Time in Broadcasting: Automatic time switching of BBC External Service programmes 

21 October 2022 tbs.pm/76347

It is not, perhaps generally realised that, in one sense, time is the basis of all broadcasting, for time is the reference against which is measured the frequency of the electromagnetic waves used for radio transmission. Broadcasting frequencies are expressed as so many kilocycles (thousands of cycles) per second, or so many megacycles (millions of cycles) per second.

Time has, of course, many other applications in broadcasting and this little booklet, produced by the British Broadcasting Corporation especially for the Centenary Exhibition of the British Horological Institute, describes some of these. It is hoped that it will be of interest to horologists and others visiting the exhibition.

From ‘Time in Broadcasting’, published in October 1958 as part of the centenary exhibition of the British Horological Institute

In the BBC’s External Services, a number of different programmes have to be sent out at different times and on different transmitters, according to a very complicated schedule. At the headquarters of the BBC’s External Services at Bush House, London, and also at the multiple transmitting stations at Skelton in Cumberland and Rampisham in Dorset, most of the switching operations necessitated by these changes of programmes and transmitters are now carried out automatically.

The programme schedules in the External Services are rigidly divided into quarter-hour periods or multiples thereof. The automatic programme switching can therefore be controlled by uniselector switches which are moved one step every quarter hour by pulses derived from a master clock, in similar fashion to the automatic time-signal equipment described in Part 1.

At the External Services headquarters at Bush House the various programmes must be routed to the lines serving the transmitting stations where the incoming programme lines must also be switched to the correct transmitters at the correct times. This latter switching, too, is carried out automatically. Another automatic mechanism connects the programme from each of the incoming lines, and from each transmitter radio output in turn, for five seconds at a time, to a monitoring loudspeaker for checking purposes.

Programme Source Switching Equipment

The programmes start at any quarter-hour instant G.M.T. (known as “zero”) and close at twenty seconds before a zero. The nominal instant of switching over is eighteen seconds before zero, and five seconds are allowed for the switching operations. There is thus a tolerance of plus two and minus three seconds in synchronization of clock times at the different switching points.

The switching equipment at the External Services headquarters is controlled by the two-second and half-minute pulses from the master clock. Since a programme must be switched eighteen seconds before the beginning of any quarter-hour when a change is due, counting circuits have been devised which make use of both sets of pulses. Starting at the exact beginning of each quarter-hour period, the half-minute pulses are made to move a uniselector switch one step at a time until 14½ minutes have elapsed, at which moment the stepping of the uniselector switch is taken over by the two-second pulses until 14 minutes 42 seconds, when the programmes are actually switched.

The routing of the programme sources to the lines is preset by inserting specially-made “combs” at the correct positions in a group of uniselector switches. The wipers of these switches move one step each quarter of an hour, and each position of the wipers corresponds to a particular time, which is indicated on a time scale attached to the stationary part of the switch. The combs connect short-circuits across certain contacts on the switch, and different arrangements of the short-circuits determine the routing of the programme to the lines in accordance with a code. Thus, having settled the routing for any particular quarter-hour period, it is only necessary to select the appropriate comb and insert it in one of the switches at the correct time-position.

The master clock at Bush House is automatically corrected by the Greenwich Time Signal once per hour, and thereby maintains an accuracy better than ± one-fifth second per day. The mechanism which performs the automatic correction is comparatively simple. A platform is fitted to the pendulum above its centre of gravity, and a small weight is attached by a very light chain to a lever which allows the weight to rest on the platform when the lever is lowered, and suspends the weight clear of the platform when it is raised. When the weight is on the platform, the centre of gravity of the pendulum is raised and the clock runs slightly fast. Conversely, when the weight is lifted by the lever, the centre of gravity is lowered and the clock runs slightly slow. This lever is controlled by a d.c. voltage derived from the time signal which operates one of two solenoids. One solenoid raises the lever, and latches it, thereby lifting the weight from the pendulum and making the clock run slow. The other solenoid releases the latch and allows the weight to rest on the platform, making the clock run fast. A changeover switch diverts the d.c. voltage from one solenoid to the other at the exact moment of the clock pulse marking the zero position of the minute and seconds hands. For thirty seconds before this “zero” pulse, the d.c. voltage is directed to the solenoid which makes the clock run fast, and for thirty seconds after the “zero” pulse, the d.c. voltage is directed to the solenoid which makes the clock run slow.

If all the time signal “pips” occur before the “zero” pulse, the clock is running slow. In this case, all the d.c. voltages from the pips will be directed to the “fast” solenoid, the weight will stay on the platform, and the clock will gain for the next hour. If the clock is fast, one or more of the pips will occur after the “zero” pulse, the corresponding d.c. voltages will be directed to the “slow” solenoid, the weight will be lifted, and the clock will lose for the next hour.

Line-switching Equipment and Sequential Monitoring

As already mentioned, the automatic switching equipment at the Skelton and Rampisham transmitting stations must provide for switching the programmes from each of the incoming lines, and from each transmitter in turn to a monitoring loudspeaker. Each incoming line which is carrying a programme is “sampled” for five seconds, then each of the transmitters taking this programme is sampled in turn for a five-second period. The switching system then samples another incoming line for five seconds, followed by five-second samples of each of the transmitters connected to this line. This sequence is continued until all the lines and transmitters working at any particular time have been sampled. The maximum time for a complete sampling cycle, when six lines and twelve transmitters are working, is thus ninety seconds, and the longest time for which a fault can remain undetected by the operator is slightly less than this. This system is known as sequential monitoring, and is less distracting to the operator than other systems involving more frequent changes of programme, besides allowing the shortest possible time-interval between successive samplings of any one line or transmitter.

The time control system uses one-second and half-minute interval pulses from the master clock. Integrating relays produce pulses every minute by “doubling” the half-minute pulses. These relays start working at “zero” time (i.e. the exact beginning of each quarter-hour period), so that their first output pulse occurs thirty seconds after zero. The one-minute pulses are passed to another group of integrating relays which count fifteen pulses. The fifteenth count will therefore occur at fourteen minutes thirty seconds after zero, or thirty seconds before the next zero. At this moment another group of relays, which counts the one-second pulses, is started working, and produces the pulse for initiating the programme switching at eighteen seconds before zero, as required. During the next three seconds until fifteen seconds before zero, the counting circuits are reset to their starting positions, and the system is then ready for the next zero pulse from the one-second pulses to start the cycle again.

The five-second interval pulses for the sequential monitor are produced by another group of integrating relays which count five one-second intervals.

Fig. 3. Part of the selection grid

The method of presetting the programme switching is different from that used at Bush House. The basic element is the same, namely a group of multi-pole multi-way time switches (sixteen-pole fifty-two-way in this case) of the uniselector type, seen at the top of Fig. 3, which move one step for each quarter-hour period. The use of a coded switching system is also common to both installations, but it is the method of setting the coding which differs.

Fig. 3 shows the left-hand half of the selection panel, and Fig. 4 a rear view of the panel and selector wiring. It will be seen from Fig. 3 that there is a series of horizontal metal bars arranged in groups of five, and a row of holes on each bar. Each group of five bars corresponds to one transmitter, and each vertical row of holes (running across the bars) corresponds to one quarter-hour period. The time scale can be seen at the bottom. Behind each of the holes on any one bar, there are sockets connected to the successive quarter-hour contacts on one level of a uniselector switch (the term “level” means a row of contacts swept by one particular wiper). A metal plug inserted in any one of the holes therefore makes a connection between the bar and one of the contacts on the uniselector. The presetting operation for one transmitter for a given quarter of an hour consists in putting plugs into one or two of the five holes corresponding to that time, in accordance with a code. This determines, by means of a system of relays and another uniselector, which line shall be fed to the transmitter for that quarter hour. The portion of the selection panel shown in Fig. 3 covers twelve hours 1 another similar panel on the right covers the remaining twelve hours. Provision is made for manual overriding of the automatic switching when necessary.

Fig. 4. Rear view of programme routing equipment showing wiring to uniselector switches

The five-second sequential monitoring requires automatic switching to connect the monitoring loudspeaker to each incoming line which is actually carrying a programme. After each connection to a line, the switching must connect the loudspeaker to the received signal from each of the transmitters, in turn, which are taking the programme from that line. This is achieved by an intricate system of relays, and by two motor uniselectors which “search” and stop moving when they find an electrically “marked” contact, as in automatic telephone practice.

Each of the lines, and each of the monitoring receivers, is linked with contacts on the two uniselectors. As soon as one uniselector has searched and found a line which is carrying a programme, the other will find the lowest-numbered transmitter radiating that programme and sample it after the five seconds sampling of the line has ended. In the meantime, the first uniselector will have found the next transmitter (in the numbered sequence) carrying the same programme. This sampling sequence continues until all the working lines and transmitters have been sampled, and the cycle then repeats itself. Although the selector switches are called upon to operate much more frequently than is usual in telephone exchange practice, they have been found able to sustain this extra load.