Marston, Peter G. (East Gloucester, MA)
Nolan, John J. (Randolph, MA)

05/338176

11/18/1975

03/05/1973

Download PDF 3920543       PDF help

Click for automatic bibliography generation

Magnetic Engineering Associates, Inc. (Cambridge, MA)

209/222

209/232, 210/222

B03C1/029; B03C1/02; B03C1/26

209/214,223,218,219,220,222,224,232 210/222,223 335/299 336/225-227,232

3326374	Magnetic separator with washing and scouring means	June 1967	Jones
3375925	Magnetic separator	April 1968	Carpenter
3405807	FILTER CARTRIDGE SEALING MEANS	October 1968	Burkhardt
3627678	MAGNETIC SEPARATOR AND MAGNETIC SEPARATION METHOD	September 1969	Marston
3822016	MAGNETIC SEPARATOR HAVING A PLURALITY OF INCLINED MAGNETIC SEPARATION BOXES	July 1974	Jones

Halper, Robert

Iandiorio, Joseph S.

What is claimed is

1. A moving matrix magnetic separator comprising:

2. The magnetic separator of claim 1 in which said electromagnetic coil means includes first and second electromagnetic coils each encircling one of said pole members.

3. The magnetic separator of claim 1 in which said electromagnetic coil means includes at least one electromagnetic coil including four interconnected elements.

4. The magnetic separator of claim 3 in which all of said elements are in the same plane, and all of said elements are proximate one of said pole members.

5. The magnetic separator of claim 1 in which said inlet means includes a plurality of spaced, parallel members arranged transverse to the direction of motion of said matrix member.

6. The magnetic separator of claim 1 in which said outlet means includes a plurality of spaced, parallel members arranged transverse to the direction of motion of said matrix member.

7. The magnetic separator of claim 1 in which said inlet means includes distributor head means between said first pole member and said matrix member for introducing fluid to said matrix member.

8. The magnetic separator of claim 1 in which said outlet means includes collection head means between said second pole member and said matrix member for removing fluid from said matrix member.

9. The magnetic separator of claim 1 in which said inlet and outlet means include a feed zone having a feed inlet and feed outlet, and a seal inlet arranged adjacent at least one side of said feed inlet and outlet in the direction of motion of said matrix unit for providing a hydrostatic seal about said feed inlet and outlet.

10. The magnetic separator of claim 1 in which said inlet and outlet means include a feed zone having a feed inlet and feed outlet, and a rinse zone having a rinse inlet and rinse outlet, adjacent said feed inlet and feed outlet, respectively, in the direction of motion of said matrix member.

11. A moving matrix magnetic separator comprising:

12. A moving matrix magnetic separator comprising:

13. A moving matrix magnetic separator comprising:

14. The magnetic separator of claim 13 in which said inlet means passes through one of said pole members.

15. The magnetic separator of claim 13 in which said outlet means passes through one of said pole members.

16. The magnetic separator of claim 13 in which said inlet means includes a plurality of spaced, parallel members arranged transverse to the direction of motion of said matrix member.

17. The magnetic separator of claim 13 in which said outlet means includes a plurality of spaced, parallel members arranged transverse to the direction of motion of said matrix member.

18. The magnetic separator of claim 13 in which said inlet and outlet means includes a feed zone having a feed inlet and feed outlet, and a seal inlet arranged adjacent at least one side of said feed inlet and outlet in the direction of motion of said matrix member for providing a hydrostatic seal about said feed inlet and outlet.

19. The magnetic separator of claim 13 in which said inlet and outlet means include a feed zone having a feed inlet and feed outlet, and a rinse zone having a rinse inlet and rinse outlet, adjacent said feed inlet and feed outlet, respectively, in the direction of motion of said matrix member.

20. A moving matrix magnetic separator comprising:

21. A moving matrix magnetic separator comprising:

22. A moving matrix magnetic separator comprising:

23. A moving matrix magnetic separator comprising:

24. A moving matrix magnetic separator comprising:

25. A moving matrix magnetic separator comprising:

26. A moving matrix magnetic separator comprising:

27. A moving matrix magnetic separator comprising:

28. A moving matrix magnetic separator comprising:

FIELD OF INVENTION

This invention relates to a moving matrix magnetic separator, and more particularly to such a separator in which the field and flow are parallel and the source of magnetic field is located in the vicinity of the poles and gap.

BACKGROUND OF INVENTION

The process capacity of magnetic separators may be increased by the use of a moving matrix technique wherein the separation process may operate continuously as the matrix moves. In one such technique the feed flows vertically between a pair of horizontally spaced poles through which passes a portion of an annular matrix rotating in a horizontal plane. The flow, field, and matrix motion are each in mutually perpendicular directions. Since increasing cross-sectional area by increasing the dimension measured along the direction of motion of the matrix is not practical, any increase process capacity also increases the distance or gap between the poles; and, conversely, any decrease in gap for the purpose of increasing the field intensity also results in a decrease in flow cross-sectional area.

In another approach, a matrix in the form of an annulus is rotated about a horizontal axis. The matrix moves through the working magnetic field volume or gap at the low point of its travel. Feed is introduced at the inner diameter of the annulus at that point and pours through the matrix to exit at the outer diameter of the annulus. In such an arrangement, it is difficult to provide a uniform magnetic field across the working gap in the area where the feed is submitted to the matrix. In fact, such uniformity could only be approached if the radius of curvatures of the inner and outer circumferences of the annulus were very large and if, in addition, the magnetic poles were curved on either side of the annulus to match the curvature of the matrix. Such requirements would make an unduly large and expensive machine and even then the uniformity of the field provided would be less than optimum.

Typically, in such moving matrix machines, flow of the fluid feed is controlled by the force of gravity. If then the flow characteristics in the matrix along the direction of motion are to be uniform through a matrix which is rotating about a horizontal axis, the radius of curvature of the inner and outer circumferences of the annulus containing the matrix would once again have to be very large and the size would increase with increase in the process capacity and flow cross-sectional area of the device: uniformity of field and flow may be required for many applications to obtain an acceptable separation. In addition, if a rinse region is required, adjacent to the feed region within the working volume, an annular matrix of even larger radius would have to be used to preserve any sort of uniformity in flow through the matrix. Further, the use of gravity as the primary force for moving the feed and the rinse through the matrix limits the usable area of such a matrix to the bottom portion and perhaps the top portion.

Another disadvantage of the prior art is that typically one or more of the electric magnetic coils or other magnetic field generating means used to produce a magnetic field in the working volume or gap are located on a return frame remote from the working gap. The field in the working volume or gap of an electro-magnet is the sum of the direct energizing coil contribution (Biot-Savart effect) and the integrated dipole contribution, i.e., magnetized iron contribution of the proximate magnetized ferromagnetic return frame. Thus, an electromagnetic coil or any source of magneto-motive force (MMF) which is located remote from the working gap has some elements with direct contributions which detract, and some elements with direct contributions which add, to the magnetic field produced in the working gap. That is, while in some arrangements most of the coil elements provide a positive direct field contribution to the magnetic field in the working volume, at least some of those elements provide a negative direct contribution which subtracts from the positive direct contribution leaving a net direct contribution to the field in the working volume which is less than the total field which can effectively be generated by that coil. In addition, the direct field contribution of any coil element is increased by reducing the distance between that element and the working volume. Reduction of this distance also tends to reduce the cost of a source of a given MMF.

SUMMARY OF INVENTION

It is therefore an object of this invention to provide a moving matrix magnetic separator in which the process capacity can be optimized by varying the flow cross-sectional area of the matrix and the field intensity in the working magnetic field volume can be increased by reducing the gap between the poles each independently of the other.

It is a further object of this invention to provide a moving matrix magnetic separator in which the magnetic field and the flow of feed through the matrix are parellel.

It is a further object of this invention to provide a moving matrix magnetic separator in which the magnetic field and the flow of feed through the matrix are uniform.

It is a further object of this invention to provide a moving matrix magnetic separator in which there is maximum utilization of the source of the magnetic field and all electromagnetic elements of that source produce a positive direct contribution to the magnetic field in the working gap.

This invention results from the realization that in a moving matrix magnetic separator processing capacity is a function of the flow cross-sectional area of the matrix and that the intensity of a magnetic field between two poles of a separator is more a function of the distance between the poles rather than their cross-sectional area for a given MMF, and, that by making the direction of the field and flow parallel, the field intensity can be increased by decreasing the distance between the poles and the process capacity can be increased by increasing the flow cross-sectional area of the matrix each without detracting from the improvement gained by the other and the further realization that the field intensity provided by a given source of magneto-motive force in the working magnetic field volume between the poles can be significantly improved by locating the source of the magnetomotive force at or very close to the working field volume in such a way that all electromagnetic elements of the source produce a positive direct contribution to the magnetic field in the working gap and that further improvemments can be effected by optimizing the uniformity of the magnetic field in the working gap or volume and the uniformity of the flow in that magnetic field in the working volume.

These realizations resulted in the construction of a moving matrix magnetic separator which includes a magnetic pole unit having a first ferromagnetic pole member, a second ferromagnetic pole member spaced from the first and a working magnetic field volume formed by the space between the first and second pole members. Electromagnetic coil means, in which each element has a positive direct contribution to the magnetic field in the working volume, encircles the magnetic pole unit proximate the working magnetic field volume, and produces a magnetic field extending in the first direction through the working magnetic field volume between the pole members. A moveable matrix member moves through the working magnetic field volume between the first and second pole members in a second direction transverse to the first direction. Fluid provided by inlet means proximate one of the pole members flows through the matrix member and the working magnetic field volume in the first direction and is removed by outlet means proximate the other of the pole members. In one preferred embodiment the matrix member is generally horizontal and rotates about a vertical axis and the magnetic field and the flow of feed through the matrix are vertical and are of optimum uniformity.

DISCLOSURE OF PREFERRED EMBODIMENT

Other objects, features and advantages will occur from the following description of a preferred embodiment and the accompanying drawings, in which:

FIG. 1 is a schematic, axonometric view of a moving matrix magnetic separator according to this invention;

FIG. 2 is a diagrammatic, plan view of the separator shown in FIG. 1;

FIG. 3 is an enlarged, diagrammatic sectional side view of a feed station and flush station as shown in FIG. 1;

FIG. 4 is a diagrammatic end view of the feed station of FIG. 3;

FIG. 5 is a diagrammatic side view of the feed station shown in FIG. 3 showing in more detail the position of the coils along the side of the station parallel to the direction of matrix motion;

FIG. 6 is a diagrammatic, axonometric view of a support for the matrix of FIG. 1;

FIG. 7 is a side view of a feed station similar to that shown in FIG. 5 wherein the electromagnetic coils are constructed in one plane without upturned ends and encircle the pole units proximate the respective poles and not the working volume between;

FIG. 8 is an end view of the feed station shown in FIG. 7;

FIG. 9 is a sectional, schematic view taken along lines 9--9 of FIG. 8;

FIG. 10 is a schematic sectional view of alternative inlet and outlet means which do not pass fluid through the poles;

FIG. 11 is a schematic diagram showing a pole unit and one form of coil that may be associated with either pole member or both of them;

FIG. 12 is a schematic plan view showing four elements of a rectangular coil;

FIG. 13 is a schematic plan showing four elements of a circular coil;

FIG. 14 is a schematic sectonal diagram of a pole unit showing an alternate form of coil;

FIG. 15 is a schematic sectional view showing another alternate coil form;

FIG. 16 is an axonometric diagram showing still another coil form;

FIG. 17 is a schematic flow chart of one interconnection system which may be used with a separator of this invention;

FIG. 18 is a schematic flow chart of an alternative interconnection system which may be used with a separator of this invention;

FIG. 19 is an axonometric view of an alternative moving matrix magnetic separator according to this invention; and

FIG. 20 illustrates an alternative pole unit structure using permanent magnets for the source of the magnetic field.

In one preferred embodiment, FIG. 1, a moving matrix magnetic separator 10 includes a horizontal matrix member 12 rotatable about its center in the direction of arrow 14 by drive means not shown. Spaced about the path of matrix member 12 are a plurality of feed stations 16, 18, 20 and 22, FIG. 2, and a plurality of flush stations 24, 26, 28 and 30.

Each feed station exemplified by feed station 18, FIG. 1, includes a feed inlet 32 and a rinse inlet 34 which are fed by feed pipe 36 and rinse pipe 38, respectively, as well as a feed outlet 33 and rinse outlet 35, FIG. 3, which have corresponding feed outlet pipe 40 and rinse outlet pipe 42. Within housing 44, FIG. 1, is a split coil or a pair of coils 46 and 48 whose ends 50, 52 and 54, 56 are bent backwardly to provide apertures 120, 122, FIG. 3, at each end of housing 44 to permit the movement of matrix member 12 therethrough. Each flush station as exemplified by flush station 24, FIG. 1, includes a housing 58, FIG. 3, a flush inlet 60 connected to flush inlet pipe 62 and a flush outlet 61, connected to a flush outlet pipe 64. Raw feed is supplied to the feed inlet pipes which are connected to the feed reservoir 66, FIG. 1. Feed reservoir 66 may receive the raw feed from external sources through inlet pipe 68 or through inlet pipes 70 and 72 from the feed, rinse and flush outlets of various stations of the machine depending upon the system designed. Similarly, rinse inlets and flush inlets may receive clean water, or outputs from previous or successive stations or any other fluid or combination of fluis through pipe 74 or other pipes in accordance with the system design. Two detailed flow charts are shown in FIGS. 17 and 18, infra, to illustrate two sepcific system designs which may be used with the magnetic separator according to this invention.

Matrix member 12, FIG. 6, may be formed with an inner peripheral member 80 connected to an outer peripheral member 82 by means of interconnection elements 84 between which, in spaces 86, is located the matrix medium such as steel wool, steel balls, tacks or the like, here omitted for clarity. In a machine such as machine 10, FIG. 1, where the matrix member 12 is an annulus, members 80 and 82 are circular rings and the matrix member is constructed as a single continuous annulus.

Each feed station as exemplified by feed station 18, FIG. 3, includes a pole unit including a first ferromagnetic pole member 90 and a second ferromagnetic pole member 92 aligned with the first pole member 90 and spaced from the first pole 90 and a working magnetic field volume or gap 94 formed between pole members 90 and 92. Located in each pole member 90 and 92 are inlet means 95 and outlet means 96 for permitting the introduction and removal of feed or rinse or any other fluid to the portion of the matrix member 12 presently within the working volume 94. Inlet means 95 is shown specifically as a plurality of ferromagnetic members or plates 98 spaced from each other in the direction of motion of matrix member 12 and extending transversely across the path of matrix member 12. Outlet means 96 is similarly formed from ferromagnetic members or plates 100 similarly spaced from each other in the direction of motion of matrix member 12 and transverse to the direction of motion of matrix member 12. Plates 98 and 100 are arranged to direct the flow of the fluid in the matrix so that it is parallel to the magnetic field extending in gap between poles 90 and 92. Following feed station 18 in sequence is flush station 24 in which the housing 58 may include, FIG. 3, simply a box in which the flush liquid entering through inlet 60 may be passing through the portion of the matrix member then present in housing 58.

Electromagnetic coil 46 has two elements 50 and 52, FIGS. 3, 4 and 5, which are transverse to the direction of motion of matrix member 12 and two elements 102 and 104 which extend along the direction of motion of matrix member 12. Similarly, electromagnetic coil 48 has two elements 54 and 56 which are transverse to the direction of motion of matrix member 12 and two elements 106 and 108 which extend in the direction of motion of matrix member 12. The elements 102, 104 of coil 46, and 106, 108 of coil 48, which extend in the direction of motion of matrix member 12, abut each other and adjoin the working volume or gap 94. The other elements of coils 46 and 48, i.e., elements 50, 52, 54 and 56, are bent out of the way of working volume 94 to form apertures 120, 122 so that matrix member 12 can pass through. Thus, elements 50, 52, 54 and 56 of coils 46 and 48 adjoin pole members 90 and 92, respectively, and not working volume 94. Because of the position of coils 46 and 48, each element of them produces a positive direct contribution to the magnetic field in the volume 94. The direction of the magnetic field is shown by arrow 110, the direction of the fluid flow by arrow 112, and the direction of motion of matrix member 12 by arrow 114 in FIGS. 3, 4 and 5. The directions of the field and flow are parallel to one another and the direction of the motion of matrix member 12 is transverse to their direction. Seal inlet 116 and seal outlet 118 may be provided adjacent feed inlet 32 and feed outlet 33, respectively, to provide a hydrostatic seal which prevents feed entering through inlet 32 and leaving through outlet 33 from leaking out of station 18. Typically, fluid such as water is introduced through inlet 116 at equal or greater pressure than the feed is introduced at inlet 32. This prevents the feed from moving laterally in working volume 94, such as in a portion of matrix member 12, so that the feed is maintained in a feed zone 89 commensurate with feed inlet 32 and feed outlet 33, and the water or other fluid is maintained in the sealing zone 91 commensurate with seal inlet 116 and seal outlet 118. Thus, any leakage which might occur would only be of the water or other sealing fluid and would not interfere adversely with the efficiency of the process. A similar seal may be provided adjacent the rinse zone on the down stream end of the rinse zone. Uniformity of flow through the matrix 12 is contributed to by a number of factors: the use of plates 98 and 100, the geometry of inlets 32, 34 and outlets 33, 35, the sealing arrangements and the uniformity of the matrix shape where it passes through the stations.

The use of the concepts of parallel field and flow directions and uniformity refers to average conditions neglecting local perturbations such as caused by matrix elements.

Although each feed station is shown as including a feed zone and a rinse zone 93, this is not a necessary limitation of the invention, as a station may include the feed zone without the additional rinse zone. Without the use of a rinse zone, the absence of rinse inlet 34 and rinse outlet 35 may require that a second sealing zone with a second sealng inlet and sealing outlet be used to prevent leakage of the feed in that area. Although the coils in FIGS. 1 through 5 have been shown as having a pair of opposite ends bent out of the primary plane of the coil in order to provide apertures 120 and 122 at each end of working volume 94, this is not a limitation; for as shown in FIGS. 7, 8 and 9, where like parts have been given like numbers primed, coils 46' and 48' may be made entirely in one plane and disposed encircling pole members 90' and 92', respectively, so that all four elements lie in the same plane and there is not a pair of elements which dip down adjacent to working volume 94' along the direction of motion of matrix member 12. The location of a coil at the pole unit proximate the working gap enables that coil to produce the maximum effective field in the gap because each element of the coil is providing a positive direct field contribution to the magnetic field in the gap. When the coil or coils are arranged with their central axes generally parallel to that of the pole members and their median planes generally parallel to the median plane of the working gap, they produce a field which traverses the gap from pole member to pole member; each element will typically have such a positive contribution in accordance with the Biot-Savart principle.

Although in the description thus far all of the inlet and outlet means are located in the pole themselves, this is not a necessary limitation of the invention. For example, in FIG. 10, where like parts have been given like numbers double primed with respect to previous figures, there is shown an inlet means or distributor means 130 divided into a feed distributor head 132 and a rinse distributor head 134. Distributor heads 132 and 134 receive their respective inputs through inlet pipes 36" and 38" and distribute it into matrix 12 by means of ports 136. The outlet means or collection means 138 may include two collection heads 140, 142 which collect fluids in the feed zone and rinse zone, respectively, from matrix 12 and dispose of it through outlet pipes 33" and 35". Collection heads 140 and 142 may be in the nature of open, shallow pans.

The structure of a pole unit and of the electromagnetic coil means and their relationship is discussed in more detail in FIGS. 11 through 16 where like parts have been given like numbers accompanied by lower case letters. Pole units 91, FIG. 11, includes a pole member 90a and pole member 92a spaced from pole member 90a and aligned with it along the axis A of the pole members, and a working magnetic field volume or gap 94a formed by the space between the pole members. The median plane G of gap 94a is transverse to axis A and is typically perpendicular to it. The electromagnetic coil means may include a coil 46a proximate pole member 90a or a coil 48a proximate pole member 92a or one proximate each pole member; a coil may enricle the pole unit anywhere along a pole member or the working gap.

For convenience in describing the geometry of a coil and its position relative to the pole unit, each coil is thought of as having four interconnected elements, L ₁ , L ₂ , L ₃ , L ₄ , FIG. 12. This is so regardless of the shape of the coil, e.g., in FIG. 13 the circular coil 46c also has four elements L ₁ , L ₂ , L ₃ , L ₄ . The number of elements used to describe a coil is typically a function of the number of sides of the associated pole member, e.g., if the pole had five sides it would be more convenient to refer to the coil as having five elements.

In FIG. 11, all four elements of coil 46a are in the same plane P and proximate the same pole member 90a, but this is not a necessary limitation. For example, FIG. 14, element L ₁ may be in one plane, P ₁ , proximate one pole member 90a, element L ₂ in a second plane, P ₂ , proximate the other pole member 92a and elements L ₃ and L ₄ in a third plane, P ₃ , proximate working gap 94a. Or, FIG. 15, element L ₁ may be in a plane, P ₄ , proximate pole member 90a, and element L ₂ in plane P ₅ proximate pole member 92a and elements L ₃ and L ₄ proximate working gap 94a and in plane P ₆ which intersects planes P ₄ and P ₅ . Both elements L ₁ and L ₂ may be in the same first plane proximate pole unit 90a, FIG. 16, and elements L ₃ , L ₄ in the same second plane proximate working gap 94a. In FIG. 16, a second coil which is a mirror image of the one shown, could be used alone or together with the one shown.

In a cycle of operation, matrix member 12, FIG. 1, rotates in a first direction through the working volume 94 wherein it encounters a magnetic field transverse to its direction of motion. In that magnetic field in the feed zone 89 of working volume 94 matrix member 12 is first subjected to a flow of feed in the same direction as the magnetic field, and is next subjected to a rinsing fluid, which may be clear water, in the adjacent rinse zone 93 so that loose particles not adhering to the magnetic matrix while still in the magnetic field are rinsed out of the matrix. Following this, after the matrix has cleared the magnetic field of the feed stations, the matrix enters the flush station which may be simply a hollow housing having no magnetic field wherein a flushing fluid such as clear water may be used to flush away the magnetic particles which had previously adhered to the matrix because of the presence of the magnetic field. The flush station may include a shell 25, FIG. 3, of magnetic material to shield the interior from the neighboring magnetic fields. The purpose of a magnetic separator is to separate more-magnetic from less-magnetic particles. The less magnetic particles leave the separator via the feed outlet 33. The more magnetic particles leave via the flush outlet 61. The material leaving via the rinse outlet 35 might be immediately mixed with the material from the adjacent feed outlet, or might be treated as a middling fraction to undergo further treatment.

The interconnection of the various feed, rinse and flush inlets and outlets of machine 10 enable a large number of different flow schematics to be implemented on the machine. For example, in FIG. 17, station 22 receives raw feed at its feed inlet, and station 20 receives at its feed inlet the output of the previous flush station 28, while feed station 18 receives at its feed input the output of the previous flush station 26; feed station 16 and flush station 30 are unused. All of the rinse and flush inputs at stations 22, 28, 20, 26, 18 and 24 are clear water as their rinsing and flushing fluid. The output of the final flush station in the series, flush station 24, is considered the product and the output of the feed and rinse zones of each of feed stations 22, 20 and 18 are considered the tailings. A somewhat more elaborate example is shown in FIG. 18 wherein the feed input of feed station 22 comes from feed tank 66, see FIG. 1; the feed input to feed station 20 is derived from the feed output of feed station 22; the feed input to feed station 18 is derived from the flush output from the preceding flush station 26, and the feed input to feed station 16 is derived from the flush output of the previous flush station 24. The rinse input to feed station 22 and the flush input to flush station 28 is clear water, while the output of flush station 28 is considered a product, and the rinse output from feed station 22 is recycled back to feed tank 66. The rinse input to feed station 20 is derived from the rinse output of feed station 18; both the feed output and the rinse output of feed station 20 are considered tailings. The rinse output of feed station 16 is submitted to the input of flush station 26 and the output of flush station 26 supplies the feed input to feed station 18. The feed output of feed station 18 is considered tailings. The rinse input to feed station 18 and the flush input to flush station 24 are both clear water as are the rinse input to feed station 16 and the flush input to flush station 30. The feed output of feed station 16 is considered tailings while the rinse output of feed station 16 supplies the input to flush station 26. The flush output of flush station 30 is considered a second product. In other situations, the more magnetic particles are the tailings and the less magnetic particles are the product, or all outputs may be considered products.

An alternative construction for a moving matrix magnetic separator which does not require a circular or continuous matrix member is shown in FIG. 19. As illustrated, it includes two feed stations 220 and 222 and two flush stations 224 and 226 which service a matrix member 228 which includes a plurality of matrix segments 230 which are submitted in series to the feed and flush stations by means of a conveyor.

Thus far, the sources of magnetic field in each of the embodiments illustrated has been an electromagnetic coil or coils. However, that is not a necessary limitation of the invention for the source of the magnetic field may as well be one or more permanent magnets as shown in FIG. 20. In FIG. 20, there is a magnetic frame 248 having a pole unit 249 which includes two pole members 250 and 252 which are also permanent magnets. A working magnetic field volume 258 is produced between poles 250, 252 for receiving matrix member 260.

Other embodiments will occur to those skilled in the art and are within the following claims: