US 2273110 A
Abstract available in
Claims available in
Description (OCR text may contain errors)
Feb. 17, 1942. c. N. KIMBALL. ETAL.
FREQUENCY MODULATED WAVE RECEIVER Filed June 14, 1940 Patented F eb. 17, 1942 FREQUENCY ivronntarnn WAVE RECEIVER Charles N. Kimball, East Orange, N. J., and George C. Sziklai, Bloomington, Ind., assignors to Radio Corporation of America, a corporation of Delaware application .time 1i, i940, serial No. 340,550
Our present invention relatesto reception of frequency modulated (F. M.) waves, and more for, employing degenerative feedback in an F. M. particularly to a novel method of, and means receiver.
F. M. reception, as is well known, is restricted at the present time to frequencies of the order of l-50 megacycles (ma), a maximum frequency deviation of the carrier of i100 kilocycles (kc.) being permitted. There are many instances When it is desired to receive amplitude modulated waves in the standard broadcast band of 550-1'790 kc. Receivers which are able to receive the amplitude modulated signals in the standard broadcast range, as well as the F. M. signals in the short Wave range, would normally have to embody two distinct receiving systems. This is particularly so since in the standard broadcast range, and in the case of superheterodyne reception, the intermediate frequency (I. F.) band width is approximately l0 kc. For reception of both F. M. and amplitude modulated signals in the standard broadcast range it would be highly desirable to utilize the same I. F. networks, while restricting the change-over switching operations to the networks prior to the rst detector and to the discriminator network.
Again, even where receiving solely F. M. Waves, it may be desirable to receive F. M. waves not only of a 100 kc. frequency deviation, but also waves having a frequency deviation of the order of -80 kc. In such cases as well it would be highly desirable to utilize common I. F. networks for both types of frequency deviation, and to restrict the switching operation to the networks preceding the rst detector. For example, in the last case, it would be most desirable to change over from signal frequency networks having a band width of say 50 kc. to 200 kc., and yet utilize the succeeding I. F. networks which would retain the 50 kc. band width.
Accordingly, it may be stated to be one of the main objects of our present invention to provide a superheterodyne receiver for F. M. waves utilizing I. F. networks of a fixed band width value, but Whose pre-rst detector networks are adapted to be of a substantially greater band width, or if desired, of the same band width asthe I. F. networks; there being employed an audio degenerative feedback in the receiving system when the pre-first detector networks are of a substantially greater band width.
Another important object of our present in-l vention is to provide an F. M. superheterodyne receiver whose I. F. networks' are of a substan- (Cl. Z50-20) tially smaller band width than the pre-first detector networks, wherein a portion of the audio output voltage of the second detector is utilized to produce frequency modulation oftheflocal oscillator output waves, and the latter frequency modulation being heterodyned so that its envelope of frequency variation is in phase with that of the received F. M. sig-naL'so that the resultant I. F. frequency deviation is reduced to the difference of the frequency deviations 0f the F. M. wave and the frequency-modulated local oscillations.
AStill another object of this-invention is to provide an audio degenerative feedback in a superlieterodyne receiver of the F. M. type, wherein as a result of sufficient degenerative feedback to the local oscillator the output of the second detector is substantially independent of the receiver input signal amplitude, thereby resulting in Asuppression of noise and other undesired forms of amplitude modulation, and also effecting a reduction in the amplitude distortionproduced by non-uniformity in the I. F. response and by discriminator non-linearity over the reduced I. F. band.
In a system of the'type described above, wherein audio degenerative feedback to the local oscillator is kemployed for reducing the frequency deviation of the applied F. M. wave, there results an appreciable time delay in the system, which causes a phase difference between the frequency deviation envelope appearing at the antenna terminals and that resulting from modulatingA the local oscillatorfrequency. This phase difference is negligibly small for low audio frequencies, and increases with audio frequency until at some audio frequency the phase delay between the antenna input frequency deviation envelope and the envelopeof the local oscillator is degrees at which point neither degeneration or regeneration exists. If this variation of phase delay with audio modulating frequency is permitted to eX- ist, it will result in a variation of effective I. F. band width with audio frequency. Itis theoretically possible to compensate for vthe phase delay encountered in the I. F. circuits by inserting corrective networks in the feedback path which result in a phase advance for audio frequencies which nullies the phase delay in the I. F. networks. It is, also, possible to increase the oscillation frequency deviation greatly vat high audiofrequencies thereby to produce an essential in-phase degenerative frequencydeviation component to beat with the input signal.
Accordingly, it may be stated to be another object of this invention to provide phase advancing networks in the degenerative feedback path so as to compensate for the phase delay caused by the resonant I. F. networks of the system.
Still other objects of this invention are to improve generally the efiiciency of superheterodyne receivers adapted for F. M. waves, and more especially to provide receivers of the latter type which are adapted for flexible operation, and are economically manufactured and assembled.
The novel features which we believe to'be characteristic of our invention are set forth in particularity in the appended claims; the invention itself, however, as to both its organization and method of operation will best be understood by reference to the following description taken in connection with the drawing in which we have indicated diagrammatically a circuit organization whereby our invention may be carried into effect.
In the drawing:
Fig. 1 shows an F. M. receiving system embodying the invention,
Fig. 2 shows graphically the second detector characteristic.
Referring now to the accompanying drawing, and particularly to Fig. 1, there is shown an F. M. receiver which is of the superheterodyne type. The conventional networks of this receiver are schematically represented, since those skilled in the art at the present time are fully aware of the manner of constructing such a receiver. The signal collector I may be a di-pole, a grounded antenna circuit, or even a radio frequency distribution line. It is assumed that the collector I collects F. M. signals in a range of approximately 42 to 50 mc., although it is to be clearly understood that the present invention is not restricted in any way to that signal frequency range. The collected signals are fed to a tunable radio frequency amplifier 2 which may embody one or more tunable stages of radio frequency amplification. The numerals 3 and4 designate the coupled tunable signal circuits feeding the amplifier 2, and it is to be understood that these tunable circuits have a band width which is equal at least to twice the maximum frequency deviation of the F. M. wave. By way of illustration, it is assumed that the band width is 200 kc., since F. M. waves may have a permissible maximum frequency deviation of i100 kc. The amplified signals are then transmitted to a first detector network 5 through a second tunable band pass network which comprises the coupled circuit 6 and l. This latter band pass network, also, has a band width of 200 kc.
The locally produced oscillations which are heterodyned with the signal energy to provide the I. F. energy, are produced by a local oscillator which comprises the tube 8 having a tunable tank circuit 9. The cathode of the local oscillator tube is returned to a tap on the tank coil, and the local oscillator itself is generally conventional in construction. It will be understood that the tunable tank circuit 9 is varied through a frequency range which is generally higher than the signal frequency range by the I. F. value at all corresponding points of the two ranges.
For the purposes of this application the I. F. value is assumed to be 456 kc., which is the magnitude of the I. F. generally employed for receiving signals in the standard broadcast range of 55C-1700 kc. The letter M denotes the coupling between the first detector circuit 5 and the local oscillator. It is to be understood that the local oscillations may be injected into the first detector circuit in any desired and well known manner. The numeral l0 designates the usual mechanical uni-control mechanism which adjusts the tuning elements of the oscillator tank circuit 9 and the various signal circuits 3, 4, 6 and 1, and this control device is represented by the dotted lines I0.
'Ihe numeral II designates the resonant output circuit of the rst detector, and, as stated previously, this circuit is assumed to be tuned to a center I. F. value of 456 kc. This assumed value is, of course, the case where it is desired to employ the same I. F. networks for receiving the amplitude modulated waves in the standard broadcast range. In this case the various signal networks and the local oscillator tank circuit would be adjusted so as to receive standard broadcast signals. Where, however, solely F. M. waves are to be received, and it may be desired to receive F. M. waves having a frequency deviation of kc. or 50 kc., then it would only be necessary to vary the band width of the coupled circuits 3-4 and 6 1. In this latter case, the operating I. F. value may be of the order of 3 mc., and, therefore, the oscillator frequency range would be appropriately chosen to provide such an I. F. value. However, for the ypurposes of the present application let it be assumed that the receiving system, as shown in Fig. 1, is the result of adjustment to impress F. M. waves having a frequency deviation of |100 kc. on a superheterodyne receiver whose I. F. networks are of a band width of 50 kc. and wherein the I. F. value is 456 kc.
'I'he I. F. amplifier tube I2 is of the usual construction, and has its signal grid coupled to the high potential side of the resonant input circuit I3, the latter being tuned to the operating I. F. value. Circuits II and I3 are reactively coupled so as to provide an effective band width of 50 kc. The low potential side of input circuit I3 may be connected to the grounded end of the grid bias resistor I4. A common source of positive potential may be provided for the screen grid and the plate of tube I2. The numeral I5 designates a resistive load arranged in the plate circuit of the I. F. amplifier.
An aperiodic amplifier follows the tuned I. F. amplifier I2. The aperiodic amplifier comprises a tube I6 whose signal grid is coupled to the plate end of resistor I5 through a direct current blocking condenser Il. The signal grid is connected to the grounded end of the grid bias resistor I8 by the grid leak resistor I9. A common source of positive potential supplies both the screen grid and plate I6, and the numeral 2U designates the resistive load in the plate circuit of tube I6. Due to the fact that in a 50 kc. double-tuned bandpass stage the time delay, assuming an phase shift over the band, would be equal to, or larger than, l0 microseconds, bandpass couplings are used only in two stages. network is used between the first detector and the first I. F. amplifier, and the second bandpass coupling network is used between the limiter tube and the second detector tube. Between the other I. F. stages resistance couplings are utilized.
The capacities and resistors for these resistance-coupled I. F. stages are chosen to produce a very broad characteristic with maximum gain at approximately 456 kc. Tubes of the GSK? type may be used in these apericdic stages, and it is to be understoodl that more than one aperiodic amplifier stage may be employed. A gain of A bandpass coupling approximately Sli-.per stage can be obtained with a time delay of less than one microsecond per stage. The total time delay of the I. F. amplifier network, measured between the first detector and the second detector input, may be approximately 22 microseconds.4 In other words, inherent time delay occurring in the I. F. network may be minimized by employing only as many resonant I. F. circuits as may be required for selectivity purposes, while the remaining I. F. gain is obtained with resistance-coupled circuits, or with other low time delay I. F. stages.
Prior to detection of the F. M. signals, the latter are passed through a limiter stage. The limiter employed in the receiving system utilizes a tube of the SSK'I type, and which tube is designated by the numeral 2 l. The signal grid of the tube is coupled to the resistor element by the direct current blocking condenser 22, and the signal grid is connected to the ground by the grid leak resistor 24. A common source of positive potential is connected to the screen grid and plate of the limiter tube. The resonant primary circuit of the second I. F. bandpass network is disposed in the plate circuit of tube 2|.
The second bandpass network comprises the primary circuit 215 and the secondary circuit 25. The symbol M1 denotes the reactive coupling between these circuits which provides a bandpass network with a band width of kc. Each of the primary and secondary circuits 25 and 2,6 is, of course, resonated to the I. F. Value, and the high potential side of circuit 25 is connected to the midpoint of the coil of circuit 26 through the condenser 21.
The constants of limiter 2| are so chosen that the tube functionsto limit any amplitude Variations in the F. M. signals. The grid condenser 22 and resistor 24 are so chosen as to permit developed grid bias to follow the variations in F. M. grid signal amplitude. For example, va time constant of 5 microseconds could be used for 22-24, since such value would permit following any audible audio frequency.
The limited I. F. signals are impressed upon The resultant voltage across resistorv 3| is ernployed as the audio modulation voltage, and is transmitted to one or more audio amplifier networks.
The discriminator network 2'5--26 has a peak separation of 50 kc. and slope of 0.3 volt per kc.
maximum. The output of the discriminator is detector network shown in Fig. 1 has been disclosed and claimed by S. W. Seeley in U. S. P.
2,121,103, granted June 21, 1938. Particular reference is made to Fig. 5 of the said Seeley Ypatent which shows a second detector network of the present type. From the ungrounded end of the output resistor 3| of the detector there is derived the audio voltage which corresponds to the frequency deviations of the F. M. wave. It is not believed necessary to recapitulate in the present description the functioning of the second detector network, since those skilled in the art are fully aware of the manner in which such a network functions. It is believed sufficient to point out that the frequency detector shown has the characteristic presented in Fig. 2. The discriminator characteristic slope direction depends on many factors. It is important, however, that it nally be such that the phase of audio feedback Voltage causes the frequency deviation envelope of the oscillator to be in phase with that of the F. M. signals. From this characteristic it is seen that as the frequency of the F. M. wave swings to and fro with respect to the center frequency (fc) of the I. F. band, there flows through the output resistor a uni-directional current whose-magnitude and polarity depends upon the magnitude and direction of frequency deviation.
' denser 3l at l5 kc.
fed through a. two-section, low-pass filter 32 which has a cut-off frequency of 200 fkc. and a time delay of approximately 3 microseconds. The audio output of the filter 32 passes to the audio utilization network. It is also fed through the degenerative audio feedback 'path 33. The reason for the selection of a cut-olf frequency for lter 32 which is as high as 200 kc., is the desirability of having a small time delay. The lower the cut-oif frequency of a filter, the higher is the time delay. Hence, a Value is chosen which minimizes time delay and is consistent with adequate filtering of the I. F.
The audio voltage fed through the feedback path 33 is impressed upon a tube 34 which is included in a phase correction stage. This phase correction stage provides a phase shift approximately equal in magnitude but opposite in phase to that existing between vthe input signal and the signal in the feedback path. This phase advance is accomplished by arranging an inductance 35 and resistance v315 in series in the plate circuit of tube 34. The ratio of the inductance of coil 35 to the resistance of resistor 36 is chosen to be equal to 28 microseconds. In order to cut down the gain of this correction stage for high frequency, the inductance 35 is resonated by coninput signal grid coupled to the line33 through condenser 3'8, while the grid is also connected to the grounded end of the bias resistor 39 through the grid leak resistor 40. The plate of the tube S4 is connected to a source of positive potential by a resistor 4| of approximately 10,000 ohms, and the direct current blocking condenser-G2r connects the upper end of coil 35 to the plate of tube 34. On a steady state basis the stage 34 should compensate for phase delay caused in the I. F. networks, the discriminator and in the audio filter network.
The output signal energy of the correction stage is applied to the input grid of an electronic reactance tube 50. The function of this tube is to provide across the tank circuit 9 of the local oscillator a reactive effect which will vary the frequency of the oscillator circuit. The reactance tube circuit appears as a negative inductance across the tank circuit 9. This is accomplished by connecting the plate of the oscillator tube 8 to ground through a path which includes the series condensers 5I and 52.` 'Ihe cathode of tube 50 is connected to ground through a bias resistor 53, while the grid 54 is connected to ground through the grid leak resistor 55. The grid-54 is connected to the junction of condensers 5I-52, and hence alternating voltage dveloped across condenser 52 is impressed on grid 54.- 'I'he input grid of tube 50 is denoted by the numeral 55'., and the latter may be surroundedby a positive screening eld. The plate 60 of tube 50' is connected to a source of positive potential through a radio frequency choke coil, and the plate is also coupled to the high potential side of tank circuit 9 through the condenser 70. The tube 50 may be of 6L? type, while the oscillator tube 0 may be of 1852 type.
It can be'readily demonstrated that tube 50 appears asanegative inductance acrosstankcir- The correction stage has its oscillator frequency at an audio rate.
inductance is inversely proportional to the gm (mutual conductance) of tube 50.
When the audio voltage on grid 55 of tube 5U swings toward the positive direction the reflected negative inductance across tank circuit 9 decreases. Since the reflected inductance is in shunt with the positive inductance of tank circuit 8, it reduces the total inductance in the circuit and the oscillator frequency decreases.
Thus, for example, if the audio frequency voltage applied to grid 55' is of 400 cycles, the effective inductancemegative) produced by the tube 50 changes amplitude in accordance with instantaneous amplitude changes in grid 55' four hundred time a second. The resultant frequency change in the oscillator may be readily computed from a knowledge of the effective L and C of the tank circuit. Thus, if the mutual conductance of tube 50 increases simultaneously on a positive audio modulation peak, the effective L of the tank circuit increases and the frequency decreases.
Considering the action more specifically, across condenser 52 there is developed a control voltage which is in quadrature with the voltage across the tank circuit 9. This quadrature voltage is impressed on grid 54, and tube 50 feeds back into the tank circuit 9 through condenser 10 alternating current which is in quadrature with the tank circuit voltage. The grid 55' is varied at the rate of the audio voltage impressed on it through the audio feedback path, and, therefore, the simulated reactive effect across tank circuit 9 is varied at the rate determined by the fed back audio voltage. That is, the quadrature current fiowing into the tank 9 is varied in amplitude at an audio rate, and changes the As a result of this arrangement the local oscillations which are fed to the first detector are frequency modulated. 'Ihe frequency deviation provided is 150 kc. total, or l75 kc. on either side of the unmodulated oscillator frequency. The frequency modulated local oscillations which are fed into the rst detector are arranged so that its frequency deviation envelope is in phase with that of the F. M. signals impressed on the first detector at circuit 1. Hence, when the two signals beat in the first detector the resultant total I. F. peak to peak frequency deviation is reduced to the difference of 200 kc. and 100 kc.
As a result there is developed across the I. F. output circuit Il F. M. waves which are at the I. VF. value, but whose frequency deviation has a maximum of $25 kc. In other words the maximum frequency deviation of the resulting I. F. signals is half the band width of the I. F. networks. The frequency modulated oscillator voltage is injected into the first detector in any desired manner.
It will now be observed that the phase correction stage 36 counteracts the infiuence of the time, or phase, delay in the I. F. circuits and the phase lag between the envelopes of the frequency deviation inthe input signal and in the heterodyned oscillator output voltage. The correction stage functions to advance the phase in the feedback path between the audio detector andthe oscillator reactance control tube. In designing the phase advancing stage, it is necessary to cut off any feedback at frequencies which produce a net phase shift of 90 degrees or more. This precaution prevents regenerative feedback at such high frequencies. It may be desirable to employ cuit 9, and that the magnitude of such negativeI more than one phase correcting stage prior to the reactance tube 50 in order to compensate for the time delay which is added by utilization of the low pass filter 32. Generally speaking, however, the phase correction stage in the audio feedback path should substantially correct for the time delay occurring up to the audio detector, while having a sharp cut-off characteristic so as to prevent oscillation at the high audio frequencies.
While we have indicated and described a system for carrying our invention into effect, it will be apparent to one skilled in the art that our invention is by no means limited to the particular organization shown and described, but that many modifications may be made without departing from the scope of our invention, as set forth in the appended claims.
What we claim is:
l. In a method of receiving frequency modulated carrier waves, heterodyning collected frequency modulated waves with local oscillations of a frequency different from the frequency of the collected Waves by an operating intermediate frequency, deriving from the intermediate frequency waves modulation component frequency voltage which corresponds to the frequency modulation of the collected Waves, and varying the frequency of the local oscillations with the said modulation voltage thereby to produce frequency modulated local oscillations having a frequency deviation which is differentfrom the frequency deviation of the aforesaid collected Waves, amplifying said intermediate frequency waves whereby a phase delay occurs, and advancing the phase of the aforesaid audio voltage prior to frequency modulation of said local oscillations to an extent sufficient to compensate for said phase delay.
2. In a method of receiving frequency modulated carrier waves, heterodyning collected frequency modulated waves with local oscillations of a frequency different from the frequency of the collected waves by an operating intermediate frequency, deriving from the intermediate frequency waves modulation component frequency voltage which corresponds to the frequency modulation of the collected Waves, and varying the frequency of the local oscillations with the said modulation voltage thereby to produce frequency modulated local oscillations having a frequency deviation which is different from the frequenny deviation of the aforesaid collected waves, and advancing the phase of the aforesaid audio voltage prior to frequency modulation of said local oscillations to an extent sufficient to maintain the envelope of said collected Waves and the envelope of said frequency modulated local oscillations in phase.
3. In a receiving system of the type adapted to receive frequency modulated rcarrier waves, a first detector circuit having an input circuit upon which such waves are impressed, said input circuit having a band width which is relatively wide, a demodulator network, a network coupling said first detector to said demodulator, said coupling network being tuned to an operating intermediate frequency and having a relatively narrower band Width than said first mentioned band Width, a local oscillator network' arranged to produce local oscillations which differ from said impressed frequency modulated waves by said intermediate frequency value, means for controlling the frequency of said local oscillator network, said last means comprising an electronic device connected to said oscillator network to provide an effective reactance control therefor, and a degenerative modulation voltage feedback path between the demodulator output and said electronic device whereby the local oscillations are varied in frequency over a frequency range which diiers from the frequency range of said relatively wide band Width by the value of said relatively narrower band Width, and means for impressing said modulated local oscillations upon said first detector