Devices and methods for determining a magnetic field distribution of a magnet along a main surface of said magnet
09797964 · 2017-10-24
Assignee
Inventors
Cpc classification
International classification
Abstract
Embodiments described herein relate to devices and methods for determining a magnetic field distribution of a magnet along a main surface of said magnet. An example device for determining a magnetic field distribution of a magnet along a main surface of said magnet includes an arrangement of at least two independent magnetic field camera modules being arranged in a fixed relative position with respect to each other, each magnetic field camera module being adapted for measuring a magnetic field distribution to which it is exposed by means of a respective detection surface. The device also includes a means for providing a predetermined relative movement between the main surface and the arrangement to thereby scan the magnetic field distribution of the magnet along the main surface.
Claims
1. A device for determining magnetic field, the device comprising: two or more independent magnetic field image sensors arranged in a fixed relative position with respect to each other, wherein each magnetic field image sensor measures a magnetic field distribution along a main surface of a magnet; and a scanning stage configured to provide a predetermined relative movement between the main surface of the magnet and the two or more independent magnetic field image sensors, thereby scanning the magnetic field distribution along the main surface of the magnet.
2. The device according to claim 1, wherein the two or more independent magnetic field image sensors are arranged such that detection surfaces of each of the two or more independent magnetic field image sensors lie within a single plane.
3. The device according to claim 2, wherein the detection surfaces of each of the two or more independent magnetic field image sensors are arranged and aligned along a single line.
4. The device according to claim 2, wherein the detection surfaces of each of the two or more independent magnetic field image sensors are arranged and aligned along two parallel lines, such that an orthogonal projection of the detection surfaces on a virtual line parallel to the two parallel lines provides a single, uninterrupted portion of the virtual line.
5. The device according to claim 3, wherein the scanning stage is configured to provide one or more relative translational movements between the two or more independent magnetic field image sensors and the main surface of the magnet, and wherein the detection surfaces and the main surface of the magnet are maintained parallel during the one or more relative translational movements.
6. The device according to claim 5, wherein at least one of the one or more relative translational movements between the two or more independent magnetic field image sensors and the main surface of the magnet bridges a previously existing dead zone between at least two predetermined detection surfaces.
7. The device according to claim 3, wherein the scanning stage is configured to provide a relative rotational movement between the two or more independent magnetic field image sensors or the main surface of the magnet around a rotation axis.
8. The device according to claim 1, wherein a first subset of the two or more independent magnetic field image sensors is arranged such that their detection surfaces lie within a first plane, and wherein a disjoint second subset of the two or more independent magnetic field image sensors is arranged such that their detection surfaces lie within a second plane, and wherein the second plane is different from the first plane.
9. The device according to claim 8, wherein the detection surfaces of the first subset are arranged and aligned along a first line, wherein the detection surfaces of the second subset are arranged and aligned along a second line, and wherein the first line and the second line are parallel.
10. The device according to claim 9, wherein the first plane and the second plane are parallel.
11. The device according to claim 9, wherein the first plane and the second plane form an angle different from 0° or 180°.
12. The device according to claim 9, wherein an orthogonal projection of the detection surfaces of the first subset and the detection surfaces of the second subset on a virtual line parallel to the first line and the second line covers a single, uninterrupted portion of the virtual line.
13. The device according to claim 8, wherein the scanning stage is configured to provide a relative rotational movement between the two or more independent magnetic field image sensors or the main surface of the magnet around a rotation axis.
14. The device according to claim 8, wherein the first plane and the second plane are arranged such that, when the scanning stage provides the predetermined relative movement between the main surface of the magnet and the two or more independent magnetic field image sensors, at least a portion of the main surface of the magnet is scanned by a magnetic field image sensor of the first subset and by a magnetic field image sensor of the second subset at different distances from the main surface of the magnet.
15. A method for determining magnetic field, the method comprising: providing a magnet; providing two or more independent magnetic field image sensors arranged in a fixed relative position with respect to each other, wherein each magnetic field image sensor measures a magnetic field distribution along a main surface of the magnet; and providing a predetermined relative movement between the main surface of the magnet and the two or more independent magnetic field image sensors, thereby scanning the magnetic field distribution along the main surface of the magnet.
16. The device according to claim 1, wherein the magnet is a permanent magnet.
17. The device according to claim 1, wherein the predetermined relative movement is of a distance smaller a size of any of the two or more independent magnetic field image sensors, thereby providing an overlap region between images recorded by each of the two or more independent magnetic field image sensors.
18. The device according to claim 1, wherein the magnet is mounted on a rotor.
19. The device according to claim 18, wherein measuring the magnetic field distribution along the main surface of the magnet comprises measuring a radial component of the magnetic field distribution over 360° around the rotor and along a full axial length of the rotor.
20. The device according to claim 19, wherein at least one of the two or more independent magnetic field image sensors comprises a Hall sensor surface, perpendicular to a radial direction of the rotor, that measures the radial component of the magnetic field distribution.
Description
BRIEF DESCRIPTION OF THE DRAWINGS
(1) The disclosure will be further elucidated by means of the following description and the appended figures.
(2)
(3)
(4)
(5)
(6)
(7)
(8)
DETAILED DESCRIPTION OF PREFERRED EMBODIMENTS
(9) The present disclosure will be described with respect to particular embodiments and with reference to certain drawings but the disclosure is not limited thereto but only by the claims. The drawings described are only schematic and are non-limiting. In the drawings, the size of some of the elements may be exaggerated and not drawn on scale for illustrative purposes. The dimensions and the relative dimensions do not necessarily correspond to actual reductions to practice of the disclosure.
(10) Furthermore, the terms first, second, third and the like in the description and in the claims, are used for distinguishing between similar elements and not necessarily for describing a sequential or chronological order. The terms are interchangeable under appropriate circumstances and the embodiments of the disclosure can operate in other sequences than described or illustrated herein.
(11) Moreover, the terms top, bottom, over, under and the like in the description and the claims are used for descriptive purposes and not necessarily for describing relative positions. The terms so used are interchangeable under appropriate circumstances and the embodiments of the disclosure described herein can operate in other orientations than described or illustrated herein.
(12) Furthermore, the various embodiments, although referred to as “preferred” are to be construed as exemplary manners in which the disclosure may be implemented rather than as limiting the scope of the disclosure.
(13) The example embodiments which are illustrated below are based on a measurement principle which is based on a high resolution and high speed quantitative 2D mapping of the magnetic field distribution of the magnet, using a sensor chip with an integrated 2D array of 128×128 (=16384) microscopic Hall sensors. The sensors have a pitch (spatial resolution) of 0.1 mm in both X and Y directions. Each Hall sensor had an active area of 40 μm×40 μm and is adapted for locally measuring the perpendicular component (B.sub.z) of a magnetic field to which it is exposed. All sensors were electronically scanned at high speed, resulting in a quantitative high resolution magnetic field map over an area of 12.8 mm×12.8 mm. A full resolution magnetic field map could be captured in less than 1 second.
(14) The magnetic field camera chip had a built-in flexibility for recording any sub matrix of the 128×128 sensor matrix. Both in X and Y directions, the start pixel, stop pixel and step size can be specified. This allows not only to e.g. operate the sensor in half resolution mode (measuring every second pixel), but also allows to e.g. measure one single pixel continuously, or record one line continuously. The latter configuration will be used later in this description to construct a large range line scan embodiment, as well as a rotor inspection embodiment.
(15) The pixels on one sensor chip (i.e. in one magnetic field camera module) were read out in sequential order, whereby each pixel needed about 50 micro seconds measurement time. This means that one could calculate the total needed time for recording one frame by simply multiplying the total number of pixels in the frame by the time per pixel of 50 microseconds.
(16) The measurement times for some special cases are: Complete sensor array: t=128*128*50 μs=0.8 seconds Half resolution, full range: t=64*64*50 μs=0.2 seconds One line, full resolution: t=1*128*50 μs=6.4 ms One line, half resolution: t=1*64*50 μs=3.2 ms
(17) The magnetic camera modules had a size of about 24 mm×24 mm×24 mm. The camera modules comprise an upper surface 2 comprising a detection surface 21 and a portion 22 next to this detection surface where detection is not possible. The lateral sizes of the magnetic camera module (24 mm×24 mm) were such that it was allowed to placing multiple modules next to each other, while the dead measurement zone 6 (or dead zone) in between the modules (24 mm−12.8 mm=11.2 mm) is smaller than the active measurement size of the camera itself (i.e. 12.8 mm). This principle allows filling up this dead zone 6 by moving the set of modules with one single step of 12 mm in each direction. As illustrated below, this principle makes possible a number of different module configurations that are suited to perform fast large area magnetic field mapping measurements.
(18) The most straightforward way of covering a large are using a magnetic field camera is to mount it onto a XY(Z) scan stage (or on a robot) and to sequentially measure multiple small-scale magnetic field maps which are subsequently stitched together to obtain a large area image. The advantage of this method is that only a single magnetic field camera module would be needed.
(19) In order to accurately stitch the different measured images into a larger image, several approaches can be taken. Either the scan step (in X and Y directions) can be taken to be exactly equal to the corresponding size of the measurement area (i.e. 12.8 mm). In this case there is no overlap between consecutive images. The quality of the resulting stitched image depends then on the accuracy of the scan stage and the alignment of the sensor X and Y axes to those of the scan stage.
(20) Another technique, which is also found in optical image stitching, is to take the scan step somewhat smaller than the sensor size, as to assure an overlap region between adjacent images. This way, image stitching algorithms can be applied to the images, where the optimal overlap position is automatically detected. This method can correct for the lack of accuracy of the scan stage or of the alignment of the sensor to the scan stage axes.
(21) Since the measurement speed of a magnetic field camera can be relatively high (i.e. 12.8×12.8 mm.sup.2/0.8 seconds in full resolution) large areas can quickly be measured using this technique. However, for some applications, such as fast inline inspection, the speed may not be sufficient, especially for larger areas. The speed in this configuration is limited by for instance the following factors: The measurements are performed sequentially, i.e. not in parallel The number of mechanical steps scales with the measurement area Roughly, the total measurement time for an area A is equal to
T.sub.total=(T.sub.single measurement+T.sub.scanstep)λA/A.sub.sensor (1)
where T.sub.total is the total measurement time, T.sub.single measurement is the time needed to measure a single 12.8 mm×12.8 mm image, A is the area to be measured, A.sub.sensor is the area of the sensor, i.e. 12.8 mm×12.8 mm.
(22) In the considerations below, A.sub.sensor has been set to 12 mm×12 mm instead of 12.8 mm×12.8 mm for the following reasons: first, usually one measures only at half resolution, meaning that the last pixel in the array will not be recorded, diminishing the effective area to 12.7 mm×12.7 mm. Second, one allows for some overlap between the different images to make it possible to use automatic stitching algorithms. In practice, a scan step of 12.0 mm has therefor been used.
(23) Expression (1) shows that the total measurement time increases proportional to the area to be measured. Based on this formula, quantitative measurement times can be calculated for a few realistic cases.
(24) Concerning the time per single measurement, the following approach was followed. In practice, magnetic field camera measurements are often performed with half spatial resolution (i.e. 0.2 mm) instead of full resolution (i.e. 0.1 mm), especially for larger magnets. This means that only each second pixel is measured in X- and Y-directions, resulting in only ¼ of all pixels being effectively read-out. This also means that the measurement time per frame is four times smaller, i.e. about 0.2 s/frame. On the other hand, multiple frames are usually recorded and averaged in order to reduce measurement noise. A realistic number of averages would be 5, resulting in a total measurement time per frame of 1 second, which is comparable with the time for a single frame at full resolution.
(25) The second parameter in the expression above is the time needed for a mechanical scan step (T.sub.scanstep), the scan step being 12 mm in either the X or Y direction. Depending on the type of scan stage used, this time can vary. Setting this to 0.5 s, corresponding to a speed of 24 mm/s, totaling the time for a single combined measurement and scan step to 1.5 s.
(26) TABLE-US-00001 TABLE 1 Total measurement times for different areas when scanning a single magnetic field camera Measurement # Time Total area steps per step time 12 × 24 mm.sup.2 2 1.5 s 3 s 24 × 24 mm.sup.2 4 1.5 s 6 s 24 × 48 mm.sup.2 8 1.5 s 12 s 48 × 48 mm.sup.2 16 1.5 s 24 s
(27) For large areas the measurement time becomes relatively long, as is clear from Table 1. This measurement time can be considerably shortened by using a configuration of multiple camera modules, according to any of the embodiments of the present invention.
(28) Some example embodiments are described now.
(29) According to a first example embodiment, the device comprises a 2D array of magnetic field camera modules, for instance mounted on a support S. An example of a regular 2×2 array or matrix of four magnetic camera modules I, II, III, IV (having for instance total measurement area of 48 mm×48 mm) is depicted in
(30) Some advantages of this embodiment are that all camera modules 1 in the 2D array are measured in parallel, decreasing the total measurement time; and that the ‘dead measurement zone’ 6 between the modules can be filled up with only three mechanical scan steps of the complete configuration of 12 mm (one in X, one in Y and one in X+Y directions), independent of the number of camera modules used and thus of the actual measurement area. To this end, the camera array can for instance be mounted onto a motorized XY(Z) scan stage, or it can be moved manually, or can be moved based on a state of the art movement platform.
(31) This principle is illustrated in
(32) The total measurement time in this configuration is
T.sub.total=4×(T.sub.single measurement+T.sub.scanstep) (2)
(33) Note that expression (2) is independent of the measurement area. Although only three mechanical scan steps are required, the returning (fourth) step to the initial position has been included, thereby covering the full measurement period, i.e. after T.sub.total a new measurement can immediately be performed.
(34) To show that this configuration allows considerably faster measurement cycles, table 2 gives measurement times for a few measurement area values.
(35) TABLE-US-00002 TABLE 2 Total measurement times for different areas using an array of camera modules. # camera Measurement # Time Total modules area steps per step time 1 × 1 12 × 24 mm.sup.2 2 1.5 s 3 s 1 × 1 24 × 24 mm.sup.2 4 1.5 s 6 s 1 × 2 24 × 48 mm.sup.2 4 1.5 s 6 s 2 × 2 48 × 48 mm.sup.2 4 1.5 s 6 s
(36) From table 2 the following can be concluded: For areas up to 24 mm×24 mm only one camera module is needed. For areas larger than 24×24 mm.sup.2 (i.e. starting from 24×48 mm.sup.2) the measurement time is always 6 seconds, independent of the area. Adding an additional camera increases the measurement area with 24×24 mm.sup.2.
(37) From the above it is clear that for larger areas, significant speed gains are realized by using a camera array configuration, in comparison to scanning a single camera over the area. For example, an area of 48×48 mm.sup.2 is scanned 4× faster using a 2×2 camera array (6 seconds) than using one single camera (24 seconds).
(38) Moreover, the measurement speed could for instance be further enhanced by: Recording less averaging frames Reducing the spatial resolution per frame Increasing the mechanical scan speed
(39) For example, a NdFeB motor magnet with a lateral size of 40 mm×20 mm can be measured using a 2×1 array of camera modules (measurement area of 48×24 mm.sup.2). For a fast inline measurement cycle, a spatial resolution of 0.2 mm is certainly sufficient. Usually also no averaging is required, given the strong magnetic field produced by the magnet (i.e. large signal/noise ratio). One camera shot hence only takes 0.2 s. When the mechanical scan speed is then also increased to the order of 120 mm/s, the 12 mm step is performed in 0.1 s, bringing the time for one measurement phase down to 0.3 s. The complete area is thus scanned in 4×0.3 s=1.2 seconds.
(40) This time, in the order of 1 second, is compatible with typical measurement times required in inline inspection stations. It becomes therefore a realistic option to e.g. perform a 100% inline quality control of large (motor) magnets.
(41) Note that the resulting time of 1.2 seconds above is independent of the measurement size. I.e. a larger area is measured in exactly the same time by simply adding extra camera modules.
(42) According to a second example embodiment, the device is of the line scanning type with ‘1.5-dimensional’ array of magnetic field camera modules I, II, III, as depicted in
(43) The inherent flexibility of the magnetic camera module 1 and the modular approach, allow constructing yet another magnetic field camera configuration, in which one or more camera modules can be used to function as a line scanner. Indeed, the spatial range of the sensor chip of the magnetic camera module 1 can be programmed very flexibly, whereby any sub matrix of the 128×128 sensor pixel matrix can be selected for readout. Some special cases of such sub matrices are: Any single sensor pixel (a 1×1 matrix) The full sensor array (128×128) The full range, but with half (or less) resolution Any single line in either the X or Y direction
(44) When the sensor array is programmed to only read out a single line, the functionality of a line scanner can be embodied, which can be advantageous in a number of situations.
(45) For example, one can imagine a very long magnet (such as a band magnet for sensor applications), which, in a production line, continuously passes over the magnetic field camera. It is not practical in this case to perform a stepped scanning scheme. Rather it makes sense to continuously read the same line of the magnetic field camera (which covers the complete width of the band magnet) and stitch the lines into one or multiple 2D images for further analysis.
(46) Another application would be in a production line, where individual magnets 5 continuously pass by on a conveyor belt, and where it is not desired to stop the belt for a magnetic field camera measurement. There the magnet could pass over (or under) the magnetic field camera line scanner at a constant speed and this would also generate a 2D image of the magnetic field distribution.
(47) Moreover, such scheme perfectly lends itself for full automation, since no manual magnet manipulation is needed.
(48) In this line scanner configuration, it is clear that there is no practical limit to the length (in the movement direction) of the magnet 5 to be measured. However, the question arises how one can measure a magnet that is wider than the camera's dimension (of 12.8 mm). A solution is shown in
(49) In this case a 2×1 array provides a line length of 36 mm.
(50) The ‘1.5 dimensional’ solution consists of placing a second row of camera's behind the first row, whereby both rows are shifted half a period (=12 mm) with respect to each other. As can be seen in
(51) The measurement speed of this embodiment/configuration can be calculated as follows. Supposing that the desired spatial resolution is equal in X and Y directions. This resolution in turn determines the measurement speed of one line, since it determines the number of sensors that are recorded in the line. The convention was used that the X direction is the direction along a recorded line, whereas the Y direction is the movement direction of the magnet.
(52) For example, a full (0.1 mm) resolution line consists of 128 sensor pixels, that each take about 50 microseconds, giving a total time of 6.4 ms per recorded line in the X direction. In order to obtain the same 0.1 mm resolution in the Y direction, the magnet must have moved 0.1 mm in the Y direction within the 6.4 ms timeframe. This requires a movement speed of about 16 mm/s in the Y direction. The movement speed (and hence measurement speed) can be increased by using a lower spatial resolution. For a half (0.2 mm) resolution image, one line consists of only 64 sensors that need a total time of 3.2 ms to be recorded. In this time, the magnet should move 0.2 mm, resulting in a required movement speed of 64 mm/s, which is four times faster than with full resolution. This figure of 64 mm/s can be used in most applications, since measurements at half spatial resolution are common in practice, and certainly for larger magnets.
(53) It is clear that this ‘1.5D’ configuration is easily scalable, whereby longer lines can be obtained by adding camera modules along the X-direction in both rows, without affecting measurement speed.
(54) A third example embodiment is described now, in relation with
(55) Although the planar nature of the magnetic field camera's 2D sensor array is fundamentally incompatible with the curved surface of the rotor, the solution here lies again in using the camera in line scan mode as described in the previous section. Indeed, contrary to a plane, a sensor line in the axial direction that has the Hall sensor surface perpendicular to the radial direction does measure the radial component of the field.
(56) Of course a few adaptations must be made with respect to the planar line scanner solution shown in
(57) First, instead of a magnet moving linearly over the magnetic field camera, i.e. the arrangement of a plurality of magnetic field camera modules, the rotor is mounted with its spindle in a rotatable fixture (rotatable around axis 4). The spindle is attached to e.g. a stepper motor which can accurately rotate the rotor over small incremental angles. Secondly, all magnetic field camera modules are preferably perpendicular to the radial direction R of the rotor. For a single row of camera modules R1 this is not a problem. Since there is a 2D array of sensors available on each camera module, there will always be a line that is perfectly radially oriented. This makes the positioning of the rotor also non-critical in the lateral direction (of course the height above the camera module is preferably accurately controlled using the spindle height and parallelism to the camera modules). The second (shifted) row of camera modules R2 however must be taken out of the plane of the first row R1 in order to make the cameras in the second row perpendicular to the radial direction of the rotor. This can be achieved by rotating the second row of camera modules over an angle α, for instance 90°, as shown in 5. In principle, the second row can be placed at other angles, such as 45° or 135° or 180°. In certain embodiments, care must of course be taken to place this shifted row of camera modules R2 at the same measurement distance from the rotor as the first row R1, and to place them parallel to the rotor surface in the axial direction.
(58) A measurement sequence can for instance be performed as follows: 1. The rotor angle is set to its initial position. 2. All camera modules I, II, III record one or more line scans in parallel. Multiple lines can be used for averaging. 3. The rotor is rotated over a predefined step angle. 4. All cameras are read out. 5. Etc. until the full 360° (or other angle) has been measured 6. The images of the different cameras are stitched together in order to obtain a large 2D magnetic field image. In this process, the image recorded by the second row needs to be shifted with a certain angle value relative to the images of the first row, namely the angular offset of both camera module rows (e.g. 90° in the case of
(59) Alternatively to performing the scanning in steps, the cameras can be read out continuously while the rotor is rotating at a constant speed, which is equivalent to a magnet moving under the line scanner camera arrangement at constant speed.
(60) The measurement speed of the rotor inspection configuration can be expressed in analogy to the linear line scan configuration. Here however, it is not the linear resolution in the Y direction that is relevant, but the desired angular resolution in the direction of the rotor's rotation. One line (at a resolution of 0.2 mm in the axial direction) is recorded in 3.2 ms.
(61) The complete time for a 360° scan is then equal to
T.sub.total=T.sub.line*360°/α.sub.step (3)
where
T.sub.line is the time for one line (in our case 3.2 ms) and
α.sub.step is the angular resolution (in units of degrees).
(62) The required rotation speed of the rotor is then equal to
ω.sub.rotor=360°/T.sub.total=α.sub.step/T.sub.line. (4)
(63) For a realistic resolution of e.g. 1°, the following values are obtained:
(64) T.sub.total=1.2 s
(65) ω.sub.rotor312°/s=0.9 rps.
(66) The above results show that a magnetic field image of a full rotor can be recorded in about one second. Again, this time is compatible with inline inspection requirements.
(67) It will be appreciated that a large area magnetic field camera system according to aspects of the present invention, opens up new possibilities for fast and accurate inspection of large magnets, as they are usually found in drive applications, allowing the flexibility to perform automated inline magnet inspection in production lines, automated or manual quality control and R&D for a wide variety of magnetic systems.