|Publication number||US3906415 A|
|Publication date||Sep 16, 1975|
|Filing date||Jun 14, 1974|
|Priority date||Jun 14, 1974|
|Publication number||US 3906415 A, US 3906415A, US-A-3906415, US3906415 A, US3906415A|
|Inventors||Baker Richard H|
|Original Assignee||Massachusetts Inst Technology|
|Export Citation||BiBTeX, EndNote, RefMan|
|Patent Citations (3), Referenced by (23), Classifications (11)|
|External Links: USPTO, USPTO Assignment, Espacenet|
United States Patent 11 1 Baker APPARATUS WHEREIN A SEGMENTED FLUID STREAM PERFORMS ELECTRICAL SWITCHING FUNCTIONS AND THE LIKE  Inventor: Richard H. Baker, Bedford, Mass.
 Assignee; Massachusetts Institute of Technology, Cambridge, Mass.
22 Filed: June 14, 1974  App]. No: 479,244
 US. Cl. 335/47; 200/214; 335/49; 335/51; 335/55  Int. CL I-lOll-I 29/24  Field of Search 335/47-58; 200/192, 214; 137/807; 310/11; 417/50; 204/200  References Cited UNITED STATES PATENTS 2,744980 5/1956 Bellamy 335/51 3289126 11/1966 Hurvitz 335/47 1451 Sept. 16, 1975 1308, 105 3/1967 Hurvitz 335/55 Primary Examiner-Harold Broome Attorney, Agent or Firm-Arthur A Smith, Jr;
Robert Shaw; Martin M. Santa  ABSTRACT Apparatus wherein a segmented fluid stream consist ing of a plurality of contiguous segments of, for example, mercury and ferrofluid, are caused to move within an insulating tube whose inner volume is filled with the fluid and to perform electric-related functions by virtue of the different electro-magnetic characteristics of the fluids that compose the stream The system described in greatest detail is an electric switch wherein, in its simplest form electric current between two electrodes will flow or not depending upon whether the space between the electrodes is bridged by mercury or ferrofluid. respectively 66 Claims, 27 Drawing Figures PATENTEUSEP 1 61975 3,906, 11 5 SHEET 2 UF 8 i I Q I 'r H FF FF- "9 FF d Hg FF i SIC] Hq F/G. 4/]
r was Wm FF 5% I FIG. 45
SOURCE F E CTR PATENTED SEP I 81975 SHEET 3 UP 8 6A I/ FLOWv-j 6E [48 Flow-+1 FIG. 7 6F-7 GI 4D LL I FLOW- I Fzs POLYPHASE 2w SOURCE FF LOAD 6A 27 etc. 26A Hg r 6B 268 SPECIAL SOURCE 0F 60 FF PURPOSE ELECTRIC 6C 26C COMPUTER POWER 6F etc. I) GE flail PROGRAMMED POWER SOURCE A I7 10 /|2 [I a A l I X x FLOW PATENTEI] SEP I 8 I975 FIG. 14
HIGH VOLTAGE HIGH POWER CURRENT l l SPLTT RING E I ND OR} 5 3 i 6 ULTRA ULTRA HIGH HIGH VOLTAGE CURRENT LOOP A LOOP B a} 4B LOOP c m 40 LOOP 0 4D LOOP B CW LOOP A LOOP D LOOP c ccw ccw PATENTEUSEFI 5197s 3.906.415
SHEET 6 F 8 l0 l7 4 H7 4- .4 .4 4; 1- FLOW LMERCURY MAGNETIC FLUID 4 r34 I E EJ MANEFc FLUID 4| RESERYQR: f MECHANICAL MECHANICAL 36 ACTUATOR ACTUATOR 37 L ZZZZL LL F/G 16 3 5s 5p 59 5s 0 6 um i AWN w 60 64 J l l l l iq/l L l l /j] MAGNETIC 55A '5 FLUlD 1o IOA 52 4 l2MERCURY 1 I [\Alhfll f n? n? V M I 1 I MAGNETIC FLUID 1 p MERCURY l T r 528 520 NORMAL MERCURY fi zg e s gA 44 Y MAGNETIC MAGNETIC FLUID 5 A 708 TOC'" PATENTEDSEPIBISYS 3.906.415
saw a 0? a FORCE I MAGNETIC FLUID MAGNETIC 77 FLUID e 75 [IO 4 II -q MAGNETIC FLUID MERCURY] MAGNETIC FLUID MERCURY MERCURY FIG. 19
SOURCE OF ELECTRIC g3; 82 SOURCE OF ELECTRIC F G. 20 POWER 47 6A jlolD x MECHANICAL INSULATING ACTUATOR FLUID MERCURY F I /&'// 92 I0) IOA eB 93 F/G. Z1
APPARATUS WHEREIN A SEGMENTED FLUID STREAM PERFORMS ELECTRICAL SWITCHING FUNCTIONS AND THE LIKE The invention described herein was made in the course of a grant from the Agency for International Development, an agency of the United States Government. The present invention relates to a system wherein a segmented fluid stream having adjacent segments that differ from one another in electromagnetic characteristics, performs electrical switch functions, electrical generator functions, electrical motor functions, and pump functions.
There accompanies herewith a report entitled Parietal Shunt Sequencer by the present inventor. The report goes into great detail on background, mathematical analysis, technical considerations, and the like concerning the subject matter of this invention. In order to keep the present specification in reasonable bounds while, nevertheless, presenting to the world in greatest detail the concepts herein disclosed, said report is hereby incorporated herein by reference. Attention is called also to the following U.S. Pat. Nos. as well as several foreign patents and one technical bulletin: 617,38 1 (Wilson); 691,739 (Batault); 792,571 (Christman); 1,441,250 (Smith); 1,982,717 (Wilhelm); 2,437,225 (Fiedler et a1); 2,465,066 (Corliss); 2,645,279 (Rossman); 2,844,687 (Gottfried et al); 3,453,462 (Yin-Yun Hsu et a1) 3,539,921 (Cohen et a1); 3,555,312 (Bidard); 3,701,868 (Lucien); 3,760,245 (Halvorsen); French Pat. No. 386,817 (1909); Czechoslovakian Pat. No. 74,449 (1944); Dutch Pat. No. 286,621 (1962); Swiss Pat. No. 295,858 (1954); German Pat. No. 495,194 (1930); British Pat. No. 1,096,861 (1967); French Pat. No. 394,858 (1909); Swiss Pat. No. 395,393 1965); and Magnetic Fluids Engineering Kit and Applications Sketches" of Ferrofluid Corporation, Burlington, Massachusetts.
It is noted above that the concepts herein disclosed have use in apparatus to perform generator functions, motor functions, and pump functions, but in what follows the main emphasis is directed to apparatus for switching electric circuits. In this connection, most switching functions could best be performed by an ideal switch that in one condition has zero resistance and in another condition has infinite resistance and changes from one condition to the other in zero time. The ideal switch should also handle large currents when ON and withstand large voltages when OFF. Since the ideal is not attainable, workers in the art have strived toward the ideal situation. Mechanical switches have very favorable resistance characteristics but are relatively slow to operate, have wear problems, are expensive to make, install and operate, and, particularly in high to very high voltages (138 kV and up), have arcing problems. Semiconductor devices, while fastacting, are lossy, have electric current and voltage restraints and, particularly in the case of thyristors, have turn-off problems. Fluid switches such as, for example, mercury tilt switches, have very good resistance characteristics, but are slowacting and, for this and other reasons, have arcing problems. Other liquid switches such as the electrocapillary type in the Lucien patent, are not so much switches as they are timers--the capillary action upon which they depend acts that slowlyv Accordingly, it is an object of the present invention to provide a novel liquid switch that combines the favorable electrical resistance characteristics of mechan- 2 ical switching with the high-speed switching characteristics of the best of the semiconductor devices.
Another object is to provide a switch whose characteristics permit, by sealing and other techniques, its use in high-voltage transmission systems (e.g., 138 kV and up), distribution systems (e.g., 5 k\/ to 25 kV), and industrial power systems (240/480 volts), but which is useful, as well, to switch low-power signals in logic circuits with milliampere and millivolt requirements.
Still another object is to provide a switch that is useful in a system wherein switching is occasional (e.g., circuit breakers for power transmission systems or distribution systems) or quite frequent (e.g., logic circuits).
A further object is to provide a switch that meets the foregoing objects and yet is inexpensive to build, simple in concept, and requires minimum maintenance.
A still further object is to provide apparatus of more general use.
These and still further objects are apparent hereinafter.
The foregoing objects are attained by apparatus wherein fluid segments are contained within a housing which requires any movement of the segments to be such that all move in the same direction at any instant of time. Contiguous segments are composed of materials that have different electromagnetic properties from the immediately adjacent segment or segments. The cross dimensions of the space within which the segments move must be large enough to render insignificant capillary action therein with respect to the materials of the fluid segments. Furthermore, it is necessary that there be a stable transverse interface between contiguous segments. The stable interface can be attained by having cross dimensions of said space that are small enough to insure that surface tension of the segments maintains the integrity of the segments, or mechanical or other keepers can be used. Means is provided to pro pel the segments in said same direction. In the electrical switch configuration, contiguous areas are composed of a liquid electrical conducting material and a fluid insulating material, respectively, there being provided electrodes, the gap between the electrodes being bridged by the fluid segments. The switch conducts or not depending upon Whether the gap is bridged by conducting or insulating fluid, respectively.
The invention is hereinafter explained with reference to the accompanying drawing in which:
FIG. I is a diagrammatic representation partly block diagram in form, of a fluid switch of the present invention comprising an annular tube within which flows in a circular path a plurality of segments of mercury (Hg) separated from one another by segments of a magnetic insulating fluid (marked FF for ferrofluid);
FIG. 2 is a longitudinal section view, greatly enlarged, of a portion of the tube of FIG. I and shows a segment of mercury with a partial segment of magnetic fluid at either end thereof;
FIG. 3 is a schematic representation of a modification of the switch of FIG. I showing only two of said segments and, whereas in FIG. 1 the segments are pro pelled along a closedloop path around the tube by interaction between permanent magnets and the magnetic fluid segments, in FIG. 3 two coils perform the propulsion function;
FIGS. 4A and 4B show schematically mechanical keepers for maintaining the integrity of the segments;
FIG. 5 shows schematically a modification of the switches shown in FIGS. I and 3;
FIG. 6 shows schematically a portion ofa composite fluid switch wherein four tubes, segments and electrodes similar to the same elements in FIG. I, are connected in a series/parallel arrangement to provide AND/OR functions;
FIG. 7 is schematic of an extension of the ANDfUHCk tioning elements of FIG. 6;
FIG. 8 shows schematically a fluid switch in which the segments are propelled in a circular path within a tube by a polyphase magnetic field;
FIG. 9 shows schematically a portion ofa fluid switch in which propulsion is effected by a magnetic field that is single phase or, the equivalent for present purposes, a pulsed d-c field;
FIG. 10 shows schematically a portion of a fluid switch with a single tube but rniltiple splitring electrodes, radially spaced;
FIG. 10A shows the schematic of FIG. I0 symbolically;
FIG. 11 is an isometric schematic of a portion of a fluid switch in which the electrodes are axially-spaced ring electrodes;
FIG. 12 is a schematic representation of a d-c electric system that includes a fluid switch (only partly shown) of the present invention having three electrode-pair combinations that act in concert with the other circuit elements to reduce arcing in an operating switch;
FIG. 12A shows schematically a system similar to that shown in FIG. 12, except for a-c switching;
FIG. I3 is a schematic showing of a fluid switch to show available electrode configurations;
FIGS. l4 and 15 show schematically possible loop patterns available when a plurality of loops are combined to form a composite switch;
FIG. I6 is a longitudinal section view. partly block diagram in form, of a portion of a fluid switch showing a mechanical mechanism for changing the lengths of the segments;
FIGS. I7, 18, 18A, 18B and 1) illustrate schematically mechanisms for changing segment length, FIGS. 18A and 188 being enlarged and more detailed views of a portion of FIG. 18;
FIG. 20 is a schematic representation of a fluid switch which is switched ON and OFF by electromechanical means and which contains an overload switching mechanism thereby to permit its use as a dc current circuit breaker in high-power and high-voltage systems;
FIG. 2| is a longitudinal section view, partly in block diagram form, ofa fluid switch having one mercury segment and one insulating fluid segment, the two being moved axially along the tube by a mechanical actuator that works against a spring force, the switch in FIG. 2 I,
like that of FIG. 20, being particularly useful in connection with high-power switching systems; and
FIG. 22 shows schematically a system capable of functions other than switch functions.
In the detailed description that now follows, the invention is discussed first in connection with the electric switch aspects thereof. Turning now to FIG. I, there is shown at 101 a fluid switch for switching electric current between a source of electric power I and either of the loads labeled 2 and 3. The switch I01 comprises a tube 4 of an electrically non-conductive material such as, for example, glass, plastic or ceramic. Contiguous regions 10, I], I2, l3, l4, 15, 16 and 17 within the tube 4 have respectively liquid electrical conducting material segments (e.g., mercury at the regions 11, l3, l5, and 17) and fluid insulating material segments (e.g., magnetic fluid or ferrofluid comprising an electrical insulator material such as transformer oil, for example, and magnetic particles at the regions l0, 12, 14 and 16). The regions or segments 10, 1],... are disposed axially or longitudinally along the tube 4 and, in most situations of interest, in the form of a closed-loop path. (In the description hereinafter, for simplicity of explanation, the elements 10-17 are always called segments," the conducting segments being mostly the liquid metal mercury and the insulating segments being mostly a liquid insulating magnetic fluid.) In order that the apparatus 101 function in its required way, the transverse interfaces labeled 10A, 1 1A, 12A... between contiguous segments 10 and 11, ll and l2, l2 and 13, etc., must be stable; in order that such stability exits, the liquid electrical conducting material and the fluid insulating material must be immiscible with respect to one another, but further conditions must exist to prevent mixing of the two liquids. Later, there is a discussion of mechanical and magnetic keepers, but, for now, the means to prevent mixing is the combination of the surface tensions of the mercury and the magnetic fluid and the inner cross dimensions of the tube 4 (which is assumed, for now, to have a circular cross section), as explained later with respect to FIG. 2, but there follows now a brief explanation of the further elements in FIG.
The switch I01 includes a plurality of electrodes 6A, 68, 7A, 78 that act in pair combinations. Thus, the electrodes 6A and 6B act together and 7A and 78 act together. The electrodes 6A and 6B of the pair combination 6A-6B are disposed transversely across the tube from one another, as are, also, the electrodes 7A and 7B of the pair combination 7A-7B, but that configuration is not always used, as later explained. The electrodes 6B and 7B are connected to ground G; ground in this specification denotes a common or return conductor but can be actual earthing. The electrodes 6A, as shown in FIG. 2, are in contact with the mercury segments II, and with the magnetic insulating fluid segments 10, ;electric current will flow between the electrodes 6A and 6B (and 7A and 73) when the gap therebetween is bridged by a mercury segment and current will be interrupted when the gap is filled with magnetic insulating fluid. It should be noted that the axial length of the magnetic insulating fluid segments must be greater than the axial length of the electrodes in the electrical arrangement of FIG. I. It should also be noted that the annular inner cavity of the tube 4 is completely filled by the liquid segments, which is important in that the circuit, when interrupted, is done so by filling the gap with the liquid fluid insulator as interrup tion occurs; and such complete filling transmits axial forces between the incompressible liquids.
The axial forces to effect axial flow of the segments 10, in a stream around the closed-loop path within the tube 4, are provided by magnets 8A, 8B, 8C and BI) whose magnetic fields couple to the segments I0, 12, I4, and 16, respectively; and thereby move the segments axially with respect to the electrodes in the direction of the arrow numbered 5. At any instant all segments 10 to 17 move in the same directional sense of flow, i.e., either clockwise or counterclockwise within the tube 4. An electric motor 9 drives the magnets.
The segments shown in FIG. 2 are accurate portrayals of a portion of an actual switch built and tested, the switch being somewhat similar to that shown schematically in FIG. 1. The cross dimensions designated (1 of the tube 4 must be small enough to permit surface tension of the mercury and surface tension of the magnetic fluid to retain both as segments within the tube. In the actual switch, d is about 3 to 4 mm. It will be appreciated, therefore, that there is an upper limit on the d dimension; and, as noted elsewhere, there are constraints on how small the cross dimensions of the tube 4 can be.
In the fluid switch labeled 101A in FIG. 3, one segment, the segment 10, of magnetic fluid is shown and one segment, the segment 1 l, of mercury, the interface A therebetween being the only interface labeled. In FIG. 3, and in later figures, elements that perform similar functions to that performed in the system of FIG. 1 generally are given the same or similar designations. Axial movement of the segments is effected by coils l9 and when the coils are energized by a dc or other power source 18 (see Baker US. Pat. No. 3,748,492), as now explained. At the time depicted in FIG. 3, cur rent can flow between the electrodes 6A and 6B. The condition thereshown exists when the coil 20 is cner gized and the coil 19 is not. In that situation the seg ment 10 of magnetic fluid will seek to position itself within the solenoidal coil 20. The segment 10 will tend to center axially within the solenoidal coil 20. If the coil 20 is deenergized and the coil 19 is energized, the segment 10 will travel axially in the direction of the arrow 5 to rest within the central opening of the solenoid 19, at which time electric current between the two electrodes 6A and 6B is interrupted. The interruption time can be quite fast as is discussed in said report, since. among other things, judicious placement of the various switch elements can result in a situation in which the interface between segments moves across the electrodes fast enough to prevent initiation of an arc that normally occurs when a circuit is interrupted (see FIG. 21 in said report). Thus, while the interface 10A in FIG. 2 can move either clockwise or counterclockwise, depending on relative positions of the coils l9 and 20 and the length of the segment 10, it is possible to create a situation wherein the interface 10A will always move counterclockwise, thereby permitting it to build up substantial angular velocity before it crosses the electrodes 6A and (:8. ln closing the circuit, the reverse situation will prevail, that is, fast segment velocity causes the switch to close quickly.
It is above noted that the inside cross dimensions of the tube 4, in the embodiments so far described, must be small enough to permit surface tension of the two liquids to retain both as discrete segments that fill the tube transversely; and the whole interior volume of the tube is filled by the combined segments. that is, no air gap exists therein. The latter situation is vital in the case of very high voltage switching operations, but, conceptually, the mercury segments can be separated axially or longitudinally from one another by a gas (cg, air) for low voltage switching and mechanical axial propulsion used. It should he noted that the magnetic fluid is wetting as to the tube material (glass here), and this is important to provide lubrication for the mechanical keepers later discussed, but the liquid carrier of the magnetic fluid should be non-wetting as to the electrode material (which is silver or copper in the apparatus built so far). The mercury (or other conductive liquid) should be wetting ofthe electrode material, particularly for low-voltage systems, to supply a low resistance electrical connection thercbetween. A further consideration is of great importance: the switch of the present invention in almost all uses for which it is intended, must be capable of sharp, fast, make-break operations. Thus it is necessary that axial movement of the liquid segments be fast and be effected only when the switch is actuated; it is necessary, therefore, that the interior cross dimensions of the tube 4 be large enough to render insignificant axial forces due to capillary action therein between the tube and either the mercury or the magnetic fluid. Such capillary action, if it were present in sufficient amount, would retard axial flow of the segments, thereby increasing propulsion problems, and would or could cause creep of the segments, thereby effecting a switching operation even in the absence of an external propulsion force upon the segments.
Mention has been made of keepers to prevent mixing of the two liquids when, for example, the cross dimensions of the tube 4 exceed that which permits surface tension alone to maintain the integrity ofthe segments. The permanent magnets 8A, 8B, act as magnetic keepers since at all times a magnetic field couples to the segments l0, l2, l4, and 16, thereby giving those fluid segments a high apparent surface tension and the semi-solid structure that magnetic fluids assume in the presence of a magnetic field. A small bias current through the coils l9 and 20 will also act a magnetic keeper; in this situation the switching current in the coils would be of much greater magnitude than the bias current. Mechanical keepers are of greater importance for present purposes because they allow the use of large tubes with low drag and, thus, large liquid velocities, etc; a number of such keeper configurations is shown in FIGSv 4A and 4B.
The mechanical keepers can be associated directly with the magnetic fluid, as are the keepers labeled 20A, 20B, 20C and 20D in FIG. 4A, or with the mercury, as are the keepers labeled 21A, 21B, 21C and 2lD in FIG. 4B. The bobbindype keepers 21A, permit increasing possibilities. For example, the keeper 218 shows, schematically, a hollow keeper with air in the interior and 21C a keeper with magnetic fluid in the interior. In this way, the weight of mercury can be somewhat bal anced with the weight of the a less massive fluid, which tends to balance the system and reduce inertia. The magnetic fluid can be contained within the keeper, as is the case of the keepers 20A and 208. Also, the latter two keepers can have appropriate retentive magnetic capability. In all the situations illustrated in FIGS, 4A and 4B the interface, c.g., the interface numbered 22, between the mercury and the magnetic fluid is sharp--a requirement here. The cross dimension or dimensions shown at e of the keepers must be sufficiently smaller than the tube inner cross dimension or dimensions to maintain friction to acceptable limits, but it must be sufficient to permit surface tension to maintain the integrity of the segments so that the cross space of the tube is filled with either a segment of mercury or a segment of magnetic fluid.
FlG. 2 shows the effect of gravity, the magnetic field exerting force on the ferrofluid, capillary forces and frictional drag, which all combine to set the shape of the boundary between the ferrofluid and the mercury.
The interface between the glass tube 4 and the mercury exhibits a characteristic non-wetting shape at points I; and 12'', while the oil-base magnetic fluid wets 7 the glass at b and b'. The static non-moving shape of the mercury slug is basically a result of the balance between the force due to its high surface tension and the force due to gravity. However, the slopes at the liquid interfaces bca and b"c"'a" are both more nearly vertical than one would expect because the magnetic field increases the apparent surface tension of the magnet fluid and causes it to approach the shape of a near-perfect cylinder. If the gravitational field were equal to O, the interface would be perpendicular to the tube axis.
When the fluids are in motion there are frictional losses (mostly in the mercury) caused by viscous et fects and turbulence (note that in a practical situation mercury has only turbulent flow). The force exerted on the slug of mercury by the magnetic fluid, together with the force due to surface tension and gravity, causes the magnetic fluid to displace (burrow under) the mercury at an. Using a magnetic fluid with a higher permeability in a stronger magnetic field will give it a higher apparent surface tension, causing it to become rigid (semirigid) and, therefore, it can push against the mercury with less distortion per unit of force. (The segments I0, I2, are still considered fluid for purposes of this specification even in this rigid or semirigid state particularly since ferrofluids will in fact flow even in this rigid or semirigid state.) This burrowing effect, along with the maximum permeability that can be obtained, is related through a complex set of parameters. The viscosity of the base fluid, fluid temperature, magnetic doping material, and the size of the tube 4, etc. used, all combine to set the maximum liquid velocity that can be obtained. Another parameter that limits the maximum fluid velocity is the physical size of the annulus. Because the fluids are flowing in a circular path there is distortion of the liquid inte rface; the fluid at the smaller radius must flow at a greater velocity. Finally, the frictional drag of liquids flowing in a pipe is in general a function of the roughness of the internal surface of the pipe and is proportional to the square of the velocity and inversely proportional to the diameter (ID) of the pipe. Obtaining a high liquid velocity is important in many of the intended uses, and a more detailed discussion of the parameters affecting this velocity is given in said report. Based on initial tests, however, it appears that velocity up to 10 meters per second is practical with an annulus of It) cm diameter made with 4 mm (ID) tubing.
A simple electromagnetic means for propelling the segments 10, in a closed-loop, circular path is illustrated in the fluid switch shown at 1018 in FIG. 5. The switch 1018 has, in addition to elements previously dis cussed, electrodes 23A and 23B which serve as switch ing electrodes for the coils l9 and 20. It is assumed at the instant of time depicted in FIG. that the circuit is energized by closing an ON-OFF switch in the source 18. The source 18 is assumed to be, d-c although it need not be. Current will flow from the source, through the coil 19, the switch or commutator 23A23B, and the coil 20 to ground G. (In this circumstance the segment 14 will be magnetized but will have no tendency to move.) The interface A will tend to move clockwise with the geometry shown; the circuit between contacts 23A23B will open; and momentum will carry the segment I] to that now occupied by the segment 13. The segment-stream propulsion circuit will be then energized. In this way the stream of segments can be propelled in a circular path; as alternate conducting and non-conducting segments move between the elec 8 trodes 6A-6B a train of electric pulses 24A is generated, consisting of voltage pulses from a d-c source 1A.
The above discussion was of course simplified to demonstrate the idea. Like the commutator arrangement for a d-c motor, the angular displacement of conducting-non-conducting segments is such that the proper coils are energized in the proper time sequence for a net rotational torque to be applied to the liquid. This action is the same as the combination of the brush and commutator segments in a d-c motor, which allow the proper armature coils to be energized in the proper time sequence such that a net rotational torque is applied to the armature.
In FIG. 6, it is assumed that the tubes shown at 4A, 4B and 4C contain respectively, segments 10E, 11E 10F, 11F 10G, 11G that are like the segment l0, l1 respectively, that flow toward the right, as shown, to form a composite switch lOlC. With switches S and S open, current (which can be a-c or d-c) will flow from the source I, through the electrode 6A to the electrode 68, thence to an electrode 6E and a further electrode 6F to an output 25. The signal at the output 25 can be similar to the train of pulses 24A, but there is a difference. The electrodes 6A to 6F are in series; thus the width of the pulses can be changed by appropriate phasing of the segmented streams in the tubes 4 and 4B. The system acts as an AND gate. An OR logic function can be provided by closing the switches S, and S and the pulse width can again be modified in a desired function. The OR function produces other possibilities which are touched upon in the next paragraph.
In FIG. 3 the switch 101A acts by an interchange of position of the segments 10 and II, but essentially the segments, once positioned, remain so for periods of time. The switch 101C in FIG. 6 can provide the same function as the switch IOIA of FIG. 3 except that the former has a continuously moving stream of segments. Thus, for example, if the source I is d-c, current can flow through the electrodes 6A to (SF and out or through the series of electrodes shown at 6C, 60, 6G and 6H and out; current can be interrupted by moving the relative positions of the segments so that one pair of the series electrodes is always insulated from another (eg, 6A is insulated from 68; 6E is insulated from 6F, etc. It will be appreciated that the open switch function will require here very precise segment lengths. This can be overcome by having a further tube 4D in FIG. 7, like the tubes 4 and 4B, with electrodes 6I and 6,]. If each stream of segments is driven, say, by a variable frequency polyphase source such as that shown at 18' in FIG. 8 and programmed appropriately by a special purpose computer 27, proper phasing can be accomplished. A simpler programing scheme, however, is shown in FIG. 9 which shows the tube 4 only. (The closed-switch condition presents a similar-type problem; both conducting and non-conducting can be solved. however, by having three of the electrode arrangements of FIG. 7 in parallel to form a three-part, series-parallel, AND/OR switch.)
In FIG. 9, the axial length of the segments is slightly longer than the length of the solenoidal coils l9 and 20. If the coil 19 at the instant depicted is energized by a programmed power source l8 (see the Baker US. Pat. No. 3,748,492 for a suitable low-power source for such purpose) the interface 10A will move to the right; now the coil 19 is de-energized and the coil 20 is energized. Proper timing of electric current in the coils I9 9 l and 20 will establish the segment flow. Similar propulfilled with liquid segments, and by having insulating sion systems associated with the tubes 4A etc.,establish segment or segments that consist of high dielectric the timing of the segments past the associated elecstrength oil (e.g., transfonner oil). Also, arcing is retrodes. (The three-part series arrangement of FIG. 7 duced by the auxiliary electrical components in FIG. would, of course, also be used in combination with the 5 12, which interconnect and protect each pair combinatubes 4A and 4C to overcome the problem of the tion when current is switched, in the manner now dislength of the insulating segments, above noted, and the cussed. three-part parallel arrangement would be used to over- In the system of FIG. 12, each segment of insulating come the same problem as to the conductive segments, fluid such as the segment 10, must have an axial length as also noted.) greater than the axial length of the electrodes 6A,
Some further arrangements are shown in FIG. 10 in addition, each mercury segment, such as the segwhich shows a single-tube ring relay tree; the mercury ment ll, normally has an axial length sufficient to span segments only are shown, the magnetic fluid insulating the electrode pairs, that is, the segment 11 must be long segments being implied. enough to short electrically the electrode pair 6A'-6B In the embodiments discussed so far, the electrodes 15 to the pair 6E6F or 6A'6B' to 6C-6D'. Both merare disposed across the tube 4 from one another. It will cury segments and magnetic fluid insulating segments be appreciated that in the system of FIG. 10, the elecmay be long enough to span the axial distance between trodes 6A and 78 can be paired, for example, but that electrode pairs (SE-6F and 6C'6D. Electrode posiin such case the mercury segments would have an axial tioning is such that the first electrode pair 6A'-6B' is length sufficient to span the axial gap between the segdisposed, in axial location, between the second elecments 6A and 7B; or some other interconnection trode pair 6C'6D and the third electrode pair scheme can be employed. Also, the electrodes so far 6E'6F', axial movement of the conductive and the inare curved plates, but they can be ring electrodes as sulating segments having the effect, upon switching the shown schematically in FIG. ll wherein the electrodes electric current, to time sequence the opening and clos- 6A and 6B are axially displaced along the tube 4. ing events of said electrode pairs. The opening and Whereas with the transverse electrode arrangement the closing events sequencing is shown in Table 1 below insulating segments must be greater in axial length than wherein the electrodes 6A6B', 6C6D' and 6E6F' the electrode axial length (unless the AND arrangeare designated switch 1, switch 2, and switch 3, respecment of FIG. 7 is used), with an axial electrode distively, and C and 0 mean closed and opened, respecplacement, as later discussed, the mercury segments tively.
' TABLE I time ---l Sequence Switch No. I 2 3 4 5 h 7 x 9 3 C-0 O-C (-0 1 00 O-( t 0 Eu 2 (-0 0-6 C 0 must be sufficiently long to span the axial gap between The auxiliary components previously mentioned electrodes, but the axial length of the magnetic insulatcomprise a capacitance 30, a transformer 31 and a ing fluid segments need not. It might further be pointed diode 32. The primary, labeled 3lA, of the transformer out at thisjuncture that by having different-length segis connected in series with the electrodes of the third ments in either electrode configuration, the pulse width electrode pair 6E6F and the combination thereof is of the pulses 24A in FIG. 5 can vary from one another-- connected in parallel with the capacitance 30, as -and, as mentioned, that width can be varied. By way of shown. The anode of the diode 32 is connected to the example, it should be appreciated that by varying the positive side of the first electrode pair 6A'6B', that is, lengths of the mercury segments then, makebeforeto the electrode 6A, and the cathode side of the diode break (long segments) and break-before-make (short 32 is connected to one terminal of the capacitance and segments) switching action can be obtained. This fur- 5U thence being connected through the second electrode ther demonstrates the flexibility of this invention. pair 6C'6I) to the negative side of the first electrode In the system of FIG. 12, a source of d-c power lD pair as well as to the negative side of the source ID. energizes the load 2 through a liquid switch that com The transformer secondary labeled 31B, is interconprises a plurality of electrode pairs 6A and 6B, 6C nected through a further diode 33 back to the source and 6D, and 6E and 6F, which acts as pair combina' ID of d-c current. thereby to allow energy stored in the tions to perform switching functions. The load current capacitance 30 to be returned to the source as a normal flow is normally between one electrode of the pair part of the switching operation. In this way, the energy (e.g., the electrode 6A) and the other electrode of the stored in the arc-protecting capacitance 30 is not pair (e.g., the electrode 6B) and occurs when a merwasted (dissipated) but instead returns to the d-c cury segment is disposed between the pair combinasource 1D. tion. The switching arrangement of FIG. 12 is for a d-c The switching system of FIG. 12 is intended for arcpower system. A modification can be made for as less switching ofd-c power systems. Thus, the first elecpower systems, as shown in FIG. 12A wherein an a-c trode pair 6A'-6B, the second electrode pair 6C-6D', power source 110 feeds inductive loads 1 II and I I2. A
and the third electrode pair 6E'6F are subject to arcload current I passes between electrodes 6A" and 6B" ing which is minimized by having the fluid segments when the gap therebetween is bridged by mercury and move axially past the electrodes at a high angular vethe load current I is interrupted when the gap is filled locity, by having the whole inner volume of the tube with an insulating liquid segment. To reduce or minimize arcing, further circuit elements, capacitances C,, C resistances R R and electrode pairs 6C')\ 6D" and 6E"6F" are used. The capacitance C, C When the electrodes 6A" and 6B" are shorted by mercury, the other circuit elements C etc., are effectively out of the system since there is almost no voltage drop between the electrodes. Now, however, if an insulating liquid segment entering from the left-hand side fills the gap between the electrodes 6A" and 68', a current I,., flows into the capacitance C and through the electrodes 6C and 6D" (it is assumed that the load waveform, also designated I, in FIG. 12A, is Since electrodes 6C" and 6D" are still shorted little current will flow to the capacitance C Now the insulating segment is made to Isolate ooth 6A" and 6C" from 6B" and 6D respectively. A current I will flow into the capacitance C; (since C, C I will be much smaller than was I and so forth, for further electrodccapacitance combinations. It should be seen that l, I,., Since the fluid segments stream is moving rapidly, whether impelled by the electromagnetic drive of FIG. 3 or the later-discussed mechanical actuator of FIG. 21, for example, the time frame of events is small. The resistances R R act to dissipate energy stored in the capacitances C C in preparation for circuit reclosure and then subsequent circuit interruption. If the load current I, is negative at the time of interruption, the same sequence of events will occur except that the direction of current flow is reversed and C C will charge to the opposite polarities.
Before going into a further detailed discussion of further modifications of the invention, a brief review of the various switching possibilities is made with reference to FIGS. l3, l4, and I which, in view of the previous explanation, are largely self-explanatory. In FIG. l3 the single closed-loop tube 4 has associated with it different contact configurations and interconnecting patterns to supply the purposes or functions indicated. The arrows indicate flow of the segment stream, which can be clockwise (CW) or counterclockwise (C'CW). Two loop patterns are shown in FIGS. 14 and 15 wherein the tubes are designated 4A to 4D.
Mention is made above of changing the effective lengths of the segments by employing AND/OR configurations; it will be appreciated with reference to FIGS. I4 and IS, that changes in the angular velocity of the fluid in the tubes 4A, etc., relative to one another can change the effective lengths of the segments as re flected in the pulse width of the output. (Together FIGS. l3, l4 and I5 demonstrate the flexibility of the possible switching combinations and configurations.) Actual changes in segment lengths can be made as now explained with reference to FIGS. l6, l7, 18. 18A, 18B and I9.
In FIG. 16, valved inlets 34 and 35 to the tube 4 permit simultaneous introduction and removal of either mercury or magnetic fluid; that is, when one fluid segment is reduced in size, another must be increased since the tube 4 is filled with incompressible fluids. Arrows 36 and 37 indicate that the flow is from a reservoir 38 of mercury to the tube 4 and from the tube 4 to a magnetic fluid reservoir 39. Mechanical actuators 40 and 41 move the valves; the dotted line 42 between the reservoirs 38 and 39 indicates interaction so that a pressure head is created in the reservoir 38 at the same time as a reduced pressure is created in the reservoir 39.
Another scheme for changing segment length is shown in FIG. I7 wherein it is assumed for now that the normal (i.e., in normal switching operation) segment flow is to the right. In FIG. 17, as is later noted in some detail, coils 62A and 63A perform liquid gating functions, coils 62 and 63 perform liquid motoring functions, coils 56, 57, etc., associated with upper reservoir tube 64 perform liquid gating functions, in connection with the reservoir tube 64 and coils 47, 48, etc., per form a liquid gating function in connection with lower reservoir tube 44. The liquid gating function is discussed later in connection with FIGS. 18, 18A and 18B, but, essentially, the coils 62A and 63A act to stop the flow of segments in the main tube 4. In the circumstance depicted in FIG. 17 the mercurymagnetic fluid interface 15A, in a condition of normal flow, is leaving the inner space of the solenoidal coil 62A. If new the coil 63A is energized, the mercury-magnetic fluid interface will be stopped at the left-hand end of the coil 63A. It will be appreciated that if the coil 62A, rather than the coil 63A had been energized, that the interface lSA would have been stopped by the coil 62A. (Also, the coils 62A and 63A can apply motor forces to the segments by proper sequencing.) In order to make the effect of the coils 62A and 63A sharp, low reluctance paths around the solenoidal coils is provided by m l-metal or other magnetic circuits 62B and 63B, respectively. In this connection, all the coils in FIG. 17 would normally be provided with such magnetic cir cuits but such is not shown in FIG. 17 to reduce cluttering of an already cluttered figure. Further coils, like 62A and 63A, can be placed around the tube 4 to stop flow in a situation when normal flow is to the left. Here and in other situations discussed in this specification, the further possibilities of the concepts shown are many, but, generally, exemplary systems only are shown.
Changing the lengths of segments 10 and II is accomplished by withdrawing one fluid from one of the reservoir tubes 44 or 64 and by simultaneously adding a like amount of the other fluid (in some circumstances it can be the same fluid from some axially displaced segment) to one of the reservoir tubes. The particular reservoir tube employed will depend on the particular segment positioning at the time of fluid transfer. Assuming now the situation of FIG. [7, if the coil 62 is deenergized and the coil 63 is energized a force will be exerted on the fluid in the tube 4 tending to move the magnetic fluid-mercury interface 10A into the interior of the coil 63, that is, to the right. It is intended here that the segment It) in FIG. 17 be increased in length and the segment 11 be decreased. To do this magnetic fluid 61 in the tube 64 is allowed to flow into the tube 4 and mercury from the segment 11 is allowed to flow into reservoir mercury in the tube 64. Interchange of flow between the main tube and the reservoir tube is affected by a number of elements. Thus. small mesh (-one to two mils) screens 52, 52A, 52B and 52C pre vent flow of mercury into magnetic fluid because of surface tension of the mercury; similarly when coils 49 and 54 are energized the increase of apparent surface tension in the magnetic fluid prevents flow of magnetic fluid through screens 52C and 52A, respectively, to a mercury segment. In either situation, however, mercury will pass through the screen at a mercury-mercury interface and magnetic fluid will pass through a screen at a magnetic fluid-magnetic fluid interface.
If now in FIG. 17 the coil 49 is energized and the coil 56 is energized (coils 54, 55, 57, S8, 59 associated with the upper tube 64 are de-energized and coils 47, 48, S and 51 associated with the lower tube are de-ener gized), the interface 10A will move to the right (and 55A will move to the left, as later noted The segments in the closedloop main tube cannot circulate because they are stopped (i.e.. maintained in a stable or fixed axial position) by the energized coil 63A; mercury of the segment II cannot move into the tube 44 because it is stopped by the screen 52C, and it is further stopped by the energized coil 49 which acts to gate flow in the other direction at the time depicted; so the combination prevents flow from the magnetic fluid reservoir shown at 46 or the mercury reservoir showing at 45, to the main tube 4. In this circumstance the magneticfluid-mercury interface labeled 55A will move to the left to about the dotted position of 55A shown. The summary of events are these: the coil 63A is energized, the coil 63 is energized, and the coil 56 is energized; mercury flows from the segment I], through the screen 52 to the reservoir 60; and magnetic fluid flows from the reservoir 6I, through the screen 52A to the segment I0. The segment 10 in FIG. I7 is thereby made larger and the segment I] shorter. If at some later time the segment positions in the main tube 4 are changed so that the segment 10 and II are replaced by the segments l and 10, respectively, mercury can be taken from the reservoir 45 and added to the mercury segment 15 while the segment is discharging to the res ervoir 46. In normal operation the coils 49 and 54 are energized, but there may be situations in which they are not.
It is in order at this juncture to discuss the gating function mentioned above, again with reference to FIG. 17. It will be recalled that with the normal segment flow to the right, the interface A will be stopped at the left-end of the coil 63A. If, now, the nor mal flow of segments in the tube 4 is reversed and the coil 62 energized, the interface 10A will stop at the right side of the solenoid coil 62. Coils like the coils 62A and 63A can be placed, like 62A and 63A, outside the right-hand side of the reservoir area. Indeed, a sin gle coil at an appropriate location will suffice. It should be seen at thisjuncture that a segment of magnetic fluid tries to fill the interior of an energized solenoid coil: if an attempt is made to reduce the volume within the energized solenoid, that is occupied by the magnetic fluid, the electromagnetic forces engendered resist the change. (Similarly, in motor action the electromagnetic forces cause the required position change in appropriate circumstances) In the system of FIG. I7 there are what might be termed mechanical gates 52, 52A electro-mechanical gates 52, 52C and coils 54 and 49 in combination, and electromagnetic gates (eg, the coils 62 A further ramification of the concept just described is illustrated in FIG. 18 wherein, again, segment flow in the tube 4 is toward the right. The segments are unequal in length, and it is assumed continue beyond the segment 17 to some indefinite number N, a segment of mercury being always separated by a segment of mag netic fluid and vice versa. If at the instant depicted in FIG. I8, the coil shown at 67 is energized, then the fluid segment I2 (which is magnetic fluid) will tend to stay where it is and mercury from the segment 11 will tend to flow through an orifice 68 into the mercury section 65 of a tube 64'. The tube 64' also contains magnetic fluid 66 which tends to flow into the magnetic fluid segment N. The amount of mercury flow is determined by energizing appropriate solenoidal coils 70A, 70B 70N, as before. The amount of mercury that flows out of the segment 11 in the tube 4 is equalled by magnetic fluid that flows into the magnetic fluid segment N through an orifice 69. The orifice 68 acts like a valve for mercury in that when a I to 2 mm orifice is used surface tension of mercury prevents mercury flow therethrough when magnetic fluid is on one side of the orifice and mercury on the other; when mercury appears on both sides, there is free flow. The orifice 69 will prevent flow of mercury from the main tube 4 into the reservoir 66 of magnetic fluid. A coil 71, when energized, also acts in the same manner, that is, it acts to prevent the flow of mercury into the reservoir 66; this is because the magnetic fluid resists leaving the volume inside the coil 71. To prevent opposite flow, that is, flow of magnetic fluid through the orifice 69 into the mercury in the main tube 4, it is necessary to prevent flow into the mercury reservoir through the orifice 68. This is effected by stopping movement of the inter face labeled 65A to the right in FIG. 18 by energizing an appropriate coil of the coils 70A 70N.
Mention is made above of the lower reluctance magnetic circuits 62B and 63B. Similar low reluctance circuits are shown in FIGS. 18, 18A and 188 where they are labeled 70A, 70B and each is associated with a solenoidal coil 70A, 70B respectively, that receives the non-magnetic tube 64' If the coil 67 is energized, as before, and any or all the coils 70A, 70B are energized, the mercury-magnetic fluid interface marked 65A will stop at the left end of the coil 70A. More particularly, since, as shown best in FIG. 188, the magnetic circuits 70A, 70B pass through the wall of the tube 64' to be in physical contact with the magnetic fluid therein, the interface 65A will stop in the axial region designated 70A" in FIG. [8A. The amount of change in the lengths of the segments I I and N is determined here by the position finally taken by that interface 65A and that position can be progressively moved to the right in FIG. I8 by, for example, starting with all the coils 70A 70N energized and successively de-energizing the coils, beginning with the coil 70A.
It would unduly prolong this specification to detail the possibilities implied in the situations depicted in FIGS. 17 and I8, but a few comments in this respect are in order. If the segments [0, ll, in FIG. 18, for example, serve to program events in a communication system, that program can be changed by modifying the lengths of the segments; such modification can be accomplished by computer control of the various program coils such as the coils 67, 70A and 71 in FIG. 18. In a system there can be many tubes like the tube 64 with associated elements. Such tubes can contain multiple mercury and magnetic fluid segments to give the necessary proper interface for liquid gating. And the functions of the coils 62 and 63 and 62A and 63A can be introduced to the system of FIG. 18.
The system of FIG. 1) permits simultaneous changes in lengths of two mercury segments or two magnetic fluid segments. If at the time depicted in FIG. 19, coils 75' and 76' are energized and a force is applied as shown to a bellows 77 filled with magnetic fluid, there will be a flow of fluid into the segment 10 and from the segment 12 into a bellows 78. At some other time when the positions of the segments of magnetic fluid and mercury are interchanged by a flow of the liquid stream to the right, for example, then interchange of mercury can be effected by use of bellows 79 and 80. In the latter situations coils 75 and 76 are energized as a gate to prevent magnetic fluid from entering the mercury segments. To provide good magnetic coupling, the coils 75, 76, 7S and 76' would be wound around the orifice between the bellows and the tube 4, not the bellows, as shown: and low reluctance shields as shown in FIG. 18 can when necessary be employed to increase the magnetic coupling to the magnetic fluid.
Bellows can also be employed to propel the fluid axially past the electrodes, as is shown in FIG. 20. To ener' gize a load L, the coil 81 of a solenoid is energized by an electric source 82, force being transmitted upward through a spring 83 to bellows 84 filled with mercury. The bellows 84 compresses causing the interface IOA to move to the left, thereby permitting current to flow from the source I through the load L and electrodes 6A and 68' to ground G. Insulating fluid in the figure moves into a bellows 85 that has appropriate mechanical biasing. In the event ofa fault on the system of FIG. 20, a high fault current in the conductor numbered 86 is picked up by a coil 87 which energizes a coil 88 that forces upward a plunger 89. The plunger 89 overcomes the mechanical bias on the bellows 85, thereby forcing insulating fluid (e.g., transformer oil or the like) into the tube 4 and the interface 10A to the right. Mercury moves into the bellows 84 overcoming the mechanical bias therein provided by the spring 83. In this way the circuit to the load will open and, if the source 82 is deenergized, will remain open. The electrodes 6A, 6B, 6C and 6D function somewhat similarly to their operation in FIG. 12 in that the two extra electrodes serve to reduce arcing. Thus, for example when the load circuit is opened by first isolating the electrodes 6A'-6B' in FIG. 20, the inductive load L feeds into a capacitance C during interruption of current flow between the electrodes 6A and 68'. A resistance R serves to dissipate energy stored by the capacitance C and reverse flow is prevented by a diode 90. It will be appreci ated that the arrangement of FIG. 20 is best adapted for d-c, but the liquid switch thereshown is useful for we as well (FIG. 12A).
The fluid switch shown at 10") in FIG. ZI is somewhat similar to the switch in FlG. 20 in that a mechanical actuator, the actuator marked 91, moves the segments l and ll (there can be more) toward the right, axially past the electrodes 6A and 68 (there can be more). The actuator 91, which can be air, hydraulic, etc., moves a piston 92 to the right; a piston 93 under biasing influence of a spring 94 applies a force toward the left in FIG. 21, upon the segments. It will be appreciated that an electromagnetic force to the left upon the fluid segments I0, ll, etc, in FIG. 21 will exert a force to the left upon the piston 92 in which case the system of FIG. 2] acts as an actuator in which an electromagnetic force is converted to a mechanical force through the piston 92; the function in such system can be characterized as a motor or pump function with the piston 92 serving as a link to some outside load. It can be seen that the tube 4 in the apparatus 101D can be millimeters across; there can be many, many such tubes; the segments 10, 11 etc., in each tube can be ac tuated by coils in the manner discussed previously in connection with FIGS. and 8, for example, thereby to provide a plurality of fluid actuators. all individually controllable to very accurate force levels.
In FIG. 22 the internal volume in the tube 4 again is filled with magnetic segments 10, 12, N. etc. (marked FF for ferro-fluid) separated by mercury segments 11, 12, N-l, etc. A mechanical or other force moves the segments to the right; the tube in FIG. 22 can form a closed-loop path. The mercury segments with low permeability and the magnetic fluid segment with high permeability move past magnetic circuits 95 and 96 external to the tube 4, except for the pole pieces marked N and S in each case, the pole pieces (or, more accurately, the pole faces) being preferably in physical contact with the high permeability magnetic segments l0, l2, (and, thus, of course, the low permeability segments l1, l3, A d-c bias current in a coil 97 will result in an a-c voltage out ofa coil 98; similarly, an a-c bias on a coil 99 will result in a modulated a-c current out of a coil 100. Further, if a current I is passed through the mercury segment shown at N-l which is subjected to a magnetic field intensity B, there is a force on the segment Nl,f I X T3. In this way a propulsion force can be applied to the segment N-l and thence to the other segments. The current is introduced through an electrode not shown, and removed through the electrode shown at 102. If this current I from the electrode 102 is connected to flow through a solenoid 104, then the magnetic segment N will be pulled under the solenoid 104. In this way both the mercury segment Nl and the magnetic fluid segment N are pumped simultaneously. The simultaneous pumping of both kinds of segments can be termed dual-phase pumping". This dual-phase multisegment pumping can be extended by adding more stages.
A few additional remarks of a general nature are contained in this paragraph. The previous discussion, for simplicity of explanation, is directed to a multi-segmented-fluid apparatus wherein a stream of mercury and magnetic fluid segments flow within a glass tube, circular in cross dimensions. It should be noted: that the conductive liquids other than mercury can, and often preferably should (depending upon the use), he used (e.g., Cerrlow-ll7 alloy a tin eutectic); that the magnetic fluids need not always be ferrofluids, i.e., the particles within the fluid carrier can be larger than those found in ferrofluids, for example; that elliptical or other cross dimensions can be employed for the tubes; that the tube material need not be insulating in connection with some functions (e.g., pumping); that the course traveled by the fluid stream in apparatus operation can vary from the various shapes shown, etc. The term tube is used throughout herein to denote any member having an elongate chamber, channel or duct within which the fluid segments can move in the same directional sense. These and still further modifications will occur to persons skilled in the art and all such modifications are deemed to be within the spirit and scope of the invention as defined by the appended claims.
What is claimed is:
l. A fluid switch for switching electric current that comprises, in combination:
a tube of an electrically non-conductive material;
contiguous regions within the tube having respectively a liquid electrical conducting material and a fluid insulating material, said regions being disposed axially along the tube, there being a stable transverse interface between the two materials at said regions;
a plurality of electrodes in contact with the liquid electrical conducting material and the fluid insulat- 17 ing material; and
means for propelling the liquid electrically conducting material and the fluid insulating material to cause the liquid electrical conducting material and the fluid insulating material to move axially relative to the electrodes. thereby to permit electric currents to flow between the electrodes and to be interrupted depending upon whether the liquid electrical conducting material or the fluid insulating material, respectively, is positioned between the electrodes, the axial movement of the liquid electrical conducting material and the fluid insulating material being always in the same directional sense of flow; the interior cross dimensions of the tube being large enough to render insignificant axial forces due to capillary action therein between the tube and either the liquid electrical conducting material or the fluid insulating material.
2. A fluid switch as claimed in claim 1 in which the liquid electrical conducting material is mercury, in which the fluid insulating material is a magnetic fluid, and in which the means for propelling comprises means for creating magnetic fields at locations along the tube displaced axially from one another, said fields being magnetically coupled with the magnetic fluid, electromagnetic interaction between the magnetic fluid and the magnetic field serving to effect said axial movement and serving as well to stop said axial movement at predetermined axial regions, thereby to position the metcury and the magnetic fluid axially within the tube.
3. A fluid switch as claimed in claim 1 in which the liquid electrical conducting material is a liquid metal segment, in which the fluid insulating material is a liquid insulating material segment and in which the means for propelling is adapted as well to maintain the segments in stable axial positions.
4. A fluid switch as claimed in claim 3 in which means is provided to prevent mixing of the liquid metal and the liquid insulating material.
5. A fluid switch as claimed in claim 4 in which the means to prevent mixing is a tube having inside cross dimensions small enough to permit surface tension of the liquid metal and surface tension of the liquid insulating material to retain both as segments within the tube.
6. A fluid switch as claimed in claim 1 in which the means for propelling comprises mechanical means to impart an axial force to at least one of the liquid electrical conducting material and the fluid insulating material, said axial force acting to effect propulsion of the electrical conducting material and the fluid insulating material axially along the tube as well as to establish stable axial positions for the same along the tube.
7. A fluid switch as claimed in claim 1 in which the means for propelling comprises electromagnetic means operable to impart an axial force to at least one of the liquid electrical conducting material and the fluid insulating material, said axial force acting to effect propulsion of the electrical conducting material and the fluid insulating material axially along the tube as well as to establish stable axial positions for the same along the tube.
8. A fluid switch as claimed in claim 7 in which the fluid insulating material is a magnetic liquid and in which the electromagnetic means comprises permanent magnet means magnetically coupled to the mag netic fluid and having means to move the permanent magnet means axially relative to the tube.
9. A fluid switch as claimed in claim 7 in which fluid insulating material is a magnetic liquid and in which the electromagnetic means comprises solenoid means which, when energized, provides a magnetic field magnetically coupled to the magnetic liquid.
10. A fluid switch as claimed in claim 1 in which the tube is a closed-loop tube wherein the liquid electrical conducting material and the fluid insulating material move axially in an endless closed-loop path.
11. A fluid switch as claimed in claim 10 in which the fluid insulating material is a liquid and in which the whole internal volume of the tube is filled with the combination of the liquid electrical conducting mate rial and the liquid insulating material.
12. A fluid switch as claimed in claim 1 in which the plurality of electrodes comprises a plurality of electrode pairs which act as pair combinations, electric current flow being between one electrode of the pair and the other electrode of the pair when the liquid electrical conducting material is disposed between a pair combination, said plurality of electrode pairs being positioned axially along the tube, and in which said means for propelling is operable to effect said axial movement but is adapted as well to maintain stable axial positioning of the liquid electrical conducting material and the fluid insulating material.
13. A fluid switch as claimed in claim l in which the liquid electrical conducting material is in the form of a segment and in which the fluid insulatiing material is also in the form of a segment, the segment of the liquid electrical conducting material being axially contiguous to the segment of the fluid insulating material, both ma terial segments being axially movable relative to said electrodes, and in which said means for propelling is adapted to maintain axial positioning of said segments.
14. A fluid switch as claimed in claim 13 comprising a plurality of segments of liquid electrical conducting material, each being separated from the other by a segment of fluid insulating material.
15. A fluid switch as claimed in claim 14 in which the lengths of at least some of the segments of either the liquid electrical conducting material or the fluid insulating material differ from one another.
16. A fluid switch as claimed in claim 15 in which the segment lengths are alterable and in which means is provided to alter said segment lengths in situ,
17. A fluid switch as claimed in claim l4 in which the plurality of electrodes comprises a plurality of electrode pairs which act as pair combinations, electric circuit flow being between one electrode of the pair and the other electrode of the pair when the liquid electrical conducting material is disposed between a pair combination, said plurality of electrode pairs being positioned axially along the tube.
18. A fluid switch as claimed in claim 13 in which the plurality of electrodes comprises a plurality of electrode pairs which acts as pair combinations, said combinations being electrically interconnected through auxiliary energy storage component means that acts. during switching, to arrest arcing so that each pair combination is protected from electrical arcing when current is switched, the electrodes of each pair combination being disposed transversely across the tube from one another.
19. A fluid switch as claimed in claim 18 in which the fluid insulating material comprises a high dielectric strength oil in which there are a first electrode pair, a second electrode pair and a third electrode pair displaced axially from one another along the tube, and in which the liquid electrical conducting material segment has an axial length sufficient to span the axial distance between two adjacent electrode pairs but not sufficient to span the axial distance between two non-adjacent electrode pairs, the switching sequence being such that the switching sequence is first electrode pair, second electrode pair and third electrode pair, and in which capacitances of diminishing magnitude act to limit current through the second electrode pair and the third electrode pair.
20. A fluid switch as claimed in claim 18 having a first electrode pair disposed, in axial location, between a second electrode pair and a third electrode pair, axial movement of the liquid electrical conductive fluid act ing upon switching the electric current, to time sequence the opening and closing events.
2 l A fluid switch as claimed in claim 20 in which the electric current is d-c, in which the auxiliary compo nents comprise a capacitance. a transformer and diode means, in which the primary of the transformer is connected in series with the electrodes of the third electrode pair and the combination thereof is connected in parallel with the capacitance, and in which the anode of the diode means is connected to the positive side of the first electrode pair and the cathode side is connected to one terminal of the capacitance, the other terminal of the capacitance being connected through the second electrode pair to the negative side of the first electrode pair, the secondary of the transformer being interconnected through further diode means back to the source of the d-c current thereby to allow energy stored by the capacitance to be returned to the source of the d-c current.
22. A liquid switch having, in combination, an elongate chamber, the walls of the chamber comprising an electrically insulating material, a plurality of adjacently disposed liquid segments within the chamber adapted to move as a segmented stream axially within the chamber, contiguous segments comprising materials of dif ferent electromagnetic properties, means for applying an axial force upon the liquid segments to effect propulsion thereof axially within the chamber, the cross dimensions of the chamber being sufficiently large than the axial propulsion force upon the segments is far greater than the effect of any net capillary action axial forces between the segments and the walls of the chamber, means for introducing electric energy to the segments in the chamber and or removing electric energy from the segments, and means to prevent mixing of the materials comprising said adjacently disposed segments, the means to prevent mixing comprising mechanical keeper means that serves to maintain the integrity of the liquid segments.
23. A liquid switch as claimed in claim 22 in which the mechanical keeper means comprises bobbin-type keepers.
24. A liquid switch as claimed in claim 22 in which the mechanical keeper means comprises hollow mechanical keepers.
25. A liquid switch as claimed in claim 24 in which the hollow mechanical keepers are disposed between segments and contain in the hollow interior liquid of the type in one of the segments.
26. A liquid switch as claimed in claim 22 in which the mechanical keeper means comprises mechanical keepers whose shape in cross section is identical to that of the elongate chamber, whose cross dimensions are 2O sufficiently smaller than the inner cross dimensions of the elongate chamber to maintain friction at acceptable limits but whose cross dimensions are sufficiently large to permit surface tension of the liquid to maintain the integrity of the segments so that the cross space of the tube is filled with liquid.
27. A fluid switch having, in combination, an elongate chamber, the walls of the chamber comprising an electrically insulating material, a plurality of adjacently disposed liquid segments within the chamber adapted to move as a segmented fluid stream axially within the chamber, the segments forming said fluid stream comprising a liquid conductive segment adjacent a liquid insulating segment, means for applying an axial force upon the liquid segments to effect propulsion thereof axially within the chamber, the cross dimensions of the chamber being sufficiently large that the axial propulsion force upon the segments is far greater than the effect of any net capillary action axial forces between said segments and the walls of the chamber, means for introducing electric energy to the segments in the chamber and for removing electric energy from the segments, and means to prevent mixing of the materials comprising said adjacently disposed liquid segments, the liquid insulating material being a magnetic fluid and the means to prevent mixing comprising magnetic keeper means.
28. A fluid switch as claimed in claim 27 in which the magnetic keeper means comprises permanent magnets disposed at predetermined axially spaced positions along the tube.
29. A fluid switch as claimed in claim 27 in which the magnetic keeper means comprises electromagnetic solenoid means,
30. Apparatus that comprises, in combination: a tube of electrically non-conductive material; a plurality of liquid segments within the tube adapted to move as a segmented stream therein, contiguous segments comprising materials having different electromagnetic properties from the immediately adjacent segment or segments, there being a stable transverse interface between contiguous segments; means for propelling all the liquid segments axially in the same directional sense along the tube; and means for introducing electric en ergy to at least some of the segments in the tube and for removing electric energy from at least some of the same segments in the tube; axial propulsion forces upon the segments along the tube by said means for propelling being far greater than the effect of any net capillary action axial forces between said materials and the tube.
3!. Apparatus as claimed in claim 30 in which the means for introducing electric energy and for removing electric energy comprises electrodes in electrical contact with the liquid segments and in which the means for propelling, in addition to effecting axial movement of the segments along the tube, serves to stop the segments at precisely positioned axial regions of the tube.
32. Apparatus as claimed in claim 31 in which some of the segments comprise a liquid electrical conducting material and other segments comprise a liquid insulating material, a segment of liquid electrical conducting material being adjacent a segment of liquid insulating material, the liquid electrical conducting material being non-wetting as to the tube and wetting as to the electrodes.
33. Apparatus as claimed in claim 31 in which sone segments comprise a liquid electrical conducting material and others comprise a liquid insulating material, a segment of liquid electrical conducting material being adjacent a segment of liquid insulating material. said tube being in the form of a closed loop so that the seg ment movement therein is along a closed-loop path.
34. Apparatus as claimed in claim 33 in which the liquid electrical conducting material is mercury and in which the liquid insulating material is magnetic fluid, the stream thereby formed being in a pattern is alternately a mercury segment, a magnetic fluid segment, a mercury segment. and so forth.
35. Apparatus that comprises, in combination: a tube of electrically non-conductive material; a plurality of liquid segments within the tube adapted to move as a segmented stream therein; contiguous segments comprising materials having different electromagnetic properties from the immediately adjacent segment or segments, there being a stable transverse interface be tween contiguous segments; means for propelling all the liquid segments axially in the same directional sense along the tube; and means for introducing electric energy to at least some of the segments in the tube and for removing electric energy from at least some of the same segments in the tube; axial propulsion forces upon the segments along the tube by said means for propelling being far greater than the effect of any net capillary action axial forces between said materials and the tube one of said segments comprising a liquid having magnetic particles therein, said appartus having electric coils wound around the tube and disposed axially along the tube, and means for exciting the coils to provide a travelling-wave magnetic field which moves axially along the tube and which couples electromagnetically with the liquid segment containing the mag netic particles, thereby propelling the liquid segment along the tube.
36. Apparatus that comprises, in combination; a tube of electrically non-conductive material; a plurality of liquid segments within the tube adapted to move as a segmented stream therein, contiguous segments com prising materials having different electromagnetic properties from the immediately adjacent segment or segments there being a stable transverse interfac oetween contiguous segments; means for propelling all the liquid segments axially in the same directional sense along the tube;
and means for introducing electric energy to at least some of the segments in the tube and for removing electric energy from at least some of the same segments in the tube; axial propulsion forces upon the segments along the tube by said means for propelling being far greater than the effect of any net capill-ary action axial forces between said materials and the tube.
one of said segments comprising a liquid having mag netic particles therein to provide a segment ot'one level of reluctance and another of said segments having a reluctance that differs from the firstnamed segment, said apparatus further including a low reluctance magnetic circuit magnetically coupled to the segments within the tube and affected by the segments as they move past the region of the magnetic circuit. thereby changing the reluctance of the magnetic circuit as the liquid stream moves axially along the tube and relative to the magnetic circuit.
37. Apparatus as claimed in claim 36 having a group of segments at each reluctance level and in which the magnetic circuit comprises a high permeability material in the form of a partial annulus but having a gap to form two pole pieces, a length of the tube being disposed within the gap so that as segments moving axially along the tube within the gap they introduce a variable reluctance there. the length of each segment being at least substantially the axial extent of the pole pieces.
38. Apparatus as claimed in claim 37 in which the faces of the pole piece are in intimate contact with the fluid segments.
39. Apparatus as claimed in claim 38 having first coil means to apply a magnetomotive force to the magnetic circuit and second coil means to extract a-c energy from the magnetic circuit.
40. Apparatus that comprises, in combination: a tube of electrically non-conductive material; a plurality of liquid segments within the tube adapted to move as a segmented stream therein, contiguous segments com prising materials having different electromagnetic properties from the immediately adjacent segment or segments, there being a stable transverse interface between contiguous segments; means for propelling all the liquid segments axially in the same directional sense along the tube; and means for introducing electric energy to at least some of the segments in the tube and for removing electric energy from at least one of the same segments in the tube; axial propulsion forces upon the segments along the tube by said means for propelling being far greater than the effect of any net capillary ac tion axial forces between said materials and the tube, said plurality of liquid segments comprising a first group of segments comprising a liquid with magnetic particles therein to provide segments of one level of reluctance and a second group of segments having a reluctance that differs from that of the first group. a segment of the first group being adjacent only to segments of the second group and vice versa, the segments of the second group being an electrical conducting material, said apparatus further including electromagnetic means for introducing axial propulsion forces directly to the segments of both groups.
4|. Apparatus that comprises, in combination: a tube; a plurality of liquid segments within the tube adapted to move as a segmented stream therein, contiguous segments comprising materials having different electromagnetic properties from an adjacent segment or segments, there being always maintained a stable transverse interface between contiguous segments; electromagnetic means applying an axial force upon the segments by electromagnetic interaction therewith to move the same along the tube; and mechanical actuator means that is actuated by axial movement of the liquid along the tube and thereby receives the force transmitted by the fluid and mechanically transmits the same to a system external to the tube; the interior cross dimensions of the tube being large enough to render any axial forces due to capillary action therein, small as compared to the force exerted upon the segments by said electromagnetic means.
42. A fluid switch for switching electric current that comprises, in combination: a tube of an electrically nonconductive material; contiguous regions within the tube having respectively liquid electrical conducting material and fluid insulating material, said regions being disposed axially along the tube, there being a stable transverse interface between the two materials at
|Cited Patent||Filing date||Publication date||Applicant||Title|
|US2744980 *||Jan 5, 1953||May 8, 1956||Cook Electric Co||Circuit control apparatus|
|US3289126 *||Jan 10, 1963||Nov 29, 1966||Fifth Dimension Inc||Mercury switch employing magnetizable fluid|
|US3308405 *||Aug 30, 1965||Mar 7, 1967||Fifth Dimension Inc||Magnetic piston mercury switch|
|Citing Patent||Filing date||Publication date||Applicant||Title|
|US4064409 *||Jul 28, 1976||Dec 20, 1977||Redman Charles M||Ferrofluidic electrical generator|
|US4167726 *||Nov 7, 1977||Sep 11, 1979||Ezekiel Frederick D||Magnetic liquid protector|
|US4384761 *||Jun 30, 1980||May 24, 1983||International Business Machines Corporation||Ferrofluid optical switches|
|US4418346 *||May 20, 1981||Nov 29, 1983||Batchelder J Samuel||Method and apparatus for providing a dielectrophoretic display of visual information|
|US4505539 *||Sep 7, 1982||Mar 19, 1985||Siemens Aktiengesellschaft||Optical device or switch for controlling radiation conducted in an optical waveguide|
|US5190304 *||Dec 13, 1991||Mar 2, 1993||Dee Prentiss||Back saver hand truck|
|US5912606 *||Aug 18, 1998||Jun 15, 1999||Northrop Grumman Corporation||Mercury wetted switch|
|US5969278 *||May 11, 1998||Oct 19, 1999||Tsai; Yu-Chu||Music producing beverage cup with electronic switch in a straw|
|US6318970 *||Mar 12, 1998||Nov 20, 2001||Micralyne Inc.||Fluidic devices|
|US6489694 *||Jan 22, 2002||Dec 3, 2002||Jacob Chass||Ferrofluidic, electrical power generator|
|US6571642 *||Mar 15, 2002||Jun 3, 2003||Murray F. Feller||Magnetic flow sensor with annular sensing path|
|US7105935 *||Jan 26, 2005||Sep 12, 2006||Industrial Technology Research Institute||Device of micro vortex for ferrofluid power generator|
|US7422709||May 23, 2005||Sep 9, 2008||Crosby Gernon||Electromagnetic rheological (EMR) fluid and method for using the EMR fluid|
|US8764410||Jun 10, 2009||Jul 1, 2014||Robert Bosch Gmbh||Capacitive device and method for the electrostatic transport of dielectric and ferroelectric fluids|
|US9293909 *||Apr 20, 2012||Mar 22, 2016||Abb Research Ltd||Passive circuit for improved failure mode handling in power electronics modules|
|US9339850||Aug 17, 2015||May 17, 2016||Cytonome/St, Llc||Method and apparatus for sorting particles|
|US20030075428 *||Oct 17, 2002||Apr 24, 2003||Tsutomu Takenaka||Electrical switches|
|US20050258090 *||May 23, 2005||Nov 24, 2005||Crosby Gernon||An electromagnetic rheological (emr) fluid and method for using the emr fluid|
|US20060110260 *||Jan 26, 2005||May 25, 2006||Industrial Technology Research Institute||Device of micro vortex for ferrofluid power generator|
|US20110005978 *||Jan 13, 2011||Cytonome/St, Llc||Method and apparatus for sorting particles|
|US20110220330 *||Jun 10, 2009||Sep 15, 2011||Thomas Proepper||Capacitive device and method for the electrostatic transport of dielectric and ferroelectric fuilds|
|US20150085415 *||Apr 20, 2012||Mar 26, 2015||Abb Research Ltd||Passive circuit for improved failure mode handling in power electronics modules|
|WO2010003752A1 *||Jun 10, 2009||Jan 14, 2010||Robert Bosch Gmbh||Capacitive device and method for the electrostatic transport of dielectric and ferroelectric fluids|
|U.S. Classification||335/47, 335/51, 335/55, 200/214, 335/49|
|International Classification||H01H36/00, H01H29/00|
|Cooperative Classification||H01H29/00, H01H36/00|
|European Classification||H01H29/00, H01H36/00|