The BBC Time Signals 

18 September 2017

From the BBC Year-book for 1 November 1929 to 31 October 1930

Long before the existence of any records the human race used the Sun, Moon, and stars to measure time. It is, therefore, quite natural that astronomers should be the time-keepers of the modern world. They have made very careful and extended studies of the three natural divisions of time — the day, the month and the year. They find that the day, i.e. the interval between instants when the Sun is due south on successive days, varies by three-quarters of a minute in the course of the year; that the interval from New Moon to New Moon varies more than twelve hours; and even the year varies slightly. When freed from certain calculable irregularities they have found three nearly uniform units of time :—

  • The period of the Earth’s rotation on its axis.
  • The period of the Moon’s revolution about the Earth.
  • The period of the Earth’s revolution about the Sun.

The friction of the tides causes a lengthening of the day by about 1/1000th of a second in a hundred years, and there may be other slight variations in the length of the day in the course of years. But within this limit the daily rotation of the Earth provides a uniform measure of time.

The astronomer makes the rotating Earth his standard clock. The stars are the dial and are brought successively across his meridian, i.e. the vertical plane stretching north and south through his place of observation. By building a wall truly north and south and looking along it he can tell the moment at which stars cross his meridian, and the same star may be watched night after night. This crude method was improved by the astronomer Römer, who devised the transit instrument, by which a telescope is mounted on an axis perpendicular to its length. The axis ends in accurately turned pivots which rest in bearings placed truly east and west. In the focal plane of the object-glass a fine wire, usually a spider’s thread, is placed in a direction at right angles to the horizontal. When the telescope is turned on its axis to the right elevation, the star is seen as a bright dot which moves across the field of view and in due course crosses the thread. The observer in this way can tell the exact moment at which the star crosses the meridian.

To note this and keep a record of it he needs something terrestrial which will move uniformly. Hour-glasses, the rate at which candles burn, water clocks, might all be used. But nothing of the requisite accuracy was discovered till the pendulum was applied by Huyghens to regulate the motion of a train of wheels. The dead-beat escapement invented by Graham made the astronomical clock a worthy partner of the transit instrument. By the co-operation of these two valued servants the astronomer not only determines the time but also measures the intervals at which different stars follow one another across his meridian. It is convenient to use a clock which keeps “Sidereal Time,” i.e. should show 24 hours between consecutive passages of the same star across the meridian. Let us suppose that on one night his clock shows exactly 4 h. 30 m. 0 s. when Aldebaran crossed the wire of his telescope, and the next night this occurred at 4 h. 30 m. 1 s. Then he would say that the clock was gaining 1 second in 24 hours. If on the first night Sirius was noted as crossing the meridian at 6 h. 40 m. 11.2 s. by his clock, he would make allowance for the amount the clock had gained in 2 h. 10 m., and thus find exactly how much Sirius was behind Aldebaran. By the co-operation of transit-instruments and clocks the accurate positions of a large number of stars have been found, and any of these are available for determining the time when required.

THE TRANSIT INSTRUMENT AT GREENWICH by which Greenwich Mean Time is ultimately determined.

Clocks have gradually become very perfect time-keepers. The Shortt clocks at Greenwich seldom change their rate by more than one hundredth of a second in a day. But this may accumulate and in ten days the clock be one-tenth of a second in error. So it must be constantly controlled by astronomical observations. A good observer in a couple of hours from ten stars can be relied on to tell the error of the clock to one-fiftieth of a second. So the clock is checked every four or five days when weather permits. Its errors are booked regularly but the clock itself is not touched.

Sidereal Time is only used by astronomers and is unsuitable for ordinary domestic purposes. Owing to its revolution round the Sun, the Earth makes one more turn on its axis with reference to the stars in a year than with reference to the Sun. So there are 366¼ sidereal days to 365¼ mean solar days in the year. The Sidereal and Mean Solar Clocks start level on March 21, but the Sidereal Clock gains about 4 minutes a day. After a fortnight the clocks are one hour apart, after a month two hours, and so on. We must keep solar time if we want the clock always to show 12 hours about midday, unless we are prepared at some part of the year to lunch in the middle of the night.

So the astronomer keeps a Mean Solar Clock and makes a little calculation to tell the Mean Solar Time from the Sidereal Time. He makes the necessary allowance for the amount his Sidereal Clock is in error, and so finds the error of his Mean Solar Clock. He has to set this clock right before time signals are sent out. The error is generally very small, but how is he to correct the clock for a small fraction of a second? A simple plan is to put a bar magnet on the pendulum of the clock and fix a solenoid a little below the pendulum. An electric current sent in one direction causes the solenoid to attract the magnet and so slightly increases gravity and quickens the swing of the pendulum, and the reverse effect is produced when the current is reversed. By keeping the current on for a short time, perhaps half a minute, the clock is quickened up or slowed down the necessary fraction of a second.

When the clock has been corrected its signals are all sent out automatically. These consist of hourly telegraphic signals to the G.P.O., which are transmitted to all parts of the country. Six signals are sent every quarter of an hour to the B.B.C. by a special wheel on the clock, and distributed as required. These give the seconds 55, 56, 57, 58, 59 and the 0 of the exact quarter. Finally, at 10 h. and 18 h. G.M.T., a series of signals are transmitted from Rugby which can be received in most parts of the world under good atmospheric conditions. It is seldom that the signals are one-tenth of a second in error, and our aim at Greenwich, though not always achieved, is to keep the error less than one-twentieth of a second.

BBC Time Signal Chart

Week-day service

Station. 10.15 a.m. 10.30 a.m. 12.0 noon. 1.0 p.m. 4.45 p.m. 6.30 p.m. 9.0 p.m. 10.15 p.m. 11.30 p.m.
National 1554.4 m. Big Ben G.T.S. † Big Ben ❊ G.T.S. G.T.S. G.T.S. G.T.S. G.T.S.
London National 261.3 m. Big Ben ❊ G.T.S. G.T.S. G.T.S.
London Regional 356 m. Big Ben G.T.S. Big Ben ❊ G.T.S. G.T.S.
Midland Regional 479.2 m. Big Ben ❊ G.T.S. G.T.S.
Provinces G.T.S. G.T.S.


Sunday service – Time Signals

Station. 10.30 a.m. 3.0 p.m. 3.30 p.m. 9.0 p.m.
National 1554.4 m. G.T.S. G.T.S. G.T.S.
London National 261.3 m. G.T.S. G.T.S.
London Regional 356 m. and Midland Regional 479.2 m. G.T.S. G.T.S.
Provinces G.T.S.


Big Ben

If circumstances are favourable, Big Ben will be broadcast at the beginning of any programme emanating from London. The day’s programme on week-days will also be concluded, when possible, with Big Ben.


G.T.S on 1554.4m is compulsory and will always be broadcast even if this means its super-imposition.

❊ Saturdays excluded.
† Greenwich Mean Time.

Sir Frank Watson Dyson, KBE, FRS, FRSE (1868-1939) was Astronomer Royal from 1910 until 1933 and introduced the “pips” to the BBC in his capacity as Director of the Royal Greenwich Observatory in 1923. The pips have been generated by the BBC itself since 1990.

You Say

5 responses to this article

David Heathcote 18 September 2017 at 11:39 am

I wonder what the point was of the 10.15am chimes. Only the first quarter would have rung, and I don’t think there was any real precision to that, as there was to the first “bong” on the hour. It’s almost like a jingle (!) Was the Daily Service at 10.15am, in which case that junction would have been deemed Important?

Brian Winter 18 September 2017 at 11:47 am

I know this article was written some time ago, but it might just be worth updating to say that the atomic clock that was at Rugby is now located in Cumbria.

Horace Linden 18 September 2017 at 11:32 pm

In fact, the NPL installed not just one but three atomic clocks for the launch in 2007 of the relocated service at the Anthorn, Cumbria transmitter site.

The site is operated by Babcock International and transmits three services — GQD on 19.6 kHz VLF for NATO, MSF on 60 kHz as the national time station, and an e-Loran service (a time and geolocation service primarily for mariners) on 90 – 110 kHz LF.

Paul Mason 22 September 2017 at 1:30 am

In the 1970s the current longer pip at 00 was introduced which was a good idea. At the end of some years a leap second has been added to account for variances due to the Earths speed. So the pips go 55, 56, 57, 58,59, 60, 61 (long pip Happy New Year!).

Paul Mason 22 September 2017 at 1:36 am

One snag with the pips is DAB radio, which is a second behind LW/AM/FM which is a slight nuisance but one can’t have everything!

Your comment

Enter it below