US2003428A - Volume control for transmission circuits - Google Patents

Description

June 4, 1935. F. A. COWAN VOLUME CONTROL FOR TRANSMISSION CIRCUITS Filed Sept. 20, 1952 10 20 39 40 5 0 60 Livss/mdeabbels LinearRecgfiar Out J IZZZZLEF 5 Beatox 5 In 4 [tienualbr (3 Patented June 4, 1935 UNITED STATES PATENT OFFICE VOLUME CONTROL FOR TRANSMISSION CIRCUITS a corporation of New York Application September 20, 1932, Serial No. 634,083

7 Claims.

This invention relates to volume control for amplifier and other electrical circuits by means of a new form of controllable attenuator to be introduced in the circuit.

Its purpose is to provide an attenuator of a type which is simple and economical in construction, and continuously adjustable to any value of attenuation, instead of by steps.

Another purpose is to adapt this attenuator to electrical circuits, such as transmission lines,

in a manner so that it will act to control the output volume of the circuit; that is, to serve as a volume compressor or to serve as a volume and power limiting device.

Still another purpose of my invention is to replace rectifier or non-linear elements of the active type which have been used in the art heretofore for such purposes, by passive rectifiers or non-linear elements, thus bringing about a simplicity in operation and an increase in stability.

The properties of this attenuator are such that it is possible, and one purpose of my invention is, to provide a circuit in which, over a wide range, the output volume without regard to the input volume shall be substantially constant, or shall even vary inversely as the input volume.

The invention will be better understood by reference to the following specification and the accompanying drawing, in which Figure 1 is a diagram of my attenuator circuit; Figs. 2, 3 and 4 show curves giving the characteristics of a type of passive rectifier which I prefer to use in this invention; and Figs. 5 and 6 show circuits incorporating my attentuator for definite purposes.

Referring more specifically to Fig. 1, there is shown a four-terminal attenuating circuit with input terminals 3* and 4 and output terminals 5 and 6. Between the terminals 3 and 5 there are shown connected in series three resistances ll, l2 and I3, and between the terminals 4 and 6 there are connected in series an identical sequence of resistances H, I2 and i3. These resistances are given values which are appropriate to the characteristics of the transmission line or other electrical circuit with which it is to be associated. Bridged across the network thus described between the resistances II and I2 is a shunt element l5 which is to be considered as a resistance, anda resistance of such character as to be controllably variable. A similar resistance I6 is bridged across the circuit, connection being made between the resistances l2 and [3.

One form of resistance which I find particularly suitable for carrying out my invention is the so-called thyrite. Still another form is the so-called rectox rectifier element which, as is well known in the art, consists of alternate discs of copper and lead, the surface of one or both of the metals being treated in a manner to give a rectifier of highly valuable properties. The shunt element I 5 is shown as consisting of four such rectox units in series, the two upper ones being so directed as to permit current to fiow downward in the direction indicated by the arrow, and the two lower ones being connected in the opposite direction. Similarly, in the shunt element IS the two upper rectifiers are connected to permit current to flow upwards, as indicated by the arrow, and the other two are connected in the reverse direction. While two units are shown in each part of the element l 5, it is to be understood that any number desired may be so connected in series, the number to be used depending upon the voltages which are impressed on the circuit. Direct current from battery l8, with potentiometer i9,

is supplied to the rectifier elements, connection being made at the midpoints of the shunt elements l5 and I6. By appropriate poling of the battery [8 it is seen that current will flow through each part of both shunt elements, as indicated by the arrows, and the value of the resistance offered by these shunt elements to alternating currents arriving at the input terminals 3 and 4 is controlled in a manner to be described hereinafter. Also, for purposes hereinafter to be described, I connect in series with the direct current from the battery l8 choke coils 2| and 22, and in shunt across the direct current supply I connect condensers 23 and 24, these preferably being in series and with the midpoint connected to ground. An ammeter A may be introduced to give infor-- mation as to the size of the current flowing through the rectifier units. An inspection of Fig. 1 will make it evident that, for currents from the terminals 3 and 4, the attenuator is a ladder-type structure and may be looked upon as made up of two I sections of network in series, the one section comprising the resistances i I, the element I5 and half of the resistances l2, and the second section comprising the other half of the resistances l2, the resistances l3 and the element 16. It will be observed, also, that the system is a balanced structure, which I find advantageous in its use on transmission lines.

This invention will be better understood by conobserved that the relationship is non-linear in a manner characteristic of all rectifying devices. The reciprocal of the slope of this curve at any point gives the value of the impedance of the unit at that portion of its characteristic, and upon investigation I find that in the case of these particular rectifiers, the impedance over a consider able range is inversely proportional to the current flowing through the unit. This is illustrated in Fig. 3, where the impedance, given by is plotted as ordinates, and the reciprocal of the current in milliamperes is plotted as abscissae. The full line represents the experimental results, and it was found for a particular rectox element that the straight portion of the curve can be represented by the equation a In other words, in the circuit of Fig. 1, the resistance of'the shunt branches may be altered by changing the direct current from the battery I8 flowing through them, and in particular the impedance of the shunt elements may be reduced to very low values so that the circuit ofiers rela-. tively high attenuation for signals to be passed therethrough. This is illustrated, more fully by Fig. 4, in which curve (a) shows the relationship between the current I' and the loss introduced by this attenuator. For this particular case the resistances II and I3 were each given values of 250 ohms, and the resistances I2 were given a value of 100 ohms. In this case, also, the attenuator looked intoan impedance of 600 ohms in both directions. The current is measured in milliamperes and the loss is measured in decibels, where the loss in decibels is defined as ten times +1.75, I=current in milliamperes.

the logarithm of the ratio of the input to the output energy, or twenty times the logarithm of the ratio of the input to the output current of the attenuator.

The behavior of the rectox element is still better indicated by curve b of Fig. 4, in which twenty times the logarithm of the direct current flowing through the unit is plotted as ordinates and shown to the right-hand side of the figure. It will be observed that over a considerable portion of its length the curve bis substantially a straight line, as indicated by the dotted line. For one particular element on which observations were made the equation of this curve is L=20.5+26.4 log I where L represents the loss introduced by the attenuator when the resistance I2 has a value of 100 ohms; If this resistance were changed, then the steepness of the curves of Fig. 4 would be changed. In particular, if the resistance I2 were gradually reduced in value, these curves would become steeper, thus showing a smaller loss or attenuation for a given current through the rectifier' elements.

The manner in which the attenuator functions can be described effectively in connection with the curve I; of Fig. 4. Thus, if the direct current through the rectifier elements, as derived from a battery IE5 or from independently rectified current coming in at the terminals 3 and 4, should be increased by a factor of ten this would be indicated on curve b by a change in ordinates from, say, +10 to If this increase is due to an increase in volume of an incoming signal, it

would correspond to a volume gain impressed at 3 and 4 of 20 decibels. Such an increase in the current through the rectifiers, however, so reduced the impedance of the rectifier elements as to introduce an additional loss in the attenuator of approximately 27 decibels, as appears from Fig. 4. Thus, if the direct current were derived from an incoming signal, an increase in volume of this signal at the terminals 3 and 4 would actually give a decrease in volume of the signal at the terminals 5 and 6. By increasing the slope of curve b this condition could be modified so that the loss due to the attenuator is not so great as the corresponding increase in incoming signal which produced the increase in loss of the attenuator. In that event the attenuator serves as a volume compressor in the sense that the volume changes at the output terminals 5 and 6 are not so great as the volume changes at the input terminals 3 and 4. Furthermore, it is evident that by suitable adjustment of the resistances I2 or other constants of the attenuator, one could obtain a condition between the two given above such that any increase in volume of the-incoming signal would bring about an equal increase in the loss of the attenuator, and in that event the output over a considerable range would be independent of the input volume. This particular condition is one which is especially useful in alternating current telegraphy in which it is desirable that the magnitude of the signals operating on the receiving apparatus shall be independent of variations in the transmission characteristics of the transmission line or. of other effects which might alter the intensity of the signal as it arrives at the receiving station. Such variations, as is well known in the art, give rise to so-called telegraph bias, and one of the functions of my attenuator would be to overcome these troublesome efiects.

To accomplish such results with my attenuator, circuits of the form shown in Figs. 5 and 6 may be used. Referring particularly to Fig. 5, there is shown a transmission line L in which is introduced the four-terminal attenuator of Fig. 1. In this case, however, instead of using the battery I8 and potentiometer I9 as a source of direct current across the terminals I and 2, the'direct current is derived from a supplementary amplifier-rectifier unit. This supplementary unit consists of an amplifier SI the input of which is bridged across the incoming line L and the out put of which is connected to a rectifier 32, preferably a linear rectifier; that is, a'rectifier in which the magnitude of the rectified current is proportional or substantially proportional to the voltage impressed on the input of the amplifier 3!. The direct current from the rectifier 32 is impressed on the attenuator through the terminals I and 2. While any suitable form of rectifiers may be used in the element 32, the preferred form is again the rectox element. The figure shows four such elements connected in a manner which is well known in the art to deliver direct current to the terminals I and 2. While such a unit is not linear in itself, it may be combined with an amplifier of such characteristics that the overall efiect is linear; that is, the current flowing from the element 32 to the attenuator is proportional to the current in or the voltage across the line L. By a suitable adjustment of the amplification of the element 3!, or other characteristics, the direct current through the attenuator may be adjusted to any desired average value. Variations in the strength of the incoming signal will then alter this direct current and correspondingly alter the loss due to the attenuator, and this may be of such magnitude as to compress the output volume or to keep the output constant in magnitude or to cause it to vary inversely as the input volume.

It is evident, of course, that the attenuation should not vary with a frequency equal to that of the incoming signal but only in accordance with variations in the intensity of said signal; that is, in accordance with the envelope of the incoming wave. Provision for this is afforded by the inductances 2| and 22 and the condensers 23 and 24 of Fig. 1 which serve to wipe out the signal frequency from the direct current supplied to the attenuator.

Fig. 6 shows a modification of the circuit of Fig. 5 in that the direct current for the attenuator is derived from the output terminals of the line L, here indicated as the output of the amplifier 34. If the output of this amplifier 34 increases, then the direct current from the rectifier 32 supplied to the attenuator is correspondingly increased, thus increasing the loss in the circuit leading to the input amplifier 34 and bringing about results which are analogous to those described in connection with Fig. 5.

Certain variations in the circuits may be pointed out. Thus, by the use of a supplemental battery 33 in Fig. 5, which may be introduced by an appropriate switch, it is possible to shift to different points of the characteristics of the attenuator. If curve b of Fig. 4 were strictly a straight line, no change in volume compression would be introduced by such shifting, but inasmuch as the curve changes its direction, particularly for smaller values of direct current, it is possible to change the attenuation characteristics of the device as a whole by such supplemental battery 33. For example, this battery might be so poled that for weak incoming signals the net direct current would be zero or would hold one on a portion of the characteristic where the loss of the attenuator is not suflicient to neutralize the increase in volume of the incoming signal, until that incoming signal has reached a. fairly large value, after which the operation would be over a part of the characteristic in which the increase in loss of the attenuator is as great as the increase in volume of the incoming signal. Thus, the device as a whole would operate as a power limiting device, the output signal being substantially proportional to the input signal up to a certain maximum value. From what has been said before, it is apparent that the adjustment may be such that after reaching a certain value, the output would actually decrease for further increase of input volume. In Fig. 6 a similar battery 33 is shown, but in this case it is connected in parallel to the rectifier 32 instead of in series, and the amount of current which is supplied by the battery can be controlled by the resistance r.

I find that the device of Fig. 6 is particularly useful in monitoring program circuits in noisy locations, such as, for example, in test rooms. In such places it sometimes occurs that the room noise masks line noise and cross-talk so that the monitor normally does not get a true picture of circuit noise conditions. With the compressor of Fig. 6, however, as soon as there is a short pause in the program, such as between sentences of a talk, the gain automatically increases whereupon the undesirable noise conditions are made evident. The quality of the program is not impaired seriously by the compression, and if the monitor desires to listen for quality the key for battery 33 may be closed, thus converting the device into a linear amplifier.

While the invention has been described largely in terms of rectox elements, it is to be understood that it is not limited to these. Thus, any non-linear passive resistance may be used, such as thyrite in which the condition is the same for positive and negative voltages but in which the current is not linearly proportional to the voltage.

What is claimed is:

1. A four-terminal attenuator comprising two balanced I sections in series, the series elements being pure resistance and the shunt elements being rectifiers of non-linear characteristic, the first shunt element consisting of two rectifiers in series and opposed to each other, the second shunt element consisting of an identical pair but poled in the opposite direction, and means associated with the rectifiers to change their effective resistance.

2. A four-terminal variable attenuator comprising two balanced I sections in series, the series elements being fixed resistances and the shunt elements being rectifiers the resistance of which is dependent on the current, the first shunt element consisting of two rectifiers in series and opposed to each other, the second shunt element consisting of an identical pair but poled in the opposite direction, and a direct current supply for the rectifiers to change their effective resistance.

3. The combination of claim 2 characterized by the fact that the direct current to the rectifiers is made variable to vary the resistance of the rectifiers.

4. The combination of claim 2 characterized in this, that the direct current supply is connected across the midpoints of the shunt elements.

5. In a circuit adapted for volume control of a signal a line with input and output terminals, an attenuator as described in claim 2 connected in the line, the direct current supply consisting of rectified current controlled by and proportional to the incoming signal at the input terminals.

6. In a circuit adapted for volume control of a signal a line with input and output terminals, an attenuator as described in claim 2 connected in the line, the direct current supply consisting of amplified and rectified current proportional to the signal volume, the direct current being adjusted to such volume by amplification and by resistance that the output volume is independent over a wide range of the volume of the incoming signal.

7. A four-terminal attenuator having an attenuation which is a function of the input power, said attenuator comprising an I-section, the series element being a pure resistance and the shunt element consisting of two rectifiers of nonlinear characteristic in series but opposed to each other, and means associated with the rectifiers to change their effective resistance.

FRANK A. COWAN.

Images