US2702935A - Method of forming transformer cores - Google Patents

Description

March 1, 1955 w. D. KYLE, JR

METHOD OF FORMING TRANSFORMER CORES 3 Sheets-Sheet 1 Filed Oct. 13, 1951 INVENTOR.

MZZi am D. K Ze .71. BY ,qzzorn March 1, 1955 w KYLE, JR

METHOD OF FORMING TRANSFORMER CORES 3 Sheets-Sheet 2 Filed Oct. 13. 1951 llllll ll..- l I I i I lh l l l lH H L INVENTOR.

Williqm D. [1 7] J1: 1 flZZor-ne y March 1, 1955 w D; L JR 2,702,935

METHOD OF FORMING TRANSFORMER CORES Filed Oct. 13, 195] 3 Sheets-Sheet 3 INVENTOR.

Wig; m 12 @z ur. BY ,wfl/

AZZorr-nzy United States Patent a William D. Kyle, Jr., Milwaukee, Wis., McGraw Electric Company, Milwaukee, poration of Delaware Application October 13, 1951, Serial No. 251,208 3 Claims. (Cl. 29-15557) assignor to Wis., a cor- This invention relates to a method of winding magnetic cores and is particuarly directed to winding cores for a transformer.

Great strides in the making of better transformers was possible with the advent of silicon steel, principally with the later invention of cold-rolled oriented silicon steel. Although the greatest advantage of its use was due to the reduction of hysteresis loss, ultimate gains of total losses have been limited by the eifect of the eddy current loss which has been reduced in the past by using the material in thin gauge form and in providing a suitable coating of high electrical resistance on each lamination.

Another advance in transformer design was made with the round-wound shape, -i. e. it is formed of a continuous spiral ribbon of flux conducting metal. This provided a continuous core circuit with virtually no airgaps, thus making it a good conductor of magnetic flux.

Replacing hot-rolled steel with cold-rolled oriented steel reduced the excitation characteristic. This made it possible to obtain a much higher flux density with the same magnetomotive force resulting in reduced amounts of steel and copper.

Certain steps in the steel manufacturing process tend to orientate the steel, but the orientation and ultimate crystalline grain growth is not fully obtained until the final high-temperature anneal. To further improve the transformer core, an annealing process is used in order to relieve all strains imposed by the winding process and to promote, where possible, the growth and orientation of the crystalline grains in the steel.

Generally, the best transformer cores are made in two ways, i. e. the core is wound of a continuous spiral ribbon which has had its final high temperature anneal before winding of the core. After this core has been wound, it is given a low temperature anneal to relieve stresses, as any stress produced in the winding operation deleteriously affects the electrical characteristics of the core. In the foregoing, the problem of adhesions, i. e. flux short circuiting is not of much moment. But, the core material is, as it is expensive, very hard to procure and must be handled very carefully in order to avoid setting up stresses in the winding operation. Nevertheless, very fine transformers are provided by this method.

In the other method, the ribbon which does not have the high temperature anneal treatment is wound in a core shape and then the core is subjected to the high temperature anneal treatment to produce the best electrical characteristic in the core. Here, the problem of adhesions arises as the flux short circuits from layer to layer to decrease the efiiciency of the core and its transformer. It is very diflicult to avoid the problem of adhesions in this later method and at the same time avoid setting up stress subsequent to the high temperature anneal treatment. The crux of this invention avoids the bad features of both methods and produces a core which is as good or superior to those produced by the other methods.

It can be seen from the foregoing inventive improvements that whatever can be done to prevent eddy current from forming anywhere in the core is an inestimable contribution to the making of transformers and magnetic cores.

I, therefore, offer this improved method of winding a magnetic core which has for its principal object to pro- 2,702,935 Patented Mar. 1, 1955 vide a means of breaking loose any adhesions that might occur during the high temperature annealing process.

Another object of my invention is to provide a means of winding a core whereby the probability of imperfection or of unevennesses being adjacent to each other is lessened.

A further object of my invention is to abolish adhesions which might form in the core during annealing.

Another object is to avoid setting up unnecessary stresses subsequent to the high temperature anneal treatment.

Other objects will occur from time to time in the course of the specification and claims.

I illustrate my invention more or less symbolically in the accompanying drawing in which:

Fig. 1 is a sectional view of a core showing the spiral wind.

Fig. 2 is an elevation of Fig. 1.

Fig. 3 is a sectional view of a core after the spiral has been flattened. v

Fig. 4 is an elevation of core being wound on a mandrel with a screw lead causing the strip to be wound spirally.

Fig. 5 is an elevation with a part cut away, showing the spiral winding formed or held by a cone shaped seat and placed in an oven to be annealed.

Fig. 6 is an elevation of the spirally wound core in the process of being collapsed between two plates.

Fig. 7 is an enlarged section of the spirally wound strips showing adhesions that might occur during annealing.

Fig. 8 is an enlarged section of the wound core after the adhesions have been broken.

Like reference figures illustrate like parts throughout the specification and claims.

Figs. 1, 2, and 4 illustrate the manner in which I make a magnetic core 1 by winding a coil 2 around a mandrel 3 so that each successive layer of coil material is moved progressively downwardly achieving a lead effect of the spiral. This can be done on a mandrel 3 with a screw lead 4 moving the coils sufficiently to form a spiral as illustrated in Fig. 4. However, any other practical method may be used to achieve this effect.

While in this spiral form, the coil is subjected to a high temperature annealing process. This is shown more or less symbolically in Fig. 5. The coil 2 can be left on the mandrel during annealing or placed on a cone shaped form 5 when placed in the annealing furnace.

Another method of achieving the spirally wound form of the core is to wind the core laminations directly on top of the one preceding it so that the sides are flat, placing the flat core on the cone shaped form 5 in the annealing furnace 6 so that the application of high temperatures during the anneal will cause the laminations to slump and form a spiral shape when the slumping %amina5tions are stopped and supported by the cone shaped orm This annealing process is an important stage in the making of transformers. The inherent properties of orientation and large crystalline grain growth in the steel do not mature until the steel is subjected to the high temperature final anneal. However, high temperatures are apt to cause adhesions 7 as shown in Fig. 7. Adhesions are a merging or melting together of adjacent layers of coil 2. They are extremely harmful to transformer or other type cores because they constitute a shorted path between laminations which permits higher eddy current losses in the core.

To overcome the danger of adhesions, I devised this inventive method of making a core. After the annealing has taken place, the core 1 is subjected to some pressure means such as two plates 8 shown in Fig. 6. The plates 8 are pressed toward each other causing the spiral form to collapse.

In Fig. 6 the spiral is partially collapsed and in Fig. 3 the coils are shown in their completely collapsed position.

This collapsing of the spiral form means that each layer is moved in parallel relationship from an over-lapping position to a completely adjacent position. This slight movement is sufficient to break any adhesions between layers that might have occurred during annealing. What imperfections or unevenness there might be, disappear or are reduced in magnitude when the layers are moved in relationship to adjacent layers, thus minimizing the possibility of increased eddy current losses.

Figs. 7 and 8 illustrate in enlarged form what happens to the adhesions 7 when the layers are moved from the overlapping position to the adjacent position. The adhesions 7 split in line with the coil edge and each half 7-a and 7-b move away from each other. This breaks the connection or short circuit between adjacent layers, and allows the high resistance coating of one layer to become adjacent to the former adhesion and thus prevents the short circuit. This will, therefore, limit the eddy current loss to the minimum set by the thickness of the steel and the resistance of the coating so that full advantage can be taken of the low hysteresis loss inherent in cold-rolled oriented silicon steel.

I claim:

1, A method for forming a core for an electro-magnetic apparatus comprising spirally winding 'a continuous strip of magnetic material on a mandrel, extending a part of each turn of said spiral beyond the edge of the preceding turn, maintaining said spiral so that there is no possilbe movement between layers, subsequently annealing the core at a temperature sufiicient to produce grain growth and orientation, which temperature causes adhesions between layers of the strip, collapsing said core form until each layer is conterminous with the next layer, and maintaining said core so there is no movement between layers, whereby an electro-magnetic core is formed without adhesions and the formerly adhered areas are non-adjacent in said core.

2. A method for making a transformer core comprisupon formerly ing spirally winding said core from continuous coil strip in such manner that adjacent layers overlap, subsequently annealing the core at a temperature sufficient to produce grain growth and orientation, which temperature causes adhesions between the layers of the strip, moving all adjacent layers so that a tightly wound core is produced having all edges of the laminae lying in the same plane, and maintaining said core in a tightly wound position, thereby precluding movement between layers thereof while preventing flux short circuiting because formerly adhered areas are now non-adjacent.

3. A method for making a wound core for an electromagnetic induction apparatus comprising the steps of winding a coil from continuous strip into conterminous layers, placing the coil on a conical form with the axis of the coil aligned with axis of the form, subsequently annealing the core at a temperature suflicient to cause slumping of the core into a conical spiral, such temperature also being suflicient to produce grain growth and orientation, which temperature causes adhesions between layers of the strip, collapsing said coil after annealing whereadhered areas are moved non-adjacent, and maintaining said core in a tightly wound position, thereby precluding movement between adjacent layers in the final core.

References Cited in the file of this patent UNITED STATES PATENTS 2,157,050 Bilger May 12, 1939 2,260,398 Otte Oct. 28, 1941 2,387,099 Vienneau Oct. 16, 1945 2,390,937 Holland Dec. 11, 1945 2,406,130 Boyce Aug. 20, 1946 2,544,871 Wiegand Mar. 13, 1951

Claims (1)

1. A METHOD FOR FORMING A CORE FOR AN ELECTRO-MAGNETIC APPARATUS COMPRISING SPIRALLY WINDING A CONTINUOUS STRIP OF MAGNETIC MATERIAL ON A MANDREL, EXTENDING A PART OF EACH TURN OF SAID SPIRAL BEYOND THE EDGE OF THE PRECEDING TURN, MAINTAINING SAID SPIRAL SO THAT THERE IS NO POSSIBLE MOVEMENT BETWEEN LAYERS, SUBSQUENTLY ANNEALING THE CORE AT A TEMPERATURE SUFFICIENT TO PRODUCE GRAIN GROWTH AND ORIENTATION, WHICH TEMPERATURE CAUSES ADHESIONS BETWEEN LAYERS OF THE STRIP, COLLAPSING SAID CORE FORM UNTIL EACH LAYER IS CONTERMINOUS WITH THE NEXT LAYER, AND MAINTAINING SAID CORE SO THERE IS NO MOVEMENT BETWEEN LAYERS, WHEREBY AN ELECTRO-MAGNETIC CORE IS FORMED WITHOUT ADHESIONS AND THE FORMERLY ADHERED AREAS ARE NON-ADJACENT IN SAID CORE.

Images