US2542915A - Transformer - Google Patents

Description

M. G. FAVRE Feb. 20, 1951 TRANSFORMER 3 Sheets-Sheet 1 Filed April 4, 1945 M. G. FAVRE TRANSFORMER Feb. 20, 1951 5 Sheets-Sheet 2 Filed April 4, 1945 Feb. 20, 1951 FAVRE 2,542,915

TRANSFORMER Filed April 4, 1945 s Sheets-Sheet :5

Patented Feb. 20, 1951 TRANSFORDIER Marcel Georges Fayre, Wettingen, Switzerland, assignor to. Patelhold Patentverwertungs- & Elcktro-Holding A.-G., Glarus, Switzerland Application April4, 1945, Serial No. 586,611 Iii-Switzerland. April13, 1944 4 Claims.

It is known to usetransformers with a primary winding divided into two symmetrical halves for the passage of modulator stages in push-pull connection to high frequency stages, the centre of the primary winding and one end of the secondary winding being. at the same potential, for instance earth potential. Such an arrangement is shown in Fig. l of. the accompanying drawing where m, az are the two tubes of a low-frequency push-pull class B amplifier and b is the modulation transformer by means of which the anode voltage of transmitting tube is varied in rhythm with the low frequency. Choke coil and condenser q; serve to relieve themodulation transformer b'from the direct anode current of stage 0. By. this means a transformer 12 with minimum leakage is. obtained. Due to, the in complete coupling of the two primary winding halves the capacitancesdistributed between the primaryand secondary side cause an unsymmetrical leading of both primary winding halves at high. frequencies.

A screen is generally arranged. between the primary and secondary side for the purpose of obtaining symmetry; Such screening; increases the leakage, however, as well as the capacitance and size of the transformer. It can only be employed with transformers for small powers where the leakage can be kept sufiiciently low by using an iron core made of special nickel alloy laminations. With power transformers such as modulation transformers. where the iron core can consist only of silicon laminations the increase in leakage and capacitance due to the increase in the dimensions is so great that the frequency range is greatly restricted.

In order to reduce this assymetry without.

screening, it has been proposed to employ additional capacitances and windings and particularly various arrangements of transformer windings. By means of the Iastmamed means the leakage is reduced but at thesameetime the ca:

pacitance, increased. Furthermore the proposed.

winding arrangements are generally very complicated.

The present invention concerns a. transformer for passing from. a symmetrical system to. an U11." symmetrical one. and vice versa, where the centre of. the. winding connected to the symmetrical system and one end of the. winding connected to the unsymmetrical system. are at the same sotential, for. instance earth potential.

According to the invention at. least. parts of both windings of the transformer serve for a mutual. screening of both transformer sides.

Constructional examples of the invention are. illustrated in the accompanying drawings which refer to a transformer for passing from a symmetrical system to an unsymmetrical. one.

In the drawings,

Fig. l is a schematic View of a prior art transformer arrangement; Fig. 2 is a. schematic view of a transformer embodying this invention; Figs. 3a-3b arev graphs relating to the transformer arrangement in Fig. 2; Figs. 4-6 show modified forms of the invention; Figs. Tat-7b are graphs relating to the transformer of Fig. 6; Fig. 8 is a view showing the core and coil arrangement of the transformers of Figs. 5 andv 6; Figs. 9-15 show still other modifications of the invention; and Fig. 16 illustrates a typical. application for a transformer embodying the invention;

In the arrangement shown in Fig. 2 the part of the winding. on the primary side P serving as a screen consists of two equal winding sections (11, d2 each connected to the earthed centre point. Thescreening winding portion e of the secondary side S is connected direct to the earthed end of the latter. The rectangles ,f and g indicated by broken lines and representing those portions of the primary and secondary winding respectively which are not concerned with the screening, are each arranged on the side of the screening wind ing parts d1, (Z2 and e which is furthest removed from the other transformer winding. The nume ber of ampere turns of the winding parts d1, (Z2 and e are so selected that between the latter a partial transmission ratio exists which is at least approximately 1:1. The potential distribution in the winding parts d1, d2 and e is indicated by the potential values shown in Fig. 2 where U is the potential difference between the ends of one of the winding portions d1, d2. The diagrams (Figs. 3a., 31)) show the potential distribution in the part windings d1, d2 and e for an increasing number of turns Z. The straight lines 71.1, 71.2 indicate the potential gradient in the part windings d1. (Z2 and the, straight lines 2'1, ii the potential. gradient. in the part winding e, the index I referring to a positive and the index 2 to a negative transformer alternating field. The parallel course of the straight lines hi, ha and i1, i2 shows that with transformers according to the invention asymmetry is completely avoided.

With the arrangement shown in Fig. 4 the screening part of the winding 6 is divided into two sections e1, c2 and the winding sections d1, 2, e1, 62 are so arranged that there is a potential distribution as indicated by the potential values shown in Fig. 3..

Fig, Eshows an arrangement of a transformer according to the invention with a prescribed transmission ratio of 1:721, such as occurs with modulation transformers used with such an ar-- range-merit as is illustrated in Fig. 1; f1 and f2 are the sections of the primary winding which are not concerned with the screening. 0n the sec-- ondary side S the screening is undertaken by the entire. winding, e. The arrangement shown in Fig. 6 differs from that of Fig. 5 only in so far as the four primary winding sections d1, dz, f1, fz are all selected to be of equal magnitude. This simplifies the construction of the transformer. In this case there is, however, a slight asymmetry, as can be seen from the potential values given in Fig. 6 and the diagrams of Figs. 7a and 7b. The reference letters of Figs. 7a and 7b correspond to those of Figs. 3a and 3b.

The arrangements shown in Figs. 5 and 6 are specially suitable for output transformers for push-pull class B amplifiers according to Fig. 1. With such output transformers it is particularly important that both halves of the primary winding should be closely coupled, so that the higher harmonics of the anode current distort the anode alternating current as little as possible. With the arrangements shown in Figs. 5 and 6 the coupling between the primary winding halves is as close as is required for push-pull class B amplifiers in order to avoid distortions. To avoid the occurrence of smaller unsymmetries care must be taken that the oscillation circuits formed by the winding sections d1, dz, f1, I'z and their parallel capacitances are all tuned to the same resonance frequency.

Fig. 8 shows in longitudinal section the winding system of the transformers used for the arrangements according to Figs. 5 and 6. The

broken lines I01, 702 indicate the limbs of the iron core and d1, dz, 62, f1, f2 are the tubular part windings of the transformer. The secondary winding e is divided into two sections c1 and 62, one on each core.

For a fine tuning of the oscillation circuits formed by the winding sections (21, d2, f1, fz and their parallel capacitances, adjustable condensers pl l can be located between the centre of the primary winding and the ends of the winding sections d1, dz, f1, f2, as shown in Fig. 9.

When the measures according to the invention are adopted the self-capacity of both halves of the symmetrical winding is to a great extent equal, as has been proved by tests. At the same time the leakage inductances of both winding halves are equal with reference to the secondary winding.

These two properties, namely equal leakage inductance and to a great extent the same selfcapacity are necessary conditions for distortionless operation, particularly when the winding halves have to serve as part of the anode circuit of a push-pull class B amplifier according to Fig. 1. Leakage inductance and self-capacity should furthermore be as small as possible. Distortion can also be reduced by keeping the leakage from one anode winding to the other anode winding as low as possible.

Both halves of the winding associated with the symmetrical system can therefore also be divided into winding sections and so arranged that successive winding sections starting from the windin centre are alternately next to and removed from the winding of the unsymmetrical system. The winding sections adjoinin the winding of the unsymmetrical system thus screen this latter from the more distant winding sections.

Fig. 10 shows a winding arrangement which is derived from the arrangement in Fig. 2. It differs from the latter in so far as the primary winding sections 111 and dz are interchanged. With this arrangement the voltage distribution is as shown in Fig. 11. The difference between this voltage distribution and that shown in Fig.

i- 3a is that in with the former there is a voltage jump.

Fig. 12 shows a winding arrangement with a two-fold division of the primary winding. It has a great similarity with the arrangement of Fig. 5, with the difference, however, that the primary winding halves are interchanged. One half (11 of the primary winding comprises the part winding d1 adjacent to the winding e and the part winding d1 which is displaced and furtherremoved from the winding e. The winding halves dz and dz" are arranged correspondingly. Tests have shown that this arrangement of the windings is particularly favourable for a symmetrical distribution of the capacitances. Also in this case small deviations from the theoretical optimum transmission ratio of 2 1:1 are admissible without etrimental chan es in the distribution of the capacitances becoming noticeable.

The arrangement shown in Fig. 13 refers to an extension of the idea which forms the basis of the arrangements according to Figs. 10-12. Here the winding is subdivided into still more sections. By this means corresponding parts d1 di" and dz dz"" of both winding halves are screened or capacitively coupled to the same extent with regard to the unsymmetrical winding e. This subdivision is, how-- ever, limited by practical considerations.

With an especially fine tuning it is also in this case possible to arrange equalizing capacitances parallel to the individual part windings. Fig. 14 shows the arrangement of such condensers parallel to the winding parts. It is not necessary that each winding section must have an additional capacitance. This depends on the construction and the quality of the required symmetry.

Fig. 15 shows the division of the windings of a transformer whose windings are distributed over two cores. Also in this case care is taken that the part windings of each winding half lie alternately directly opposite the secondary winding and further away from the latter.

The transformer according to the invention can also be used when passing from an unsymmetrical to a symmetrical system. In the arrangement shown in Fig. 16 the transformer serves for connecting an unsymmetrical amplifier stage m to the next push-pull amplifier stage n.

The invention also covers high frequency transformers without iron cores.

I claim:

1. In a transformer for use in passing from an electrically symmetrical system to a nonsymmetrical system and vice versa, a core, primary and secondary windings arranged concentrically on said core and wound in the same direction and adapted to be connected respectively to the two systems, the mid-point on said primary winding being at the same potential as one end of said secondary winding and which point divides said primary winding into two halves, each half of said primary winding beicomprised of at least two cylindrical winding sections connected in series one of which lies adjacent said secondary Winding and the other spaced radially from said secondary winding by a winding section of the other half of said primary winding, all winding sections of each half of said primary winding being distributed longitudinally of said core and alternately in two radially spaced primary winding cylinders and said secondary winding being constituted by a third winding cylinder spaced radially from said primary winding cylinders.

2. A transformer as defined in claim 1 and which further includes condensers for balancing connected in parallel to the winding sections of said primary winding.

3. A transformer as defined in claim 1 wherein one end of said secondary winding and the outer ends of the winding sections of said pri- 10 mary winding adjacent said secondary winding are grounded, and the inner ends of the winding section of said primary winding adjacent said secondary winding are cross-connected with the inner ends of the other two winding sections of said primary winding.

4. A transformer as defined in claim 1 wherein one end of said secondary winding and the inner ends of the winding sections of said primary windin adjacent said secondary winding are grounded, and the outer ends of the winding section of said primary winding adjacent said secondary winding are cross-connected with the outer ends of the other two winding sections of said primary winding.

MARCH; GEORGES FAVRE.

REFERENCES CITED The following references are of record in the file of this patent:

UNITED STATES PATENTS Number Name Date 1,559,850 Casper Nov. 3, 1925 1,534,086 Scriven Apr. 21, 1925 2,013,140 Friis Sept. 3, 1935 2,272,452 Whittle Feb. 10, 1942 FOREIGN PATENTS Number Country Date 760,016 France Dec. 1933 212,712 Switzerland Mar. 17, 1941 868,091 France Dec. 15, 1941

Images