US3292021A

US3292021A - Superconductive device

Info

Publication number: US3292021A
Application number: US274726A
Authority: US
Inventors: Ethan D Hoag
Original assignee: Avco Corp
Current assignee: Avco Corp
Priority date: 1963-04-22
Filing date: 1963-04-22
Publication date: 1966-12-13
Anticipated expiration: 1983-12-13
Also published as: DE1464774B2; CH439512A; GB1073960A; DE1464774A1

Description

E. D. HoAG 3,292,021

SUFBRCONDUCTIVE DEVICE Dec. 13, 1966 7 Sheets-Sheet 2 Filed April 22, 1963 H NC WT m W DS G mm A TW O E RT s G A o H D N A H T. E

INVENTOR.

BY u /Y ATTORNEYS Dec. 13, 1966 E, D. HOAG 3,292,021

SUPERCONDUCTIVE DEVICE Filed April 22.' 196:5 7 sheets-sheet s TO ROTARY DRIVEN STEPPING SWITCH 99 To BATTERY sa '04 TO ROTARY DRIVEN STEPPING SWITCH 99 Eig 'E ETHAN D. HOAG INVENTOR.

7 Shee tss-Sheet 4.

:mm umm nml oml vmml omml nml oml ETHAN D. HOAG INVENTOR.

,5MM/rfi: (f-,Q11

ATTORNEYS E. D. HOAG VI l V F V r hgw SUPERCONDUCTIVE DEVI CE :ml 3m l :ml :ml ml ovm .vml ovm r f Il Il rl l VI .wml wml wml wml uwm wml @wml uwm/ l V l *l H Il. f H. F F H. m l u mm Ol Dec. 13, 1966 Filed April 22, 196s Dec. 13, 1966 E. D. HoAG 3,292,021

SUPERCONDUCTIVE DEVICE Filed April 22, 1963 7 Sheets-Sheet 5 ATTORNEYS 7 Sheets-Sheet 6 Filed April 22. 1963 lOvnov-

ETHAN D. HOAG a INVENTOR ATTORNEYS United States Patent O 3,292,021 SUPERCONDUCTIVE DEVICE Ethan D. Hoag, Watertown, Mass., assignor to Aveo Corporation, Cincinnati, Ohio, a corporation of Delaware Filed Apr. 22, 1963, Ser. No. 274,726 13 Claims. (Cl. S10-40) This invention relates generally to superconductive devices and more particularly to means for and methods of inducing electric current ow in superconducting circuits.

It has been known since at Ileast early as 1911 that the electrical resistance of certain bodies of material disappears at a given temperature. Various scientific investigations in this field have resulted in many findings, some of which were not readily predicted, if predictable at all. These findings in some instances are related by various writers to classical theories for explanation and in other instances the findings are presented phenomenologically since some classical theories failed to provide a complete explanation. The characteristic of certain elements, numerous compounds and countless alloys to change from a resistive or normal state to a condition of zero electrical resistance at given temperatures is referred to as superconductivity. When a material undergoes such a transition, it is appropriately termed superconductive and the temperature at which the transition takes place in a material is referred to as the critcial temperature. Current fiowing in a closed superconductive circuit such as a coil intended to provide a given magnetic field may be designated persistent current, hence, a superconductive circuit having such a current fiowing therein may be referred to as operating in the persistent mode. The critical temperature varies with the different materials and for each material, this temperature decreases as the intensity of the magnetic field around the material is increased from zero. Once a body of material is rendered superconductive, it may be restored to the resistive or normal state by the application of a magnetic field of given intensity. The magnetic field necessary to destroy superconductivity is designated the critical field. Once a body of material is rendered superconductive. it may also be restored to the resistive or normal state if the current density therein exceeds a given value. The current density necessary to destroy superconductivty is designated the critical current density. In practice, temperature, current, and magnetic field are all interdependent and cannot be varied completely at will. For example, current and magnetic field are so intimately related that with most practical materials, it is not possible to achieve a critical field without first inducing a critical current density; however, for purposes of discussion, considerable latitude or emphasis in this respect is possible since in some cases a critical field may well be achieved without inducing a critical current density.

Magnetic field intensity is considered to be a controlling inuence in the destruction of superconductivity. Many writings with a thorough and detailed presentation of the phenomena and theories relating to superconductivity are available, one of which is Cambridge Monographs on Physics" (Superconductivity) Second Edition by D. Schoenberg. A description of one practical arrangement for securing low temperatures is presented in an article entitled The Cryotron-A superconductive Computer Component by D. A. Buck in the proceedings of the I.R.E. for April 1956.

A concept which is pertinent to the present invention is that a magnetic field applied to either a superconducting plane or an area enclosed by a closed superconducting loop cannot cause any net change in fiux through such plane or loop. In the case of superconducting loop, the net flux through the loop is maintained at zero by equal 3,292,621 Patented Dec. 13. 1966 ICC and opposite tiux lines which are supported by a circulat ing current around the loop. When the density of the circulating current exceeds the critical current density of any part of the superconductor comprising the loop, superconductivity is destroyed and the circulating currents are dissipated through 12R losses in the loop. Thus, when as a result of lowering its temperature a superconductive material passes from its normal state to its superconductive state in the presence of an externally applied magnetic field, it becomes a perfect diamagnetic and excludes an applied field entirely except in a thin surface layer. Presumably, during the course of transition from the normal to the superconducting state, multiple connected parts within the substance may develop which have the general form of closed superconducting regions surrounding cores of normal material. Such cores of normal material will have magnetic fiux passing through them. The perfect conductivity of the enclosing superconducting regions make it impossible for this fiux to change. Accordingly, a substance may retain a small magnetic moment proportional to the amount of flux trapped n this fashion even after an externally applied field has been reduced to zero. A persistent current exists in the superconductor around these cores of normal material and it is this current that maintains the trapped flux.

U.S. Patent No. 2,981,933, for example, discloses the provision of Iholes in a thin film of superconductive material wherein a magnetic field links two or three or more closely spaced holes. By pulsing a drive coil placed over the third hole, the flux linking the first two holes is made to transfer from one of them to the third hole. The result, after termination of the driving pulse, is the trapping of flux linking the third hole to one of the original two holes.

Although the manner in which fiux is trapped" is not fully understood as yet, one theory which has attempted to explain trapped" fiux (a concept also pertinent to the present invention) is the following:

If a single current carrying conductor is disposed over two holes in a superconductive film and is appropriately arranged to produce a magnetic eld that attempts to link the holes, this attempt is initially unsuccessful due to an opposing magnetic field established by circulating currents induced in the superconductive film immediately around the holes. As noted previously, the induced circulating currents prevent the fiux from penetrating the superconductive film. So long as the density of the circulating currents flowing in the small volume of superconductive film between the holes are less than the critical Current density of the film, the applied magnetic field is prevented from linking the holes by the opposing magnetic field produced by the circulating currents. However, when the density of the circulating currents exceeds the critical current density of the superconducting film between the holes, the area between the holes becomes resistive, the circulating currents will be dissipated due to the resistance of the film between the holes, the-re will now be only a minute opposing magnetic field, and the applied field will link the two holes. The heat generated by the transition from the superconductive to the normal resistive state and the heat generated by the circulating currents flowing through the resistive area will raise the temperature of the area between the holes to a temperature above the critical temperature of the superconducting film so that the latter will remain in the normal resistive state for a short period of time. If the applied current is maintained during the aforementioned short period of time, the produced magnetic field will remain and link the two holes. After the generated heat is dissipated by the superconductive environment surrounding the film such as, for example, liquid helium, the film returns to its superconductive state and if the applied current is removed, the magnetic field maintained by the applied current will attempt to collapse.

However, the attempted collapse of the magnetic field will induce circulating currents around the two holes which will maintain the field, thus trapping the field linking the two holes. t

According to the present invehtion, there is provided a unique and novel method of and apparatus for inducing current flow in a superconducting circuit. In one of its basic forms, the invention includes a plate of superconductive material wherein a normal region may be provided without substantially affecting persistent or dominant current ow in the plate and means for moving the normal region from an outer edge of the plate toward the opposed edge of the plate.

Broadly speaking, the choice, in some cases singly and in other cases in combination, of the manner in which a normal region is provided, whether the superconductive plate forms only part or all of a superconducting circuit, the direction in which the normal region is moved and the distance it is moved in the superconducting plate, determines the nature of the device. Thus, with the provision of magnetic lines of ux in a stable normal region at the outer edge of a plate of superconducting material and movement of the normal region with its magnetic ux to the inner portion of the plate, a predetermined amount of persistent current may be induced in the plate each time the normal region is moved, thereby providing from one point of view a permanent magnet the strength of which may nevertheless be changed, or, from another point of view, storage of predetermined amount or amounts of current in the superconducting plate. Alternately, if a dominant current flows in a plate per se, movement of the normal region from the center of the plate to the outer periphery thereof will decrease the magnitude of the dominant current. On the other hand, movement of the normal region with flux existing therein entirely across a superconductive plate forming part of a closed superconducting circuit will induce a current flow in the superconducting circuit. This latter embodiment of the invention is particularly useful for energizing closed superconducting circuits such as, for example, superconducting magnets.

Some of the maior problems encountered with superconducting magnets are those connected with energizing the magnet. Energization of superconducting magnets becomes increasingly difficult as the inductance of the magnet is decreased because of the heat leak encountered with the massive leads required for low inductance magnets. For example, a wire wound superconducting magnet may typically require a maximum current ow of only ten amperes to provide a given magnetic eld. Thus, its energization does not present any particular problem; however, a superconducting magnet wound from superconductive material in tape or strip-like form may require 1,000 amperes or more to provide a given magnetic field. Low inductance superconducting magnets and particularly such magnets wound from superconductive tape or striplike material are desirable from both a structural and electrical point of view since such magnets have far fewer turns, withstand higher J x B loadings, and have much lower transient voltages than equivalent wire wound superconducting magnets. The present invention facilitates the energization of superconducting magnets having a low inductance as well as those having a high inductance since apparatus in accordance with the present invention operates at liquid helium temperatures and its input impedance is completely independent of the inductance of the superconducting magnet it energizes. Since only small leads need be brought into the helium irrespective of the magnitude of the persistent current that it is desired to induce inthe magnet, heat leak due to leads is minimal. Devices in accordance with the present invention need not have any moving parts and in any event do not have any inherent power limitations so far as the rate at which energy is induced into a superconducting circuit is concerned.

Accordingly, it is an object of the present invention to provide apparatus for and a method of inducing current flow in a superconducting circuit.

Another object of the present invention is to provide apparatus for inducing current ow in a closed superconducting circuit.

It is a further object of the present invention to provide energizing apparatus for energizing closed superconducting circuits such as superconducting magnets wherein the input impedance of the energizing apparatus is completely independent of the inductance of the circuit being energized.

A still further object of the present invention is to pro'- vide means for and a method of energizing superconducting circuits and particularly superconducting circuits having a low inductance and large persistent currents wherein heat leak is maintained at a minimum.

Yet another object of the present invention is to provide energizing apparatus having no moving parts and no inherent power limitations for energizing closed superconducting circuits.

Other objects of the invention will be pointed out in the following description and claims and illustrated in the accompanying drawings, which disclose, by way of example, the principle of the invention and the` best mode which has been contemplated, of applying that principle.

The novel features that are characteristic of the invention itself, however, both as to its organization and method of operation, together with additional objects and advantages thereof, will best be understood from the following description of specific embodiments when read in conjunction with the accompanying drawings, in which:

FIGURE 1 shows a superconductive loop to illustrate the principle of the present invention;

FIGURE 2 illustrates one embodiment of the present invention having no moving parts for inducing current in a closed superconductive circuit;

FIGURE 3 illustrates another embodiment wherein a movable electromagnet is utilized to induce current in a superconductive circuit;

FIGURE 4 shows metallic strips added to a superconductive plate to facilitate establishment and maintenance of a stable normal region in the superconductive plate;

FIGURE 5 shows another arrangement of the present invention utilizing an iron core and which does not have any moving parts;

FIGURE 6 is a fragmentary view of part of the iron core shown in FIGURE 5 containing matrix wires;

FIGURE 7 is a schematic diagram showing the connection of the matrix wires in the legs of the iron core in FIGURE 6;

FIGURE 8 shows still another arrangement of the present invention which utilizes air core coils and that does not have any moving parts;

FIGURE 9 is a schematic diagram illustrating one way of energizing the air core coils of FIGURE 8; and

FIGURE l0(a)-l0() are graphic representations of the variation of current density in the air core coils of FIGURE 8.

Referring to FIGURE 1, there is shown for the purpose of illustrating the principle of the invetnion, a closed ring or loop of strip-like superconductive material forming a conductor 1. It has been previously noted that a magnetic l'leld of less than critical eld strength applied to either a super-conducting plane or an area enclosed by a superconducting loop such as, for example, area 2 of FIGURE l, cannot cause any net change in magnetic lines of flux 3 through such a plane or loop. Accordingly, application of a magnetic eld of less than critical eld strength cannot cause any net change in magnetic lines of ux 3 passing through the area 2 enclosed by conductor l. As noted previously, whereas a superconducting loop or plane is a perfect diamagnetic and excludes an applied magnetic eld completely, fixed normal regions may exist in a superconducting plane through which magnetic ux can pass. It therefore follows (as is taught by the prior art) that the mere existence of magnetic ux in an interior normal region cannot cause any net change in flux through such a plane or loop. However, a net change in flux through a superconductingloop or plane or a net change in current owing in the loop or plane, which are one and the same thing, may be simply and efiiciently achieved in accordance with the present invention. It is important to note that the dimensions of the conductor 1 illustrated in FIGURE 1 are such that a normal region may be provided through the yconductor without substantially affecting dominant current ow in the conductor when it is superconductive, i.e., operating in the persistent mode. Otherwise stated, the dimensions (thickness, width and length) of the conductor 1 are such as to permit not only the establishment of a stable normal region 4 through the conductor as more fully described hereinafter, but also the establishment of a stable normal region without affecting such persistent or dominant current ow as may exist or be induced in the conductor. Thus, the dimensions of the major surfaces 5 and 6 of' conductor 1 may easily be made greater than the greatest dimension of the stable normal region 4 established therein. Accordingly, if the normal region 4 illustrated in FIGURE l is first established at a point 7 which includes the edge of the conductor, magnetic lines of fiux 8 may be made to pass through the normal region and, hence, through the conductor 1. Since the conductor is driven normal on a local basis only, so that its superconductive continuity taken as a whole is never interrupted, the normal region 4 may be moved not only within but across the conductor without substantially affecting the flow of dominant current therein. Still further, because a superconducting region always surrounds the stable normal region when moved to a point within the periphery of the conductor, the aforementioned magnetic fiux passing through the stable normal region may be made to move within the conductor by moving the normal region containng the magnetic flux. Accordingly, the normal region 4 with its fiux 8 may be moved from the point 7 at the edge of conductor l across the conductor until it arrives at point 9 which includes the opposite edge of the conductor without any undesirable effects. The attempted collapse of the magnetic lines of fiux 8 upon arrival at a point 9 on the opposite edge of the conductor results in the inducement of persistent current fiow in the conductor. Obviously, if the normal region 4 is successively moved across the conductor in the manner described above, persistent or dominant current flow in the conductor may be increased until the critical current is reached. Similarly, if either the direction of movement of the normal region or the direction of the fiux is reversed, the magnitude of current flow previously induced in the conductor will be reduced. Movement of the normal region 4 is illustrated by arrows 11. As will now be obvious, the present invention permits selectively increasing (or decreasing) the net ux through an area enclosed by a superconducting loop, a result heretofore considered impossible. Otherwise stated, the present invention permits the establishment, reduction, or variation of a single dominant current flow in a closed superconducting circuit.

As will now be evident, persistent current may be nduced in a superconducting circuit represented by conductor 1 by creating a stable normal region through the conductor whereby magnetic lines of flux may pass through the normal region, providing magnetic lines of flux through the normal region so established, moving the normal region with the liux existing therein across the superconducting circuit, and removing the original source of magnetic lines of fiux. Further, if the superconducting circuit is a plate, movement of the normal region with its magnetic flux from the periphery of the plate to another point within the plate such as, for example, the center of the plate, in the same manner disclosed a-bove, will induce a circulating current in the plate, the density of which current may be increased to the critical current density of the plate by consecutively permitting the magnetic lines of fiux to become trapped by stopping them and allowing the normal region to recover its superconducting state.

Attention is now directed to FIGURE 2 which illustrates one` embodiment of the present invention for inducing in a superconducting circuit substantially any current density less than the critical current density of the superconducting circuit. As shown in FIGURE 2, a plate or strip 21 of superconductive material such as, for example, hib-25% Zr (niobium-zirconium alloy) is provided in the air gap 22 of a laminated iron core 23. The laminations are not shown for clarity. The laminated iron core 23 is comprised of an elongated base portion 24, an end portion 25, an upper portion 26 parallel to and spaced from the base portion 24 and a plurality of separate depending finger-like portions 27-32 (six as shown) carrying respectively electrical field coils 33-38. The depending portions 27-32 terminate adjacent the base portion 24 to form the aforementioned air gap 22 (actually a plurality of air gaps) and are spaced apart a distance such that the magnetic field in the air gap 22 produced by each field coil 33-38 includes a portion of the area encompassed by the magnetic field produced by an adjacent field coil. The field coils 33-38 disposed on the depending portions are essentially identical one to another as to the direction of winding, number of turns and resistance, i.e., each coil produces essentially the same number of ampere turns when connected to the same source of current. The first and last depending portions 27 and 32 are respectively disposed at least adjacent the

edges

41 and 42 of the superconductive strip 21. The base portion and the depending portions are chamfered to concentrate the magnetic lines of iiux therebetween in the air gap. The

end portions

43 and 44 of a continuous superconductive wire 45 forming a coil 46 wound on a mandrel 50 are electrically connected as by spot welding to respectively the

end portions

47 and 48 of the ssuperconductive strip furthest from the iron core whereby any persistent current flowing or induced in the superconductive circuit comprising the strip 21 and the superconductive coil 46 flows through the strip 21 in a direction transverse of a plane passing thruogh each of the depending portions of the iron core. The broken line 49 surrounding the iron cere 23 and the superconductive circuit, designated generally by the numeral 5l, indicates a superconductive environment such as a dewar containing liquid helium. Identical terminals of each of the field coils are connected through a common conductor 52 to one terminal of a source of current represented by battery 53 and the remaining terminals of the field coils are connected through a conventional rotary-driven stepping switch 54 such as, for example, the rotary-driven wafer type manufactured by the Oak Manufacturing Company, to the other terminal of the battery 53 whereby as the wiper or the like of the stepping switch is rotated the field coils 33-38 are sequentially pulsed such as, for example, from left to right in FIGURE 2. With the exception of the outermost field coils 33 and 38 respectively disposed adjacent the

edges

41 and 42 of the superconductive strip, the electrical circuit to each successive field coil is completed before the circuit to the preceding field coil is broken. Thus, before the circuit to the first field coil is broken, such as, for example, field coil 33, the circuit to the next succeeding field coil such as, for example, field coil 34 is completed until the circuit to the last field coil (coil 38) is completed. The circuit to the first field coil (coil 33) is not completed again until after the circuit to the last field coil (coil 38) has been broken. The circuit to the last field coil is broken before the circuit is completed to the first field coil to permit the stable normal region and the magnetic lines of fiux established in the superconductive strip by the last coil to disappear. y

In a device successfully tested and essentially identical to that disclosed in FIGURE 2, the core was composed of 4 mil laminations of transformer grade steel. Each of the eld coils on the core contained 200 turns of .004 inch diameter Nb wire and provided approximately 200 ampere turns. The superconductive strip was comprised of Nb-25%Zr. The dimensions of the strip were approximately .002" x 1" x 2". The coil of the superconductive magnet was comprised of 600 turns of .020 inch diameter Nb-25%Zr wire wound on a l/z" x l mandrel. The coils on the core were sequentially pulsed in the man` ner described hereinabove with a basic period ranging from .3 to 1.0 seconds. Approximately 136 minutes were required to provide a 4 kilogauss magnetic field at the center of the coil and a persistent or dominant current of approximately 2O amperes before the point of quench was reached. The superconductive magnet was spaced less than 2 inches from the core.

A stable normal region is provided through the plate strip 21 and caused to move across the strip in the following manner. Sequential energization of the coils 33- 38 by means of the rotary switch 54 causes a magnetic eld to be progressively moved across the strip. The moving magnetic field induces local current densities in the portion of the strip subjected to the magnetic iield which exceed the critical current density of the strip. Accordingly, only the portion of the strip in the air gap subject to the instantaneous magnetic eld is driven normal thereby providing a normal region that has a predetermined and essential constant size (i.e., the normal region is stable) and that does not extend to such an extent as to prevent the ow of dominant current in the strip.

Upon creation of the stable normal region at the outer periphery or edge 41 of the strip, magnetic lines of ux pass through the strip in the stable normal region. Pro gressive movement of the stable normal region toward the inner periphery or edge 42 of the strip resulting from sequential pulsing of the coils 33-38 causes the magnetic lines of ilux to move with the normal region until the inner periphery or edge 42 of the strip is reached. At this point, all of the eld coils 33-38 are open circuited. When the magnetic lines ot' flux which have now been introduced into the area enclosed by the superconductive loop which includes the plate 21 and coil 46 attempt to collapse because the field coils 33-38 have all been open circuited, the attempted collapse of these magnetic lines of ux induces a persisitent current in the superconductive circuit 51 which maintains at least part of the field, thus trapping the lines of llux introduced into and passing through the superconducting loop. Accordingly, the net tlux through the superconducting loop and, hence, dominant current ow in the loop is increased each time the stable normal region is moved from the outer periphery or edge 41 to the inner periphery or edge 42 of the strip. Thus, each time the above-described cycle is repeated, the magnitude of the persistent current in the superconductive circuit 51 will be increased until the limit of the critical current density is reached. If the critical current density is reached, the superconductive circuit 5l will of course be driven normal and the previously induced current will be dissipated as 12R losses.

In addition to inducing a predetermined amount of current in the superconducting coil 46 and thereby energizing or starting it up, the apparatus disclosed in FIGURE 2 may also be utilized to de-energize or shut down the superconducting coil when operating in the persistent mode. For shutdown, the stable normal region may be created at the inner edge 42 of the strip 21 and moved to the outer edge 41 such as, for example, by reversing the direction of rotation of the rotary switch 54 used for start-up. Alternately, the direction of the lines of magnetic flux may be reversed such as, for example, by reversing the battery connections used for start-up. It should also be noted here that the present invention is not limited to the use of mechanical means such as a rotary-driven switch for pulsing the coils 33-38 and, hence, effecting movement of the stable normal region. Any one of a number of other conventional and well known mechanical and/ or electrical schemes may be used with equal facility.

For start-up, the direction in which the stable normal region is moved and the direction of the magnetic lines of flux is generally not important since this merely determines the direction of ow of the persistent current that is induced in the superconductive coil 46. However, shutdown of a coil operating in the persistent mode will not occur if both the direction of movement of the stable normal region and the direction of the magnetic lines ot' ux are reversed as compared to that used to start up the coil.

Energization of a closed superconductive circuit 51 as disclosed in FIGURE 2 is somewhat sensitive to the velocity of the stable normal region. For example, for any given velocity of the stable normal region, there appears to be a maximum persistent current that may be induced in a closed superconductive circuit. When this maximum persistent current for a given velocity of the stable normal region is reached, continuous movement of the stable normal region at the same or a greater velocity is ineffective. However, if the velocity of the stable normal region is decreased, current will again be induced into the circuit -until a new maximum persistent current is reached. Accordingly, for energization of large coils and the like, high velocities may be most efficiently used when the persistent current is low, progressively lower velocities being used for persistent currents of progressively greater magnitude.

A word may also be said at this point regarding the location of an iron core or the like similar to iron core 23 with respect to a coil or the like similar to coil 46. Since the strip 21 must of necessity be superconductive, inasmuch as it forms at least part of the superconductive circuit, the iron core 23 of FIGURE 2 with its associated eld coils is most conveniently located adjacent coil 46, the only leads emerging from the superconducting environment 49 being the small leads connected between the iield coils 33-38, the rotary-driven stepping switch 54 and the battery 53. The iron core need only be spaced from the coil 46 a sufcient distance that the maximum magnetic field provided by coil 46 does not result in saturation of the iron core since this may render the iron core inoperative for its intended purpose.

Notwithstanding the fact that most ferromagnetic core materials saturate at around 15,000 to 24,000 gauss and that the coil 46 may not provide a magnetic lield of such strength, the iron core disclosed in FIGURE 2 may well become saturated if it is placed too close to the coil 46 because of the tendency of the lines of ilux generated by coil 46 to travel through the low reluctance path provided by the iron core. Of course one way to prevent saturation of the iron core is to shield it in any one of a number of suitable and well known ways. However, shielding of the iron core will normally not be required since it usually may be placed a sufficient distance from the coil as to be in a magnetic eld less than that which will cause saturation. Such a location can be easily determined. For example, the field strength of the coil 46 at various points for normal operating conditions can be ploted or empirically determined thereby permitting the initial selection of a presumably suitable position. Having thus selected a location for the iron corewhich is believed to be suitable, one can begin to energize the coil. If the etort is unsuccessful, it is only necessary to move the iron core further from the coil -until the desired magnitude of current can be induced in the coil.

As will now be evident, FIGURE 2 discloses apparatus which utilizes a plurality of eld coils on an iron core, the position of which is fixed relative to the superconductive plate in which it is desired to provide a moving stable region. The converse of FIGURE 2 is illustrated in FIG- URE 3 wherein a generally rectangularly shaped magnetic circuit comprising an iron core having a central opening and an air gap in one leg is movable with respect to a superconductive plate in which it is desired to provide a stable normal region. As shown in FIGURE 3, a field coil 6l surrounds one leg. of an ircn core 62 having a central opening and which is provided with an air gap 63 in another leg to receive a superconductive plate 64. A source of current for the field coil 61 is represented by battery 65. The ampere turns of the field coil 64 must be sufficient to induce in the superconductive plate 64 local currents having a density greater than the critical current density of the superconducting plate 64. This is, of course, equally true for the field coils 33-38 disclosed in FIGURE 2.

Movement of the iron core 62 with respect to the superconductive plate 64 is indicated by the double-headed arrow 66. Whereas one extreme position of the iron core 62 is indicated in phantom which suggests that the iron core 62 is moved rather than the plate 64, it is to be understood that the iron core 62 can be fixedly supported and the plate 64 moved by any suitable and conventional means. The particular means of supporting and/or actuating the iron core or the superconductive plate, as the case may be, is not essential to the present invention. Those skilled in the art will have no difficulty in selecting any one of a number of well known and suitable means for accomplishing this purpose, hence, it is not considered necessary to describe such conventional apparatus in any detail. However, by way of example, the superconductive plate may be rigidly supported by leads 67 and 68 (reinforced if necessary). On the other hand, the iron core may be supported and moved transverse of the plate by a conventional cam and shaft arrangementy or, for that matter, the iron core may be electromagnetically actuated. Electromagnetic actuation is perhaps preferable over a mechanical arrangement, the driving means for which is locoated outside of the superconductive environment, because the electromagnetic scheme maintains heat leak at a minimum. In the embodiment illustrated in FIGURE 3, the stable normal region is established in the plate 64 and functions in the same manner as discussed in connection with FIGURE 2.

If desired, an alternating magnetic field may be superimposed on the constant magnetic field provided by the apparatus of FIGURE 2 or FIGURE 3 to facilitate establishment and maintenance of the stable normal region by inducing A C. eddy currents in a normal metal, thereby producing heat adjacent the superconducting material and lowering its critical current density. y

In FIGURE 4, there is shown a pair of electrically conductive, normal and non-magnetic strips 7l and 72 such as, for example, copper carried by the plate of superconductive material 73 to facilitate establishment of the stable normal region. The copper strips 71 and 72 are attached to the opposed

major surfaces

74 and 75 of plate 73 and disposed in the aforementioned air gaps. Eddy currents induced in the copper strips by the moving magnetic field are helpful in establishing the stable normal region in the superconductive strip between these copper strips.

At this point, it will be helpful to note that separate and distinct means, while they may he used, are not necessary to establish the stable normal region and the magnetic lines of ux through the normal region. Accordingly, for example, a point source of heat (not shown) sufiicient to drive the superconductive strip normal on a -local basis only may be provided in combination with means for providing a magnetic field, both of which are simultaneously moved across the superconductive strip as I and for the purposes set forth hereinabove.

It has been previously noted that saturation of the iron core is undesirable. While the above statement is generally true, par-tial saturation of an iron core to provide moving low reluctance or ux conducting regions may,

in accordance with the present invention, be advantageously utilized to provide a moving stable normal region.

With reference now to FIGURE 5, there is shown a generally U-shaped iron core 81 having a field coil 82 disposed at the bight of the U. The

legs

83 and 84 of the iron core are enlarged and extend inwardly toward each other to provide a rectangular air gap 85 for receiving a strip of superconductive material 86. The air gap 85 extends outwardly past both

edges

87 and 88 of the strip of superconductive material. The field coil is connected in series through a switch 89 to a source of cur. rent represented by battery 90. A matrix, designated generally by the numeral 91, of current carrying wires preferably composed of superconductive material, is provided through the portions of the

legs

83 and 84 of the iron core adjacent the air gap 85. The matrix 91 of current carrying wires in combination with a source of current and suitable switching means as more fully described hereinafter (see FIGURE 7), functions to provide a moving unsaturated path in the legs of the iron core nonnal to the major surfaces of the superconductive strip.

Broadly speaking, the matrix of current conducting wires is so arranged and adapted that the net magnetic field provided by the matrix is zero but is strong enough .locally to saturate the iron cone at all points adjacent the superconductive strip.

As shown by way of example in FIGURE 5 and FIG- URE 6, a plurality of closely spaced current carrying wires (ten as shown in FIGURE 5) are provided in the inner portion of each leg adjacent the air gap. A portion of each wire is disposed in each leg and, as -best shown in FIGURE 6, each such portion `beginning at a point adjacent the air gap passes back and forth through each leg in a plane normal to the air gap and parallel to the direction of dominant current flow. The current ow and direction of magnetic tiux surrounding two adjacent wires are shown in FIGURE 6. As will now be evident, if current flows in opposite directions in adjacent wires, the net magnetic field produced thereby in each leg is zero so far as the liux created by field ooil 82 is concerned but is sufficient to saturate the iron core adjacent the air gap when current is supplied simultaneously to all of the wires. It can be shown quite easily that the conditions for saturation can be met using transformer grade steel for the core and Nb-25%Zr for the matrix wires. With suicient current flowing in each of the matrix wires in each leg of the iron core to saturate the portions of the iron core containing the matrix wires, it will be obvious that the lines of magnetic liux provided lby the field coil 82 will, for all practical purposes, be prevented from passing through the air gap. However, if the supply of current is broken to the portions of any one matrix wire lying in the same plane, the region adjacent these two particular portions will no longer be saturated and therefore the magnetic lines of flux generated by the field coil may pass therethrough and will be concentrated in a relatively small region of the air gap located between the aforementioned portions of the wire. Further, if the supply of current to an adjacent matrix wire is now broken and thereafter current is again supplied to the first mentioned matrix wire, the aforementioned unsaturated regions will move to a new position. Accordingly, it will now Ibe eviden-t that oppositely disposed unsaturated regions may be initially created in the legs of the iron core at the outer edge 87 of the superconductive strip and thereafter caused to move through the legs of the iron core to the opposite edge 88 of the superconductive strip in the manner described hereinabove. Such a condition is shown by way of example in FIGURE 5 at a given time during one cycle. The regions a, 100b, 10la and 101b enclosed respectively by broken lines 102a, 102b, l03a, and 103b designate saturated regions in the legs of the iron core. The oppositely disposed unsaturated regions intermediate ythe oppositely disposed saturated regions 100a-l01a and l00b-101b are designated by respectively the

numerals

104 and 105.

1 1 Attention is now directed to FIGURE 7 which schematically illustrates one way of providing a moving unsaturated region in the iron core shown in FIGURE 5. The portions of the matrix wires in leg 83 are designated by the numerals 92a--92i and the portions of the matrix wires in leg 84 are designated by the numerals 93a-93j.

lf desired, the portions 94a-94i of the matrix wires may be formed into a cable or the like and lbrought around the back of the air gap 85 to permit insertion and removal or extension of the superconductive strip 86. One terminal of the serially-connected portions of the matrix wires are alternately connected respectively through

common conductors

96 and 97 to one terminal of a current source illustrated yby battery 98 to provide the necessary reverse current ow in adjacent wires, the remaining terminals of the matrix wires being connected to the other terminal of the battery 98 through a conventional rotarydriven stepping switch 99. The arrows adjacent the portions 94a-94i of the matrix wires indicate the direction of current ow in these Wires. The switch 99 is of the rotary-driven type wherein all contacts are normally closed and as a wiper arm or the like is rotated, the electrical circuit to each succeeding contact is broken whereafter the circuit to the preceding contact which was previously open is closed. Accordingly, a stalble normal region may be established and caused to move entirely across the superconductive strip in the following manner:

Upon energization of the field coil and the provision of current to all of the matrix wires, the circuit to

portions

92a and 93a of the first matrix wire is broken. This permits the lines of ux generated by the field coil to be applied to the edge 87 of the superconductive strip 86 and drive it normal at this point. Field coil 82 provides lines of ux in the iron core 8l sufiicient to induce critical current densities in the superconductive strip located in the air gap 85 (thereby creating a stable normal region) but insuflicient to saturate the aforementioned unsaturated portions of the legs of the iron core. Of course, when the portion of the superconductive strip subjected to the magnetic field concentrated by the relatively small volume of the unsaturated portions of the iron core

adjacent portions

92a and 93a is driven normal, lines of magnetic ux pass through the superconductive strip 86 at this point. The circuit to the next succeeding

portions

92b and 93b is now broken and after a short delay, the circuit to

portions

92a and 93a is completed. As a result, the stable normal region containing lines of ux is caused to move from edge 87 toward edge 88 of the superconductive strip to a new location between

portions

92b and 93b.

The above-described procedure is repeated for all of the remaining matrix wires, thereby causing the stable normal region to move entirely across the superconductive strip from edge 87 to edge 88 and induce a small amount of persistent current in the superconductive circuit. After the circuit to the last matrix wire (portions 921' and 93j) is broken, the circuit to the first matrix wire (

portions

92a and 93a) is again broken and the cycle repeated until the desired magnitude of persistent or dominant current has been induced in the superconductive circuit.

Attention is now directed to FIGURE 8 which illustrates an embodiment of the invention utilizing air core coils to provide the moving normal region. Whereas the use of an iron core imposes certain limitations because of the possibility of saturation of the iron core, no such limitations are encountered when air core coils, now to be described, are used. As shown in FIGURE 8, a super` conductive strip 111 is disposed between two

groups

112 and 113 of current conducting coils designated respectively by the numerals 112a-112y and 113a-1l3y which are preferably wound from superconducting wire. Supporting means for the coils and the superconductive strip may be of any conventional form and are not shown for purposes of clarity. Each group of coils is coaxial about a separate axis parallel to the major surfaces 114 and of the superconductive strip 111 and normal to the direction of dominant current ow in the strip. Both axes preferably lie in a single plane normal to the major surfaces of the superconductive strip. As shown in FIGURE 8, each group of current conducting coils is adjacent respectively one each of the major surfaces of the superconductive strip (coils 112a-1l2y are adjacent major surface 114 and coils 113a-113y are adjacent major surface 115) and extend thereacross to at least opposite edges 116 and 117 of the major surfaces. The wire forming the coils is preferably insulated and hence, the coils may rest on the superconductive strip.

The coils comprising each group are separately connected through switching means, generally designated 118, to a source of current as more fully described hereinbelow effective to supply current to the coils to cause magnetic flux generated by these coils to be applied to and move entirely across the superconductive strip 111. Attention is now directed to the fact that the portions of the coils forming each group and adjacent the superconductive strip 111 define

regions

121 and 122 in which current density of a given magnitude may flow in a given direction at a given time. Current density in

regions

121 and 122 only is essential in this embodiment of the invention. As will become more evident hereinafter, the use of coils or toroids is a simple and efficient way of providing the desired current density in all parts of the

regions

121 and 122. Accordingly, it is to be understood that the invention is not limited to the use of coils. Current density in the

aforementioned regions

121 and 122 adjacent the

major surfaces

114 and 115 of the superconductive strip 111 is in one direction, designated by

arrows

123 and 124, only at the beginning (or end) of a cycle. Accordingly, the current supplied to the coils defines a current density having a predetermined magnitude and direction in the

regions

121 and 122 occupied by the conductors forming these coils. To facilitate discussion of the operation of the embodiment disclosed in FIGURE 8, X, Y, and Z axes are indicated. The magnitude of the current density in

regions

121 and 122 is essentially determined by the source or sources of current. As previously noted, current density in both regions is in one direction at all points only at a given point in time. This produces lines of magnetic ux 125 that encircles the superconductive strip as shown. If these lines are now caused to be applied to the

major surfaces

114 and 115 of the superconductive strip as indicated at 125a and 125b, instead of merely encircling it, a current flow 126 on a local basis will be induced in the superconductive strip 111. Accordingly, the current density in a region or regions adjacent the superconductive strip may be utilized to induce current density in the strip on a local basis. Since the cross sectional area of the superconductive strip 111 can be made quite small as compared to the cross sectional area of the region or

regions

121 and 122, which latter region or regions can be made quite large, critical current density may be easily induced in the superconductive strip. Accordingly, the cross sectional area of the superconductive strip and/or the cross sectional area of the coils adjacent the superconductive strip is selected to provide a critical current density in the superconductive strip.

Obviously, if the point of application of the lines of magnetic liux to the superconductive strip is limited in size and is made to move, a stable normal region 127 may be established in the superconductive strip and made to move across it. Briefly, lines of magnetic ux are applied to the superconductive strip and the point of application varied by sequentially reversing the ow of current in opposed pairs of coils (one each being in each group of coils such as, for example, coils 112a and 113a) from one edge to the opposed edge of the superconductive strip. Again, note that the outermost coils (coils 112a- 113a and 112y-113y) extend outwardly past the edges 116 and 117 of the superconductive strip 111. The dimension of each coil in the Y direction is made sufficiently small compared with the dimension of the superconductive strip in this same direction (its width direction) to make the movement of the stable normal region 127 across the strip as constant as is reasonably possible.

Directing attention now to FIGURE 9, it will be seen that a normal path comprising a resistance (not shown in FIGURE 8) is provided across the terminals of each coil to facilitate reversal of current ow in each coil. The ratio of L/R where L is the inductance of each coil and R is the resistance of the resistor connected across the terminals of this coil) must be greater than the time required to actuate the switch associated with each coil and less than the time that elapses between the actuation of one switch and the next succeeding switch discussed hereinbelow. This insures that the current in a given coil actually reverses before the switch associated with the next succeeding or adjacent coil in each group is actuated to permit reversal of current in these coils. If this were not true, then the actual time required for the current to reverse in each coil could be sufiiciently great that in the limit the current in all coils would reverse simultaneously and thereby render the device inoperative for its intended purpose.

One end of each coil as shown in FIGURE 9 is connected through a common conductor 141 to opposite terminals of two sources of current represented by

batteries

142 and 143. Two batteries, the polarities of which are reversed, permit, in combination with the switching means 118, reversal of current ow in the coils. The remaining terminals of the batteries are separately connected through

common conductors

144 and 145 to respectively the terminals b and a of switches 146-171', there being one switch for each pair of coils as shown in FIGURE 8. Each switch has two terminals designated by the subscripts a and b and a switch arm designated by the subscript c which may be actuated in any conventional manner, such as, for example, by properly phased rotary-driven cams (not shown). The switch arm c of each switch is connected to one each of the remaining terminals of the coils. For of clarity, only the connection of the parallel combination of coils 112a-112y and resistors 140a140y is shown in FIGURE 9-the coils 113a and 113y of the second group are'respectively connected in parallel across the coils shown in FIGURE 9 to provide the proper current ow in

regions

121 and 122. With all of the switches in the same position, for example, the a position as shown in FIGURE 9, it will be readily seen that current will fiow in the same direction through each coil and that reversal of each switch arm to the b position will reverse the direction of current iow in each pair of opposed coils and thereby change the polarity of the current density J in the region or regions adjacent the superconductive strip.

FIGURES 10a-10i illustrate the variation of current density J along the Y axis in each region as a result of actuation of the switches associated with each coil. For example, assume that at time t, the current density is negative and constant throughout the region, i.e., current flows in the same direction at all points in

regions

121 and 122 and each coil provides the same number of arnpere turns. This is illustrated in FIGURE 10a. At time t1, after the first switch arm such as, for example, switch arm 146e has been actuated to reverse current flow in the first coil or coils (coils 112a and 113a), the current density along the Y axis (also the width dimension of the superconductive strip 111) is indicated in FIGURE lOb. At time t2, after sequential actuation of, for example, switch arm 147C. the current density will be as shown'in FIGURE 10c. Thus, as the switches are sequentially actuated, the polarity of the current density along the Y axis progressively reverses until at time t5 it is of the same magnitude as at time lo but is of opposite polarity all along the Y axis. At time t5, all of the switches have been actuated from, for example, temrinal a to terminal b. If all of the switches which are now in the up or b position) are simultaneously actuated to their down or a position, indicated by solid lines in FIGURE 9, the polarity of the current density all along the Y axis will vary substantially as indicated in FIGURES lOf-lOi until at time t@ the magnitude and polarity of the current density is identical to that existing at time tu. At this time, the cycle is repeated.

Sequential reversal of current ow in the coils causes at least part of the lines of magnetic flux which initially encircled the superconductive strip 111 (see FIGURE 8) to now pass through the superconductive strip 111 at, for example, the point 172 where current fiows in opposite directions in each group of coils. FIGURE 8, assuming that the first ten switches 146-155 have been actuated, eurent ow in the first ten coils 112:1- 112i and 11M-113i is in one direction and current ow in the remaining coils 112k-l12y and 113k-113y is in the opposite direction. The concentration of the magnetic lines of flux at point 172, for example, which is to say at the point cf reversal of polarity in current density (see FIGURES 10b-10e) induces critical current density 126 in the superconductive strip 111, thereby establishing a stable normal region 127 which moves at a rate determined essentially by the rate at which the switches are actuated. A stable normal region 127 intermediate the edges 116 and 117 of the superconductive strip is illustrated in FIGURE 8.

The various features and advantages of the invention are thought to be clear from the foregoing description. Various other features and advantages not specifically enumerated will undoubtedly occur to those versed in the art, as likewise will many variations and modifications of the embodiment of the invention illustrated, all of which may be achieved without departing from the spirit and scope of the invention as defined by the following claims:

I claim:

l. In combination:

(a) a superconductive member having dimensions whereby a normal region may be provided through said member without substantially affecting dominant current ow in said member when superconductive; l

(b) means for providing a stable normal region through sai dmember in a direction subntantially normal to the direction of current flow in said member when said member is superconductive; and

(c) means for causing said normal region to move from the periphery of said member in a direction in said member substantially normal to both the axial direction of said normal region and the direction of said dominant current flow.

2. In combination:

(a) a superconductive plate-like member having two opposed and generally flat surfaces, dominant current ow in said member being in a predetermined direction when said member is operating in the persistent mode;

(b) means for providing magnetic ux through said surfaces in a stable normal region when said member is superconductive; and

(c) means for causing said. flux to move entirely across said member and dominant current tiow in said member, said surfaces having dimensions greater than the greatest dimension of said normal region whereby dominant current flow in said member is substantially unaffected by the presence of said normal region.

3. In combination:

(a) a superconductive plate-like member having two opposed and generally fiat surfaces, dominant current ow in said member being in a predetermined direction when said member is operating in the persistent mode, said member forming part of a superconducting loop whereby inner and outer edges of said member are defined with respect to said loop;

Thus, as shown in` (b) means for providing magnetic fiux through said surfaces in a stable normal region when said member is superconductive; and

(c) means for causing said iiux to move from the outer edge of said member entirely across said member and dominant current tiow in said member, said surfaces having dimensions greater than the greatest dimension of said normal region whereby dominant current ow in said member is substantially unaffected by the presence of said normal region.

4. In combination:

(a) a superconductive plate-like member having two opposed and generally at surfaces, dominant current flow in said member being in a predetermined direction when said member is operating in the persistent mode;

(b) magnetic means for simultaneously providing a stable normal region and magnetic fiux through said surfaces when said member is superconductive, said ux passing through said normal-region; and

(c) means for actuating said magnetic means to cause said ux to move entirely across said member and dominant current ow in said member, said surfaces having dimensions greater than the greatest dimension of said normal region whereby dominant current fiow in said member is substantially unaffected by the presence of said normal region.

5. A device including:

(a) a plate of superconductive material;

(b) means for creating a stable normal region through said plate when it is superconductive whereby magnetic lines of fiux may pass through said normal region;

(c) means for providing magnetic lines of ux through said normal region; and

(d) means for moving said normal region with said ux existing therein from one point at the periphery of said plate to an opposed point located at the periphery of said plate.

6. In combination:

(a) a superconductive plate-like member comprising part of a closed superconductive circuit, said member having two opposed and generally at major surfaces, dominant current ow in said member being in a predetermined direction when said member is operating in the persistent mode;

(b) magnetic means including a field coil and an iron core having an air gap for simultaneously providing a stable normal region and magnetic fiux through said surfaces when said member is superconductive, said member being disposed in at least part of said air gap and said fiux passes through said normal region; and

(c) means for actuating said magnetic means to cause said flux to move entirely across said member and dominant current flow in said member, said major surfaces having dimensions greate than the greatest dimension of said normal region whereby dominant current iiow in said member is substantially unaffected by the presence of said region.

7. The combination as defined in claim 6 wherein said magnetic means includes a plurality of field coils, said air gap is defined in part by a plurality of spaced fingerlike projections forming an integral part of said iron core, and a field coil is disposed on each of said projections.

8. In combination:

(a) a superconductive plate-like member comprising part of a closed superconductive circuit, said member having two opposed and generally at major surfaces, dominant current fiow in said member being in a predetermined direction when said member is operated in the persistent mode;

(b) magnetic means for simultaneously providing a stable normal region and magnetic ux through said surfaces in said normal region when said member is superconductive, said magnetic means including first and second groups of current conducting coils adjacent respectively one each of said major surfaces and extending to at least opposite edges of said major surfaces, each group of said coils being coaxial about a separate axis parallel to said major surface and normal to the direction of said dominant current ow; and

(c) means for selectively supplying current to each of said coils to cause said fiux to move entirely across said member and dominant current tiow in said member, said major surfaces having dimensions greater than greatest dimensions of said normal region whereby dominant current fiow in said member is substantially unaffected by the presence of said normal region.

9. The combination as defined in claim 6 wherein said iron core is generally U-shaped and has a pair of opposed legs which define said air gap, said air gap extending to at least opposed edges of said major surfaces, said magnetic means additionally including a matrix of current carrying wires in said legs for saturating predetermined portions of said legs, and said means for actuating said magnetic means selectively supplies current to different portions of said matrix whereby the portions of said legs adjacent said major surfaces are saturated except for opposed and relatively narrow unsaturated regions which effectively move in unison from one to the other of said edges of said major surfaces as said current is selectively supplied to said different portions of said matrix.

10. The method of inducing current flow in a superconductive circuit wherein at least a part of said circuit includes a superconductive member having dimensions whereby a normal region may be provided through said member without substantially affecting dominant current ow in said member when superconductive comprising:

(a) providing a substantially stable normal region at an edge of and passing through said member when said member is superconductive;

(b) passing magnetic flux through said normal region;

(c) moving said normal region with said fiux into said member in a direction at a angle to said dominant current ow; and

(d) thereafter causing said normal region to disappear.

11. The method of inducing current ow in a superconductive circuit at least a part of which includes a plate-like superconductive member having dimensions whereby a normal region may be provided through said member without substantially affecting dominant current tiow in said member when superconductive comprising:

(a) providing a stable normal region at an edge comprising part of the periphery of said member and passing through said member when said member is superconductive;

(b) providing magnetic ux through said normal region; and

(c) moving said normal region with said flux across said member and dominant current fiow in said member.

12. The method of inducing current tiow in a closed superconductive circuit wherein said circuit includes at least in part a superconductive plate-like member comprising:

(a) creating a stable normal region through said member whereby magnetic lines of liux may pass through said normal region;

(b) providing magnetic lines of tiux through said normal region; and

(c) moving said normal region with said fiux existing therein between two points in said member, one point being located at the periphery of said member.

13. The method of inducing current ow in a closed superconductive circuit wherein said circuit includes spect to each other and located at the periphery of at least in part a thin superconductive plate-like memsaid member. ber comprising:

(a) creating a stable normal region through said mem- References Cited by the Examiner ber whereby magnetic lines of ux may pass through 5 UNITED STATES PATENTS said "omai mgm 3,094,628 6/1963 Jiu 307-885 x b rovidin ma netic lines of ux throu h said norfnal region? andg g 3,201,765 s/196s Pearl 340-1711 (c) consecutively moving said normal region with said flux existing therein from one point to another point MILTON O' HIRSHFIELD Pr'mary Examme" in said member, said points being opposed with re- I. A. SILVERMAN, Assistant Examiner.

Claims

1. IN COMBINATION: (A) A SUPERCONDUCTIVE MEMBER HAVING DIMENSIONS WHEREBY A NORMAL REGION MAY BE PROVIDED THROUGH SAID MEMBER WITHOUT SUBSTANTIALLY AFFECTING DOMINANT CURRENT FLOW IN SAID MEMBER WHEN SUPERCONDUCTIVE; (B) MEANS FOR PROVIDING A STABLE NORMAL REGION THROUGH SAID MEMBER IN A DIRECTION SUBSTANTIALLY NORMAL TO THE DIRECTION OF CURRENT FLOW IN SAID MEMBER WHEN SAID MEMBER IS SUPERCONDUCTIVE; AND (C) MEANS FOR CAUSING SAID NORMAL REGION TO MOVE FROM THE PERIPHERY OF SAID MEMBER IN A DIRECTION IN SAID MEMBER SUBSTANTIALLY NORMAL TO BOTH THE AXIAL DIRECTION OF SAID NORMAL REGION AND THE DIRECTION OF SAID DOMINANT CURRENT FLOW.