US3748570A

US3748570A - Adjustable output voltage transformer

Info

Publication number: US3748570A
Application number: US00266185A
Authority: US
Inventors: S Martner
Original assignee: AMCO PROD CO
Current assignee: AMCO PROD CO
Priority date: 1972-06-26
Filing date: 1972-06-26
Publication date: 1973-07-24
Anticipated expiration: 1990-07-24
Also published as: CA982667A

Abstract

An alternating current transformer having an effective variable ratio of output to input voltage is made up of two ferromagnetic cores, one of which is linked by a winding to the input, whereas the second is linked by another winding to the output. A third coupling winding, which is closed on itself through an adjustable impedance, electromagnetically links the two cores. When the impedance in the coupling winding is low, the voltage ratio of the arrangement is essentially the turns ratio of the output to the input windings; when the adjustable impedance is high, the coupling between the two cores is weak and the output voltage is much below its maximum possible value. There is no need for voltage adjustment by other means such as, for example, magnetic saturation of the cores through the interposition of a direct current on one of the windings.

Description

ited States Patent [1 1 Martner 1 July 24, 1973 1 ADJUSTABLE OUTPUT VOLTAGE TRANSFORMER [75] Inventor: Samuel T. Martner, Tulsa, Okla.

[22] Filed: June 26, 1972 [21] Appl. No.: 266,185

[52] US. Cl 323/44 R, 323/50, 323/54,

336/145, 336/182 [51] Int. Cl. H01! 39/00 [58] Field of Search 323/44, 46, 47-50,

FOREIGN PATENTS 0R APPLICATIONS 607,082 12/1934 Germany 323/88 Primary Examiner-Gerald Goldberg Attorney-Paul F. Hawley [57] ABSTRACT An alternating current transformer having an effective variable ratio of output to input voltage is made up of two ferromagnetic cores, one of which is linked by a winding to the input, whereas the second is linked by another winding to the output. A third coupling winding, which is closed on itself through an adjustable impedance, electromagnetically links the two cores. When the impedance in the coupling winding is low, the voltage ratio of the arrangement is essentially the turns ratio of the output to the input windings; when the adjustable impedance is high, the coupling between the two cores is weak and the output voltage is much below its maximum possible value. There is no need for voltage adjustment by other means such as, for example, magnetic saturation of the cores through the interposition of a direct current on one of the windings.

5 Claims, 2 Drawing Figures ADJUSTABLE OUTPUT VOLTAGE TRANSFORMER BACKGROUND OF THE INVENTION 1. Field of the Invention This invention finds one prominent use in smooth and relatively stepless variation in the output voltage of a transformer. In the past, this has been accomplished by such devices as a variable tapped transformer, an autotransformer, or a controlled rectifier in which cur rent is allowed to pass during a selected portion of the voltage cycle. To a certain degree, it has also been accomplished by the so-called saturable core reactor, that is, a reactor or a transformer which carries a direct current as well as an alternating current.- In this case the direct current ampere turns on any coil linking the ferromagnetic core tend to saturate the core of the reactor or the transformer. Due to the saturated core operation, there is a proportionately less variation in the magnetic flux in the core for a given variation in the ampere turns of any alternating current winding. Thus, as the direct current ampere turns linking the coil increase, the effective impedance decreases or, in the case of a saturagle core transformer, the effective turns ratio decreases, which, in effect, reduces the output voltage.

Such arrangements have been widely used but they have certain well-known disadvantages. In the case of a saturable core reactor or transformer, a major disadvantage is the poor waveform of the output voltage due to the saturation effect with the consequent power loss. In the case of an adjustable tapped transformer, including autotransformers, a principal difficulty is the contact troubles that are experienced with the sliding contacts and the consequent undesirable local heat losses at the taps. There is also a possibility of less than desired fine control of the output voltage, which only changes in steps. In the case of a controlled rectifier, a major disadvantage is the poor waveform of the output voltage and current, with the resultant power loss, besides that of the inherent instability of the device.

, I have found that it is possible to have a transformer arrangement which permits a smooth variation of the output voltage with an essentially sinusoidal voltage and current output. There is no necessity for providing rectifiers and the associated control circuits, as is the case with either the saturable core reactors or the phase-controlled rectifiers.

Essentially, the arrangement consists of two separate ferromagnetic cores. One of the cores is electromagnetically linked to the input by the input winding, whereas th'esecond core is electromagnetically linked to the output by the output winding. A third coupling winding links the two cores; its circuit is closed through an adjustable impedance. Accordingly, the magnetomotive force transferred from the first or input core to the second or output core depends upon the value of the impedance included in the third coupling circuit. A low circuit impedance corresponds to a high output voltage and vice versa.

2. Description of the Prior Art A search of the literature has revealed a large num ber of patents superficially related to the arrangement of my invention. A closer study, however, indicates that none of these inventions is, in 'fact, based on my arrangement. It is to be emphasized that my invention in no way makes use of the phenomenon of magnetic saturation in ferromagnetic cores. Accordingly, any arrangement which varies a voltage in an alternating current coil-core system by employing a direct current flowing in a winding magnetically coupled to the core, operates on a completely different principle. This covers such arrangements as the Stoekle U.S. Pat. No. 1,426,123; the Lee U.S. Pat. No. 1,815,516; the Dowling U.S. Pat. No. 1,910,381; the Overbeck U.S. Pat. No. 2,062,037; the Edwards U.S. Pat. No. 2,142,837; the Weis U.S. Pat. No. 2,332,879; the I-Iolt U.S. Pat. No. 2,586,567; the Stammerjohn U.S. Pat. No. 2,843,813; the Kelley U.S. Pat. No. 2,886,769; the Zelina U.S. Pat. No. 2,911,586; and the Manteutfel U.S. Pat. No. 3,037,160.

Nor does the arrangement of my invention necessarily involve a capacitor for the adjustment of the phase of a particular current, as shown in the I-Iaug U.S. Pat. No. 2,441,814, or in the magnetic trigger system of the Walsh U.S. Pat. No. 2,603,771. Another difference between these patents and my invention is that both the Hang and the Walsh patents use really only one ferromagneticcore with three legs, whereas there are two separate cores in my invention. The Haug system also uses an open ferromagnetic core, i.e., one with an air gap, which is not required in my arrangement. In fact, it is this air gap which is adjusted to provide the voltage control, a concept totally different from the one employed in my arrangement. It can also be said that Walshs system does not afford positive control over the output voltage with respect to the input voltage, since it uses a triggering system which involves abrupt change of output voltage, an arrangement completely foreign to the operation of my invention.

Both the Rump U.S. Pat. No. 1,807,797 and the Swiss Pat. No. 320,384 of Licentia use two separate closed ferromagnetic cores and three windings linking the cores, as does my invention. I-Iowever, in both these cases, the arrangement is a current transformer which, by general nature, is intended to have a constant ratio of input to output current for metering purposes. Ac-

cordingly, there is no arrangement for adjusting the output voltage. Physically, both these patents fail to teach the use'of a passive impedance in series with and closing the circuit of the third coupling winding.

The closest prior art found is in FIG. 1 of the Von Ohlsen U.S. Pat. No. 2,001,557. The two separate closed ferromagnetic cores are present, and the input and the output windings are also present, as in my invention. However, the third coupling winding linking both the coils is connected in a closed parallel circuit consisting of an inductance 45 and a condenser 46. This forms a frequency-dependent circuit, that is, a circuit which Van Ohlsen teaches is in parallel resonance at a frequency somewhat above any used in the circuit. In other words, the circuit operates as a frequencydependent kind of amplifier. Van Ohlsen does not teach a passive adjustable impedance in the third winding. Accordingly, he is unable to produce an adjustable output voltage transformer. The arrangement I have invented is, therefore, different in design and function from that of Van Ohlsen. It is more linear over a wider range, and it is frequency-independent.

SUMMARY OF THE INVENTION This invention comprises an alternating current transformer arrangement permitting smooth adjustable variation in the output voltage at any desired load within the allowable heating range of the windings. It employs two separate closed ferromagnetic cores, one of which is linked to the input only by an input winding and the other is linked to the output by an output winding. The coupling between the two cores is provided by a third coupling winding linking the two cores, and

forming a complete circuit through a passive adjustable.

impedance. Preferably, the third coupling winding contains more turns than either of the other two, and also, preferably, the adjustable impedance is a variable resistor of high ohmic value.

BRIEF DESCRIPTION OF THE DRAWINGS The attached drawings form a part of this specification and are to be read in conjunction therewith. In these drawings:

FIG. 1 represents in diagrammatic form one arrangement of a variable ratio transformer in accordance with my invention.

FIG. 2 illustrates in isometric form a toroidally constructed variable ratio transformer, which is a preferred form of this invention.

DESCRIPTION OF THE PREFERRED EMBODIMENT In FIG. 1 there are shown two separate closed ferromagnetic cores 11 and 12. These are typical transformer cores, which, in the case of ordinary power frequencies, would be made of thin, laminated silicon steel sheets stacked together, as has been known for a long time. The design of such cores is by now so well known that no particular description is merited. If this transformer arrangement is to be employed at higher frequencies, appropriate changes in core material may be made to a ferromagnetic material usable at such higher frequencies, and, if low radio frequencies are to be employed, the closed cores would conventionally be made of ferrite. It is important that the cores l1 and 12 be closed, that is, that there be no substantial air gaps in either coil. This requirement insures a close magnetic coupling between the input winding 13 and one part of the third winding 14 on the one side on core 11, and between the secondpart of the third winding 15 and the output winding 16 on the other core 12. Close adjustment is not easily accomplished unless such closed ferromagnetic cores are employed.

The input winding is connected across input terminals 17 leading to a alternating current source, and the output winding across whatever load 18 is to be used.

With this arrangement, the voltage across coil 14 tends to be directly related to the voltage across coil 13 by the mutual turns ratio, as is common in any ordinary power transformer. A similar arrangement holds between the voltage across coil 15 and that across the output winding 16.

It is to be noted that the coils l4 and 15 form the third coupling winding linking the two cores l1 and 12. The circuit, which includes this third winding, is completed through a passive adjustable impedance 19. Therefore, the voltage across the points is given, simply, by the product of the current flowing in either of the

coils

14 or 15 and the impedance of the adjustable impedance 19.

Since there is a voltage drop across the terminals 20, it is clear that the voltage across coil 15 is not identical to that generated across coil 14, the latter being the higher. If impedance 19 is adjusted to be essentially zero, the voltage applied to coil 15 is essentially that generated across coil 14, and the maximum possible voltage is present across the output coil 16 and hence across the load 18. As the impedance of element 19 is increased, the voltage drop across it increases, and the voltage drop across coil 15 decreases accordingly, decreasing the output voltage across coil 16. Accordingly, the output voltage of this double transformer arrangement can be made to change smoothly from a maximum (for the case of the impedance of element 19 being a minimum) to a minimum (when the impedance of element 19 is a maximum). If the minimum impedance of element 19 is negligible compared to the open circuit impedance of the

windings

14 and 15, the open circuit output voltage is determined essentially by the input voltage and the turns ratio between

coils

16 and 13, as in the case of a conventional transformer, except for small voltage drops in

coils

13, 14, and 15 due to the negligible magnetizing currents in these coils.

The smoothness of the control is determined by the smoothness with which the adjustable impedance 19 can be varied. Thus a continuous control, such as a variable resistor, is more smooth as a control element than a tapped coil or the like. For this and other reasons, I prefer to have the adjustable impedance 19 be a variable resistor. Those reasons are that such an arrangement provides a frequency-independent element, and that a wide variation in impedance is more easily obtained than is possible with either a tapped inductance or capacitance. The advantage of making this circuit relatively frequency-independent is most apparent when the input voltage across terminals 17 contains more than one frequency component. If the output voltage, under these circumstances, is required to have the same waveform as the input voltage, it is necessary that the frequency components should be in the same proportion in the output voltage as in the input voltage. If the adjustable impedance 19 is a resistor, especially with a large ohmic value in comparison with the open circuit impedance of

coils

13, 14, and 15, such a result is essentially insured; this, however, is certainly not the case if a reactance is used. Also, if my circuit is to be used for adjustment of load voltage in a power circuit, a frequency-dependent impedance 19 would serve no useful purpose, since power frequencies are maintained constant to a considerable degree of precision.

With this type of control arrangement, it is important that the currents in the windings be not so large as to saturate either of the two ferro-magnetic cores. This also insures a good waveform at the output of the circuit if there is a good waveform of the input alternating voltage applied across terminals 17.

It is not to be assumed that the variable impedance 19 is necessarily only a manually adjustable one. Certainly this is a very simple arrangement. However, it is equally possible to adjust this resistance by a servocontrol circuit or by other electronic devices. Such circuits have been known for so many years, and have accomplished so many different functions that it appears unnecessary to describe them in detail here. The output voltage may be adjusted to have a flat or rising characteristic as desired. Thus, if it is desired to use this arrangement as a constant output voltage device, it is only necessary to apply the voltage across terminal 16 to the input of any of the. several servo mechanisms which decrease the impedance of impedance 19 when the output voltage decreases, and vice versa. Such an arrangement can work over a wide range of loads, and also for considerable variations in the input voltage.

While not absolutely necessary, I find it desirable to have the number of turns on coils 14 and be essentially identical. With such an arrangement it is desirable to have the number of turns of each of these

coils

14 and 15 be large compared to the number of turns of the input coil 13 and the output coil 16. While the turns on

coils

14 and 15 should be at least equal to the lesser of the number of turns of the output and

input windings

16 and 13, it is preferable that the number of turns of this third coupling winding be considerably greater than the number of turns of the input winding 13 and of the output winding 16. The reason is that the impedance of element 19 for a specified range of control is directly related to the number of turns in coils l4 and 15. A high impedance can be used if the number of turns on the coils l4 and 15 is large. The power dissipated in element 19 decreases as this impedance increases, and, therefore, the desired variation in output voltage can be obtained with a lower power dissipation in the adjustable passive impedance 19. Thus I prefer to have the number of turns of the third coupling winding to be at least five times greater than the larger of the number of turns of the input and the output windings. For the same reason I prefer to have the variable resistance of element 19 have a maximum value of at least the open circuit impedance of this third coupling winding.

The ratio of maximum to minimum resistance of element l9 depends, of course, on the degree of control desired on the output voltage. A ratio of maximum to minimum resistance of at least 10 permits a wide variation in the output voltage of the transformer, particularly when the maximum resistance of this element has been chosen as stated immediately above.

One particular form of this adjustable output voltage transformer is especially desirable. This is shown in FIG. 2. Here the two magnetic cores 11 and 12 are essentially identical in shape and size, so that, after the input winding 13 and the output winding 16 have been placed on the respective cores 11 and 12, the third winding coupling the two cores can be just a single winding'wound about both the magnetic core structures. In this form, a particularly suitable shape for the magnetic cores is the toroidal one. Being essentially of the same dimensions, these cores can be fixed coaxially adjacent to each other, asshown in FIG. 2, and-the third winding 21. (which replaces coils l4 andlS) can be helically wound about the adjacent cores. In this arrangement, essentially all of the magnetic field of each core is confined within the respective windings. Thus there is very little leakage flux from either core 11 or core 12, which, besides superior performance characteristics, means very little magnetic energy loss due to coupling of the leakage flux with any external circuit. This, of course, includes any conducting material near the coils. Also, for precisely the same reason, there would be very little pickup or voltage induced in the transformer from extraneous magnetic fields. Thus there would be little cross feed into the load or the output terminals. Obviously, this minimization of magnetic field linking is particularly important in communications circuits and the equivalent. Incidentally, in this drawing the adjustable passive impedance is shown as the adjustable resistor 22.

I claim:

1. An adjustable output voltage transformer comprising:

1. Two separate closed ferromagnetic cores,

2. An input winding linking only one of said two cores,

3. An output winding linking only the other of said two cores,

4. A third coupling winding linking both said two cores, and

5. A passive adjustable resistance in series with and closing the circuit with said third winding;

the currents in said windings being essentially only alternating in nature and without appreciable direct current, said currents being insufficient to saturate said cores.

2. An adjustable output voltage transformer in accordance with claim 1 in which the number of turns of said third coupling winding on each of said cores is at least equal to the lesser of the number of turns of said input and of said output winding.

3. An adjustable output voltage transformer in accordance with claim 2 in which the number of turns of said third coupling winding is considerably greater than the number of turns of said input and of said output winding.

4. An adjustable output voltage transformer in accordance with claim 2 in which the number of turns of said third coupling winding is at least five times greater than the larger of the number of turns of said input and of said output winding and said variable resistor has a maximum resistance at least equal to the open circuit impedance of said third coupling winding and a ratio of maximum to minimum resistance of at least l0.

5. An adjustable output voltage transformer in accordance with claim 3, in which the cores are toroidal in shape and are at least approximately of the same dimensions, said cores being fixed coaxially and adjacent each other, whereby the magnetic field in each core is maximized and the leakage flux is minimized.

Claims

1. An adjustable output voltage transformer comprising:

1. Two separate closed ferromagnetic cores,

2. An input winding linking only one of said two cores,

3. An output winding linking only the other of said two cores,

4. A third coupling winding linking both said two cores, and

4. An adjustable output voltage transformer in accordance with claim 2 in which The number of turns of said third coupling winding is at least five times greater than the larger of the number of turns of said input and of said output winding and said variable resistor has a maximum resistance at least equal to the open circuit impedance of said third coupling winding and a ratio of maximum to minimum resistance of at least 10.

5. A passive adjustable resistance in series with and closing the circuit with said third winding; the currents in said windings being essentially only alternating in nature and without appreciable direct current, said currents being insufficient to saturate said cores.