US1973037A - Amplification system - Google Patents

Description

Sept. 11, 1934, G. L. BEERS 1,973,037

AuPLIFIcA'rIoN SYSTEM Original Filed Nov. 30, 1926 2 Sheets-Sheet l 5 In B E (t l w U v 2:

6K) 800 I000 I200 I400 I500 WITNESSES: Fre uenc Kiloc des INVENTOR George L. Beers.

Paten .1! Sept. 11, 1934 AMPLIFICATION SYSTEM George L. Beers, Merchantville, N. .12, assignor to Westinghouse Electric & Manufacturing Company, a corporation of Pennsylvania Original application November 30,1926, Serial No. 151,722. .Divided and this application February 1, 1932, Serial No. 590,163

12 Claims. (Ci. 175-359) My invention relates to amplification systems, and it has particular relation to a system designed for voltage amplificatiomatradio frequencies.

This application is a division of my application Serial No. 151,722 filed November 30, 1926 now Patent 1,907,478; May 9, 1933.

One object of my invention is to provide a coupling means for improving the selectivity characteristics of a radio-frequency amplifying system. i

Another object of my invention is to improve the voltage amplification characteristics of a radio-frequency amplifying system.

Another object of my invention is to provide an improved radio-frequency transformer to obtain the above results.

Another object of my invention is to provide a radio-frequency amplifying system that will be reasonably selective over the entire range of frequencies now used in radio broad-casting, and, at the same time, give a substantially equal voltage amplification over the same range.

Anotherobject of my invention is to provide a radio-frequency amplification system that is inherently stable and does not tend to oscillate over the range of frequencies for which it is designed.

Another object of my invention is to provide a coupling transformer the primary of which will have a resonant period substantially outside of the frequency band within which, it is intended to function.

Other objects of my invention will become apparent from the following description of my invention.

In the construction of radio-frequency amplifiers according to the teachings of the Alexanderson Patent No. 1,173,079, it has been found, by experiment, that the inductance of theprimaries of the tuned transformers should be kept low in order to lessen the tendency of the amplifier to go into self-oscillation. It is customary to so proportion the primaries of such tuned transformers that the natural period thereof is somewhat higher than any frequency to which the secondary is intended to be tuned, with the result that, although the amplification at the higher frequencies is satisfactory, the selectivity, by reason of primary resonance and high losses in the secondary, is less good at the high-fre quency end of the tuning range than at the low-frequency end.

On the other hand, a radio-frequency transformer of the usual type, having a low-.induc tance primary, is extremely selective at the lower-frequency end of the range, since the radio-.

frequency resistance of the secondary is many In order to efliciently amplify both side bands 05 I resulting from modulation by voice and music, it is necessary that the resonance curve, at any given radio-frequency over the tuning range, shall be from 8 to 10 kilocycles wide at or 90% of the peak or resonance-current.value.

With radio-frequency transformers of the usual type, input frequencies one, or one and one half kilocycles above or below the frequency to which the secondary may be tuned at the low-frequency end of the range will give ap- 75 proximately 90% of the resonance current, while, at the high-frequency end of the range, the input frequencies may vary as much as fifty kilocycles above or below resonance and still give secondary currents 90% of the resonance current. The selectivity at high frequencies may be somewhat improved by using several stages, but the improvement is not sufficient to be satisfactory. Even a single-stage amplifier is much too selective for good quality reproduction at low frequencies and this condition is greatly exaggerated in a multi-stage amplifier.

If, therefore, an attempt is made to operate a receiving set of the usual multi-stage, tuned radio-frequency type in the vicinity of a highpowered broadcasting station operating on a frequency which lies toward the high-frequency end of the tuning range, much difiiculty is experienced in receiving signals from stations opthe inductance and distributed capacity of the primary of each inter-tube, radio-frequency transformer that it has a natural period considerably below the lowest frequency to which the secondary is intended to be tuned. In addition, I may so relatively arrange the primary and secondary inductors that, at high frequencies, the inductive transfer of energy therebetween is reinforced by energy transferred through the capacity existing between such inductors by reason of their space relationship.

The novel features which I consider characteristic of my invention are set forth with particularity in the appended claims. The invention itself, however, both as to its organization and method of operation, together with further objects and advantages thereof will best be understood by reference to the following description, taken in connection with the accompanying drawings, in which:

Figure 1 is a diagrammatic view. of a preferred form of my invention incorporated into a multi-stage radio-frequency amplifying system,

Fig. 2 is a sectional view of a preferred form of inter-tube transformer, and

Fig. '3 is a diagram illustrating graphically the difference between the voltage amplification characteristics of my improved coupling system and systems of the prior art,

Fig. i is a diagram showing curves illustrative of the selectivity characteristics of the usual low-inductance primary radio-frequency transformer, and

Fig. 5 is a diagram showing curves illustrative of the selectivity characteristics of my improved transformer.

Referring specifically to Fig. 1, a thermionic device 1 having a filament 2, a grid 3 and a plate is provided with an input circuit comprising an inductor 5 shunted by a tuning condenser 6. The inductor 5 may be a loop, or it may be a coupling inductor of any well known type, associated with an antenna-ground circuit (not shown). One end of the inductor 5 is-connected to the grid by a conductor 7; a conductor 8 leads from an intermediate portion of the inductor 5 through a grid biasing battery 9 to the filament 2, and the opposite end is connected to the plate 4 through a small variable or fixed condenser 11; The purpose of the intermediate tap and condenser, in radio-frequency amphfiers of the usual type, is to prevent the generation of self-oscillations in the tube and accompanying circuit network, as is explained in the Rice Patent No. 1,334,118. In my system, however, the condenser 11 has a somewhat differentfunction, which will be later explained. a

An inductor 12, which is the primary of a radio-frequency transformer 13, is connected between the plate 4 and the filament-2, a battery 14 supplying plate potential. A battery 15 in series with a rheostat 16 supplies filament power, and the same battery also may be utilized to energize the filaments of other tubes in the system, as indicated in the drawings.

A. second thermionic device 18 having a filament 19, a grid 21 and a plate 22, has its grid 21 connected to one end of the secondary 23 of the radio-frequency transformer 13,- and its plate 22 connected to the other end thereof through a variable or fixed condenser 24. A condenser 25 is connected across the inductor 23 in order that it maybe tuned to a desired frequency, and a conductor 26 leads froman intermediate point thereon, through a grid-biasing battery 27, to the filament 19.

Plate potential for the thermionic device is supplied from the battery 14, the primary 28 of a radio-frequency transformer 29 being connected between the battery and the plate. The secondary 31 of the transformer 29 is shunted by a tuning condenser 32 and is connected to the remaining portion of the system by conductors 33 and 34.

The two tuned radio-frequency amplifying stages may be followed by additional radiofrequency stages, or by a detector and audiofrequency amplifier. The additional apparatus is optional, and, since it may follow conventional lines, if desirable, is merely indicated by a rectangle 35.

In order that the inter-tube transformers be sumciently selective to reasonably eliminate interfering frequencies, it is only necessary that the ratio of reactance to radio-frequency resistance of the secondary be made at least so high that primary frequencies five kiiocycles on each side of resonance do not induce currents of more than of the resonance value in the secondary.

With the usual radio-frequency transformers, having low-inductance primaries, the approach toward resonance conditions in the primary at the higher frequencies effectively adds resistance to the secondary at such frequencies. Inasmuch as the radio-frequency resistance of the secondary is also higher at high than at low frequencies, it is seen thatthe total damping at high frequencies is considerably greater than it is toward the low-frequency end of the scale, and the selectivity is consequently poorer.

The results of a large number of tests on radio-frequency transformers of the usual type are summarized by the curves in'Fig. 4. Curve C represents the voltage across the secondary tuned to 600 kilocycles, with varying input frequencies, and it is significant to note that a frequency five kilocycles away from resonance gives but 50% of the resonance voltage. Curve D illustrates the fact that the high-frequency end of the range in the usueltransformer is much less selective than at thelow-frequency end, frequencies five kilocycles away from resonance at 1450 k'ilocycles giving approximately 95% of the resonance voltage.

In order, therefore, to improve the selectivity at high frequencies, it is necemary to greatly reduce the losses at such frequencies. These losses may be reduced, according to my invention, by giving to the primary a resonance frequency such that itcannot be near resonance at the high frequencies. This may-be accomplished by either greatly increasing or greatly decreasing the resonance frequency of the primary with respect to highest frequency to which the secondary is tunable; If, however, the resonance frequency of the primary is increasedby lowering its inductance, the amplification at the low frequencies is poor.

Therefore, -1 preferably lower the resonance frequency of the primary to a point -near or below the lowest frequency to which the secondary is tunable. Such a primary does not increase the efiective resistance of the secondary at high frequencies but does increase the effective resistance thereof at low frequencies. The effective resistance thus introduced compensates the decreased damping resulting from the decreased radio-frequency resistance of the secondary at 1,978,087 lower frequencies and broadens-the tuning atsuch frequencies.

Curves E and F, in Fig. 5, summarize the results of a large number of tests made on my improved transformer. It will be noted from these curves that the selectivity at both the high-frequency end and the low-frequency end of the range is substantially the same, the relative displacement between the two curves never becoming greater than one tenth of the maximum resonant voltage.

By properly choosing the inductance and distributed capacity of the primary, and by carefully adjusting the mutual inductance of the primary and secondary, I am thus able to construct a transformer that will be satisfactorily selective at both the high and low-frequency ends of the tuning range.

In order to cover the present broad-cast-frequency range, I have found that a transformer having a primary inductance of approximately 5000 micro-henries and having a natural period of approximately 400 kilocycles, when in circuit, coupled with a secondary inductance of 175 micro-henries, is quite satisfactory. When using a transformer of these dimensions, the tuning condenser shunting the secondary should have a maximum capacity of 550 micro-microfarads and a minimum capacity of 15-20 micro-microfarads or less.

Referring particularly to Fig. 2, my preferred transformer comprises a primary 40 of 1250 turns of #30 double-cotton-covered wire, in the form of a duo-lateral coil, and a secondary 41 of 72 turns of #24 double-,cotton-covered wire wound as a single-layer solenoid. The inside diameter of both primary and secondary is two inches and they are preferably separated to inches axially.

The curve A in Fig. 3 illustrates graphically the voltage-amplification characteristics of a transformer constructed according to my invention, while the curve B illustrates the characteristics of a transformer having the usual low-inductance primary.

According to my invention, therefore, highervoltage amplification may be obtained at the lower frequencies than with transformers having primaries of relatively low inductance. This condition is considered desirable, inasmuch as the better broadcasting stations are, at persent, operating at frequencies toward the lower end of the broadcasting range.

An amplifying system constructed according to my invention is relatively stable. At the higher frequencies, the reactance of the highinductance primaries in the plate circuits of the tubes is predominately capacitive and, consequently, the conditions for negative feed-back are present. In order, therefore, that the systemshall amplify efiiciently, a certain amount of positive feed-back should be supplied and this feed-back is most conveniently introduced by utilizing a small condenser between the plate electrode and one end of the inductor associated with the grid of the same tube.

As already mentioned, the action of this arrangement is different from that of the Rice Patent 1,334,118, although diagrammatically they are similar in appearance. In the Rice patent, the neutralizing condenser and that portionof the grid inductor included between the connection from the filament and the connection to the condenser cooperate to place, on the grid of the tube, potentials equal and opposite in phase to potentials placed thereon by reason of the tube-capacity coupling between the input and the output circuits in order that regenerative feed-back shall be compensated. In my invention on the other-hand, the normal reactance of the plate circuit is capacitive, tending toward de-generation", instead of toward regeneration and oscillation, and the condenser 11 in combination with the portion of the winding with which it is connected, act to produce a regenerative feed-back. While, therefore, the circuit arrangement is diagrammatically like that of the Rice patent, the effect of the condenser and its associated coil is regenerative after the well known Weagant regenerative circuit.

Although not specifically shown in the drawings, it is believed obvious that my invention may be applied to systems in which the plate or output coil for each stage is inductively associated with another coil that is comprised in a feedback system. The feed-back coil may be a continuation of the output coil itself, in which case the said output coil could be characterized as split".

I have found, in addition, that the necessary amount of positive feed-back may be supplied by utilizing certain other well known feed-back systems, in which either capacitive or inductive back-coupling, or a combination of the two, is

the respective windings so they are adjacent,

provided the relative direction of the windings is such that the capacity coupling is in proper phase.

By varying the spacing between the primary and these secondary as by shifting the primary winding on the coil form 42 toward or away from the secondary winding, the voltage-amplification characteristics of the transformer may be appreciably altered, inasmuch as there is a consid-- erable transfer of energy at the higher frequencies across the inter-winding capacity. An optimum spacing can be determined for each transformer, the spacing 42 for the specific transformer illustrated in Fig. 2 being approximately inches.

By reversing the relative direction of the windings, the capacity coupling will then oppose the electromagnetic coupling, and the amplification at the higher frequencies will be considerably lessened.

I have accordingly provided a radio-frequencyamplification system which has a substantially straight-line selectivity-characteristic over the range of frequencies for which it is designed. My improved coupling system may also be given an approximately straight-line voltage-amplification characteristic, although it is somewhat preferable to have a higher voltage step-up at the lower frequencies.

My improved system is also remarkably free from parasitic oscillations, and, although preferably utilizing a feed-back means analogous to the usual neutralizing condenser and associated network, such means is not primarily for th pm.

lations.

Although I have illustrated and described only one specific embodiment of my invention, it is believed obvious that numerous modifications will be apparent to those skilled in the art. My invention, therefore, is not to be limited except insofar as is necessitated by the prior art and by the spirit of the appended claims.

I claim as my invention: l. A radio-frequency transformer comprising a single-layer solenoid having a predetermined tunable frequency range, and a multi-layer coil co-axial therewith and totally external thereof.

2. A radio-frequency transformer having a secondary. comprising a single-layer solenoid, and a primary comprising a concentrated inductance co-axial therewith and totally external thereof, the natural period of said primary being low in comparison with the average frequency which the transformer is designed to handle.

3. A radio-frequency transformer having a secondary comprising a single-layer solenoid, and a primary comprising a concentrated inductance co-axially therewith, the natural period of said primary being low in comparison with the average frequency which the transformer is designed to handle.

4. A radio-frequency transformer having a primary winding, and a secondary winding coupled thereto, said primary winding having a greater inductance than said secondary winding, said windings being wound with respect to each other to constitute a capacity coupling between the two.

5. A radio frequency transformer comprising a primary winding and a secondary winding,

said secondary having a predetermined tunable frequency range, said primary having a natural period below the lowest frequency of said frequency range, said windings being so disposed with respect to each other as to constitute a capacity coupling between them.

' 6. A radio-frequency transformer comprising a helical-wound secondary winding having a length at least as great as' its diameter, and a primary winding of smallamal length relative to that of said secondary winding located adjacent one end of said secondary winding, said primary winding comprising a larger number of turns than said secondary winding whereby a circuit in which said primary winding is connected is rendered resonant at a frequency which is below the range of frequency to be received by said circuit.

7..A radio-frequency transformer according to claim 6 in which'said primary coil is located at the low-potential end of said secondary coil and the high-potential end of said primary coil is at its outer circumference, whereby the capacitive coupling between the windings is negligible.

8. A high-frequency transformer comprising a helical-wound secondary coil and a closely wound multi-layer primary coil, said primary coil being located near the high-potential end of said secondary coil, the portion of said primary coil which is nearest the secondary coil being the high-potential portion, whereby the 9:; coupling between said coils is substantially capacitive as well as magnetic.

9. A high-frequency transformer according to claim 8 in which the efiective direction of said primary winding is opposite that of said second- 131". ary winding, whereby said capacitive coupling aids said magnetic coupling.

10. A high-frequency transformer according to claim 8 in which said primary coil is effectively wound in the same direction as said secondary winding, whereby said capacitive coupling opposes said magneticcoupling.

11. A coupling system comprising two windings, the first of said windings being a cylindrical coil having an axial length at least as great as its diameter, the second of said windings being a multi-layercoil having an axial length oo-axial therewith and totally external thereof,

said coils being spacially adjustable relative to each other.

' GEORGE L. BEES.

its

Images