DE1110249B - Push-pull modulator - Google Patents

Description

The invention relates to a push-pull modulator with a push-pull amplifier, a carrier voltage source and a modulation voltage source, both of which are connected to the modulator in such a way that they are decoupled from each other.

Such modulators are known from radio technology (e.g. British patent specification 443 803, USA patents 2447701 and 2 832 651). In these circuits of radio technology, the carrier voltage to the input of the push-pull amplifier because a larger carrier energy is required at the output with a relatively small degree of modulation.

In carrier frequency telephony, on the other hand, there is no great carrier energy at the output of a modulator but the transmitted sideband should contain as much energy as possible. In carrier frequency telephony systems conventional rectifier-modulators are generally used, which always work with noticeable losses and therefore an additional amplifier must be provided. The carrier energy at the carrier entrance is generally large compared to the signal voltage, whereby the problem arises of having a larger carrier residue on To avoid output of the modulator.

In the Belgian patent 562224 modulators with transistors are described. These Circuits have the advantage that, through the transistor properties, the level of the carrier voltage, the should be on the carrier input terminals of the modulator, can be significantly reduced. this fact is of considerable importance with regard to the carrier supply generators. With those in the Belgian However, no amplification of the signal voltage is possible in the circuits described in the patent specification.

A push-pull modulator for carrier frequency telephony with only a small residual carrier at the output which requires only a small carrier input level and which has considerable gain of the signals in contrast to the attenuation that these in the conventional rectifier modulators experienced, is described, for example, in German patent specification 752409.

In the case of a push-pull modulator with a push-pull amplifier, a carrier voltage source and a Modulation voltage source, both of which are connected to the modulator in such a way that they are mutually exclusive are decoupled, this is achieved by connecting the modulation voltage source to the input of the Push-pull circuit and the carrier voltage source are connected to a high-impedance input of the modulator and that both voltage sources have a common point of fixed potential.

Push-pull modulator

Applicant:

International Standard Electric Corporation, New York, N.Y. (V. St. A.)

Representative: Dipl.-Ing. H. Ciaessen, patent attorney,
Stuttgart-Zuffenhausen, Hellmuth-Hirth-Str. 42

Claimed priority:
Netherlands from February 26, 1959

Jean Victor Martens and Etienne Fideel Poppe,

Antwerp (Belgium),
have been named as inventors

In contrast to the circuit arrangement according to US Pat. No. 2,832,051, which uses a push-pull transistor modulator relates, arrangements are also known in which the two signals are fed to different branches of the amplifier will. In this United States patent specification, the supply takes place of both signals, the input signal and the carrier signal, to the bases of two transistors that are in Emitter circuit work, the carrier signals are in push-pull on these two electrodes. the Carrier and input signals are due to the high gain of the emitter circuit of a transistor favorably influenced, and both signals are on a relatively high-resistance circuit. But transistor push-pull amplifiers in common emitter circuits tend to have large differences in the transmission characteristics of the two individual amplifiers, since with this type of circuit the difference between the current amplification factors α of the transistors is of greater importance is than with common base connection of the transistors. An emitter feedback must therefore be provided, and the input energy of the arrangement for this output must be large.

Also known are modulators with a push-pull amplifier made of two identical transistors, whose Emitter are connected to a winding of a push-pull input transformer, whose center tap is connected to a point of fixed potential, one end of the carrier voltage source also is led to the point of fixed potential, while

109 620/295

the other end is connected to the bases of the transistors.

Such an arrangement has the advantage that the input signals are applied in push-pull mode without being impermissible large differences in the transmission characteristics of the two amplifiers occur, and at the same time the carrier voltage is applied to the high-resistance base circuits of the two transistors, whereby only a low carrier input energy is required. Such arrangements are for example in French patent 1145 796 has been described.

According to the invention, the properties of the previously described arrangements, e.g. B. that the required output impedance through the basic circuit the transistor amplifier is generally very large, from which - with the input circuit open - a very small ratio between input and output voltages can be eliminated.

It is an object of the invention to provide a push-pull transistor modulator to realize, in which the load resistance can be chosen to be very small, so that there is not only an optimal power gain, but also an optimal output power.

Another feature of the invention is that the output transformer with a symmetrical Feedback winding is provided, the center of which with one end of the carrier voltage source while the outer ends of the feedback winding are connected to the bases of the transistors are switched.

This gives an output impedance that leads to an optimal power gain, without to be inadmissibly high, because now the transformation ratio of the windings of the feedback transformer is decisive instead of, as before, the corresponding transistor parameter.

In the modulators described above, it is desirable that fluctuations in the level of the Carrier voltage have as little influence as possible on the gain of the modulation circuit. The non-linear distortions should also be as small as possible. This also applies to crosstalk regarding the carrier supply.

It is also an object of the invention to take advantage of the small carrier level required to evaluate the best possible stabilization of the carrier supply.

According to a feature of the invention, one end of the carrier voltage source is via a rectifier connected to a point of fixed direct voltage potential, and this point is for the frequency the carrier voltage source is decoupled from earth, and the rectifier is polarized so that it is during the Half periods of one polarity of the carrier wave is conductive, so that when the rectifier is conductive, the fixed DC voltage potential is applied as a bias voltage to the circuit with high input resistance. With a Such an arrangement gives a well-defined bias to the two transistors during of the half-cycles of a polarity of the carrier wave to make conductive at the same time.

Another object of the invention is to make the modulator even more of the level fluctuations make the carrier wave independent, d. H. an increase in the transistor direct currents as a function of to avoid the carrier wave.

According to one feature of the invention, the common point of fixed potential of the two voltage sources is via a resistor-capacitor combination connected to the point of fixed direct voltage potential, and the amplitude of the DC voltage lying on the capacitor depends on the level of the input carrier wave.

The direct current through the transistors, which corresponds to the mean value of the emitter currents, is which flow during the mentioned half-waves (where the two transistors are conductive at the same time), stabilized, the operating point of the transistors remains constant, and damage to the transistors occurs An impermissible increase in the carrier input voltage is avoided.

The invention will now be explained in more detail with reference to the drawings.

Fig. 1 shows the block diagram of a modulator according to the invention;

Fig. 2 shows an arrangement with negative feedback, by means of which it is possible to reduce the output impedance of the to reduce amplifiers used in a modulation system according to the invention;

Fig. 3 shows an embodiment of a modulator according to the invention, which is particularly as Channel demodulator is suitable in a carrier frequency system.

The principle of the push-pull modulator according to the invention is first described in connection with the analysis an active quadrupole described. This is carried out in the manner in which V. B e 1 e ν it c h in an article entitled "Linear theory of bridge and ring modulator circuits" in the journal "Electrical Communication ”, March 1948, p. 62.

Fig. 1 shows a modulator in the form of an active four-pole network N with an input and an output terminal and a terminal which is grounded via the contact U. The input terminal is connected to a current source which supplies the current / with the amplitude / and the angular frequency Co _1. This current source has an internal resistance A ₁ . The output terminal is connected to a load resistor R. _z . The contact U is alternately open and closed during the half-waves of a square-wave voltage, which has the angular frequency ω _{& , whereby both angular frequencies oj 2} = co ₃ - Co ₁ and ω ₃ + O) ₁ and other modulation products appear at the output terminal of the modulator .

The four-pole network JV is defined by four parameters. Z is the input impedance and A is the current gain with the output short-circuited. B is the ratio of input to output voltage and Y is the output conductance with an open input. The relationship between the input voltage V ₁ , the output current Z ₂ , the output voltage V ₂ and the input current I ₁ is as follows:

V ₂ (1)

i - A · i AY ■ ν (9Ϊ

h ~ ^A h + * ^V 2 W

In _{the ideal case, R 1} and R ₂ are equal to zero, with the exception of R ₁ at the angular frequency Oj ₁ and A ₂ at the angular frequency ω ₂ . Then we get for V ₁ and v ₂ :

V ₁ = V ₁ - cos OJ ₁ 1 (3)

v ₂ = V ₂ cos OJ ₂ 1 (4)

Here, V ₁ and F _{2 are} the amplitudes of the input and output voltages.

Using a conductance matrix, equations (1) and (2) become:

V ₁ .Sv ₂ _.

h =

Av ₁

D represents the relationship between the input current /, the internal resistance R ₁ and the load resistance R ₂ and is

= Y-Z AB

Cl)

The behavior of the four-pole parameters in the various positions of the contact U is now to be examined. The parameters, A, Y and D of network N are valid when contact U is closed. If the contact is open, these parameters become 1, —1, 0 and 1. In addition, the parameter Z (with open contact) becomes zaZ + R and can easily be changed

V ₁ -

= Z + R

and can be calculated from (1) and (2), provided that Z ₁ = -Zg. R is then calculated as:

_{R =} IL + ^ HL * 1

The parameters in the equations (5) and (6) are replaced by modified parameters berücfcsichtigen the operation of the modulation contact U. This gives:

1 1 + U Z 2 A. 1 + U 7, ?,

I Lr JL) _v _ / ^B _L ±

Z + i? 2 j ^Vl \ Z 2

i?

1 \ Z 2

i?

1 -u

(10)

(H)

Here w is a square wave function, the magnitude of which is alternately +1 and −1 during successive half periods of the modulation frequency (with the angular frequency ω _3).

The function u is defined by the following Fourier series:

u - - sinco _z t + -Tr- sin 3 ω ₃ t + .... (12)
π \ 3 /

Since [in equations (3) and (4)] it was assumed that the input voltage V ₁ and the output voltage V _{2 are} pure sinusoidal voltages with the angular frequencies Co ₁ and ω ₂ , and since the current source 1 is also a pure sinusoidal voltage with the angular frequency W ₁ , the terms of the angular frequency W ₁ in equation (10) and the terms of the angular frequency CO ₂ in equation (11) can be equated. This creates two new equations that link the amplitudes V ₁ and V _{2 of} the input and output voltages with the amplitude / current source.

J =

R ₁

2 \ ZZ + R V ₁ - \ *
^{1 line}

Z + R

(13)

Z + R

(14)

The power gain G of the modulation circuit is defined as the ratio between the output power at the load resistor R ₂ and the maximum power that can be drawn from the current source / with the internal resistance R ₁ :

Since the ratio between V ₂ and / [in equation (15)] can be determined from the equation pair (13) - (14), the power gain is now obtained as a function of the various parameters Z, B, A and Y as well as the internal and load resistors A ₁ and R ₂ . It is easy to see that the power gain G reaches its maximum when -R ₁ and R ₂ assume the values defined as follows:

R,

D (Z + R) + Z R + 2Z

(16)

B. ~ Z

Z + R

Z + R

(17)

In many cases the parameter Y, which is defined as the master value with an open input, is very small and can be neglected. This is the case when the network is embodied by a transistor that operates in common base. With Y equal to zero, equations (16) and (17) lead to simple ones

1 HO

Express, for the optimal sizing of the interior can be what working on a fairly small

and load resistances R ₁ and R ₂ taking into account- ~ "'~ - -. _ - .. -

the remaining network parameters Z, B and A:

AB \ 1

A ₁ = - ± £ = - (18)

(19)

If Z is positive here, the positive sign applies to R ₁ and the negative sign to R _2.

In such a case, an acceptable positive optimum for R ₂ can only be achieved by a negative amount of A · B , which is the case with a common base transistor.

For such maximum amounts of R ₁ and _{R 2} , where Y is zero, the maximum gain G _M can easily be derived from equations (13), (14) and (15):

25th A.

W) -

(20)

= 1.02 neper or 8.9 decibels.

Load resistance R ₂ allowed. The described feedback arrangement allows B to be made practically independent of the gain criteria (e.g. the transistor or the parameters). The special type of negative feedback (shunt-shunt negative feedback) is mostly used to reduce the output impedance of an amplifier, but in the present case of a modulation amplifier according to FIG. 1, this type of negative feedback cannot be used because the input circuit has a small resistance at the output frequency .

Figure 2 shows that parallel-series negative feedback can be used to achieve a useful reduction in output impedance. The network N in Fig. 2 is shown without modulation contact, and it can be seen that the feedback transformer TR has a first winding which is parallel to the load resistor R ₂ and a second winding whose number of turns is κ times the number of turns of the first winding and is inserted between the lower terminal of the network N and earth. The parameters of the modified network, which has the input voltage V ₁ 'and the output voltage V ₂ ' as well as the input current Z ₁ (as before) and the output current i _z ' , can now be derived from the basic network equations (1) and (2) as well from the composite network equations, which are easy to set up for the circuit according to FIG. 2, are calculated.

It can be seen that the numerator of this equation is equal to the ratio between A and B , which corresponds exactly to the maximum power gain of the quadrupole N when there is no modulation, as in a normal amplifier. The denominator of the above equation represents the losses in the presence of modulation and is

40

This is exactly the loss of a conventional rectifier-modulator between two filters works and the arrangement described above has the advantage that a modulation by means of an amplifier is made possible, whereby components are saved, while at the same time an optimal design the circuit arrangement is possible, as well as the maximum gain and a minimum of losses is present during modulation.

If the internal resistance R ₁ and the load resistance R _{2 are} determined by equations (18) and (19), not only is the amplification of the modulation arrangement a maximum, but the terminations at the input and output also assume optimal values. The maximum amount of the load resistance is inversely proportional to the parameter B , which is very small for a common base transistor. However, such a high load resistance is usually undesirable. However, it will generally not occur in connection with a maximum output power in the load. It is therefore desirable that the parameter B has an amount such that optimum power gain and maximum output power are achieved.

2 shows an arrangement with negative feedback, by means of which a larger value of B '= V ₁ + η · V _{2 is obtained}

■ ν »

O = i ₂ '- i ₂

• μ OV

(21)
(22)
(23)

The first two equations above represent the relationships between the new input and output voltages V ₁ and V ₂ and the voltages V ₁ and V ₂ , while in equation (23) the ampere-turns of the transformer are set to zero, which is the ideal case applies. The voltages V ₁ , V ₂ and the current / ₂ are eliminated from equations (21), (22) and (23), and the modified network equations result:

v / = Z I ₁ + (B-Bn + ή) V ₂ ' (24)

i ₂ '= (4-An n) Z ₁ + (1 - «) * · Y ■ v ₂ ' (25)

From the above pair of equations it can be seen that the parameter Z has not been changed, and since Y was previously assumed to be zero, Y is also zero in the modified network. However, parameters A and B have been changed. But for the special case of the basic circuit of a transistor, A is equal to - 1, from which it can be seen that the parameter A does not change compared to earlier. If the parameter β of the network N is very small and approximately of the order of magnitude of η , it can be set equal to η . From equation (19) it can be seen that if Z is positive and A is negative, the value B will be positive, which means that η will also be positive, which is the case when the two windings of the transformer TR are connected in series but against each other ( s. draw).

If one applies the equations (18) and (19) with the feedback arrangement according to FIG. 2 to the arrangement according to FIG. 1, B can be replaced by η

ίο

and a useful value for n, less than one, can be chosen, achieving a desired amount for the load resistance R ₂ and a corresponding power gain for the modulation arrangement. The load resistance R ₂ should be specifically chosen so that it corresponds to an optimum, whereby a maximum output power in connection with a maximum power gain is achieved.

which the demodulator has with the transistors T ₁ and T ₂ , said hybrid circuit can be a resistor hybrid circuit which is afflicted with a loss of 10.7 decibels instead of 3 decibels as in a coil fork circuit.

The push-pull output transformer TR ₃ has

a third feedback winding, derenMittelanzapfung of the transformer TR ₂ directly connected to the ungrounded end of the secondary winding of a detailed design according to the invention, io, and the outer ends of these feedback Taking into account the ideas which are development in comparison to the bases of the transistors T ₁ and linkage with Figures 1 and 2 discussed is ~
now described with reference to FIG.

3 shows a push-pull transistor demodulator, part of a channel modulator-demodulator unit for carrier frequency transmission. The input terminals P ₁ and P ₂ are connected to the primary winding T ₂ .

The described arrangement basically corresponds to the working principles that are related with FIGS. 1 and 2 were explained. The carrier voltage.

which is connected to terminals P ₃ and P ₄ , Ver

a symmetrical input transformer TR ₁ connected. Parallel to this winding is the cause that the bases of the transistors T ₁ and T _{2 are} either negative or positive with regard to the emitter potentials, so that both transistors develop.

tuning capacitor C ₁ , so that the input circuit on 20 neither amplify nor blocked. In the latter case, the center frequency of the incoming signals is matched, ie to the angular frequency ω _ν
Terminals P ₃ and P ₄ represent the

Input terminals for the carrier voltage with the angular frequency CO ₃ represent. These terminals are connected to the primary winding of the carrier input transformer TR ₂ via the decoupling resistors R ₁ and R ₂ . The decoupling resistors also allow the carrier voltage via terminals P ₃ and P ₄ to be fed via similar resistors to the modulator (not shown), which is part of the channel-modulator-demodulator unit.

One end of the secondary winding of the transformer 7'.R ₂ is connected directly to the center tap of the secondary winding of the transformer TR ₁ , and this common point is grounded. The outer ends of the symmetrical secondary winding of the transformer TR ₁ are connected to the emitters of the transistors T ₁ and T ₂ , which work in push-pull with respect to the input signal applied to the terminals P ₁ and P _2. The collectors of these two transistors are each connected to the negative pole of a 48-volt battery via one of the two primary windings of the push-pull output transformer TR ₃ and each via one of the two primary windings of a further symmetrical output transformer TR _1. Each half of the primary winding of the transformer TR ₃ is matched to the center output frequency of the lower speech frequency sideband, ie specifically to the angular frequency ω, of the carrier wave, with the aid of capacitors C ₂ and C _3.

The two primary windings of the symmetrical output transformer TR ₁ are tuned to a signal frequency with the help of capacitors C ₆ and C ₇ , which has a higher frequency than the speech frequency band and via the secondary winding of this transformer TA ₄ to a signal receiver (not shown) that is connected to terminals P ₇ and P _{8 are} connected.

The secondary winding of the output transformer TR ₃ has the terminals P ₅ and P ₆ , to which a low-pass filter (not shown) is connected, with which the desired sideband is filtered out. The filter output can be connected to a hybrid circuit and this in turn to the modulator (not shown) and can also be used to couple the modulator and demodulator with a two-wire circuit. With regard to the essential amplification, there is a very high level of attenuation _{between terminals P 1/2} and P _5/6.

The arrangement of FIG. 3 is of particular interest in connection with a channel demodulator. In such a case, the frequency of the carrier voltage source is quite large compared to the speech frequency bandwidth, and the spectrum of the speech frequency band is close to the carrier frequency. After the output _{terminals P 5} and P ₆ , the speech frequency sideband is screened out, which is very different either from the carrier frequency or from a frequency of the upper sideband. This means that if the output circuit is matched to the desired sideband by means of the capacitors C ₂ and C _3, any remaining carrier residue is also automatically reduced. The output terminals P ₅ and P ₆ may - as mentioned before - be connected to a low-pass filter which has a small impedance at the input frequency bandwidth. The carrier frequency applied to terminals P ₃ and P ₄ is automatically included in the input frequency band when the carrier frequency is close to the frequency range of the input signals.

It should be noted that the analysis with respect to FIG. 1 contains the special condition that the input impedance, except for the frequency of the input signals, is zero and the output impedance, except for the frequency of the desired output signals, also assumes the value zero.

In the demodulator arrangement according to FIG. 3, however, while the output impedance of the demodulator except for the desired lower speaking frequency sideband, practically still as zero can be considered, the input impedance approximately only in terms of the desired lower Sideband can be assumed to be a short circuit. If you take z. B. assume that the input signals in a Channel from 60 to 64 kHz and that the channel carrier is 64 kHz, the lower sideband is obtained with 0 to 4 kHz, while the upper sideband appears in the frequency range 120 to 124 kHz. at With an input tuned to 60 kHz, this tuned input circuit can be used for the upper sideband can no longer be viewed as a short circuit, but operates as such with respect to the lower sideband. The voting conditions adopted in connection with Fig. 1 are no longer valid, a greater gain than in equation (20)

109 620/295

can now be achieved, although the new optimal values for the internal resistance .R ₁ and the load resistance R ₂ do not differ significantly from the values defined by equations (18) and (19).

One of the advantages of transistor modulators over rectifier modulators is that im In contrast to the latter, a very small carrier power controls a much higher signal power.

FIG. 3 shows how this advantage is realized in a modulator that operates almost ideally.

A voltage divider consisting of the resistors R ₃ and i? ₄ , is connected between the negative pole of the battery and earth. A capacitor C ₄ is located in parallel with the resistor R _t . The connection point of the resistors R ₃ and i? ₄ is led via the rectifier G _L to the center tap of the feedback winding of the transformer TR ₃ .

The base-emitter path of the transistors T ₁ and T ₂ work as a full wave rectifier in conjunction with the RC element R _{5 ZC 5} . Since the transistors are of the PNP type, the capacitor C _{5 is} charged negatively and the voltage across it becomes higher negative as the carrier voltage increases. On the other hand, since the bases of the two transistors are led to a fixed, predetermined negative potential via the feedback winding of the transformer TR ₃ and the rectifier G _L , the potential at the cathode of the rectifier G _{L is} lower than the negative one during the negative half-waves of the carrier voltage Potential at the junction of resistors R ₃ and i? ₄ , whereby this rectifier is conductive to the potential at the center tap of the feedback winding of the transformer TR _{3 is} limited to the bias voltage at the junction of R ₃ and R ^ . If the rectifier is conductive, a circuit is connected to the secondary winding of the carrier transformer TR ₂ , which consists of the impedance of the rectifier in series with the parallel connection of A ₃ , A ₄ and C ₁ and the parallel connection of the resistor R ₅ and the capacitor C ₅ . A further increase in the carrier level can be prevented by this circuit (via the carrier transformer TR ₂ ) .

The normal level of the carrier voltage is preferably selected so that it is above the threshold at which the rectifier G _L becomes conductive, whereby, caused by the voltages at the bases of the transistors T ₁ and T ₂ , the transistors become conductive at the same time, whereby the voltage at the bases during a half cycle is always equal to the predetermined maximum value of the potential prevailing at the connection point of the resistors R ₃ and R _i. The limiter arrangement described has the advantages that the carrier level has only a slight effect on the amplification, small non-linear distortions occur and, as already mentioned, there is little crosstalk over the carrier supply. It can be assumed that the carrier level fluctuations have a small influence on the emitter current of the transistors, since the amplitude of the carrier wave at the transformer TR _{2 is} very much smaller than the potential at the junction of the resistors R ₃ and R _t , e.g. B. 0.5VoIt versus -16VoIt. In percentage terms, the fluctuations in the potential at the resistor R ₅ , which result from the level fluctuations of the carrier, are much smaller, which also corresponds to very small changes in the emitter current of the two transistors T ₁ and T ₂ .

From the above description it can be seen that the circuits described are based on the inventive concept can be varied in many ways. For example, more than two can Transistors are used. If you connect two transistors in parallel, you can use the circuit If you draw a higher power, if you connect two transistors in cascade, you get a higher one Reinforcement. If desired, add any input frequencies to the output of the modulator suppress, it is advantageous to arrange four transistors instead of the two in FIG. Such Suppression or any significant attenuation of the input frequencies at the modulator output advantageous, since this allows the filter arrangements at the exit to be significantly simplified.

Claims (4)

PATENT CLAIMS: 1. Push-pull modulator with a push-pull amplifier, a carrier voltage source and a modulation voltage source, both of which are connected to the modulator in such a way that they are decoupled from each other, and in which the push-pull amplifier consists of two identical transistors, the emitters of which are connected to a winding of a push-pull input transformer , whose center tap is connected to a point of fixed potential, while the carrier voltage source has one pole at the bases of the transistors and the other at a point of fixed potential, characterized in that the output transformer is provided with a symmetrical feedback winding, the center of which with the one end of the carrier voltage source is connected, while the outer ends of the feedback winding are connected to the bases of the transistors. 2. Push-pull modulator according to claim 1, characterized in that one end of the Carrier voltage source via a rectifier to a point of fixed direct voltage potential is switched and this point is decoupled from earth for the frequency of the carrier voltage source is and that the rectifier is polarized so that it during the half periods of a polarity of Carrier wave is conductive, so that the fixed DC voltage potential when the rectifier is conductive as a bias on the circuit with high input resistance. 3. push-pull modulator according to claim 2, characterized in that the common point fixed potential of the two voltage sources via a resistor-capacitor combination is connected to the point of fixed direct voltage potential and the amplitude of the capacitor lying DC voltage depends on the level of the input carrier wave. 4. push-pull modulator according to claims 1 to 3, characterized in that the push-pull modulator is used in a carrier frequency system as a channel demodulator, the Input circuit for the modulation voltage source, d. H. the input transformer to which Frequency of the carrier voltage source is tuned and a low-pass filter is provided at the output of the demodulator, which is used to filter out the lower speech frequency band and represents a small resistance outside the transmission range, and that the frequency of the carrier voltage source is greater than the speech 14 frequency band is, e.g. B. 64 kHz for a voice frequency band of 4 kHz. Considered publications: German Patent Nos. 752 409, 945 853, 679;
French patent specification No. 1145 796. 1 sheet of drawings