of The Register? Create a new account here.

  • Enter your comment

  • Add an icon

Anonymous cowards cannot choose their icon

Other stories you might like