TV CAMERAS AND ANALOGUE COLOUR ENCODING
Ultimately the transmission of a full colour TV picture requires three separate and simultaneous streams of information; those for the red, the blue and the green light components of the picture. Although these signals exist at each end of the chain (i.e. at the pick-up and display tubes) they are seldom conveyed along separate channels, even where the link is a short one consisting merely of cable. The reasons for this are many: much of the information in the three channels is identical, completely so on black-and-white scenes; the human eye is insensitive to coloured detail and discerns most fine detail as a black- and-white image; the channels available for transmission of a TV signal (whether in radio, cable or recording-media form) are not wide enough to handle three full-bandwidth signals; any differential treatment in regard to phase or amplitude response of the three would lead to degradation of the reproduced picture; and the system would not be compatible with monochrome signals and equipment.
For these reasons colour TV signals are almost invariably coded at source and decoded only in the circuits immediately preceding the colour picture tube. There are four main analogue encoding systems, NTSC, PAL, SECAM and MAC. The latter was dealt with briefly in Chapter 4; here the NTSC and particularly the PAL system will be described. The concept of these encoding systems is to start with the basic luminance waveform and add to it extra signals to describe the colours in the picture. A single subcarrier is added, along with a refer- ence signal during the back-porch period of the line waveform. In the receiver’s decoder the timing (phase) of this chrominance (chroma for short) signal is compared with that of the reference (burst) signal to indicate the hue of the colour being transmitted; and the amplitude of the chroma signal is compared with that of the burst signal to indicate the saturation of the colour being transmitted. The term hue distinguishes between different colours like yellow, green and red, while saturation describes the brightness of the colour – in practical terms, how much it is diluted by white light, or how different it is to a black-and-white reproduction of the same object.
The signals chosen for transmission of the chroma information are not primary-colour signals at all. It is more convenient to use colour- difference signals, each representing the difference between one primary-colour signal and the luminance signal. To arrive at a difference value a subtraction process is involved. Thus subtracting the luminance (Y) signal from each of the primary-colour signals renders the colour-difference signals G−Y, B−Y and R−Y. It is only necessary to send two of these signals; the third can be derived from them in an add-matrix at the receiver. Because the human eye is most sensitive to colours in the green region of the light spectrum, most of the G signal is conveyed in the Y channel, and of the three colour difference signals (on an average scene, not a snooker table or a cricket field) G−Y is the smallest. For this reason the G−Y signal is chosen to be left out of the transmitted signal, and to be recovered from the other signals at the receiver. In effect, then, the full colour picture is sent in three ‘packets’, i.e. Y, R−Y and B−Y signals.
ENCODING
In an image pick-up system which renders RGB signals directly a Y signal must be derived from them; in accordance with the sensitivi- ties of the human eye to each, 59% of the G signal, 30% of the R signal and 11% of the B signal are derived from the primary-colour signals by simple resistive potential dividers and then added together to form a properly balanced Y signal, see left-hand side of Fig. 6.1. Further matrices add inverted Y (i.e. −Y) to each of R and B to produce separate R−Y and B−Y signals. These colour-difference signals are fed to modulators where each amplitude-modulates a locally generated subcarrier signal at 4.433619 MHz (PAL) or 3.575611 MHz (NTSC), frequencies chosen to minimise the dot- pattern on the display due to the chroma signal. The local subcarrier feeds to the R−Y and B−Y balanced modulators have a phase differ- ence of 90°, a timing delay of one-quarter of one cycle. The effect of
this is that one subcarrier’s instantaneous value is at its zenith while the other is passing through zero – an important point in the subsequent decoding process, and the key to separating the encoded R−Y and B−Y signals.
The modulators used are special suppressed carrier types, in which the carrier itself is cancelled internally, leaving only the sideband products of the amplitude-modulation process. This balanced- modulator technique ensures that no chroma carrier is present to cause dot patterns in picture areas with low or zero colour content, and that only on highly saturated scenes do large subcarrier signals appear. The outputs of the R−Y and B−Y modulators are together added to the Y signal.
Burst insertion
A sample of subcarrier signal must also be added to the Y signal in the form of a colour burst, consisting of ten cycles of subcarrier with a phase corresponding to −(B−Y). Since B−Y phase is regarded as 0°, the burst is thus at 180°, and this sample signal is also derived from the subcarrier generator. It is passed to the Y amplifier via a gate whose opening coincides with the back porch: the burst gating pulse to operate it comes from a sync pulse generator which is also responsible for timing and forming sync and blanking pulses.