Compact pseudorandom scale and read head for an inductive absolute position sensor

Abstract

Electronic absolute position encoder (100), comprising: a scale (104) extending in a measuring axis direction and comprising signal modulation scale patterns, comprising: a periodic pattern (110') having a spatial wavelength Wf extending in the measuring axis direction in a periodic scale track on the scale (104); and an absolute position code pattern (150') having a code bit length Wcode in the measuring axis direction extending in the measuring axis direction in a code scale track on the scale (104), the absolute position code pattern (150') being configured to provide an absolute code range ACR, each group of N consecutive code bits uniquely identifying a corresponding absolute position within the absolute code range ACR, a detector (105) configured to be mounted proximate to the scale (104) with relative movement between the detector (105) and the scale (104) in the measuring axis direction, wherein the detector (105) comprises a field-generating configuration (112) comprising at least one conductor loop generating an alternating magnetic field, and sensor elements (157) comprising respective conductor loops that receive at least a portion of the alternating magnetic field and generate a corresponding signal dependent on the signal modulation scale pattern at an adjacent portion of the scale (104), wherein the sensor elements (157) comprise: a set of sensor elements (157) for a periodic pattern (110') arranged to be superimposed on the periodic pattern (110') and to provide a set of spatially periodic detector signals dependent on the periodic pattern (110') on the adjacent portion of the scale (104), each element of the set of sensor elements (157) for a periodic pattern (110') providing a respective periodic signal having a respective spatial phase; and M sets of sensor elements (157) for a code pattern (150') arranged to be superimposed on the code pattern (150') of the absolute position and to provide M respective sets of code detector signals, where M is an integer at least equal to two, and the M sets comprise at least first and second sets of sensor elements (157) for a code pattern (150') comprising at least N elements spaced in the measuring axis direction to sense at least N consecutive code bits on the adjacent portion of the scale (104) and to provide corresponding first and second sets of code detector signals; and where: Wcode is greater than Wf and is at most M*Wf; the sets of sensor elements (157) for a code pattern (150') are located in the measurement axis direction in respective code alignment positions configured such that, as the absolute position code pattern (150') moves in a single direction along the measurement axis with respect to the detector (105), it moves by successive alignment intervals to align or realign with each successive alignment position, and each successive alignment interval is at most Wf; and the electronic absolute position encoder (100) further comprises a signal processing configuration that inputs the M respective sets of code detector signals and determines the absolute position of the adjacent portion of the scale (104) with a coarse resolution at least as fine as Wf based on the M respective sets of code detector signals, and further determines the absolute position of the adjacent portion of the scale (104) with a fine resolution at least as fine as 0.1*Wf based at least on the spatially periodic detector signals.

Description

BACKGROUNDTechnical FieldThe invention relates generally to precision measurement instruments and, more particularly, to absolute position encoders using inductive measurement principles.General State of the ArtVarious optical, capacitive, magnetic and inductive transducers and motion or position transducers are available. These transducers use various geometric configurations of a transmitter and a receiver in a read head to measure motion between the read head and a scale. As is known, inductive sensors are one type of sensor that is highly insensitive to contamination by particles, oil, water and other fluids. US 2001 / 0 003 422 A1 discloses a generic absolute position transducer system with two elements movable relative to one another, a code track transducer and at least one fine wavelength transducer The U.S. Pat. No. 6,011,389 (patent '389), which is hereby fully adopted for reference, describes an induction current position transducer which can be used in highly accurate applications. U.S. Patents 5,973,494 and 6,002,250 all of which are hereby incorporated by reference in their entirety, describe incremental induction calipers and linear scales including signal-providing and signal-processing circuits. U.S. Patent Nos. 5,886,519, 5,841,274, 5,894,678, 6,400,138, and 8,309,906, each of which is hereby incorporated by reference in its entirety, describe absolute induction calipers and electronic measurement tapes using the induction current transducer. As described in these patents, the induction current transducer can be readily manufactured using known printed circuit technology.Various implementations of the induction current transducer (and other types of transducers) can be implemented either as incremental or absolute position transducers. In general, incremental position sensors use a scale that allows the displacement of a read head with respect to a scale to be determined by accumulating incremental units of displacement from a starting point along the scale. However, in certain applications, such as those where encoders are used in low power devices, it is more desirable to use absolute position encoders. Absolute position sensors provide a unique output signal or a combination of signals in each position (of a read head) along a scale. They do not require continuous accumulation of incremental displacements to identify a position. Thus, absolute position sensors allow for various energy saving methods, among other advantages. In addition to the aforementioned patents, U.S. Patents U.S. Pat. Nos. 3,882,482 A, 5,965,879 A, 5,279,044 A, 5,237,391 A, 5,442,166 A, 4,964,727 A, 4,414,754 A, 4,109,389 A, 5,773,820 A, and 5,010,655 A disclose various encoder configurations and / or signal processing techniques relating to absolute position encoders, and are hereby fully incorporated by reference.As used herein, the terms "track" or "scale track" generally refer to a region of the scale or scale pattern that extends in the measurement axis direction and has an approximately constant width and position in the direction transverse to the measurement axis. A scale track generally lies below and is aligned with a particular set of detectors during relative movement in the measurement axis direction. The detectors are responsive to a pattern of scale elements in the underlying scale track to generate position signals that are dependent on the detector position along the track.A common technique for encoding the absolute (ABS) position in an encoder is to use a set of binary encoded scale tracks arranged in parallel with an incremental or periodic scale track. To avoid position ambiguity and the use of the interpolated analog measurement of the incremental or periodic scale track for a high resolution measurement, the binary encoded scale track must have a spatial resolution or measurement resolution that is at least as fine as the spatial wavelength of the periodic scale track. In this case, each unique code value uniquely identifies a particular adjacent periodic wavelength. Accordingly, the least significant binary track usually has a code bit length equal to or less than the periodic wavelength of the incremental track.In compact applications where it is desired to reduce the width of the scale across the measurement axis direction, it is common to use an "N-bit" pseudo random code, which is a serial binary code in which each set of N adjacent code bits arranged in the measurement axis direction represents a unique position in the measurement axis direction (where N is an integer). (This is unlike "parallel" binary code which uses separate code tracks for each binary location that are arranged on a broad scale in a direction transverse to the measurement axis direction.). For a serial binary code, a movement in the measuring axis direction moves one bit to the next code value (comprising N bits arranged in the measuring axis direction). Similar to the binary code description above, in order to avoid position ambiguity and the use of interpolated analog measurement of the incremental or periodic scale track for high resolution measurement, it is generally necessary for the pseudo random code scale track to have a spatial resolution or measurement resolution that is at least as fine as the spatial wavelength of the periodic scale track. In this case, each unique code value uniquely identifies a particular adjacent periodic wavelength. Otherwise, the actuation may not always be reliable, or undesirably complicated signal processing at the read head may be necessary, or the like. Accordingly, a pseudo-random scale track usually has a code bit length (referred to herein as Wcode) that is equal to or less than the periodic wavelength of the incremental track (referred to herein as Wf), particularly in encoders using inductive sensor technology. Various implementations of pseudo random code systems in an inductive absolute position transducer are described in detail, for example, in previously incorporated U.S. Patent No. 5,841,274.However, the known techniques, which consist in converting binary serial code tracks (e.g. pseudo-random code tracks) into absolute position encoders, are not suitable or ideal for all applications. Users desire improvements to the aforementioned known encoder systems to provide improved combinations of compact size, measurement range, resolution, reliability of operation, reduced electronic complexity, and lower cost in various implementations. Configurations for absolute position encoders that provide such improved combinations would be desirable.SUMMARYThe present summary is provided to introduce a selection of concepts in a simplified form that are further described below in the detailed description. The summary is not intended to identify key features of the claimed subject matter, nor is it intended to be used as an aid in determining the scope of the claimed subject matter.An electronic absolute position transducer is provided, which can be implemented, for example, as a high-precision electronic digital indicator, as a micrometer, as a linear scale, as a linear measuring device, etc. It can also be adapted for various rotary implementations (in which the measuring axis direction describes a circle or an arc, for example, and the scale is a circular or curved track located on a rotary element).The encoder includes a scale extending in a measurement axis direction, a detector, and a signal processing configuration that determines an absolute position of the detector along the scale based on detector signals provided by the detector.The scale includes signal modulation scale patterns including a periodic pattern and an absolute position code pattern. The periodic pattern has a spatial wavelength Wf and extends in the measurement axis direction in a periodic scale trace on the scale. The code pattern of the absolute position has a code bit length Wcodeand extends in the measurement axis direction in a code scale track on the scale. The absolute code pattern is configured to provide an absolute code range ACR in which each group of N consecutive code bits uniquely identifies a corresponding absolute position within the absolute code range ACR.The detector is configured to be mounted near the scale with relative movement between the detector and the scale in the measurement axis direction. The detector generally comprises a field generating configuration comprising at least one conductor loop that generates an alternating magnetic field and sensor elements comprising respective conductor loops that receive at least a portion of the alternating magnetic field and generate a corresponding signal that is dependent on the signal modulation scale patterns at an adjacent portion of the scale.The sensing elements include a set of sensing elements for a periodic pattern and M sets of sensing elements for a code pattern, where M is an integer equal to at least two. The set of periodic pattern sensing elements is arranged to be superimposed on (e.g., superimposed on and facing the periodic scale track) the periodic pattern and to provide a set of spatially periodic detector signals that are dependent on the periodic pattern at the adjacent portion of the scale, each element of the set of periodic pattern sensing elements providing a respective periodic signal having a respective spatial phase. The M sets of sensor elements for a code pattern are arranged to be superimposed on the code pattern of the absolute position (e.g., superimposed on and facing the code scale track) and provide M respective sets of code detector signals. M sets of code pattern sensing elements include at least first and second sets of code pattern sensing elements including at least N elements spaced in the measurement axis direction to sense at least N successive code bits at the adjacent portion of the scale and provide corresponding first and second sets of code detector signals.The aforementioned combination of elements is further configured in accordance with the principles disclosed herein as follows:Wcode is greater than Wf; and is at most M*Wf;The sets of sensor elements for a code pattern are located in respective code alignment positions in the measurement axis direction that are configured such that as the code pattern of the absolute position moves in a single direction along the measurement axis with respect to the detector, it moves by successive alignment intervals to align with or re-align with each successive alignment position, and each successive alignment interval is at most Wf; andThe electronic absolute position transducer further includes a signal processing configuration that inputs the M respective sets of code detector signals and determines the absolute position of the adjacent portion of the scale with a coarse resolution that is at least as fine as Wf based on the M respective sets of code detector signals, and further determines the absolute position of the adjacent portion of the scale with a fine resolution that is at least as fine as 0.1*Wf based on the spatially periodic detector signals. Example operations or principles that may utilize the signal processing configuration are disclosed in more detail below.An absolute encoder system where Wcode is greater than Wf is unusual in pseudo random type absolute position encoders, particularly unusual with respect to magnetic or inductive type absolute position encoders. As noted above, Wcode has typically been made equal to or less than the incremental track periodic wavelength Wf to avoid the aforementioned positional ambiguity problem. However, according to the principles disclosed herein, the problem of position ambiguity is solved when Wcodeis greater than Wfby analyzing the signals provided by the M sets of sensor elements for a code pattern to determine the code position with respect to the detector at a resolution that is better than Wcode. This is accomplished using the aforementioned various features in accordance with the principles described in more detail below. Such an absolute encoder system offers certain advantages in certain implementations, particularly implementations where Wf is low (e.g., less than 5 mm or 2 mm or 1 mm or less) and a relatively large absolute range (e.g., several tens or hundreds of Wf) is desired, and / or implementations where reducing detector dimensions and / or bit processing complexity or size is particularly important (e.g., in small meter housings or the like). These and other advantages will be explained in more detail below.BRIEF DESCRIPTION OF THE DRAWINGSThe following are shown: FIGS. 1 and 2 are diagrams including portions of a known absolute encoder system using a first known code sensor configuration. FIG. 3 is a diagram showing certain aspects of a second known code sensor configuration that may be used in a variant of the absolute encoder system shown in FIGS. 1 and 2. FIG. 4 is a partially schematic isometric view showing certain features and parameter relationships in an absolute encoder system to explain various problems that may be associated with the prior art designs and the code sensor configurations shown in FIGS. 1-3, as opposed to certain advantages that may be provided in accordance with the various principles disclosed herein. FIG. 5 is a schematic diagram illustrating certain aspects of how a digital position code may be sensed and processed according to the principles disclosed herein to resolve the encoded position with a resolution that is more accurate than the resolution provided by the code bit length in a first implementation that uses two sets of sensor elements for a code pattern. FIG. 6 is a schematic diagram illustrating certain aspects of how a digital position code may be sensed and processed according to the principles disclosed herein to resolve the encoded position with a resolution that is more accurate than the resolution provided by the code bit length in a second implementation that uses three sets of sensor elements for a code pattern. FIG. 7 is a block diagram of example components of a signal processing configuration implementation that illustrates certain aspects of how a digital position code may be sensed and processed according to the principles disclosed herein to resolve the encoded position with a resolution that is more accurate than the resolution provided by the code bit length.DETAILED DESCRIPTIONFIGS. 1 and 2 are diagrams including portions of a known inductive type absolute encoder system 400 that uses a first known code sensor configuration. The diagrams of Figures 1 and 2 are disclosed in previously incorporated U.S. Patent No. 5,841,274 ('274 patent). In various implementations according to the principles disclosed herein, the sensor principles and signal processing taught in the '274 patent may be used in combination with the principles disclosed herein. Various aspects of the absolute encoder system 400 are briefly summarized herein. Additional related details, explanations, and teachings are available in the '274 patent.The encoder system 400 includes a read head 402 (including a detector including a field generator and sensing elements described below) mounted near a scale 404 with relative movement between the detector and the scale in a measurement axis direction MA. The scale element 404 extends along a measurement axis direction MA and includes signal modulation scale patterns. The signal modulation scale patterns include a periodic pattern or "fine wavelength scale" 418 having a spatial wavelength 304 and extending in the direction of the measurement axis MA in a periodic scale track. The receiver windings 414 and 416 of the read head 402 provide a set of periodic pattern sensing elements that function in conjunction with the field generating transmitter winding 412 to provide a "fine wavelength measurement transducer" 410. In short, the field generating transmitter winding 412 is energized with an alternating current to generate an alternating magnetic field through the receiver windings 414 and 416, which are "twisted" conductive loops that receive the alternating magnetic field and ideally generate a null output in response to the alternating magnetic field when the flux interrupters 170 of the periodic pattern 418 are not present. However, due to the signal modulation effect of the periodic pattern 418, the receiver windings 414 and 416 generate a spatially periodic signal that is dependent upon the position of the adjacent periodic pattern 418. It should be appreciated that each of the receiver windings 414 and 416 generates a respective periodic signal that has a respective spatial phase due to its different locations on the read head 402 (e.g., to provide "quadrature" signals in the depicted implementation, as is known in the art).The signal modulation scale patterns further comprise an absolute position code pattern or "binary coded scale" 458 extending in the direction of the measurement axis MA in a code scale track. The code pattern 458 has a code bit length along the measurement axis MA that approximately coincides with the edge-to-edge distance 308 in various implementations. The code bit length is sometimes referred to as code bit length Wcode in the following description.The "symmetric pairs" 457 of the read head 402 provide a set of code pattern sensing elements that function in conjunction with the field generating transmitter winding 452 to provide a "binary encoded transducer" 450. In the depicted implementation, an N-bit binary code is used, where N=8. In short, the field generating transmitter winding 452 is driven with an alternating current to generate an alternating magnetic field through the 8 symmetric pairs 457, which are "twisted" conductive loops (each including a positive polarity loop 454 and a negative polarity loop 456), which receive the alternating magnetic field and ideally generate a blank output in response to the alternating magnetic field when the flux interrupters 170 of the absolute position code patterns 458 are not present. Due to the signal modulation effect of the absolute position code pattern 458, the set of eight sensor elements for a code pattern (i.e., the set of eight symmetric pairs 457) produces a corresponding set of code detector signals that are dependent upon the N successive code bits at the adjacent portion of the absolute position code pattern or "binary code scale" 458. In the depicted implementation, binary code scale 458 includes an upper portion 459 and a lower portion 459'. As shown, for each code bit, a flow modulator or switch 170 may be located in one of the upper or lower portions and may be absent from the other portion to produce a signal of the desired polarity in the superimposed balanced pair 457. A positive voltage (e.g., above a threshold or reference level) may correspond to a logic "1" and a negative voltage may correspond to a logic "0".FIG. 2 shows an example implementation of the signal generation and processing circuit 240 connected to the read head 402. In FIG. 2, elements having the same number as in FIG. 1 may be similar or identical and should be understood based on the foregoing description. Various aspects of the signal generation and processing circuit 240 are briefly summarized herein. Additional related details, explanations, and teachings are available in the '274 patent.In short, with respect to the fine wavelength measuring transducer 410, the signal generator 250 drives the field generating transmitter winding 412 as mentioned above. The resulting signal in receiver windings 414 or 416 may be selected by switch 243 and amplified by preamplifier 245 and input to sample and hold circuit 260 and converted by A / D converter 246. Microprocessor 241 may then analyze the spatially periodic signal values and their relationship according to known methods for processing quadrature signals to provide a high resolution interpolated position measurement at a local spatial wavelength 304 indicated by "binary-coded transducer" 450 (or 450'), as will be described in more detail below.Briefly, with respect to binary coded transducer 450, signal generator 250 drives field generating transmitter winding 452 as previously discussed. In the depicted implementation, the transmitter windings 412 or 452 are selected by operation of the switch 242 under control of the digital controller 244. However, in other implementations, as indicated in the '274 patent, a single transmitter winding may be configured for use by the two transducers 410 and 450 (or 450'). In any event, the resulting eight signals in the balanced pairs 457 of the binary-coded transducer 450 (or 450') may be input to a sample-and-hold circuit 460, amplified by the preamplifiers 262, and simultaneously picked up by the sample-and-hold circuits 464. The digital controller 244 may then select the sampled signals by controlling the various switches in the sample-and-hold circuit 460 as depicted such that each signal may be converted by the A / D converter 246. The microprocessor 241 may then analyze the various code bit signal values (e.g., by comparing them to a binary signal threshold) and use predetermined logic to analyze the relationship between the code bit values to determine the corresponding absolute position codeword. The '274 patent teaches various alternative methods of using these code bit signal values and / or their logical relationships to identify a particular absolute positional relationship along the measurement axis MA with a resolution of a 1 code bit length. Therefore, it is not necessary to repeat these processes here. It is noteworthy, however, that the '274 patent teaches in any event that the resolution of the absolute position code (i.e., its code bit length) is not greater than, and preferably less than, the spatial wavelength 304 of the periodic pattern of the "fine wavelength scale" 418. Otherwise, the absolute code of a particular wavelength or period does not uniquely correspond to the fine wavelength scale 418, resulting in a possible position error or uncertainty of a wavelength, despite the resolution below the wavelength provided by the fine wavelength transducer 410. In the preferred example depicted in FIGS. 1 and 2, the code bit length or edge-to-edge distance 308 is only half of the spatial wavelength 304 to provide an error margin and uniquely associate a particular absolute position code with a particular period of the spatial wavelength 304 despite possible signal measurement errors or the like. Under the aforementioned conditions, the '274 patent teaches that a coarse resolution absolute position code can be combined with a high resolution interpolated position measurement within a particular period of spatial wavelength 304 to provide a high resolution absolute position measurement, as will be readily understood by those skilled in the art.The teachings of the '274 patent recognize one possible problem with the code sensing arrangement shown in Figures 1 and 2. In particular, if the eight symmetric pairs 457 are aligned with an intermediate position between the code elements of the binary code scale 458, as shown in FIG. 1, then the signal value of a symmetric pair 457 may yield an "empty" or undefined signal value at a transition between the code values. For example, the first, third, fourth, seventh, and eighth symmetrical pairs 457 from the left in FIG. 1 explain this state. The '274 patent teaches that a solution to this problem can be achieved by excluding certain code values corresponding to "no code value transition" and "all code value transitions" between the adjacent code bits in a codeword. In this case, it is possible to logically decode all code values along the scale despite the occurrence of various undefined signal values in various positions. However, this may be impractical or unreliable in various applications for various reasons. Thus, the '274 patent also teaches the solution discussed below with reference to Figure 3.FIG. 3 is a diagram showing certain aspects of a second known code sensing configuration usable in an absolute encoder system 800, which may be a variant of the absolute encoder system shown in FIGS. 1 and 2. The diagram of FIG. 3 is disclosed in the previously incorporated '274 patent to solve the aforementioned problem, where the signal value of one or more symmetric pairs 857 (similar or identical to the previously described symmetric pair 457) may generate an "empty" or undefined signal value at a transition between the code values (e.g., in the binary code scale 458). The periodic signal λ810may be adopted to represent the periodic spatial phase position along the periodic pattern of the fine scale 418 shown in FIGS. 1 and 2. As shown in FIG. 3, the code bit length or edge-to-edge distance 308 may be assumed to be equal to (or less than) the spatial wavelength 304. As described in the '274 patent, in the binary scale transducer shown in FIG. 3, the code sensing configuration includes a first set 821 of symmetric pairs 827 and a second set 823 of symmetric pairs 827. In particular, the first set 821 and the second set 823 are offset by half the code bit length or margin-to-margin distance 308. Thus, if one of the sets 821 or 823 is positioned at transitions between the code scale elements (as depicted in FIG. 1, for example), the other set 821 or 823 is aligned with the code scale elements. Thus, although one of the sets 821 or 823 may have undefined code signals, all the code signals are sufficiently defined in the other set 821 or 823.The '274 patent also indicates the method to be used to determine which of the sets 821 or 823 is to be used in a particular position. Basically, the '274 patent indicates that the fine wavelength transducer (e.g., the fine wavelength transducer 410 shown in FIGS. 1 and 2) may be used to determine which of the sets 821 or 823 is to be used. In the depicted example, assuming the fine wavelength 304 is equal to the code bit length or edge-to-edge distance 308, if the voltage amplitude of one of the receiver windings (e.g., receiver windings 414 or 416 shown in FIGS. 1 and 2 ) is positive, then a first of the sets 821 or 823 is to be used. On the other hand, if the amplitude of this receiver winding is negative, the other of sets 821 or 823 is to be used. The '274 patent does not teach another method for determining which of the sets 821 or 823 is to be used.FIG. 4 is a partially schematic isometric exploded view showing certain features and parameter relationships in an absolute encoder system 100 to depict various problems that may be associated with the designs and code sensing configurations shown in FIGS. 1-3 in certain implementations, as opposed to certain advantages that may be provided if the absolute encoder system 100 incorporates various features according to the principles disclosed herein. It should be appreciated that certain pattern dimensions (e.g., Wcode and 304-also referred to as Wf) are increased for purposes of explanation.The depicted absolute encoder system 100 includes a read head 102 including a signal processing circuit or configuration 140, a detector 105 including a field generator 112 and periodic pattern sensor elements 110' and code pattern sensor elements 150' (described below) mounted proximate a scale 104 with relative movement between the detector 105 and the scale 104 in a measurement axis direction MA. The scale element 104 extends in the measurement axis direction MA and includes signal modulation scale patterns. The signal modulation scale patterns include a periodic pattern 118 (only a portion of which is shown in FIG. 4) having a spatial wavelength 304 extending in the direction of the measurement axis MA in a periodic scale track 118'. The periodic pattern sensing elements 110' include the receiver windings 114 and 116 of the read head 102 that provide a set of two periodic pattern sensing elements that function in conjunction with the field generating transmitter winding 112 to provide a fine wavelength measurement as discussed above with respect to the corresponding elements in FIGS. 1 and 2. In short, the field generating transmitter winding 112 is energized with an alternating current to generate an alternating magnetic field through the receiver windings 114 and 116, which are "twisted" conductive loops (arranged similar or identical to the receiver windings 414 and 416 shown in FIGS. 1 and 2) that receive the alternating magnetic field and ideally generate a null output in response to the alternating magnetic field when the flux interrupters 170 of the periodic pattern 118 are not present. However, due to the signal modulation effect of the periodic pattern 118, the receiver windings 114 and 116 produce a spatially periodic signal that is dependent on the position of the adjacent periodic pattern 118. It should be appreciated that each of the receiver windings 114 and 116 (116 not shown) is configured to generate a respective periodic signal having a respective spatial phase due to its various locations along the read head 102 (e.g., to generate "quadrature" signals as depicted for the corresponding elements in FIGS. 1 and 2, and as is known in the art).The signal modulation scale patterns further include an absolute position code pattern 158 (only a portion of which is shown in FIG. 4) extending in the direction of the measurement axis MA in a code scale track 158'. The code pattern 158 has a code bit length Wcodealong the measurement axis MA. Code pattern sensing elements 150' may include at least first and second sets of N code pattern sensing elements to sense an N-bit codeword. Such a set, with N = 6, is shown in Figure 4. Additional sets of code pattern elements are not shown in Figure 4 for clarity of illustration. However, it should be appreciated that generally at least one second set is included (e.g., in a similar configuration to that shown in FIG. 3 that includes the first and second sets 821 and 823). Generally, the code pattern sensing elements 150' may comprise M sets of code pattern sensing elements arranged to be superimposed on the absolute code position pattern 158 and to provide M sets of code detector signals, where M is an integer equal to at least two. In various implementations, each code sensor element 157 may correspond to the symmetric pair 457 shown in FIGS. 1 and 2. The M respective sets of code pattern sensing elements are arranged to be superimposed on the absolute code position pattern 158 and function in conjunction with the field generating transmitter winding 112 to provide M sets of code detector signals which are dependent on the N successive code bits at the adjacent portion of the absolute position code pattern 158 in accordance with the aforementioned principles.Various parameter relationships in a practical implementation of the absolute encoder system 100 will now be described. A cavity or opening GHC for a meter housing is schematically illustrated in FIG. 4. It should be appreciated that in various implementations, the scale 104 and read head 102 may be necessary to be located in or sealed within the cavity GHC for a meter housing. It is often a practical requirement that the dimension of the meter housing and cavity for a meter housing be minimized (e.g., in gauges or linear gauges or the like). The scale is typically mounted on a bearing system and is guided in the measurement axis direction MA inside the cavity GHC for a meter housing. Thus, the measurement range in such a measurement device is at most the adjustment range TR shown in FIG. 4, which is the difference between the dimension of the cavity GHC for a measurement device housing in the measurement axis direction MA and the total scale length Lscale. The measurement range in such a measurement apparatus is also at most the operating range OR shown in FIG. 4, which is the difference between the total scale length Lsscaleand the necessary detector length Ldet. Thus, for a given dimension of the cavity GHC for a meter housing and scale length Lscale, decreasing the detector length Ldetincreases the usable measurement range of the meter, which is a highly desirable characteristic of a meter. In various implementations, it may be desirable that Ldet be at most 20 millimeters or 16 millimeters or even less.However, the above also assumes that the absolute code range ACR provided by the N-bit absolute code position pattern 158 is not a limiting factor. That is, it is also a condition that the absolute code position pattern of N bits 158 provides an absolute code range ACR equal to or greater than the desired usable measurement range of the measurement device. Generally, the approximate absolute code range ACR of an N-bit code of the type described herein is about ACR ≃ (2^N - 1)*Wcode, which means that increasing the number of bits in the code and / or increasing Wcode provides a means to increase the absolute code range ACR.However, in some applications it may be undesirable to increase the number of code bits N. For example, in various implementations, the connections and / or circuitry available to process code sensor signals in the signal processing configuration 140 may be limited by practical considerations such as space, cost, or allowable sample processing time. For example, in some applications it has been found that it may be desirable to use N-bit codes, where N=6 or less. Further emphasizing the problem in measurement device applications such as those illustrated in FIG. 4, it is that the spatial wavelength 304 (also referred to below as fine spatial wavelength Wf) of the periodic pattern may be limited by the need to provide high resolution and accuracy (e.g., on the order of 10 or 5 microns or even 1 micron or less). In this case, in view of practical signal interpolation steps, the spatial wavelength may be limited to 2 or 1 millimeter or even less. In this case, the sets of code detector signals must provide resolution and accuracy better than these small spatial wavelengths. The '274 patent and other known absolute encoder systems teach that this is accomplished by making Wcode equal to or less than the spatial wavelength 304 (i.e., Wf), which is generally considered to be on the order of 5 millimeters in the '274 patent. However, for a spatial wavelength Wf of 1 millimeter, this means that the teachings of the '274 patent and other known absolute encoder systems indicate that the absolute code range ACR for a 6-bit code is about 64 millimeters, which is too small for many applications and therefore generally unacceptable.Fig. 4 gives a solution to the problem mentioned above. In particular, FIG. 4 shows a dimension for Wcodethat is greater than the spatial wavelength Wf, which is contrary to the teachings of the '274 patent and other known absolute encoder systems that use magnetic or inductive sensor technology. In the particular implementation depicted in FIG. 4, Wcode ≃1.5*Wf. This particular implementation can be used with Wf=1 millimeters and increase the absolute code range ACR for a 6-bit code to approximately 96 millimeters, which is sufficient for many applications. This is, however, illustrative and not restrictive. Generally, according to the principles described in more detail below, an absolute position transducer using conductive loops as sensors as disclosed and claimed herein is configured such that Wcodeis greater than Wf(e.g., 1.25 Wf or 1.5 Wf or 2 Wf or more) and is at most M*WFwhen M sets of sensor elements are used for a code pattern. Various considerations regarding the arrangements of the M sets of sensing elements for a code pattern and related signal processing to address the deficiencies and constraints of the prior art and to enable the use of a configuration where Wcodeis greater than Wfwill be described in more detail below.FIG. 5 is a schematic representation 500 illustrating certain aspects of how a digital position code may be sensed and processed according to the principles disclosed herein to resolve the encoded position with a resolution that is more accurate than the resolution provided by the code bit length Wcode. FIG. 5 illustrates a first implementation that uses two sets of sensor elements for a code pattern (i.e., M equal to two). Each set of sensing elements for a code pattern is represented by a single one of its elements. The first set of sensing elements for a code pattern is represented by the representative sensing element of a first set RSES1. The second set of sensing elements for a code pattern is represented by the representative sensing element of the second set RSES2.According to the principles of the present disclosure, the M sets of sensor elements for a code pattern are located in respective code alignment positions (CAP) in the measurement axis direction configured such that when the code pattern of the absolute position moves in a single direction along the measurement axis with respect to the detector, it moves by successive alignment intervals to align with or re-align with each successive alignment position, and each successive alignment interval is at most Wf. This principle is represented in FIG. 5 by each code alignment position CAP 1 indicated in a reference position which is the center line of the representative sensor element RSES 1 and each code alignment position CAP 2 indicated in a reference position which is the center line of the representative sensor element RSES 2. As shown in FIG. 5, these code alignment positions differ by an alignment interval Dcap12, which is the interval over which the code pattern of absolute position 558 must move to move from an alignment on CAP1 to an alignment on CAP2. In general, a consecutive alignment interval Dcap21may also be defined, associated with the absolute position code pattern 558 moving further in the same direction from the alignment on CAP2to the alignment on CAP1. In the particular implementation depicted in FIG. 5, Dcap12is equal to Wcode / 2 and Dcap12=Dcap21. However, in various other implementations, Dcap 12 may be different than Dcap 21 as long as the two alignment intervals are not greater than Wf.It should be appreciated that in practice the representative sensor elements RSES 1 and RSES 2 and the respective sets of sensor elements they represent may be overlaid or interleaved in the measurement axis direction at a read head (e.g., similar to that depicted in FIG. 3 for sensor element sets 821 and 823). In FIG. 5, they are only separated in order to more clearly explain the operating principles described here. This also applies to the absolute position code pattern 558, which is to be understood as representing one and the same code pattern on a scale. Thus, the absolute position code pattern 558 depicted in FIG. 5 is located at the same position in the measurement axis direction MA in each of its representative instances. For purposes of explanation, the mapped absolute position code pattern 558 includes only code bit values alternating in the measurement axis direction MA. It should be appreciated that various codewords may have adjacent code bits that are the same. However, as indicated hereinabove, an absolute position code pattern should generally have at least one transition between adjacent code bits in each codeword. Therefore, the mapped absolute position code pattern 558 is useful to explain important principles with respect to such code.The mapped signal level SL 1 (or SL 2) is representative of a code signal magnitude at various relative positions between the representative sensor elements RSES 1 (or RSES 2) and the absolute code pattern 558 in the measurement axis direction MA. To more easily compare representative code signal magnitudes CSM1 and CSM2 at any position, a solid line at SL1 (or SL2) means the signal resulting from the mapped polarity of the code bits of the absolute position code pattern 558, whereas the dashed line at SL1 (or SL2) means a signal resulting from a reverse polarity of the code bits compared to those mapped by the absolute position pattern 558 and the "double" code signal magnitudes CSM1 and CSM2 are indicated therebetween. Because the alignment interval Dcap12 is equal to Wcode / 2, the sensitivities of the signal levels SL1 and SL2 of the representative sensor elements RSES1 and RSES2 are shifted by just this amount relative to each other when the code pattern of the absolute position 558 is moved in the measurement axis direction MA. An exemplary code reference position CRP is indicated in FIG. 5. In this code reference position CRP, the representative sensor element RSES1is located at the transition between the code elements and therefore outputs a symmetrical or undefined signal magnitude "Undef". The distance between these signal magnitudes Undef is Wcode, as shown. In contrast, in this code reference position CRP, the representative sensor element RSES 2 is aligned directly above a code element and outputs the largest possible signal magnitude. As shown in FIG. 5, the signal levels SL1 and SL2 experience a complete cycle between these signal magnitudes over a distance Wcode.FIG. 5 also shows that between a code position represented by reference line RL1 and a code position represented by reference line RL2, the signal magnitude of representative sensor element RSES2 is greater than that of representative sensor element RSES1, as represented by shaded larger signal magnitude region LS2 (which repeats periodically upon a change in position of Wcode, as shown by larger signal magnitude region LS2', and so forth). Similarly, between a code position represented by the reference line RL 2 and a code position represented by the reference line RL 3, the signal magnitude of the representative sensor element RSES 1 is larger than that of the representative sensor element RSES 2 as represented by the shaded larger signal magnitude region LS 1 (which periodically repeats upon a position change of Wcode as shown by the larger signal magnitude region LS 1', and so on).Accordingly, it will be appreciated that for this implementation, if signal processing is performed to determine which of the first and second sets of sensor elements (represented by representative sensor elements RSES 1 and RSES 2) has the largest signal magnitude(s), then the absolute code position may be determined at a resolution of about Wcode / 2 (for the case of ideal or near-ideal signals) as indicated by the position resolution or length in the measurement axis direction MA of the larger signal magnitude regions LS 1 and / or LS 2. Example operations or principles that may be used by the signal processing configuration to determine which set of sensor elements has the largest signal magnitude(s) will be described in more detail below.As stated above, each alignment interval (e.g., Dcap12, Dcap21) may be at most Wf (or preferably slightly less to provide a more robust margin of error) in various implementations according to the present disclosure. Accordingly, for this particular conversion, Wcodeis greater than Wfand may be at most 2*Wf, or preferably somewhat less. As previously noted, an absolute encoder system wherein Wcode is greater than Wf is unusual in pseudo random code type absolute position encoders, and particularly unusual with respect to the use of these codes in magnetic or inductive type absolute position encoders.FIG. 6 is a schematic representation 600 illustrating certain aspects of how a digital position code may be sensed and processed according to the principles disclosed herein to resolve the encoded position with a resolution that is more accurate than the resolution provided by the code bit length Wcode. FIG. 6 is substantially similar to FIG. 5 and is to be understood accordingly. Therefore, only substantial differences will be described herein.FIG. 6 is similar to FIG. 5 except that it illustrates a second implementation that uses three sets of sensor elements for a code pattern (i.e., M equal to three) rather than two sets as in FIG. 5. Similar to Fig. 5, each set of sensing elements for a code pattern is represented by a single one of its elements RSES1, RSES2 and RSES3, respectively. According to the principles of the present disclosure, the M sets of sensor elements for a code pattern are located in respective code alignment positions (CAP) in the measurement axis direction that are configured such that when the code pattern of the absolute position moves in a single direction along the measurement axis with respect to the detector, it moves by successive alignment intervals to align with or re-align with each successive alignment position, and each successive alignment interval is at most Wf. This principle is represented in FIG. 6 by the respective code alignment positions CAP1, CAP2 and CAP3 indicated in a reference position which is the center line of the respective representative sensor elements RSES1, RSES2 and RSES3. As shown in FIG. 6, CAP1 and CAP2 differ by an alignment interval Dcap12, and CAP2 and CAP3 differ by an alignment interval DcapCAP23. In this particular implementation, depicted in FIG. 6, DcapCAP12=DcapCAP23=Wcode / 3. Dcap12 is the interval by which the absolute position code pattern 558 must move to move from the orientation on CAP1 to the orientation on CAP2. Dcap23 is the interval over which the absolute position code pattern 558 must move to move from the orientation on CAP2 to the orientation on CAP3. In general, a consecutive alignment interval Dcap31 may also be defined associated with the absolute position code pattern 558 moving further in the same direction from alignment on CAP3 to re-alignment of a consecutive code bit or word on CAP1, and so on. However, in the particular implementation depicted in FIG. 6, Dcap 12=Dcap 23=Dcap 31=Wcode / 3.It should be appreciated that in practice the representative sensor elements RSES 1, RSES 2, and RSES 3 and the respective sets of sensor elements they represent may be overlaid or interleaved in the measurement axis direction on a read head (e.g., similar to depicted in FIG. 3 for sensor element sets 821 and 823), overlaying a single code pattern of absolute position 558. They are separated in FIG. 5 only for the sake of clearer illustration of the operating principles described here. It should be understood that the absolute position code pattern 558 represents one and the same code pattern on a scale.The mapped signal levels SL1, SL2 and SL3 and the indicated code signal magnitudes are similar to those depicted in FIG. 5. By having the alignment intervals DcapCAP 12 and DcapCAP 23 equal to Wcode / 3, the sensitivities of the signal levels SL 1, SL 2, and SL 3 of the representative sensor elements RSES 1, RSES 2, and RSES 3 are shifted with respect to each other by just these amounts when the absolute position code pattern 558 is moved in the measurement axis direction MA. Thus, FIG. 6 shows that between a code position represented by reference line RL1 and a code position represented by reference line RL2, the signal magnitude of representative sensor element RSES2 is greater than that of representative sensor elements RSES1 and RSES3 as represented by shaded larger signal magnitude region LS2 (which repeats periodically upon a position change of Wcode as shown by larger signal magnitude region LS2', and so forth). Similarly, between a code position represented by reference line RL2 and a code position represented by reference line RL3, the signal magnitude of representative sensor element RSES3 is greater than that of representative sensor elements RSES1 and RSES2 as represented by shaded larger signal magnitude region LS3 (which repeats periodically upon a change in position of Wcode as shown by larger signal magnitude region LS3', and so on.) Similarly, between a code position represented by reference line RL 3 and a code position represented by reference line RL 4, the signal magnitude of representative sensor element RSES 1 is larger than that of representative sensor elements RSES 2 and RSES 3, as indicated by shaded larger signal magnitude region LS 1 (which periodically repeats upon a position change of Wcode, shown by larger signal magnitude region LS 1', and so on.)Accordingly, it is to be understood that for this implementation, if signal processing is performed to determine which of the first, second and third sets of sensor elements (represented by representative sensor elements RSES1, RSES2 and RSES3) has the largest signal magnitude(s), then the absolute code position can be determined with a resolution of about Wcode / 3 (for the case of ideal or near-ideal signals), as indicated by the position resolution or length in the measurement axis direction MA of the larger signal magnitude regions LS1, LS2 and / or LS3. In various implementations, the set having the largest signal magnitude may be determined by comparing its summed absolute signal values or its summed squared signal values or the like. However, these comparison methods are only exemplary and not limiting.FIG. 7 is a block diagram 700 of the example components of a implementation of a signal processing configuration that illustrates certain aspects of how a digital position code may be sensed and processed according to the principles disclosed herein to resolve the encoded position with a resolution that is more accurate than the resolution provided by the code bit length.As shown in FIG. 7, the signal processing configuration inputs the M respective sets of N code detector signals (in the illustrated implementation, M=2, N=6) corresponding to first and second sets of code detector signals input as Set1 signal inputs and Set2 signal inputs, respectively. The signals may be generated from the first and second sets of code pattern sensing elements (e.g., Set 1 and Set 2, respectively) according to the aforementioned principles (e.g., the signals may be provided similar to those provided to sample-and-hold circuits 464, as described with reference to FIG. 2 ). The signals can then be routed on two parallel signal paths. One way leads to an array of comparators 771 which compare the signals with a digital signal reference voltage and digitize the signals into binary code signals. On the other hand, the Set1 signals are supplied to a Set1 rectifying circuit 773, and the Set2 signals are supplied to a Set2 rectifying circuit 774. The rectifier circuits 773 and 774 output the magnitudes of the Set1 and Set2 signals to a Set1 summing circuit 775 and a Set2 summing circuit 776, respectively. Set1 summing circuit 775 and a set2 summing circuit 776 output their set1 and set2 summing signals to a comparison and control circuit 777 which determines which of the signals is larger according to known methods. The comparison and control circuit 777 then outputs a switching control signal based on this determination to a switching matrix 778 and outputs a position code alignment signal based on this determination to an absolute position determination circuit or routine 779. For example, if the Set1 summation signal is larger, then the switching control signal sent to the switching matrix 778 causes it to output binary code signals corresponding to Set1 as the absolute position code value, which is then sent to the absolute position determination circuit 779. The position code alignment signal sent to the position determination circuit or routine 779 by the comparison and control circuit 777 is a signal that, in this case, indicates that Set1 has provided the absolute position code value, causing the position determination circuit or routine 779 to associate the code alignment position of the Set1 set of sensor elements for a code pattern with that absolute (Set1) position code value. On the other hand, if the Set2 summing signal is greater, then the switching control signal sent to the switching matrix 778 causes it to output binary code signals corresponding to Set2 as the absolute position code value, which is then sent to the absolute position determining circuit 779. The position code alignment signal sent to the position determination circuit or routine 779 by the comparison and control circuit 777 is a signal that, in this case, indicates that Set2 has provided the absolute position code value, causing the position determination circuit or routine 779 to associate the code alignment position of the set2 set of sensor elements for a code pattern with the absolute (Set2) position code value. In either case, the position determination circuit or routine 779 is configured to input the code absolute position value and the code orientation position of the corresponding set of sensor elements for a code pattern and determine the absolute position of the portion of the scale adjacent the detector (e.g., the detector 105 described with reference to FIG. 4 ). If the sets of sensor elements for a code pattern are configured according to the principles disclosed hereinabove, then the position determination circuit or routine 779 may determine that absolute position at a coarse resolution at least as fine as Wf based on the corresponding sets of code detector signals. Since this is sufficient to determine the absolute position within a particular period of the wavelength Wf of the periodic pattern at the adjacent portion of the scale, then the position determination circuit or routine 779 may further determine the absolute position of the adjacent portion of the scale at a fine resolution at least as fine as 0.1*Wf based on the spatially periodic detector signals according to known methods.It should be understood that although the signal processing configuration shown in FIG. 7 is intended for implementation using signals provided by two sets of sets of sensing elements for a code pattern (i.e., M=2), it can be easily adapted to process signals from additional sets of sensing elements for a code pattern by repeating the rectification and summation signal paths in implementations where M is 3 or 4 or more.It should be appreciated that the foregoing principles are applicable to various other embodiments. Various embodiments described above may be combined to provide further embodiments. All U.S. Patents referred to in the present specification are hereby incorporated by reference in their entirety. Aspects of the embodiments may be changed as necessary to use concepts of the various patents to provide still other embodiments.These and other changes may be made to the embodiments in light of the description set forth above. In general, in the following claims, the terms used are not intended to limit the claims to the specific embodiments disclosed in the specification, but are to be interpreted as including all possible embodiments along with the entire scope of equivalents to which these claims are entitled.

Claims

An electronic absolute position encoder (100) comprising: a scale (104) extending in a measurement axis direction and comprising signal modulation scale patterns comprising: a periodic pattern (110') having a spatial wavelength Wf extending in the measurement axis direction in a periodic scale track on the scale (104); and an absolute position code pattern (150') having a code bit length Wcode in the measurement axis direction extending in the measurement axis direction in a code scale track on the scale (104), wherein the absolute position code pattern (150') is configured to provide an absolute code range ACR, wherein each group of N consecutive code bits uniquely identifies a corresponding absolute position within the absolute code range ACR, a detector (105), Configured to be mounted proximate the scale (104) with relative movement between the detector (105) and the scale (104) in the measurement axis direction, the detector (105) comprising a field generating configuration (112) comprising at least one conductive loop generating an alternating magnetic field and sensor elements (157) comprising respective conductive loops receiving at least a portion of the alternating magnetic field and generating a corresponding signal dependent on the signal modulation scale pattern at an adjacent portion of the scale (104), the sensor elements (157) comprising: a set of periodic pattern sensor elements (157) (110') arranged to be superimposed on the periodic pattern (110') and to provide a set of spatially periodic detector signals, depending on the periodic pattern (110') on the adjacent portion of the scale (104), each element of the set of sensor elements (157) providing a respective periodic signal having a respective spatial phase to a periodic pattern (110'); An M set of code pattern sensor elements (157) (150') arranged to be superimposed on the absolute position code pattern (150') and provide M respective sets of code detector signals, wherein M is an integer at least equal to two, and the M sets comprise at least first and second sets of code pattern sensor elements (157) comprising at least N elements spaced in the measurement axis direction to sense at least N successive code bits at the adjacent portion of the scale (104) and provide corresponding first and second sets of code detector signals; and wherein: Wcode is greater than Wf and is at most M*Wf; the sets of sensor elements (157) for a code pattern (150') are located in respective code alignment positions in the measurement axis direction, which are configured such that when the code pattern (150') of the absolute position moves in a single direction along the measurement axis with respect to the detector (105), it moves by successive alignment intervals to align with or re-align with each successive alignment position, and each successive alignment interval is at most Wf; and the electronic absolute position transducer (100) further comprises a signal processing configuration that inputs the M respective sets of code detector signals and determines the absolute position of the adjacent portion of the scale (104) with a coarse resolution at least as fine as Wf based on the M respective sets of code detector signals, and further determines the absolute position of the adjacent portion of the scale (104) with a fine resolution at least as fine as 0.1*Wf based on at least the spatially periodic detector signals.The electronic absolute position transducer (100) of claim 1, wherein M is two, and the first and second sets of code detector signals each provide at least N code detector signals.The absolute electronic position transducer (100) of claim 1, wherein M is at least equal to three.The electronic absolute position transmitter (100) of claim 1, wherein each of the M sets of code detector signals provides at least N code detector signals.The electronic absolute position transducer (100) according to claim 1, wherein Wf is at most 2 mm, and N is at most 8.The electronic absolute position transducer (100) of claim 5, wherein M is two and Wcode is at least 1.25 Wf and at most 1.75*Wf.The electronic absolute position transducer (100) according to claim 5, wherein the detector (105) has a dimension Ldet in the measurement axis direction that is at most 20 millimeters.The electronic absolute position transducer (100) according to claim 5, wherein Wf is at most 1 mm and N is at most 6.The absolute electronic position transducer (100) of claim 1, wherein each successive alignment interval of at least one of a) is approximately equal or b) is approximately Wcode / M.The absolute electronic position transducer (100) of claim 1, wherein the detector (105) and the scale (104) are configured as an eddy current transducer that functions by generating alternating magnetic fields.

Images