US3278878A - Leakage reactance transformer - Google Patents

Description

Oct. 11, 1966 A. R. COLLUCCI 3,278,878

LEAKAGE REACTANCE TRANSFORMER Filed Jan. 27, 1964 H604 FIG.5

ANTHONY R. COLLUCCI INVENTOR ATTORN EY United States Patent 3,278,878 LEAKAGE REACTANCE TRANSFORMER Anthony R. Collucci, Ipswich, Mass., assignor to Sylvauia Electric Products Inc., a corporation of Delaware Filed Jan. 27, 1964, Ser. No. 340,397 2 Claims. (Cl. 336--160) This invention relates to apparatus for starting and operating electric discharge lamps, particularly those of the igh pressure mercury type containing added metals and iodine. The apparatus will, however, be useful with other devices which require similar characteristics.

Apparatus for starting and operating discharge lamps is generally called a ballast and includes a transformer and inductive reactance, the latter generally being provided as leakage reactance. The lamp will start on the open circuit voltage of the transformer, and the operating current will be limited by the reactance, causing a voltage drop which lowers the voltage across the lamp. This is desirable, because the operating voltageof the lamp is generally much less than the starting voltage.

Various expedients have been used to increase the ratio of starting voltage to operating voltage, but these have had disadvantages such as producing a poor current wave form or requiring additional components, such as condensers.

With the use of iodine-containing mercury lamps, the problem became greater, because an even higher ratio of peak starting voltage to root-mean-square operating voltage was necessary, and a considerable amount of energy had to be available at the peak of the starting voltage wave. The starting voltage, that is the open-circuit voltage of the transformer, not only had to have a high peak in its wave form, but the peak had to be maintained long enough over the cycle to supply enough energy to heat the electrodes sufiiciently and excite the gas. The peak voltage could not be just instantaneous; it had to supply enough energy to maintain the discharge until proper conditions were established for an arc to occur. This takes longer in an iodine-containing lamp than in my high pressure mercury lamp, which is more easily triggered into operation. I

I have found that the starting of such lamps can be accomplished by the use of a transformer in which a reg-ion of increased flux density as provided in the series magnetic circuit of the transformer core by spacing the ends of at least some of the laminations, at the region where they are ordinarily interleaved, a considerable distance apart, but having some other laminations extend across the gap between the ends of the first mentioned laminations. In this way, the flux density in the portion of the laminations which bridge the gap, is considerably increased, with consequent saturation effects.

However, the exact reason why these staggered gaps increases the ratio of peak open-circuit voltage to rootmean-square (R.M.S.) open-circuit voltage, and provide other desirable effects, is not clear, and I do not wish to be limited to any theory in that respect.

Further objects, advantages, and features of the invention will be apparent from the following specification, taken in connection with the accompanying drawing, in which:

FIG. 1 is a perspective view of a transformer according to the invention;

FIG. 2 is an enlarged longitudinal section of the core in the vicinity of the air gaps;

FIG. 3 is a schematic circuit diagram of the electrical connections;

FIG. 4 is an oscillogram of the wave-form of the opencircuit voltage of the transformer; and

FIG. 5 is an oscillogram of the wave-form erating current in the lamp circuit.

of the op- 3,278,878 Patented Oct. 11, 1966 In FIG. 1, the rectangular core 1 is made of lamina-,

tions bent into the form shown, preferably as shown in! Biggs et a1. United States Patent 3,096,568, issued July 9, 1963. They are made into oppositely-facing U-shaped sets, and joined together as in that application, except for one very important difference: instead of the open ends of the U being wholly interleaved where they join, in the present device only a relatively few strips are interleaved, and most of the strips in each U are spaced apart to form an air gap.

As shown in the comparatively enlarged view of FIG. 2, four strips 6, 7, 8, 9 are spaced from strips 10, 11, 12, 13 to form such an air gap 14, and then the ends of strips 15 and 16 meet in a butt joint 17 between strips 13 and 19 beyond the air gap. Strips 21, 22, 23, and 24 are spaced from strips 19, 25, 26, 27 to form air gap 28, which is bridged by strips 29 and 30 which meet in a butt joint 31 beyond the air gap 28. Succeeding sets of strips can follow the same pattern, until the desired thickness of core is achieved. In this way an air gap is provided and bridged by a few strips such as 29, 30 which will not only operate at high flux density in the saturation range but will also form a structure which will hold together mechanically because of the interleaved bridging strips.

Returning now to FIG. 1, the rectangular core 1 is seen to have four legs 3, 4, 5, 2. A primary coil 40 and a secondary coil 41 encircle leg 3, being spaced apart on that leg, which is much longer than legs 2 and 4. Similarly, primary coil 42 and secondary coil 43 encircle leg 5, on which they are placed and spaced apart.

The magnetic shunts 44 and 45 extend from leg 3 to leg 5 outside core 1 and between primary coils 40, 42, and the secondary coils 41, 43, the strips or laminations 46 running perpendicularly to the side of the core, as in copending US. patent application Serial No. 119,495, filed June 26, 1961, now abandoned, by Sheppard Cohen, and assigned to the same assignee as the present application.

The ends 47, 48 of the laminations of shunt 44 are welded in a line about inch wide across the ends of the strips as shown by the weld 49, and an additional weld 50 is similarly made on the side 51 of the shunt, and a hole drilled close to the weld about 43 inch from the weld in the specific example described. A bolt 53, of which head 52 is shown, is passed through the hole and secured on the other side of shunt 45, which is similar to shunt 44 and similarly drilled by a nut. In the example described, bolt 53 can be a /2 inch bolt. The shunts 44 and 45 are composed of a series of fiat rectangles of transformer steel attached as shown. They do not penetrate into the opening, generally called the window, in the rectangle of the core 1.

The air gaps 14, 28 and so forth, as shown in FIG. 2, are in the core at about the middle of the lengths of secondary coils 41, 43. This means that the U-shaped stack of strips having leg 2 as the bottom of the U is shorter than the stack having leg 4 at the bottom of its U.

The connection of the coils to the lamp are shown in FIG. 3. The two primaries 40, 42 are shown in parallel; in the transformer of the specific example they would ordinarily be connected in parallel for 1l5-volt operation, in series for 230-volt operation. Both secondaries 41, 43, would be connected in series, in either case, and they would also be in series with the two coil primary in the usual auto-transformer manner. The transformer is de signed for a frequency of 60 cycles per second. A condenser 60, of about 28 microfarads in the example de scribed, is also in series with the lamp 61, which is preferably a high-pressure arc discharge lamp with a filling of mercury vapor, iodine vapor, and other metals, for example, thorium.

In an example of the invention, primary coils 40, 42 were each of 378 turns of #16 B. & S. enamelled copper wires, there being 38 turns per layer. Secondary coils 41, 43 were of 275 turns of #19 B. & S. wire of the same kind, with 20 turns per layer. If these coils are connected as in FIG. 3, this gives an auto-transformer turns ratio \of 1031/275, and since the open-circuit peak voltages in ithis transformer are found to correspond closely with the ,turns ratio although the operating voltage is much lower, than for a primary R.M.S. voltage of 120 volts, the primary peak will be about 169 volts, and the secondary peak about 650 volts, as indicated in FIG. 4. The operating voltage of the transformer will be much lower, about 290 volts.

In the specific example, the outside dimensions of the main core 1 are about inches along leg 3, and about 2 5 along leg 2, and about 2% inches width of each strip. The legs are A inch thick from the outside strip to the inside strip. The bridged air gaps are about 1% inches from the outside end of the core, and the strips are welded together in positions about /2 inch on either side of each air gap. The welds like those on the shunts 44 and 45 are bead-welded to a depth into the core of about inch. The legs include about 44 strips each of about #26 gauge, that is, each being about 18 /2 mils thick, about 8 of the laminations bridging a W inch airgap to bridge it across, as previously stated. The exact number of laminations may vary slightly because of diflferences in the thickness of the commercial strip from piece to piece. In this specific example, at least 8 of the laminations had to be interleaved to give the right transformer characteristics.

A metal band will generally be fastened, by welding or in some other secure manner, to the outside of the core.

The steel strips of which the core is made should be of silicon steel as used for transformers, the silicon contact generally being about 4%.

In this application, the words strip and laminations are used interchangeably since the laminations of the core are steel strips. The words airgaps and-gap are used interchangeably, since a gap in the magnetic core is customarily called an airgap, even if it is filled with nonmagnetic material, such as fibre or a resin.

After the transformer is assembled, it can be impregnated or coated with a protective varnish in the usual manner.

What I claim is:

1. A transformer having a four-legged main magnetic core, the legs arranged to form a rectangle having a window, two of the legs being long and two short, a primary coil and a secondary coil on each long leg, the secondary coils being connected in series aiding the primary and secondary coils on each leg being spaced apart, a magnetic leakage shunt extending from one long leg to the other in the space between coils, and a partially bridged gap in the portion of each long leg inside a secondary coil, the gap being formed by laminations whose ends are spaced apart from each other and bridged by laminations which extend across the gap to meet corresponding laminations in a butt joint beyond the air gap and interleaved with the other laminations.

2. The transformer of claim 1, in which the number of interleaved laminations which meet in gaps is about 20% of the number which meet in butt joints.

References Cited by the Examiner UNITED STATES PATENTS 2,350,029 5/1944 Glass 336-212 X 2,353,511 7/1944 Short 336-234 X 2,550,501 4/1951 Sims 336--178 X 2,553,591 5/1951 Kronmiller 336165 X 2,553,596 5/1951 Mann 336-165 X 2,762,988 9/1956 Pomazal et a1 336212 X 2,830,277 4/1958 Kane 336234 X 2,850,708 9/1958 Antalis 336 X 3,128,443 4/1964 Herman et a1 336- X LEWIS H. MYERS, Primary Examiner.

ROBERT K. SCHAEFER, Examiner.

T. J. KOZMA, Assistant Examiner.

Claims (1)

1. A TRANSFORMER HAVING A FOUR-LEGGED MAIN MAGNETIC CORE, THE LEGS ARRANGED TO FORM A RECTANGLE HAVING A WINDOW, TWO OF THE LEGS BEING LONG AND TWO SHORT, A PRIMARY COIL AND A SECONDARY COIL ON EACH LONG LEG, THE SECONDARY COILS BEING CONNECTED IN SERIES AIDING THE PRIMARY AND SECONDARY COILS ON EACH LEG BEING SPACED APART, A MAGNETIC LEAKAGE SHUNT EXTENDING FROM ONE LONG LEG TO THE OTHER IN THE SPACE BETWEEN COILS, AND A PARTIALLY BRIDGED GAP IN THE PORTION OF EACH LONG LEG INSIDE A SECONDARY COIL, THE GAP BEING FORMED BY LAMINATIONS WHOSE ENDS ARE SPACED APART FROM EACH OTHER AND BRIDGED BY LAMINATIONS WHICH EXTEND ACROSS THE GAP TO MEET CORRESPONDING LAMINATIONS IN A BUTT JOINT BEYOND THE AIR GAP AND INTERLEAVED WITH THE OTHER LAMINATIONS.

Images