DE3013195C2 - Method and apparatus for gain control of a color channel in a television system - Google Patents

Description

The invention relates to a method according to the preamble of claim 1 and a device for Implementation of this procedure.

Recent developments according to the French SECAM standard have shown that some tolerances of the SECAM system must be kept closer in the future. This not only results in problems with Design of coding circuits but it also becomes more difficult to manufacture SECAM coders to match. Test facilities, such as vector display devices, which enable simple and precise comparison of NTSC or PAL decoders do not exist for SECAM. There is therefore a strong need for the Introduction of self-aligning circuits in SE-CAM encoders.

From DE-OS 24 46 539 a circuit for automatic gamma adjustment is known in which the gamma levels in the color signal channels of a color camera with the help of a test signal which is fed into two of these staining channels. This test signal is generated on one output signals corresponding to the gamma levels in these channels, and these Are compared with each other. If they differ, the gamma level of one channel will be like this regulated so that both channels match * Suitable for adjusting a SECAM encoder However, this known method is not particularly good, as will be explained in detail later, because an exact gain setting in a color channel for a deviation direction of the SECAM color carrier can lead to the fact that the gain for other deviations is no longer correct and this then fall out of the tolerance range.

The object of the invention is to provide a method - and a corresponding Vorrichiung - for gain control of a color channel for a SECAM color television system, in which the gain in the color channels is controlled so that any frequency deviations / calibrations remain within the prescribed tolerance limits.
This object is achieved by the characterizing part of claim 1 and claim 6, respectively. Some exemplary embodiments of the invention are explained in detail below, showing the specific problems involved in regulating the color channel gain in the SECAM system.

In the drawings shows

FIig. 1 is a block diagram of a SECAM encoder, F i g, 2 and 3 time waveforms,

F i g. 4 shows a block diagram according to an embodiment of the invention;

F i g. 5 some signal curves,

Fig. 6 an alternative embodiment,

Fi g. 7 a more detailed block diagram of the alternative

tive embodiment,

F j g. 8 to 11 some signal curves and

F i g. 12 is a detailed circuit diagram.

In the SECAM system, the color carrier is frequency-modulated by the color difference signals appearing in a line sequence. The center frequency of the color carrier is 4.25 MHz for the (BY) -signa \ Foij = 4.25 MHz and for the / Ä-y> signal F _O ß = 4.40625 MHz , The maximum deviation of the color subcarrier is represented for the (BY) and the fÄ-y ^ signal by Db and Dr and should be adhered to with an accuracy of 10 kHz. While the horizontal sync pulses occur , the modulator is resynchronized to Fob or For.

F i g. 1 shows a circuit of a SECAM encoder. The first part 100 processes the baseband signal and forms from three primary color signals in a suitable manner pre-emphasis and clamped line sequence signals with line identification signals. These signals are fed to a frequency control loop 102 which supplies a baseband SECAM FM color signal from the voltage-controlled oscillator 4S in accordance with the line sequence signals. The FM-Signal-1 is then fed to a high-frequency signal processing section 104 , which carries out a line-alternating phase reversal, stronger pre-emphasis and filtering.

1: The signals R, G and B coming from a suitable source, e.g.

In a preferred embodiment of the invention described here, these potentiometers are replaced by a device which brings about an automatic adjustment. One output supplies the luminance signal Y, which is fed to a delay line 14 which compensates for the inherent circuit delays to which the color signals fed to the input 17 or the adder circuit 16 are subjected. From here the luminance signal is fed to the adding circuit 16, which also receives a mixed synchronous signal at an input 17a. The matrix 12 also generates color difference signals R-Y and BY, which alternately reach a circuit 20 through the line sequence switch 18, which is supplied with a half-line frequency switching signal ( _H I2 ), which receives blanking and clamping pulses and effects pre-emphasis, clamping and blanking . In the circuit 20 receive the color difference signals, a low Preempha ^ is according to the SE-CAM standard. Then they are clamped during the blanking interval to a voltage of Ub / 2. From the circuit 20 the color difference signals are applied to a switch 22. during the vertical identification period - this is the time period during the field blanking time when signals are sent which indicate which color signal is currently being transmitted - the switch 22 controlled by a bottle insertion signal fed to its control input 23 switches the sawtooth output signal of the bottle generator 24 (which is formally known as Designated SECAM line identification signal generator is) to one input of each A ⁺ = + 21OkHz 4406.25 + 28OkHz the switches 26 and 28, at other times is the switches 26 and 28, one of the color difference signals supplied to the switches 26 and 28 are similar to the operation of the switch 18 through a / w / 2 signal controlled

During the occurrence of a (BY) -ZeWe the resistor 30 is connected via a switch 26 to a voltage source 27 of the voltage Ub / 2 , while the resistor 32 is connected via the switch 28 to the output of the switch 22 and the (BY) - The signal is fed to the non-inverting input of the amplifier 34. During the occurrence of the (RY) -ZeWe, the output of the switch 22 is connected via the switch 26 to the resistor 30 and the inverting input of the amplifier 34. At the same time, the voltage source 27 is connected via the switch 28 to the resistor 32 and the non-inverting input. As a result of this alternating supply of the color signals to the amplifier 34, the direction of deviations in the color difference signals R-Y and B-Y alternates. The resistor 35 causes negative feedback for regulating the gain of the amplifier 34. At the non-inverting input of the amplifier 34, a center frequency control signal is also added via a resistor 42 , which is fed with the aid of a phase detector 38 to which the frequencies For and Fob are fed during alternating lines , and a sample and hold circuit 40 is generated. The sampling pulses for the sample and hold circuit 40 contain n horizontal sync pulses. Characterized the oscillator 48 during the horizontal synchronization period to the frequencies For or Fob dubbed After running through an amplitude limiting stage 44, the deviations soft limited and a low pass filter 46, whose cutoff frequency is about 1.3 MHz, the output signal modulates the amplifier 34, the frequency of the voltage-controlled Oscillator 48, if this is not nachsynchroniüert. The output signal of the oscillator 48 is fed to a color subcarrier phase switch 50 whose input 51 is fed a switching signal to effect a 180 ° phase reversal for each frame and for one of every three lines in accordance with the SECAM standard.

The output signal of the switch 50 passes through a bell filter 52 to increase the color carrier amplitude during high color carrier deviations, which also corresponds to the SECAM standard. The output signal the filter 52 is fed to a bandpass filter 34 with a pass band of 3 to 53 MHz, from where it goes to a blanking circuit 56 which receives blanking pulses before the FM color signal y signal and the composite synchro signal with the help of the Addition 16 is added.

For both color difference signals, the color carrier is modulated so that a linear relationship between the There is a change in the color carrier frequency and the value of the color difference signals. The deviations for Color bars according to the European Broadcasting Union are listed below and in FIG. 2 together with the maximum deviation limits VQn 4.756 and 3.900 MHz are given:

= 4686.25 kHz R- = -28OkHz 4406.25 - 28OkHz = 4126.25kHz B ⁺ = + 23OkHz 4250 + 23OkHz = 448OkHz

B "= -Ii-OkHz 4250-23OkHz

■■ 402OkHz tolerancea: <± SkHz

Current SECAM encoders require two adjustments for the (RY) and the (B-Yy amplifiers (in Fig. 1 using potentiometers 13 and 13a). Adjusting the color gain during production is not easy for two reasons:

a) the tolerances are <± 5 kHz and

b) because of the non-linearities of the modulator
(Oscillator 48).

10

"A compromise must therefore be made, as can be seen from the example according to FIG. 3a, which shows an ideal (R- K ^ -line modulator transfer function and a real non-linear modulator transfer function

The best compromise comes in the relationship

AR - = -AR * or -AR - = AR ⁺

The same applies to the BY line. It has been shown in practice that the non-linearity of voltage-controlled oscillators, as they are normally used, is less than ± 1%, as exaggerated in FIG. 3b. Assuming an error of ± 1% for the worst case, this results in a deviation error of ± 2.8 kHz (for deviations of ± 280 kHz). It therefore makes no sense, for example, to set the color gain very precisely for the deviation R * , because the R - signal would then be outside the tolerance.It should be noted that the slope of the voltage / frequency characteristic of the oscillator also changes as a result of temperature fluctuations or aging phenomena changes An automatic color amplification circuit must not only compensate for changes in the amplification of the oscillator, but also automatically adjust the amplification to the best compromise between the deviations R ⁺ and R - or B * and B - .

FIG. 4 shows a preferred embodiment of the circuit according to the invention on the basis of a block diagram, in which corresponding elements have the same reference numbers as in FIG. 1 have. The box labeled 12, 18, 20, 22, 26 and 28 contains the elements from FIG. 1 labeled with these reference numbers. 1. For the sake of simplicity, those parts from FIG. 1, which are not changed, not shown here. 4 allows the gain control for the (R- Y? Line. A control circuit for the (BY) -ZeWe is identical and not shown here. Just after the start of the vertical blanking interval, only an R + is entered in the matrix 12 with the aid of the input selector switch 60 -Signal and an R -signal inserted, as shown in Fig. 5 and originate from the color bar generator 58, which generates the standard-compliant known EBU color bars. A frequency measuring circuit 62, which is shown in more detail in Fig. 7, measures the frequency of the deviating Color carrier from detector 38 for both signals R ⁺ and R -. Alternating reference signals For · and Fob are fed to phase detector 38 through switch 72 under control of half-hour pulses (FIG. 5e). With the aid of a coincidence phase detector 64 and an integrating circuit 66, a DC voltage is generated to control the (R-Y) color gain, so that

R * + R

is equal to 280 kHz. To ensure that the gain loop also works properly when the input of the matrix 12 is connected to an external source (e.g. the pick-up tubes of a camera), the input selector switch 60 is periodically connected to the color bar generator output ^{via an OR-1} TOr 68 only during R + and only during R - switched on. The generator 58, the gate 68 and the measuring circuit 82 are controlled by an initiation signal "CE" (FIG. 5h) for the rule line insertion to ensure that the insertion occurs at the beginning of the vertical blanking interval (see FIG. 5g). Furthermore, a switch 70 provides a manually adjustable control of the insertion of a signal from a camera or a color bar generator 58. The only R ⁺ and only R ~ signals contain R, G, B combinations in the following manner:

/? + (= Red) R = high

KJ - IilCUIIg

B = low

Ä- (= cyan) R = low
G = high
B = high

The color bar generator 58 and the input selector switch 50 are in the RCA camera model TK-47 already found.

The gain control element according to FIG. 4 has an FET 74 which is connected as a variable parallel resistor. The resistors 30 and 32 according to FIG. 1 are divided into two parts 30a / 30Z> or 32a / 32b and, together with the field effect transistor 74, form a variable attenuator that reacts to the gain control voltage from the integrator 66 that is fed to the field effect transistor 74.

This circuit works well at relatively low signal amplitudes (<30OmVs ₅ ). For higher signal amplitudes, however, the attenuation becomes signal-dependent. A slightly more complicated solution uses a linear multiplier circuit 76 in place of the field effect transistor 74, as shown in FIG. 6 is shown In the RCA camera TK-47 this multiplier circuit is already built in and can be used as a contrast compression multiplier circuit. The color gain control signal can be fed to the same linear multiplier circuit 76 very easily, so that no FET attenuator 74 is required. Either type of gain control element can be used to operate the automatic color gain control circuit of the present invention.

The in Fig. The block diagram shown in FIG. 7 shows more details of the preferred embodiment of the circuit constructed in accordance with the principles of the invention, and corresponding elements are denoted by the reference numerals already used and some of the elements already described are grouped together in blocks Phase detector 38 supplied, which can be of the type MC 4044 from Motorola Ina, Phoenix, Arizona, USA, and contains an edge-triggered phase detector 78, which is used to synchronize the oscillator 48, and an exclusive OR gate 80, which is used as a frequency mixer for the The resulting difference frequency is 28OkHz for R + and R -, if the R - amplification is correctly adjusted and if the oscillator 48 is absolutely linear

is working. The reference frequency at the input of the phase detector is for (RY) -ZsWe F ₀ R (= 4.40625 MHz) and for the (^ y> line F _O b (= 4 ₍ 25 MHz)).

The output signal of the EXCLUSIVE-OR gate 86 is low-pass filtered by the filter 82 and fed to a buffer 84, the gain of which is so high that it is saturated and generates a square wave output signal if the signal GE fed to it has a high level. The pulses are fed from the buffer 84 to a divider earth electrode 86 and divided there by 4 I o, so that a signal to be measured by the detector 64 is produced. An output signal of the divider decoder 86 is used to start a reference divider which has dividers 88 and 90 and generates a time reference pulse (TR) which is fed to the coincidence detector 64 via an inverter 92. If the output signal of the divider decoder 86 divided by 4 is in coincidence with the time reference pulse TR, then the output voltage U _{e of} the integrator 66 does not change and the (R-Y) gain is exact. Otherwise Ug changes until there is coincidence. The signal CE fed to the buffer 84 ensures that the output signal of the EXCLUSIVE-OR gate 80 is only during the two control lines (only R -, only R ^{+, see FIGS. 5f and h)} Divider decoder 86 is supplied. This signal CE is the same as in FIG. 4 and in FIG. 5h.

F i g. 8 shows forms of in the circuit of FIG. 7 occurring signals, and Fig.7 should now be in individually explained.

The first positive transition of the signal BO (buffer output signal) is fed to the set input (s) of the flip-flop 94, the output signal Q of which assumes a low value and enables the flip-flop 90 of the reference divider to divide the reference color carrier For by 2. Positive transitions of the output signal Q of the flip-flop 90 clock the circuit 88, which divides by 32. After 31 clock pulses, the output signal of the circuit 88 assumes a low value, while it assumes a high value again the next positive transition of the signal / (Fig. 9c) positive transition of the signal TR (FIG. 8d) controls the circuit 88 and likewise triggers the flip-flop 94. The signal A assumes a high value and puts the circuit 90 which is divided by 2 out of operation. At the same time, the divider decoder 86 continues to count the signal BO. After four pulses of the signal BO , the output signal of the divider decoder 86, which is the signal G to be measured, as shown in FIG. The time reference pulse TR (FIG. 8e) blocks the coincidence phase detection 64 and allows the signal GE to reach the input of the integrator 66. If the signals GE and TR are in coincidence _v then the output voltage V _{1 of} the integrator 66 does not change because its input current contains a positive and a negative part each of the same amplitude, and the (7? -Y ^ color gain is correct. Otherwise Ug changes until coincidence is reached. The division factor of 32 is calculated as follows:

Pulses of the signal BO =
1 fairbearer cycle =

28OkHz
1

■ = 14.2857 // p.

4.40625MHz

■ 63 impulses.

= 226.95 ns

From FIG. 9 it can be seen that the center r of the time reference pulse (FIG. 9b) corresponds to the 63rd pulse of F ₀ R if the reference dividers 90 and 92 are programmed for a division ratio of 32. The flip-flop 90 is used to divide For by 2 in order to obtain a broader timing reference pulse than that which must be generated in order to

= 28OkHz

to be kept constant, as will be explained below. From Fig. 8a it can be seen that two cycles (R - cycle and R + cycle) are used to generate the signal U _g . The assumption that U _g does not change when the signals TA and GE are coincident must be modified as follows: U _g does not change when the mean input current of the integrator 66 of both cycles is zero. Figure 10 illustrates three cases in which there is all a mean current value of 0 for the two cycles. 10a shows the relationships when R - and R ^{+ have} exactly 280 kHz (no non-linearity of the oscillator), while FIGS. 10b and 10c show the relationships for a non-linearity of the oscillator of about 1.5%. The correct extreme values of R + and R - are calculated in the following way. The positive-going edge of the signal GE for the signal R + in FIG. 10 coincides with the 64th clock pulse, and thus the frequency of

R ⁺ =

64

275.39 ~ - 4.6 kHz

R- = = 284.27 ~ + 4.3kHz 62

It has already been said earlier that the non-linearity of the oscillator is <1% and that consequently the positive- going edge of the signal GE is always within the positive part of the pulse Ta for both cycles R + and R - when the mean input current of the integrator 66 is zero, and therefore stable operation is obtained. Fig. 11 shows a practical example of a never linearity of 1%. It can be seen that R - then equals 282.8 kHz and R + equals 2722 kHz and thus

₌ 282.8 + 277.2 ₌
2

As has already been said, the same can be done for the regulation of the (Β -! ^ - gain found that very good results are obtained, if the fB-i ^ gain with the help of a potentiometer in the matrix circuit according to FIG. 1 is adjusted, which has stable resistances and a return-stabilized Amplifier has the (Ä-YT loop

The gain is then stabilized against the effects of temperature and aging for both the (RY) and the (fl-y> line.

Fig. 12 shows details of the circuit as shown in the /? C4 camera model TK-47 is used, where the /? C4-usual designations for integrated circuits are used. The mode of operation corresponds to the explanations above.

In addition 6 sheets of drawings

Claims (10)

Patent claims: 1. A method for gain control of a color channel in a television system, in which in a test signal is input into the channel and the signal deviation that occurs in the channel as a result serves as a measure for the gain control, characterized in that for one SECAM color channel with a frequency modulator delivering the modulated SECAM color carrier Frequency deviations occurring on the basis of the test signal at the modulator are measured and based on this measurement, the gain of the color channel is regulated in such a way that the frequency deviations stay within the tolerance norm. 2. The method according to claim I _1, characterized in that the test signal is fed in during the vertical blanking interval. 3. Procedure? ¹¹ according to claim 1 or 2, characterized in that the measurement comprises frequency division of the deviating signal to generate a signal to be measured, the delivery of a time reference pulse, the coincidence determination between the signal to be measured and the time reference pulse and the integration of the signal resulting from the coincidence determination. 4. The method according to claim 1 to 3, characterized in that in the control one from the Frequency deviation derived control signal of a multiplier circuit in the signal path of the color channel is fed. 5. The method according to As / claim 1 to 3, characterized in that when djr control one from the Frequency deviation derived * control signal in a cross-circuit in the signal path of the color channel arranged gain control element is supplied. 6. Apparatus for performing the method according to claim 1, characterized by a Device (58, 60, 68) for feeding a test signal into the color channel by means of a MeP circuit (80, 62, 64, 66) for measuring the frequency deviation occurring due to the test signal, and by a control device (74 or 76) for controlling the gain of the color channel accordingly the resulting deviation. 7. Apparatus according to claim 6, characterized in that the signal feed circuit (58, 60) feeds the test signal during the vertical blanking period 8. Apparatus according to claim 6, characterized in that that the measuring device has a divider decoder (86) for dividing the deviation signal Automatic generation of a signal to be measured, as well as a reference divider (88, 90) for dividing the reference signal and generating a timing reference pulse, a coincidence detector (64) to which the timing reference pulse and the signal to be measured are supplied, and one coupled to the coincidence detector Includes integrator (66). 9. Apparatus according to claim 6 to 8, characterized in that the gain control device one arranged in the signal path of the color channel having a linear multiplier circuit (76). 10. Apparatus according to claim 6 to 8, characterized in that the gain control device has an adjustable cross-connection element (74) arranged in the signal path of the color channel