US2151145A - Reduction of interference in electric signal transmission systems - Google Patents

Description

March 21, 1939. a PERCNAL 2,151,145

REDUCTION OF INTERFERENCE IN ELECTR IC SIGNAL TRANSMISSION SYSTEMS Filed Jan. 11, 1957 2 Sheets-Sheet l Ainp/ifude limiter ,lA/l/EA/TDR WILLIAMS. PERCIVAL A TTORNE) March 21, 1939. w s PERCWAL 115L145 REDUCTION OF INTERFERENCE IN ELECTRIC SIGNAL TRANSMISSION SYSTEMS Filed Jan. 11, 1957 2 Sheets-Sheet 2 B -T,M

I B v /TIME Y \IO. .0. Q... I.

HTTTT lNVE/V TOR A TTOR/VE) UNITED STATES PATENT OFFICE REDUCTION OF INTERFERENCE IN ELEC- TRIC SIGNAL TRANSMISSION SYSTEMS William Spencer Percival, Ealing, London, England, assignor to Electric & Musical Industries Limited, Hayes, Middlesex, England, a company of Great Britain Application January 11, 1937, Serial No. 119,990 In Great Britain November 8, 1935 3 Claims. (Cl. 250-20) The present invention relates. to electric signal amplitude; in picture transmission systems, any transmission systems, and is particularly condisturbing signal which produces a signal at the cerned with the reduction of the effect of interpicture reproducer corresponding to a picture ference in such systems, due to disturbing signals. brightness value or in the Whiter than white Considerable interference is experienced, more region may produce interference, since persistence 5 particularly in wireless and other carrier wave of vision may cause the efiect of the disturbing transmission systems, due to disturbing signals signal to be prolonge of a transient nature, such as are produced, for Other disturbing signals, such as those due to example, by the ignition systems of internal comatmospherics, are also known to be of a transient g bustion engines; these disturbing signals are nature, and may also haveavery large amplitude. often of very great amplitude, and are believed It is known to provide means whereby in the to consist of short trains of heavily damped oscilpresence of interference the signal amplitude is lations with a natural frequency of the order, apreduced to produce gaps in the signal. Thus it proximately, of 40 megacycles per second, each has beenproposed to provide in an auxiliary train lasting about 1 micro-second or less. It is channel an amplitude limiting device capable of 15 believed that the decay period of such a train, passing only signals exceeding a predetermined that is, the time taken for the amplitude to decay amplitude equal to that of the maximum desired to 1 ing device, which are interfering signals, to ree duce the amplitude of the received signal in the 20 lchannel.

(that is, about 0.368) of its initial value, is about mam slgna V 0.05 micro-second, corresponding to about 2 or 3 In g i s gg g g g i z s ggg complete cycles of oscillation, and, on account of Ion 8 ep gm er this high damping, Such a disturbing Signal may there are described means responsive to transient cause interference within a Wide frequency band Interference 9 b and less amphtude It m be Clear that if such a disturbing Si than the maximum desired signal for reducing the 8 applied to a sharply tune d lightlydampediip amplitude of the received signals during the presf the transient interference. clut, the time for which the resulting interference ence 0 lasts may be many times the duration of the It has been found that m certaln circumstances signal] itself due to the lightly damped circuit interference is rendered unobjectionable simply go Oscillating at its Own natur a1 frequency under the by gappmg as above described; 011 the other influence of the disturbing signal By Way of exother; clrc'ums 2 1 p0 g i sa is ac ory. or examp n esira e e cc gg fi ii g g figggg i fi fi g of interference arising from sparking at the comg g sappned j g f mutator of an electric motor may be removed 'by 5 sired signa to a tuned circuit the disturbing gapping alone in the absence of carrier Wave Signal will g a period In the presence of a strong carrier, however, in-

v I terference re-appears, not in its original irregular t: (1 +10 Ki) form, but as a pronounced hum caused by the I 40 g 21rf suppression of the carrier during the sparking at 40 seconds to decay to an amplitude which is each commutator segment Further, in some cases, particularly where the l interfering signals are removed, and hence the e gaps are of relatively long duration, the gaps times the maximum amplitude of the desired themselves may produce objecti na efiects 45 signal, where f is the cut-off frequency of th even in the absence of a carrier wave. For intun ed circuit measured from the resonant fre- Stance where the signals COI' GSDOHd 0 oun s, quency; thus for a single tuned circuit cutting the gaps may produce audible eifects.

ff t 15 megacycles t value of 1: when p is It is the object of the present invention to pro- 2G, for example, is 0.48 micro-second, while if vide means whereby the efiect of the gaps may 50 fis 10 kilocycles, and p is again 20, t is '72 microbe rendered unobiectionable. seconds. It is clear from these examples that According to the present invention there is since transient disturbing signals are found to provided a circuit arrangement for reducing or cause marked interference in sound transmisremoving the eiTect of interference from received sion systems, such signals must be of very large signals, in which means are provided for reducing 55 the amplitude of the received signals in the presence of said interference to produce gaps in said signal, the circuit arrangement being characterised by the provision of means for reducing the loss of signal energy which would otherwise result from the gaps. Thus in one arrangement means are provided for causing the signal wave form during the gapping periods to assume either a value which is a function of the amplitude of the received signals in the neighbourhood of the gap or else a predetermined fixed value, which is other than zero.

In an alternative arrangement, interference is removed by gapping and before or after the gaps, the amplification of an amplifier through which the signal is passed is increased, preferably to such an extent and for such a time that the loss of signal energy due to the gapping is substantially counterbalanced.

In this case it is preferably arranged that t1g1= tzgz where 251 and t2 are the times. over which the gain is decreased by the gapping and increased for compensation, respectively, and g1 and g2 are the loss and increase in gain (measured in decibles) corresponding to the gapping and the increase in amplification respectively. Thus, if the gain is reduced substantially to zero at a gap, the amplification may be doubled before or after the occurrence of the gap for a time equal to the duration of the gap.

The invention is illustrated by way of example with reference to the accompanying drawings in which Fig. 1 is a diagram of a circuit for producing gaps in a signal,

Fig. 2 is a block diagram illustrating a further gapping arrangement, I

Figs. 3, 4, 5, and 8 are curves showing the. effect of g'apping and of applying the present invention to gapped signals, and

Figs. 6 and '7 are circuit diagrams showing two ways in which the present invention may be applied to gapping arrangements such as those of Fig. 1 or 2.

Certain forms of circuit whereby the gapping", which is an essential part of the present invention, can be carried out will first be described with reference to Figs. 1 and 2.

Referring to Fig. 1, terminal I1 is the input terminal to the main channel and terminal I2 is the input terminal to an These terminals are connected to the same or different receiving aerials. The main channel is relatively. sharply tuned to the frequency of the carrier it is desired to receive, that is to say it has a relatively narrow pass band, while the auxiliary channel is relatively fiatly tuned.

The auxiliary channel comprises a circuit L101 tuned to the frequency of the desired carrier or to some other frequency. A valve V1 acts as a mixer valve for mixing the oscillations from circuit L1C1 with local oscillations from an oscillator V2. In one example circuit L1C1 is tuned to a frequency between 15 and megacycles per second and the oscillator V2 serves to produce an. intermediate frequency of 4.8 megacycles per second. The valves V3, V4, V5, and V6, are arranged to amplify at this intermediate frequency. Tuning of the intermediate frequency stages is effected with the aid of variable inductances L2 and L5 which are clamped by means of resistances R1 and R2. The tuning of the various stages may be staggered so that with the damping resistances the desired breadth of tuning is obtained. The pass-band for the intermediate frequency amplifier may be 150 kilocycles per second.

auxiliary channel.

The output of the intermediate frequency amplifier isv coupled to a diode detector V1 and the output of which passes through a filter to the inner grid of a heptode valve Vs which constitutes the first valve of the main channel. The filter comprises shunt condensers C5, C6 and C1 and series elements, each comprising a resistance and an inductance, R3L'7, E3118 and R4L9. The filter is so designed. as to remove the carrier frequency of 4.8 megacycles per second and to pass Without substantial attenuation the modulation frequencies passed by the intermediate frequency amplifier, that is frequencies below about '15 kilocycles per second. It is also designed so as to avoid resonances which might cause transients to give rise to variations of an oscillatory character on the inner grid of valve Vs. valve Va is taken from terminals P to a receiver of normal type which with the valve Vs constitutes the main channel and passes a narrower band of frequencies than the intermediate frequency amplifler V3, V4, V5, and V6- A suitable positive bias is applied to the cathode of valve V8 relative to its control grids by means of a resistance Rs shunted by a by-pass condenser C8. A suitable voltage may be impressed between terminals V to make the cathode of the diode V7 somewhat positive relatively to the anode thereof and thus to delay the diode and revent it passing current for signals below a predetermined amplitude.

The input terminal 11 of the main channel is connected to the outer control grid of the valve V11 through a band-pass delay network L10C3L11C4 which is designed to give to the signals passed therethrough' a delay equal to or slightly greater than the delay introduced in signals passing from terminal I2 to the valve Va.

The operation of the circuit above described is as follows:

It will first be assumed that, whether the input I2 is connected to the same aerial as or to a different aerial from the input I1, the circuit L1C1 is tuned to a frequency different from that to which the main channel is tuned and that the frequency chosen for circuit L1C1 is such that no carrier of substantial strength is received. Both terminals I1 and I2 however, receive the same transient interference and that in the auxiliary channel is detected at V7 if the transient is above the minimum amplitude (determined by the voltage at V) to which the detector V1 is respon- 'sive, and is applied as a negative pulse to the inner control grid of valve Vs. It is arranged by a suitable choice of valve and bias for V8 that even a small transient interference signal is capable of cutting off the valve Vs by arresting the flow of electrons therein. Since the delay in the main channel is equalto or slightly greater than that in the auxiliary channel, the amplification of valve V8 and hence that of the main channel,

The output of will have been reduced substantially to zero by the time a transient reaches the outer control grid of valve Vs from terminal I1.

It is also preferable to arrange that the con denser C5 with the resistance in parallel therewith has a sufficiently long time constant to maintain the valve Vs cut-off for a suitable short period after the transient has ceased at the input to diode V1. In this way the necessity for a precise adjustment of the delay in the main and auxiliary channels which may be difficult especially when the two channels are operated at different carrier frequencies, can be avoided.

of the main amplifier remains The voltage from the filter in the output' of valve V7 may of course be used to cut-ofi valves other than or additional to V8.

It has been found that with certain types of transient interference, such as that due to switching, a connection such as that shown, capable of passing direct current between the V7 and V8 is essential to obtain the best However, with other forms of transient interference such as that arising from the ignition systems of motor cars which is of short duration, this is not the case and the output from the detector V7 may then conveniently be amplified before applying it to cut-off the valve Va. In this case any delaying which may be required can be done with advantage in the added amplifier instead of in the detector V1.

By making the voltage V very small or zero it can be arranged that transient interference even of very small amplitudes serves to cut-off the valve Vs.

With the arrangement so far described, it has been assumed that nov carrier of substantial amplitude is present in the input to the auxiliary channel and the circuit described can then be used to reduce, if desired substantially to zero, transient interference appearing in the main as well as to transients of greater amplitude.

Even where the input to the auxiliary channel is tuned to the same frequency as the main channel, the apparatus described is capable of cutting 01f the valve V3 in response to transients which in the main channel have an amplitude somewhat less than that of the maximum desired signal, as well as in response to larger transients. This is because owing to the flatter tuning of the auxiliary channel the ratio of transient amplitude to desired signal amplitude is greater in the auxiliary channel than in the main channel. In the main channel the transient will tend to persist longer than in the auxiliary channel butby a However, in order to be able to eliminate from the main channel transient interference of substantially smaller amplitude than the maximum desired signal, it is necessary to arrange that the auxiliary channel does not pass the desired signal, or any other continuous wave, at any substantial strength,

As already stated, the diode V7 may be delayed so as to respond only to signals above a predetermined amplitude. It is of course desirable to whereby the valve V7 may be set to be responsive to a relatively low amplitude and nevertheless it unresponsive to continuous waves (whether they be of the same frequency as or a different frequency from that of the desired signal), even though the amplitude of these continuous waves may change due to fading for example.

,For this purpose a special form of automatic volume control circuit is shown in Fig. 1. This comprises an additional intermediate frequency stage including a valve V9 and a tuned circuit L12R5 connected to the output of valve V9. The valve V9 is of such type that when given a low screen-grid voltage it operates as a limiter, partly by grid current damping and partly by anode current cut-off. The output of this valve V9 is fed to a diode detector V10 and the output of the latter is fed through a filter F to the grids of one or more of the valves of the intermediate frequency amplifier V3, V4, V5, Vs.

The valve V9 ensures that only signals of relatively low amplitude have any efiect upon the A. V. C'. voltage developed. Since the energy content of transients of the same amplitude as a continuous wave is much smaller than the energy content of the continuous wave, transients of the same amplitude as the continuous wave have but little influence on the A.

pendent upon the continuous waves and it can be arranged that the amplification of the inter mediate frequency amplifier V3, V4, V5, V6 varies as the amplitude of continuous waves varies in such a Way as to maintain the output of conplitude.

It may often be advantageous to associate the two channels with different aerials. For example the aerial supplying the main channel may be arranged so as to be more sensitive than the other aerial to the desired signal. The latter aerial on the other hand may be made relatively more sensitive to transient interference by suitable location or orientation thereof. For example, the aerial of the auxiliary channel may be a length of wire taken close to a source of interference, the effect of which on the main channel it is required to eliminate.

The two channels may if desired both be arranged to convey useful intelligence. For instance the main, sharply tuned channel may form part of a wireless receiver for a carrier modulated with sound signals whilst the broader auxiliary channel may form part of a television receiver for reproducing pictures to which the sounds are an accompaniment. A detector separate from V7 may be provided to rectify the output of valve Va and pass the rectified signals to a television reproducer.

The circuit L10, C3L11, C4 may be preceded if desired by an amplifier having a broader pass hand than the rest of the main channel followed by an amplitude limiter set to pass only signals having an amplitude equal to or less than that of the maximum desired signal. In this way the tendency for shock excitation of the sharply tuned circuits of the main channel amphfier is reduced because only transients of relatively small amplitude reach the sharply tuned It is not necessary that the auxiliary channel should have a continuous pass band, nor is it necessary that it should pass equally all frequencies within its pass band. For example one or more bands of continuous wave interference or be substantially suppressed in the auxiliary channel with the aid of one or more suitable band stop filters or the like.

An alternative gapping arrangement will be described with reference to Fig. 2 A main channel extends between input terminals I1 and output terminals P and contains an amplifier A. An auxiliary channel is branched from the main channel and contains an amplifier B, an ampliauxiliary channel may, if desired, be fed with signals from a different source from the main channel as described in connection with Fig. 1.

The limiting device Dis arranged to pass only signals of amplitude exceeding the maximum desired signal amplitude. Any transient or other interference of amplitude greater than that of the maximum desired signal is rectified at El and causes a direct current to flow in the winding W of a relay X and to break relay contacts Y which are in the main channel. A gap is thus produced in the signal at the terminals P in place of the interference.

The diagram of Fig. 2 is only intended to indicate the general way in which gapping may be obtained in response only to signals exceeding the desired signal amplitude. Thus the current from E, instead of operating a mechanical switch, may be used to operate a thermionic valve arranged to act as a switch. Thus the current from E can be arranged to develop a voltage which biases a valve of the amplifier A in such a way as to reduce its amplification preferably substantially to zero.

Alternatively, the rectifier may be omitted and the signals fromD may be superimposed upon the signals from A in opposition so that interference tends to be cancelled out in the main channel.

There have been described above examples of circuits whereby interference can be replaced by gaps. Although particular reference has been made to interference of a transient nature, it is to be understood that the present invention is applicable wherever interference of any kind is replaced by gaps.

The effect of gapping will be clearly understood by reference to Figs. 3 and 4. In Fig. 3 there is shown a part of a signal wave form 8. The signal in this example is a television signal and full black and full white levels are indicated by lines B and W respectively which are parallel to the time co-ordinate. At a in Fig. 3 are shown pulses of transient interference which extend considerably above full white level and consequently would produce bright flashes on the receiving screen.

With the aid of a gapping circuit, such as one of those already described, at each of the times when a pulse a of interference arrives the gain of the main channel is reduced substantially to zero and gaps bas shown in Fig. 4 are produced.

According to one feature of this invention, the signal is caused to assume, during a gapping period, a level substantially equal to the signal level immediately preceding the gap. It therefore takes the form shown in Fig. 5 where the gaps are replaced by portions 0.

One circuit with which this can be done will be described with reference to Fig; 6.

The circuit of Fig. 6 comprises a sharply tuned circuit F having effectively in parallel with it a circuit comprising a diode K, a resistance M and a bias source N, a condenser T being interposed as shown. To associate the circuit of Fig. 6 with the gapping circuit of Fig. l, the input terminals G1 and G2 of Fig. 6 are connected to the terminals P of Fig. l and the output terminals H1, H2 of Fig. 6 are connected to the inner grid of valve V8 and the lower end of resistance R6 respectively of Fig. l. The circuit F is tuned to the frequency of the desired carrier. The circuit F is normally damped by the resistance M through the diode K.

When the valve Va tends to be cut off by an interfering-signal acting through the auxiliary channel in Fig. l, the anode of the diode K, which is connected to the inner grid of the valve V8,

becomes negative and the diode insulates. The

tuned circuit F then continues to ring, or sus-,

tain oscillations, at the amplitude previously impressed upon it, that is to say at the amplitude of the signal immediately preceding the gapping period and thus maintain the signal amplitude at the desired level.

A similar arrangement can readily be applied to a gapping circuit of the kind illustrated in Fig. 2.

It will be understoodthat the circuit of Fig. 6 acts by preventing the signal level from falling appreciably below that prevailing before the interference which is being removed occurred. With this arrangement therefore no gaps are actually produced although the effect is the same as if the gaps had been formed and later filled in.

If desired, instead of arranging that the amplitude of the signal in the gap is dependent upon the amplitude of the signal preceding the gap, the amplitude may be determined by the amplitude of the signal following the gap, suitable delay circuits being included for this purpose. Further, as a development of the two arrangements described, the amplitude of the signal during gap-s can be made to depend upon the average ofthe signal amplitude before and after the gap.

The signal may also be made to assume during the gaps a fixed amplitude, other than zero, which is preferably approximately half the maximum desired signal amplitude. In a television system the signal level in the gaps would then correspond to grey.

In the elementary circuit shown in Fig. 2 as applied to television reception this may be done as follows: Assuming that passing a modulated carrier, there is provided a detector followed by a switch, which may be linked with switch Y, and which applies between the output terminals a fixed bias. This bias may correspond to grey. It is then arranged that whenever switch Y is opened the bias switch is closed so that a voltage corresponding to grey is maintained across the output terminals.

Clearly the mechanical switches can be replaced by thermionic valves acting as electrical switches.

Fig. 7 shows a circuit whereby the gaps are arrangedto be followed by an increase in gain so that the loss of signal energy which would occur due to the gaps is substantially compensated for.

Terminals P1 are connected to a main channel containing gapped signals, for example they may be connectedto terminals P of Fig. 1. Terminals P2 are connected to an auxiliary channel containing gapping pulses. For example they may be connected to the lower end of resistance R6 and the inner grid of valve V8 in Fig. 1 but in this case a valve must be interposed to reverse the sign of the pulses since the pulses are required to make the upper terminal P2 more positive.

The terminals P1 are connected to a valve V11 acting as an amplifier which has its grid connected through a condenser to the grid of a valve V12. The anodes and cathodes. of valves V11 and Vmare connected in parallel with the interposition of suitable self-biasing circuits B1 and B2.

the main channel is.

The anodes of valves V11 and V12 are connected through a common anode impedance circuit D1 to the positive terminal of a source of high tension voltage the negative terminal of the source being connected to the lower terminal P1. The impedance of the circuit D1 is arranged to be small compared with the impedance of both of the valves V11 and V12 taken separately.

The terminals P2 are connected through a .delay network E1 and an inductance L13 to the grid and cathode of valve V12 as shown.

The valve V11 is given a normal bias by means of the circuit B1 so that it acts as an amplifier of the signals fed at P1 from the main channel. The valve V12 on the other hand is arranged to be biased by circuit B2 substantially to anode current cut-off so that normally its amplification is substantially zero. When a pulse arrives from terminals P2 at the grid of valve V12 it serves to reduce the large negative grid bias thereon and the valves V11 and V12 then function as amplifiers in parallel. Because of the low value of impedance D1 relatively to the valve impedances, the efiect of the pulse is substantially to double the gain. The output is taken at terminals P3. The delay network E1 serves to delay the pulses so that the increase in gain occurs after the gapping. The purpose of the inductance L13 is to provide a high impedance path between the network E1 and the grids of valves V11 and V12.

The efiect of the arrangement of Fig. '7 is indicated in Fig. 8. Signals such as shown in Fig. 4 are assumed to be applied at terminals P1 and the pulses d are produced when the gain of the circuit of Fig. 7 is doubled as described. Assuming that the time occupied by a gap b and the following pulse at is not unduly long, the eye will be incapable of resolving the dark and white spots produced on the screen and the eiTect will therefore be the same as that produced by a signal of the form shown in Fig. 5. Similarly, where the signals represent sounds the ear will average the gap and pulse provided that they are of short duration.

It will be'clear that the increase in amplification may, if desired, be arranged to take place before, instead of after, the gaps. For this purpose, where the same pulses are used both for gapping and for increasing the amplification, suitable delay circuits are used to delay the gapped signals more than the pulses which increase the amplification.

If desired switching pulses may be used to switch one of the two valves into operation and the other out of operation, for example by producing suitable changes in the grid bias on these valves, the valves being of suitably different magnifications.

It is not necessary that the same switching signals as are used for gapping used for increasing the gain, although this is a convenient arrangement in many cases.

I claim:

1. In combination with a signal receiver having at least one tube provided with a tuned signal input circuit, a control network adapted to reduce the signal transmission efliciency of said receiver during the reception of transient intershould also be p ference, said control network comprising a rectifier having a resonant input circuit tuned to a radio frequency, means for impressing on said rectifier input circuit transient interference impressed on said signal input circuit, means responsive to the rectifier output for reducing the gain of said receiver tube, said signal input circuit being of relatively high selectivity, said rectifier input circuit having a relatively low selectivity, additional means for minimizing the loss of signal energy which would otherwise result from said gain reduction, the last means comprising a resonant circuit adapted to be coupled to said receiver tube output, the resonant circuit being tuned to the desired signal frequency, a device for normally damping said resonant circuit to prevent sustained oscillations therein, and means responsive to said gain reducing means for rendering the device ineffective thereby to permit said last resonant circuit to produce sustained oscillations of said signal frequency.

2. In a radio receiver of the type including a tube having at least a cathode, signal grid, output electrode and a gain control electrode shielded from the signal grid, a signal input circuit connected to the signal grid and cathode, an output circuit connected to said output electrode, a noise control network comprising a rectifier, an amp lifier having an input circuit tuned to said signal frequency and adapted to have combined signal and noise energy impressed thereon, said amplifier being coupled to the rectifier, means for developing potential pulses from the rectified noise energy, means for impressing the pulses on said gain control electrode in a sense to reduce the receiver tube gain substantially to zero, an auxiliary network for minimizing the loss of signal energy which would otherwise result from said gain reduction, the latter network comprising means to sustain oscillations substantially at the amplitude of the signal, a device for rendering the last means ineffective, and means responsive to said pulses for rendering the device ineffective whereby said last means produces said oscillations.

3. In combination with a signal receiver having at least one tube provided with a tuned signal input circuit, a control network adapted to reduce the signal transmission efiiciency of said receiver during the reception of transient interference, said control network comprising a rectifier having a resonant input circuit tuned to a radio frequency, means for impressing on said rectifier input circuit transient interference impressed on said signal input circuit, means responsive to the rectifier output for reducing the gain of said receiver tube, an auxiliary network for minimizing the loss of signal energy which would otherwise result from said gain reduction, said last network comprising a device coupled to said receiver tube output circuit adapted to transmit signal energy through the receiver, means normally rendering the device ineificient to transmit said energy, and additional means responsive to said gain reducing means for rendering the last means inefiective whereby said signal transmission takes place.

WILLIAM SPENCER PERCIVAL.

Images