Accuracy of UK power grid time control?

[Christopher Tidy]

Hi all,

Thanks very much for all the information. Sorry for the original
cross-post; I was expecting very few replies.

In answer to some of your questions, the clock is a WWII German
"Dehomag" slave clock, originally designed to be driven by a master
clock. It has no second hand. The minute hand appears to be balanced,
but the hour hand not. The clock is about 14" across, so not huge, but
it's exceptionally heavy as nearly everything is made from 1 mm steel
plate. Here's a picture of the clock:

http://www.mythic-beasts.com/~cdt22/dehomag.jpg

I can't decide whether to repaint it black, as it was when manufactured,
or to leave it grey. It's for my bedroom, so if it were to stop during
power cuts I wouldn't mind. It sounds like the 50 Hz grid frequency will
be a plenty accurate time keeping source, and it's an interesting
solution to the problem. We live in a pretty rural area so I'll try to
design a circuit which will be fairly resistant to noise. I had already
planned to put a Schmitt trigger on the input from the step-down
transformer.

Once again, thanks for all the advice. It's much more than I got from
the National Grid - the phone number on their website doesn't even work!

Best wishes,

Chris

 

[John G]

Hi all,

Thanks very much for all the information. Sorry for the original
cross-post; I was expecting very few replies.

In answer to some of your questions, the clock is a WWII German
"Dehomag" slave clock, originally designed to be driven by a master
clock. It has no second hand. The minute hand appears to be balanced,
but the hour hand not. The clock is about 14" across, so not huge, but
it's exceptionally heavy as nearly everything is made from 1 mm steel
plate. Here's a picture of the clock:

http://www.mythic-beasts.com/~cdt22/dehomag.jpg

I can't decide whether to repaint it black, as it was when
manufactured, or to leave it grey. It's for my bedroom, so if it were
to stop during power cuts I wouldn't mind. It sounds like the 50 Hz
grid frequency will be a plenty accurate time keeping source, and it's
an interesting solution to the problem. We live in a pretty rural area
so I'll try to design a circuit which will be fairly resistant to
noise. I had already planned to put a Schmitt trigger on the input
from the step-down transformer.

Once again, thanks for all the advice. It's much more than I got from
the National Grid - the phone number on their website doesn't even
work!

Best wishes,

Chris

You did not tell us what the drive mechanism is.
If it was similar to the Simplex/IBM master clocks of the time it most
likely advanced every minute by a pulse to an electromagnet.

Now there are several ways you could generate that.
A Small timer motor with a cam like a washing machine timer.
Or an Electronic cct synched by the mains.

If it is for your bedroom you may not want it after the first few hours
as the electromagnet noise will drive you bonkers.

 

[James Sweet]

In the U.S. the power grid is very accurate over long periods of
time. But what are the odds of running six months without a single
power interruption?

Depends where you are, I used to have outages a couple times a year, now
the house I'm in I've had exactly one outage in almost 2 years and it
was a pretty good storm that went through.

Aside from that though in both the US and UK the mains frequency has
excellent long term stability. It may gain or lose a few seconds over
the course of the day but it will be dead on over weeks/months.

 

[Olaf Peuss]

If it's critical that the clock not stop you can put it on a UPS battery
backup.

Great idea! In order to save the odd $40 for a quartz oscillator (read
the OP!), you spend at least $100 on a UPS device.

SCNR and best regards,
OP

 

[Christopher Tidy]

John G wrote:

You did not tell us what the drive mechanism is.

I did but you probably missed it in my second post. It's a stepper
motor, as I believe is common on many slave dials made in continental
Europe, but relatively uncommon in the UK.

If it is for your bedroom you may not want it after the first few hours
as the electromagnet noise will drive you bonkers.

The noise is almost all produced by a ratchet which prevents reverse
rotation of the rotor. It isn't quite as disturbing as the "clunk-click"
of the Gent mechanisms - it's more of a muffled "kerthunk".

Chris

 

[n cook]

Hi all,

I'm thinking of building an electronic clock control circuit which uses
the 50 Hz mains frequency for time keeping. The reason for this is that
the clock dial is rather large, so probably wouldn't run for long on
battery power, and I don't fancy spending £40 buying a programmable
quartz oscillator chip.

Despite doing some research online and in electrical engineering books,
I can't find a figure for the accuracy of the time keeping of the UK
power grid. Perhaps this is because there aren't official bounds set for
the time error - one of my electrical engineering books says it is a
legal requirement that the frequency be kept "as close as possible to 50
Hz" - but even if this is the case it should be possible to determine a
typical error figure.

From my point of view I'd regard an acceptable error as 5 minutes every
6 months. Does anyone know the typical time error seen on the UK grid,
or where I might find this information? Any suggestions would be much
appreciated.

Best wishes,

Chris Tidy

Some future use for the dips in mains frequency use to reduce National Grid
load
http://www.publications.parliament.uk/pa/ld200405/ldhansrd/pdvn/lds06/text/6
0324-07.htm
starting at heading
Dynamic Demand Appliances Bill [HL]

" Dynamic demand appliances contain a low-cost electronic microcontroller.
This listens to the mains hum, which runs at a frequency of around 50 hertz.
The signal can be detected through every plug socket connected to the
national electricity supply. Through this signal, the dynamic demand
appliances can sense whether the National Grid is under stress and adjust
the time at which they use electricity. The technology is suitable for
appliances that already switch on and off during the day on a "duty cycle",
such as domestic and industrial fridges, freezers and water heaters.

Millions of such appliances acting together would smooth out demand for
electricity. "

 

[Alan J. Wylie]

I wrote a more detailed article about this a few years ago, which
discusses various notable historic events, like how the power grid
had to handle the majority of the UK using the toilet at the same
instant, which resulted in the largest ever surge in demand on the
UK power grid (which with advanced planning, it handled just
fine)...
http://groups.google.com/group/sci.engr.lighting/msg/99b03c9711a4f753?hl=en&

You mention Dinorwic/Dinorwig - there's a more detailed posting about
it at

<http://groups.google.com/group/uk.rec.subterranea/msg/dd48c794775000bf>

 

[Doki]

Indeed. Part of the standard equipment in the old days was a special
'crystal oven' with tightly controlled temperature. By regulating the
temperature of the crystal inside, the accuracy its vibrations was
improved.

An old 'urban lengend' was that the first quartz watches were calibrated
assuming the temperature of the crystal was going to be controlled by the
body heat of the wearer. And that leaving your watch on the dresser over
the week-end would cause it to slow down slightly. Don't know if it is
really true, but it's a nice story.

My cheap and nasty Casio digital watch seems to have lost 2 minutes after I
left it first in my old car, and then in my old car in the shed. Neither
place was particularly warm...

 

[Rob Morley]

Hi all,

I'm thinking of building an electronic clock control circuit which uses
the 50 Hz mains frequency for time keeping. The reason for this is that
the clock dial is rather large, so probably wouldn't run for long on
battery power, and I don't fancy spending £40 buying a programmable
quartz oscillator chip.

Despite doing some research online and in electrical engineering books,
I can't find a figure for the accuracy of the time keeping of the UK
power grid. Perhaps this is because there aren't official bounds set for
the time error - one of my electrical engineering books says it is a
legal requirement that the frequency be kept "as close as possible to 50
Hz" - but even if this is the case it should be possible to determine a
typical error figure.

From my point of view I'd regard an acceptable error as 5 minutes every
6 months. Does anyone know the typical time error seen on the UK grid,
or where I might find this information? Any suggestions would be much
appreciated.

I have a vintage synchronous clock which seems to keep good time over a
period of months.

 

[Andrew Sinclair]

Some future use for the dips in mains frequency use to reduce National Grid
load
http://www.publications.parliament.uk/pa/ld200405/ldhansrd/pdvn/lds06/text/6
0324-07.htm
starting at heading
Dynamic Demand Appliances Bill [HL]

" Dynamic demand appliances contain a low-cost electronic microcontroller.
This listens to the mains hum, which runs at a frequency of around 50 hertz.
The signal can be detected through every plug socket connected to the
national electricity supply. Through this signal, the dynamic demand
appliances can sense whether the National Grid is under stress and adjust
the time at which they use electricity. The technology is suitable for
appliances that already switch on and off during the day on a "duty cycle",
such as domestic and industrial fridges, freezers and water heaters.

Millions of such appliances acting together would smooth out demand for
electricity. "

Have a look here - an on-line frequency monitor.

http://www.dynamicdemand.co.uk/grid.htm

The rest of the website covers the topic of dynamic demand control which
is quite interesting (well it is if you work, as I do, for the company
that owns 70% of Dinorwig).

Andy

 

[Andy Wade]

From my point of view I'd regard an acceptable error as 5 minutes every
6 months. Does anyone know the typical time error seen on the UK grid,
or where I might find this information? Any suggestions would be much
appreciated.

You can download more information than you ever wanted to know about the
National Grid from here:

http://www.nationalgrid.com/uk/Electricity/Codes/gridcode/gridcodedocs/

I've pasted two relevant snippets below (NGET stands for National Grid
Electricity Transmission plc).

1. From the definitions:

Target Frequency
----------------
That Frequency determined by NGET, in its reasonable opinion, as the
desired operating Frequency of the Total System. This will normally
be 50.00Hz plus or minus 0.05Hz, except in exceptional circumstances
as determined by NGET, in its reasonable opinion when this may be
49.90 or 50.10Hz. An example of exceptional circumstances may be
difficulties caused in operating the System during disputes affecting
fuel supplies.

2. From document BC3, which deals with the frequency control process:

BC3.4.3 Electric Time
---------------------
NGET 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.


All very well, but ISTM that the biggest problem with using the mains
alone as a time standard is power cuts, after which you'll always need
some absolute standard such as MSF, GTS, NTP, etc. to reset your clock.

 

[n cook]

Some future use for the dips in mains frequency use to reduce National Grid
load
http://www.publications.parliament.uk/pa/ld200405/ldhansrd/pdvn/lds06/text/ 6
0324-07.htm
starting at heading
Dynamic Demand Appliances Bill [HL]

" Dynamic demand appliances contain a low-cost electronic microcontroller.
This listens to the mains hum, which runs at a frequency of around 50 hertz.
The signal can be detected through every plug socket connected to the
national electricity supply. Through this signal, the dynamic demand
appliances can sense whether the National Grid is under stress and adjust
the time at which they use electricity. The technology is suitable for
appliances that already switch on and off during the day on a "duty cycle",
such as domestic and industrial fridges, freezers and water heaters.

Millions of such appliances acting together would smooth out demand for
electricity. "

Have a look here - an on-line frequency monitor.

http://www.dynamicdemand.co.uk/grid.htm

The rest of the website covers the topic of dynamic demand control which
is quite interesting (well it is if you work, as I do, for the company
that owns 70% of Dinorwig).

Andy

What a swiz, I wanted to see what the real time meter would do at 19.15
tonight

Reason


V





V






V






V








V






V

Dr Who returns

 

[Derek ^]

That Frequency determined by NGET, in its reasonable opinion, as the
desired operating Frequency of the Total System. This will normally
be 50.00Hz plus or minus 0.05Hz, except in exceptional circumstances
as determined by NGET, in its reasonable opinion when this may be
49.90 or 50.10Hz. An example of exceptional circumstances may be
difficulties caused in operating the System during disputes affecting
fuel supplies.

2. From document BC3, which deals with the frequency control process:

BC3.4.3 Electric Time
---------------------
NGET 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.


All very well, but ISTM that the biggest problem with using the mains
alone as a time standard is power cuts, after which you'll always need
some absolute standard such as MSF, GTS, NTP, etc. to reset your clock.

The traditional approach assuming the outages were brief, was to use a
cheap oscillator (an astable mutivibrator) running from a back up
battery when the mains feed was down.

DG

 

[Andrew Sinclair]

What a swiz, I wanted to see what the real time meter would do at 19.15
tonight

Wouldn't see any difference. The TARDIS has an internal power supply
and is not reliant on a connection to the National Grid.

Whilst it doesn't cover frequency, this site;

http://www.bmreports.com/bwx_reporting.htm

covers system demand (takes a few seconds usually for the graphs to
load) and other commercial parameters, today capped out at just over 38
GW demand.

Cheers,

Andy

 

[Christopher Tidy]

You can download more information than you ever wanted to know about the
National Grid from here:

http://www.nationalgrid.com/uk/Electricity/Codes/gridcode/gridcodedocs/

I've pasted two relevant snippets below (NGET stands for National Grid
Electricity Transmission plc).

1. From the definitions:

Target Frequency
----------------
That Frequency determined by NGET, in its reasonable opinion, as the
desired operating Frequency of the Total System. This will normally
be 50.00Hz plus or minus 0.05Hz, except in exceptional circumstances
as determined by NGET, in its reasonable opinion when this may be
49.90 or 50.10Hz. An example of exceptional circumstances may be
difficulties caused in operating the System during disputes affecting
fuel supplies.

2. From document BC3, which deals with the frequency control process:

BC3.4.3 Electric Time
---------------------
NGET 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.


All very well, but ISTM that the biggest problem with using the mains
alone as a time standard is power cuts, after which you'll always need
some absolute standard such as MSF, GTS, NTP, etc. to reset your clock.

Thanks very much for those useful snippets, Andy. When it refers to a 10
second error in electric time, do you know if it refers to a 10 second
error from the true time at any instant (assuming that the sychronous
clocks were set to the correct time at an instant when the electric time
was correct), or a 10 second cumulative error week on week, month on
month, etc.? It isn't immediately clear to me. You're also right to
point out that you need some kind of absolute time standard if you care
about that sort of accuracy. I don't: all I'm interested in is whether
the grid is accurate enough to make this rather cool clock keep time for
day-to-day purposes, and it seems that the consensus is that it will be
fine.

Best wishes,

Chris

 

[The Natural Philosopher]

Hi all,

I'm thinking of building an electronic clock control circuit which uses
the 50 Hz mains frequency for time keeping. The reason for this is that
the clock dial is rather large, so probably wouldn't run for long on
battery power, and I don't fancy spending £40 buying a programmable
quartz oscillator chip.

Despite doing some research online and in electrical engineering books,
I can't find a figure for the accuracy of the time keeping of the UK
power grid. Perhaps this is because there aren't official bounds set for
the time error - one of my electrical engineering books says it is a
legal requirement that the frequency be kept "as close as possible to 50
Hz" - but even if this is the case it should be possible to determine a
typical error figure.

From my point of view I'd regard an acceptable error as 5 minutes every
6 months. Does anyone know the typical time error seen on the UK grid,
or where I might find this information? Any suggestions would be much
appreciated.

IIRC it is actually 100% spot on in the long term.

However it tends to go plus minus several seconds during the day as peak
loads tend to slow the generators..they then overrun a bit in the off
peak hours to catch up.

BUT as a matte of standard and poossibly even law, they always do get it
right over the long period.

 

[The Natural Philosopher]

For any type of accuracy, depending on the power grid is not a dependable
option! There can be noise, interference, and power dips. On the short
term, the power grid will be accurate, but over the long term, it can be out
by a fair amount.

All wrong.

Clocks essentially filter out all the trash, the long term accuracy is
guaranteed by the generating companies, and the one thing that gets you
is really power outages only.

But that was teh one thing you don't mention.

 

[Tony Williams]

The traditional approach assuming the outages were brief, was to
use a cheap oscillator (an astable mutivibrator) running from a
back up battery when the mains feed was down.

On big tower clocks (which often had difficult access
for adjustment) one standard solution was to make all
power cuts last a multiple of 12 hours exactly, either
manually or with a battery-powered crystal oscillator.

 

[n cook]

On big tower clocks (which often had difficult access
for adjustment) one standard solution was to make all
power cuts last a multiple of 12 hours exactly, either
manually or with a battery-powered crystal oscillator.

Any connection with the tradition/obligation ? that any broken civic clocks
should have their hands set to 12.00 ?

Seriously aside, is there any problem for civic clocks in seriously leaning
towers?
One near me , if picture outside of a file is downloadable here
http://www.nutteing.freeukisp.co.uk/triangle1.jpg
with my assistant holding a plumbob to show the lean
or otherwise part down on
http://www.divdev.fsnet.co.uk/graff.htm

 

[Andy Wade]

Thanks very much for those useful snippets, Andy. When it refers to a 10
second error in electric time, do you know if it refers to a 10 second
error from the true time at any instant (assuming that the sychronous
clocks were set to the correct time at an instant when the electric time
was correct), or a 10 second cumulative error week on week, month on
month, etc.? It isn't immediately clear to me.

I assumed it means the former, i.e. ±10 s 'absolute' error, but
"electric time" is not defined in the extensive definitions/glossary
section, nor do the strings "electric time" or "electric clock" appear
anywhere else in the complete Grid Code document (546 page PDF!). For
your latter interpretation they'd have to specify the accumulation
period, and I can't see any such specification.