US857560A - Process for generating high-frequency alternating currents. - Google Patents

Description

PATBNTED JUNE 18, 1907'.

M. LEBLANG. PROCESS FOR GENERATING HIGH FREQUENCY ALTERNATING GURRENTS.

' APPLICATION PILBDIEB. 6, 1904.

8 SHEETS-SHEET 1.

IN ENTOR; fi Mrcc? .ZeZZano, I

I Attorney,

80.}.857560. PATENTED JUNE 18, 1907.

- M. LEBLANG. PROCESS B0B. GENERATING HIGH FREQUENCY ALTERNATING GURRENTS.

a sums-Sum a.

APPLICATION FILED EBB. 6. 1904.

INVg/VTOR: 772 Qewwoe ACZZQTLC), 3, J y

Atlameys,

No. 857.560- PATBNTED JUNE 18I 1907.

M. LEBLANG.

P300388 FOR GENERATING HIGH FREQUENCY ALTlEiRNAIIN'G GURRENTS.

APYLIOATIOII FILED P113. 6, 1904.

a sunk-51mm a.

WITNESSES. INVENTOR.

I 22460406 fiw) y w Attorneys UNITED STA S P TENT OFFICE.

MAURicE LEBLANC, or PARIS, FRANCE, ASSIGNOR To WESTINGHOUSE ELECTRIC & MANUFACTURING COMPANY, A CORPORATION or PENN- SYLVANIA.

Specification of Letters Patent.

Patented June 18, 1907.

Application filed February 5, 1904. Serial No.' 192.227.

To all whom it may concern:

Be it'known that I, MAURICE LEBLANo, a citizen of the Republic of France, and a resident of Paris, France, have invented a new and useful Process for Generating High- Fre uency Alternating Currents; of which the ollowing is a specification.

In an application No. 183,246 heretofore filed byme on November 30, 1 903, I have described and claimed a means for producing uni-directional currents, which comprises a dynamogenerating an alternating electromotive force having its maximum value of one sign greater than its maximum value of the opposite sign, and an electric exploder which is adjusted to establish an arc and to permit current to pass only with the electroniotive-force of that sign which has the greater maximum va ue. The. uni-directional currentsthus produced may be made continuous by means of a condenser-in shunt of these parts. b In another application No. 183,247 heretofore filed by me on November 30, 1903 I have shown a means for changing the continuous currents thus produced into alternating currents traversing the primary coil of a transformer,"*wln'ch means consists, generally speaking, of a resonator connected to the line carrying the continuous currents and an electric exploder in shunt of the resonator, the parts being electrically proportioned so that the continuous current pcing circuitwhich is connected to the transriodically charges the resonator when the exploder arc is not established and the resonator periodically discharges as an alternating current through the exploder when its arc is established, these alternating currents being led through the primary coil of the transformer. But in the application No. 183,247 the alternatin current thus set up in the primary coil of the transformer is rectified into a continuous current in the workfornier secondary.

Now in the present application, I may utilize the means for generatin a continuous current above reterred to and may furthermore employ the means for converting this continuous current into an alternating current in the rimary circuit of a transformer as above escribed. But. the alternating currents generated in the secondary coil of this transformer, instead of being rectified,

I now propose to supply as alternating currents of hi h fre uency to the Working 011'- cuits, and furt iermore propose to make both the effective intensity and the frequency of these high frequency currents as constant as those of currents furnished by alternators, so that one may utilize with them all the roperties of electrical resonance. To this end l have devised an exciter for the dynamo or prime source of current, which eXciter takes its energy from the working circuit and acts in. such a manner, as will be more specifically pointed out later 'on, that any departure from the normal fre uency 'of the alternating working currents rings about such a change in "the'excitation of the prime source of electrical energy as to restore thisfrcquency to its normal value.

In the drawings Figure 1 shows a complete diagram of my system. Fig. 2'shows the electroniotive-force and current curves of the prime source of electrical energy which I prefer to employ. setof-current curves of this prime source. Fig. 4 shows the current curve of the apparatus which I prefer to use to transform the continuous currents into alternating currents and Figs. 5, 6, 7, and 8, show types of electric explodcrs which 1 may employ.

. In Iig. 1, I have indicated by I the prime source of electrical energy which I prefer to employ and which, taken with the electrical exploders shown as spark gaps 6, 7, S, and the condenser 10, generates a continuous current of practically constant intensity. I have designated b II the electrical resonator and the exploder 1n shunt thereof, for transforming thiscontinuous current into altcrnating currents. I have furthermore designated by III the frequency regulator which Fig. 3 shows another I have devised, by means of which any detaken-from the working circuit, into the continuous currents which are necessary to modify the excitation of the prime generator.

Coming now to a more specific description of the dynamo I which generates an alternating electromotive-force having its maximum value of one sign greater than its maximum value of the opposite si 11, it will be found to consist of a air of fie d'poles having the circumferentia extent of the positive pole, say, much less than that of itsnegative pole. The armature, in the case shown, consists of six coils connected in diametrically opposite pairs, and connectedto the rings 1, 2, 3, 4, in such a manner that the brush which rests on the ring 1 acts as a common return for all the pairs of coils, and the brushes which rest on the rings 2, 3, 4, are each respectively connected to the other end of one oi the three pairs of coils. The armature is supposed to turn at constant velocity and to be separately excited by means of field coils 33, 33', in a circuit with a source of continuous current, which impresses a constant diiference'of otential upon the terminals of the field 005s. .o

It is manifest that instead .of using two field poles, Lma use a number of pairs of field poles that Instead of using three pairs of armature coils I may use any other number of such pairs; that I may use either a'series or a parallel connection between the coils I constituting a pair of coils; and that I may connect the armatures coils in sets ofthree or more instead of in (pairs.

Since the circumferential extent of the ositive field pole of my prime generator is ess than that of the negative field pole,'-it follows that the fluxintensity is much greater under the narrow pole than under the broad pole. 'The armature being turned at a constant velocit within the ma etic,

field thus produced, 1t further follows that.

" law something like that represented by the curve A of Fig. 2. That is to say, the maximum value of the positive electromotiveforce will be much greater than the maximum value of the negative electromotiveforce whereas the duration of the positive electromotive-force eriod will be much less than the duration'of tive-force period.

By referring to Fi 1, it will be seen that the electrical explo ers- 6,7, and 8, which each consist of a air of separated balls a, b, are connected to t e armature circuitsof the prime generator I so that the balls I), b, b, are all connected through the condenser 10 to one side of the armature coils by'means of the common return conductor 5, and that the balls (1 a, a, by means of the brushes 2, 3, and

ed on the surface of the.

the negative electromo-- sensed i 4, are each irespectively connected to the other side of the armature coils. The condenser 10, in fact, is in shunt of the set of exploders 6, 7, and 8, and-the armature circuits of the prime generator; and each armature 'circuit is in series with one exploder. Since the electric exploder or spark gap 0, b, may be of common form, I need only remark that in order to establish an are between such exploders it is necessary to develop between them a voltage of a given magnitude; that the are being once established, its resistance becomes quickly negligible, and that the arc is immediately exti uished when the intensity of the current w 'ch traverses it is zero or becomes inferior to a certain limit. I designate by H the voltage necessary to establ1sh or spring an arc in the exploders 6, 7, 9, and by h the voltage normall maintained between the plates of thecon enser 10. It is to be observed that the magnitude of h is always less than the ma nitude of H. I fur. thermore assume, tofix ldeas, that the maximum positive value of the electromotive- .force E su plied by any given'armature circuits of the d amo, is greater than the sum H+h, an that its maximum negative value is smaller than the difference l-I- h.

one armature circuit, say that connected with the exploder 6. Since now, during the eriod of ositive electromotive-iorce, there is availab e, to establish the arc in the exploder, by reason of the resence of the condenser 1.0, a voltage of h, it will be clear that some time during the period in which the armature coil is the seat'of positive electroinotive-force, there will be available a voltage. greater than H h it, that is to say H, to establish this are in the ex loder. It follows that the exploder arc wil always be established sometime dur' the period of poslitive electromotive for'ce 1n the armature col Referring now to Fig. 2, in which I have traced a curve showing the variation of the electromotive-force E develo ed between the balls of the exploder 6, I wil assume that. ,at the momentt,, the volta e developed between the balls of the explo er is e ual to H,

that is that it is at this moment s establish the arc. Starti at this moment cient to 1 t the armature circuit will e traversed by a current, the intensity of which will gradually increase, as is shown by the curve B in Fig. 21, If we neglect the energ lost inthe circuits ofthe dynamo and in t e exploder, that is to say if we suppose negligible the resistance of the armature circuits and of the exploder, when the latter is in action, the intensity of the current which traverses the exploder, once it is in action, will only be limited by the action of the self-induction of tho armatureicircuit under COIlSIdfiIfi-i-hYH. This means, in effect, that the current B will 'lag and below the line t.

- 90 degrees behind the electromotive-force A,

measured positive and ne ative above and below the axis 0 t along t e curve A. The

electromotive-force across the terminals of the exploder 6, by reason of the action of the condenser having a normal voltage of h, is to be measured positive and negative above Y Since the maximum positive electromotive-force enerated by the armature circuits is greater t an H h as be fore assumed, the curve A will always have a part of its peak above the line in dashes 0 t which is parallel to 0 t and at a distancc Il therefrom. Therefore, during some portion of the period of'positive electromotiveforce, there Will' always be enou 11 potential across the ex loder 6 to establis 1 an are, as has been sai before. Since the maximum negative electromotive-force generated by the armature circuits. of the rime generator is always less than Hh, it follows that the curve A will never reach the line 0" t-drawn at a distance of II below the axis 0 t. This, put in other words, means that the maximum ne ative electromotive-force across the termina s of the ex loder 6, which is measured downward from t e line 0 t, can never be equalto H, which means that it will always be insufficient to initially establish an arc across the exploder.

l. have shown above that at'an epoch of positive electromotive-force t a currentpasses across, the exploder which increases in intensity until the epoch t,. Thereupon the intensity of this current diminishes and becomes zero at the epoch t,, which epoch, aceordin to the fundamental laws of current induction, is determined by the fact that the area S shown in hachures is equal to the area S shown in hachures. It can readily be shown that the distance t t is always less than the distance T between two adjacent peaks of the curve A, which means that the current represented by the curve B will always become zero before the end of that complete period of electromotive-force in which the current was initially started. But when the current represented by the curve B ,that is the current which traverses the exploder 6, becomes zero, the arc in this exploder will be immediately extinguished, and since, during the period of negative electromotive-force, this electromotive-force never reaches a value suflieient to initially establish the arc, as has been shown above, it follows that the exploder arc will not be re-established until the next positive period of electromotivc-force A, that is at an epoch t +1. '0 see, therefore, that the action ofthe armature circuit of the prime generator which is connected to the exploder 6 is to send into the condenser 10 a series of positive current waves 13,, B,, as shown in Fig. 3, and to send into this condenser no negative current waves. Similarly the action of that armature circuit which is connected with the exploder 7 is to send into the condenser 10 a series of positive current waves 15,, B Finally the armature circuit which is connected with the exploder 8 sends into the condenser 10 a series of positive current waves B B By using a sulficiently large number of dephased armature circuits on-the' prime generator, it is manifest that the condenser 10'will receive what -is practically an uninterrupted charge of con-- tinuous current. The apparatus will operate most eiliciently if the voltage 7t, maintained across the condenser 10, is small with reference to the voltage H necessary to initially establish the exploder are.

I now come to a description of the means 11 for transforming the continuous current, which is supplied by the devices above described, into an alternating current of high frequency in the primary circuit 12 of the 'transformer 13, such means consisting essentially of an electrical resonator of a specified electrical design with an electric exploder in shunt thereof to which an automatic circuit breaker may be added for insuring the steadiness of the operation.

The primary circuit 12 of the transformer IOC 13 passes through a self induction coil 14 and a condenser 16 in series therewith. If the primary circuit 12 has suilicient self induction in itself, the coil 14 is unnecessary. Since the primary circuit 12, the self induction coil 14, and the condenser 16 constitute a circuit having both capacity and self induct/ion, they will constitute an electric resonator 1n shunt of which, as is seen, '18 mounted an electric exploder 17 which may IIO also be formed of two metallic balls 0,, b, separated by an.'air space. The continuous current supplied by the hne 9, 1n the manner already fully described passes into this res- ,onator and charges the condenser 16 for a certain time, at the end of which an arc is established in the exploder 17. This are once established, the resistance of the exploder becomes practically zero so that we have the well known phenomena of a charged condenser discharging through a small external resistance. This means that an oscillatory discharge will be set up giving rise to an alternating current. The electrical constants are so chosen that before the end of one con iplete oscillation of this discharge,

that is shortly before the end of one correlate period of the alternating current set up by established. The condenser 16 is thereupon again charged by the continuous current' supplied by the line 9 until matters are in circuit breaker '18 which, when the solenoid charge of thecondenser.

I Thereu on: .the solenoid again raises its'core the same state in which they were when the arc in the exploder was first established. But matters being in the same state in which they were at the moment when the exploder arc was established for the first time, it follows that the exploder arc will be again established, that there will be another oscillating discharge in the condenser and another alternating current comprising one complete positive and one complete negative period and so on. The alternating current, thus generated, will not gradually die down, as in the discharge of the ordinary'Leyden jar, for the reason that the condenser is recharged from the main line at the end of each alternation; I here remark that the self induction coil 11 hel s to maintain constant theinte'nsity I of t e continuous current inv the line despite the establishment of the exploderarc. It is thus seen how, by the arrangement thus far described, a continuous current is transformed into an alternating current in the primary 12. The continuous current being of constant intensity, it can be shown that the alternating current intowhich it is transformed ,.is also of effective constant intensity.

It remains to briefly refer to the automatic circuit breaker 18. It will be noticed that, as shown, this automatic circuit breaker consists of a solenoid in series with the primary '12 and-condenser 16, there being a branch circuit through the flexible conductor 155, the core of the solenoid and the mercury is de-energized, short circuits the condenser 16 and practically short circuits the explode]- 17. This, automatic circuit breaker is designed to maintain its-core raised and the short circuits broken throughout the normal operation ofthe'a paratus. Itdoes not go into and out of action at each oscillatory dis- But should, for any accidental reason, such as a surcharge on the working circuits, the arc in the exploder not disestablish itself at the proper a):

time toward the end of any com lete osci tion, the energy contained in t eresonator will rapidly dissipate, so that the solenoid will permit itscore to drop and short circuit the exploder are 17 The exploder arc is therefore now extinguished as it should be.

to brea i the short. circuits about the condenser 16 ,and' the exploder 17 sothat the system may renew its normal operation. The automatic circuit breaker has therefore the function of preventing a disarrangement of the normal 0 eration' of the system by reason of the fai ure of theexploder to disestahlish its are at the proper time.

I now briefly refer to the principles which are to be employed in the electrica design of the apparatus II, by reference to Fig. 4, in which the'axis of abscissa denotes time and the axis of ordinates denotes current strength or charge. I

So long as the resonator is shortcircuited by the exploder are,- it is the seat of a current of intensity J the variation of which may be represented by a curve resembling a sinusoid, such as the curve J shown by full lines in Fig. 4. It is also manifest that the variation of the charge Q of the condenser 16 may be represented by a'sinusoid such as the curve Q in dots and dashes, in Fig. 4, this curve being dephased by 90 degrees from the curve J.

Now it has been proposed, in the prior art, to convert a continuous current into an alter- .nating current of high frequency by charging a resonator with the continuous current and then discharging the resonator through an exploder. But such method has never been perfected to a. oint at which it can be commercially use and certainly not for purposes requiring electrical resonance, for the reason that it was not known how to make the alternating currents of effective constant frequency, nor how to make them, practically speaki sinusoidal in form. l have discovered ow to do these two things. Without explaining, at length, how 1 have reached a knowledge of the requirements necessary to effect these two objects, I may say. that if we designate b R the ohmic resistance of the resonator, y l the coefficient'of its apparent self induction, by c the capacity of its condenser 16, and by E the quantity of electrical energy stored in the resonator at any instant of time t, then the rules are these: In order that the frequency of the alternating current shall be constant, it is necessary that In order to transform the alternating currentsof eonstant effective intensity in the primary circuit 12 into the alternating currents which are to be utilized on the working circuit, Iuse the secondary 19'of the transformer 13. The alternating currents in the primary circuit 12 are thus transformed into alternating currents in the secondary circuit 19 from It shall be small with respect to which the are supplied to the working circuit 20. bince the effectiveintensity of the alternating current in the primary circuit 12 is constant, the efiective intensity of the secondary currents in the. secondary circuit 19 will ordinarily be constant, the transformer 13 being'then of the constant current type and acting to transform an alternating'currentof constant effective intensity into another alternating current of constant effective intensity, 'If it be desired to transform the primary current of efiectiv'e constant intensity into a secondary current of effective constant voltage, it is merely necessary to insert a condenser in series in the secondary circuit, of a capacity to render zero the coefficient of apparent self induction of the secondary.

In the above description I have assumed the power factor of the working circuit to be constant so that if the prime generator is regulated to produce a continuous current of constant intensity, the frequency .of the alternating currents will be constant. I now come to a description of my regulating exciter III which is intended to regulate the excitation of the prime generator I in such a manner that the frequency of the currents on the working circuits remains constant despite variations in the power factor. Now an increase in the excitation of the field of the primegenerator acts to augment the intensity of the charging current of the reso-- nator in diminishing the duration of this charge. A diminution of the duration of the charge will have for its effect the diminution of the interval of time which separates the initiation of two consecutive establishments of the arc in the exploder 17, thus augmenting the frequency a: and vice-versa. I have thus been led to rapidly augment the excitation of the generator if the frequency diminishes and to rapidly diminish this excitation if the frequency increases; In this way I bring about the result that the frequency on the working circuit does not vary except between limits as close together as I please.

Let us assume that the working circuit is the ordinary series circuit traversed by a current of effective constant intensity. I intercalate in series in this circuit, the primaries 21 and .22 of two transformers having secondary circuits 23 and '24. But if the working circuit 20 is of the multiple arc type, supplied withcurrent at constant voltage, the primary c rcuits 21 and 22 will be mounted in multiple therein. But Whether the working circuit is of the series or the multiple arc ty e, I mount a primary circuit 26 of a transformer 200 and aco-ndenser 25 in series with the secondary circuit 23 and I furthermore mount the primary circuit 28 of another transformer 300 and a'conglenser 27 in' series with the secondary circuit- 24. The secondary circuits 29 and 30' the trans- The prime generator I has exciting mutual induction with their primary circuit 26 are equal and of op osite signs. A similar remark applies to t e secondary circuits 31 and 32 of the transformer 300. vI designate by y a frequency which is small with re-- spect to the frequency as. I furthermore determine the capacity of the condenser 25 in such a manner that the apparent self induction of the circuit 23, 25, 26, is zero when it is traversed by currents of frequency a:y. l furthermore determine the capacity of the condenser 27. in such a manner that the apparent self induction of the circuit 24, 27, 28, is zero when it is traversed ,by a current of frequency 90+y. From this it follows that the electromotive-force developed in the secondary circuits 29, 30 and 31, 32, will be equal when their frequency is :10. When the frequency of the current in the working circuit diminishes and falls below a; and thereby gets nearer to ac-y and farther from n+1 the electromotive-force developed in the circuits 29, 30, will increase, and the electromoti ve-force developed in the circuits 31 32, will diminish. On the other hand, when the frequency of the alternating current in the working circuit increases above as, and gets nearer to m-i-y and farther away from 03-11,

the electromotive-force in the secondaries 31 32,- will increase and that in the secondaries 29, 30, will decrease.

field coils 33, 33, interconnected in any suitable manner and traversed by a continuous current furnished by any local source of elec trical energy, which maintains a constantdifference of potential between the conductors 36 and 37. This is a common way of exciting the field magnets of agenerator, and I need only say that I regulate the intensity of this local source of current in such a manner as to normally excite the prime generator to. furnish to the working circuits currents of fre 'uency as at mean load.

he coils 34, 34, which are connected to' connection of the secondaries 29, 30,-with the field coils 34, and of the secondaries 31, 32, with the field coil 35, it will now at once be apparent that should, for any reason, the frequency of the alternating current on the. working circuit 20 increase above the normal frequency w, the electromotive-force, in the secondaries of the. transformer 300 will increase, which means that there will be more end 'of the secondary 29 is connecte current in, the field coils 35, subtracting from the effect of the exciting coil 33. At the same time the electromotive-force inthe secondaries of the transformer 200 ,will decrease, so that there will be less current in the field coils 34, to re-inforce the normal field excitation of the coils 33. Both these actions, then, combine to decrease the excitation of the prime generator I, which means,

as has before been explained, that the fre- .quency of the alternating currents generated electrode 40. The correspondin end of the secondary 30 is connected to the e ectrode 39. One end of the secondary 31 is connected to the electrode 43 and the corres onding end of the secondary 32 is connecte to the electrode 44. Op osite these electrodes just mentioned an co-operating with each of them is the electrode .400, which is connected to the wire 37 and' thereby to one terminal of I each of the. field coils 34, 35. The other terminals of the field coils 34, 35, are connected, respectively, to the secondaries 29, 30, and 31 32, by means of conductors 38 and 42. A circuit is thus established from .the secondaries 29, 30, to the electrodes 39, 40, and across the vacuum to the common terminal 400, thence to the conductor 37, through the field coil 34,. the wire 38, and back to the secondaries'29 and 30. A similar circuit may readily be traced connecting the secondaries '31, 32 through the electrodes 43, 44, and the common terminal 400, with the field coil 35.

The electric valve IV which I have shown comprises terminals 39, 40, 43, 44, which are i composed of bells of steel or iron, of one sign,

located in.a vacuum tube and a terminal 400, of opposite sign, composed of a drop of. mercury. Such a construction has the remarkable roperty that once an arc is established in the vacuum tube, it will permit the current to pass from the steel bell to the mercury but will oppose-a practically infinite resistance to the current passing in the opposite directions'from the mercury'to the steelbell. I may say immediately, however, that although this construction is very efiicient for rectifying alternating currents, it is only here given by way of example; Explaining the action of this electric valve a little more fully,

it to be noted that an alternating current the traversing the primary circuit 26 of the transformer 200 generates, say, a positive wave of electromotive-force in the. secondary 29 and a simultaneous negative wave of electromotive-force in the secondary 30, this being due to the fact that these secondaries are oppositely wound. An are being assumed as established in the ,electric valve, we will suppose that the connections are so fixed that the positive electromotive wave is allowed to generate a current passing from the electrode 40 to the electrode 400. At this time, the negative wave of electrolnotiveforce in the secondary 30 produc'es no effect. But in the next half period of alternating current in the primary 26, there will be a-negative wave of electromotive-force .in the secondary 29, and a positive wave of electrornotive-force in the secondary 30 so that a current now passes from the electrode 39 to the terminal400, and no current passes from the terminal 40 to the terminal 400. This means that there will always be a current in a given direction circulating through the coil 34, assing in one half-periodfrom 39 to 400, and in the next half period from 40 to 400. The oppositely wound secondaries 29, 30, taken together with the terminals 39,40, and the common terminal 400, of the electric valve have caused a rectified current to pass through the'field coil 34, such rectified current being necessary for the proper differentialexcitations of the prime generator. A similar remark manifestly applies to the field coil 35, the secondaries 31, 32, the terminals 43, 44, and the co-operating terminal 400.

While it is true that, 'the are once established in the electrical valve IV, the ga between the terminals 39, 40, 43, 44, an the co-operating terminal 400, ofi'ers a practical negligible resistance to currents flowing in a given direction and prevents the flow of currents in the opposite direction, it is yet necessary, in order to initially establish the arc, to impress a high voltageacross the terminals. To-thi's end I add a fifth electrode 45, of iron or steel, to the electric valve,'and place it in circuit with the impedance coil 46 which is thereupon connected with the conductor 36. I also run a tap through the resistance 47 and the circuit breaker 48, to the conductor 37, in the manner shown. By closing the circuit breaker 48, a current is passed through the impedance coil 46. When the circuit breaker is opened, the im edance coil 46 gives off a high tension disc arge in a circuit which contains the field coil 33,which jumps from the terminal 45 to the terminal 400 and initially establishes the arc in the electric valve. This are is thereupon maintained by theiaction of the other terminals which are'sufiiciently close to ether for the urpose. While I have descri (1 my regul ator III rzi types of electric eXplo for use with the prime generator I and iits '00.- o crating apparatuses, it is clearly manifest t at it may be used in other relations. It is furthermore to be understood that I' have merely described what ma be called generic ers and electric valves, and that I may use other types. In order to insure the instantaneous extinction of the arc in such exploders when the current traversing them becomes zero,I may send a current of air between the balls a, b, as

indicated in Fig. 5. Or I may replace the balls by two horn shaped parts a, b, situated between the polar extremities of an electro .magnet, as indicated in Fig; 6. Again I may inclose the balls a, b, in a vacuum tube as shown in Fig. 7, such exploders going out of action instantaneously as soon as the Ill? tensity of the current traversing them sinks below a certain limit. One of the most effective forms of electric exploder which may be employed consists of an inverted U tube in which a vacuum is maintained, the ter-' minals being constituted by globules of mercury in the inverted ends of the tube, all as shown in Fig. 7. The effective resistance of such tubes is very small, after the arc has i once been established; they go out of action instantaneously and the surface of their electrodes is not injuriously affected.

The apparatus shown herein is claimed in my copending application No. 306,033, filed March 14, 1906, which is a division of this case. Y

' I claim 1. The process of transformi unidirectional currents into currents whic are alternating and of constant frequency, which consists in charging a resonator, hav1ng R small with respect to with unidirectional currents for one ortionof each alternating eriod, and disc iarging the resonator throug a path of low resistance during the other ortion of each alternating period, substantially as described. a y

2. The process of transformi unidirectional currents into currents whic are alternati and, sinusoidal, which consists in charging the resonator with unidirectional current for one portion of each alternating period so that it shall store an amount electrical energy which is large with respect to the energ consumed in the system in one period, and ischarging the resonator through a path of low resistance duri the other portion of this alternating perio substantially as described.

3. The process tional currents into practically sinusoidal alternating currents of effectively constant frequency, which consists in charging a resof transforming unidireconator having R "small with respect to with unidirectional currents for one portion of each alternating period, and so that it shall store an amount of electrical energy which is large with respect to the energy consumed in the system in a period, and discharging the resonator through a path of low resistance during the other portion of this alternating period, substantially as described.

4. The processof generating and maintaining upon a working circuit alternating currents of a constant normal frequency, which consists in generating unidirectional currents, transforming these into alternating currents by. periodically charging a resonator therewith and discharging the resonator through a low resistance path,causing a de arture from the normal alternating current requency to bring ab out a variation in the current flowing in anelectrically tuned circuit, and utilizing such variation of current excitation of the prime generator of unidirectional cuTrents to restore the alternating frequenc to normal, substantially as described.

5. he rocess of maintaining at a normal value the requency of an alternating current transformed from a unidirectional current, which consists in causing a departure from this frequency to respectively increase. and decrease the currents flowing in two different ly tuned electrical circuits, and utilizing such variation of current flow differentially to change the excitation of the generator of unidirectional current to restore the alternating frequency to-normal, substantially as described.

6. The process of generat'n and maintaining upon aworking circuit :ilternating currents of a constant normal fre( uency, which consists in generating unidirectional currents, transforming these into alternating currents by eriodically charging a resonator therewith and discharging the resonator through a low resistance path, causing "a departure from the normal frequency of such a ternat ing currents to respectively increase and decrease the currents flowing in two differentl tuned electrical circuits, and utilizi such variation of current flow difierentia l r to change the excitation of the generator of unidirectional current to restore the alternating freqpency to normal, substantially as descri. ed.

flow to change the In testimony whereoi I have signed my MAURICE LEBLANC. Witnesses: I

Hanson C. Coxn, J EAN Co'r'rmn.

Images