US2278669A - Frequency changing device - Google Patents

Description

FREQUENCY CHANGING DEVICE Filed Dec. 11, 1939 auRRE/v r 1 Geo/9 E Pgfim INVENTOR.

l We) //VPUf; au7% I A 0RNEYS..

' A complete frequency atented Apr. 7, 1942 FREQUENCY CHANGING DEVICE George H. Polnn, Olmsted Falls, Ohio, assignor of one-half to E. M. Heavens and one-half to Closman P. Stocker Application December 1 1, 1939, Serial No. 308,671

Claims.

This invention relates to static frequency changers and more particularly to frequency changers which depend upon the saturation of a magnetic core for operation. a

In the past frequency changers have been made for converting one frequency to another by means of saturable magnetic cores used in tunedcircuits. However, it has been necessary to utilize output filters for obtaining the desired frequency whenever an approximate sine wave output was desired. Also, frequency changers of the past have relied upon a construction which made it advisable to supply the changed frequency from a non-linear or saturable impedance.

By means of my invention I depart from the prior art by supplying the changed frequency from a linear impedance which is a part of the frequency changing circuit rather than from a filter circuit which is fed by the saturable element. I have developed a static type frequency changing system which is much simpler than the frequency changer circuits known in the prior art. By the use of my invention 1 am able to obtain wave forms on the output of the frequency changer which approximate a sine wave.

changer capable of producing this result can be made according to my invention by arranging in a novel circuit relation three simple elements, that is, a saturable inductance, a linear inductance and a capacitor. It is well to observe that the frequency is changed directly without the use of auxiliary exciting circuits and that the changed frequency having a good wave form is obtained without the use of filters. Furthermore, I am enabled to obtain even or odd harmonics of the input frequency from such an arrangement without the use. of direct current excitation, the use of double core construction on the saturable element, or the use of a magnetic bias on the iron core. In either case the wave form of the output frequency is good.

It is an object of my invention to change the frequency of an alternating current to either an even multiple or an odd multiple of the supply frequency.

It is another object of my invention to produce even harmonics of an alternating current supply by using in a novel circuit relation a capacitor, a non-linear inductance and a linear inductance, without an external bias being applied to the core of the non-linear inductance.

It is another object of my invention to produce even harmonics of an alternating current supply by using in a novel circuit a capacitor, 9. nonlinear inductance and a linear inductance and by utilizing 'a self-exciting phase shifting mechanism to generate a new frequency.

It is another object of my invention to provide a frequency changer in which the output fre-' quency is supplied from a linear inductance in the circuit.

It is another object of my invention to provide a frequency changer in which a capacitor is used as a tuning element common to two tuned circuits each of different frequency.

It is an object of my invention'to provide a complete static type frequency changer having only three necessary circuit elements arranged in a novel circuit relation. j

It is an object of my invention to provide a parallel ferro-resonant circuit in a frequency changer to generate a new frequency and to assist in separating the applied frequency from the new frequency.

It is another object of my invention to cause substantially the whole voltage of the applied frequency to appear across a substantially antiresonant circuit included in a frequency changer.

It is another object of my invention to provide a frequency changer circuit having a capacitor as a commontuning element for at least two circuits tuned to different frequencies and in which the two tuned circuits are operated as parallel combinations with respect to their respective exciting means.

lit is another object of my invention to substantially eliminate the voltage of'the supplied frequency in the output of a static frequency changer without the use of output filters.

It is a further object of my invention to utilize a self-exciting phase shifting mechanism to generate a new frequency.

It is another object of my invention to make use of a phase displacement of the voltage between two elements in'a frequency changer to separate the supply frequency from the output frequency.

It is another object of my invention to balance out and separate voltage of the supply frequency from the voltage of the output frequency in a frequency changer by connecting an output impedance at approximately null points or equipotential points with respect to voltage of the supply frequency but which points are not at equipotential with respect to voltage of the output frequency.

It is a further object of my invention to cause, in a frequency changer, a large current of the changed frequency and a small current of the supplied frequency to flow through the impedance supplying the voltage to be utilized.

It is still a further object of my invention to provide a new method or arrangement of separating the changed frequency from the supply frequency in a static type frequency changer circuit.

Other objects and a fuller understanding of my invention may be had by referring to the following description and claims taken in conjunction with the accompanying drawing in which like parts are designated by like reference characters and in which:

Figure 1 shows one embodiment of my fretion on the saturable inductive element. Other quency changer which is capable of supplying either even or odd harmonics of the input frequency depending upon the adjustment of the various circuit elements.

Figure 1A shows an assumed circuit approximately equivalent to the circuit shown in Figure l; the non-linear element of Figure 1 is represented in Figure LA by a linear element and a harmonic generator.

Figure 2 shows a modification of the frequency changer as shown in Figure 1 in which the saturable element has been changed to use a smaller value of capacitance.

Figure 3 is a modification of Figure 2 in which the condenser connection has been changed to give an alternative method or arrangement of balancing out the supply frequency from the output.

Figure 3A shows voltage-current characteristics of Figure 3, and

Figure 3B shows the input voltage-current characteristics which manifest a change in predominant frequency with variation in input voltage.

with reference to Figures 1, 1A, 2, 3, 3A and 3B in which corresponding parts are indicated by the same reference characters, it will be seen that my frequency changer utilizes in a novel circuit a capacitor, a non-linear inductance and a linear inductance.

It is well known in the prior art that new frequencies may be obtained from circuits using the three elements mentioned above, but a frequency changer made in accordance with my invention operates in a fundamentally different manner and produces a changed frequency of flood wave form without the use of additional parts or auxiliar circuits.

'Iypical frequency changers of the static type shown in the prior art may be found in such circuits as shown and described in U. '8. Patents Nos. 1,587,512, 1,698,284, 1,782,348, 1,877,703, or 2,150,386; or in som of the more recent publications such as the article entitled Magnetic Generation of a- Group of Harmonics" by E. Peter: son J. M. Manley and L. R. Wrathall, pages 995- 1001, in Electrical Engineering, August 1937.

In reviewing the prior art it will be evident that in general the frequency changer circuits contain many parts which are intended to act as filters 'or which in some other way are used to reduce the distortion of the output wave obtained from the frequency changer. Also it will be noted that in general. the output frequency or changed frequency is supplied to a load which is taken directly from the non-linear inductive element. In order to produce even harmonics of the supply frequency by circuits of the prior art, it was necessary to apply an external bias to the non-linear inductance. This was done circuits use rectifiers in the filter circuit as a means of obtaining even harmonics of the applied frequency.

By my invention as shown in Figures 1, 2 and 3, I am able to get even or odd harmonics having excellent wave form by the use of only three reactive circuit elements.

With reference to Figure 1, the frequency changer circuit comprises winding II on saturable core it connected in parallel relation with capacitor 12 and this combination is connected in series with winding I4 on a substantially linear core l3. Core II has an additional winding II for supplying power of the changed frequency to a load. Source of alternating current i0 supplies powerto this combination and depending upon the tuning of the various circuit elements, it is possible to obtain even or odd the input frequency from output winding ll. As will be recognized non-linear inductance II is the element which generates the new frequencies in co-operation with the other circuit elements. In the circuit shown in Figure 1, an approximately anti-resonant condition t the applied frequency is obtained by tuning non-linear inductance II with capacitor I2. Capacitor I! also cooperates with linear inductance It to form a tuned circuit for the new frequency. I prefer to have the impedance of inductance l4 linear over the operating range. My circuit contains only three essential elements and may have one or two different tuned circuits. Where two tuned circuits ar used the capacitor is utilized to tune blothirircuits with an inductive element in each c rcu I shown as ap proximately equivalent to non-linear inductance H, although many details have been omitted for purposes of simplifying theexplanation.

the re r values of circuit constants are selected it 1; p

sible to have capacitor I! act as a common tunlinear inductance ll, to tune the circuit comprising generator ll, inductance I4 and capacitor I! to be substantially anti-resonant to the changed or new frequency.

In the frequenc changer ure 1, non-linear inductance II and capacitor I! are tuned to approximately an anti-resonant condition with respect to the frequency of source In and the circuit operates to supply the changed frequency from output winding l5. As previously ointed out in connection with Figure 1A, the impedanc of winding I4 is adjusted with capaciwr I: to form an approximately anti-resonant circuit to the output frequency. By referringto Figure 1A, it will be observed that the anti-rescnant circuit to the new frequency is connected in parallel circuit relation with respect to generator ll of th new frequency. By operating the circuit comprising inductive impedance I4 and capacitive impedance I! in the neighborhood of anti-resonance I obtain a distinct advantage when supplying energy of the new frequency to a unity power factor load because with th circuit connection as shown a load connected to output winding I5 is effectively connected across the anti-resonant circuit and variations in load resistanc do not tend todetunc the circuit. Therefore, variations in load resistance between zero and infinitive do not appreciably affect the generation of the new frequency.

By way of example it might be assumed that 180 cycles per second is desired from output winding l5 and that source H) has a frequency of 60 cycles per second. Capacitor l2 and nonlinear inductance are then tuned to approximately an anti-resonant condition with respect to the 60 cycles and inductance I4 is tuned with capacitor l2 to approximately an anti-resonant condition with respect to the 180 cycles. Inductance II is proportioned to saturate within the range of voltage supplied by source III while the impedance of inductance H remains substantially linear. In such a circuit, saturation of the nonlinear inductance ll generates harmonics of the input frequency and a current flows in the 180 cycle tuned circuit. Tests show that by maintaining an anti-resonant circuit to the 180 cycles a small current of the harmonic frequency flows through saturable inductance ll while a large current of the harmonic frequency flows in the anti-resonant circuit consisting of condenser l2 and linear inductance element H. Therefore, in operation the anti-resonant circuit comprising non-linear inductance H and capacitor l2 limits the 60 cycle current flowing through linear inductance H to a very low value and the voltage induced in inductive element M will be, for practical purposes, only the harmonic voltage. With the circuit operating as explained the predomi nant harmonic current through linear inductance M will be of the desired frequency and energy of this frequency may be supplied to a load connected to output winding l5.

With the circuit as shown in Figure 1 in operating condition, measurements indicate that practically the whole of the 60 cycle voltage of source I!) appears across capacitor [2 and nonlinear inductance H and that very little appears across the inductance element l4.

Observations made by means of an oscillograph element in series with inductance l4 and source In show that substantially all of the current flowing through inductance I4 is 180 cycle current. Because capacitor I2 is common to both circuit shown in Fig- I anew tuned circuits it has both 60 cycle and 180 cycle circuit shown in Figure 1 there is some residual voltage impressed across it. Non-linear inductance I I being connected in parallel with capacitor I! also has voltages of both frequencies impressed across it, controlling its saturation and the generation of the new frequency. From this observation, I find that the new frequency is automatically separated from the supply frequency by the novel construction and operation of a frequency changer made in accordance with my invention, and filters are unnecessary for practical purposes.

Although I have explained the operation of the circuit shown in Figure l by assuming capacitor l2 and inductance l4 tuned to 180 cycles, I have been able to change the tuning of these two elements to obtain cycles which is an even har monic of 60 cycles or 300 cycles which is the fifth harmonic of {50 cycles. Any of these new frequencies may be obtained with wave forms which are good approximations of the sine wave whenv taken directly from output winding l5. In my invention, the resistance of the circuit elements is preferably low. I

It is possible to obtain even harmonics from the without the use of external bias or any equipment other than that shown in the diagram. In view of the fact that it is generally considered necessary to bias the magnetizable core to produce even harmonic distortion in a non-linear inductance such as element II, a question is presented as to the exact method of operation when the even harmonics are generated. One possible explanation is that magnetism in core l6 of inductive element H, tesimal even harmonic distortion. This even harmonic distortion causes in the first instance a feeble 120 cycle current to flow through capacitor l2, inductive element and saturable inductive element H. phase to increase further the even harmonic distortion in inductive element ll thereby causing an even increasing 120 cycle current to flow until a normal operating value is reached. The action is cumulative and the 120 cycle current through the tuning elements l2 and It quickly builds up to a large value.

When 120 cycles is obtained from a circuit as shown in Figure 1,- the wave form is good and approximates a sine wave. Furthermore, under certain circuit conditions it is possible to gradually load output circuit l5 and supply 120 cycles to the load until an approximately short circuit load condition exists. The generated frequency disappears only when the output winding i5 is short circuited and then just as soon as the short circuited condition is removed the generated frequency almost instantaneously reappears and builds up to its original strength.

Although the. above explanation of the operation of my frequency changer seems to be substantiated by observations and measurements, I do not wish to be bound or to limit the scope of my invention by this explanation.

The frequency changing circuit shown in Figure 2 uses the non-linear inductance 20 as a stepup transformer which permits the use of a smaller capacity of higher voltage rating compared with the capacitor used in the circuit shown in Figure 1. However, the frequency changer in Figure 2 operates the same as explained in connection with Figure 1.

It can be shown that with this type of con struction the same anti-resonant circuits are present in Figure 2 as are shown in the simplified which introduces infini- This current is in the proper and assumed diagram of Figure 1A. Laboratory data tends to indicate that there is one difference between the circuits of Figures 1 and 2. This difference resides in the fact the internal impedance ofthe generator of the new frequency is not the-gsame for Figure 2 as for the circuit shown in Figure 1. This change in internal impedance does not seem to make any difference in the operation, except where output winding I5 is connected directly into another similar static type frequency changer in cascade. In this case if the first frequency changer is constructed as shown in Figure 2, the output of the second frequency changer may be more distorted than if the first frequency changer is constructed as shown in Figure 1. It is therefore evident that in some cases where space is a factor, some advantage is gained by the use of the circuit shown in Figure- 2, butin general where cascade operation of fre quency changers constructed in accordance with my invention is contemplated, I have found that the one shown in Figure 1 is preferably employed except for the last stage.

In Figure 2 I have shown leads 38 and 39 which are connected across linear inductance I4 for the purpose of supplying power of the new frequencies to a load. Linear inductance Il may be used as an auto transformer to supply different voltages to a load by a tap such as 31 on winding M. It is to be understood that the linear inductance H, as shown in Figures 1, 2, and 3, may be used as a choke, an auto-transformer or an insulating transformer.

Another modification of my frequency changer is shown in Figure 3 in which non-linear inductance is connected in series relation with capacitor l2 and source of supply iii. In Figure 3 linear inductance H is connected between intermediate tap 2| canon-linear inductance 20 and generator ID in such a way that a portion of the winding of non-linear inductance 20 and linear inductance H are connected in series relation with generator l0. Although it is not necessary for intermediate tap 2| to be spaced exactly in the middle of the winding of non-linear inductance 20, I have found that very good results can be obtained when intermediate tap 2| .is at the mid-point. I have, however, moved tap 2| to an off-center position and by changing the circuit constants, obtained good results. i

At first glance, Figure 3 seems to be entirely different from the frequency changer circuit shownin Figure 2, but the only structural change is that capacitor I2 is connected to the opposite side of source l0. With the circuit connected as shown in Figure 3,- good operation may be obtained with only one resonant circuit. In general, I prefer to tune' the circuit comprising capacitor l2, inductive element I, together with the leakage reactance of non-linear inductive 20 to the new frequency; An operating difference exists between the -ircuit shown in Figure 3 and the circuits shown' in Figures 1 and 2. in that the voltage of the supply frequency isbalanced out of inductive element in Figure 3 in a new and novel manner.

When non-linear inductive element 20 has a higher impedance to the frequency of source l0 than capacitor |2, the voltage across non-linear element 20 is substantially in phase with the voltage of source l0 and the'voltage across capacitor I2 is approximately 180 out of phase with the voltage of source l0. Under this condition there exists a point on the winding of non-linear element 20, which is approximately at l2 was approximately half the valu through any other the same potential with res of source in as the side of nected to capacitor i2. Tap 2| is preferably placed in or near this point on the winding of non-linear element 20 which gives the minimum voltage of the frequency of source ID across inductive element [4. Thus I make use of a phase displacement of the voltage between two elements in the frequency changer to separate the supply frequency from the output frequency.

Although voltage of the supply frequency is balanced out of'output impedance H by connecting impedance H to points which are at approximately equipotential with respect to the supply voltage, the voltage of harmonics of the frequency of source In generated across nonlinear element 20 by the current flowing through it will appear across element It. As in the previousv explanations, the circuit is arranged so that the generator of the new frequency works into an approximately anti-resonant circuit so that there is no excessive current drain on the generator of the new frequency to decrease its voltage. However, the circuit is tuned so that the current of the new frequency flowing through non-linear element 20 aids in the generation of the voltage of the new frequency. The fact that current flow of the new frequency through non-linear element 20 aidsin the generation of voltage of the new frequency across non-linear element 20 is evidenced by the fact that even harmonics of the frequency of source H) are readily obtained in my frequency changer by proper tuning of inductive element H with the other circuit elements.

Test results indicated that with the circuit as shown in Figure 3, excellent operation can be obtained when tap 2| is at the mid-point of the winding of non-linear element 20. Tests made upon a device using this center tapped arrangement showed that the voltage of the frequency of source It! across non-linear element 20 is approximately twice the voltage of the frequency of source l0 acrosscapacitor l2. The current of the new frequency flowing through capacitor e of the current of the new frequency flowing through linear element I. It can be seen that if non-linear element 20 is considered the generator of the new frequency and linear inductive element It is connected across only half as many turns as capacitor I2, then to maintain a condition of anti-resonance the inductive current through linear element I4 would have to be twice the capacitive current through capacitor i2. Experimental results indicate that the circuit is near a condition of anti-resonance with respect to the new frequency. Under this condition inductance M will alwayshave a greater current of the new frequency through it than will flow single portion of the' circuit. I have observed the current flow in various pect to the frequency the generator Ill conbranches of the frequency changer as shown in 'source i0 and 180 c cycles flowed in all of the branches of the cirmately twice the found in anyoth Test results showing the variations in current flow through linear inductance H with variation of the voltage supplied by source ID, are shown in Figure 3A. Curve a shows the effective current through inductive element H which is a current of the same fundamental frequency as ource l0 together with harmonics of that frequency. Curve 1) shows the value of current flowing through inductance H which has the same fundamental frequency as source l0.

With reference to curve b it can be seen that the value of fundamental current, having the same frequency as source l0, passes through a minimum value betwen values of input voltage represented by c to d. It will be noted that the effective current represented by curve a is relatively large compared with the current of the fundamental frequency within the input voltage range 0 to d. This means that inoperating a frequency changer tion the harmonic currents are greatly in excess of the fundamental currents. Because inductive element H has a much desired frequencies than to the fundamental frequency and also because the flow of the desired harmonic is much larger than the current of the fundamental frequency the output voltage delivered to a load will be practically free of voltage of the frequency supplied by source In. I have found that the best operating range is between the limits 0 and d shown in Figure 3A.

Figure 3B shows the variation of the effective value of current supplied to my frequency changer as the voltage of source Ill varies. A rather unusual and unexpected feature of my invention ,is shown by the curve in 313 which is that as the input voltage is gradually increased the effective current assumes and different output'frequency predominates. It is togbe understood that this result is obtained without readjusting any of the circuit elements. In one of the devices tested the supplied frequency was 60 cycles. The predominate frequencyof the output frequency changer between the limit m and n and 0 and p was 180 and .300 cycles respectively. From this result it can be seen that my frequency changer can be used quencies without altering any of its elements.

Throughout the description and in the following claims I refer to linear inductance. By this I mean a substantially linear inductance element. A suitable linear inductance may be made with an iron core having an air gap in series with the magnetic path and may comprise a choke coil, a transformer with insulated windings, or an auto transformer. The term "non-linear inductance means a saturable inductance or an inductance which changes its impedance for various values of current flowing through it. Also I mention resonant and connection with inductances which are either substantially linear or non-linear. It 'is-t'o be understood that such circuits need not be sharply tuned to the desired frequencies because the inductive elements contain iron or other magnetic material and the impedance may currents through the inductive elements. ever, in the case of cults involving the non-linear inductance elements these may be understood to be either in series ferro-resonance or parallel ferro-resonance as the case may be.

Although I have described my invention with a Howin accordance with my invenhigher impedance to the the shape 'of a series a of plateaus and as each plateau is reached a new voltage delivered by my vary with the re onant or anti-resonant cirby way of example, and that to produce two distinct frequency,

anti-resonant circuits in I it is understood been made only numerous changes in the details of construction on the combinations and arrangement of parts may be resorted to without departing from the spirit and scope of the invention as hereinafter claimed.

I claim asmy invention:

1. A frequency changing circuit for obtaining a new frequency from an alternating current source comprising, in combination, a linear inductance, a capacitor and a non-linear inductance having an intermediate tap forming a first winding means and a second winding means, said linear inductance and said first winding means being connected in series with the alternating current source, said capacitor and said second winding means being connected in series with each other and shunting the linear inductance, and output means connected to the linear inductance.

2. A frequency changing circuit for obtaining a new frequency from an alternating current source comprising, in combination, ductance, a capacitor and a non-linear inductance, said linear inductance and said non-linear inductance establishing a series circuit with the alternating current source, said non-linear inductance generating a new frequency different from the frequency of the supply source, said capacitor being connected to the alternating current source in circuit relation with the said linear and non-linear inductance and establishing two substantially anti-resonant circuits, the capacitive effect of the capacitor and the reactive effect of the non-linear inductance building up an antiresonant condition with respect to the frequency of the alternating current source and the capacitive effect of the capacitor and the reactive effect of the ,linear inductance building up an antires'onant condition with respect to the new frequency, and output means connected to the linear inductance for giving the new frequency, said capacitor and one' of said inductances being connected in closed circuit with the supply source and the other said inductance being in a closed electrical circuit including the capacitor.

3. A .Z'equency changing circuit adapted to be connected to a source of alternating current and generate new frequency comprising, in combination, a linear inductance, a capacitor, and a non-linear inductance, said capacitor and nonlinear inductance being connected to form a parallel combination and generating the new freand means for connecting said parallel combination in circuit with ,said linear inductance and the source of alternating current, said capacitor and said linear inductance forming a closed circuit with the supply source and having substantially equal and opposite impedance values with respect to the new frequency to cause substantially all of the voltage of the new frequency to appear across the linear inductance.

4. A frequency changing circuit adapted to be connected to a source of alternating current and certain degree of particularity, that the present disclosure has generate a new frequency comprising, in combination, a linear inductance, a capacitor, and a non-linear inductance, said capacitor and nonlinear inductance being connected in circuit relation to form a parallel combination and generating the new frequency,

said parallel combination being connected in series with said linear induct! ance and the source of alternating current and delivering to the linear inductance the new generated frequency, said capacitor and said linear inductance forming a closed circuit with the supa linear in- I ply source and having substantially equal and op posite impedance values with respect to the new frequency to cause substantially all of the voltage of the new frequency to appear across the linear inductance, and output means for obtaining the new frequency from said linear inductance.

5. A frequency changing circuit adapted to be connected to a source of alternating current and generate a new frequency comprising, in combination, a linear inductances non-linear inductive element and a capacitor forming two substantially anti-resonant circuits to separate the new frequency from the frequency of the source of alternating current, said non-linear inductive element and said capacitor being connected in circuit relation to form a parallel combination and generating the new frequency, said parallel combination comprising one of said anti-resonant circuits and being anti-resonant to the frequency of the source of alternating current, said parallel combination being connected in series with the linear inductance and the source of alternating current and delivering to the linear inductance the new frequency, said capacitor and said linear inductance forming a closed circuit with the supply source and being anti-resonant to the new frequency and comprising the other anti-resonant circuit, said capacitor constituting a tuning element common in both anti-resonant circuits, and output means for supplying the new frequency from the linear inductance.

6. A frequency changing circuit adapted to operate from an alternating current source utilizing three reactive elements to form two anti resonant circuits for generating a new frequency and separating the new frequency from the fre-- quency of the alternating current source com,- prising, in combination, a linear inductance, a non-linear inductance, and a capacitor, said nonlinear inductance and capacitor forming a first circuit anti-resonant to the frequency of the altemating current source and generating a new frequency, said linear inductance and capacitor forminga second circuit anti-resonant to the new frequency, said first circuit and the linear inductance being connected in circuit relation with each other and the alternating current source to cause substantially all the voltage of the new frequency to appear across the linear inductance, and output means to supply the new frequency from the linear inductance, said ca pacitor and one of said inductances being connected in closed circuit with the supply source and the other said inductance being in a closed electrical circuit including the capacitor.

7. A frequency changing circuit adapted to be connected to a source of alternating current i and generate a new frequency comprising, in

combination, a linear inductance, a capacitor, and a non-linear inductance having an inter mediate tap, said non-linear inductance and said capacitor being connected in series with each other and with the source of alternating current and generating the new frequency, said linear inductance being connected to the intermediate tap on the non-linear inductance to a point common to the capacitor and the source of alternating current, said non-linear inductance and said capacitor delivering to the linear inductance the new frequency,.and output means for supplying the new frequency from the linear inductance.

8. A frequency changing circuit adapted to be connected to a source of alternating current and generate a new frequency comprising, in combination, a linear inductance, a capacitor, and a non-linear inductance having an intermediate tap, said non-linear inductance and said capacitor being connected in series with each other and with the source of alternating current and generating the new frequency, said linear inductance being connected to the intermediate tap on the non-linear inductance to a point common to the capacitor and the source of a1- ternating current, said non-linear inductance and said capacitor delivering to the linear inductance the new frequency, and output means for supplying the new frequency from the linear inductance, said intermediate tap on the nonlinear inductance being electrically selected to cause substantially all the voltage of the new frequency to appear across the linear inductance.

9. A frequency changing circuit adapted to 0perate from an alternating current source utilizing three reactive elements to form two antiresonant circuits for generating a new frequency and separating the new frequency from the frequency of the alternating current source comprising, in combination, a linear inductance, a non-linear inductance, and a capacitor, said nonlinear inductance and capacitor being connected in parallel and generating the new frequency and forming a first circuit anti-resonant to the frequency of the alternating current source, said linear inductance and capacitor forming a second circuit anti-resonant to the new frequency. said first circuit and linear inductance being connected in circuit relation with each other and the alternating current source to cause substantially all the voltage of the new frequency to appear across the linear inductance, and output means to supply the new frequency from the linear inductance.

10. In combination with an alternating current source, a linear inductance and an antiresonant circuit connected in circuit relation with each other and with the alternating current source to form a frequency changing circuit, said anti-resonant circuit being substantially anti-resonant to the frequency of the alternating current source nd comprising a non-linear inductance and a capacitor and causing harmonics of the frequency of the alternating current source to appear across said linear inductance, said capacitor and one of said inductances being connected in closed circuit with the supply source and the other said inductance being in a closed electrical .circuit including the capacitor.

GEORGE H. POHM.

Images