US1694791A - Induction electric furnace - Google Patents

Description

Dec. 11, 1928.

- 1,694,791 E. F. NORTHRUP 4 INDUCTION ELECTRIC FURNACE Filed Feb. 14, 1925. 2 Sheets-Sheet 1 coymtcnona. Z I Z 11 4 9 18 l 7 7 4% V g 4 --z 1&- g E I g -.16 &

Dee. 11, 1928. 1,694,791

E. F. NORTHRUP C v INDUCTION ELECTRIC FURNACE I Filed Feb. 14, 1925 2 Sheets-Sheet 2 W jmjjzwb Patented Dec. 11, 1928.

UNITED STATES PATENT OFFICE.

EDWIN F. NO RTHRUP, OF PRINCETON, NEW JERSEY, ASSIGNOR TO AJAX ELECTRO THERMIC CORPORATION, OF TRENTON, NEW JERSEY, A CORPORATION 01' NEW JERSEY.

Application filed February plied from multipolar alternators of standard desi n. a p

A further purpose is to connect the inductor of a high temperature furnace free from transformer iron in series with an alternating current dynamo'of standard construction and of frequency commercial or within the lower range of high frequency currents.

1 A further purpose is to maintain an interrelation between 'the size of'the furnace, the frequency and the ampere turns applied to make the size large enough to adapt it to operate on frequencies available from rotary alternators and small enough to Operate on their voltages.

A. further purpose is to increasethe frequency as compared with frequencies in common use to such a high frequenc tha'tthe increasing cost of production 0' standard generators balances against the reducing cost of condensers to correct power factor securing minimum cost for the entire equipment.

A further purpose is to so interrelate features ofstructure and physical and electrical constants as to adapt the ironless induction furnace to vary much lower frequencies than had previously been believed to be possible. A further purpose is to adjust the current automatically so as to pass it through a portion of the coil corresponding to the depth of the charge.- a p A further purpose is to use edgewise windings in multiple to secure maximum cooling for a turn of conductor having considerable I extent parallel to the axis of the coil.

' Further urposes will appear in the specification an in the claims.

I have preferred to illustrate my invention diagrammatically in aasingle form, whose dimensions are subject to wide modification,

INDUCTION ELECTRIC FURNACE.

14, 1925. Serial No. 9,312.

and diagrammatically show two of the ways v1n which it may be applied to a commercial circuit, selecting a form of furnace and ways andwhich at the same time well illustrate the principles involved. i

- Figure 1 is a sectional elevation, largely diagrammatic, of a furnace embodying my invention.

Figures 2 and 3 are diagrammatic views llustrating different ways" of applying the lnvention to a commercial circuit.

Figure his a fragmentary :ection showing i of connecting which-are practical and efficient i preferred -edge-wound inductor structur with multiple path water cooling. v F1gure5 1s a corresponding section showlng air cooled edge-wound coilsr Figure 6 is a diagrammatic view of an in ductor having automatic adjustment of primary current flow to depth of charge.

Flgure 7 is-a diagrammatic view illustrate ing by curves the frequency-cost relation existing with respect to condensers and'generators.

like parts.

' Describing in illustration and not in limitation and referring to the drawings:

Prior to my present invention, I have ap plied high frequency induction to the melting of electrically conducting metals but without full appreciation of the fact that the frequency could be reduced to come withinthe range of direct supply from generators of standard multipolar construction in application to non-magnetizable metals or to. magnetizable metals at temperatures above their recalescence point. induction coils surrounding the bath with frequencies of a" lower order had so far failed to teach the conditions essential to their successful use as never to have been practical and to have led to the belief that a lower order of frequency could not be used commercially for such a purpose. I 1

"I have discoveredthat this belief of the in adaptation of a lower order of frequency to the'heating of non-magnetic materials to high temperatures and iron and steel above the recalescence point has been due to a lack of conception of the true inter-relations of dimensional, physical and electrical constants of the furnace and circuit with a consequent In the drawings similar numeralsrefer to i Prior proposals to use misapprehension of an experiments with 'frequencies too low for t e size of charge.

I have discovered that increase in the lincar dimensions of a furnace capable of use upon high frequency induction is quite an im portant factor in adapting the furnace to operation upon a much lower order of frequency. The linear dimensions of a high frequency ronless induction furnace are relatively small and when it has been sought to operate a l'urnace of this character upon materially lower frequencies than that for which it is adapted there have been such a falling off in furnace efficiency and drop in power factor as have led to the belief of an utter inadaptation of the lower order of frequencies to heat nonmagnetic material to high temperatures without inter-linkage of transformer iron.

I have found that the power factor of a furnace of this character apparently rapidly drops away with any extension of the magnetic induction beyond the center of the charge, and that if the furnace is to maintain a power factor approaching its best power factor there must be such inter-relation between the impressed frequency and the resistivity and diameter of the charge as will prevent this extension of the magnetic induction across the center.

It is readily shown that the ratio between the reactance and the resistance of an inductor solenoid of the form shown in Figure 1 is a direct function of the product AFN where A is the radius (or other linear dimension) of the inductor and N the frequency of the impressed current.

The effective inward extension of the magnetic induction in centimeters into the charge can be shown to approximate 25,000 (p/N) which is about four times what is known technically as depth of penetration. This is that extent of penetration of the induced current into the mass which with the same current density existing at the surface would give the same total current as exists in the mass. Its

- numerical value is from one-third to one-fifth as great as the actual distance of induction penetration beneath the surface of the mate-.

rial.

For a theoretical discussion of depth of penetration consult Transient Electric Phenomena and Oscillations, Chapter VII, by Steinmetz.

If the conducting material being heated has a diameter suggested, equal to or greater than twice the distance of effective inward penetration of the induction, that is, e ual to or greater than about 50,000 (p/N) the inductor resistance'be very low as compared with its reactance, the power factor shouldapproach its best value.

Figure 1 illustrates a desirable form of ironless inductive furnace. The conducting mass 15 to be heated having radius B, height and H=H A and resistivity is surrounded and supported by a crucible 16 of wall thickness d. This in turn is surrounded by the inductor 17. A thin electrically insulating sheet 18 and a space 19 filled with a heat insulating substance are shown between the crucible and the inductor.

The inductor is most desirably of single layer winding, preferably edgewound and has a radius A, a radial dept 6 1%, and an average diameter preferably equal to the height H=H A. It is desirable to have H =2 and H QA.

The coupling 4 between the material 15 heated within the crucible and the inductor will vary both with the area ratio 13 /0 where C is the inside radius of the inductor, and somewhat with the ratio of H to A. \Vhen the crucible is charged to depth such that H=2A, a normal full charge, the coupling factor closely approximates a value 0.8B /C The inductor space 17 is partially occupied by electrical insulation and cooling provision, of which the most effective is water and only a certain proportion f of the total volume of the space 17 is occupied by the copper of the conducting element. The resistivity of the conducting element within the inductor is and the element has n turns of total resistance R. The voltage is proportional to the number of turns (a) and can be kept down by reducing a. This becomes of more importance in the larger furnaces.

Under these conditions it is readily shown that "the volume of the inductor is amflA that the direct current ohmic resistance R equals rfftfNI-UY that the 11 R loss within the inductor is equal to af- 25; (ni)'-'1\' and finally that the FR loss per unit volume within the inductor is (l/ i) f t A"( lti)". When the coil carries alternating current the direct current resistance is somewhat increased due to skin effect but this increase does not affect the general principles here set forth, though it may somewhat affect the final value which must be assigned to the furnace efficiency.

I find usual values for p; and desirable values for f and t, to be respectively 2X 10*, 1/2 and 1/6, and with these values the i R loss in the inductor becomes 75.4X1O-(oi) 1\", and the PR loss per unit volume within the inductor becomes 36 X lO A' (m') The power input in afinal analysis is llmited alone by my ability to extract heat from the coil by water flow as fast as it is generated; a far off limit.

In practice the permissible power input as regards the heating is greatly increased by the use of cooling fluid, the intensity of heating the inductor, as measured by the 51 R loss per unit volume within the inductor being then limited by what might be called the specific li'ti power to cool it as measured by the maximum rate at which each unitary volume may be cooled. v This s ecific power to cool the inductor, as measure by how many heat units per unit volume of the inductor may be removed each unit of time, will depend somewhat on circumstance and mechanical facilities, yet it may without undue error be taken as'approxiinately independent of the actual volume or sectional area of the inductor (provided the furnace proportions above expressed are The HR loss per unit volume has been shown.

maintained) within such liinits'as are likely to be met in usual practice, that'is, be taken as a constant independent of the size of the inductor and of the frequency of alternation of the voltage impressed upon the inductor.

Thisconstant will be the maximum 5 R loss per unit volume of the inductor that can be adequately removed by cooling the inductor.

to be measured b A' (m') The heat deve oping within the'inductor per unit volume of the inductor for a given a form of furnace is thus s'hown to-be measured by A"(m') so that any number of furnaces similar otherwise, but variant with respect to size and with respect to the frequency of the alternating voltage impressed upon the inductors, should normally at full load operate withthe same value for A (M1).

Considering these similarfurnaces, I find from mathematical analysis that if A? (at)? has the sanievalue in all, that then the respective values of-A N in the diiferentfurnaces may be taken as measuring the respective in,- tensities of 'heat input into the different charges; also that ,when these values areinulipgied by a suitable constant they .may be on as measuring the respective ratios be-fi t t tween reactance'and resistance of the respective inductors; also that by dividing the re.

- spective values of AN by the respective resistivities of the different charges values are obtained which determine in each case the ratio of the distance induction is penetrating -'into the charge to the radiusof the charge.

It is thus clear that furnaces o crating with the same values of A"(n'i) un er condit ons giving e uivalent values of A N in the? respective urnaces, should pour heat into their respective charges with equal intensities; i. e., the heat ttbSOlPlilOIl per unit volui'neof charge should be the same in all the furnaces; also,

" that in all the furnaces there should be the same ratio between the inductor reactan'ceand inductor resistance; also if'the furnaces are charged; with the same material or with (life frent "materials having equal resistivities,

then the ratio between the distance induction penetrates beneath the surface of the charge and the-radius of the charge will be the same in all. c

If the furnaces differ in size, as is here presumed to be the case, the frequencies of the p The power same current. electromagnetic induction will be converted into heat wintliin a relatively small outer higher thanin the larger furnaces, while. conditions with respect to power factor, furnace efficiency and intensity of heat delivery should be approximately the same in all.

A small furnace operating on at equency of 50,000 cycles should operate in t e same way as a furnace having ten times its diameter and operating on a frequency of 500 cycles, the charge and requisite power for the larger furnace being however 1000 times that of the smaller.

Consider the effect of. endeavoring to op- ,erate the smaller furnace upon 500 cycles instead of 50,000. The current, limited by the permissible heat withdrawal from the inductor coil'will be the same in each case. The heat-loss in theinductor will have substantially the same value as for 50,000 cycles, while the heat deliveredto the charge, .even.

were it possible to'm'aintain the same power factor, would drop to 1% of its former value; factor however would also drop away, both from fall in the reactance of the inductor and from the'induction spreading across the center of the charge, so that the power usefully converted into heat would become quite insignificant.

It is little wonder, in view of-the foregoing,

that it has hitherto been believed by those skilled in the art that commercial frequencyinduction and high frequency induction within the reasonable range of multiple pole gen erators of standard type (i. e., within the range of direct. generation by increasing the speed and the number of poles of standard types of multipolar generators) and frequency changducfting non-magnetic materials.

It is obvious" that the'larger furnace above is adapted to operate on frequenciesmuchhigher than 500 cycles. Subject to increase in R due to intensity of heating though unfortunately the 'nductor,appr0xiinately in accord with a re ation EaAn.

If we operate the larger 50,000 9 sity wit furnace upon but with corresponding increase in the power required and in the voltage to maintain the It should be noted that the shell only of the metal, losing the advantage.-

' ers were utterlyunadapted to heating con-' skin effect there may be a con siderable gain in elect-rical'efliciency and in cles instead of 500 cycles the'intenwhich heat is poured into the meltwill'be increased very nearly a hundredpfold,

ous generationof the heat throughout the mass which is obtainedwhen the depth'ofthe electromagnet induction extends nearly or quite to the center.

The larger furnaces are thus adapted to operate .upon much wider ranges with respect to intensity of heating and therefore are adapted to operate witlrreasonablo efiiciency upon much wider and higher ranges of temperature and frequency than the smaller furnaces. Also the larger furnaces should be better adaptedto operate equally well upon a wider range of materials (i. 0., materials in which there is a greater range of resistivity variation) than the smaller furnaces.

Comparing the two furnaces of different size for the same voltage the smaller furnace normally will have a larger number of turns and a higher frequency with which the induction, we will say, roaches but does not effectively cross the center of the charge, while the larger furnace for the same kind of material and same kind of heat absorption will have a smaller number of turns and lower frequency.

For the same values of A"(m) and AN the frequency for the most desirable operation of different furnaces will vary inversely as the square of the radius, or if the same fre( uency he maintained the rate of input varies irectly with the square of the radius.

In considering the question of size we must not overlook the fact that size cannot be increased indefinitely.

A natural relation exists between voltage,

number of turns and radius for furnaces hav:

ing AN equal and A"(m') equ'althat' is, operating at the same intensity of heat input and eificiency the values of E/An will be equal. From this we see that the voltage varies directly as the radius and directly as the number of turns. "Any limitation or preference in voltage therefore can be taken care of by adjustment of the number of turns.

In the operation of any furnace reduction of frequency below the value of N1:':A2N/A12 will result not only in less intensive heating but also in a drop in power factor by reason of the induction flowing across the center of the charge. Wit-h increase in frequency without a. corresponding increase in voltage, there will of course be a reduction in the current through the inductor with corresponding reduction in power.

My invention is well adapted for electric furnaces without the use of transformer iron within the coil and for direct supply of current from rotor generators of standard multipolartype and at voltages available from such generators. 7

Though I prefer not to use either trans formers or frequency changers, I recognize that both are adjustable by static apparatus, the voltage through condensers and the frequency through one or more frequency changers. successively doubling the frequency, though with corresponding successive reductions in the energy available. I have not considered it necessary to illustrate either of these devices in my diagrammatic showing to the cnera.tor will be substantially in phase.

with tie generator voltage. I I. As above indicated, in either of these cts the voltage or frequency could be adjiisted to that required by transformers or frequency changers, the furnace coil being fed from the transformer or tlequcncy changer orboth.

I have indicated the inductor spacerathcr than the inductor coil in Figure 1 to cmphasize the fact that some features of my invention are inde ndent of details of'character as to-the in uctor which fills the space; for example, as to whether it be watei -cooled or air-cooled, edge wound or otherwise, whether the different turns be in series or parts in parallel and whether each turn represents a single conductor or a pluralityof omductors in multiple.

Because of the desirability in large furnaees of having a small number of turns with a considerable vertical space occupiedb the turns and the necessity for a high rate o heat withdrawal from the inductor, I have-"inastrated in Figure 4 both the cdgewise winding and the water cooling which is so desirable where the rate of heat withdrawal must be a maximum. I have also included in thisaligure the multiple grouping of edgewise wmmamaductors by which a turn, in cross"section extending over a distance )arallel to theaxis greater thauitsi radial dept may stillbe edgewonnd and water-cooled. These features lit in with the multiple flow of cooling fluid applied by me in other furnaces.

In Figure 4 the individual conductors 25' or 25 (which may be air or water cooled) are edgewound and are shown as water cooled. In the fragmentary Figure 5 these conductors are shown as edgewound solid and air-cooled.

Both the water-cooled and air-eooled conductors are shown as connected in multi )lC, the unit of multiple shown beim three, so that each set of three conductors is electrically connected in multiple to com rise oneturn, the diagrammatic connections icing shown at 20 and 20'. The cooling water is supplied through insulating tubes to inlets 21 and is vented to discharge at 22, the several inlets and discharges being not connected turn by turn in the illustration but obviously being adapted to have as many inlets and as many outlets, electrical and fluid, for the entire coil as mayzbe preferred.

ill

Because the individual conductors are grouped in sets the conductors of each set require no insulation from each other, greatly increasing the unit volume of copper-per unit length axially of the inductor space; but the turns, comprising a number of conductors each, are insulated from each other'as at'23.

Air cooling space is shown at 24 and a heat insulatin and electrical insulating lining 18 is placed between the coil and the crucible, preferably againstthe coil. p

In Figure 6 I show automatic re ulating mechanism foradjusting the input 0 electric energy proportionally to the axial extent of f the charge, which in the case of a crucible becomes the depth of charge.

In this figure the inductor coilis divided into a plurality of sections shown in the illustration as two; these sections of inductor being connected in multiple with the current supply and,"if water cooled, also preferably with multiple water passages andthe usual tube insulation from the connecting pipes. jUbviously the number of such sections. electrically in multiple can be increased without varying from the theory of my invention though necessarily with practical limits as to the expediency. v

The inductor coil sections 17 and 17 which are in parallel will operateprecisely as the series coil when on full charge, except that their parallel connection will correspondingly increase the current flow. However, the current flow in each section is-limited by the impedance of the section, and there will be a very much reduced impedance within the section when it surrounds a conducting charge. As a result with a depth of charge equal to half the furnace depth the impedance of the lower coil 17v would be substantially reduced by reason of the charge whereas the reactunce of the upper coil 17 Would'remain a maximum. There would result from this a natural automatic adjustment of the current by which a relatively small current only would flow through the coil 17, whereas the coil 17 would take a large current.

The current is thus automatically much" restricted to'the portion of the coil surrounding the charge, greatly improving the coupling and reducing the 2' 7 inductor loss. At

the same time'the power factor will be im- I proved and the efiiciency will be increased.

Subject to requirements of simplicity and strength the conditions of intended use as they afiect the depth of furnace charge must be, considered largely in determining InFigure 1 it is of course apparent that the crucible itself may be the heating element,

the cruciblewalls being then of conducting material. Either conducting or non-conducting material would then be heated within the crucible.

the .numberand size of sections or sub-divisions of the llldllCtOICOll 1n each glven case.

Throughout the discussion, the resistance of the inductor has been taken constant whereas to a certain extent it will be a func-- tion of frequency.

It was shown by men the General Electric Review Nov. 1922, p. 663, that thepower absorbed by'the mass is approximately expressed by the Equation (10) on that page,

0.636E P 'z(+22 (I) where P is the power in watts, the coupling factor, E the volts at inductor terminals and m the reactance of the inductor when empty.

when the condenser is in shunt to the load. In Equations (3) and .(4) i 1 %m negative reactance,

R=the equivalent resistance of the'load and X=theequivalent reactance of the load. C

is the capacity in far-ads of the condenser. It

was shown by me in the General Electric Re- View, Nov. 1922, p. 662, that Since the coupling factor is constant, we can write,

Xaw (6) Hence, when the condenser is series connected (which is the more simple case to consider) r for complete power factor correction.

Hence, we can write from relation (2),

Pug-mow s It follows from relation (9) that the CE rating of a condenser required for correcting power factor varies directly as the power absorbed, but inversely as the frequency used.

From a commercial or economic standpoint this is a very vital conclusion. It sets a high premium upon the advantage of inductive heating with high frequency current.

While it is true that condensers are now priced as proportional to their k. v. a. rating,

Quotations received on alternators of different capacities and frequencies prove that the cost of such alternators per k. w. output increases with the frequency they must supply. Since condensers cost less as the frequency increases and alternators cost more as the frequency increases, it is of economic importance to pick the frequency point for the minimum cost of the combination of generator and condenser which operates at unity or near unity power factor.

In Figure '2' I show in a general way by curves the cost of generators and condensers at difierent frequencies based upon one CE output. In this figure curve 26 represents the cost of generators, which is seen to in crease with the frequency, and 27 represents the cost of condensers, seen to decrease with the frequency. In both of these, particularly the condenser cost, there are factors entering into the cost which may be treated as constants and which change the curves somewhat from their purely theoretic forms. Other factors, including costs for materials, improvements in design and advances in methods of manufacture will change the values of these curves. Howeven'at the time when any installation is contemplated the values, actual and comparative may readily be ascertained.

Curve 28 represents the sums of the ordinates of curves 26 and 27. This curve is seen to be high at both ends and to have an intermediate low section representing the range of economical equipment.

If the frequency at which two such curves give minimum combined ordinates ishigh enough to meet other considerations of efficient inductive heating, then this is the frequency to use. As previously shown the othfir considerations indlca'te that we should ma e If the diameter D of the 'mass to be heated is a foot or more, and the resistivity of the metal is about that of molten brass, the permissible frequency will be well below that shown by application'of the current as the range of frequencies for minimum combined cost of equipment.

With values at the present time I find that the lower range of minimum cost is about 360 cycles and that this range extends with little variation in the cost to somewhat over 500 cycles Without the use of frequency changers. The combinations of generators of lower frequency with frequency changers, making due allowance for loss of energy in the frequency changers would give a different curve for current supply cost which is to be considered in connection with the cost of condensers in reaching conclusion as to the most advantageous equipment from the standpoint of combined costs.

By making sure that the furnace is large enough for the Equation (12) to be satisfied it is clear that we can take our frequency for most commercial quantity of production solely upon an economic basis and from data such as used in plotting theleurves in Figure 7.

The curves shown in Figure 7 have been plotted for one value of powerinput, thesamc for both. There will be some variation particularly in generator costs according to the size of the equipment but it will not make much difference in the comparative cost.

Throughout the claims where I refer to the furnace coil as fed from a generator or use similar language I intend to include therein the feeding of the coil from the generator whether itbe through a transformer or frequency changer as well as when connected with the generator without the interposition of any such apparatus.

In view of my invention and disclosure, variations and modifications will doubtless become evident to others skilled in the art, to meet individual whim or particular need, and I therefore claim all such in so far as they fall within the reasonable spirit and scope of my claims.

Having thus described my invention, what I claim as new and desire to secure by Letters Patent is:-

1. In inductive heating free from interthreading by transformer iron, the process of efiiciently using alternating currents with in the range of frequency of generation by rotary niultipola-r generators which consists in confining the current path to a single layer of turns about the charge and in maintaining such a relation of frequency to diameter of the charge that the induced electromagnetic energy shall not penetrate substantially be yond the axis of the charge.

I 2. In inductive heating free from interthreaded transformer iron, of non-magnetic charges and of magnetic charges above the recalescence point, the process of efliciently using the current from rotary 'multipolar generators while maintaining the power factor which consists in making the radius of the charge not less than three times the depth of penetration of the electromagnetic energy into the charge.

3. In'inductive heating free from trans-r former iron by a coil surrounding a charge,

making the radius of the 'pool not less than the novelty which the furnace encies which are not high by enlarging thec arge for such a frequency to a point at which the induction induced withinthe charge does not substantially pass the center of the charge.

4. In inductive heating free from trans,- former' iron of a charge surrounded by the inductor, the noveltywhich consists in maintaining the} frequency within the range of a standard rotary multipolar generator, and in three times the depth ofpenetration.

. 5. In an induction furnace, a rotary generator ofmultipolar type, an edgewise wound coil is reached.

water-coole energy helical i'nductor furnace. coil free from included transformer iron, fed from the generator and static capacity correcting the power' to fre-- factor, the relation of size of charge quency being such that the energy induced in the char is substantially all converted into heat be ore the axis of the coilis reached.

6. In an induction furnace, a rotary generator of multipolar t an e'dgewise wound helica inductor furnace coil free from included transformer iron fed from the generator, connections'for'passing the cooling liquid through different parts of the conductor in mu1tiple,and static capacity correcting the power factor, the relation of size of charge to all converted into heat before the axis of the 7. In an inductive electric furnace, an

alternating current generator of rotary multipolar type having a frequency in'excesspf 7 so that the energy in substantially all converted into heat beforeexisting commercial'line freq encies,-an mductor heatingecoil fed thereby, the diameter of the charge ing sufficient at thefrequency so that the energy in the secondary shall be substantially all converted into heat before the axis of thecoil is reached, and capacity consists in accommodating.

frequency being such that the induced in the charge is substantially" inductor coil being erator of rotary multipolarty at the frequency used.

alternating current generator of rotary multipolar type having a frequency in excess of existin commercial line frequencies, an in-- ductor eating coil'free from enclosed transformer iron fed thereby, the diameter of the :charge being suflicient at the frequency used and at the resistivity of the metal treated so that the energy in the secondary shall be substantially all converted into heat before the axis of the coil is reached, and capacity for power factor correction in the circuit.

510. In an inductive electric furnace, an

alternating current generator of rotary. multlpolar type having a frequency in excess of existing commercial line frequencies, a single layer edgewise wound helical inductor heating coil fed thereby, and free from enclosed transfogmer iron, the diameter of the charge beinggsufiicient at the frequency used so that the energy in tiall allconverted into heat before the axis of t e coil is reached and capacity in series with the supply circuit for power factor corthe secondary shall be substan-- rection in the circuit and to raise the-voltage.

11. In an inductive electric furnace, a generator of rotary multipolar form havin a frequency above existing commercial line requencies and a single layer, edgewise wound, hollow,- water-cooled, helical inductor heating coil fed thereby, and free from inclosed transformer iron, the diameter of the charge being large enough. with respect to the freuency so'that the energy in the secondary s all be substantially all convertedinto heat before the axis of the coil is reached.

12. In an inductive electric furnace, a generator of rotary multipolar type, an edgewise wound, helical conductor coil fed thereby and w free from enclosed transformer iron, and static capac ty for power factor correction of the circuit, a plurality of thel'edgewise wound conductors of the inductor coil being connected in multiple toform each effective turn thereof.

13. In an inductive electric furnace, a 'generator of rotarymultipolar type, an edgewise wound helical inductor coll, free from en'- elosedtransformer lIOIl and sta'tie capacity v for powerfactor correction of the circuit, the split up into a plurality of sections which are electrically connected in multiple with the source of current supply.

14. In an inductive electric furnace, a gena furnace coil supplied therefrom, free rom enclosed .transformeriron, thecoil having parts of its length electrically in multiple one with the other and power-factopcorrective static capacity for the circuit.

15. In an inductive electric furnace, a. generator of rotary multipolar type and a fur nace coil fed former iron within the (2011, the eoilbemg edge-wound and water cooled and having parts of its length electrically in multiple therefrom, free from transhaving t one with another and power-factor-corrective devices for the circuit.

10. In an inductive electric furnace, a generator of rotary multipolar type and a furnace coil fed therefrom, free from interlinkage of transformeriron withinthe coil, the coil being helical, single layer, edge-wound and water cooled and parts of the coil being in multiple with each other both electrically and with respect to the water cooling.

17. In an inductive electric furnace, a generator of rotary multipolar type and a ,furnace coil fed therefrom, free from interlinkage of transformer iron within the coil, the coil comprising cdgewound conductors, a plurality in multiple to the single turn.

18. In an inductive electric furnace, a generator of rotary multipolar type and a furnacc coil fed thercl'ronn-free. from interlinkage of transformer iron within the coil, the coil comprising cdgewound and water cooled comluctors, a plurality of conductors being in multiple both electrically and with respect lo the water cooling to make u each turn.

19. A high frequency indm-tivo furnace using static capacity for power fzu-tor(: n'- rection, having the frequency such that the sum of the costs er generator kw. for the generator and static capacity for power factor correction shall be a minimum.

20. A high frequency inductive fnrnacc free from interthreading of the inductor coil by transformer iron, in combination with a source of hi h frequency current including a 'enerator o rotary multipolar type, and conenser power factor correction having the frequency such that the sum of the costs for thecurrentsupply and for the condensers shall be a minimum per kw. used. c

21. A high frequency'induetion furnace free from interthreading of the inductor coil by transformer iron in combination with a source; of high frequency current including a generator of rotary multipolar type, and

condenser power factor correction in series with the generator and the coil of the furnace,

6 sum of the costs for current supply and condensers a minimum'per kw. used.

22. A high fre uency induction furnace free from inter-1m age of transformer iron with the inductor coil, an mductor coil, a

rotary multipolar type of high 'l're uency generator supply for the inductor coil and tain the power factor in the circuit, the frerpiency o the supply being such that the 1nuccd energy in the charge substantiall has .power-factor-corrective condensers to mainbeen absorbed before the axis of the merge I .factor-corrective condensers for the circuit in which the frequency is such that the induced ener y is substantially absorbed in the secondary before the axis of the furnace is reached and the combined cost of current supply and condensers is a minimum.

24. In a commercial frequency hi h temperature furnace, a single layer solenoidal inductor, an alternator energizing the inductor, means for removing heat developing in the inductor, and a non-magnetic conductor surrounded and heated by the inductor and having an inter-relation between its diameter, its resistivity and the applied frequency such that' the product of its diameter in centimeters into the square root of the ratio between the frequenc in cycles or secg, end and the resistivity in 0 ms per cu ic centimeter--is approximately as reat as 50,000; i. e., expressing it algebraical y charge is substantially all converted into heat before the axis of the coil is reached. r- EDWIN F. NORTHRUP.

Images