US2125119A - Coupling transformer - Google Patents

Description

Juy 26,

193s. H, T; MAN 2,125,119 COUPLING TRANSFORMER s sheets-sheet 1 Filed June 2, 1936 Harold? Lyman S A tor-wey July 26, 19381. H. T. LYMAN COUPLING TRANSFORMER 5 Sheets-Sheet 2 Fiied June 2, i956 Inventar:

Patented July 26, 1938 UNITED STATES PTET OFFICE l 2,125,119 coUPLING TRANsFoRMER York Application June 2, 1936, Serial No. 83,057

11 Claims.

My invention relates to coupling devices for electric circuits and more particularly to coupling units which are adapted to be used in coupling successive stages of an electron discharge ampliy, er system.

The advantages of transformer coupling in high frequency electric circuits over other forms of coupling are well known. This type of coupling is particularly advantageous in connection with electron discharge amplifiers wherein it is desirable to obtain the highest degree of undistorted amplification with a minimum number of circuit elements. However, the conventional type of coupling transformer now employed for this purpose is open to the limitation that a uniform amplification response characteristic can not be obtained over a Wide band of frequencies. This limitation is due to the fact that series resonant conditions are produced in the windings of a transformer at certain frequencies which lie Within the operating range and which produce attenuation of the components of current having frequencies corresponding to such series resonant frequencies. It has therefore been the practice in the past to employ resistance coupling in those circuits wherein it was desired to transmit currents having a wide band of frequencies between two electrically coupled circuits.

It is an object of my invention to provide in a high frequency circuit'a transformer which is capable of transmitting between two circuits electric oscillations having component frequencies extending over a wide range with equalized transmission eiciency for all of the frequency components within the range.

It is a further object of my invention to provide a transformer having the above operating characteristics which is self-contained, which is small in size, and which is of economical construction.

In accordance with my invention, the above objects are attained by dividing the primary and secondary windings of the transformer into coil sections. These sections are designed to have different resonant frequencies and the sections of each of the two windings are so disposed on the transformer core and so connected that a high leakage reactance obtains between the respective coil sections.

Further, in accordance with my invention, the effects of series resonance between any two coil sections of either of the windings are obviated by shunting the sections having the higher resonant frequencies with resistances having values 55 equal to the reactance values of the coil sections which they shunt at the frequency where resonance obtains between the shunted coil section and the coil section having an immediately lowei` resonant frequency.

A transformer constructed in accordance with p my invention may be considered as comprising a plurality of frequency transmission channels, each capable of transmitting a predetermined portion of the operating frequency band. The coil sections described above included in each of these channels are so connected and so dist posed about the core of the transformer that the respective channels operate substantially independently of each other and no undesired interaction between the different channels, such as loading effects, are produced. In this manner, all of the frequency components existing in the output from a source of `oscillations may be impressed on any form of utilizing circuit with a uniform transmission efficiency.

The novel features which I believe to be characteristic of my invention are set forth with particularity in the appended claims. My invention itself, however, both as to its organiza` tion and the method of operation, together with further objects and advantages thereof, will best be understood by reference to the specification, taken in connection with the .accompanying drawings, in which Fig. l illustrates a circuit having one form of my improved coupling unit included therein; Figs. 2 and 3 illustrate curves relating to the unit shown in Fig. 1; Fig. 4 illustrates a modification of the unit shown in Fig. 1; Fig. 5 illustrates a physical embodiment of my improved transformer; Fig. 6 illustrates the circuit connections for the transformer shown in Fig. Figs. '7 `to 13 inclusive illustrate the equivalent circuits of the coupling unit shown in Fig. 5 when different frequency bands are considered; Fig. 14 illustrates another physical embodiment of my improved transformer, and Fig. illustrates the circuit connections for the transformer shown in Fig. 14.

Referring to Fig. 1 of the drawings, I have illustrated the simplest form of my improved coupling unit at I as a device for impressing the output from an electron discharge amplifier 2 on the input circuit of an electron discharge amplifier 3. As shown, the coupling unit I comprises a core 4 upon which are wound primary and secondary windings 5 and 6. The primary winding 5 is illustrated as comprising a pair of series connected sections 'l and 8 and the secondary winding is comprised of a pair of series connected sections 9 and I0. In order to prelil vent electrostatic or capacity coupling between the windings 5 and B an electrostatic shield Il is provided which is maintained at ground potential.

In certain applications, as for example in television modulation circuits, the oscillations which it is desired to transmit between the two coupled circuits may have frequency components extending over an exceedingly wide range. Thus, the discharge amplifiers .2 and 3 may form a part of an amplifying system for amplifying the signal currents generated in a circuit whose energizetion is controlled in response to the energization of a television pick-up device.

It is known that to operate satisfactorily a coupling device or coupling network between two electrically coupled circuits must be capable of transmitting between the circuits all of the frequency components of the alternating voltage impressed across terminals of the coupling system with a uniform transmission efficiency. Obviously, to obtain this uniform transmission eiciency the impedance of the coupling device or network must be maintained above a certain predetermined value for all of the frequency components of current flowing therein.

As will be pointed out in detail hereinafter, certain difficult problems are involved in the construction of a transformer offering the desired high impedance to currents of all frequencies within a wide range. One such problem which arises in the construction of a transformer suitable for transmitting between windings currents having a wide band of frequencies is that of eliminating undesired reductions in the impedance of the windings produced by series resonance effects between the inductance and distributed capacity of the windings at certain frequencies.

Such resonance effects may be overcome by dividing the primary and secondary windings into sections having different resonant frequency ,characteristics thereby to provide a plurality of channels for the transmission of energy between the two coupled circuits. Thus the sections 3 and if! and the coupling therebetween may be considered as comprising one channel of the coupling device of Fig. 1, and the inductively related sections l' and S may be considered as the other channel. By dividing the windings in this manner each of the inductively related pairs of sections may be designed to transmit a particular portion of the operating frequency range. For example, the winding sections 8 and iii may each comprise a small number of turns as compared to the number of turns in the sections 'i and 9 respectively, and may be designed to have a very low distributed capacity and a higher resonant frequency than that of the sections 'l and 9. As thus constructed, these inductively related sections operate to transfer between the two ampliner stages the high frequency components of the oscillations generated in the output circuit of the amplifier 2 and the relatively larger sections i and 9 operate to transfer the low frequency components of the oscillations.

One difficulty encountered in utilizing this multi-channel method of coupling is that produced by resonance effects occurring between the reactances ci' the coil sections at certain Vfrequencies within the operating frequency range. These resonance effects reduce the effective iinpedance of the coupling device at certain frequencies to such an extent that the transmission of currents having these frequencies may be materially reduced or completely prevented.

The effect of resonance between the reactance of the winding sections to reduce the impedance between the terminals of the windings to currents of certain frequencies may more clearly be understood by reference to Fig. 2 wherein the impedance characteristics of the winding sections 'i and 3 of the primary Winding 5 are plotted as a function of frequency, assuming no load conditions of the transformer. In this figure, the curves i2 and I3 represent thercalculated internal inductive reactances of the winding sections l and 8 respectively as a function of frequency and the curves iii and l5 illustrate the apparent external reactance of the same circuit elements including the effects of Vdistributed capacity of these elements. The curves I and Il show the apparent resistances of the windings l and 8 for the various frequencies within the operating range.

it will be observed that due to the distributed capacity of the coil section l, this coil is in a resonant condition at a frequency F1 and at higher frequencies it is at first increasingly capacitively reactive and later decreasingly capacitively reactive. This capacitive reactance is sufficient at frequencies F2 (slightly higher than frequency F1) and F3 to form a series resonant circuit with the inductive reactance of the section ii. However, at the frequency F2, the apparent resistance of the section l, as shown by curve i6, is suflicient to maintain the impedance between the terminals of the primary winding 5 above that predetermined value below which the source of impressed voltage is overloaded. At F3, however, the apparent resistance of the winding 'i is negligible and the series resonance produced between the reactances of the sections 'i' and 8 reduces substantially to Zero the overall impedances of the primary windings to currents of these frequencies. Hence, at this frequency, F3, and in the absence of additional means for maintaining the overall terminal impedance above the predetermined value necessary to produce equalized transmission efficiency, substantially no transfer of energy having this frequency will occur between the primary and secondary windings.

In order to obviate the above diiculty and provide the necessary impedance in the intermediate frequency zone at which neither of the windings 'i or 8 is otherwise effective to transmit energy, the resistance I8 is provided in shunt with the primary winding section 8.

At the frequency of resonance between the reactances of the windings l and 8 the resistance i8 should be very low in order to prevent series resonance. However, if such a value of resistance be employed, substantially the entire effective primary section would be short-circuited at frequencies above this series resonant i'requency. Accordingly, it has been found desirable to use a compromised value of resistance IB, the value being determined to be equal to the inductive reacta-nce of the winding section B at the resonant frequency where this inductive reactance is equal to the apparent capacitive reactance of the winding section '1.

The presence of distributed capacity in and load across tlie secondary windings 9 and lli affects the apparent reactances of the primary windings to the extent that the resonant frequencies F1 and Fi are shifted slightly toward decreasing frequency and furthermore the peak values of apparent reactances and resistances may be reduced, but in general the relations eX- pressed above between the values of the primaryk that the overall impedance between the terminals of the primary winding does not decrease below' a predetermined minimum value at any frequency up to the resonant frequency F4 of the winding 8. The curve further shows that at low frequencies the greater portion of the impedance between the terminals of the primary winding is due -to 4the impedance of the winding section l andat high frequencies the impedance of the winding section 3 is predominant. At the intermediate frequencies where the reactance of the winding l is partially or totally in series resonance with the inductive reactance of the winding 8, the impedance between the primary winding terminals is due almost entirely to the apparent resistance of the shunt resistor i8 in parallel with winding 3.

As shown by curve I9 of Fig. 3, if the winding sections l and S be constructed in the manner described above and the resistance i3 be provided in shunt with the section 8, the impedance is sufficient to transmit currents having all frequencies Vless than a frequency slightly higher than the resonant frequency of the high frequency winding 8.

Although I have described only the effects of series resonance in the primary winding, it will, of course, be understood that these effects also occur in the secondary winding and, accordingly, it is desirable to provide resistors connected in shunt `with one secondary winding or in shunt with each of the secondary winding sections for the purpose of preventing series resonance effects between the winding sections from lowering the impedance of the winding to currents having frequencies corresponding to the frequencies at which such resonance effects occur. The value of these resistors is determined in accordance with the principles discussed above in connection with the primary section shunting resistance. Thus, in the two section secondary winding of the device illustrated in Fig. 1 a resistor E@ is provided having a value equal to the inductive reactance of the winding section i@ at the frequency where resonance obtains between this reactance and the apparent capacitive reactance of the section 9, inclusive of eiIects on this reactance of such coupling as may exist between all windings.

The frequency range may be extended by adding additional winding sections having resonant frequencies increasingly higher than that of the high frequency sections 8 and il) and shunting each of these sections with a resistance having a value equal to the inductive reactance which it shunts at the frequency of resonance of the reactance of this winding with the reactance of the next lowerV resonant frequency winding section.

Thus, in Fig. 4, I have illustrated a transformer constructed in accordance with the principles outlined above in which three channels are provided for transmitting the low, intermediate and high frequency portions of the frequency band between the two windings. This transformer is illustrated as including primary winding sections 2|, 22 and 23 and secondary wndingsections 2li, 25 and 26, the sections being so designed that the inductively coupled sections 2l and 24 constitute the low frequency transmitting channel, the sections 22 and 25 constitute the intermediate frequency transmitting channel, and the sections 23 and 26 constitute the high frequency transmitting channel. Each of the higher resonant frequency sections 2-2, 2?, 25 .and .2S is shunted by a resistor having a value determined in the manner set forth above. It will, of course, be understood that in practiceadditional frequency channels may be provided by increasing the number of winding sections and .providing additional shunt resistors.

In the construction of a transformer wherein all of theinductively coupled coil sections constituting the various frequency transmitting channels are mounted upon the same core, it is necessary, if satisfactory operation is to be obtained, so to arrange the sections of each winding that a considerable leakage reactance is obtained between the respective sections of the primary and secondary windings. This necessity arises from the fact that the low capacity reactance of certain of the windings may constitute a heavy load on the other windings if the winding sections are in closely coupled inductive relation.

In accordance with my invention, the above difficulty is obviated by mounting the winding sections on the transformer core in the manner illustratedin Fig. and connecting the elements of the coupling unit in accordance with the circuit shown in Fig. 6 wherein elements identical to those of Fig. 5 are identified by like reference characters.

Referring to Fig. 5 of 'the drawings I have illustrated the transformer as comprising a core 2l upon which are mounted primary winding sectons 28, 29 and 3l), and secondary winding sectionsl, 32, 33, Sli and 35. The primary winding section 28 and the secondary section 3l are arranged in closely coupled inductive relation upon the center leg of the core and constitute the Channel for transmitting currents having frequencies extending over the low portion of the operating frequency range. The high frequency transmitting channels are comprised of the primary sections 29 and til and the secondary sections 32, 33, @il and 35. It will be observed that the winding sections included in the high frequency channels are mounted on the outer legs of the three-legged core 2l thereby to provide a high leakage reactance between these winding sections and the sections of the low frequency channel.

As will be more clearly brought out hereinafter the respective primary winding sections are connected in such a manner that when the respective winding sections are energized the sections of the high frequency channels including the primary winding sections 29 and 3E] are magnetically isolated. Thus, the windings are connected in such l anner that when the primary winding 28 is enev 'ized a ux p1 indicated by the dash lines is produced in the core which traverses the core in the direction indicated by the arrows appended to the identified dash lines. Similarly the winding sections 29 and .30 are connected so that when they are energized a flux p2 is produced in the core having the directon indicated. It will be observed that the two fluxes 1 and (p2 traverse the core in such a directon that the voltage induced in the sections 29, 30, 32 and 33 are opposite and, hence, the two high frequency windings 29, 32 and 3U, 33 are in effect magnetically isolated from the windings 28 and 3l The secondary winding sections 39 and 39 are connected so that the flux p1 induces equal and aiding voltages in each of these sections, while the flux (p2 induces canceling voltages. In this manner the two sections 34 and 35 are magnetically isolated from the windings formed by the winding sections 29, 32 and 39, 33.

The Complete circuit diagram for one form of a coupling unit having a transformer such as described above embodied therein is shown in Fig. 6. In the circuit illustrated the transformer is shown as being embodied in a coupling unit for impressing the output oscillations of an electron discharge amplifier 2 on a load illustrated as a resistance The respective primary and secondary winding sections 28 to 35 inclusive are shown diagrammatically as being connected in a manner which will produce the desired ilux interaction as described in the preceding paragraph. In addition, a pair of circuits are provided for further insuring complete isolation of the respective frequency band transmission channels, One of these circuits comprises a condenser 36 connected in series with the primary winding sections 29 and 39 across the primary winding section 28. A resistance 31 is provided for preventing an undesired lowering of the operable primary winding impedance due to series resonance between the winding sections 29, 39 and 58. Similarly, a resistance 49 is connected in shunt with the sections 34 and 35 for the same purpose. The other circuit comprises a condenser 38 connected in parallel with a resistance 39 in series with the secondary winding sections 32 and 33 across the series connected secondary winding sections 3l, 34 and 35.

While the connections shown in Fig. 6, and described in detail hereafter, pertain to a transformer with multi-section windings some of which are effectively paralleled, this is not to be construed as any limitation on the theory of operation of dthe transformers described in connection with the totally series-connected winding transformers illustrated in Figs. 1 to 4 inclusive. No structural changes in the transformers shown in Fig. 5 and Fig'. 14 are involved thereby.

The operation of the coupling unit illustrated in Figs. 5 and 6 will best be understood by reference to Figs. '7 to 13 inclusive wherein I have shown the equivalent circuits for the net work of Fig. 6, when particular frequency bands are considered.

For a band of frequencies extending over the lower portion of the operating frequency range the tightly coupled primary and secondary sections 28 and 3l respectively are effective to provide a transmission channel for currents having frequencies within this portion of the operating range. Over the effective transmission range of these two winding sections the condensers 36 and 38 and the resistance 39 together with the high leakage reactance between the respective primary and secondary winding sections effectively isolate the transmitting channels comprising the sections 519, 32 and 39, 33 and prevent these channels from affecting the transmission efficiency of the channel formed by the sections 28 and 3|. For currents having frequencies eX- tending over this low portion of the operating frequency range the impedance of the secondary sections 34 and 35 is very low and hence the ratio of voltage transformation is effectively determined by the turn ratio between the sections 28 and 3l.

When the range of frequencies referred to in the preceding paragraph is considered, the equivalent circuit for the coupling unit and tube 2' becomes as shown in Fig. '7 wherein e0 represents the voltage available a't the input terminals of the electron discharge device multiplied by the amplification factor of the `discharge device, p represents the internal anode to cathode resistance of the discharge device 2', R is the equivalent resistance formed by referring the resistance 3 to the primary side of the transformer and X1 is the primary magnetizing reactance.

At a slightly higher band of frequencies the leakage reactance between the winding sections 28 and 3l becomes appreciable as compared to R and the equivalent circuit becomes as shown in Fig. 8. In this circuit the reactance :r2 represents the leakage reactance between the sections 28 and 3l. Over this band of frequencies the primary magnetizing reactance of the transformer becomes so large as to have a negligible effect on the operation of the network and, accordingly, it is omitted from the equivalent circuit shown.

At a still higher band of frequencies the equivalent circuit becomes as shown in Fig. 9 wherein c1 represents the capacity of the condenser 36, c2 represents the capacitance of the condenser 38 referred to the primary side of the transformer, the resistance R1 represents the resistance 3l, R2 represents the resistance 39 referred to the primary side of the transformer, and the reactance x3 represents the magnetizing reactance of the winding sections 29 and 3l).k Over this band of frequencies the leakage reactance :r2 is of a magnitudeV such that the voltage drop therethrough tends to lower the voltage across the terminals of the equivalent resistance R. However, the condensers 36 and 38 become effective over this frequency band to open up the transmission channels 29, 32 and 39, 33 thereby to maintain the desired value of impressed voltage across the terminals of resistance R. It will of course be understood that the turn ratio between the sections 29, 32 and the sections 30, 33 must be the same as the turn ratio between the sections 28, 3| if the voltage across the resistance R is to be maintained at the `desired value.

By making the reactance of condenser c1 approximately equal to .r3 in this portion of the frequency range and further by making R1, x3, R2 and wz all equal to each other and greater than R, the network will be broadly tuned, thereby increasing the current flowing in 9:2 and R by an amount sufficient to compensate for the voltage drop produced by the resistance R1. The resistance 31, or in Fig. 9 resistance R1, operates to prevent series resonance, between the capacitive reactance of condenser 36, in Fig. 9, ci and the inductive reactance .ra at any frequency. Similarly the resistance 39, or in Fig. 9 resistance R2, prevents series resonance between the secondary magnetizing reactance of the winding sections 32 and 33 and the low frequency blocking condenser 38.

At still higher frequencies the magnetizing reactance x3 of the network shown in Fig. 9 becomes so large as to have practically no effect on the operation of the circuit. Further, capacitive reactance of condenser C2, becomes so small with respect to the resistance R2 that the latter has little effect on the circuit. Thus, the equivalent circuit for a band of frequencies covering this portion of the operating frequency range becomes as shown in Fig. 10. Partial resonance of the network voccurs over the band of frequencies which is sufficient to compensate for the voltage drop produced by the resistor R2.

At a still higher band of frequencies the equvalent circuit for the coupling unit illustrated in Fig. 6 may be accurately represented by the circuit of Fig. l1 wherein R3 represents the resist- 'ance Ml of Fig. 6, :r4 represents the inductive reactance of the winding sections 3i and 35, C3 represents'the equivalent distributed capacity of the winding sections 2S and 3i referred to the primary windings 29 and 3o and X5 is the leakage reactance between primary windings 29 and 3o and secondary windings 32 and 33. Within this band of frequencies the equivalent distributed capacity c3 of the winding section 3l becomes of sufficient magnitude. to constitute a considerable load across .resistance 3) which tends to reduce the voltage available at the terinlz-ials of the resista-nce or in Fig. l'l.- R. However, at such frequencies the winding sections 34 and 35 shunted by the resistance dii maintain the overall impedance of this branch of the equivalent circuit at a value sufficient to prevent the undesired reduction in voltage. If the circuit elements are selected having the correct impedance values, it is possible to obtain the same voltage transfer ratio in this portion of the operating frequency range as is attained in the lower portions of the range.

In the arrangement of Figs. 5 and 6 it is necessary that the distributed capacity of winding 28 be sufficiently low as not Ito constitute an objectionable load on discharge device 2.

At the lower portion of the frequency range in which coils 34 and 35 are operative the equivalent inductance mi of the winding sections 3d and 35 is substantially equal to the equivalent capacitive reactance corresponding to the dis-- tributed capacity of the winding section 3i. Thus, the apparent resistance R3 corresponding to the resistance lil as viewed from the input terminals of the coupling unit is the impedance which prevents excessive loading of the windings 32. and 33. over this band of frequencies. It becomes. apparent from a consi-deration of this equivalent circuit that the resistances R1 and R3 must be large as compared to the resistance R if a uniform response characteristic is to be obtained over a frequency range including the band at present being considered.

When a still higher band of frequencies is considered the equivalent circuit becomes as shown in Fig. 12 since the reactances represented by C3 and Cae-re now negligible and x4 extremely high.

At a still higher band of frequencies covering a range extending to the upper limit of the operating frequency range the equivalent circuit becomes as shown in Fig. 13. In this frequency band the distributed capacity of the winding sections 29, 3ft, 3.2 and 33 begins to have a loading effect due to the capacitive reactance thereof, which is effectively in shunt to the resistance R. The equivalent capacity representing these components of distributed capacity is represented. by the condenser c4. At higher frequencies the loading effect of this condensive reactance is increasingly greater until a frequency is reached where this reactance is equal to the leakage inductivev reactance frs. 'I'his frequency represents the absolute upper frequency limit for the operating range of the structure shown in Fig. 5 and the voltage ratio characteristic falls off rapidly at frequencies above this value.

It has been found that in certain cases the distributed capacity of the winding 28 tends to overload the source at frequencies extending over the upper portion of the operating range. This undesired effect may be overcome by employing the 4ransformer illustrated in Fig. 14 wherein a pair ol primary winding sections M and 42 are pro vided in addition to the elements of the transformer illustrated in Fig. 5 vthese windings being connected asshown in Fig. 15. The sections 4I and i? are mounted on the outer legs of the transformer cere and are connected to be magnetically isolated from the winding sections 29, 30, 32 and They are designed to have a very low distributed capacity and are shunted by a resistor 43 for preventing series resonance effects between the inductance thereofand the equivalent capacitive reactance formed by the distributed capacity of the winding sections 28 and 3 l. This obviates the necessity of designing winding 28 to have low distributed capacity.

It has been found that a small coupling unit constructed in the manner described above is capable of transmitting electric circuit currents extending over an exceedingly wide band of frequencies as for example, from 20 cycles to 100 kilocycles or even more. By employing the winding section arrangement illustrated and connecting these sections in the manner shown, the respective frequency band transmission channels thus formed may be magnetically isolated thereby to provide a high leakage reactance between the sections included in each channel, without employing a core structure of unduly large size. It will be apparent to those skilled in the art that any number of channels may be provided by an obvious extension lof the structural arrangement and circuit connections described in the preceding paragraphs and the band width may be even further extended.

While I have shown a particular embodiment of my invention, it will of course be understood that I do not wish to be limited thereto since many modifications in the structure may be made, and I contemplate by the appended claims to cover all such modifications as fall within the true spirit and scope of my invention.

What I claim as new and desire to secure by Letters Patent of the United States is:

1. A coupling unit for coupling together two electric circuits for the transmission therebetween A. y

ing the inductive coupling between the primary and secondary sections havingV like resonant frequencies for providing a plurality of channels for transmitting currents having frequencies eX- tending over predetermined portions of said frequency range.

electric circuits for the transmission therebetween of alternating currents having component frequencies extending over a wide range, comprisinga transformer having a core, sectional- 70 2. A coupling unit for coupling together two ized primary and secondary windings on said core, each of the sections of each of said windings having a resonant frequency different from the resonant frequency of each of the other sections of the same winding and substantially the same as that of one section of the other winding, the primary and secondary winding sections having the same resonant frequencies being coupled in close inductive relation thereby to provide a plurality of channels for transmitting currents having frequencies extending over predetermined portions of said frequency range.

3. A coupling unit for coupling together two electric circuits for the transmission therebetween of alternating currents having component frequencies extending over a wide range comprising a transformer having a core, sectionalized primary and secondary windings on said core, each of the sections of each of said windings having a resonant frequency different from the resonant frequency of each of the other sections of the same winding and substantially the same as that of one section of the other winding, the primary and secondary sections having the same resonant frequencies being coupled in close inductive relation thereby to provide a plurality of channels for transmitting current having frequencies extending over predetermined portions of said frequency range, and means to produce relatively large leakage reactance between sections having different resonant frequencies whereby loading of one of said channels due to the low capacity reactance of one of said sections corresponding to another of said channels is prevented.

4. A coupling unit for coupling together two electric circuits for the transmission therebetween of alternating currents having component frequencies extending over a wide range comprising a transformer having a core, sectionalized primary and secondary windings on said core, each of the sections of each of said windings having a resonant frequency different from the resonant frequency of each of the other sections of the same winding and substantially the same as that of one section of the other winding, the primary and secondary sections having the same resonant frequencies being coupled in close inductive relation thereby to provide a plurality of channels for transmitting currents having frequencies extending over predetermined portions of said frequency range and being arranged on said core so that the leakage reactances between sections having different resonant frequencies are sufficient to prevent loading of any one of said channels by the other of said channels, and means for preventing lseries resonance effects between the respective sections of each of said windings from lowering the transmission efficiency of said transformer for currents having frequencies corresponding to the frequencies at which such resonance effects occur.

5. A coupling unit for coupling together two electric circuits for the transmission therebetween of alternating currents having component frequencies extending over a wide range, comprising a transformer having a core and including sectionalized primary and secondary windings, said primary and secondary windings each including a section having a low resonant frequency and a plurality of sections having successively higher resonant frequencies, said primary and secondary windings sections being coupled together in close inductive relation in the order of their resonant frequencies thereby to provide a plurality of separate transmission channels each effective over a predetermined portion of said frequency range, and means including a resistance connected in shunt with each of said high resonant frequency sections for preventing series resonance effects between the respective sections of each of said windings from lowering the transmission eiciency of said transformer for currents having frequencies corresponding to the frequencies at which such resonance effects occur.

6. A coupling unit for coupling together two electric circuits for the transmission therebetween of alternating currents having component frequencies extending over a wide range comprising a, transformer having a core and including sectionalized primary and secondary windings on said core, said windings each including sections having different resonant frequencies and being inductively coupled to provide a plurality of transmission channels each effective over a predetermined portion of said frequency range, and means for preventing series resonance effects between the respective sections of each of said windings from lowering the transmission efficiency of said transformer for currents having frequencies corresponding to the frequencies at which such resonance effects occur.

'7. A coupling unit for coupling together two electric circuits for the transmission therebetween of alternating currents having component frequencies extending over a wide range comprising a transformerhaving a core, sectionalized primary and secondary windings on said core, said windings each including sections having respectively high and low resonant frequencies, said high resonant frequency primary and secondary sections being inductively coupled to provide a transmission channel effective over the upper portion of said frequency range and said low resonant frequency primary and secondary sections being inductively coupled to provide a transmission channel eifective over the lower portion of said frequency range.

8. A coupling unit for coupling together two electric circuits for the transmission therebetween of alternating currents having component frequencies extending over a wide range, comprising a transformer having a core, sectionalized primary and secondary windings on said core, said windings each including sections having respectively high and low resonant frequencies, said high resonant frequency primary and secondary sections being coupled in close inductive relation thereby to provide a transmission channel effective over the upper portion of said frequency range, said low resonant frequency primary and secondary sections being coupled in close inductive relation thereby to provide a transmission channel effective over the lower portion of said frequency range, and means to produce relatively large leakage reactance between said high and low frequency sections whereby loading of said first-named channel by the other of said channels due to the low capacity reactance of said low frequency section in said upper portion of said range is prevented.

9. A coupling unit for coupling togetherY two electric circuits for the transmission therebetween of alternating currents having component frespectively high and low resonant frequencies, said high resonant frequency primary and secondary sections being inductively coupled to provide a transmission channel effective over the upper portion of said frequency range, said low resonant frequency primary and secondary sections being coupled to provide a transmission channel effective over the lower portion of said frequency range, said high and low resonant frequency sections of each of said windings being arranged on said core so that the leakage reactances between said sections are suiiicient to prevent loading of one of said channels by the other of said channels, and means for preventing series resonance effects between the respective sections of each of said windings from lowering the transmission efciency of said transformer for currents having frequencies corresponding to the frequencies at which such resonance effects occur.

10. A coupling unit for coupling together two electric circuits fo-r the transmission therebetween of alternating currents having component frequencies extending over a wide range, comprising a transformer having a core, sectionalized primary and secondary windings on said core, said windings each including sections having respectively high and low resonant frequencies, said high resonant frequency primary and secondary sections being inductively coupled to provide a transmission channel effective over the upper portion of said frequency range and said low resonant frequency primary and secondary sections being coupled to provide a transmission channel effective over the lower portion of said frequency range, said high and low resonant frequency windings being connected to form a series resonant circuit of substantially zero impedance at a frequency Within said range, and means including resistances connected respectively one in shunt with each of said high resonant frequency sections and having a resistance value equal to the reactance value of the shunted section at the frequency where said series resonance occurs for preventing said series resonance effects from lowering the transmission efficiency of said transformer for currents having frequencies corresponding to the frequencies at which such resonance effects occur. n

1l. A coupling unit for coupling together two electric circuits for the transmission therebetween of alternating currents having component frequencies extending over a wide range comprising a coregsectionalized primary and secondary windings on said core, said windings each including sections having respectively high and low resonant frequencies, said high resonant frequency primary and secondary sections being inductively coupled to provide a transmission channel effective over the upper portion of said frequency range, said low resonant frequency primary and secondary sections belng inductively coupled to provide a transmission channel effective over the lower portion of said frequency range, and means for preventing series resonance effects between the respective sections of each of said windings from lowering the transmission efficiency of said transformer for currents having frequencies corresponding to the frequencies at which such resonance effects occur.

HAROLD T. LYMAN.

Images