US1964522A - Phase control system - Google Patents

Description

Patented June 26, 1934 OFFICE PHASE CONTROL SYSTEM Harold M. Lewis, Allenhurst, N. J.

Application June 13, 1929, Serial No. 370,696

7 Claims.

This invention relates to modulation of alternating current energy for communication purposes and more particularly to the control of the phase of alternating current energy by modulation processes. One object of my invention is to provide a plurality of bands of energy having the same frequency components rotated uniformly in relative phase. Another object of the invention is to reproduce a given band of frequencies, such as a band of voice frequencies, with the component frequencies rotated in phase in a predetermined manner. Other objects of the invention are to utilize such bands of frequencies for signalling purposes.

bands provides for the transmission of a carrier wave and a single side band with automatic neutralization of the other side band. The economy of such transmission is well known. Another method of utilizing rotated frequency bands permits the correction of phase distortion such as occurs in telephony.

Still other objects and advantages of my invention will be apparent from the specification.

The features of novelty which I believe to be characteristic of my invention are set forth with particularity in the appended claims. My invention itself, however, both as to its fundamental principles and as to its particular embodiments, will best be understood by reference to the specification and accompanying drawings, in which:

Figure 1 shows diagrammatically a simple form of circuit for the reproduction of a predetermined frequency band rotated uniformly in phase.

Figure 2 shows a similar circuit utilizing three element vacuum tubes for accomplishing the purposes of Fig. 1.

- Figure 3 shows circuits arranged to provide several new frequency bands identical to a given band in frequency components but uniformly differing in relative phase in a predetermined manner, or to provide for the single reproduction of the given band having certain of the frequency components rotated in phase in one predetermined manner and other frequency components rotated in some other predetermined manner.

Figure 4 illustrates circuits providing for the use of frequency rotated bands generated as in Fig. 3, together with a new carrier wave source and vacuum tube modulators to give a modulated wave having a single side band for communication.

Figure 5 illustrates the method of producing corresponding new bands of frequencies alike in One method of using such rotated frequency their component frequencies but differing in phase, and the reproduction from these bands of the original band shifted in phase uniformly as desired.

Figure 6 illustrates circuits in part as in Fig. 5 and in part as in Fig. 4 to again produce new side bands of an original band and so controlled as to permit transmission of a single side band with automatic suppression of the other side band.

The advantages of this invention may be obtained by modulation processes, for example, al though it will be understood that other specific ways may be used if desired, and that term is used herein to include any terms defining the process and result of applying electrical energy to a distorting or nonlinear conducting device such as the two or three element vacuum tube, saturated iron core inductances or transformers, crystal and other rectifiers and the like. Such actions as modulation, demodulation, rectification, detection, heterodyne action, harmonic generation, etc., are examples of processes which I will hereafter refer to simply as modulation processes or modulation.

In the drawings, where it has been desired to indicate a particular type of modulator, the three element vacuum tube has been shown, since advantages of amplification and simplicity of circuit arrangement are thus obtained. It should be borne in mind, however, that other types of modulators can be substituted and that the distorting action is due to a nonlinear relation between the applied voltage and resultant current of the plate-filament or grid-filament circuit, either of which may be used as desired. My invention, however, is described as applying to generalized circuits, for purposes of simplicity.

It will be shown in mathematical terms that the applicationof energy of a single frequency and energy of a band of frequencies to a modulator results in new bands of frequencies, commonly termed side bands. It will be shown also that performing this same operation on a number of modulators simultaneously, applying to each modulator the same band of frequencies and the same single frequency in different phase, results in the production of corresponding side bands having a uniform relative phase displacement between the component frequencies, which corresponding side bands can be selected and utilized for transmission purposes. It will also be shown that from a single side band, produced by modulation of a given frequency band, the original band can be reproduced, with its frequency 1 10 components differing in phase uniformly from the corresponding components in the original band, by a second modulation.

Arrangements and methods for rotating the phase of alternating currents or voltages of a single frequency by reactances and coupled circuits, etc., are well known in the art and are not explained in detail herein. This invention, in its fundamental aspect, utilizes the rotation of a single frequency together with modulation to accomplish the uniform relative phase rotation of a band of frequencies. In the demonstrations which follow only second order modulation is dis-. cussed. It should be remembered, however, that the higher orders of modulation can equally well be used to accomplish the desired results by utilizing the proper carrier in its proper phase in the several modulation processes.

Considering now Fig. 1, G1 indicates a generator of alternating current energy of a single frequency, and its voltage as applied to the modulator M1 may be represented as 1=E1 cos wit. G2 indicates a generator of alternating current energy of a band of frequencies, such as a microphone generating a band of voice frequencies, and its voltage as applied to the same modulator M1 may be represented as e2:E2 cos (w2t+0) Thus But one frequency of the band f2 from generator G2 is indicated, since the results obtained for this single frequency apply to any and all of the component frequencies which might have been selected for the demonstration. The phase angle 0 is added to make the case more general.

In the output of M1 is a filter F1 which may be regarded as a device which passes a predetermined band of frequencies to the exclusion of others or as an output load. for the modulator M1 whereby certain component frequencies of the modulation process are developed and selected to the exclusion of others. F1 may therefore be an audio transformer, tuned circuit, band pass filter or the like, depending upon the requirements of a given case.

Considering now the modulator M1 we can write as the relation of input to output voltage,

cm is the output voltage from the modulator M1 and e=e1+e2, a0, a1, 112, as, are constants depending upon the structure of M1 and its adjustment. Since we are interested in the frequency components and their phase due to modulation we can omit these constants which relate to the magnitude of voltages only. Further terms of higher powers than (we can be dropped since they can be made relatively small, and components due to their presence are mostly of frequencies which would be excluded by F1. (It can be noted, however, that the higher terms of even power produce components of second order modulation useful in the results sought.) We therefore use for the characteristic of the modulator 1e.111=e+e Substituting in this equation we ob- The first two terms indicate that the impressed voltages are repeated. The third and fourth terms expand into the double frequency (second harmonic) of the frequencies impressed on the input, plus terms indicating direct current. The

product term is of interest and expands as follows,

These last terms are the sum frequency and difference frequency of the impressed frequencies and are termed the side frequencies, or where (02 is a band of frequencies they become the side bands. In Fig. l the filter F1 selects or develops one of these side bands to the exclusion of all other components, and delivers voltage of the side band frequency to the second modulator M2. In the output of G1 is a phase rotator P to control the phase of energy from G1 delivered to M2. Letting c be the voltage applied to the input of M2, and using in this case the upper side band, we have where (1) is the phase angle determined by the adjustment of P. For modulator M2 the same considerations hold as for M1, and we can therefore use for its characteristic Substituting in this equation the value of e, we again obtain the impressed voltages repeated, the second harmonics of these impressed voltages and a product term 'which is,

2E1 E2 COS (wit-hp) COS ([wi-l-wsli-l-B) which expands into The last term written is seen to be of the same frequency as originally supplied by G2 but shifted in phase by the angle Had the lower side band been supplied to M2 instead of the upper one as we assumed, then our result would have been which is also the original frequency from G2 rotated by the phase angle +4 If we had supplied to M2 both side bands simultaneously, then both of these terms would have been produced, in which case we would have which is seen to be the original frequency from G2 imchanged in phase, cos 4) merely determining the amplitude.

F2 is a filter or load for M2 designed to develop and select the band of frequencies generated by G2. Its output is therefore a new source of the frequency band from G2 of any phase as predetermined by the adjustment of P.

It is clear from the foregoing that either one, but not both, of the side bands produced by M1 may be used in reforming the original frequency band in the desired phase. It is also clear that for a fixed adjustment of P any and all frequencies generated by G2 are rotated by the same phase angle 4:, since f2 may be any frequency com ing from G2. The output from F2 is thus composed of frequencies having a relative uniform phase difference from those supplied by G2. It should be noted that the output from G1 is constant in frequency and in amplitude. Hence E1 and also E1 are constants. The output from G2 is a band of frequencies as occur in telephony, for example, and hence E2 may be different for each frequency component of the original band. Since our equation shows E2 to be linear (of the first power) it is clear that the reformed band is directly proportional to the original band and has suffered no amplitude distortion.

Figure 2 performs the same function as Fig. 1 and in the same manner. The forms of modulators illustrated are, however, three element vacuum tubes. M1 is shown to consist of two such tubes arranged in a form termed a balanced modulator. The tubes are alike and the circuits symmetrical to cause suppression of the carrier wave supplied from G1. The signalling or modulating wave from G2 is applied in differential or opposite phase to the input grids. One of the side bands generated by M1 is selected by F1 (the tuned circuit) to the exclusion of the other. This side band energy, together with energy from G1 of phase determined by the phase rotator P is applied to the input of modulator M2, which is here shown as a three element vacuum tube. F2 is the output load or filter for developing energy of frequencies generated by G2 and is therefore a new source of frequencies as generated by G2 of phase determined by P.

Figure 3 is like Figure 2 up to the output of F1, except that source G2 is shown as an incoming telephone line, terminated by a transformer feeding an amplifier. The side band developed by F1 is delivered to the input of several modulators M2, M3, and M4. The energy of carrier frequency from G1 is also applied to the several modulators but in different phase in each case as predetermined by the setting of P2, P3 and P4. It follows that if F2, F2 and F1 are alike in developing (i. e., acting as a load for) frequencies generated by G2, then we arrive at three additional sources of the frequency band coming from G2, and the new bands have uniform relative phase displacement of the component frequenciesfrom each other and from those of G2. If however the filters F1, F2 and F3 are different from each other in that each selects a portion of the frequencies generated originally at G2, and if together their output gives all of this band of frequencies, then the output of these filters together constitute a new source of the desired frequency band in which certain portions of the band have one uniform relative phase relation to the original band, and other portions have different uniform relative phase difference from the original band. Such a new source is useful for the correction of phase distortion as it occurs in telephony due to the telephone line impedance differing for the different voice frequencies, causing the relative phase of frequencies as generated and as received at the end of the telephone circuit to differ. The circuit of Fig. 3 would therefore give correction of this distortion by the phase readjustment in groups of portions of the band that had undergone the same relative phase distortion.

Figure 4 is like Fig. 3 up to the output of filters F2 and F2, since these filters evidently deliver the band of frequencies generated by G2 of uniform relative phase determined by P2 and P3. (F2 and F3 are here assumed to be identical filters for the band from G2. M4 and M5 are three element vacuum tubes, each acting as modulators in the balanced arrangement shown. They are assumed to be alike in structure, in their characteristics and in adjustment.) The output of F2 is applied to M4, and the output of F3 is applied to Ms. A new source of carrier frequency G3 is provided, having in its output circuit the phase rotators P4 and P5. The output of phase rotator P4 supplies the input of M4, and the output of P5 supplies the input of M5. The output of modulators M4 and M5 is shown supplying an antenna circuit for radiating the high frequency energy. The arrangement of output circuit shown is one commonly termed subtractive. By that it is meant that the output of M4 and M5 subtract algebraically in the antenna. By a different poling of the output inductances the output energy may be made to add.

Let us now consider the action of these modulators. Again, for our demonstration, we consider the case to be one of second order modulation, and being here interested in the component frequencies and their phase, and not in their amplitude as determined by the tube structures and their adjustment, we write for the output voltage of the modulators the relations E2 cos wst we can write for the output from P4 and P5 respectively p4=E3 COS (wst-i-X) where X is phase angle due to P4 6p5 E3 cos (wst-l-X) where X is phase angle due to P5. The output of M4M5 is adjusted to develop .in the antenna circuit energy of the frequency of the new carrier wave Let the adjustment ofthe phase rotators P2, P3, P4 and P5 be such that I Then the outputs from M1 and M5 become respectively Subtracting we obtain as the voltage form of the energy in our antenna and we note that the lower side band cancels out, leaving the carrier and the upper side band of double amplitude.

By making our output additive the result would be one of addition and be In each case we have the carrier and a single side band. Our choice of phasing was such as to make the relative phase of the carrier wave, as applied to the modulators, ninety electrical degrees. Also we made the modulating or signalling frequencies as applied to the two modulators differ by ninety electrical degrees. Such rel ative phase conditions were shown to cause the cancellation of one side band and thus avoid the requirement of a filter separation. By the further choice of the phasing of the modulating or signalling wave to have it differ by forty-five electrical degrees from the original wave generated at G2, we have predetermined conditions so that ordinary detection (second order modulation) of the radiated wave will reproduce the original frequencies from G2, all in their original phase relation.

It should be noted that the advantages of automatic suppression of one side band are realized at high carrier frequencies where the filter separation useful with low frequency carriers is ineffective. The reason for this lies in the fact that with high frequency carriers the modulating frequency side bands usually differ by a small percentage from the carrier frequency. Especially is this true when the desired side bands are voice frequency side bands and it is desired to transmit a single side band together with the carrier. The carrier and single side band is a most desirable type of transmission since it occupies one-half of the frequency channel required to transmit both side bands. It requires no unusual type of receiving equipment and is inherently the required form of wave to give the original signals free from distortion.

Proceeding now to Fig. 5, we have a somewhat different arrangement of modulators operating upon a similar theory, and for our example they will be considered as second order modulators. Here G2 is again the source of a band of frequencies E2 cos mat and G1 is again the source of a carrier wave E1 cos mt. Two modulators M1 and M2 are supplied directly from G2 and indirectly from G1 through phase rotators P1, P2. The filters F1 and F2, if designed for the selection of one of the second order side bands (for example, the upper side band) develop these side bands in their outputs, viz:

where 1 is the phase angle determined by P1 and 2 is the phase angle determined by P2. It is clear that from the original band of frequencies we have produced in the output circuits of F1 and F2 two new sources of frequency bands. These new bands differ from the original band generated at G2 but are identical to each other in frequency components, and these frequency components in the two new bands have a uniform relative phase displacement predetermined by the settings of P1 and P2. These new side bands thus generated are applied to the modulators M3 and M4 together with energy of the carrier wave from G1 unaltered in phase. The output circuits of the modulators contain the filters F3 and F4 responsive to the original frequency band generated by G2 (which is to say that they are responsive to the lower side bands from second order modulation of the input frequencies). Therefore they develop the voltages which are seen to be of the same frequency as the original frequency (15-1 fg=2fl but of relative phase q 1 and 2 as predetermined by P1 and P2. Thus the system of Fig. 5 can be used to reproduce an original frequency band in uniform relative phase as desired, though with more apparatus required than in the system of Fig. 1, Fig. 2 and Fig. 3.

Figure 6 illustrates that half of Fig. 5 to the left of the line a: :0 used as a source of frequency bands to supply modulators M3 and M4. A new carrier source G3 also supplies each modulator M3, M4 through phase rotators P3 and P4 with voltage of the carrier frequency E3 cos wst rotated in phase in an amount z and 4, predetermined by P3 and P4. The same demonstration which, when applied to Fig. 4, showed that a carrier and single side band could be formed with cancellation of the other side band, applies here and need not be repeated. Depending on the output transformer connections the circuit is made additive (or it may be made subtractive) and the modulated wave form in the output will be causes one side band (in this case the upper) to balance out, thereby leaving the carrier is and side band f3(f1+f2) alone to be radiated. A subtractive output would have left I; and 13+ (f1-l-f2) to be radiated.

It should be noted that the side band frequencies in this case may be made of a frequency range well separated from the carrier through a choice of frequency at G1 of much higher frequency than the band of frequencies generated at G2. The use of such a wave is advantageous when the final carrier frequency from G3 is a very high frequency, since it permits the separation of the side band from the high frequency carrier by filtering when only the side band is to be transmitted. Also if the carrier and widely separated side band is to be radiated to a receiving station using a double detection receiver, the circuit of Fig. 6 will be advantageous.

While I have shown and described certain preferred embodiments of my invention, it will be understood that modifications and changes may be made without departing from the spirit and scope of my invention, as will be understood by those skilled in the art. It is to be understood that suitable amplifiers, attenuators, biasing batteri s, transformers, etc. would in practice be inserted between the various operat ing parts shown in the drawings as required to bring the alternating current energy to proper levels for efiicient functioning of the parts.

I claim:-

1. The method of controlling the phase of a band of frequencies which comprises modulating said band with a carrier wave, selecting a product of said modulation, separately modulating said product by a plurality of waves resulting from shifting said carrier Wave through differing amounts, and selecting the desired products of the last mentioned modulations.

2. In high frequency apparatus, in combination, means for producing a band of frequencies and a carrier frequency, means for modulating said carrier with said band, means associated with the output of said modulating means for selecting one of the side bands generated by the process of modulation, a plurality of means for altering the phase of the carrier frequency to the extent desired, a plurality of modulating means each supplied with the side band selected from the output of the first modulating means, and with carrier frequency altered the desired amount in phase, and means associated with the output of said last modulating means for selecting the original band of frequencies altered in phase.

3. In signalling apparatus, means for producing a band of frequencies and a carrier frequency, a modulator, means for supplying said band and said carrier to said modulator, means associated with said modulator to select one of the side bands generated, a plurality of means for shifting the phase of the carrier, a plurality of additional modulators to which the selected side band and the shifted phase carrier are applied, means associated with each of said second modulators to select the desired components of modulation, a double balanced modulator, means for producing a second carrier frequency, means for shifting the phase of said second carrier, and an output circuit for said balanced modulator so arranged as to develop the second carrier frequency and a single side band.

4. The method of producing a two phase audio frequency current which comprises producing a two phase intermediate frequency current, modulating one of said phases of said intermediate frequency by the audio frequency current, the phase of which it is desired to split, selecting one of the side bands thus produced, detecting said side band with each of the phases of said intermediate frequency current, and thus producing a two phase audio frequency current.

5. The method of producing a two phase audio frequency current which comprises producing an audio current representative of intelligence to be transmitted, generating an intermediate frequency current, push-pull modulating said intermediate frequency current by said audio frequency current, selecting one of the side bands produced by said modulation, splitting the phase of said intermediate frequency current, and demodulating the selected side band by combining with each of the phases of said intermediate frequency current, and thus producing a two phase audio frequency current.

6. Means for the production of two phase audio frequency current which comprises a source of intermediate frequency energy, a source of audio frequency energy, means for modulating said intermediate frequency energy by said audio frequency energy, means for eliminating from the result of said modulation all except one of the side bands thus produced, means for detecting the audio frequency existing in said side band in combination with said intermediate frequency, and means for independently detecting said audio frequency existing in said side band by said intermediate frequency energy at phase quadrature to that used in said first mentioned detection, thus producing two quadrature related audio frequency components of said side band.

7. The method of producing a two phase audio frequency current which comprises producing an audio frequency current representative of intelligence to be translated, generating an intermediate frequency current, push-pull modulating said intermediate frequency current by said audio frequency current, selecting one of the side bands produced by said modulation, splitting the phase of said intermediate frequency current, and de- 11 modulating by the plate rectification method the selected side band by combining with each of the phases of said intermediate frequency current as thus produced, and thus producing a two phase audio frequency current.

HAROLD M. LEWIS.

Images