US2065884A - Tuned radio frequency amplifier - Google Patents

Description

R. A. BRADEN TUNED mmc FREQUENCY AMPLIFIER l Dec; 29, 1936.

`I5 Sheets-She'et l Original Filed May 16, 1923 lNvENToR RENE A. BRADEN BY '7 MM ATroR EY Dec. 29, 1936. R. A. BRADEN TUNED RADIO FREQUENCY AMPLIFIER Original Filed May 16, 1923 3 Sheets-Sheet 2 f5 /zi INVENTOR. RENE A. BBADEN AT'IIORNEY Dec. 29, 1936.

R. A. BRADl-:N

vTUNED RADIO FREQUENCY AMPLIFIER Oiginal Filed May 1e, 1928 3 Sheets-Sheet 5 INVENTOR RENE A. BRADEN/` BY A l )MM ATrRNEY Sii Patented Dec. 29, 1936 UNITED STATES PATENT OFFICE TUNED RADIO FREQUENCY AMPLIFIER Rene A. Braden, Mcrchantville, N. J., assigner to Radio Corporation ci' America, a, corporation of Delaware 6v Claims.

This invention relates to radio signalling apparatus and especially to a tuned radio frequency amplifier for a radio receiver and is a division of application Serial No. 278,105, filed May 16, 1928.

An object of my invention is to make a radio receiver for receiving modulated radio frequency waves which will respond with substantially equal amplification to all of the frequency cornponents contained in the modulation within the limits of a certain band of frequencies.

Other objects as well as advantages of my invention will appear as the explanation and description thereof, which will be given with the aid of the accompanying drawings, proceeds.

In the drawings,

Fig. 1 is a resonance curve of a theoretically perfect` radio frequency amplifier,

Fig. 2 shows resonance curves of an ordinary radio receiver,

Fig. 3 illustrates a circuit with which I shall analyze the operation of my invention,

Figs. 4 and 5 illustrate resonance curves of tuned circuits;

Fig. 6 represents a single stage selective amplifier, suitable for use in a multi-stage receiver to be built according to my invention,

Fig. 7 is a group of resonance curves made by utilizing the circuits shown in Fig. 6,

Fig. 8 shows the eifect on the resonance curve G of Fig. 7 of adding resistance to the circuits shown in Fig. 6,

Figs. 9 to 16 inclusive, show, according to my invention, various methods of coupling tuned circuits between amplifier stages, and

Fig. 17 shows a receiver built according to my invention.

Radio telephone signals are transmitted by means of a carrier current (a high frequency current of constant frequency) whose amplitude is modulated by low frequency currents which are produced by the action of sound waves on a microphone or telephone transmitter. This modulated carrier current can be shown mathematically to be equivalent to a carrier current of constant amplitude and two groups of currents, one having frequencies above the carrier frequency and the other having frequencies below the carrier frequency. A

The differences between the carrier frequency and the various frequencies in each group are equal to the frequencies contained in the original audio frequency modulating current. The two groupsl of currents of higher and lower frequency respectively are generally known as the upper and (Cl. 17E- 44) lower side bands. To receive such a modulated wave, and preserve the relative intensities of the original modulating currents, it is obvious that a receiver will be required which will respond with equal amplification to all frequencies contained in the two side bands. As indicated before, it is an object of this invention to provide such a receiver.

It is well known that speech and music contain frequencies from below 100 cycles to above 5000 cycles per second. High quality reproduction requires that this full range of frequencies be transmitted and reproduced with equal efficiency. In a radio receiver, this means that the amplification must be substantially uniform over a range of at least 5000 cycles above and below the carrier frequency, and at the same time good selectivity of the receiver must be had.

In broadcast transmission it is common practice to set different stations on carrier frequencies l0 kilocycles apart. Each station transmits a modulated wave containing all frequency components between (F-5000) and (F4-5000) where F is the carrier frequency of the station. The receiver must not respond to any frequency more than 5000 cycles away from the carrier frequency to which it is tuned. An ideal receiver for broadcast reception would be one having a relation between frequency and amplification such as shown in Fig. 1.

Such a curve, namely, a frequency vs. amplification curve is generally called a resonance curve. it is common practice to make resonance curves for various receivers and radio frequency amplifiers as a test of their performance, the procedure being to impress on the input terminals of the receiver or amplifier an unmodulated radio frequency voltage of constant amplitude, and measure the amplified voltage at the output while the input frequency only is varied.

Resonance curves for an ordinary receiver using three stages of simple tuned radio frequency amplification are shown in Fig. 2, curve l illustrating the resonance curve made at a carrier or mean wave frequency of 600 kilocycles, f

the range of frequencies being from 580 to 620 kilocycles; and curve B illustrating a resonance curve made at a carrier wave frequency of 1400 kilocycles, the variation being from 1360 kilocycles to 1440 kilocycles.

My invention will be further understood by considering Fig. 3. Numeral l denotes a vacuum tube of the four electrode, screen grid type, 3 and ll are inductance coils, 2 and 5 are variable tuning ccndensers. Coils 3 and 4, and condensers 2 and 5 need not be of the same size, the only requirement being that circuits 2, 3 and 4, 5 are capable of being tuned to the same frequency. 6 is a battery for supplying plate and screen voltage to the vacuum tube, l is a grid battery and 8 is a filament battery. 9 indicates the control electrode and IG the screen grid. The coils 3 and 4 are coupled together inductively.

The operation of the circuit of Fig. 3 may be explained by assuming that a radio frequency voltage is impressed on the input terminals of that figure, that is, on the control grid 9 of the vacuum tube I, and by considering the effect of Varying the frequency of the impressed voltage. The circuit consisting of the coil 3 and the condenser 2 must be tuned to some particular frequency, that is, the resonance frequency. The coil 4 and condenser 5 must be tuned to the same frequency.

Now assume that the input voltage is started at a frequency below resonance and gradually raised to a frequency above resonance. As the frequency -approaches resonance, the current in 2 and 3 will rise. In the absence of the secondary circuit 4 and 5 the current would follow curve A of Fig. 4, reaching the highest point at exact resonance. With the secondary circuit tuned and coupled as described above, the primary current follows the curve B.

The current flowing in the coil 3 induces a voltage in the coil 4 which is equal to the product of the frequency, the mutual inductance between 3 and 4, and the ciurent in 3. As the variation in frequency considered is very small compared with the mean frequency, it is sufficiently accurate to say that the voltage induced in coil 4 is proportional to the current in the coil 3. The induced voltage therefore varies with frequency as curve B of Fig. 4.

The current in the secondary circuit is equal to the product of the induced voltage in 4 multiplied by the admittance of 4 and 5 in series. If 4 and 5 are assumed for convenience to be identical with 3 and 2, respectively, the admittance curve of the secondary circuit will be the same as that of the primary, and will be the same as curve A of Fig. 4. The secondary current can therefore be determined as a function of frequency by multiplying together the two curves, A and B. rIhe result of this operation, multiplied by certain constants depending on the values of inductance and capacity used in the two circuits, gives a curve showing the voltage developed across the secondary coil and condenser when a constant voltage of variable frequency is impressed on the grid of the vacuum tube l. Such a curve for a circuit like that shown in Fig. 3 is given in Fig. 5.

Curves similar to those illustrated were made by utilizing circuit constants as follows: The vacuum tube I was a screen grid tube having an amplication constant of approximately 200 and an internal plate resistance of 500,000 ohms. rIhe coils 3 and 4 were of 225 microhenries inductance, and the ratio of resistance to reactance at the frequency at which the curve was made was .007. The frequency used was '700 kilocycles. The condensers 2 and 5 were variable air condensers whose maximum capacity was about 850 micro micro-arads. The coupling between the coils 3 and 4 was 4 microhenries.

I have discovered that the width X of the resonance curve as illustrated in Fig. 5 is controlled by the amount of coupling between the primary and the secondary circuit.

Referring now to Fig. 6, which illustrates a single radio frequency amplifier stage built according to my invention, 3 and 4, as in Fig. 3 are coils inductively coupled together, II and I2 are small non-inductive resistances, I3 and I4 are by-pass condensers connected across the plate and grid batteries I5 and I6, and Il is a coupling condenser. I8 is the filament battery. rIhe purpose of the condenser IT in Fig. 6 is to control the electrostatic coupling between the primary and secondary circuits 2--3 and 4-5.

Fig. 7 shows a group of resonance curves made with a circuit like that in Fig. 6 with the coupling adjusted to various values, curve A being the case of weakest coupling and curve I the case of strongest coupling. In my circuits, close coupling, that is, coupling sufliciently close to cause double humped resonance curves, is utilized.

Fig. 8 shows the effect of adding resistance to each tuned circuit when the coupling is adjusted as in curve G of the preceding figure. Curve G results from the addition of a small amount of resistance, while G was produced by adding a large -amount of resistance.

The circuits 2, 3 and 4, 5 of Fig. 6 are coupled together in one of the ways shown in Figs. 9

to 16 inclusive. In Fig. 9 the two coils 3, 4 are placed close together with their axes parallel, the distance between them being regulated according to the amount of coupling desired. An electrostatic screen I9 is placed between the coils to reduce the electrostatic capacity between the windings.

As shown in Fig. l0, the screen I9 of Fig. 9 takes the form of a slotted plate or comb-like structure.

In Fig. 11 the coils are placed end to end, with an electrostatic screen I9, 20 between and around them.

In Fig. 12 the same arrangement is used, except that the electrostatic screen ls placed differently, there being a single screen 2l surrounding the coils.

In Fig. 13 reversed coupling coils 22, 23 are connected in series with the two main coils 4, 3 respectively there being a single screen 2l surrounding the coils.

In Fig. 14 the two coils 3 and 4 are in separate electrostatic and electromagnetic shields, and are coupled by the coupling coils 24 and 25. The two tuning condensers 2 and 5 are so placed that the end plates of the stators 26, 21 constitute the two plates of a coupling condenser between the coils 4 and 3, the active area of these plates being the portion which is not covered by the rotor plate.

By suitably shaping the end plates of the rotor groups and by placing between the stator groups a shield in which there is a suitably shaped aperture, the coupling capacity between the condensers may be made to have any desired value and to vary in any desired manner as the condenser rotors are turned to tune the receiver t0 different wave frequencies. If desired, it is possible to have the capacitive coupling between the end plates of two rotor groups.

In Fig. l5 both coils and condensers are in shielded compartments 28, 29 and coupling is provided by the coupling coils 24 and 25 and the coupling condenser 30. In Fig. 16 coupling between the two coils 3 and 4 is provided by LlA J the coupling coils 3| and 32, the remaining portions of the tuned circuits being shielded. In Fig. 16 no capacitive coupling is shown but may be added in one of the ways described above.

The shielding is in the cases of Figs. 9, 10, 11, 12 and 13 preferably of the type shown in Fig. 10. However, it may take other forms Such as thin strips of metal pasted or glued on paper and wound about, and if needed, wound within the coils. The shielding as shown should be grounded at convenient points. Where the shielding surrounds both coils and condensers, it may take the form of metallic boxes.

The amount of shielding will depend of course on the amountl of electrostatic coupling as compared to the inductive coupling necessary to give the desired band width.

As shown in Fig. 9, the inductive coupling is such that the voltages induced in coil 4 by the inductive coupling is opposite in direction to that induced by the electrostatic coupling. Such also is the case in Figs. 10, 11 and 12. In Fig. 13 the voltages induced in coil 4 by the inductive coupling between coil 4 'and coil 3 is also opposite in direction to that induced in coil 4 by the electrostatic coupling. To reduce the inductive `coupling between. the coils 3 and 4 which in the structure shown may be too large, reversed coupling coils 22-23 shown as single turns are utilized. These coils have the eiect of reducing the inductive coupling to a desired value.

In Figs. 14 and 15 the direction of the coupling coils 24--25 with relation to coils 3-4 respectively controls the direction of the magnetic coupling which should be chosen such that the voltages induced are opposite in direction to the voltages caused by the electrostatic coupling between the end plates of the stators 26-21 in the case of Fig. 14 and condenser 30 in the case of Fig. 15.

Fig. 1'7 illustrates still another embodiment of my invention, in which the coils are coupled as shown in Fig. 13, while capacity coupling to control the received band width is secured by the method illustrated in Fig. 14.

In this receiver the primary and secondary coils 3, 4 respectively of one stage are wound on a Bakelite tube 6 inches long and 2 inches in diameter. Each coil is approximately 1.6 inches long, and the distance between the adjacent ends of the two coils is 2 inches. The mutual inductance between the coils is reduced to the proper value by a single turn of wire connected in series with the primary coil, placed -136 inch from the nearer end of the secondary coil, and connected so that its coupling to the secondary, is opposite in direction to the coupling of the primary to the secondary coil as shown in Fig. 13. The resulting total mutual inductance is about 1 microhenry. The circuits are tuned by variable condensers 12 having a maximum capacity of about 340 micro microfarads. The electrostatic capacity between the high potential ends of the coils is reduced to a suitable value by a grounded electrostatic shield in the form of a slotted metallic tube 2|, 3 inches in diameter which is placed concentrically over the coils. Electrostatic coupling between the condenser stator plates is adjusted to a suitable value by means of a partial electrostatic screen between the condensers. Each stage is suitably shielded by metallic boxes 13.

What I claim is:

1. A high frequency coupling system comprising an input circuit and an output circuit, each of said circuits including an inductance tuned by a capacity to a frequency which is the carrier frequency to be transmitted, said inductances being located in close proximity to each other, and an electrical conducting shield closely surrounding said inductances whereby the effective coeiiicient of coupling between said coils is of a sufficiently small magnitude so that said system as a whole tunes to the same frequency to which each of said circuits is individually tuned.

2. A high frequency coupling system for transmitting a signal between a pair of vacuum tubes, comprising an inductance in the output of the rst of said tubes and an inductance in the input of the second of said tubes, a capacity shunted across each of said inductances of such value as to cause each of said inductances to be resonant at the same frequency, said inductances being located in close proximity to each other so that acting independently there would exist such a high degree of magnetic coupling therebetween as to produce a double resonance characteristic, and shielding means so situated in relation to said coils that the coupling which would exist between said coils in the vabsence of said means is reduced to the proper effective value to provide a single resonance characteristic.

3. A high frequency coupling system comprising an input circuit and an output circuit, each of said circuits including an inductance element, said inductance elements being placed so close together that there is a strong magnetic field therebetween when said inductances act alone, and means associated with said inductance elements for producing a magnetic iield which is weaker than, and which opposes, said strong magnetic field, whereby the resultant magnetic iield interlinking said inductances is relatively weak as compared with said strong magnetic field.

4. A high frequency transformer comprising a pair of coils, each of said coils being wound in at least one layer, said coils being co-axially situated and having the adjacent ends thereof 'separated from each other by a distance not exceeding the outside diameter of said coils, and a shielding ring circumferentially surrounding said coils, the inside diameter of said ring being less than twice the outside diameter of said coils.

5. A coupling system as defined in claim 1, said inductances being co-axially positioned, and the distance between the adjacent ends of the inductances not exceeding the diameters thereof.

6. A high frequency coupling system comprising an input circuit and an output circuit, each 0f said circuits including an inductance element, said inductance elements being placed so close together that there is a strong magnetic field therebetween when said inductances act alone, and coupling coils associated with said inductance elements for producing a magnetic field which is weaker than, and which opposes said strong magnetic eld, whereby the resultant magnetic ileld interlinking said inductances is relatively weak as compared with said strong magnetic field, and means surrounding said inductance elements and coupling coils for reducing the electrostatic coupling between the said circuits by a desired amount.

RENE A. BRADEN.

Images