CA1301884C - Magnetic position sensor - Google Patents

Abstract

ABSTRACT OF THE DISCLOSURE

A position sensor and method for detecting and providing an indication of the position of a movable member comprising a ferromagnetic element oriented substantially parallel to the path of travel of the movable member, the element including a segment thereof having spaced-apart ends defining the positional limits between which movement of the movable member is to be monitored, the segment comprising a pair of contiguous, oppositely polarized, remanently magnetized regions defining an intersection along the length of the segment; a magnet positioned proximate the intersection and mounted in known spatial relationship to the movable member for movement therewith, and, simultaneously along the ferromagnetic element, the magnet providing a magnetic field presenting a constant polarity to the element and having a strength sufficient to locally polarize the element, whereby movement of the magnet means along the element alters the length of the respective regions; and, means for sensing the position of the intersection without altering that position and for providing an indication of the position of the movable member based thereon.

Description

13~8~

IÇ ~Q~L~IQ~ QR

~ch~ ld The present invention relate6 generally to position sensors and to methods employing ~uch 6ensors for detecting and providing an indication of the position of a movable member and, more particularly, to position sensor6 and methods employing magnetic means for non-contact po~ition 6en~ing.

~CI9~QU~ t The ability to obtain and indicate, in a useful manner, highly accurate information regarding the position of a movable member is very important in controlling tools, equipment and apparatus, monitoring processes, determining liquid level, and for many other purposes. Typically, due to the nature of the tools, equipment, apparatus or process, the environment in which they operate and the need for continuous position information, the determination of position i~ usually accompli~hed by a position sensor in~talled proximately to the member whose position i~ being sensed but out of contact therewith in order to avoid interfering with or influencing in any way the movement of the member.
Moreover, the po~ition ~ensed will, mo~t usually, be transmitted to a location remote from the movable , ~

:~3~

member where it may be observed, recorded or used to control the operation of equipment or processes.
One well known way of detecting the position of a movable member without actually establishing physical contact therewith is by the use of magnetic means ~ounted, directly or indirectly, on the movable member for magnetically interacting with other means to produce a signal indicative of the position of the movable member. For example, in U.S. Patent No. 4,071,818 - Rrisst there is disclosed a method and apparatus in which a magnetic field generator is ~ounted on the movable member in order to be movable therewith and an elon~ated ferromagnetic element is positioned adjacent and parallel to the path of movement of the member. The generated magnetic field produces a change in the Youngls modulus of elasticity in an adjacent region of the ferromagnetic element such that sonic strain pulses launched along the ferromagnetic element will be partly reflected from this region of Young's modulus discontinuity. The time required for a sonic ~train pulse to travel from a given point on the ferromagnetic element to the region of Young's modulus discontinuity, be partially reflected therefrom and return to a known detection point provides a measure of the position of the field generator along the ferromagnetic element. Inasmuch as the positional relationship between the field generator and any point of interest on the movable member is fixed and known, when the position of the field generator is known the corresponding position of any point of interest on the movable member is al~o known.
The shortcoming of the ~risst apparatus is that it requires the measurement of very short times. When the length of path travelled by the sonic strain pulse is relatively long, the apparatus may be acceptably accurate for certain purposes. However, as the length of path travelled becomes short, the accuracy of the ~3~18~4 ~risst apparatus decreases rapidly. Moreover, the response speed of the Kris~t apparatus i~ relatively slow since thi~ apparatus, like all such apparatus utilizing sonic pul~es, i8 limited by the relatively slow travel of reflected sonic pulse~ and the rate of decay of echoes. Moreover, the accuracy and reliability of the Rrisst apparatus is adversely affected by the fact that the Young's modulus of the element varies with stress and magnetization.
Therefore, it varies over the length of the member at least because the stress is greater along the upper portions of the element than along its lo~er portions.
In addition, it may vary due to differing magnetization along its length as a result of treatments to which those portions of the member may have been subjected at various stages during its manufacture.
In U.S. Patent No. 4,194,397 - Yasuda there is shown a liquid level indicator in which a bypass flow column is affixed to the side of a tank and a magnet-containing float i~ disposed within the column. A
housing affixed to the column but remote from the tank contains a plurality of vertically spaced-apart magnet rotors which have an initial magnetic orientation in a first position wherein the south pole i~ on the bottom and the north pole is on the top. As the liquid level in the column rises, the north pole on the float causes the rotors to rotate about their horizontal axes to a second position wherein the north pole is on the bottom and the south pole is on the top. Conveniently, the rotors are painted in such a manner that different colors are displayed when the rotors are in the first and second positions. In this way, the magnetic orientation within the housing changes as the float moves up and down, and such changes are visually apparent from the displayed color change. The Yasuda method and apparatus is of limited usefulness and applicability. A visual indication of position by ~L3~

color changa is impractical for a wide variety of poten-tial sensor uses and th~ resolution and accuracy of such an indication is severely limited. Moreover, the Yasuda apparatus does not permit the remote indication of posi-tion or the ready application of the position information to the control of tools, equipment or processes.
It is therefore, apparent that despite the many advances in the use of magnetic means in the detection and indication of the position of a movable member, there still exists a need for a magnetic position sensor which is significantly more economical than previous sensors, which is extremely accurate and reliable and which is readily adaptable to sensing the position of a member, irrespective of its configuration or the environment in which it operates.
Disclosure of the _nvention In accordance with an embodiment of the present invention there is provided a position sensor for detec-ting and providing an indication of the position of a movable member comprising: a ferromagnetic element oriented substantially parallel to the path of travel of the movable member, the element including a segment thereof having first and second spaced-apart ends defin-ing the positional limits between which movement of the movable member is to be monitored, the distance along the element between the ends defining the length L of the segment, the segment comprising a pair of contiguous, oppositely polarized, remanently magnetized regions, each of the regions being uniformly magnetized along its length, the first region including the first end and having a length X wherein 0 < X < L and the second region including the second end having a length L-X, the contig-uous regions defining an intersection along the length of the segment: magnet means positioned proximate the inter-section; one of the magnet means and element mounted in association with and in known spatial relationship to the movable member for movement therewith and relative to and ~3t318~34 along the other of the magnet means and element at a substantially constant distance from one another, the magnet means providing a magnetic field presenting a constant polarity to the element and having a strength sufficient to locally polarize the element, whereby the relative movement between the magnet means and the element alters the length of the respective regions:
means associated with the element for sensing the position of the intersection without altering that position and for providing an indication of the position of the movable member based thereon.
In accordance with another embodiment of the present invention there is provided a method for detecting and providing an indication of the position of a movable member comprising the steps of: orienting a ferromagnetic element substantially parallel to the path of travel of the movable member, the element including a segment thereof having first and second spaced-apart ends defin-ing the positional limits between which movement of the movable member is to be monitored, the distance along the element between the ends defining the length L of the segment, the segment comprising a pair of contiguous, oppositely polarized remanently magnetized regions, each of the regions being uniformly magnetized along its length, the first region including the first end and having a length X wherein 0 5 X < L and the second region including the second end having a length L-X, the contig-uous regions defining an intersection along the length of the segment; mounting magnet means in association with and in known spatial relationship to the movable member for movement therewith, and, simultaneously, along the ferromagnetic element at a substantially constant dis-tance therefrom, the magnet means being positioned proxi-mate the intersection and providing a magnetic field presenting a constant polarity to the element and having a strength sufficient to locally polarize the element, whereby movement of the magnet means along the element alters the length of the respective regions; sensing the ,, .

.13v~
- 5a-position of the intersection without altering that posi-tion and providing an indication of the position of the movable member based thereon.
In accordance with a further embodiment of the present invention there is provided a method of making a non-contact position sensor for detecting the position of a movable member comprising the steps of: providing a ferromagnetic element adapted to be oriented substan-tially parallel to the path of travel of the movable member, the element including a segment thereof having first and second spaced-apart ends defining the posi-tional limits between which the position of the movable member is to be monitored, the distance along the element between the ends defining the length L of the segment;
positioning a magnet means proximate the element for providing a magnetic field presenting a constant polarity to the element and having a strength sufficient to loc-ally polarize the element, whereby relative movement between the magnet means and the element alters the polarity of the portions of the element proximate the magnetic means; uniformly magnetizing the segment in a first direction along its length L by moving one of the magnet means and the element relative to the other in a single direction along the entire length of the segment between its ends: and positioning the magnet means along ths segment while maintaining the magnet means proximate the element by moving one of the magnet means and the element relative to the other in the opposite direction a distance X wherein O < X < L, the movement reversing the polarity in the portions of the element along which it moves for defining within the segment a pair of contig-uous, oppositely polarized, remanently magnetized regions, each of the regions being uniformly magnetized along its length, the first region having a length X and the second region having a length L-X, the contiguous regions defining an intersection therebetween at the position of the magnet means.

~3~
- 5b-Brief Description of the Drawings The invention will be better understood from the following description taken in conjunction with the accompanying drawings in whicho Figure 1 is a schematic representation of one embodiment of the magnetic position sensor of the present invention.

~L3~18~3~

Figure 2 is a ~chematic repre~entation of another e~bodiment of the magnetic po~ition sensor of the present invention.
Figure 3 is a graphical representation of the major and selected minor hysteresi6 loops of a ferromagnetic material useful in the sensor of the present invention.
Figure 4 i~ a circuit diagram showing one form of circuitry useful in sensing position in aocordance with the present invention.
Figure 5 i8 a graphical representation of the relationship between magnet position and output signal for the circuitry illustrated in Figure 4.
Figure 6 is a ~ircuit diagram showing another form of circuitry useful in sensing position in accordance with the present invention.
Figure 7 is a schematic representation of another e~bodiment of the magnetic position sensor of the present invention.
Figure 8 is a series of graphical representations of the relationship between magnet position and magnetic field for representative tilt angles of the magnet in accordance with the present invention.

MQ~ ~QL Carrying Out ~h~ Invention In accordance with the present invention there is provided a method and apparatus for detecting and providing an indication of the position of a movable member in which a ferromagnetic element is positioned substantially parallel to the path of travel of the movable member and a magnet means is mounted in association with and in fixed and known spatial relation to the movable member for movement therewith and, simultaneously, along the length of the ferromagnetic element. The position of the magnet _ 7 _ means defines a pair of contiguoust oppositely polarized, remanently magnetized regions in the ferromagnetic element. ~y appreciating that the two regions are contiguous and intersect along the length of the element; that the magnet means i~ always positioned at and defines the position of the inter~ection; that each of the two regions constituting the ferromagnetic element has a different incremental permeability; and, that the effective overall permeability of the overall length of the ferromagnetic element is dependent upon the position of the magnet means and, hence, of the movable member;
the position of the magnet means at the intersection of the regions can be determined. When the position of the magnet means is known the corresponding position of the movable member can readily be determined. It will be appreciated that in another form of the invention, the same result can be achieved by mounting the ferromagnetic element in association with and in known spatial relationship to the movable member for movement therewith and relative to the magnet means at a ~ubstantially constant di~tance from the magnet means.
Referring to the drawings to better understand the invention, and particularly to Figure 1, a movable member the position of which i6 to be monitored is shown schematically at 10. Member 10, as illustrated, is capable of linear movement in a horizontal direction as indicated by arrows P. Mounted on one ena 12 of member 10, in such a manner as to be movable with m ember 10, is magnet means 14, functioning as a localized magnetic field source. It will be appreciated that magnet means 14 need not be affixed at end 12 of member 10, but may be located at any point along its length. It may even be indirectly mounted to member 10, as by mounting it on support means which is itself m ounted on member 10. The important ~3~ 84 consideration is that the spatial relation~hip between member 10 and magnet means 14 is known at all times for all positions of the movable member. In the illustrated embodiment, as member 10 move~ alonq its path of travel (~hown by arrows P~, magnet means 10 moves along a corresponding path of travel (shown by arrows R).
A ferromagnetic element 16, which i~ desirably in the form of an elongated rod, tube, ribbon, strip, wire or thin f ilm on a supporting ~ubstrate, is positioned substantially parallel to the paths of travel P,R of member 10 and magnet means 14 ~uch that magnet means 14 moves along element 16 as the movable member moves along its path of travel. In its movement along element 16, magnet means 14 traverses a segment lû of element 16 def ined by segment ends 20 and 22, representing the limits between which movement of member 10 will be monitored. The length of segment 18 i~ denoted by L.
Initially, segment 18 is magnetically preconditioned by uniformly magnetizing (polarizing) the segment in a first axial direction along its length "Ln. This can conveniently be accomplished, for example, by positioning magnet means 14 at segment end 20 and moving it from left to right in Figure 1, toward segment end 22, to align the magnetization within the ~egment in a single direction. When magnet means 14 reaches segment end 22, all magnetization within the segment, random or otherwise, has been realigned and the segment has been polarized. Subsequently, it may be convenient to move magnet means 14 back to position 24, i.e., toward segment end 20. Such movement has the effect of reversing the polarization in the portion of the segment from 22 to 24, i.e., magnetizing portion 22 to 24 in a second and opposite direction. With magnet means 14 situated at 24, two oppositely polarized and remanently magnetized regions, A and B, are defined ~3(~88~
g along the length L of the segment 18. Region A extends from 20 to 24 and ha~ a length X. Region B extend6 from 24 to 22 and has a length L-X. The length of region A is either less tban or equal to the length of segment 18, i.e., X~L. ~t the extremes, magnet means 14 may be moved to a position corresponding to end 20 or 22. In the position corresponding to end 22, X-L
and L-X=O, i.e., the entire segment 18 corresponds to region A. In the position corre~ponding to end 20, X=O
and L-X=L, i.e., the entire segment 18 corre~ponds to region B. The intersection of regions A and B, which are contiguous, is at 24, which is always the position of magnet means 14. It will be appreciated that any further movement of magnet means 14 to the left or right serve6 to alter the position of the intersection.
~owever, irrespective of the movement of magnet means 14 between segment ends 20 and 22 inclusive, ~egment 18 remains polarized into two regions, region A
of length X which is magnetized in the first direction and region B of length L-X which is magnetized in the ~econd and opposite direction. All that changes with the movement of magnet means 14 i8 the relative length of each region. No matter whPre magnet means 14 is located along segment 18 it i5 always positioned at and defines intersection 24. Therefore, if the location of intersection 24 can at all times be determined then the position of magnet means 14 is known and the corresponding position of any point of interest on ~ovable member 10 can be determined therefrom. The location of intersection 24 is determinable by sensing a characteristic of element 16 which varies with changes in the remanent magnetization of the regions, e.g., permeability, speed of sound, resistance.
Magnet means 14 is a localized field source. As hereinbefore described its purpose is to initially ~agnetize or polarize, or to reverse the polarization in an already polarized portion of, ferromagnetic 13V~8~

element 16. ~o accomplish this, magnet means 14 desirably provides a magnetic field having a constant polarity and a strength sufficient to polarize, i.e., either locally polarize or reverse the polarization of, element 16. Inasmuch as the position of magnet ~eans 14 is indicative of the position of the movable member and since determining the position of magnet means 14 requires mean~ for highly accurately determining the position of intersection 24 it is important that the magnetic field emanating from magnet means 14 influence the polarization in as local a manner as possible. It is undesirable, for example, from the standpoint of obtaining highly accurate position information, to utilize a magnet means 14 having a magnetic field which is sufficiently strong at a substantial distance from its ~ource that it will influence the polarization of element 16 at a substantial distance from the actual position of the magnet means. On the other hand, it is particularly desirable to utilize magnet means 14 having a magnetic field which is only strong enough to influence polarization of element 16 immediately adjacent the magnet means and which has a field gradient characteristic which causes the field to drop off in intensity ve~ rapidly at even small distances from the actual position of the magnet means. Magnet means which have these desirable characteristics are advantageously positioned closely adjacent element 16 in order to minimize the necessary field strength and to localize the effect of the field. As a minimum, the field strength at element 16 must exceed the coercive force of the element in order to reverse the polarization thereof. More practically, at least where bias fields are employed, the field strength at element 16 should be at least twice the coercive force of the element.

13V~8~4 In addition, it should be apparent that the magnet means must provide a constant polarity to element 16.
In other words, only one pole of the magnet means can be permitted to influence the polarization of the element. Thu8, if the magnet means compri~es, as in a preferred embodiment, a conventional permanent magnet, such a6 an elongated bar magnet of any convenient cross-section, typically circular, it is de~irably positioned with one end or pole thereof in clo6e proximity to element 16 and with the other end or pole remote from the element. If desired, for example to increase the field gradient and to narrow the transition range between the two regions A and B, the pole adjacent the element may be conical or chisel-shaped. Moreover, the magnet means need not be a single simple bar magnet shape. Rather, it may comprise two or more magnets symmetrically distributed around element 16 in a plane normal to its axis, with like poles of each magnet toward element 16, or may be ~-shaped or C-shaped, or even configured as a fully closed structure to reduce the stray field.
Alternatively, in some applications the magnet means may advantageously be an electromagnet.
In addition, the magnet means may be specially configured or specially magnetized for a particular purpose. For example, a magnet means configured as a short, hollow cylinder which has been radially magnetized in such a manner that the inner diameter surface is one pole while the outer diameter surface is another pole has been found to be extremely useful for measuring liquid levels. In this particular application the magnet means may be embedded within a hollow cylindrical float means adapted to float on the liquid surface with the ferromagnetic element (optionally surrounded by a cylindrical protectivP
tube, e.g., of stainless steel) extending vertically through the central opening in the float means to a 13~ 4 point adjacent the bottom of the liquid container.
When ~o configured, as the liquid level chanyes, the float and magnet means move longitudinally along the element with the single pole comprising the inner diameter of the magnet means proximate the element (or its surrounding tube).
It is generally desirable for the magnet mean~ to be oriented with its axi~ or moment sub~tantially normal to the element'~ longitudinal axi~. When ~o oriented the influence of the magnetic field on the magnetization of element 16 is most localized and thi is generally preferred. However, the axis of the magnet means may be tilted, as hereinafter described, or the magnet means may be asymmetrically shaped or shielded on a single ~ide to expose element 16 to unequal field gradients in order to deliberately produce asymmetric magnetizations in regions A and B.
This would be desirable in those instances where it is sought to obtain unegual incremental permeabilities in regions A and B without refiort to the use of bias currents, as hereinafter described.
Referring to Figure 2 there is illustrated an alternative arrangement for ferromagnetic element 16 and magnet means 14. In this arrangement the path of travel of the movable member is indicated by bi-directional arrows S. Element 16 consists of a pair of parallel, elongated ferromagnetic elements 100,102, which may be independent parallel elements or elements joined at their respective end portions to form an endless ferromagnetic element (shown in phantom)~
Magnet means 14 is positioned between the elements 100,102 with one pole proximate element 100 and the opposite pole proximate element 102. As magnet means 14 movefi, with the movable member, along the path of travel indicated by S, the refipective poles of the magnet have oppo~ite polarizing effects on the respective elements 100,102. As a result the available 3V~

~ignal from such an arrangement is double that which ~ould be available from a ~ingle element arrangement.
~.oreover, this arrangement reduces the leakage flux from magnet means 14, thereby narrowing the axial length of the transition between regions A and B within each element. In addition, by positioning the magnet ~eanfi between and ~paced from elements 100,102 the effect of the air gap distance between the end of the magnetic pole and the element becomes inconsequential.
Increa~ing one air gap 104 decreases the other air gap 106, allowing common mode rejection of magnet position normal to the axis of the elements. In addition, the end effects attributable to the use of one or more independent elements may be minimized by closing the ~agnetic circuit, as indicated in phantom at 108,110.
Even better performance can be obtained from the arrangement of Figure 2 and multiple position sensors can be simultaneously employed close to one another ~ithout mutual interference by reducing even further the leakage field of the magnet. This can be accomplished by partially or substantially completely surrounding elements 100,102 and magnet means 14 with a aenerally cylindrical, elongated casing of soft ~agnetic material to provide shielding from the earth's and other ambient fields as well as from the fields associated with other position sensors. Depending upon the amount of shielding desired the casing may have a cross-section varying between C-shaped and circular (a circular casing may require the provision of means, such as a longitudinally extending slot, for mounting the magnet means, directly or indirectly, to the ~ovable member).
Operation of the magnetic position sensor of the present invention, in one preferred embodiment, as hereinbefore described, is dependent upon the unegual incremental permeabilities of the two regions A and B
of the ferromagnetic element, which permits the 13~ 4 determination of X, i.e., the intersection position, f rom measurements which ~en~e the effective overall incremental permeability, over the length L, of Eeg~ent 18. To better understand the principles underlying operation of the present magnetic position sensor reference is had to Figure 3 which illustrates a major hysteresi~ loop of a typical ferromagnetic material useful in the present invention. The hys~eresis loop is a plot of applied magnetic field (~) against r,agnetic flux density (B) (a comparable graphical relationship would result if magnetization ~.) was plotted as the ordinate since B and M are related by the equation B=H+41rM). Point J on the hysteresis loop of Figure 3 corresponds to the magnetization at a point along ~egment 18 when magnet means 14 is proximate to that point, e.g., in moving f rom its position at 24 towards segment end 22 to enlarge region A. After the cagnet means has moved away from that point a suff icient distance that its magnetic field no longer influences the magnetization at that point, the applied field becomes zero, and the magnetization at that point drops to its remanent value, corresponding to A'.
Likewise, point R corresponds to the magnetization at a point proximate to magnet means 14 as the magnet means r~oves from its position at 24 towards segment end 20 to ~nlarge region B. However, after the magnet means has rioved away from that point a sufficient distance that its magnetic field no longer influences the r;agnetization at that point, the magnetization drops to its remanent value corresponding to B'. The incremental permeability at any point on the loop is defined as the slope, ~B/~H, of the major axis of the lenticular-shaped minor hysteresis loop which results by applying incremental changes in field, ~H, at the point of interest. This change in field can be achieved by applying very small AC currents, sufficiently small not to disturb the magnetic state of 13(118~34 the ferromagnetic element in any irreversible manner, i.e., of small enough amplitude to naintain re~ulting ~agnetization excursions within reversible limit~
without altering the remanent magnetization~, at that point. It will be appreciated that the incremental permeabilitie~ at pointE A' and B' (slope ~B/~ at points A' and B') are very ~imilar and inRufficiently distinguishable for the differences in theEie incremental permeabilities to permit a determina$ion of the extent of the respective regions A and B. ~owever, the application of a small DC bias field (~hown in dashed lines) to the ferromagnetic element shifts the corresponding magnetization to points A and B in Figure 3. The minor loops at points A and B are shown adjacent the major hy~teresis loop. It can be Eieen that the slopes of the major axis of these loops, ~BA/~H and ~BB/hH, representino the incremental permeabilities ~A and ~B at pointE A and B, re~pectively, are very different and readily distinguishable. Tbus, in the DC bias field the rever~ible incremental permeability ~B of region B i8 much greater than the incremental permeability ~A f region A and ~B=cpA, where c iE a con~tant. The effective overall incremental permeability of segment 18, as shown in Figure 1, is given by the relationship X ~ ~B(L-X) ~E =
.

from which it can be seen that X, the position of the intersection of regions A and B, is a linear function f PE as follow6:

X = PE ~ B
~lA ~B )~A )~B

~3~

Expres6ing X in term~ of permeability ratios by substituting rB=c~A, yields (YA ) 1-C
Since ~E varie~ from a value ~A when X=L to a value rB when X=o, the ratio ~E/pA can vary from 1 to c. It will therefore be appreciated that, as hereinbefore &tated, the operation of the magnetic po~ition sensor of the pre~ent invention to detect X, the position of the intersection, is dependent on there being unequal effective permeabilities in regionæ A and B. Moreover, inasmuch as operation of the magnetic position sensor improves with increasing ratio of these permeabilitie6, desirable ferromagnetic materials exhibit substantial variation in incremental permeability with magnetization.
Materials exhibiting typical ferromagnetic hy~tere~is loops, as illustrated in Figure 3, generally exhibit an incremental permeability which initially decrea~es at a moderately rapid rate, but ~ubse~uently decreases at a more gradual rate, with increasing ~agnetization from ~0, the incremental permeability at ~=0 (which i~ typically 100 to 100,000 for ~soft~
~agnetic materials), to usat and which approaches a value of 1 at M8at. The remanent magnetization, MR, is the magnetization remaining in the ferromagnetic .aterial after being exposed to M8at and the remanence ratio MR/MSat, which varies from O to 1, is a measure of the squareness of the hy~teresis loop. Neither of these extremes is desirable. Since the application of a biasing field to the element causes one of the regions A and B to be biased toward M8at and the other to be biased toward less magnetization, the respective incremental perm eabilities will be altered correspondingly by the biasing field. Specifically, the incremental permeabilities will va ~ from the value ~l3~

at MR in opposite directions with the increase in permeability (increasing toward~ yO along the moderately rapidly changing part of the u vs. M curve) generally being larger than the decrease (decreasing toward ~5at along the more gradual part of the~u vs. M
curve). As a result, the application of a bias field to a suitably ferromagnetic element has the effect of increasing the difference between and the ratio of ~A
and ~B-Bias fields will generally be obtained with directcurrent except in very short stroke devices when permanent magnets might be used. To preserve power and to avoid heating due to such currents, it is desirable that the ferromagnetic material be selected such that satisfacto~ operation can be obtained with small bias fields. On the other hand, the magnetic position sensor should not be so sensitive that it reacts to ambient fields from the earth or nearby magnetized objects. Thus the desirable ferromagnetic materials for use in the present invention should be neither magnetically soft, as might be reguired for transformer cores, motors and other electrical machinery, nor magnetically hard, as might be used in permanent magnets. Materials that are in the desirable range from not especially soft to semihard, i.e., materials having coercivities from about 3 to 100 Oersteds, appear to be ideal and are particularly preferred.
xemplary of such materials are AWS 502 (5.0% Cr, 0.06% C, 0.8% Mn 0.4~ Si, 0.5% Mo, balance Fe), AWS 410 (12.3% Cr, 0.2% Ni, 0.08% C, 0.9% Mn, 0.4% Si, 0.4% Mo, balanae Fe), *52 Alloy (52% Ni, 49% Fe), *Kanthal 70 (70% Ni, 30% Fe), *Unimar 300 (17-19% Ni, 7-9.5% Co, 3-5.2% Mo, 0.1-0.8% Ti, .05-.15% Al, up to 0.03~ C, balance Fe) and *Teledyne Vasco 9-4-20 (9.8% Ni, 3.62% Co, .15% C, balançe Fe). Best results to date have been obtained with AWS 502 which appears to be stable against environmental magnetism but otherwise *Trade Mark 13~

sufficiently soft to provide the necessary sensitivity to the magnetic position ~ensor of the pre~ent invention.
Ideally, the ferromagnetic material~ ~elected for use in the present invention ~hould have the aforementioned desired properties after annealing or other heat treatment rather than from cold working since it is important that the properties be homogeneous along the entire length of the element and the distribution of properties resulting from cold work mzy be difficult to control. If desired, stress due to tension or torsion may be suitably instilled in the element to provide a desirable magnetic anisotropy for m odifying the permeability of an otherwise less satisfactory material.
It is well-known that the total inductance, LT, of a coil is a linear function of the permeability of the material it encloses. It follows that inductance is one very convenient way of obtaining a measure of the overall effective incremental permeability of the ferromagnetic material comprising the core of the cGil and, thereby, of the differing remanent magnetization of the regions A and B. In accordance with the present invention it has been found that inductance i8 a very effective way of obtaining a measure of the overall effective permeability of the segment L, comprising the two regions A and B of the ferromagnetic element 16 of Figure 1, and of determining the distance X, the location of the intersection of regions A and B.
Referring to Figure 1, if element 16 of cross-sectional area A is overlaid with a solenoidal winding having ~nn turns per unit length, the inductance of the portion of the winding in region A (i.e., to the left of the position of magnet means 14) is given by the relationship ~3~38~

LA ~ nXd~/di where t is the magnetic flux. Expres~ing the same relation~hip in terms of flux den~ity (B) and magnetic field tH) yields:

LA = nXAdBA/di LA = nxAp~d~/di Substituting for the magnetic ield yields:

LA = nXApA(.4~ ~di/di) ~pressing all constants as the single constant R:

LA = KX~A

In like manner, the corresponding Pxpression for region B is:
LB = R(L-x)~B

The total inductance of the winding, LT=LA~LB, is:
LT = R (XPA + (L~X),UB) Since PA = b~uB, substituting in the expression for LT
and simplifying, yields LT = R(XbyB ~ (L-X)~uB) LT = ~yB[(b-l)X~L]

This relationship confirms that the inductance of a coil wound about ferromagnetic element 16 in Fiqure 3 i-~ a linear function of the position X of magnet means 1~ and can be advantageously used in designing 13~i884 electrical circuitry which will produce an electrical signal indicative of the position of magnet mean~ 14 along element 16.
One ~uch exemplary circuit is ~hown in Figure 4.
This circuit, which includes a ~ingle operational amplifier OA and coil 30 wound about element 16 oscillates at a frequency that varies with inductance.
The period (l/f) i8 a linear function of the inductance if RLoss~ the resistive 1088 in solenoid 30, is con~tant. In actuality, RLoss increa~es with frequency; however, the effect of this change can be made negligible by making the resistance R large relative to the largest expected value of RLoss (typically about 7 ohms). The circuit includes a variable voltage source 32 in series relationship with R3 and Ll and milliammeter 34 to provide the desired DC
bias current through the solenoid 30. It is desirable to make the inductance of Ll large compared to the inductance of ~olenoid 30 (which i~ typically about 0.2 mH) to effectively block any alternating current from flowing through the biasing circuit. At the same time, R3 is set to a value which only negligibly affects the DC current flowing through the solenoid.
Capacitor Cl serves to smooth the output wav~ shape.
Exemplary component parameters for the satisfactory operation of this circuit in accordance with the present invention, with a power input to the circuit at terminals 36 of + 15 volts, are the following:

R = 120 ohms Ll = 100 mH
Rl = 2700 ohms Cl = ~47 ufd R2 = 1000 ohms R3 = 370 ohms To demonstrate the use of the circuit of Figure 4 in a magnetic position ~ensor such as is illustrated in Figure 1, a sensor was constructed using a 0.8 mm ~3~81~4 ciameter length of AWS 502 wire as ferromagneticelement 16. The wire was stress relieved by heating ~-ith a conducted current of 6 amperes RMS 60 Bz for 6 ~inutes under which conditions the wire reached an estimated temperature of 500C.
A solenoidal winding of ~34 AWG with heavy formvar insulation was closely wound onto the element 16. The ~-inding length was 39.6 cm with 51 turns/cm. The element was approximately 10 cm longer than the coil and was allowed to protrude therefrom 5 cm at each end to avoid strong demagnetizing fields near the coil ends. The element with its solensidal winding was laid into a narrow, close fitting groove milled into a ~-ooden board. This board was mounted ~or support on a ~illing machine table. A small alnico magnet, 4 mm in diameter by 12 mm long, was mounted in the spindle of the milling machine. ~he spindle, and therefore the ~agnet, was oriented normal to the element and approximately centered over it with the end of the ~agnet spaced from the solenoid by about 1 mm.
With the magnet fixed in place the table was moved relative to the spindle to move element 16 relative to ~agnet 14. The element was fir~t moved, in a single continuous pass, from one coil end to the other under the magnet to achieve the required initial polarization (magnetic preconditioning). Subsequent relative r ovement of the element and the magnet was monitored by ~easuring the period of the output wave from the circuit shown in Figure 4. A DC bias current of 30 mA
~as supplied by variable voltage ~ource 32. Although the coil was 39.6 cm (15.6 inches) long, mea urements ~-ere taken over a segment thereof of length L=38.1 cm (15 inches) at points 2.54 cm (1 inch) apart. This interval was selected since the controls for the ~illing machine table were calibrated in inches and it ~-as relatively easy to accurately traverse 1 inch lengths of the element with each movement.

~3~1~384 ~ he data resulting from this demonstration aregraphically illustrated in Figure 5. It will be appreciated that, with only minor departures, the relationship between the magnet position X, i.e., the distance from segment end 20, and the measured period of the circuit output wave is linear. The large~t linearity error in the data, i.e., departure from the illustrated line in ~igure 5, was 1.6~. The large~t hysteresis error, obtained when the direction of movement of the magnet, after reaching segment end 22, ~as reversed and measurements were taken at 1 inch intervals as the magnet returned to segment end 20, was 1.1%. This data is believed to be excellent and justifies use of the sensor in such varied applications as control elements in machinery and liquid level devices, among others. However, the foregoing manner of data evaluation is relatively harsh. It is common in evaluating position sensing devices to use the center as the reference point. If this is done, the performance of the tested sensor improves remarkably.
Horeover, if only the center section of the coil is used for position sensing to avoid the adverse influence of end effects, such as fall off of the field within the solenoid _(only half as large at the coil ends as at the center) and increase in the demagnetizing field (greatest at the ends), the data improves remarkably. Even if the coil ends are used, these end effects can be minimized by closing the magnetic path, as shown at 108 and 110 in Figure 2, and winding additional layers of solenoid at the ends.
The linearity of the relationship shown in Figure 5 is seen to be excellent, except possibly at the very ends where the loss of mutual inductance diminishes the inductance per unit length and the demagnetiæing fields of element 16 tend to equalize the intensities of the two remanent states. Since the inductance per unit length varies inversely with the ~ .. ':

~3~11384 square of the ~inding pitch, it is po~sible to compensate for these end effect~ by concentrating more turns in the end regions of the ~olenoid. Moreover, by appropriate variation of the winding pitch over the length of element 16, ~pecific, functionally desirable transfer functions for the ~ensor are attainable. ~or e~ample, using emprical techniques, the position vs.
inductance transfer function can be altered to follow a desired relationship, ~uch as the relationship between liquid level and volume. This approach finds utility in liquid level devices utilizing a floating magnet, where the output ~ignals are made to be proportional to the volume remaining in vessels of non-uniform cross-sections. Unlike vessels such as vertical cylinders, where the volume of liquid in the vessel is a constant function of liquid level, with vefisel~ such as spherical tanks or cylindrical tanks with di~hed, oval or spherical end caps, the volume of liquid in the vessel is a function of ve~sel radius as well as liquid level. To account for vessels, such as spheres, whose volume increases or decreases rapidly with only small changes in liquid level, the winding pitch over the length of element 16 may be varied with more or less turns (decreased or increased pitch) in sections of element 16 corresponding to areas in which such large volume changes occur. The increased (decreased) concentration of windings provides an increased (decreased) output ~ignal to reflect the increased (decreased) volume change notwithstanding that there has only been a small linear movement of magnet 14.
Another exemplary circuit useful in accordance with the present invention is illustrated in Figure 6 in connection with the use of twin, parallel ferromagnetic elements 100,102, as illustrated in Figure 2. Solenoids 50,52 are oppositely wound upon elements 100,102 such that they are biased in opposite directions by current flowing from source Vcc lnto ~3~

- 2~ -adjacent ends. The magnet means 14 is oriented 6uch that the same pole, indicated as the N pole, influences the polarity in each element 100,102. Thus, as magnet ~eans 14 moves along the solenoids from top to bottom, as shown in Figure 6, the magnet influences the polarity in each element in the same manner but, due to the opposite windings in each element, the inductance of one coil increase while the inductance of the other coil decreases. In the multivibrator circuit shown, only one of the transistors Ql~ Q2 conducts at a time, thus allowing a square wave voltage to create a cyclically time varying magnetic field for application to elements 100,102. As the inductance of one coil 50,52 increases while the inductance of the other decreases, this difference in inductance produces different voltage signals Vl, V2, the magnitude of which linearly depend upon the relative inductances in the respective collector circuits. The difference in the magnitudes of Vl and V2 appear as VOUt.
Interestingly, with a circuit as illustrated in Figure 6, there i8 no need to apply a separate DC bias current. This is because the current in each coil 50,52 varies from zero to a finite amount in one direction only and is the equivalent of a DC bias and an AC component.
The very same desired effect, i.e., 6imultaneously increasing the inductance of one coil while decreasing the inductance of the other coil, can be achieved in other circuit configurations similar to that of Figure 6. Por example, using the same magnet configuration, the coils 50,52 may be wound in the same direction if current from source V~c is made to flow into the respective coils from opposite ends. If the magnet configuration were modified to the configuration shown in Figure 2, wherein opposite poles of the magnet influence the polarization in each of the elements 100,102, then the desired effect may be achieved with coils 50,52 wound in the ~ame direction with current entering e~ch of the coils from the 6ame end.
Alternatively, with the magnet configuration of Pigure 2, the coils 50,52 can be wound in oppo~ite directionfi if the current from ~ource Vcc flows into the respective coils from opposite ends.
It may be desirable for reducing the complexity and c06ts of the sen6ing circuits a~ well a6 their electrical power requirements to eliminate the bias field used to upset the equality of the magnetization in regions A and B. With an electrically ~upplied bia6 field there is also some ambiguity in position indication if the magnet is moved while the power i~
off. The need for a bia6 field can be eliminated, in accordance with the present invention, without sacrificing the de~ired intensity differential between the oppositely polarized element region~ A and B by tilting the magnet 14, ~uch that its moment is at a tilt angle ~, away from a po~ition normal to the longitudinal axis of element 16, as can be seen from Figure 7. Tilting disrupt~ the symmetry between the oppositely directed magnetizing fields acting on the element, as can be seen in Figure 8 which graphically illustrates the relationship between the field, ~, and ~ which is a normalized distance from magnet 14 along element 16 for representative tilt angles of ~=0, a=300, ~=60~ and ~=90. The unegual maxima in left and right directed fields is clearly apparent for values of greater than 0.
It should be appreciated that except for the small portion of element 16 underlying magnet 14, there is no continuous field of conseguential magnitude acting on element 16. Since magnet 14 is the ~ource of the field, those portions of element 16 traversed by movement of the magnet are effectively swept by a characteri6tic sequence of field maxima as determined b~ the 6trength and tilt angle of magnet 14. It can be ~3t~

under~tood from Pigure B that magnetization re~er~al requires that the field maxima of major lobes 1 and 2 ~ajor field maxima) both exceed the coercive force of element 16. At the 6ame time, the field maxima of lobe 3 must be ~ubfitantially le~s than the coercive force. Contrary to the ~ituation where an applied bias field i~ used to upset the equality of the magnetization in the two region~ A and B of element 16, where a tilted magnet is u~ed the field maxima of the ~eaker of the two major lobes, lobe 2 in Figure 8, while large enough to exceed the coercive force of element 16, must not be 60 large that after magnet 14 has passed any point along element 16, the re~anent incremental permeability (and remanent magnetization) at that point approaches the ~aturation level. This is to assure that the lobe 1 field maxima and the lobe 2 field maxima acting on element 16 produce distinguishable remanent incremental permeabilities.
To accomplish this the field maxima of the wea~er of the two major lobes i6, mo~t de~irably, in the range 1.2 to 2.0 time~ the coercive force of element 16 and is sufficiently smaller than the field maxima of the other major lobe to produce distinguishable re~anent incremental permeabilitie~. Thus, the ~election of an optimum tilt angle is interrelated with the strength of the magnet employed. As long as this interrelationship is observed, the selection of a tilt angle, ~, in the range 20 to 70 appears to be sati~factory.
It will be appreciated that notwithstandins that the foregoing illustrative examples have utilized solenoidal windings, the interactive conductor r~uired to provide the bias and alternating fields for ~ensing the effective permeability of the element ~ay be configured in other ways than solenoidal winding~. For example, the ferromagnetic element itself may be used as a conductor or it may be provided with a more conductive core (as by plating the ferromagnetic 13~31~8~

element onto a conductive wire, strip or ribbon or by passing a conductor through a hollow, e.g., tubular, ferromagnetic element) or it may be positioned proximate to a colinear conductor. For example, the ferromagnetic element may be magnetically coupled to a ~aturable core and the windings positioned on the ~aturable core. In thi~ way an external magnetic structure may be used to determine the magnetization of the element 16, rather than employing the element itself, which is particularly useful in connection with ~mall devices. If the ferromagnetic element also serves as a conductor, it may be desirable to have an insulating interface between the element and the cond~ctor.
It will also be appreciated that notwithstanding that the present invention has been described primarily in terms of measuring inductance as a convenient means for sensing the position of the intersection between oppositely polarized, remanently magnetized regions A
and B, any characteristic of element 16 which varies ~ith changes in the remanent magnetization of the regions may be measured. Thus, it is known that the ~elocity of sound waves in element 16 varies with ~agnetization. Inasmuch as the magnetization of regions A and B are different, sound will travel at a different velocity in each of regions A and B and, tberefore, depending upon the length of each region the velocity of sound through element 16 will vary.
Therefore, if a sound wave generator 26 is located at 20 and a corresponding detector 28 is located at 22, the time required for sound waves to travel along element 16 can be measured and correlated with the position of magnet 1~ along element 16. In still another way of sensing the position of the intersec$ion, it is known that the resistance of element 16, due to the magnetoresistance effect, varies with magnetization. Inasmuch as the ~agnetization of regions A and B are different, the .; .
.

13q~1884 resistance of element 16, as indicated by resistance sensing means 29 schematically illustrated in Fig. 1, depends upon the length averaged resistance of each region. Therefore, the sensed resistance of element 16 can be correlated with the position of magnet 14 along element 16.
Whether or not solenoidal windings are used, the excitation (AC and bias) and sensing circuits may be separated. In the simplest case three colinear con-ductors, for AC excitation, bias and sensingl may pass through the hollow opening of a tubular ferromagnetic element for providing isolation of the separate circuits.
In addition, combinations of linear conductive elements and solenoidal windings may be used. In this connection, it is important to note that interaction between the circular field of a centrally conductive current carrying wire and an external solenoidal winding would be enhanced if the ferromagnetic material were endowed with helical anisotropy, as from torsion, heat treating in a magnetic field or plating techniques. Among the advantages con-ferred by using magnetic anisotropy is that the voltage generated as a result of the Matteucci effect, the inverse Wiedemann effect, or any combination of the two, may be sensed to detect the postiion X of the magnet at the intersection of regions A and B. In a ferromagnetic element having helical anisotropy instilled therein, a coaxial conductor and a solenoidal winding thereabout, the flow of AC current through the solenoid causes a voltage, the Matteucci signal, to appear at the ends of the coaxial conductor. The flow of AC current through the coaxial conductor causes a voltage, due to the inverse Wiedemann effectl to appear across the solenoid.
Either of these voltages have a linear correlation with incremental permeability and, hence, with position X.
While the present invention has been described with respect to particular embodiments thereof, it will be appreciated that numerous modifications may be made ~3(~8~3 b~ tho~e ~killed in the art without actually departing from the ~cope of the claimed invention. Accordingly, all modification~ and equivalents may be resorted to which ~all within the 6cope o~ the invention a~
claimed.

Claims (34)

1. A position sensor for detecting and providing an indication of the position of a movable member compri-sing:
a ferromagnetic element oriented substantially parallel to the path of travel of said movable member, said element including a segment thereof having first and second spaced-apart ends defining the positional limits between which movement of said movable member is to be monitored, the distance along said element between said ends defining the length L of said segment, said segment comprising a pair of contiguous, oppositely polarized, remanently magnetized regions, each of said regions being uniformly magnetized along its length, said first region including said first end and having a length X wherein O
< X < L and said second region including said second end having a length L-X, said contiguous regions defining an intersection along the length of said segment;
magnet means positioned proximate said intersection;
one of said magnet means and element mounted in association with and in known spatial relationship to said movable member for movement therewith and relative to and along the other of said magnet means and element at a substantially constant distance from one another, said magnet means providing a magnetic field presenting a constant polarity to said element and having a strength sufficient to locally polarize said element, whereby said relative movement between said magnet means and said element alters the length of the respective regions;
means associated with said element for sensing the position of said intersection without altering that position and for providing an indication of the position of said movable member based thereon.

2. A position sensor, as claimed in claim 1, where-in said means for sensing comprises means for sensing a characteristic of said element which varies with changes in remanent magnetization, said sensed characteristic being correlatable with the position of said inter-section.

3. A position sensor, as claimed in claim 2, wherein said means for sensing comprises means for sensing the permeability of said element.

4. A position sensor, as claimed in claim 2, wherein said means for sensing comprises means for sensing the speed of sound in said element.

5. A position sensor, as claimed in claim 2, wherein said means for sensing comprises means for sensing the resistance of said element.

6. A position sensor, as claimed in claim 1, wherein said means for sensing comprises means for sensing the effective overall incremental permeability over the length of said element.

7. A position sensor, as claimed in claim 6, wherein said means for sensing further includes means for applying a small DC bias field to said element.

8. A position sensor, as claimed in claim 1, wherein said means for sensing includes coil means wound about said ferromagnetic element and electrical circuit means associated therewith for sensing the inductance of said coil and for providing an indication of the position of said movable member based thereon.

9. A position sensor, as claimed in claim 1, wherein said magnet means has a field strength at said element which exceeds the coercive force of said element.

10. A position sensor, as claimed in claim 7, wherein said magnet means has a field strength at said element at least twice the coercive force of said element.

11. A position sensor, as claimed in claim 1, wherein said magnet means is oriented with its moment tilted at an angle of 20° to 70° from a position normal to the longitudinal axis of said element.

12. A position sensor, as claimed in claim 11, wherein the magnetic field strength of said magnet means is such that both of the major field maxima exceed the coercive force of said element.

13. A position sensor, as claimed in claim 12, wherein the weaker of the major field maxima is 1.2 to 2.0 times the coercive force of said element and is sufficiently smaller than the other major field maxima to produce a remanent incremental permeability which is distinguishable from the remanent incremental permeability produced by the other major field maxima.

14. A position sensor, as claimed in claim 1, wherein said magnet means comprises a permanent magnet.

15. A position sensor, as claimed in claim 1, wherein said magnet means is oriented with its moment substantially normal to the longitudinal axis of said element.

16. A position sensor, as claimed in claim 1, wherein said ferromagnetic element has a coercive force in the range from about 3 to 100 Oersteds.

17. A position sensor, as claimed in claim 1, wherein said ferromagnetic element comprises a single elongated ferromagnetic element.

18. A position sensor, as claimed in claim 1, wherein said ferromagnetic element comprises a pair of parallel, elongated ferromagnetic elements, each said element comprising a pair of contiguous, oppositely polarized, remanently magnetized regions.

19. A position sensor, as claimed in claim 18, wherein said magnet means is positioned between said elements for movement substantially parallel thereto, said magnet means presenting a single, constant polarity to each said element.

20. A position sensor, as claimed in claim 18, wherein said magnet means is positioned between said elements for movement substantially parallel thereto, said magnet means presenting opposite polarity to each said element.

21. A position sensor, as claimed in claim 1, where-in said ferromagnetic element is endowed with intention-ally instilled magnetic anisotropy.

22. A position sensor, as claimed in claim 8, where-in the pitch of said coil means is uniform along the length of said element.

23. A position sensor, as claimed in claim 8, where-in the pitch of said coil means is smaller adjacent the ends of said element than along the length of said element intermediate said ends.

24. A position sensor, as claimed in claim 8, where-in the pitch of said coil means is non-uniform along the length of said element.

25. A position sensor, as claimed in claims 1, 2, 6, 7, 8, 11, 12 or 24, wherein said magnet means is mounted in association with and in known spatial relationship to said movable member for movement along said ferromagnetic element.

26. A method for detecting and providing an indica-tion of the position of a movable member comprising the steps of:
orienting a ferromagnetic element substantially parallel to the path of travel of said movable member, said element including a segment thereof having first and second spaced-apart ends defining the positional limits between which movement of said movable member is to be monitored, the distance along said element between said ends defining the length L of said segment, said segment comprising a pair of contiguous, oppositely polarized remanently magnetized regions, each of said regions being uniformly magnetized along its length, said first region including said first end and having a length X wherein 0 ? X ? L and said second region including said second end having a length L-X, said contiguous regions defining an intersection along the length of said segment;
mounting magnet means in association with and in known spatial relationship to said movable member for movement therewith, and, simultaneously, along said ferromagnetic element at a substantially constant distance therefrom, said magnet means being positioned proximate said intersection and providing a magnetic field presenting a constant polarity to said element and having a strength sufficient to locally polarize said element, whereby movement of said magnet means along said element alters the length of the respective regions;
sensing the position of said intersection without altering that position and providing an indication of the position of said movable member based thereon.

27. A method, as claimed in claim 26, wherein said sensing step comprises sensing a characteristic of said element which varies with changes in remanent magnetiza-tion, said sensed characteristic being correlatable with the position of said intersection.

28. A method, as claimed in claim 27, wherein said sensing step comprises sensing the permeability of said element.

29. A method, as claimed in claim 26, wherein said sensing step comprises sensing the speed of sound in said element.

30. A method, as claimed in claim 26, wherein said sensing step comprises sensing the resistance of said element.

31. A method, as claimed in claim 26, wherein said sensing step comprises sensing the effective overall incremental permeability over the length of said element.

32. A method, as claimed in claim 26, wherein said sensing step comprises providing coil means wound about said ferromagnetic element, sensing the inductance of said coil and providing an indication of the position of said movable member based thereon.

33. A method of making a non-contact position sensor for detecting the position of a movable member comprising the steps of:
providing a ferromagnetic element adapted to be oriented substantially parallel to the path of travel of said movable member, said element including a segment thereof having first and second spaced-apart ends defin-ing the positional limits between which the position of

34 said movable member is to be monitored, the distance along said element between said ends defining the length L of said segment;
positioning a magnet means proximate said element for providing a magnetic field presenting a constant polarity to said element and having a strength sufficient to locally polarize said element, whereby relative move-ment between said magnet means and said element alters the polarity of the portions of said element proximate said magnetic means;
uniformly magnetizing said segment in a first direc-tion along its length L by moving one of said magnet means and said element relative to the other in a single direction along the entire length of said segment between its ends; and positioning said magnet means along said segment while maintaining said magnet means proximate said ele-ment by moving one of said magnet means and said element relative to the other in the opposite direction a dis-tance X wherein O ? X ? L, said movement reversing the polarity in the portions of said element along which it moves for defining within said segment a pair of contig-uous, oppositely polarized, remanently magnetized regions, each of said regions being uniformly magnetized along its length, said first region having a length X and said second region having a length L-X, said contiguous regions defining an intersection therebetween at the position of said magnet means.
34. A position sensor, as claimed in claim 1, wherein said ferromagnetic element is elongated and said magnet means comprises a hollow, generally cylindrical means having a generally centrally disposed opening defined therein for relative movement along said element with said element extending through said opening, said generally cylindrical means including an inner diameter surface proximate said element and an outer diameter surface remote from said element, said inner diameter surface comprising one pole of said magnet means and said outer diameter surface comprising the other pole of said magnet means.