US2452197A

US2452197A - Induction furnace for variable heat patterns

Info

Publication number: US2452197A
Application number: US584201A
Authority: US
Inventors: Theodore R Kennedy
Original assignee: Ajax Electrothermic Corp
Current assignee: Ajax Electrothermic Corp
Priority date: 1945-03-22
Filing date: 1945-03-22
Publication date: 1948-10-26
Anticipated expiration: 1965-10-26

Description

Oct. 26, 1948. T. R. KENNEDY INDUCTION FURNACE FOR VARIABLE HEAT PATTERNS Filed March 22, 1945 FIG-1.

INVENTOR. /2 /mmwf ATTORNEY.

Patented Oct. 26, 1948 INDUCTION FURNACE FOR VARIABLE HAT .PATTERNS Theodore It. Kennedy, Lower Makefield Township,

Bucks` County, Pa., assigner to Ajax Electrothermic Corporation, Trenton, N. J., a corporation of New Jersey Application March 22, 1945, Serial-N0. 584,201

2 Claims. 1 The. invention herein described deals with high frequency induction furnaces Yfor the heating and melting of metals and has todo with methods and. means for controlling the pattern or degree of heating over selected portions of a charge piece within such a furnace.

An .object of the invention is to provide a l simple and eflcient method and means for selec- Y a simple and efcient method for selectively varying. vthe induction effect over a selected portion of' a charge therein without making any physical changes in the furnace coil.

Four figures have been used to illustrate the invention.

Eigure 1 is a sectional view of an induction furnace and charge piece, with diagrammatic representation of the electrical apparatusv used therewith, in accordance. with the present invention.

Figure 2. is va vector .diagram of the voltage and currents in the respective parts of the apparatus .s hownin Fig. .1. y

Figures 3 and 4 are views, generally similar to Fig. 1, but illustrating methods cf varying the heating pattern in an .induction furnace charge piece prior to applicantsv invention described herein.

In the drawing like numerals designate like parts..

Induction heating is valuable to industry hecause of. the great control it affords in regulatthe concentration of heating in a given part lfor .partsof a charge piece tobetreated. The

heating energy may be spread out to heat .a

large charge to a uniform temperature.; it may be concentrated to heat a small section or area very quickly toa high temperature, or it may be divided to heat a portion of a charge piece to a certain temperature and other parts of the same charge piece to a different temperature.

In many forging operations, the heating pat- .tern of the charge piece is of extreme importance.

" A part to be Workedto a greater degree than f ltei'nperature than the other part.. In forging a shell nose, for instance, the end. usually .mustl be made hotter than the rest of the shell, not only to facilitate the :shaping ofthe 4nose but to prevent distortion of the shell body which would result if the heating pattern were not tapered off gradually toward the body of `the shell.

In induction heating in general, and especially Where irregular heating patterns are required, .it is often difIicul-t to attain the desired pattern without a great ydeal of cut and try experimentation. Even in simple furnace assemblies lthe heating pattern cannot be entirely predicted because of variation in the metal or magnetic qualities of the charge pieces, the couplingand end efect of coil and charge piece and other considerations, v

Heretofore, coils have been calculated, `using prepared formulas, with' duel consideration to details known to aiect the application under consideration. Where it hasl been desired to concentrate heating energy, the inductor turns have been crowded closeA to the-charge piece vvand where less 4heating has been desired, the coil turns have been spaced from the charge piece or spaced from each other.

Heretofore, also, when an inductor coil has been tried and has been found to produce an. un.- desi-red heating pattern in a charge piece, lit Ahas been physically torn apart, 'the coil turns have been respaced and a new trial has been made. By eicient juggling of the turn. spacing ya prevferr-.ed heating pattern has been attained. Such procedure, however, is diflicult and uncertainthe heat insulation 3, and guides or spacers 4 k(Figs. .3 andv 4), between the charge piece and coil must be .broken ou-t and replaced, and only an experienced operator can make the necessary electrical changes.

By applicantrs new invention, the heating gradient of.- a charge piece` may be varied with great easel by ordinary mechanics. Instead of -varying the .coupling .between the furnace. coil and charge piece or of. vrespaeing.- the lfurnace vcoilturns, thev operator needs only to add a shunting capacitor to the inductor coil over the yparti-of the charge piece Where it is desired to--inture change. Several capacitors of varying size may be employed with a single coil, allowing almost any desired pattern to be attained without making physical changes in the furnace coil itself.

Referring again to Fig. 1, Vif it is desired to heat the threel sections A', B and C", of the load l, to different temperatures, the procedure, according to the present invention, would be as follows. First, a furnace coil 2 would be calculated, which with a, given voltage, obtainable from a power source G, and with a given coupling and turn spacing withrespect to the load, would induce the desired amount of power into the charge piece as a' whole. The inductance of the coil would be balanced by a capacitor C placed across its extremities to effect a unity power factor condition on the generator supply mains. If a greater or less power were obtained than was desired, the generator tap would be varied along the coil for adjustment.

Under the above-noted conditions, and assuming a load extending out of the ends of the coil, the heating of the charge piece within the coil would be substantially uniform, and the vector .diagram would be approximately as shown by stant as to length and direction, while the heating pattern of the load is changed, if the generator is to be operated to its best` advantage.

If Iit is desired to heat the section B' of the charge piece to a degree considerably in excess of that of C', and to heat section A' more than C', but less than B', then the voltages across these sections must be determined and capacitors Yof appropriate rating must be tapped across the sections in question. It must be borne in mind,

however, that the total Kv.a. of the capacitors for the coil as a whole must be kept substantially constant, and as units are added at B and A otherunits must be taken away from the coil as a whole.

Returning to the vector diagram, it is known that to get more heating in section B', or in section' A', the current, which varies as the square `root of the heating effect, must be increased in coil sections B or A. Since the vector sum of the current in the inductance and capacitor of section B or of section A must be equal to the current in the coil of section C, it is only necessary lto determine the length of the current vector required in each case, and in general to couple it in angle and phase to agree with a capacitorl vector drawn along the line Ico, the length of which makes the resultant vector substantially equal to Ici., and to determine from the new capacitor vector, what size capacitor to add vto 'effect the required heating.

.The foregoing might appear to be misleading 'ain that the terminals of the'inductance current -vectorsIcn ImJ and IBL are all shown as falling along the same straight line and the terminals yof the capacitor current vectors also are shown vas falling along a straight line.

This is not the case in actual practice as the phase angles of each, together with their associated voltages, do vary somewhat. However, the diagram is sufciently accurate to obtain a close approximationl of the capacitors needed Iand is very much less complicated than if the vectors were shown at their proper phase angles.

Actually, in practice, the selection of the proper capacitors, sections and number of turns would probably be a matter of trial and error. Once the main coil has been calculated and built, any competent engineer or mechanic can easily make the changes necessary to shift the heating effect along the charge or to any portion of the charge desired.

One relatively simple rule must be borne in mind. Namely, that the maximum capacity in microfarads which can be used eiectively across any igiven section of the inductor winding is equal to the total microfarads of the bank originally used to resonate the winding as a whole multiplied by the inverse ratio which the section in question bears to the original length of the coil as a whole. For instance, if a quarter length of the coil is assumed, the maximum capacity which can be effectively connected across its terminals is four times the capacity used originally to resonate the coil as a whole. If this amount of capacity is used, then the current through that portion of the winding will be quadrupled and the heating will be sixteen times as great. This, of course, would be an extreme case and the generator would be greatly overloaded.

In practice, usually, the original coil will not be far wrong and the required energy shift toward an end or a center section will be slight.

The vector diagram does not take into effect the voltage vectors across coil sections C, Band A because they are relatively unimportant, so long as the capacitive effect across any section does not equal or overbalance the inductivereffect of that same section. If the capacitive eect exceeds the inductive elect in any section, the heating in that section diminishes as the capacity is further increased. Although, it is possible to work in this range, applicant prefers to work in the more positive range where the capacitive effect is always less than the inductive eifect.

Although applicant has limited his discussion to the bridging of furnace turns by capacitive reactance, it must be understood that a generally similar control can be obtained by using an inductive reactance which would act in an opposite direction, i. e., increasing the current in the turns not bridged. However, the use of capacitors lis much to be preferred because the power losses are very slight as compared with the'heavy losses which would have to be taken into account with inductive reactances.

From the foregoing it may be seen that by the simple expedient of supplying a sufficient number of taps on the furnace coil and by adding capacitor units across selected coil turns, it is possible to obtain almost any desired temperature pattern in the charge piece being treated.

The invention herein described has been used to vary the heating pattern of munitions bodies to determine the best conditions for forging operations and has afforded considerable saving over the old cut and try method where the coil has to be torn apart after each try, and made over.

Applicant claims as his invention:

1. An induction furnace comprising a continuous unidirectional helical inductor winding adapted to surround a charge to be heated, said winding having taps arranged throughout its length, a main capacitor connected across substantially all of the charge enclosing portion of the winding, a source of alternating current energy connected across substantially the same turns of said Winding as are connected by said main capacitor ,and an auxiliary capacitor con nected to certain of said taps across intermediate turns of said winding, the auxiliary capacitor being of a value which is proportional to the heating effect desired in the portion of the winding Which it shunts, and the collective effect of all the capacitors being such as to produce substantial 10 main capacitor, and a plurality of auxiliary capacitors connected across intermediate turns of said winding, the auxiliary capacitors each being of a value which is proportional to the heating eiect desired in the portion of the winding which it shunts, and the collective eiect of all the capacitors being such as to produce substantial resonance with the shunted turns of the Winding.

THEODORE R. KENNEDY.

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

UNITED STATES PATENTS Number Name Date 1,946,876 Northrup Feb. 13, 1934 1,986,353 Ncrthrup Jan. 1, 1935 Sherman Sept. 4, 1945