422,914

PATENT SPECIFICATION

Application Date: July 11, 1933. No. 19691/33.

Complete Specification Left: June 26, 1934.

Complete Specification Accepted: Jan. 11, 1935.

-------------------------

PROVISIONAL SPECIFICATION

Improvements in and relating to Television and the like Apparatus

We, ELECTRIC & MUSICAL INDUSTRIES LIMITED, a British Company, of Blyth Road, Hayes, Middlesex, CECIL OSWALD BROWNE, a British Subject of 29, Monkís Drive, West Acton, London, W.3, JOHN HARDWICK, a British Subject, of 66, Drayton Gardens, West Drayton, Middlesex, and ALAN DOWER BLUMLEIN, a British Subject, of 32, Woodville Road, Ealing, London, W.5, do hereby declare the nature of this invention to be as follows:-

The present invention relates to television and the like apparatus.

In television, in order to reproduce accurately a picture of an object, it is necessary that the picture signals supplied to the reproducer should be substantially identical with those generated by scanning at the transmitter and the latter signals may comprise oscillations of any frequency between some maximum value and zero or at least a very low value. The transmission of extremely low frequencies offers considerable difficulties particularly so far as the amplification thereof is concerned.

In the specification of co-pending Patent Application No. 11204/33 (Serial No. 422,906) there are set forth methods and apparatus whereby direct and low frequency components of a signal can be re-inserted at a desired point or at a number of points in a transmission system and the signals may therefore be transmitted without these direct and low frequency components. To enable the re-insertion to take place, the signal amplitude with all components present is made to attain periodically but no necessarily regularly a fixed maximum value in one direction. Owing to elimination of direct and low frequency components, these maxima may not lie at a uniform separation from the electrical zero line which is assumed by the signal since this zero line is such that the areas enclosed by the signal wave form above and below the zero line are equal to one another. The missing components are inserted by causing the signal variations to take place about a zero line which is coincident with or is displaced by a predetermined fixed amount from the peaks of the maxima above mentioned.

In the prior specification referred to, the means described for inserting the missing components comprise essentially a unidirectionally conducting device, or at least a device which conducts better in one direction than in the other, and a circuit having a time constant which is long compared with the intervals between successive signal maxima. The unidirectionally conducting device or rectifier may be the control grid circuit of an amplifying valve or of the cathode ray device used for reproducing the pictures, or a separate rectifier such as a diode may be provided. In all cases, when a steady state of operation has been reached, the rectifier serves automatically to adjust conditions so as just to pass current at the peaks of the maxima. The circuit of long time constant ensures that these conditions are not greatly departed from in the intervals between successive maxima.

In some television systems, synchronising signals are transmitted, superimposed upon the picture signals, in the form of trains of oscillation usually of a frequency higher than the highest frequency required for the representation of the picture. These trains of oscillation are liable to be superimposed upon the recurrent maxima and where this happens they tend to prevent the correct re-insertion of direct and low frequency components and are themselves mutilated by the action of the rectifier upon their peaks.

According to the present invention, in television or the like apparatus having means for the r-insertion of direct and low frequency components into electrical signal variations comprising picture signals, recurrent maxima or minima and synchronising signals in the form of re-current trains of oscillation, in which the r-inserting means cause the signal variations to take place about a zero line coincident with or separated by a predetermined fixed amount from the peaks of said maxima or minima, there are provided means for preventing the said trains of oscillation from affecting the re-inserting means. Thus the re-inserting means may be located in an auxiliary channel branched from the main channel through which the signal variations are transmitted and a filter circuit or the like may be provided in the auxiliary channel to remove the trains of oscillation from the signals fed to the re-inserting device. The latter develops corrective signals which are red into the main channel, the corrected signal variations in this main channel still contain the trains of oscillation which are required for synchronising purposes.

In carrying the invention into effect we may proceed as follows.

A television transmitter has any suitable means for scanning an object in contiguous strips and for developing thereby picture signals corresponding to the light and shade values of the picture. It will be assumed for convenience that an increase in brightness is represented by an increase in amplitude of the picture signals in a negative direction. When a strip has been scanned the picture signal amplitude is arranged to assume for a short interval a value which is a predetermined amount more positive than that corresponding to complete black. During this interval there is also generated by the scanning means a synchronising pulse and this pulse is used to modulate a high frequency oscillation which will be referred to as a sub-carrier. The sub-carrier oscillations are generated only during the existence of the pulse and the envelope of the oscillations is substantially the same as the wave form of the pulse. The train of sub-carrier oscillation is mixed with the picture signals and is therefore superimposed on that part of the picture signal wave form which corresponds to the interval between the scanning of successive strips.

The composite signal variations comprising the picture signal, the recurrent maxima or minima, representing the intervals between strips, and the trains of sub-carrier oscillation are transmitted in any suitable way, for example by using these signal variations to modulate a carrier wave.

It will be assumed that at some point either at the transmitter or receiver the signals have been deprived of direct and low frequency components, for example due to passing the signals through a resistance-capacity coupled amplifier, and it is desired to re-insert these missing components. For this purpose the output circuit in which the signals appear has one terminal connected through a coupling condenser to the control grid of a thermionic valve and the other terminal connected to the cathode of the valve. Between the grid and cathode is bridged a circuit comprising in series a filter, a diode and a biassing battery, the cathode of the diode being connected to the negative terminal of the battery. This terminal is also connected through a resistance to the control grid of the valve. The filter may comprise an inductance and a condenser in parallel forming a circuit resonant at the sub-carrier frequency.

The coupling condenser, resistance and diode constitute the re-inserting means, the condenser and resistance having a time constant which is long compared with the intervals between successive signal maxima or minima and short compared with the periodicity of the lowest frequency present in the signals to be corrected. For example the sub-carrier frequency may be 400 kilocycles per second, the picture signals to be corrected may extend from 100 cycles to 350 kilo-cycles per second, the maxima or minima may occur at the rate of 3000 per second and the time constant of the condenser and resistance may be about one five hundredth of a second.

The conditions will be assumed to be such that the pulses constituting the re-current maxima or minima make the control grid of the valve more positive. If this is not the case the above described circuit should be modified to invert the diode. In the absence of signals the control grid of the valve is maintained slightly negative relatively to the cathode by the bias battery through the resistance. In the preliminary explanation of the operation of the circuit it will be assumed that the bias battery has been removed and the terminals to which it is connected short-circuited. The potential difference between the grid and cathode of the valve in the absence of signals will therefore be zero. When a positive maximum pulse arrives it increases the grid potential and current flows through the diode. When the pulse has ceased the grid becomes negative because of the current which flowed through the diode. The excess negative charge on the coupling condenser gradually leaks away through the resistance but, owing to the relatively long time constant of the condenser and resistance, has not fallen greatly before the next positive pulse arrives. Current will flow through the diode on the arrival of each positive pulse until the grid has assumed a negative potential such that, at the peak of each positive pulse, the grid potential just becomes zero or slightly more positive than this, so that current just flows through the diode. If, due to a change in the direct or in a low frequency component of the signal, the amplitude of the pulses changes, the normal grid potential will change also in such a manner that current only just flows through the diode at the peaks of the pulses. In this way the missing signal components are inserted with reference to the peaks of the positive pulses.

If now the grid bias battery be assumed to be connected as already described, the only effect will be to reduce the grid potential under all conditions by a fixed amount equal to the voltage of the bias battery. By means of this bias battery the valve can be arranged to work on the desired part of its characteristic curve and flow of grid current can be prevented.

The filter serves to prevent oscillations of sub-carrier frequency from affecting the function of the re-inserting device but the trains of sub-carrier oscillation are nevertheless passed undistorted to the grid of the valve.

The anode circuit of the valve comprises a suitable source of direct current. In the case of a television receiver the anode may be connected directly to the control grid of a cathode ray tube, this grid being maintained at a suitable normal potential relatively to the cathode of the tube. Suitable means may be provided for separating out the trains of sub-carrier oscillations, rectifying them and using the synchronising pulses so obtained to control the movement of the cathode ray across the screen in one direction, for example horizontally. The necessary vertical motion may be imparted to the ray under the influence of suitable framing oscillation sent from the transmitter in the intervals between successive complete scannings of a picture. These framing oscillations may also, if desired, be superimposed, in the form of trains of sub-carrier frequency oscillation, upon the picture signals.

The apparatus above described, although having the advantage of simplicity, has the disadvantage that the filter does not operate into a satisfactory terminating impedance and its filtering action may not therefore be sufficiently sharp for some purposes. An alternative circuit which will now be described does not suffer from this disadvantage.

The circuit containing the signals deprived of certain direct and low frequency components is coupled by a series condenser and a shunt grid leak resistance to the grid circuit of an amplifier valve constituting the first valve of what will be called the main amplifier. The grid of this valve may be negatively biassed by means of a battery connected between the cathode of the valve and the grid leak. The anode of the valve is connected through a resistance and a source of anode voltage to the cathode thereof and through a coupling condenser to the grid of a second amplifying valve, constituting the second valve of the main amplifier. The anode circuit of this valve may be suitably connected to the cathode ray tube or other circuit in which the signal with its restored components is required.

The oscillations to be corrected are also fed to an auxiliary channel. For this purpose one terminal of the circuit containing the oscillations to be corrected is connected through a filter element and a coupling condenser arranged in series to the grid of an auxiliary amplifying valve, the other terminal of the circuit being connected to the cathode of this valve. The filter element comprises an inductance in parallel with a condenser and a second filter element comprising an inductance in series with a condenser is connected from the point between the first filter element and the coupling condenser to the cathode of the valve. The grid of the auxiliary valve is connected to the cathode thereof through a grid leak and a suitable grid bias battery. The anode of the auxiliary valve is connected through a resistance to a source of anode voltage and through a condenser to the anode of a diode, the cathode of the diode being connected through a bias battery to the cathode of the auxiliary valve. The bias battery serves to maintain the cathode of the diode suitably negative relatively to the cathode of the auxiliary valve. a resistance is connected between the anode and cathode of the diode and this resistance and the last mentioned condenser together constitute a circuit of time constant which is long compared with the intervals between successive signal maxima or minima and form part of the re-inserting device. The anode of the diode is connected through a resistance to the grid of the second valve of the main amplifier, the resistance constituting the grid leak of the valve.

This apparatus functions very similarly to that previously described. The signals applied to the auxiliary channel are deprived of their components of sub-carrier frequency by the filter comprising the series and shunt filter element. This filter is terminated by the grid leak resistance of the auxiliary valve and may therefore be made to eliminate the sub-carrier frequency quite sharply. The signals free from sub-carrier frequency components are amplified by the auxiliary amplifier valve and fed through the condenser in the anode circuit of this valve to the diode. The signal maxima in the form of positive pulses cause current to pass through the diode until the normal voltage of the anode of the diode (that is the voltage in the intervals between positive pulses) is such that current only just passes through the diode at the peaks of the positive pulses. This voltage acts as bias voltage on the grid of the second valve of the main amplifier and since it is maintained at a value (relative to the cathode of the valve) which differs by a predetermined fixed amount from that of the peaks of the recurrent maxima the effect is to re-insert the missing components in the signals arriving on the grid of the second main amplifier valve. Although the filter prevents sub-carrier oscillations for flowing through the auxiliary channel and so affecting the re-inserting device, the sub-carrier oscillations nevertheless pass, unaffected, together with the picture signals through the main channel.

The time constant of the condenser and resistance of the re-inserting device is made longer than that of the coupling condenser and grid leak in the grid circuit of the second main amplifier valve and the latter time constant is made longer than the interval between successive maxima. The former time constant is, however, usually made shorter than those of other circuits through which the signals have passed since they contained their direct and low frequency components. The capacity of the coupling condenser in the grid circuit of the second main amplifying valve is preferably made considerably smaller than that of the condenser of the re-inserting device.

The two valves of the main amplifier and the valve of the auxiliary amplifier may be any suitable type of amplifying valve such as triode, screened grid or pentode for example.

Dated this 11th day of July, 1933.

REDDIE & GROSE,

Agents for the Applicants,

6, Breamís Buildings, London, E.C.4.

COMPLETE SPECIFICATION

Improvements in and relating to Television and the like Apparatus

We, ELECTRIC & MUSICAL INDUSTRIES LIMITED, a British Company, of Blyth Road, Hayes, Middlesex, CECIL OSWALD BROWNE, a British Subject of 29, Monkís Drive, West Acton, London, W.3, JOHN HARDWICK, a British Subject, of 66, Drayton Gardens, West Drayton, Middlesex, and ALAN DOWER BLUMLEIN, a British Subject, of 32, Woodville Road, Ealing, London, W.5, do hereby declare the nature of this invention and in what manner the same is to be performed, to be particularly described and ascertained in and by the following statement:-

The present invention relates to apparatus for television, picture transmission and reception and like purposes.

In television, in order to reproduce accurately a picture of an object, it is necessary that the picture signals supplied to the reproducer should be substantially identical with those generated by scanning at the transmitter and the latter signals may comprise oscillations of any frequency between some maximum value and zero or at least a very low value. The transmission of extremely low frequencies offers considerable difficulties particularly so far as the amplification thereof is concerned.

In the specification of co-pending Patent Application No. 11,204/33 (Serial No. 422,906) there are set forth methods and apparatus whereby direct and low frequency components of a signal can be re-inserted at a desired point or at a number of points in a transmission system and the signals may therefore be transmitted without these direct and low frequency components. To enable the re-insertion to take place, the signal amplitude with all components present is made to attain periodically but not necessarily regularly a fixed maximum value in one direction. Owing to elimination of direct and low frequency components, these maxima may not lie at a uniform separation from the electrical zero line which is assumed by the signal since this zero line is such that the areas enclosed by the signal wave form above and below the zero line are equal to one another. The missing components are inserted by casing the signal variations to take place about a zero line which is coincident with or is displaced by a predetermined fixed amount from the peaks of the maxima above mentioned.

In the prior specification referred to, the means described for inserting the missing components comprise essentially a unidirectionally conducting device, or at least a device which conducts better in one direction than in the other, and a circuit having a time constant which is long compared with the intervals between successive signal maxima. The unidirectionally conducting device or rectifier may be the control grid circuit of an amplifying valve or of the cathode ray device used for reproducing the pictures, or a separate rectifier such as a diode may be provided. In all cases, when a steady state of operation has been reached, the rectifier serves automatically to adjust conditions so as just to pass current at the peaks of the maxima. The circuit of long time constant ensures that these conditions are not greatly departed from in the intervals between successive maxima.

In some television systems, synchronising signals are transmitted, superimposed upon the picture signals, in the form of trains of oscillation usually of a frequency higher than the highest frequency required for the representation of the picture. These trains of oscillation are liable to be superimposed upon the recurrent maxima and where this happens they tend to prevent the correct re-insertion of direct and low frequency components and are themselves mutilated by the action of the rectifier upon their peaks.

According to the present invention, in apparatus for television and like purposes having means for the re-insertion of direct and low frequency components into electrical signal variations having recurrent maxima or minima and having combined therewith a modulated carrier oscillation, in which the re-inserting means cause the signal variations to take place about a zero line coincident with or separated by a predetermined fixed amount from the peaks of said maxima or minima, there are provided means for preventing said carrier oscillation from affecting the re-inserting means. Thus the re-inserting means may be located in an auxiliary channel branched from the main channel through which the signal variations, and usually also the modulated carrier oscillation, are transmitted and a filter circuit or the like may be provided in the auxiliary channel to remove the carrier oscillation from the signals fed to the re-inserting device. The latter develops corrective signals which are fed into the main channel, the corrected signal variations in this main channel still containing the carrier oscillation which in the case of television is used for synchronising purposes.

The present invention further provides television or picture telegraphy apparatus for use with electrical signal variations comprising picture signals, re-current maxima or minima having their peaks constantly related to "picture black" but varying in relation to the electrical zero of said signal variations in accordance with changes in the average amplitude of said picture signals, and synchronising signals in the form of recurrent trains of oscillation, said apparatus having a main channel or path adapted to transmit said electrical signal variations and a second channel or path containing means for re-inserting direct and low frequency components into the signals in said main channel or path and means for preventing said trains of oscillation for affecting said re-inserting means.

The invention will be described by way of example with reference to the accompanying drawings in which

Fig. 1 shows a circuit diagram of an arrangement in accordance with the present invention,

Fig. 2 shows an alternative arrangement according to the invention and

Figs. 3 to 6 are diagrams illustrating the invention.

Referring to Fig. 1, a television transmitter has any suitable means for scanning an object in contiguous strips and for developing thereby picture signals corresponding to the light and shade values of the picture. This picture signal generating apparatus is represented diagrammatically at 1 and the nature of the signals generated is shown in Fig. 3, the picture signals being designated a. In this case it is arranged that an increase in brightness is represented by an increase in amplitude of the picture signals in a negative direction, that is downwards in Fig. 3. When a strip of the picture has been scanned the picture signal amplitude is arranged to assume for a short interval a value, represented in Fig. 3 by b, which is a predetermined amount more positive than that corresponding to complete black, represented by the line c. During this interval there is also generated by the scanning means a synchronising pulse, shown at d in Fig. 4, and this pulse is used to modulate a high frequency oscillation which will be referred to as a sub-carrier. The sub-carrier oscillations are generated only during the existence of the pulse and the envelope of the oscillations is substantially the same as the wave form of the pulse. The resulting synchronising signals preceding from the apparatus designated 2 in Fig. 1 are shown in Fig. 5. The spaced trains of sub-carrier oscillations, represented by e in Fig. 5, are mixed with the picture signals and are therefore superimposed on those parts of the picture signal wave form which correspond to the intervals between the scanning of successive strips. The resulting composite signal is shown in Fig. 6.

The dotted line f in Figs. 3 and 6 represents the electrical zero of the signals, that is to say the datum line about which the oscillations will take place if the D.C. signal component is removed. The line f is so situated that the area enclosed by the wave form a, b (Fig. 3) or a, e (Fig. 6) above it is equal to that below it.

The composite signal variations are then amplified in an amplifier represented by 3 in Fig. 1 and this amplifier is of a type, for example a resistance-capacity coupled amplifier, which is incapable of passing the D.C. signal component and components of very low frequency. Such an amplifier is used because of the extreme difficulty or impossibility of producing the necessary high gain with an amplifier capable of handling the D.C. component. The signals of Fig. 6 are to be transmitted by modulated carrier and it is desired to re-insert the missing components before they are applied to the modulator represented by 4. For this purpose one of the output terminals 5 of amplifier 3 is connected through a coupling condenser 6 to the control grid of a thermionic valve 7 and the other terminal 8 connected to the cathode of valve 7. Between the grid and cathode is bridged a circuit comprising in series a filter 9, a diode 10 and a biasing battery 11, the cathode of the diode being connected to the negative terminal of the battery. This terminal is also connected through a resistance 12 to the control grid of the valve. The filter may comprise as shown an inductance and a condenser in parallel forming a circuit resonant at the sub-carrier frequency.

The coupling condenser 6, resistance 12 and diode 10 constitute the re-inserting means for re-inserting the D.C. and low frequency signal components, the condenser 6 and resistance 12 having a time constant which is long compared with the intervals between successive signal maxima or minima (b, Fig. 3), or compared with the longest interval if these intervals vary, and short compared with the periodicity of the lowest frequency present in the signals to be corrected. For example the sub-carrier frequency may be 400 kilocycles per second, the picture signals to be corrected may extend from 100 cycles to 350 kilocycles per second, the maxima or minima may occur at the rate of 3000 per second and the time constant of the condenser 6 and resistance 12 may be about one five hundredth of a second.

The conditions will be assumed to be such that the pulses constituting the re-current maxima or minima make the control grid of the valve 7 more positive. If this is not the case the above described circuit should be modified to invert the diode 10. In the absence of signals the control grid of the valve 7 is maintained slightly negative relatively to the cathode by the bias battery 11 through the resistance 12. In the preliminary explanation of the operation of the circuit it will be assumed that the bias battery has been removed and the terminals to which it is connected short-circuited. The potential difference between the grid and cathode of the valve 7 in the absence of signals will therefore be zero. When a positive maximum pulse, b Fig. 3, arrives it increases the grid potential of valve 7 and current flows through the diode 10. When the pulse has ceased the grid becomes negative because of the current which flowed through the diode. The excess negative charge on the coupling condenser 6 gradually leaks away through the resistance 12, but, owing to the relatively long time constant of the condenser and resistance, has not fallen greatly before the next positive pulse arrives. Current will flow through the diode 10 on the arrival of each positive pulse until the grid has assumed a negative potential such that, at the peak of each positive pulse, the grid potential just becomes zero or slightly more positive than this, so that current just flows through the diode 10. If, on account of a change in the direct or in a low frequency component of the signal, due for example to a change in average brightness of the picture, the amplitude of the pulses changes, the normal grid potential will change also in such a manner that current only just flows through the diode 10 at the peaks of the pulses. In this way the missing signal components are inserted with reference to the peaks of the positive pulses b, Fig. 3.

If now the grid bias battery 11 be assumed to be connected as shown, the only effect will be to reduce the grid potential under all conditions by a fixed amount equal to the voltage of the bias battery. By means of this bias battery the valve can be arranged to work on the desired part of its characteristic curve and flow of grid current can be prevented.

The filter 9 serves to prevent oscillations of sub-carrier frequency from affecting the function of the re-inserting device but the trains of sub-carrier oscillation are nevertheless passed undistorted to the grid of the valve 7.

The anode circuit of the valve 7 comprises a suitable source of direct current 13 in series with an anode resistance 14. The anode is conductively coupled to the modulator and transmitter 4 so that the D.C. and low frequency components are present up to the modulating point. The re-inserting means of Fig. 1 may be used at other points in the system where it is desired to re-insert the missing components. For example at a television receiver, the signals (after detection if the transmission is by way of modulated carrier) are applied to an amplifier incapable of amplifying the D.C. component and these signals after amplification appear at terminals 5 and 8 of the re-inserting means. The anode of the valve 7 may in this case be connected directly to the control grid of a cathode ray tube, this grid being maintained at a suitable normal potential relatively to the cathode of the tube. Suitable means may be provided for separating out the trains of sub-carrier oscillations, rectifying them and using the synchronising pulses so obtained to control the movement of the cathode ray across the screen in one direction, for example horizontally. The necessary vertical motion may be imparted to the ray under the influence of suitable framing oscillations sent from the transmitter in the intervals between successive complete scannings of a picture. These framing oscillations may also, if desired, be superimposed, in the form of trains of sub-carrier frequency oscillation, upon the picture signals.

The apparatus of Fig. 1, although having the advantage of simplicity, has the disadvantage that the filter 9 does not operate into a satisfactory termination impedance and its filtering action may not therefore be sufficiently sharp for some purposes. An alternative circuit which will now be described with reference to Fig. 2 does not suffer from this disadvantage.

In Fig. 2 like parts are given the same references as in Fig. 1.

In Fig. 2 the circuit is shown as applied to a receiver. Modulated carrier signals are received and detected by receiver 15. The signals from the receiver 15, of the character shown in Fig. 6, are amplified in an amplifier 3 incapable of amplifying D.C. and very low frequency signals. The output of the amplifier 3 is coupled by a series condenser 16 and a shunt grid leak resistance 17 to the grid circuit of an amplifier valve 18 constituting the first valve of what will be called the main amplifier. The grid of this valve 18 may be negatively biased by means of a battery 19 connected between the cathode of the valve and the grid leak 17. The anode of the valve 18 is connected through a resistance 10 and a source of anode voltage 21 to the cathode thereof and through a coupling condenser 22 to the grid of a second amplifying valve 23, constituting the second valve of the main amplifier. The anode circuit of this valve 23 may be suitably connected to the cathode ray tube or other circuit in which the signal with its restored components is required. In the case illustrated the anode of valve 23 is connected to the cathode thereof through a resistance 24 and a source of current 25 and also to the cathode of a cathode ray tube 26. The control electrode of the cathode ray tube 26 is connected, through a biasing battery 27 if necessary, to the cathode of the valve 23. The reversal of connections to the cathode ray tube compared with those described in connection with Fig. 1 is due to the use of two stages of amplification instead of one following terminals 5, 8.

The oscillations from terminals 5, 8 are also fed to an auxiliary channel. For this purpose one terminal 5 is connected through a filter element 28 and a coupling condenser 29 arranged in series to the grid of an auxiliary amplifying valve 30, the other terminal 8 being connected to the cathode of the valve 30. The filter element 28 comprises an inductance in parallel with a condenser and a second filter element 31 comprising an inductance in series with a condenser is connected from the point between the first filter element 28 and the coupling condenser 29 to the cathode of the valve 30. The grid of the valve 30 is connected to the cathode thereof through a grid leak 32 and a suitable grid bias battery33. The anode of the valve 30 is connected through a resistance 34 to a source of anode voltage 35 and through a condenser 36 to the anode of a diode 37, the cathode of the diode being connected through a bias battery 38 to the cathode of the valve 30. The bias battery serves to maintain the cathode of the diode 37 suitably negative relatively to the cathode of the valve 30. The bias battery serves to maintain the cathode of the diode 37 suitably negative relatively to the cathode of the valve 30. a resistance 39 is connected between the anode and cathode of the diode 37 and this resistance and the condense 36 together constitute a circuit of time constant which is long compared with the intervals between successive signal maxima or minima and form part of the re-inserting device. The anode of the diode 37 is connected through a resistance 40 to the grid of the second valve 23 of the main amplifier, the resistance constituting the grid leak of the valve.

This apparatus functions very similarly to that previously described. This signals applied to the auxiliary channel are deprived of their components of sub-carrier frequency by the filter comprising the series and shunt filter elements 28 and 31. This filter is terminated by the grid leak resistance 32 of the auxiliary valve 30 and may therefore be made to eliminate the sub-carrier frequency quite sharply. The signals free from sub-carrier frequency components are amplified by the valve 30 and fed through the condenser 36 to the diode 37. The signal maxima in the form of positive pulses cause current to pass through the diode 37 until the normal voltage of the anode o the diode (that is the voltage in the intervals between positive pulses) is such that current only just passes through the diode at the peaks of the positive pulses. This voltage acts as bias voltage on the grid of the second valve 23 of the mains amplifier and since it is maintained at a value (relative to the cathode of the valve) which differs by a predetermined fixed amount from that of the peaks of the re-current maxima the effect is to re-insert the missing components in the signals arriving on the grid of the second main amplifier valve 23. Although the filter 28, 31 prevents sub-carrier oscillations from flowing through the auxiliary channel and so affecting the re-inserting device, the sub-carrier oscillations nevertheless pass, unaffected, together with the picture signals through the main channel.

The time constant of the condenser 36 and resistance 39 of the re-inserting device is made longer than that of the coupling condenser 22 and grid leak 40 in the grid circuit of the second main amplifier valve and the latter time constant is made longer than the interval between successive maxima. The former time constant is, however, usually made shorter than those of other circuits through which the signals have passed since they contained their direct and low frequency components. The capacity of the coupling condenser 22 is preferably made considerably smaller than that of the condenser 36 of the re-inserting device.

The valves 18, 23 and 30 may be any suitable type of amplifying valve such as triode, screened grid or pentode for example.

The re-inserting circuit of Fig. 2, although described in relation to a receiver, can be used in place of that of Fig. 1 at any point in the system where it is desired to re-insert the D.C. and low frequency components.

It is not essential that the trains of sub-carrier oscillation should be superimposed upon the picture signals asymmetrically with respect to picture black as shown in Fig. 6. The maxima b may if desired have their peaks lying in the line c corresponding to picture black and they still constitute recurrent signal maxima in one direction and are suitable for use in re-inserting the missing components.

The invention has been described in some detail as applied to television and is clearly applicable also to picture telegraphic apparatus. It is also applicable to other apparatus which is required to handle electrical signal variations having recurrent maxima or minima (which may recur regularly or irregularly) and combed therewith modulated carrier oscillations, the apparatus having means for re-inserting direct and low frequency components with reference to the peaks of the maxima or minima. Such a case may arise for example in dealing with signal variations of any kind having a direct component and/or an alternating component of very low frequency which it is desired to re-insert with reference to signal maxima or minima and having combine with the signal variations a carrier modulated with some other form of intelligence.

Having now particularly described and ascertained the nature of our said invention and in what manner the same is to be performed, we declare that what we claim is:-

  1. Apparatus for television and like purposes having means for re-inserting direct and low frequency components into electrical signal variations having re-current maxima or minima and having combined therewith a modulated carrier oscillation, in which the re-inserting means cause the signal variations to take place about a zero line co-incident with or separated by a predetermined fixed amount from the peaks of said maxima or minima, wherein there are provided means for preventing the said carrier oscillation from affecting said re-inserting means.
  2. Apparatus according to claim 1, comprising a main channel along which said electrical signal variations are transmitted and an auxiliary channel containing said re-inserting means, a filter being provided in said auxiliary channel for preventing said carrier oscillation from affecting said re-inserting means.
  3. Apparatus according to claims 1 or 2, wherein said main channel is adapted to transmit both said signal variations and said modulated carrier oscillations.
  4. Television or picture telegraphy apparatus for use with electrical signal variations comprising picture signals, re-current maxima or minima having their peaks constantly related to "picture black" but varying in relation to the electrical zero of said signal variations in accordance with changes in the average amplitude of said picture signals, and synchronising signals in the form of re-current trains of oscillation, said apparatus having a main channel or path adapted to transmit said electrical signal variations and a second channel or path containing means for re-inserting direct and low frequency components into the signals in said main channel or path and means for preventing said trains of oscillation from affecting said re-inserting means.
  5. Apparatus according to any of the preceding claims, wherein said re-inserting means comprise in combination a unidirectionally conducting device and a circuit of time constant which is long compared with the interval between successive signal maxima or minima.

Dated this 26th day of June, 1934.

REDDIE & GROSE,

Agents for the Applicants,

6, Breamís Buildings, London, E.C.4.

-----------------------------------

Leamington Spa: Printed for His Majestyís Stationery Office, by the Courier Press. - 1935.