US2434705A - Gas compressor - Google Patents

Description

3.111,20, 194s.y -FQLAGO l 2,434,705

' GAS COMPRESSOR med sept. 9, 1944, 5 sheets-sheet 1 Jn'.2o,1948. '4 F. LAGO 2,434,795

' GAS COMPRESSOR Filed Sept. 9, 1944 5 Sheets-Sheet 2 v Jan. v20, 1948. F, LAGO 2,434,705

GAS COMPRESSOR Filed sept. 9, 1944 5 sheds-sheet` s Jan. 2o, 194:3.l F. LAGO 2,434,705

I GAS COMPRESSOR Filed Sept; 9, 1944 5` Sheets-Sheet 4 Jan.20,` 1948. Y F, LAGO` l 2,434,705l

Y GAS COMPRESSOR Filed sept. 9, 1944 5 Sheets-sheet 5 lNvENToR ATTORNEY Patented Jan. 20, 1948 -I i Y FrancisLagBronx, N. Y., assgnorto Henry W. v 1 Jarrett and Philip H. Sellew, New York, N. Y.

` applicati@ settanta 9, 1944, serial No. 553,359

' 33Claiins.

(c1. 230V-ss) This invention relates'y primarily to a pumping device, and more particularly toY a compressor adapted for use in variousv special applications, as for example in a refrigerator ror the like.

This invention is an improvement on the invention disclosed in my U. VS. patent, No. 2,258,415, in which the compression `of aA fluid is effectedby the reciprocal movement `oi" 'a' column ofliquid, such as mercury, back and forthiromonev compression chamber to another under the in'iiuence` of electrical currents. This invention isA animprovement upon the inventionV of that patent involving simplification orthe'apparatus Vand improvement in the eiiiciencywand reliability. j It is an object" of'thisinvention toprovidean apparatus'by which the liquid maybe made Yto 'w3 flow back andrforth through a conduit to per; form its function'so that the compression ,is obtained without the need of any mechanically moving parts and with regard to some of its embodiments without the need oi anycontacts Aor switches except such as areemployedto discennect'the apparatus from the line.l l

It is a further vobject to provide a device vofthe character described inlwhich the'number of parts are reduced to a minimum and the vspace required for the apparatus is minimized, thussaving expense of manufacture and reducing upkeep and making the device more compact. 'v

The invention accordingly comprises a device possessing the features; properties and 'the relation of elements which-willbe exemplifiedin the article hereinafter described and the scopeof the application of which Will be indicated in the claims. v'

It is a further `object of this inventionto provide a mechanism in which the moving element may be started in movement by an oscillation of limited amplitude while the device is not under load and may gradually be brought up to a full amplitude, gradually increasing its pressure until it is able to assume the load without requiring a heavy starting current.

1t is a further yobject to provide a'device in which the circuits themselves without any auxiliary elements maybe designed to eiect a power factor correction as iar as 'it is necessary or desirable to do so. y

It is av further object to provide a device which by reason of its construction will giveno Vradio in- Fig. 9 is a of the nature and objects of'the invention, reference should be had to the following detailed description, taken in connection with the accompanying drawings, in which: v

n Fig. 1 -is an elevation of apparatus embodying this invention.

Fig-2 is a section onl the line II-II of Fig. 1.

Figi 3is a cross section of the conduit employed',1d rawn to a larger scale. v

Figs. 4, 5,v 6 and 7 are vector diagrams of the electrical and magnetic quantities at differentV points in the operation.r` Y

fFig. 8-'is an elevation of a modied form of the device showing a separately excited secondary.

side" elevation of the construction 'shown in Fig. 8, part being removed.

Fig. 10 is a section ofthe conduit along the lines Ill-l0 of Fig. 9, with themajor portion of the core removed.

Figs. 11, 12, 13 and 14 are vector diagrams of thequantities involved in the operation of the construction of Figs. 8 and 9.

*Figy15is an elevation of'iurther modification in i'whichl the exciting coil for the secondary is placed upon the same core with the primary.

` Fig.' 16 Lis alvector diagram of the quantities involved in the operation of the construction as shownin Fig. 15. i

Fig. 17 is an elevation of a further modification ofthe construction shown in Fig. 15, in which the secondary excitingA coil is placed on a different portion ofthe core, thereby creating the eiect A,known as'the constant current effect.

, respectively to operation of the constructions of terference and Whichwill produce noiickering of lights and which will draw from the linea uni.-

v form current with a high power factor.

The device of this invention may be constructed to operate withv its elements directly energized Figs. `1'? :and 18. y

Fig. 21 is a construction similar to that of Fig. 9,

save that the coils are placed in parallel.

. Fig.Y 22 is a viewV similar to Fig.y 10 of a modied construction.

Fig.23is a section on the line 23--23 of Fig. 24

showing anV alternative construction in whichinsulating barriers are introduced into the conduit torlocalize the secondary currents.

Fig. 24 is a section of the conduit on the line 24-24 of Fig. 23.

Fig. 25 is an alternative construction in which the conducting barriers are introduced into the tube to direct the flow of the iluid.

i Fig. 26 is a modification of the device having a sweeping field directly energized by the line current.

Fig. 27is asection on'the linel21-2@1"ofig.'26.

Figs. 28, 29 and 80 are vector diagrams of the construction shown in Fig. 26.

Fig. 31 is a perspective elevation of a modified form of the device.

Fig. 32 is a detail of the apparatus 'fof Fia-31.

Figs. 33, 34, 35 and 36 are vector :d showing different phases of the operation of the device 1n Fig. 3l.

Fig. 3'/ is a view identical with Fig, 3'1-.except that it shows auxiliary chamber connected to the compression side of .the pump.

Reierring .now to ".Eigs. 1 to `3, Ythe.ruirnerall-lill designates a. core, G-shaped in cross section, of

laminated iron, having a primary coil I `upon -it,Y

wound to take the line voltage. The lnumerals 52 and a3 designate -auxiliary coresisimilar in shape tothe core tu, buthaving no primary coils thereon. Between the poles 54,55 and-56 of the cores 50, 52 and 53, there is placed a conduit 51, which, as will be seen in Fig. 3, is composed of two .insulated plates 58 and 59 Vheldinspaced relation by and sealed to a pair of conducting plates 60 and 6|.

This conduit 51 at its-ends bends upwardly and is joined to compression chambers 62 and-63,.si'm ilar to the compression chambers shown inmy former patent, U.1S. Ser. No.12,2587,4'15, `of October 7, 1941. These compression chambers* are alike and'each comprises generally alower expansion portion-64 in the formrof an upwardly expanding cone, a central portion N65 which .is -.generally cylindrical, and-a top .por-tion $6 which .is in the form of an inverted cone. The purpose off-the shape of these `parts is'to avoid loss of energy through sudden changes of speed of `thelow of liquid within the conduit. y Y

At its open upper end=each compression chamber is provided with an inlet -51 for vgas .under control of a -valve 68. Above each compression chamber is a cushioning chamber 59 connected with the -compression chamber through a valve seat under controlofa-ball valve 1I..` An outlet l12 for the compressed'gases.extends into the cushioning chamber 69 far enough so :that as-the liquid rises in 'the cushioning chamber vbeyond the lower-endof the outlet, theoutletvisclosedby the ball 1| and the force of the movementzoi the liquid is absorbed in compressing thegas trapped in the cushioning chamber. The twoirilet openings are joined to the source of gas by conduit', andthe two outlet openings are similarly ,joined by conduit '14. Y

Auxiliary 'conduits 1:5, 16 'and "11 extend from the center of the pole faces 54,5155 and, where they 'communicate with `4the conduit upwardly through the cores 50, y52 and T53 respectively, where they are joined together by acnimon conduit 18.

The conduit '51 and ione `of the compression chambers zare illled with a vconducting fllquid *'50, preferably having a good magnetic .permeability and electrical conductivity. Mercury'maybensed as the liquid, but certain-mercury r-alloysfhavingfa higher permeability, .sucht as those =containing iron, nickel or cobalt, fareiorthatreasonpreterable.

With the above construction, when the coll 5I is placed upon the line, the poles 54 of the core 50 are directly excited. They will hereinafter be called the main poles. The poles 55 and 56 of the cores 52 and 53 are in space quadrature and they are excited by currents within the secondary, that is. Within the conduit itself and the liquid itconttins. We use the fterm space quadratunvsfparticmarly 'apprnpriate when a plurality of primary poles are placed in series having inpposite polarity, with these consequent poles between them. In such case, if we regard the priniary'pdles as spaced 180 apart, in accordance .'with theirelectril-pOIarity, the auxiliary poles `willheunildway"between them and so may be considered as spaced 90. The whole arrangement maybe consideredlas a huge stator of infinite radiumwith 180 electrical angle from main pole to main pole. In any =case the auxiliary poles mustbeso spaced relative to the main Vpole that theycanfreact with the vcurrentsgenerated herein at 90 .phase angleto lthemain flux. The same principles `apply -andwhence the same language is-appllcable by analogy when only three .poles are'usedsince Athe .functioning of the device is similar and Vthe'spacing of the 4poles and the timing ofthe currents has the same relation. 'They willherein be called the auxiliary poles, It 4is possible. .to` employ a .plurality of these poles alternately They need not be salient poles, andmay take any form.

-Coil 5I acts .asqthe primary -of -a .transformer and transmits-power from the vline to the shortcircuit which consists of the conductorplates, Sil-Bi and the iiuidd. We may call this secondary current Iz. Motion ofthe duid past thepoleM oncestarteii generatesan E. M. F. and the :resulting current Isx circulates about theauxiliary .poles 55 `and liliy `and thereby excites them. Sincel'exds -mainly an .exciting current, it -is approximately in time Yquadiurture with -its E. M. F. The `el`d-of the-auxiliary poles isnearly initime-quadrature withthemaindeld and it is Ain space quadrature. -Asweepingiield is thus producedand it :moves inthe Adirection of motion of the rfluid. APower :is developed .by the reaction betweenlz.'thercurrentln the secondary, .and the auxiliary eldfiux.

-Motion of the fluid -under poles 55 .and 5B also generates anE.- MJ?. which acts in .approximate opposition -to the transformer secondary voltage at :pole -54 therefore `is yof the .nature of a counter-E. M. and limits the value of Iz. A transformer voltage about poles 55 and 5t similarly limits =the currents liu.

Motioncontinues until the iiuid reaches 'the cushioning .chamber 68 `whereitis deceleratedand stopped. -Sincewthe .auxiliary poles .B5-56 -owe theirsexitation to `Athegmotion of 'the duid, they are .de-energized when .themoton 'is stopped. The forcesnow :acting :on the fluid .are those .due to theenergy of compression in chamber "EB and the :difference .iin -elevation of the surface of Athe nid- :which @is now at the `bottom .of vone cylinder and fat the top of theother. This stored energy initiatesmotionin theopposite direction. When the direction of motion is reversed, `the direction ot'fI-u fis reversed-and .hence the polarity of vthe auxiliary polespisreversed. .The direction oflz is independent of motionand lis therefore not reversed. Since it ,reacts withthe=reversed-fle1d of poles 55 and\56,`the-.developed poweris reversed and maintainsrthe reversal :of flow .of the uld. VThiscontinues until rtheothencushioning chamber is reached whereupon the cycle repeats itself.

In the preferred formof the device, an electrically conductive huid which is magnetic is used. The reluctance of the air gap is therebyV decreased. Asmall exciting current is thus needed andthe efficiency and power-factor of kthe device are improved, the flux across the air gap will be increased'and the developed power is increased.

The short-circuit currents circulating` about the pole faces have a reactivecomponent that reacts with the iiux and a force is developed that is directed toward the center of the pole. Tubes 15, 16 and 11 (Fig. 1) are shown located at the center ofthe poles and are connected by a common tube 18. The fluid tends to flow through these tubes and away from conduit 51. During periods of maximum velocity in tube 51, the poles54, 55 and 56 have nearly the same strength. The force along tubes 15 and 11 will be nearly equal to that in tube 15, and little motion will take place inr common tube 18. As the fluid in` conduit 51 is decelerated, the strength of poles 55--55 is decreased and the force along 16 and 11 is then decreased. The fluid will then circulate from 51 through l5 and 18 and back through 16-11 to 51. The motion in this auxiliary circuit is a maximum when that in the main tube is a minimum. The auxiliary flows is in approximate timequadf rature with the main flow.

The motion of the fluid across the face of pole 54 to tube 15 generates a counter-E. M. F. that limits the current'Iz, since this E. M. F'. is in approximate time piiase with that due to flow along tube 51. Since the flows are in approximate time quadrature, there always exists a counter- E. M. F. to limit I2 and the power demand from the line is made more uniform through the entire cycle of operation.

As before mentioned, there is always some motion inthe auxiliary circuit directed along 18 away from pole 54. As the uid flows from 15 and 11 to 15 along 51, it is moving across the face of poles 52 and 53 in a direction opposite to motor motion and we have generator action. Since there is no outside connection, the generated energy is used in exciting poles 55 and 55. Motor energy for this generator is received from pole 5G Where the fluid moves toward the pole center. The motor takes a relatively small power current from coil 5| by transformer action. Since the generator current is largely an exciting current, it circulates in a reactive circuit and is in approximate time quadrature referred to the motor current. For the same power consumption we need a small current in the motor end and we get a large exciting current from the generator end.y Since a small additional line current increases greatly the excitation of the auxiliary poles, we greatly improve the power-factor of the f device, its emciency, and the output power developed.

The excitation of poles 55 and 55is obtained from the motion of two iiuid circuits; that in conduit 51 and that in tubes 15-11. Since these motions are in approximate time quadrature, we have a more uniform excitation throughout the operating cycle except at the point where the polarity of the poles is reversed. The output does not fallaway as the fluid in tube 51 is decele'rated, but vremains more nearly uniformwith consequent improvement` in efciency.

As shown; these tubes are balanced because this makes for the best correction during operation. If, however, one of these tubes, either 16 or 11, is omitted, an unbalanced condition will exis which providesa starting torquethrough eA.

The ux in the auxiliary poles 55 and 56, when excited by motion of the fluid along conduit 51, lags nearly in time quadrature with respect to that of pole 54. Motion in the auxiliary circuit increases theangle of lag of the resultant auxiliary pole flux. This corresponds to the condition generally obtained for power-factor compensation. The resultant primary current is rotated more nearly in phase with the transformer secondary voltage. Consequently the line current is rotated more nearly in phase with the line voltage and the power-factor is improved. By sufficient auxiliary pole iiux lag, over-compensation or a leading power factor may be obtained. Y

The various magnetic and electrical quantities may be shown as vector quantities on Figs. 4, 5, 6 and '1 of the drawings, but for the purpose of illustration these diagrams have been simplied by omitting resistance and reactance voltages.

In Fig. 4 pM is the flux due to the winding connected to the line. EMs is a voltage generated by the motion of the mercury through this flux. The corresponding current Iex excites a field A in time and space quadrature which is the eld of the auxiliary poles. This eld induces a transformer voltage EAT that is nearly in opposition to EMs. Their resultant Eex determines the value of Iex. EMIl is the transformer voltage induced by the main eld and is nearly in opposition with EAs which vis generated by motion of the mercury Their resultant E2 determines the value of the secondary current I2 which reacts with A to sustain motion. I2 also reacts with the primary and is balanced by a component current I1 from the line by transformer action. The line supplies the power which is proportioned to needs by a counter-E. M. F, EAs. The line current I1 is the sum of I2 and the exciting current Iex. Note that qm and I2 react to give power, that the auxiliary field is excited by motion of the mercury, and that the polarity is such that motion is sustained.

When the mercury has risen to the top of one leg of the U, it is decelerated and stopped by the cushioning chambers. The combinedrforces of the compressed gas and the unbalanced legs of U-tube initiate motion in a direction opposite to the previous direction.- This motion now excites the auxiliary poles with opposite polarity and the motion is sustained in the new direction until the fluid is again stopped and reversed by the cushioning chambers. The device is self-reversing and does not need extraneous devices `as reversing switches. It generates its own armature current and it excites its own field with the proper polarity.

The auxiliary path for the mercury comprising passages 15, 16, 11 and 18 is provided from the center of one pole to the center of the next. This means, as has been explained, prevents the line current from varying to locked-rotor values as wouldotherwise obtain when the mercury column is stopped.

As has been referred to above, the reaction between the secondary current circulating beneath the pole face and the fluxA of the pole produces the auxiliary path. However, when the velocity that the velocity in one path increases While thatY intheother decreases. The two moti-ons may, inra sense, be considered as being in time quadrature with each other. Since each motion must generate a counter-E. M. F., the currentV is -al- Ways kept within limits.

The vector diagram for conditions during inotlon is shown in Fig. 5; qiM is the main ux and EcMsisthe voltage generated by motion of the mercury to the center of the pole. The corresponding current IoM is largely an eXcitingcurrent and excites ilux oma. Reasoning as before, we-get the generated voitage Enos and the induced voltages Ecivrr and EMQT. ECM determi-nes the value of the current component IcM.. Similarly EicM determines the-value of the power component IaeM. Since all the vectors are inspace phase, they are added and we get a resultant flux bivio and current ICM which reacts with the stator anda balancing component Ilo appears in' the line current.

It has been pointed out that there is always some motion of the iiuid in the auxiliary path. Obviously, the main poles will act as motors and the auxiliary poles as generators since in the case of the latter, the direction of oW is reversed, in the sense that it is counter to the diirection o increasing lag of flux toward the polo center which is shown as (pso. The generator voltages are plotted in Fig. 6. oA is the flux of the auxiliary poles. Eens is a generated voltage andI since the motion is opposite, theI voltage isl opposite in direction and is in phase opposition with ci.. The volta-ge generated hy cutting this flux isEAos. Een and EAQT are induced volt-v ages. Adding ail the voltages We get ECA and the value oi ICA .is deternined. It is nearly entirely an exciting current since there is no connection with any other circuit and therefore aids one. psov adds with or. and the resultant is (psc. Since this has been rotated clockwise, the direction off the voltages in the mercury column: must be changed. These changes are shov/'n inl Fig.. 7.

We note that Ens has now been rotated clockwise andl the resultant E2 has been rotatedcoun ter-cioclcwise. Consequently, Iz now leads The component *he pr.. i, I1, nowleads andi thereby brings the l cu ent, IL, rnorenearly in phase with the voltage. We thereby have power-factor correction. W e nor.y know thatthe' device generates its own. exciting. power. Since the motion through the auxiliary tap. isi aminimum when the auxiliary pole flux is a maxi-muni and Vice-versa, this compensation canV beheld to limits.

In the form of' invention disclosedA in Figs. 8i to 10J there is provided but a singie core which is here designated as corresponding'to. the core 50 in the first embodiment. This coreV is also C- shaped and between its poles is placed; the. conduit 5T as in the preceding embodiment. In this embodiment, however, the conducting plates 8E' and 81 formingl the sides of thev conduit 5l are connected across the secondary 88 of a transformer 89, the primary' @il of which in series with the primary 9! on the core 35i. I prefer to connect this secondary S to the plates 35 and'!E by means of a plurality oi leads 92', these leads being connected to the transformer in multiple in orderto insure: as uniform a density oicurrent acrossthe pole'faces as possible.

' By this construction a-secondarr currentV is caused@ tonw; inthe fluid. which-'lsin phaselwlththeilux across the. conduit `hetween-the poleof theA coxief..V These. oonditioris-orefsuehas to causante'A umd tozuv through'-l the:V conduit 1in' e; direction the4 relative ,polarity civ coilsv 90 and .,Sl. Withvw thisY arrangement there. is provided a.. re-

versingiswitchrof; any conventionalzforxn', for

reversing. the' polarltyof one of these coils. This switch may1 be opemted'by :inv actuating member 93 mounteasshcwn, on` the cushioning chamber `69 andV having astem. extending into the pathofthe ballfvalvel 1l, sorthat the actuating member. V83 will operate the reversing switch as soon as thebal-l' valve strikes stem SI.

Inthe construction shown h1..Fig;.9thc transformer 89 is shown as of the constant current varietyandi its coil 9U is` shown as connected in series the' excttingcuil .9| .of the coro. The coil Sthas inherently ahigh reactance and. the transformer coil is deliberately designed for high reactance. The angle betwe'enthe primary current andthe flux of each of these unitscan thus be: made nearly the same.

Figr 10 isa plan view of the conduit 51 and illustrates the portion in theigap. .The.po1eface of core is shown in dotted .It is tobe noted that the mercury and. the conducting end walls constitute short-crcuited seccndaries in parallel with eachother and reacting through transformer action on coi-1 9| which thereby becomesy the primary of the transformer. The currents inthe short-circuits I'z and Iz. are not only a source of energy loss, but also adversely' affect the vphase relationship'between theY mercury current and the flux in gap 2 and thereby' seriously affectsthe motor action above described.

In Figs, 111-4 there: are shown the: approximate. vector diagrams of the conditions of this current, the Various quantities being assumedV to be representedby sinewaves.v

Fig. V11 indicates approximately the condition-s' obtained in core if no short-circuited cur'- rents existed. M is the flux, Eu the voltage, and I'they current inr the coil 9i. I andv 46u. are

nearly in phase. The phase angle a between I andv EM is nearly 90 and the'.A power consumed in. ei:-v citation iscorrespcndingly'low.

A constant-voltage:.transformer in series with the coil 9| under the` abovel theoretical conditions would perform approximately vas shown in Fig. 12. where all quantities are referred tothe prim-ary.- The transformer primary current is I. The secondary current IT is nearly inphase oplosition to- I andV nearly in phase with its voltage Fig. 13 combinesthose vectors. of Figs. -11 and 12'which react with each other inithe gap of4 core 85. -IT is nearly in phase with the ux ou and* hence, since they are alsoin space phase, a. m'otive. force. is` developed. that is: proportional tcy the projection of --ITY on qm. IT' indicates the' condition Where itis in phase. opposition t'o ouf which obtainswhen the current (or' qw) is reversed. a is the phase angle between. -IT- and the line voltage v.

Fig- 14 showsfthefactual conditions which; exist when: thei short-circuiti currentsf cxisfr'in the: conduit 51 where it. passes; under the pole faces of core 85 if thesame voltage drop En. is permitted across coil'U. I2 is the vector sum of I"z and Ifz (Fig. 1051. Inow4 is the vector-'sum of both I-and I""ex. Since- I is rre-longer lthe totalI exciting current, the fluxfdecreusesfrom aval-'ue of onuto rent.; The coil is connectedto the electrode transformer.

VFig. 9.

M, the latter corresponding If'ex which is less than I. A study of Figs. 11 and l2V indicates that IT is nearly in phase with I. Referring to ,Fig. 14, since I'is rotatedV away from c'M, so would IT be thrown further of phase. is then reduced by both a decreasein flux anda decrease in the projection of I'r on the iiux.,V In addition, I2 dissipates energy in the form of heat. The short-circuit currents should be minimized Vor utilized. Means and methods for achieving these results will be later described.

As is evident from the description of the operation of the device, the mercury column recipro- Cates in the conduit; itsspeed varies froml a maximum to standstill just before reversal. As it moves along the gap, a counter-E. M. F. is generated which limits the current flowing from transformer secondary 88. At a standstill, Lthe Ysecondary is, ineifect, short-circuited. The current demand from the line varies in accordance with the motion of the mercury and is greatest at ythe .dead period during reversal. Objectionable iiicker of ,lighting and energy losses might thus result. This effect is substantially corrected by installing a constant current type of transformer. The current demand isgmade more nearly uniform by utilizing the varying reactance of the constant current transformer.

Asimplifiedfform of such a device isshown in Fig. in which 'the transformer has been eliminated by placing a secondary coil |00 on a commoncore 10| with exciting coil |02 which now becomes the primary coil also and isconnected directly to a source of alternating curtube 51 cfa form lhereinbefore described fand is inl series with a reversing switch 92 which -is connected to a pilot circuit operated by the; rod 94. Thevsecondary current isapproximately in` phase with the primary current. Due tothe air, gap in the core I0|,` the leakage flux is great and the4 reactance is thereby increased.` .The voltage characteristic drops sharply and the performance approaches that of aconstant current Besides limiting the current rkand holding itfapproximately constant,` the current `across the electrode tube is-more nearly in phase with the `flux and the electromotive forcelinthc mercury is correspondingly increased. These conditions are vectorially plotted in Fig. 16. In lags considerably behind E, and is more nearly in phase (or phase opposition) with the ux c. Another form of this device is shown inFig, 17 where the :core is divided into three legs |21, |28 and |29. The coil |30 sets up a magnetomotive force whichA establishes the flux in legs |21 and |29 .which are magnetically in parallel with each other. The flux inleg |28 is a mutual ux Vand a voltageis induced insecondary'coilJSl which is connected across mercury tube 51 andv The motive force r vice.'"The ,power factor of the unit is apparently lowered at the standstill conditions, butit improves.greatly.V when theunitis in motion and in series withjga reversing switch 92. Coil |30 is connected to a source of alternating current. Switch 92 is actuated by thevrise and fall of mercury,as was described for the device shownin Thefcharacteristics of this device are similar to those of Fig- 15 except that the reactance is lgreater and the current regulation is better.

Fig, 18shows Va similar device except that an auxiliary air gap |34 has been added and the constant current characteristic is thereby improved. A movable piece |35 may be added for adjusting the length of the gap and consequently the transformer characteristics. f

The' vector diagrams of Figs. 13 to `16 have since the, tour-rent is substantially constant.; the

overall performance isimproved.

'.@v'Ihe vector diagram under static conditions for the devicesfof Figs. 15, 17 and 18 is shown in Fig. .11;9. lso isthe shorticircuit current in the mer;- cury, Imis fthe'.liliamsformery secondary current rand is theuseful current. I; is the sum1-of these currents. addedvectorially to -Iz to` give the total primary icurrentdngcoil |30., E Aisthe voltage nducedtby In;` is the exciting` current which, is

`the iiuxqs.V visithejimpressed voltage and is 4equal-to -TE'addedvectoriallyf-.to Ez, therprirnar'y -0 is the power factor angie.Y

impedance drop. The projection of IT on yis is a measure off the motiveforce developed in the mercury. Thefvector diagram for theV above devices when thefmercury is ymoving is shownin Fig. 20.` When the mercury .is-moving", along the air gap 'of= an excited` core, Eit* is jcutting;` lines ofA force and a voltage, Ec, 4is generated which is in opposition-.to thatV `voltage or ita Vcompunent that causes. It:` 4to ow across the mercury. This .componentis in phase `with the flux 1p. The line :voltage Vy, being a constant-must now supply Ec,-and(f-E,l-Ez). The lattervoltage,-V1, must thereforadecrease @as ,-E decreases, which in turn causes ,Etofdecrease.` With constantfvoltage transformationjE and. consequently. Ii would decrease. Remem- -bcring-that withthe mercury at rest at the Aend `of the strokeno .counter-E.M. F. exists andk also that as motion'is resumed in the opposite ...direction by the combined action of thel cushioning chambers, gravity, and the reversed secondary current, compression has;just,started and were 4,it not `for lthe inertia of the-mercury, a `.no-load condition would obtain.` With` such .transforma-V tion'the` current would vary` widely. By using ctmstar'it Acurrent transformation, however, `4-Iz would remain substantially.constantsfrornl noofby a varying power-factor whichwith constant potential transformation would remain ...nearly jconstant.- AThese vector diagramafor the sake'of lsimplicity, have been drawn 4with. resistance and reactance voltages omitted.

We have' called attention in connectlonwith 1Fig.,9 to `the fact that the flux in the transformer V,core 89A and the 4flux in Vt'he2mai-n core'85 1caribe made substantially the same.u VOn thisaccount these coils can be electrically connected inparallel.. Such an arrangementis shown in Fig., 21. .'IZhermotor core'|36 contains the electrode -tube f5.1 withina gapfasheretofore described. :f-Ihe f-coreisfexcited by coil I3ltconnected to a source of*- electrical energy. i `Theielectrode tube 51 is vconnected Iby wires l|38ito the -secondary |39 of :transformer whose primarycoil IM isCQn'- knected through `a reversing switch 92=in parallel to theesame' source of `energy as the coil |31.

,Thisgfswitch is actuated bythe rise and fallof the mercury. The operation of this unit is; similar tothatof. the `unit shown in Fig. 9.g

- iIn an alternate form, reversing` switch 92. is connected inthe line ybetweenelectrode ltube .51 and secondary coil |39. The operation of the unit load tofull load conditions. Excepting thevari` depends on thereversal loi Stheseoondary oft-heprimarycurrent, butisother-wise the same.

In Illig. 10 short-circuit currents, due :to 4trans- "iormerf actlon,were shown. -In- `v1l.ig.'120 we .see that, during movement,` a counter-'E.*M."F.,1Ec. is-generated;

Fign22 shows=thesame-form of'electrodevtube as Fig. 10. -Ii lmotion is as indicated; ythea-generated E. M. F.-shortcircuit path is-represented bythearrows. The ilux-is-directed 'upward from the surface'of-the paper. It -is obvious Ithatlthe field becomes distortedldue 'to-theM.' F. 4set upbyftheseecurrents. -Ifthepoles are :near satu- V`rationythe aiding-currents do notfincrease ilux,

dbut #the bucking currents will decrease the V'i'iux @their -cnd off the pole lace. The --net resultis increased -reactance and decreased il'ux wat `the poieiaee. The/lattercmakes for a'decreased motiveforce-in the mercuryand' increased-reactance.

Toaeifectivelyminimize all shortlcircuit currents :that v circulate V4around fthe outside ofthe pole face, insulating barriers such-asshown in Figc-23; which #is-af section cut asshown-linfllr are provided. These barriers |42 -across the path Y-: 'i"a!s'liort cirwitcurrent'fandv consistof an f insulating ma- ;teriei. ,'I'preier touse metallic stripscoated Ywith an Yifnsuhitingcoating 'TheQelectrodes 4be insulated at varioussectionsto aid inaccmpIisl-i- #inglthe lsainepurpose. Such areas' are indicated $5444. flsoythe electrodes may :be whollyeontained withinthepole face areaifthe Vmagnitude oift'heshorteeircuit currents justify this.

r'Urrinsulated metallic -strips Ymaybelocated 'alongtheetube'as |45, Fig. 23. Since theyfare ,conductors V#they do not 'prevent the messageY 0i the secondary current, -l'rutV aidv -in streamlining the flow of mercury andalso-serve'as stays t0 `rcinforcethe sides |46 against internal pressure.

vztjpreierredform is shown-in'Fig. 25. An -electricalrconductor "|41 isf-insulated electrically at lthe' sections H'and also servesasastay fort-he -sidesiof the'tubeasdescribed, above for |42 and 'M5-staying 'the sides |46.

A'A-modiflcation ofthe device is shown in-Fig. v26. :Themotor marais-shown rat |49 and is excited 'by the coil |50. This jportion of the vdevice lis designed forlowreactance. 4Two additional cores iareshowngat and |52 and ktheyaie excited respectively 'by coils |53 'and |54. This Vportion-oi' the device is designed for high Vreactance.y The rcoils |53 and |54are connected either in series witheachother -or in parallel, and this circuit is Vconnectedinparallel `or in series `with `the coil |50, but one of Athese circuits, as the coil |50, is controlled by a reversing switch 92.

The electrode tube 5T is-similar in constructionfto'gthose -hcreinbefore described, but; in this embodiment/it is extended through the Aauxiliary r`i5| vand |52.

operation-of this device'is similar'tothat shown in Figs. 8 and 9, except that the motor held '|45"isre'versed with respect to both the current ofthesccondary |59 of the transformer |80 Jwhich is connected lacross the electrode Ytube and the-holdsY of dtheauxiliary 'poles' |5|^and` |52. Aexoiting'eoil Ni -of this transformera-may -be1ccnnectedeitherin series yor in parallel-with .the :coils 1150, '|53 and |54, depending on 'its 5re- .octanoev characteristics as 'herelnbefore described. Also; transformer 1&0 maynbe Iof theconstant 'potential Aor constant current type.

'I'he section of'- the -electrode-tubef'l 'shown' in Fig.2'1=where the-poles |49, 15| and 52 'gare "indieated'thegshaded areas. 'The-leads fI|;Yis-;tr.a. :tbe :noted that the short-ecircuit' Acurrent Ifsefthatflies entirely .outside the motor pole I 49 also passes. under poles 15| and 152 which are, iinei!ect,;in space quadrature with the pole |49 `from an electrical standpoint. The ux oi' |5| .andi52 areiny phase opposition with each other -andout of phase, preferably-inquadrature, with .the motorl .pole |19'. J'I'he phase sequenceis in 'the direction of motion.

The vector diagram ofthe motor pole |49 is shown approximatelyzin Fig. 28 where Isc is the Vshort vcircuit current shown in Fig. 27, Eu is the induced `voltage and'Ii' is the coil current. 'The flux-.41M lagsf'behindthe .coil currentby the angle `a1.

A' similar diagram is .shownin "Fig, 29 forthe Aauxiliary poles 'ISI and: |52.y The Aflux is 4m and rthe short-circuit current in the mercury is IAsc whichalso passes under the main pole as shown Kin .'llig.A 27. EA is the 'induced voltage and 1" is `the exeitingcurrent. Ii iis the-primary current infcoll |531and" |54. 'Since'all coils are connected in YVseries, the currents 'Ii' and I1" are identical andthe-vector diagrams of Fgsf28 and 29 can becom'ioiried as shown .inFig '30 with I1 Aas the common current `andthe fluxes ou and -A lagging-'by-thefrangllesm andan. Ii -qba is the ilux in coilflthen mills, -of course. the ilux in coil "IIIandduristher ilux'in coil .which is equal lto'fthatci''-coil453 reversed. Since the fluxes are -noto-axial, they do znot' react'upon each other. lIt isvto be notedlthat they are distinct, in space displacement, and -that there is in effect a flux progression Vfrom-pole '|51 to V'pole V|82 lwhich is also the direction of flow of mercurydue tothe flux 1ini-pole--I49V1andthesecondary current of transformer |60 across the tube. The fiuxes of'Fig. 30 represent, in effect, atwo-phase system. I prefer to makel the angle (a1-2) between and' 41M as nearly 90 as possible. This may be approximated by inserting suitable inductance -and capacitance in the circuitas is common practice.

lItlwill Vbefnoted -fromFig 27 that Ise passes under the auxiliary poles. `'Since Iie lags behind oir-(F1528). and sordoeso. lagbehind 4M (Fig. 30) ,.Ifsc and osare-nearly in phase or, as shown in"1*ig.' '30,acom'ponent Yof'Ise entirely in phase with op. and consequently we get motor action 'in the direction vthe mercury has been made to move-as mentioned above. vIn other words, the portion Aof the short-circuit current entirely around thefoutsideU-oithe pole (as Iz in Fig. 10 or Iwin Fig. 27) has been made to exert a useful force.

It -is also f to v-be-noted the short-circuit current entirely von'the outside of auxiliary poles, Ixsc,

passesfacross main pole- |48 and by similar reasoningfand as indicated in Fig. 30 exerts useful `force-in the direction of motion.

In the form of the invention disclosed in Fig. 31,- two cores, `r|80 and IBI, are provided. These coreslare-ofiidentical construction but oppositely placed.l Each is-generally of a C-shaped cross 4for lone core-only. and the numbers applied to both.

"The central pole of each core, |b or |8|b, is energized by coil |82 -directly-irom the line 'H2 are-fromthe-secondary 4ofrthe^transformer. `NS and these coils '-are preferably connected in se- 3 Q Y ries.l Each legofjthe conduit 51 terminates `in a compression, chamber, suchas we have previously described.` Abyepass conduit |83` leads from the center of the face of the magnet'pole opposite the centeroffthe central sub-po1e,` as shown at |84, to aj point opposite the center of one of the sidesub-pole's, as shown at |35.` This by-pass conduit can, if desired, also be connected to thecente'rl of the other sub-pole face, as shown in the construction ofFig. 1, butinl this embodimentl yhave chosen to illustrate it in the unbalancediform with both cores unbalanced in the same direction of'motion of the mercury, in order to afford a starting torque.

` f The vector diagram showing the reason for this starting current is shown inFig. 36, in which pM is the main pole iiux and 4u. is what we may call the leakageiiux through one-half of the auxiliary pole. 'EMT vis induced voltage circulating in n i thegmercury around Vthe main pole and ELs is the counter-E. E {d'ue to leakage flux. The resultant voltage EM and its current determine thevpo'wer'takento causevmotion of the mercury auxiliary path.V EMs is a, voltage generated in the by motion of the mercury'under 45M and Eur is a voltage induced bythe leakage iiux. The resultant vvoltage life` causes a current IG to `circuu late, and a flux dueto this current aids the leakage ilux Las may be seen from the gure. We seezthen'thatwe have strengthenedl the flux in the` auxiliary pole and that this flux is out of Vphaserwiththe main pole., As a result, we have a sweeping field even before theA mercury starts in motion` in` the main conduit, which can initiate the movement of .the mercury in the main conduit, and thus `vthe device becomes self-starting because ofthe unbalanced connections of the auxiliaryconduit. i y

It will also be seen that the power factor correction can be veiected by the by-pass conduit going through the main `pole alone,1if it be connected to themain vconduit outside the main pole, since kthe flow of mercury into the auxiliary conduitgenerates an E.M. F.`which causes a current to circulatejin' themercury in V quadrature `with the main iiux.: The flux due to this current combines withgthe main ux `to produce a resultantL flux lagging behind the main flux. The

flux in the auxiliary-poles isthusrotated with a consequent improvement in` power factor, as indicated in Fig. 7; Wewould getfthesame result, only in less degree, if the. center tap conduit be placed in the auxiliary pole instead of the main pole. Q

The auxiliary poles |80a and lcand 18m and ||c are low reluctance magnetic paths and are excited by virtue of an M. F, generated by motionf the mercury by the poles |8017 and |8|b. When Vthe mercury is in stationary Vvcondition, a starting torque is provided by the unbalanced ow of mercury through the conduit |83, which energizes the pole |80c and I8 |c thereby producing asweepingeld which Ican start the mercury in motion., t j t There is also shown in this figure an auxiliary chamber |90 connected to thevcenter of the bottom portion of theU-conduit 51.

During the operation of the device, this auxiliary chamber functions as followsrMercury descending from. one of thek legs ofthe conduit 51, say for example the left-hand leg, will` divide. Part will go up the right-hand leg,whi1e part' will go up into the auxiliary chamber, |90, compressing `the gas therein. y`Whenthe mercury inthe righthand leg has reached its upper limit and is iliary chamber stoppedforreversal, all further iiow of mercury from the left-hand legis diverted into the'aux- |90. Thus it will be seen while the mercury in the right-hand leg is stopped, that in the left-hand leg continues in motion for a time.

l will be seen that the ow inthe right and lefthand legs is out of phase with` each other, and

while the mercury in the left-hand leg is decelerating and its counter-E. M. F. decreasing,I the mercury, in the right-hand legis accelerating and its counter-E. IM. F. is increasing, r

Shortly afterwards the mercury in thenlefthandleg is stopped and reversed by the combined forces of gravity, the cushioning chamber, the force developed in the right-hand leg and that due to the gas compressedin the auxiliary chamber. The reversal is quickly affected since these forces operate upon onlyv a portion ofthe total mercury. The entire mercury column is `now reversed and the gas in the chamber |90 expands and returns its energy to the moving column. 'I'his action now continues until the mercury rises tothe top of the left-hand compression chamber 65 and the above mentioned cycle is repeated. f L The vector diagram showing the operation just described, omitting, for the moment, the ley-pass conduits, is givenin Fig.k 33 in which the main pole flux M induces' EMfi` by transformer action. The motion of the mercury by the pole generates EMs. Since there` is no outside connection, EMs

f isof thenature of `an exciting voltage for the auxiliary pole and the flux 45A lags bynearly 90.

` This ux also inducesA a transformer voltage EAT which reacts with EMs to limit its valuek to the resultant, EAR; `Motion of the mercury bythe auxiliary pole generates EAS which is of the natureof a counter-E. M. F. and it `reacts with anjd reduces the short-circuit voltage under the u main pole to EMR.

, From. 'the foregoing it'will be ,seen that the mercury .moves into and out of the auxiliary chamber |90 at double the frequency of the` mercury in the main conduit. It will also be seen thata counter-E. always exists in the line and the currentdemand is made more uniform throughout the cycle of operations.

In the foregoing description the mercury is stopped and reversed, first, inone leg of the tube, and then reversed in the auxiliary chamber |90.

,The force-,speed characteristics and the volume "of the auxiliary chamber, however, may be so chosen as to make the opposite true, so that the `mercuryis reversed, rst, in the auxiliary chamber. and ,the` low. level `of mercury leg, and therefafter` reversed in the legwhere the mercury is at its high level. The advantages above referred The auxiliary chamber may be openV to the atmosphere,\as shown in Fig. 32, it'being understood, of course, that in such instance it would behigh enough so that all the `energy stored in itis stored in the form .of a columnrof mercury. It. may on the other hand, if desired, be vented .toreithe'r the suction or the pressure side of the system.

www

15 In the ypreferred form where Athe .auxiliary chamber is `open at the .top, a iloating ball valve isfsupplied to prevent the escape of mercury into the system andalso to prevent gas from entering the electrode tube. These two functions may be combined by one ball valve, as shown in Fig. 32.

VThe auxiliary chamber k224 containing mercury 22S-is connected .to the electrode tubeconduit 226,

and floating onthe surface` of the mercury isa `ball valve 221. The venting pipe 228` is provided fwith a neck to form a valve seat at 229, and another seat is provided at the bottom, as shown at 23,0. Thus the mercury level cannot rise above the-point where the valve 221 closes the neck 229,

,nor-can it .fa-l1 below .the point where the. valve closes the neck-230.

It will also be clear that the upper venting 228 of the auxiliary chamber 22.4 may be connected by suitablevalves .with the `inlet and exhaust 13 and 14, so that the movement of mercury within this auxiliary chamber contributes a quota, toward the useful compression of the gas.

Fig..33 shows therelationship between the primaryand secondary values of the electromagnetic system of Fig. 31 .of the exciting coils |82, indicating the reection of power from the line into the mercury system..

The secondary voltage in the mercury is EMT and thesecondaryor mercury current is I2. These arethe only Vtwo that react .with the primary. All other voltages are duedirectly or indirectly to `the motion of the mercury -past the main pole.

All-other currents have acomponent that appears in Iz.

.Referring to Fig. .33, EivrrA reacts on the primary ycoils and it has a component -EMT in the primary. Under standstill Aconditions I2 would be determined by Enr andthe secondary impedance. The .velocityof the columns depends on resistance and excitation requirements, and this determines the value ofthe counter-E. M. F. which modifies Eurto some value such as EMR. This, .together with the impedance, determines the .value of `Iz duringv running conditions.

Fig. 34 is the motor-generator vector in which E is theline voltage, IM Yis the motor cur-rent, that is, the primary current required to drive the mercury, and VIe is the vgenerator current, that is a current in the mercury that circulates vabout and excites the auxiliary pole. 'The power is` the product ofthe voltageand the projection I1 of the current IM, and I2, the projection of IG, which are herein shown equal. VIior the same power ,a small motor current IM can cause a much larger exciting (low power factor) current IG to `circulate. Note the improvement in powerfactor from G to 0M in so far as the line is concerned. It is obviously better to generate IG than to draw it from the line.

Since certain changes may be made in the above construction and different embodiments of the invention could be made withoutV departing from thescope thereof, it is intended that all matter contained in the above description or shown in theaccompanying drawings shall be interpreted as illustrative and not in alimitingv sense.

It is also to be understood that the following claims are intended Yto cover all of the generic and specific features of the invention herein described, and all statements of the scope of. the invention which, as amatter of language, might be said to fall, therebetween.

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

1. Adevice ofthecharacter .described compris.- ing, incombination, a conduit. adapted to contain a conducting liquid and having a pair of electrodes within-it spacedl from each other ina direction transverse to the conduit, a magnetic core havingy its poles embracing said conduit and spaced in a direction transverse to the spacing of said electrodes, an electric coil for energizing said core to create .an alternating magnetic flux across saidconduit creating a. magnetic field, and means associated Withsaid conduit for modifying said magnetic eld to cause the same to sweep in the direction of extension of -the conduit.

2. A device of the character described-comprising, in combination, a conduit adapted to Vccntain a conducting liquid and having a pairof electrodes within it spaced from each other ina direction transverse to the conduit, a magnetic core having its poles embracing lsaid conduit and spaced in a direction transverse to the spacing of said electrodes, an electric coil for energizing said core to create an alternating magnetic ux acrossh said conduit, an auxiliary core having its pole pieces embracing said conduit adjacent to said energized core, whereby movement of a conducting liquid within said conduit through the flux of said energized core will energizesaid second mentioned core, thereby creating a sweeping field.

3. A device of the character described comprising, in combination, a conduit adaptedfto contain a conducting liquid and having a pair of elec,- trodes within it spaced from each other in a direction transverse tothe conduit, a magnetic core having its poles embracing said conduit and spaced in a direction transverse to the spacing of said electrodes, an electric coil for energizing said core to create an alternating magnetic ux across said conduit, an auxiliary core having its pole pieces embracing said-conduit adjacent to said energized core, whereby movement of a conducting liquid within said conduit through the flux of said energized core will energize said second mentioned core, thereby creating a sweeping eldl and a by-pass conduit connecting a point in said conduit at the center of the pole face of said second mentioned core with another point in said conduit.

4. A device of the character described comprising, in combination, a conduit adapted to contain a conducting liquid and having a pair-of electrodes within it spaced from each other in adirection transverse to the conduit, a magnetic core having its poles embracing said conduitand spaced in a direction transverse to the spacing of said electrodes, an electric coil for energizing said core to create an alternating magnetic ux across said conduit, a pair of permeable cores, each having pole faces embracing said conduit. one on each side of and adjacent to saidenergized core, whereby motion of a conducting liquid within said conduit through the magnetic eld of said energized core will energize said adjacent cores out of phase with the ilux in said energized core, whereby a sweeping ux is obtained.

5. A device of the character described comprising, in combinationa, conduit adapted tocontain a conducting liquid and having a pair of electrodes within it spaced from each other in a d1- rection transverse to the conduit, a magnetic core having its poles embracing said conduit and spaced in a direction transverse to thespacing of said electrodes, an electric coil for energizing said core to create an alternating magnetic ilux ,conduit connecting a point infsaidponduit at 17; across Said Conduit, ,ai n, innig-permeable each having `noie fas embracing Saidcndut. one on each side of and adjacentgto,saiduenersized Core. wherebyimotionai ,alwnduetinsriliquid within said- :conduit through the ,magnetic fieldof said energized Vcore will energize sa adjacent coresout of phase with the Vfiux, i H said ,energized core, whereby ,.a.. 'sweepingffluig Aisfobtained', anda by-,pass conduitfccnnectiriga point in said'conduit at .the center or alpole `face'cf'the energizedcore with apointinthe conduit atthe center of the pole face of one of said'othen cores.

' 6. A device of the characterdescribed comprising, in combination, a conduit adaptedvtocontain a conducting liquid and having ,a-.pair of electrodes within'. it spaced from each otherlin adirection transverse to theconduit, a'magnetic core having its poles embracin'gl saidv conduit"`{an d spaced in a direction vvtransierseto thes'pacing of said electrodes, an electric'coil for energizirigfsaid 20 core to create an alternating magneticflux across said conduit, a pair vof'p'e'rin'efable`cies1`eaclrihav;- ing pole faces embracing jsaidcon'duit, .,ofnei'on each side of and adjacent to sa'i'denergize'd' core, whereby motion of va conducting liquidwithir'isaid 2 conduit through the magnetic field oi"`said"ener gized core will energize ysaid adjacent coresoutfof phase with the flux in said'energiz'ed core, whereby a sweeping flux is obtained, and an auxiliary the center of a pole ,facefofftlieenergized core,

with points Vin saidjconduit'atthe centers of the pole faces of eachof said otherv cores.' Y 7. A device of the character described compris#- ing, in combination, a conduit adapted'tocontain 3 'a conducting liquid and l"liavingfa e pair of i electrodes vwithin' itfspaced fromeachother in a direction transverse to' the conduit, a magnetic core `havingits poles embracing said conduit `and spaced in a direction Vtransverse vto the spacing Vcf 4 said electrodes,4 janelectric coil for venergizing said core to create an alternating magnetic flux across said` conduit, and lfmearisv vfor passing` across said conduit from one ,electrode` toftheothera current substantially in phase with the flux withi- 45 in said core, including av constant 'currentjtranse former Vhaving its primaryfin series withy the primaryof the exciting coil. q e 8. yA device'of the character describedcomprising, in combination, aconduit adaptedto contain 50 a conducting liquid and* having ajpair of elec-iV trodes within it spaced from'weach otherfin 'a direction transverse to'the `conduita magnetic core having its ypoles embracing said conduit; and spaced in a direction transverse tothe spacing `oi 55 said electrodes, an electric coil for energizing'said core to create an alternating" magnetic flux across said conduit, and means for passing'across said conduit from one electrodeto the other a cur-V rent substantially in phase with the flux within 60 said core, including a "secondary coil fuponsaid main core having its terminalsconnectedtosaid electrodes.

9. A device of the character described comprising, in combination, a conduit adapted to'contain a conducting liquid andhaving a pairof electrodes within it spaced from each other" inv a direction transverse to the conduit," a magnetic core having its poles embracing saidvconduit and spaced in a direction transverse to thespacing of said electrodes, an electric coil for'eneifgizing said core to create an alternatingimag'netic flux across said conduit,and means for passingacross said conduit from one electrode to the other a current substantially in phase with the', fluxwitli-i 75 ,ini said, core, including a secondary coil upon a closed branch of said main core having its-terwz riinals connected to said electrodes. Y

)1J0.,A` device ofthe character describedcomp'rilsirig,I ir'icornbination, a' conduit adapted `to con- 4,tainacohducting,liquid andhaVing avpair of electrodes within.r it spaced fromeach other in a direction 'atransversefto the conduit,` a magnetic fcore having its poles embracing said 'conduit and A in'fa direction transverse to' tne spacing idfelectrodean electric coil for energizing Qsaid, cor'efto" create an alternating magnetic fiux id conduit; and means for passing across said frduit` from one electrode to the other a currentsubstantially lin phase. with the flux fwithin' said core; including a'secondary coil upon va separateportionofsaid main core having an 'adjustable `air` gapthereimand having its termiiials connected to ysaid electrodes.

f. A.lLjA .device of the character described comprising, in combination, a conduit adapted to contain a conducting liquid and having a pair of velec-- trades within it 'spaced from each other in a 'dre'ctionhtransverse to the conduit, a magnetic core having its poles embracing said conduit and spaced in adirection transverse tothe spacing ,of 'said`.:eleetrodes, an velectric coil for energizing .said `coreto create an alternating magnetic flux across said conduit, a pair of auxiliary cores, each having pole faces embracing said conduit between themgand jeach ofsaid auxiliary cores having a jniagnetizing" coil, Vvvljiich jmagnetizing coils are vconnected together, auxiliary means for passing an Aelectric current through said conduit from one ',electrod'efto the other, including a transformer 'having its secondary connected across said electrodes, andineans for reversing the current in 'the energizing coil for lsaid rst mentioned core. 12. A" device in accordance with claim 1, having a magnetic electrically conducting liquid in said conduit." f 1 e 1 L 13. A device in accordance with claim 2, having a magnetic electrically conducting liquid in said conduit. f `5 f Y v14. device of the character described comprising,inc`ombination, a conduit adapted to con'- 'tain a conducting liquid, a magnetic core having pole" 'faces embracing Vsaid conduit, an electric coil forenergizing said coreto create an` alter- "natirigl magnetic flux across the conduit, and means associated with said conduit for modifyingsaid magnetic ield to cause the same to sweep in the direction [of extension of the conduit.

Y -lpA'devicecfthe character described comvprising.in'coinbiiiation, a conduit adapted to contain a conducting liquid, a magnetic core having pole" faces embracing said conduit, an electric coil for energizing said core to create anjalternating magnetic ux'across the conduit, :an auxiliary -core having pole faces embracing saidfconduit adjacent'to the energized core, whereby movementofa conducting liquid within said conduit through the flux of said energized core. will energizeysaid second mentioned core, therebycreating a sweeping field.

16.1Adevicelof the character described com- 'prising,f in combination, a conduit adapted to containa conducting liquid,`a magnetic Vcore having pole, faces embracing said conduit,lan electric coil'ffor energizingsaid core to create an alternating magnetickfluxwacross the conduit, means associated withr said conduit for modifying saidfelectrical eld to cause the same to sweep in'thedirection of extension of the conduit, and

fonemas a 'magnetic electrically `conducting liquid within spaced in a Vdirection'transverse'to'thespacing of said electrodes,"anelectric'coil-for `energizing saidV core to Vcreatean alternating' magneti'cux 'across "said'con'duit, anauxilia'ry core havingits pole 'faces "embracing Vsaid conduit adjacent'to said energized core,` whereby movement `cfa con- "ducting liquid within said'rcondult through the 'flux of sai-d energized core will energize's'aidsecond mentioned core, thereby creatinga sweeping `fleld, and a by'pass conduit connecting a'point' in said conduit atthe center 'of 'thepole face of said energized core with another 'pointinsaid conduit.

18. A device Vof the character described comprising, in combination, a conduit adapted to contain `a conducting liquid and having a. pair .of electrodes within it space'd'r'om each other' in a `di 'rection transverse' tothe conduit, a magnetic core having its poles embracing said conduit and spaced in a direotiontransverse tothe spacingrf said electrodes, an electric 'coil' for energizingsaid core to create an alternatingmagneticflux'across said conduit, a permeable Y core adjacent .to vsaid iirst mentioned coreV having pole faces embracing said conduit, whereby motion ofaa conducting liquid within said conduit through the magnetic liield of. said energizedrcore willenergizesaid adjacent core out .of phase vwiththe flux infsaid energized core, whereby a sweepingeld is obtained, and a by-pass conduit connecting a point in said conduit at the centerof a pole=face of the energized core with a point in the conduit at the center of a pole face of -said permeable core.

19. A device of the character described .comprising, in combination, a conduit Yadapted to contain a conducting liquid, a. pluralityofpaths for said conducting liquid, Ymagnetic meansifor moving a liquid within said conduit along each of said paths with i alternating motions out l. of phase with each other, whereby ,-.atsubstantlally continuous flow is obtained,

20. A device of the-character descrlbedcom prlsing, in combination, .a conduit :adapted to contain a conductingliquidfsaid conduithaving means to permit oscillatory-motion of liquidin one part, out of phase with oscillatorymotion in another part, a plurality of electrical meansior inducing a reciprocating ilow of a conducting liquid within -said'conduit, through said conduit, said means being lout f of'phasewith Veach other, whereby theV reversalof 'thet liquid4 in one'portion of said conduit occurs-whiletheli'quid'in'anothcr portion of the. conduitisy moving.

21. Inansystem for-reciprocating afconiiucting liquid through .a conduit .against saloadby electromagnetic means, :including a Ymagnleticilield lreacting upon the liquid.themethodzofireducing the startingcurrent which comprisesproviding an .unloaded byfpass .for said -.conducting `liquid from within said conduit.

22. In a device forthe=reciprocation.ofa conducting fluid througha conduitbyfmeansof. an alternating Viiux passed across the conduit, .the method of 'improving the =power Sfactor which comprises impressing upon the conducting liquid a. kvoltage in suchphase'asto shift 'the "phase of 2'0 oscillatotymotion ofthe secondary current inthe :liquid in "the: direction of -Ilead, with vrespect 1 to themain-voltasa "-23.\"-In"a'de"vice foi-fthe! reciprocation of' a 'con- `duetlrig-:ill'iil through a 'conduit` byA meansf'of 'an Aalternatingflux passed `across -the conduit, the :method of reducing the starting current which comprises-providing an initially unloaded pathyway tor=sai`dconducting liquid-and starting'a 'portionpf *said liquid in 'motion Vthrough 'said pathway whilenot urrder 'load' before'starting'the Yrem'aii'fdcr11i-motion.

24. "A device or the 'character described com- 'prismg a"'coniduithavmg"a pair'ofelectrodes iongitudmaliydisposedtnerein, a'compressionciiamlier^at^each'e1'iioibsafdconduitl including avalve mechanismitorpumpingiiufd by themovement 'of 'said conductingiiuirhapairof motorunitsfeach tcomprising la lmagnetic core `having `its poles 'spaced apart in 'a Jdirection `transverse to 4the spacing 'dinard-"electrodes :to -embrace vsaid conan `electric'coii for energizing' said core to create "anlternating'magnetic -iiux across `said conduitfandmeansassociated with said conrduitior 'modifyingsaid "electric'ileld' to cause' the .same tonrove 'a "conducting liquid contained within'safcl 'condintintthe'direction of-motion of the:conduit. `anraxrauxiliary'chamber connected to'safd conduitmtermediatebetween said. motors "whereby the "movement of 'said `conducting 'iluid through 'oneof saldmotors maybe out ofphase witnthe 'movementof thei'lui'dlthrough' the other of said motors.

'25. Aldeviceof the characterdcscr'lbed comprising. a .conduit'having apairoi electrodes longitudinally 'disposed therein, 1.a compression chamber at eac'hqendof. said conduit lincluding a valvel mechanismjfor.v pumpingffluid .bythe movementor said conductingiiluid, a pair Aofmotor units, eachcomprising.magnetic corehaving its poles .spacedapart ,in .a direction transverse to lthespacing ,of -said .electrodes .'.to Vembrace `said conduit, andian electric coil'or .energizing said core.to.create an. lternatingmagnetic-'ux 4across said conduit', eachrofssatdsmotorsLhaving :a `core havingitsmole faceiiivided into `apluralityof pole sections, one o! said sections beingenergized by said coil whereby .the movement of.said conducting .'uid gpast said .energized section Awill .in- `duce secondary .currents Lto energizessaid .other section whereby, saidqelectrlcfii'eldis -caused `to sweeplin the (direction lof .extension `of the .coniiaiand anauxiliarychambenconnected. to Isaid conduit intermediate between. saidmotors wherebythemovementofsaid.conductinguidthrough .one of motorsimay-bevoutof .phase 4with .the movement of throught-the other-,of said motors.

26. bA vdevce of the n.character A.described r.comprsing` a,v conduithavng. a pair of. electrodes longitudinaLLv. disposed therein, acompression chamber at each end of said conduit includingifa-valve mechanism fonpamining` uid Aby the movement of saidlconductinghuid. a. pair` of .motor'units, each comprising. ,--a magneticrcore rhaving Aits .poles spaced -apart dn a -.dlrection ttransverse to fthe spaninglofsaid-.electrodes-to embrace said conduit, andanfelectriccoil;forenergizing.Y said core to create. an alternating-magnetic. flux across. said conduitweach of..saidtmotors..having va eorehavingf. its lpole .'fawce` divided.r into three .pole sections, the center of Vone. oi Said sectlonsnbeing energized by saidcoil whereby the movement of said conducting nuidpast said energized. section willinduce'secondary currentsto energize each o'f said othersections whereby saidelctric eld is caused to sweep inthe direction of. extension of the connduit in either direction of motion, and an auxilialry` chamber ,connected tofsaid conduit intermediate between said motors whereby the movement of said conducting iiuid through one of said motors may be out ofphase with the movement of theffluid through the other of said motors.

current substantially in phase with the flux within said core, and auxiliary poles embracing said conduit adjacent to the poles of said core wherebythe short circuit currents in said conduit are spacing of said electrodes to embrace said conduit, and an electric coil for energizing said core to create an alternating magnetic flux across said conduit, each of said motors having a core having its pole face divided into three pole sections, the center of one of said sections being energized by said coil whereby the movement of said conducting fluid past said energized section will induce secondary currents to energize each of said other sections whereby said electric eld is caused to sweep in the direction of extension o f the conduit in either direction of motion, and an auxiliary chamber connected to said conduit intermediate between said motors whereby the movement of said conducting uid through one of said motors may be out of phase with the movement of the iiuid through the other of said motors, and an auxiliary conduit connecting a point in said conduit at the center of the energized pole with a point o n said conduit at the center of one of said other poles.

28. A device of the character described comprising a conduit having a pair of electrodes longitudinally disposed therein, a compression chamber at each end of said conduit including a valve mechanism for pumping fluid by the movement of said conducting fluid, a pair of motor units, each comprising a magnetic core having its poles spaced apart in a direction transverse to the spacing of said electrodes to embrace said conduit, and an electric coil for energizing said core to create an alternating magnetic ux across said conduit, each of said motors having a core having its pole face divided into three pole sections, the center of one of said sections being energized by said coil whereby the movement of said conducting iluid past said energized section will induce secondary currents to energize each of said other sections whereby said electric eld is caused to sweep in the direction of extension of the conduit in either direction of motion, and an auxiliary chamber connected to said conduit intermediate between said motors whereby the movement of said conducting uid through one of said motors may be out of phase with the movement of the fluid through the other of said motors, said auxiliary chamber being open at its upper end and having a check valve in said opening.

29. A device of the character described comprising, in combination, a conduit adapted to contain a conducting liquid and having a pair of electrodes within it spaced from each other in a direction transverse to the conduit, a magnetic core having its poles embracing said conduity and spaced in a direction transverse to the spacing of said electrodes, an electric coil for energizing said core to create an alternating magnetic flux across said conduit, and means for passing across said conduit from one electrode to the other a reduced and a iiux is generated in said auxiliary poles, which assists in driving said liquid,

30, A device of the character described comprising,V in combination, a conduit adapted to contain a conducting liquid and having apair of electrodes within it spaced from each other in a direction transverse to the conduit, a magnetic core having its poles embracing said conduit and spaced in a direction transverse to the spacing of said electrodes, an electric coil for energizing said core to create an alternating magnetic flux across said conduit, and means actuated by Vmotion of the liquid through said conduit for giving a sweeping motion to the ilux in the direction of motion, whereby the motor force generated by the flux ceases when the motion ceases and is reversed when the motion ceases and reverses. y

31. A device of the character described comprising, in combination, a conduit adapted to contain a conducting liquid and having a pair of electrodes within it spaced from each other in a direction transverse to the conduit, a magnetic core having its poles embracing said conduit and spaced in a direction transverse to the spacing of said electrodes, an electric coil for energizing said core to create an alternating magnetic iiux across said conduit, an auxiliary core having its pole pieces embracing said conduit adjacent tosaid energized core, whereby movement of a conducting liquid within said conduit through the flux of said energized core will energize said second mentioned core, and auxilia'ry means for energizing said auxiliary core.

32. A device of the character described comprising a conduit having a pair of electrodes longitudinally disposed therein, a compression chamber at each end of said conduit including a valve mechanism for pumping uid by the movement of said conducting fluid into said compression chamber, a pair of motor units, each comprising a magnetic core having its poles spaced apart in a direction transverse to the spacing of said electrodes to embrace said conduit, and an electric coil for energizing said core to create an alternating magnetic flux across said conduit, each of said motors having a core having its pole face divided into three pole sections, the center of one of said sections being energized by said coil whereby the movement of said conducting fluid past said energized section will induce secondary currents to energize each of said other sections whereby said electric field is caused to sweep in the direction of extension of the conduit in either direction of motion, andan auxiliary chamber connected to said yconduit intermediate between said motors whereby the movement of said conducting fluid through one of said motors may be out of phase with the movement of the uid through the otherfof said motors, said auxiliary chamber connecting at its upper `end to the compression chamber and having valve (References on following page) i REFERENCES CITED fgirllm'xlg 'rfrences are df record "1n" the `l-f this patent:

Vl'JN'I'IIED` STIYFES PATENTS NlxmberN fNe (Date 630;'197- {Mershon Aug.f2o; 19o1 .9575242 "Nerg'erth 'May1o, 1910 298,664 kCflfubb 4Apr. 1,3191`9

Images