Test Card B no longer exists. A photograph exists of a studio, with a partial view of Test Card B on a stand. BBC engineer George Hersee described it in his 1967 work, but it is believed that no copies of the full card remain in existence.

It was similar to Test Card A but with an extra greyscale strip below the circle. The letterbox just above 'A' moved to the top of the card. It was never actually broadcast, but was used for camera line-up at Alexandra Palace.

Test Card C was first broadcast in January 1948. At the time, the aspect ratio of TV broadcasts was 5:4. By the early-1950s, this had changed to 4:3. The design of Test Card C was altered slightly as a consequence.

Some BBC regional transmiiter sites used slightly different versions of Test Card C. There were at least three different designs in use at the BBC. Additionally, the ITA used a slightly different design for the ITV Network.

In 1964, two new test cards were introduced. Test Card D was adopted by the BBC and ITV for use on BBC One and throughout the ITV Network. Test Card E was used for the new 625-line BBC Two service. Test Card D and Test Card E were virtually identical, bar the frequency gratings, which were finer on the latter, to cater for the higher definition offered by the 625 service.

The frequency gratings on Test Card E were sinusoidal, and thus technically more useful than the square waves on previous test cards. However, as a result, some of the gratings looked soft and fuzzy - the lowest grating in particular. TV dealers complained about this and Test Card E was consequently scrapped, after c. one week on air. It was replaced by a modified version of Test Card C. Unfortunately, we have no examples of the special BBC Two version of Test Card C.

Test Card F, designed by BBC engineer, George Hersee, was introduced in 1967, and aired on the new colour 625-line BBC Two service. The test card features Mr Hersee's daughter, Carole, pictured alongside Bubbles the clown.

Since the 1950s, BBC trade test transmissions were accompanied by music compiled from library music publishers. Over the years, virtually every music genre imaginable was featured. The vast majority of this music was not commercially-available in the UK. However, in 1953 and 1954 some commercially-available recordings were used with Test Card C; these were all on the Oriole label. In 1954, some pieces from the Xavier Cugat Orchestra and the David Carroll Orchestra included vocals. Tracks featuring vocals were a rarity during trade tests. For a period in the 1970s and early-1980s, some of the tapes included pop music. Any pop music played with the test card would have been an 'out-take' recording, produced outside of the UK. The BBC could transmit this material as 'library' music, as the recording would not have been commercially-released in the UK.

The test card was also occasionally accompanied by a test tone. Until the 1970s, the tone would be inserted for short periods at regular intervals during the day. By the 1980s, music was played with the test card for much of the day, with tone being restricted to short periods each morning, when the test card first appeared. There were different tones for BBC One and BBC Two: BBC One used 1KHz, and BBC Two 440Hz. The use of different tones on each channel was mainly just so that engineers (and viewers of a certain disposition) could uniquely identify each channel.

Originally, the tones originated at Broadcasting House (the home of radio). We're not sure exactly how they were generated there. In later years (around the start of the 1980s), some units were built that took in 5MHz from Rubidium oscillators, and synthesised the above audio frequencies precisely. The kit is still there apparently, and still works (as of early-2009). However, since the introduction of the 'VALID' test signal (colour bars with a rotating central pattern, and synchronised blip on the sound, made by Vistek (now ProBel)), the BBC uses 997Hz for the main programme stereo, and 440Hz for the second stereo - often used for Clean Effects. The problem with precise 1KHz is that only some of all the possible digital levels are used. 997, however, is such an odd number, that it cycles through all the levels, thus fully exercising any DACs. But it is close enough to 1KHz so as not to be audibly different.

Test Card F was broadcast from a 35mm slide, using a Rank Cintel slide scanner. It was actually a dual-layer slide - one layer for the monochrome areas, the other for the colour parts - very carefully aligned and sealed in a glass slide-holder. There was no BBC One-branded slide version of Test Card F - the 'BBC2 COLOUR' test card slide was used on BBC One and BBC Two. However, when Test Card F was aired on BBC One, a 'BBC 1' caption was electronically-superimposed, covering up the 'BBC 2 COLOUR' legend.

The Rank Cintel slide scanner was a large and expensive piece of equipment, but probably of more importance, technically, is that it would be near impossible to keep two scanners aligned identically to match each other. Like all standards, their absolute accuracy is not as important as the fact that everybody agrees. Hence, it was far better to use just one scanner, and overlay a 'BBC1' caption on the feed that was used for the BBC One transmissions, as this resulted in a consistent test signal for both networks. Test Card F was generally broadcast for longer periods during daytime hours on BBC Two than on BBC One; the latter's schedule often being filled by programmes for schools and children. The BBC One caption overlay was a little crude by modern standards. It was produced by a dedicated unit that had the relevant parts from an 'Anchor' machine (an early form of electronic caption generator, before the introduction of Astons and Ryleys).

In 1980, the BBC began an experimental 'Ceefax In-Vision' service. Gaps in the schedule which previously would have been occupied by trade test transmission, were now being filled by a selection of teletext pages. The 'In-Vision' broadcasts - offering a digest of news, sport, travel and TV listings - were largely restricted to short, early-morning slots, with the test card playing the more prominent role during gaps in the schedule.

The first Ceefax 'In-Vision' broadcast took place on Monday March 17 1980 on BBC One. It was Monday June 02 1980 before the teletext 'In-Vision' service appeared on BBC Two.

Ceefax pages were typically broadcast from 8.30am to 9am (weekdays) (though by 1982 the 9am finish time had been extended until just before the first programme of the day) on BBC One and 10am to 10.30am (weekdays) on BBC Two. During the summer months of 1982, the weekday BBC Two 'In-Vision' service moved to 9.35am; 'Service Information' had moved to 10.15am (rather than 10.30am), and 'Play School' was now shown at 10.30am (it's long-standing start time was 11.00am).

From 1980 to 1983, an 'In-Vision' service was also broadcast from 3.20pm to 4pm (weekdays) on BBC Two; this was immediately followed by the appearance of Test Card G, for the afternoon 'programme transfer' (see below). At c. 4.45pm, Test Card F would appear, and remain until the next programme, usually an Open University production, at 5.10pm.

On Monday February 28 1983 at 6am, 'Ceefax AM' was born, replacing Test Card F, in the lead-up to the 'Breakfast Time' programme at 6.30am. An indication that 'Ceefax In-Vision' might soon play a more prominent role in the schedules.

With TV sets becoming more reliable, there was less of a need for the broadcast of a test card. The trade could also now buy their own test signal generators. BBC editorial teams were also keen to make better use of the airtime.

In early-May 1983, daytime trade test transmissions were replaced by 'Pages from Ceefax'. The test card was now largely restricted to relatively brief early-morning slots, shortly after transmitters came back on air following overnight closedowns. Weekdays, Test Card F would be broadcast for seven-and-a-half minutes, between 5.52am and 6am on BBC One; it would also be broadcast for seven-and-a-half minutes on Saturday and Sunday mornings, before the first programme of the day. The test card would also have a seven-and-a-half minute outing on BBC Two each day: typically between 8.52am and 9am on weekdays; 'Pages from Ceefax' would take over at 9am.

The scheduling of Test Card F varied over the years:

• Contrary to popular opinion, Test Card F was not broadcast overnight, after BBC One and BBC Two closed down. Until 1997, the BBC One and BBC Two transmitter network was shut down each night, around twenty minutes after the last programme. When the announcer had bid us farewell, the screen would fade to black; we'd typically have thirty seconds of black and silence, which would be followed by ten minutes of tone on BBC One and two minutes of tone on BBC Two. Why the difference in duration between the two channels? Well, late-night regional news and weather forecasts were common in all BBC One regions in the 1970s, with the exception of London (Network). Apparently, regional centres in England could not opt back to Network if there was no valid Network feed to opt back into. So, Network had to stay online until all the regions had opted back in. This was not the case for the regional centres in Belfast, Cardiff and Glasgow, which regularly stayed up later than Network.

Local continuity in the English regions was discontinued in the early-1980s, due to BBC cutbacks. However, the ten-minute tone at closedown on BBC One continued until November 1997. This was presumably out of courtesy to the national regions in Belfast, Scotland and Wales, who may have been closing down minutes after Network; by staying online, Network will have allowed the national Ceefax service to continue in the regions (if Network went offline for the night, the Ceefax feed to the regions would have been lost; Ceefax was still a national service until 1997). Regional continuity was much less common on BBC Two, and so there was no need for a lengthy period of tone - just enough to wake up anyone that may have fallen asleep.

• From c. the mid-1980s, Test Card F would appear from 5.45am (weekdays) on BBC One. It would also appear around fifteen minutes before the first programme of the morning on BBC Two (usually an Open University production).

• Between January 1992 and 1994 (precise date to be confirmed), BBC One and BBC Two would broadcast an overnight encrypted service for specialist audiences. These programmes were usually broadcast between 2am and 6am, under the banner 'BBC Select'. Such broadcasts typically took place at least twice-a-week on each channel. Where a BBC Select programme was being broadcast, the channel in question would remain on air throughout the night. The BBC Select broadcast would be preceded and followed by Test Card F and tone. On BBC One, 'Pages from Ceefax' would appear approximately fifteen minutes before the first programme of the day. The same would apply to BBC Two, unless the first programme was an Open University broadcast, in which case, no Ceefax pages would be broadcast. However, from c. 1994, Ceefax pages also appeared prior to the early-morning OU programme, which was usually the first programme of the day.

• On Monday January 04 1993, BBC One's 'Business Breakfast' moved to a 6am start; consequently, 'Ceefax AM' moved to 5.45am. Test Card F was now broadcast for fifteen minutes from 5.30am each weekday.

• In autumn 1995 (precise date to be confirmed), the launch of the overnight 'Learning Zone' meant no overnight closedowns on BBC Two during the week, so Test Card F was relegated to weekends only.

• Following the launch of BBC News 24 in November 1997, Test Card F transmissions on BBC One came to an end, Rather than closing down, BBC One now broadcast BBC News 24 overnight. Weekend Test Card F remained unaffected on BBC Two for the time being, appearing around half-an-hour prior to the first programme of the day, and giving way to 'Pages from Ceefax' after fifteen minutes. In early-1998 (more precise date TBC), the practice of weekend overnight transmitter shutdowns on BBC Two was discontinued. For a number of months, Test Card F and tone filled BBC Two's downtime; however, by December, 'Pages from Ceefax' was being broadcast through the night, accompanied by tone.

• Since 1998, the test card has put in the odd appearance on BBC One and BBC Two (in various guises - Test Card F, Test Card J and Test Card W), mainly as part of overnight RBS tests on the analogue transmitter network. An RBS test is carried out once-a-year, on BBC One and BBC Two - usually in January.

In May 1984, BBC engineers created an electronic version of Test Card F. No major design alterations were made for the electronic version. There were a couple of cosmetic changes: a new font for the designation letter and the colour bars along the top, which now completely replaced the black and cyan blocks.

Internally, most PAL coders inserted a couple of lines of colour bars within the vertical interval. Known as IRS bars (Insertion Reference Signal), this allowed for quality checks at any time, and not just during line-up, similar to the ITS (Insertion Test Signal) on transmitted signals. The coder for the slide version of the test card was modified to insert a fat band of colour bars at the top of the picture, but had the disadvantage of hiding the top of the arrow-head that pointed to the edge of the picture. When the electronic test card was created, the colour bars were part of the frame store that made up the image, and so they were extended down to the top white bar of the frame to look tidier. The arrow-head was overlaid on top of the central magenta bar, thus restoring that feature.

During its first few months in use, the new electronic test card was broadcast without a channel identification caption. When they did finally appear, these captions were quite basic. However, it wasn't long before they were replaced by the proper channel logos.

The electronic Test Card F was a revelation, as it was perfectly linear, and had a flat grey background, unlike the old slides, which always had a bit of shading. Digitising an image was quite difficult in those days, and rather than try to scan the original photo again, the BBC's Research Department (who created the electronic version) worked with the Computer Graphics Workshop at TV Centre, and grabbed an image of the current test card slide on Quantel Paintbox. They took the data file back to their base at Kingswood, and incorporated the central picture into the computer-generated pattern.

However, so that there was a clean and precisely-defined white circle around the picture, they had to zoom in a little to hide the old one on the slide. Despite their best efforts, the picture on the electronic test card always looked a little desaturated (pale).

According to Barney Wol's website, when the widescreen Test Card W was being put together (and the revamped 4:3 card that was named 'J'), the original 2.25" slide of Carole Hersee was re-scanned. The website has more details about the BBC's electronic Test Card F and test signals in general.

The kit that produced the electronic test card also had the ability to add an on-screen countdown. Since the 1970s, a ten-minute countdown to the next programme was regularly displayed over the test card. However, this feature was rarely used in the days of the electronically-generated Test Card F, as 'Pages from Ceefax' was now filling the gaps in the daytime schedule; the test card rarely preceded programmes now. Occasionally, the 'in-vision' Ceefax generator would misbehave (particularly during the period from 1983 to 1986), and Test Card F or 'the colour test card' (i.e., Test Card G), would be brought in to cover.

Until 1986, many of the BBC's Ceefax in-vision broadcasts ran off hardware that was built in-house, using the standard Mullard chip set. In 1985, BBC Micros were employed to produce some of the Ceefax in-vision slots on both channels; the old generators were phased out completely during 1986. Specially-written software running on a BBC Micro allowed engineers to alter the header (row 0) data: from 1985 to 1987, headers reading 'BBC1 In Vision' and 'DAYTIME on TWO' were a common sight; the date would be displayed in cyan; the clock retained its standard yellow colouring. Incidentally, Presentation staff produced the 'Daytime on Two' in-vision pages that filled the gaps between schools programmes on BBC Two from September 1983 until December 1987.

From 1988, the header row reverted back to the standard Ceefax row 0 display. Later that year, the BBC began experimental Level 2 in-vision broadcasts (first TX: August 18 1988). However, the BBC Micro-generated in-vision facility continued in use until 1994, when the Level 2 generator took over completely. In 1994, the BBC made more use of Level 2 teletext features for its 'In-Vision' service, such as the increased range of colours and higher resolution graphics. However, in 1996, 'Pages from Ceefax' reverted back to the standard Level 1 format.

The first broadcast of Test Card W on BBC One and BBC Two took place on Sunday August 06 2000 [TBC]. Appearances of test cards on BBC One and BBC Two are few and far between these days, and are generally restricted to the annual RBS tests, which take place in January each year. However, these tests only affect the analogue transmitter network; they also make use of Test Card J, rather than Test Card W. BBC Two does still have periods of downtime. However, these 'gaps' in the schedule are normally filled by 'Pages from Ceefax' rather than the test card.

Test Card J - also developed during 1999 - is a modified version of Test Card F. Test Card J had some extra features: pairs of flashing dots within the greyscale frame, and a green chrom/lum delay square in the top streak-box.

Test Card J and Test Card W also include a new version of the picture featuring Carole Hersee, produced from the original 6cm square transparency. The slide was obtained from Carole's father, George, in 1998. A high resolution image was produced from the slide - just over 5,000 pixels square, with a colour depth of 48 bits (16 each from red, green and blue).

Test Card J and Test Card W were designed by senior BBC engineer, Peter Vince. The conceptual designs were converted to high-quality graphic images using software developed by Richard T. Russell of the BBC Research department.

More information about the development of Test Card W and Test Card J can be found on Barney Wol's website,

In 1971, the BBC introduced a modified version of the Philips PM5544 Test Card. This was the first electronically-generated test card to be used by the BBC. It appeared on BBC One and BBC Two from time-to-time, often appearing in slots where Test Card F would normally have been used. Test Card G was used mainly when the slide scanner used to broadcast Test Card F was undergoing maintenance, or when the slide scanner failed. It is believed that Test Card G was phased out on Network BBC One and BBC Two following the introduction of the electronically-generated version of Test Card F in 1984.

Test Card G was regularly referred to by BBC announcers as 'the colour test card'.

During the 1970s and early-1980s, BBC Two radiated a test card (mostly Test Card F) for much of the day. For a portion of this time, the Network feed of BBC Two was used to broadcast programmes and other material to the regional centres.

In these days, the signals for BBC One and BBC Two were fed to the main London transmitter at Crystal Palace, and also on 'Network Distribution' circuits to the major regional broadcast centres (Cardiff, Bristol, Birmingham, Manchester, Glasgow, Belfast etc.). Most of the time, the regional centres would route the London signals to their local transmitters, but occasionally the region would 'opt-out', and feed their own material to their transmitters. Local news bulletins are the classic example.

In the 1970s and 1980s, during BBC Two downtime, the circuit would be used to transfer various material, such as programmes, to the regions. A locally-generated test signal (usually electronic Test Card G) would be transmitted from the regional centres for the duration of the transfer.

The gentle brightness change from black to white on the 'Sawtooth' signal, if viewed on an oscilloscope is a linear ramp, with a gentle slope up, and a sudden drop to black at the end. This was (perhaps erroneously) perceived to be similar to the teeth on a saw blade. The 'Sawtooth' was a common sight on BBC One and BBC Two in the 1970s and early-1980s at closedown and when the transmitters were switched on in the morning.

The 'Composite Pulse and Bar' was used for aligning video circuits with an oscilloscope, and not by eye on a monitor. From the 1970s to 1990s, it was often the first thing to appear each morning when the BBC One and BBC Two transmitters came back online, following the overnight closedown. A time switch was used to trigger the 'Pulse and Bar' each morning - typically half-an-hour before the appearance of Test Card F - allowing sufficient time for the transmitters to warm up.

In the 1970s and 1980s, a black and white version of the 'Pulse and Bar' was often used; however, a colour version was also in use. Chrominance was added to the standard monochrome test signal to try and accurately and easily show chrominance amplitude, and the
composite pulse showed chrominance/luminance delay. It is believed that the original 'Pulse and Bar' generators didn't even use station subcarrier, but just an approximate 4.43MHz oscillator.

Until the late-1980s, during monochrome programmes, the colour burst was switched off. A colour television would detect that, and disable its colour decoder circuit, so eliminating any colour noise, which would otherwise be very apparent over a black and white picture. However, in later years, Tektronix produced the TSG271 test signal generator which always had a burst on its output, and the phase of the subcarrier on the 'Pulse and Bar' signal was as specified for that in the ITS signals (120B0 - nearly, but not exactly, magenta).

The original 'Pulse and Bar' design included an additional feature not shown here. In the white area on the right-hand-side, and not far from the centre of the screen, there was a black vertical line, running the full length of the screen. However, this line was later split into four 'dashes' of equal length. These dashes were used as a means of identifying the BBC regional centre where the signal originated. The presence of four vertical dashes (the maximum) meant the signal was being broadcast by BBC Northern Ireland in Belfast. No dashes (as in the examples above) indicated that the 'Pulse and Bar' was coming from London.

The 'Streak Test' wasn't transmitted regularly by the BBC. But, we happen to have one off-air recording. If the low frequency performance of a circuit is poor, then a constant brightness level, which is basically a DC voltage, will end up with a slope on it. By having different width white bars, different LF time-constants can readily be seen. This was useful for aligning flying-spot slide scanners and telecine machines.

From c. 1984, a new test signal appeared. The Open University test signal comprised a black screen, with white text reading 'OU' in the top left and top right corner of the screen. It was accompanied by tone. The signal was transmitted around Open University programmes, from a special transmission suite, used specifically for Open University broadcasts. The announcer had complete control over all sound and vision and was also responsible for playing out the Open University programmes from this suite.

The OU test signal would normally appear for a short period prior to the first OU programme each morning, and in the gap following the end of the last early-morning OU programme, and the appearance of Test Card F shortly before 9am, when the main BBC Two Network Control Area came fully online. The signal was merely a means of keeping the transmitters on air.

An electronic switch was set to put out 'Pulse And Bar' half-an-hour or so before the first Open University programme each morning, allowing time for the transmitters to warm up. The automated system would then switch to the OU suite, which would be sending out the OU test signal. In a situation where the announcer had overslept, resulting in the first OU programme not being broadcast or delayed, the only thing being broadcast would be this fairly innocuous test signal.

The OU transmission suite was used mainly for early-morning and late-night Open University broadcasts. Eagled-eyed viewers will have noticed a slight frame roll during any transition between BBC Two Network Control and the OU suite. If you ever wondered why this occurred - well, here's the explanation.

In those days, there were two complete transmission 'chains' - all the way from the network mixer output, via Ceefax and ITS inserters, and SIS encoders, with an auto-changeover between the two on detection of a SIS decoder fault (which would also cover failure of the Ceefax or ITS inserters, and the whole circuit from the mixer output). But, this auto-changeover was done with real relays! Thus the U-links on the inactive chain would be swung across from the main Network Control Area output, to the OU suite, and the changeover forced manually. While everything was synchronous, it was the rather large hole in the signal resulting from the relays changing over that resulted in the frame roll.

On two occasions in 1997 - June 21 and October 05 - BBC One and BBC Two transmitters remained on air throughout the night to test the contingency transmission facilities at BBC Pebble Mill in Birmingham. In the event of a major problem at BBC Television Centre in London, which prevented BBC One and BBC Two being transmitted from there, control of both networks would switch to Birmingham. The test cards shown here were broadcast from BBC Pebble Mill as part of a test of this contingency arrangement.

Related websites:
Alan Pemberton's Test Card History: UK and international test cards, including comprehensive technical detail.
Barney Wol's Site: looking at BBC test cards and various test signals.
The Test Card Circle: UK test cards and the music that accompanied them.
Thomas Bertsam's Test Card Gallery: a collection of test cards/signals from around the world.