Conveyor belt with magnetic markers readable by a magnetic linear encoder

Abstract

The conveyor (1) includes a conveyor belt (2) having magnetic markers (3), at least three (preferably 3 to 16) position detection units (51, 52, 52, 5N) that derive position information from the magnetic markers (3), and a control unit (6) that generates position information from sine wave signals and cosine wave signals generated by the position detection units (51, 52, 52, 5N). The conveyor belt (2) has an end junction zone (4) where the magnetic markers (3) are not present or are unreadable. Each position detection unit preferably includes a Hall sensor (511, 512). The Hall sensor generates a blank zone entry signal indicating that the position detection unit has entered the end junction zone when the maximum strength of the magnetic field of the magnetic marker falls below a threshold strength that is lower than a predetermined maximum strength, and generates a blank zone exit signal indicating that the position detection unit has exited the end junction zone when the maximum strength of the magnetic field rises above the threshold strength. The control unit ignores the position information from the position detector when it generates a blank zone entry signal, and reconsiders the position information from the position detector when it generates a blank zone exit signal. Methods for calculating the position information of the position detectors (51, 52, 52, 5N) and applying error correction are also disclosed. Single-pass and multi-pass inkjet printers equipped with the conveyor (1) of the present invention are also disclosed.

Description

[Technical Field]

[0001] The present invention relates to a conveyor belt having a periodic pattern of magnetic markers embedded in the belt aligned parallel to the direction of belt movement, a transport device including the belt, and, more particularly, to a single-pass or multi-pass inkjet printer including the belt. [Background technology]

[0002] Accurate conveyor belt position control is necessary in conveyor systems that are integrated with other machine components or devices and require the speed and position of substrates on the conveyor belt to be synchronized with the other devices, such as in pick-and-place, filling, or printing applications. The conventional method for determining the position of a conveyor belt during operation is to use a rotatable wheel in non-slip contact with the belt's carrying surface, with the moving belt rotating the wheel, and the wheel's rotations being translated into belt travel.

[0003] A single-pass inkjet printer is a specialized transport system that simultaneously prints an image onto a substrate. The primary colors (black, yellow, magenta, and cyan in the simplest CMYK model, and optionally more, such as light magenta and light cyan in more sophisticated models, sometimes up to nine primary colors) are ejected as small ink dots from the nozzles of individual printheads that are longitudinally spaced apart by some offset. The printheads with their color nozzles extend across the entire lateral width of the substrate, eliminating the need for lateral movement of the printheads. This arrangement allows all primary colors to be printed sequentially on the substrate in a single pass, and the substrate is transported from one printhead to the next, resulting in a fully printed image in all colors. Non-primary colors are formed by mixing two or more primary colors, by overlapping ink droplets of different primary colors. Therefore, an inkjet printer must be able to print ink droplets of different primary colors ejected from different longitudinally spaced printheads onto the same location on the substrate. For example, at a print resolution of 600 dpi (dots per inch), the ink droplet positioning tolerance for each color is ±21 μm. This tolerance must be maintained over the entire length of the print process area, which increases as the spacing between print heads increases and the number of primary colors increases. This requires the print driver to synchronize the ink ejection of the print heads and to accurately determine the position of the substrate as it is transported from one print head to the next, with the lowest possible tolerance and in any case without exceeding the print resolution.

[0004] Conveyor belts with magnetic markers whose position can be determined by an appropriate magnetic detector, and conveying devices using such belts, have been known for some time. Magnetic markers are north-south magnets that run along the thickness of the belt from one side to the other. A typical detector is a magnetic linear encoder, which has been commercially available for some time. The oldest document known to the applicant in this field is Japanese Patent Application Publication No. 55-123806. This document, which appears to have already been published in the field of printers, discloses the use of magnetic markers on the belt and two magnetic heads to determine the belt's position, presumably simply to generate "quadrature signals" (see below). Other documents disclosing belts with magnetic markers and their use in printers include Japanese Patent Application Publication No. 2001-125333, U.S. Patent Application Publication No. 2008 / 049054, and U.S. Patent Application Publication No. 2008 / 0192076.

[0005] The four aforementioned documents relate to endless belts, i.e., belts that do not require end splicing. However, the majority of belts are sold in an open-end form so that, after installation on the conveying device, the belt can be cut to the required length and pretensioned to the desired tension. Only after these steps are performed is the open-end belt installed and made endless, while still maintaining the required pretension. Such end splicing is typically performed using heat and pressure, optionally in conjunction with a hot melt adhesive. In the case of belts with magnetic markers, such end splicing can lead to deterioration or destruction of the magnetic markers, making them unreadable by a magnetic linear encoder in the end splicing zone.

[0006] EP 4060291, which has not been previously published, discloses an apparatus for determining position on a support with a periodic magnetic marker pattern. The apparatus includes two linear encoders, one "downstream" and one "upstream." The apparatus can address "transition regions" where the magnetic signal is unreliable or absent by making the "reading window" (the distance between the two linear encoders) larger than the "transition region." EP 4060291 relies on a calculated displacement to address a single "transition region." However, the document is silent on how this should work when there are multiple "transition regions" of different sizes.

[0007] Despite the fact that the maximum magnetic flux density of the marker's sinusoidal magnetic signal gradually decreases to zero rather than abruptly decreasing to zero upon entering the end junction zone, the magnetic linear encoder's determination of whether the marker is "readable" or "unreadable" is a yes / no decision. At some point, the belt position determination system must decide to completely ignore the linear encoder's output when it enters the end junction zone where the sinusoidal signal strength decreases and its output position information becomes unreliable. Similarly, at some point, the system must decide to begin reconsidering the linear encoder's output when it leaves the end junction zone and the sinusoidal signal strength increases sufficiently again that its output position information becomes reliable.

[0008] The end junction zone, which contains markers that cannot be read by magnetic linear encoders, typically has a length of 80 to 120 mm. Nevertheless, position determination using magnetic markers must be possible at any time and at any belt position. A conventional initial solution is to use two linear encoders separated from each other by a longitudinal distance greater than the longitudinal length of the end junction zone. This method ensures that there is always at least one linear encoder that is not located in the end junction zone and can provide position information from readable magnetic markers. The applicant is aware of several prior art techniques that attempt to address the above.

[0009] JP 2006-096429 A discloses belt position determination using optical (transparent / opaque) markers on the belt. The belt has a "knot 141x" (a seemingly transparent zone wider than a single marker) and uses two detectors and an "origin tab" that is detected by an "origin detector." The signal is processed by a CPU.

[0010] US Patent Application Publication No. 2012 / 124848 relates to position sensing using a code containing absolute position values, the code having at least one discontinuity and using at least two sensors. This document therefore does not rely on markers being counted by sensors, but on directly read position values and, if necessary, correcting for missing position values at discontinuities.

[0011] U.S. Patent Application Publication No. 2012 / 124849 relates to determining the angular position of a rotating shaft using magnetic markers. In this case, the magnetic markers are arranged as an open-ended strip around the shaft, and the ends of the strip are then mechanically joined. The end joint forms a magnetic marker-free zone. To sense that zone, markers are placed near the boundary of the end joint at a spatial frequency (referred to herein as a "position frequency") that is different from the spatial frequency along the remainder of the strip. This allows a linear encoder to sense the proximity of the end joint zone.

[0012] WO 2016 / 146463 addresses the problem of determining position on a magnetically marked belt, where the belt has unreadable magnetic markers in its end junction zones and uses two linear encoders. To this end, the document uses a fiducial marker and two associated fiducial marker detectors, which can switch from the first linear encoder to the second linear encoder, or vice versa, depending on which of the two linear encoders is in the end junction zone. The document also discloses a single-pass inkjet printer using such a belt.

[0013] The present invention aims to provide improved position determination in belts having magnetic markers, said belt having end abutment zones where the magnetic markers are not readable by the position determination system. Summary of the Invention

[0014] Therefore, the present invention provides a) The belt periphery, the outer conveying surface, the inner pulley-facing surface, and the length L jand magnetic markers along the entire periphery of the belt except for one or more blank zones, j being an index of the blank zone, the magnetic markers being arranged in the direction of belt movement with alternating north and south poles of the magnetic markers facing the outer conveying surface, adjacent north and south poles being spaced from each other by a pole pitch DP to generate an alternating magnetic field of a predetermined maximum strength, and in each of the blank zones the magnetic markers are either absent or generate a magnetic field of a maximum strength less than the predetermined maximum strength; b) at least two position detectors arranged along the belt periphery in spatial proximity to the magnetic markers, the position detectors detecting magnetic fields from magnetic markers not within the blank zone when the conveyor belt travels in the direction of travel, and generating 1) a sine wave voltage signal corresponding to the angular position of the position detector relative to the pattern with respect to a direction of magnetic flux density B detected from the magnetic markers, and 2) a cosine wave voltage signal corresponding to the angular position of the position detector relative to the pattern, wherein there are at least two spaced apart position detectors among the at least two position detectors, and one of the two spaced apart position detectors detects a distance of any blank zone length L from any one of the remaining spaced apart position detectors. j at least two position sensing units spaced apart by a circumferential distance in a belt movement direction greater than c) means for generating a blank zone entry signal when the maximum strength of the magnetic field detected by one position detection unit drops or is about to drop to a threshold strength that is lower than the predetermined maximum strength, and means for generating a blank zone exit signal to the position detection unit when the maximum strength of the magnetic field detected by the same position detection unit rises or rises to the threshold strength; d) a control unit that derives belt position information from the sine wave voltage signals and cosine wave voltage signals of all position detection units when the conveyor belt travels in the moving direction, the control unit ignoring the sine wave voltage signals and cosine wave voltage signals of any position detection unit from the time when the blank zone entry signal for that position detection unit is received, and reconsidering the sine wave voltage signals and cosine wave voltage signals of that position detection unit from the time when the blank zone departure signal for that position detection unit is received; A conveyor comprising:

[0015] Preferred embodiments of the inventive conveyor are according to the dependent claims.

[0016] The present invention also provides a single-pass or multi-pass ink jet printer comprising the conveyor of the present invention. [Brief explanation of the drawings]

[0017] [Figure 1] 1 is a diagram showing a preferred embodiment of a position detector used in the present invention; [Figure 2] 1A and 1B show two embodiments of the conveyor of the present invention; [Figure 3] 1A and 1B show two embodiments of the conveyor of the present invention; [Figure 4] 1A and 1B show two embodiments of the conveyor of the present invention; [Figure 5] FIG. 2 is a diagram showing the structure of a preferred embodiment of a control unit used in the present invention; [Figure 6] 1 is a schematic diagram of a printer equipped with a conveyor of the present invention; [Figure 7] FIG. 10 shows the results of a synchronization test using a conveyor of the present invention compared to a prior art conveyor using a commercially available linear encoder for position determination. [Figure 8] 10A and 10B are diagrams illustrating the results of performing phase error correction on input signals of sine and cosine waves containing phase errors. DETAILED DESCRIPTION OF THE INVENTION

[0018] For the purposes of this invention, the following definitions apply: The "periphery" of an endless belt essentially corresponds to the longitudinal length of the open-end belt or tape from which the endless belt is made by end splicing. The periphery of an endless belt typically runs along the "pitch line" or "neutral line" of the endless belt, i.e., along the interior line of the endless belt that does not change length when the endless belt is bent over pulleys during use. The term "periphery," as is customary in the art, refers to both the physical peripheral lateral edges of the endless belt and the associated length L of such peripheral lateral edges. End splicing to make the belt or tape endless is typically performed as a final step, typically using a hot press and, optionally, a hot-melt adhesive.

[0019] A "magnetic marker" suitable for this invention forms an alternating S / N pole pattern when viewed from one side of the belt. When viewed from the other side of the belt, the magnetic pole pattern is reversed because for each N pole on one side, there is one S pole on the other side. The distance between the magnetic flux density maxima of adjacent S / N poles is half the period length of the magnetic marker pattern, commonly referred to as the "pole pitch." If a magnetic sensing device is moved along a path of alternating N / S poles and senses the magnetic field from a distance, the sensed signal will be sinusoidal rather than nearly rectangular.

[0020] For the purposes of the present invention, this pole pitch is referred to as DP. For the purposes of the present invention, the DP may be any value, but is preferably in the range of about 0.5 to about 10 mm, and more preferably about 0.5 mm, about 1 mm, about 2 mm, about 2.5 mm, about 5 mm, or about 10 mm. The term "about" does not refer to the variation in the nominal value of DP itself, but rather reflects the variation due to expansion or contraction of the conveyor belt included in the conveyor of the present invention during operation. Thus, the DP may deviate from the respective nominal values by up to +2% due to expansion of the conveyor belt, and may deviate from the respective nominal values by up to -2% due to contraction of the conveyor belt. These deviations are suitable for the definition of the term "about" as used herein.

[0021] The pattern of magnetic markers is preferably essentially periodic in the direction of belt movement. "Essentially periodic" preferably means that the standard deviation of the geometric dimensions of the magnetic markers (such as length in the belt direction, width transverse to the belt movement direction, and / or thickness in the belt thickness direction) and / or the standard deviation of the magnetic flux density B generated by them at a predetermined fixed distance from the belt surface in the belt thickness direction does not exceed 50% of the mean value of said geometric dimensions or said mean magnetic flux density, respectively. More preferably, "essentially periodic" means that said standard deviation does not exceed 10% of the respective mean values. Most preferably, "essentially periodic" means periodic.

[0022] A "blank zone" is a zone on a conveyor belt where magnetic markers are absent, destroyed, or erased, or where magnetic markers outside the blank zone have a low magnetic flux density B relative to a predetermined maximum magnetic flux density B. A typical example of a blank zone in the above sense is an end joining zone of a conveyor belt, where the magnetic markers are damaged or destroyed by the action of heat and pressure during the end joining process. Another example of a blank zone is a zone where magnetic markers are identified as defective and all such defective markers are intentionally erased to make it clear that no position information can be obtained from the position detection unit. A conveyor belt used in the present invention may include one or more such blank zones, and typically includes exactly one such blank zone in the form of the end joining zone.

[0023] A "Hall sensor" detects the magnetic flux density B that is applied to the Hall effect and generates an output voltage proportional to B.

[0024] A "magnetoresistive sensor" senses the direction (tilt) of magnetic flux density B relative to the direction of the pattern of magnetic markers on the conveyor belt by anisotropic magnetoresistance (AMR), giant magnetoresistance (GMR), or tunneling magnetoresistance (TMR), preferably by AMR for the purposes of the present invention. In an AMR sensor, the resistivity ρ(α) of the magnetoresistive material depends on the angle α between the direction of current flow I in the magnetoresistive material (which remains constant in the AMR sensor) and the direction of magnetic flux density B (which varies depending on the position of the AMR sensor relative to the magnetic marker pattern), and is preferably given by:

[0025]

number

[0026] where ρ0 is the resistivity when B is perpendicular (perpendicular) to I, and ρ P is the resistivity when B is parallel to I, and usually, ρ0 is PFor GMR or TMR sensors, the fixed reference magnetic field B ref The resistance ρ(α) of a thin insulating material sandwiched between a reference layer, which is magnetized at θ and a free layer, which can be magnetized in any direction by an external magnetic field B (which varies depending on the position of the GMR or TMR sensor relative to the pattern of magnetic markers), is given by B ref and B, and is preferably given by the following equation:

number

[0027] where ρ P is B and B ref is the resistivity when ap is B and B ref is the resistivity when the two are antiparallel. P is minimum and ρ ap is a maximum. Magnetoresistive materials for AMR, GMR, and TMR sensors are known in the art.

[0028] The "position sensing unit" senses the angular position of the magnetic markers within a pattern that is preferably periodic in nature in the sense described above. Thus, typical outputs of a position sensing unit located near a magnetic marker, when the belt runs in the direction of movement, are: 1) a sinusoidal voltage signal related to the direction of the magnetic flux density B sensed from the magnetic marker, corresponding to the angular position of the position sensing unit relative to the pattern; and 2) a corresponding cosine voltage signal corresponding to the angular position of the position sensing unit relative to the pattern. The position sensing unit, in the sense described above, preferably comprises a first Wheatstone bridge including two half-bridges, each half-bridge formed of an upper leg and a lower leg, each half-bridge being constructed of resistors that are magnetoresistive sensors in the sense described above, the first Wheatstone bridge being capable of generating the sinusoidal signal, and a second Wheatstone bridge of the same structure being capable of generating the cosine signal. The two Wheatstone bridges may each include an upper leg and a lower leg composed of multiple magnetoresistive sensors in the sense described above that can be electrically switched in series and / or parallel with each other, but with different spatial orientations to enable error correction in the magnetic signal B from the magnetic marker, as is known in the art. For example, if the magnetoresistive sensors are AMR sensors, see U.S. Patent Application Publication No. 2017 / 322052. When the magnetoresistive sensors constituting the Wheatstone bridge are AMR sensors, the angular position (in radians) is within a full period of 2π corresponding to the pole pitch DP and is half the distance between two adjacent north poles (or two adjacent south poles) of the magnetic marker pattern. When the magnetoresistive sensors constituting the Wheatstone bridge are GMR or TMR sensors, the angular position (in radians) is within a full period of 2π corresponding to 2DP and corresponds to the full distance between adjacent north poles (or adjacent south poles) of the magnetic marker pattern. The first Wheatstone bridge for generating a sine wave signal and the second Wheatstone bridge for generating a cosine wave signal preferably have the following characteristics: a) When the magnetoresistive sensor included in the Wheatstone bridge is an AMR sensor a1) The upper legs of the first and second half bridges of the first Wheatstone bridge are separated from each other by a total pole pitch, i.e., a distance DP, and the lower legs of the first and second half bridges are separated from each other by the same distance DP, measured from the geometric centers of the magnetoresistive sensors that make up those legs. The same distance applies to the upper legs of the first and second half bridges of the second Wheatstone bridge and the lower legs of the first and second half bridges of the second Wheatstone bridge. a2) The first and second Wheatstone bridges are separated from each other by a spacing distance that is one-quarter of the pole pitch of the magnetic markers on the belt, i.e., DP / 4, and the spacing distance is measured from the geometric center of the magnetoresistive sensor in the lower leg of the first half bridge of the first Wheatstone bridge to the geometric center of the magnetoresistive sensor in the lower leg of the first half bridge of the second Wheatstone bridge. b) When the magnetoresistive sensor included in the Wheatstone bridge is a GMR or TMR sensor b1) Same as a1) above. b2) Same as a2) above, except that the spacing between the first and second Wheatstone bridges is half the pole pitch of the magnetic markers on the belt, i.e., DP / 2. When these characteristics are met, a position sensing unit is positioned near or spatially adjacent to the pattern of magnetic markers in the belt such that the direction of the pattern of magnetic markers, and therefore the direction of belt movement, is parallel to the spacing distance, and the conveyor belt runs in the direction of movement, and the first Wheatstone bridge generates a sinusoidal voltage signal of a predetermined magnitude as the differential voltage between its midpoints, and the second Wheatstone bridge generates a corresponding cosine voltage signal of the same magnitude within experimental error as the differential voltage between its midpoints.

[0029] However, the position sensing unit preferably includes first and second pairs of Wheatstone bridges as described above, the first pair separated by the spacing distance to generate sine and corresponding cosine signals, and the second pair separated by the spacing distance again to generate sine and corresponding cosine signals from an alternating S / N magnetic pole pattern. The two pairs of Wheatstone bridges are, in turn, separated by a full DP, the spacing measured from the geometric center of the magnetoresistive sensor in the lower leg of the first half bridge of the first Wheatstone bridge of the first pair of Wheatstone bridges to the geometric center of the magnetoresistive sensor in the lower leg of the first half bridge of the first Wheatstone bridge of the second pair of Wheatstone bridges. Furthermore, one of the two pairs of Wheatstone bridges has a power supply whose polarity is reversed relative to the polarity of the power supply of the other pair of Wheatstone bridges. Due to the precise DP spacing between the two pairs of Wheatstone bridges, combined with their reversed power supply polarity, the second pair of magnetoresistive sensors in the Wheatstone bridge simultaneously senses the north magnetic pole when the first pair of magnetoresistive sensors in the Wheatstone bridge is over the actual north magnetic pole and sensing the south magnetic pole. Similarly, the second pair of magnetoresistive sensors in the Wheatstone bridge simultaneously senses the south magnetic pole when the first pair of magnetoresistive sensors in the Wheatstone bridge is over the actual north magnetic pole. Similar considerations apply to the first and second pairs of magnetoresistive sensors. Summing and optionally averaging the signal from the first Wheatstone bridge in the first pair of Wheatstone bridges with the signal from the first Wheatstone bridge in the second pair of Wheatstone bridges can remove systematic differences in the magnetic flux density B and / or spatial dimensions between the north and south magnetic poles, providing a summed or averaged sine wave (respectively cosine wave) signal.Similarly, the signal from the second Wheatstone bridge of the first pair of Wheatstone bridges and the signal from the second Wheatstone bridge of the second pair of Wheatstone bridges can be summed and optionally averaged to again remove systematic differences in the magnetic flux density B and / or spatial dimensions between the north and south poles and provide a summed or averaged corresponding sinusoidal (respective corresponding sinusoidal) signal.

[0030] For purposes of this invention, a "sinusoidal signal" refers to an analog signal S (before conversion to a digital signal by an ADC), typically in units of volts or amperes, that can be represented as a discrete Fourier spectrum in time:

[0031]

number

[0032] where C, A1, A i (i=2, 3, ...N) are constants with the same units as S, φ=2πf has units of 1 / s and is the fundamental angular velocity of the Fourier spectrum, and γ1, γ i (i=2, 3, ...N) is the angular offset in radians. N is the largest integer and can theoretically go up to infinity, but in practice it is usually in the range of 10 to 100, more precisely, e.g., 10, 20, 50, or 100. In this definition, the magnitude of A1 is determined by the coefficient A i and / or γ1 and γ i (i=2, 3, ...N) are all equal within experimental error. Most preferably,

number

[0033] The conveyor of the present invention also includes means for generating a blank zone entry signal when the maximum strength of the magnetic field detected by the position detection unit drops or is about to drop to a threshold strength lower than the predetermined maximum strength, instructing the control unit to begin ignoring the position signal of the position detection unit.

[0034] The blank zone entry signal can be derived from the fact that the strength of the magnetic signal detected by the position detection unit itself decreases upon entering the blank zone, which is one example of a means by which the blank zone entry signal can be generated when the maximum strength of the magnetic field detected by the position detection unit falls below a threshold strength.

[0035] However, from the viewpoint of position accuracy, it is preferable to have additional means capable of generating such a blank zone entry signal, and this additional means is arranged upstream of the corresponding downstream position detection unit, so that the upstream additional means generates the blank zone entry signal before the corresponding downstream position detection unit itself enters the blank zone. This is an example of a means capable of generating a blank zone entry signal when the maximum strength of the magnetic field detected by the position detection unit is about to drop below the threshold strength.

[0036] For the additional means capable of generating a blank zone entry signal when entering the blank zone, the term "start signal generating means" is also used in this specification. The additional means capable of generating a blank zone entry signal when entering the blank zone, i.e. the start signal generating means, is typically a Hall sensor in the above sense.

[0037] The conveyor of the present invention may further comprise, in addition to each start signal generating means, a corresponding timer for allowing the lapse of time after which the blank zone exit signal is generated and the position signal of the position detector is subsequently reconsidered by the control unit (embodiment A). In the case where the blank zone entry signal is generated by the position detector itself (embodiment A1), this time period t satisfies the following inequality:

[0038]

number

[0039] Here, L is a length greater than the maximum length of the blank zone present on the belt. u is the distance from the end of the blank zone to the start of the next blank zone upstream of the blank zone (in the direction of belt movement). If there is only one blank zone, the "blank zone in question" and the "next blank zone upstream" refer to one and the same blank zone, and D u refers to the entire belt circumference excluding the length of the only blank zone present. Recall that the belt of the conveyor of the present invention is endless, and the distance is actually the peripheral distance of the belt loop. Alternatively, if the blank zone entry signal is generated by the upstream start signal generating means (embodiment A2), the time period satisfies the following inequality:

[0040]

number

[0041] where D m is the distance between the upstream start signal generating means and the associated downstream position sensing element, and all other symbols have the same meaning as in the previous inequality.

[0042] Alternatively, the conveyor of the present invention may further include, in addition to each means capable of generating a signal indicating entry into a blank zone, corresponding additional means capable of generating a blank zone exit signal when leaving the blank zone (Embodiment B). These means capable of generating a blank zone exit signal may be the position detection unit itself. This is an example of a means capable of generating a blank zone exit signal when the maximum strength of the magnetic field detected by the position detection unit rises above the threshold strength. There may also be additional means capable of generating a blank zone exit signal when leaving the blank zone. Each such additional means is located downstream of the corresponding upstream position detection unit, and the downstream additional means generates a blank zone exit signal after the corresponding upstream position detection unit leaves the blank zone. This is an example of a means capable of generating a blank zone exit signal when the maximum strength of the magnetic field detected by the position detection unit rises above the threshold strength. The additional means capable of generating a blank zone exit signal when leaving the blank zone is referred to herein as an "end signal generating means." The control unit begins to reconsider the position information from a given position detection unit upon receiving a blank zone exit signal from the corresponding end signal generating means.

[0043] When the blank zone entry signal is generated by the position detection unit itself, preferably the blank zone exit signal is also generated by the position detection unit itself (embodiment B1). These position detection units are also examples of means capable of generating a blank zone exit signal when the maximum strength of the magnetic field detected by the same position detection unit rises above the threshold strength.

[0044] However, when the blank zone entry signal is generated by a start signal generating means such as a Hall sensor, preferably the blank zone departure signal is also generated by an end signal generating means such as a Hall sensor (embodiment B2), or the additional means may generate both the blank zone entry signal and the blank zone departure signal (embodiment B3).The additional means capable of generating a blank zone entry signal when entering a blank zone and generating a blank zone departure signal when leaving a blank zone is referred to in this specification as a "start / end signal generating means."

[0045] Preferably, the conveyor of the present invention is capable of detecting both the start and end of the blank zone (e.g., embodiments B, B1, B2, B3, etc.). For this purpose, the conveyor of the present invention is provided with both a start signal generating means and an end signal generating means, and more preferably, both the start signal generating means and the end signal generating means are Hall sensors (as in embodiment B2).

[0046] When the start and end of a blank zone are detected by the position detection units themselves (as in embodiment B1 above), each position detection unit generates its own blank zone entry signal when it enters a blank zone (as detectable by a weakening or disappearing magnetic signal) and its own blank zone exit signal when it leaves the blank zone (as detectable by a reappearance or re-increase in magnetic signal). The control unit ignores position signals from any position detection unit during the period between receiving the blank zone entry signal and receiving the blank zone exit signal. Of all the position detection units present, there are typically at least two, and preferably at least three (preferably adjacent to each other, although this is not required), and any two of them are spaced apart in pairs by a distance greater than the maximum length of the blank zone present on the belt. Position detection units that satisfy this condition are referred to throughout this application as "spaced position detection units."

[0047] When the conveyor of the present invention is equipped with a start signal generating means and an end signal generating means, and particularly when both of these start signal generating means and end signal generating means are Hall sensors (as in the above embodiment B2), there is one start signal generating means and one end signal generating means for each position detecting unit, and each start signal generating means is located upstream of a corresponding one of the position detecting units, and each end signal generating means is located downstream thereof, and the start signal generating means and the end signal generating means are arranged to sandwich the corresponding position detecting unit. For at least two required spaced apart position detecting units, each of them has a length L of any blank zone in the belt moving direction. j and each of these spaced position sensing units also has an associated start signal generating means and an end signal generating means, and any pair of two (preferably, but not necessarily, adjacent) upstream and downstream spaced position sensing units is more preferably spaced apart from the start signal generating means of the downstream position sensing unit of the pair by a peripheral distance greater than the maximum length L of any blank zone present on the belt. j The control unit is configured to disregard the position information from a given position detection unit upon receiving a blank zone entry signal from the corresponding start signal generating means (when the corresponding first additional means has entered the blank zone), and to start reconsidering the position information of that position detection unit upon receiving a blank zone exit signal from the associated end signal generating means (when the corresponding end signal generating means has left the blank zone). This embodiment B2 is considered to be the most preferred.

[0048] When the conveyor of the present invention includes start / end signal generating means, particularly when the start / end signal generating means is a Hall sensor (embodiment B3 above), there may be exactly one start / end signal generating means for each position detector, the position detectors and start / end signal generating means being arranged alternately along the belt periphery in the longitudinal direction of the conveyor belt, i.e., in the direction of belt movement, with each position detector being sandwiched between two adjacent start / end signal generating means. There may also be at least two spaced apart position detectors, preferably at least three spaced apart position detectors, with any pair of two (preferably, but not necessarily, adjacent) upstream and downstream position detectors being spaced apart from each other by a distance L1 greater than the maximum length of any blank zone present on the belt. In this embodiment B3, any one of the start / end signal generating means generates a blank zone entry signal for a position detection unit downstream of the start / end signal generating means (when the start / end signal generating means enters the blank zone) (not necessarily, but preferably adjacent to it), and when the start / end signal generating means leaves the blank zone, the start / end signal generating means generates a blank zone exit signal for that same position detection unit.

[0049] In embodiment B31 within embodiment B3, the control unit ignores position information from a predetermined position detection unit when it receives a blank zone entry signal from the corresponding upstream start / end signal generation means (when the corresponding start / end signal generation means enters the blank zone), and starts to reconsider the position information of the position detection unit when it receives a blank zone departure signal from the corresponding downstream start / end signal generation means (when the corresponding start / end signal generation means leaves the blank zone). This embodiment B31 is similar to embodiment B2 except that the start signal generation means and end signal generation means are not separate, but rather there is a start signal generation means and an end signal generation means.

[0050] Also, in embodiment B32 of embodiment B3, there is one first start signal generating means that is located upstream of all the position detecting units and that can generate a blank zone entry signal for a downstream (not necessarily, but preferably adjacent) position detecting unit when entering a blank zone; one second start signal generating means that is located downstream of all the position detecting units and that uses (interprets) the blank zone entry signal generated by the start signal generating means as a blank zone departure signal for an upstream (not necessarily, but preferably adjacent) position detecting unit when entering a blank zone; and one less than the number of the position detecting units. and a number of further start signal generating means sandwiched between an upstream (not necessarily adjacent, but preferably adjacent) position detecting unit and a downstream (not necessarily adjacent, but preferably adjacent) position detecting unit, wherein the blank zone entry signal generated by each further start signal generating means is used (interpreted) as a blank zone entry signal for said downstream (not necessarily, but preferably adjacent) position detecting unit, and is used (interpreted) as a blank zone exit signal for said upstream (not necessarily, but preferably adjacent) position detecting unit. In this embodiment B32, any of the further start signal generating means is spaced a distance L from the corresponding upstream position detecting unit that generates the blank zone entry signal (the blank zone exit signal that is used / interpreted). m is greater than the maximum length of the blank zone present on the conveyor belt. Also, the second start signal generating means generates a blank zone entry (used / interpreted blank zone departure) signal by a distance L from the adjacent upstream position detecting unit. nIn this embodiment B32, the control unit ignores position information from the most upstream position detection unit from the time when it receives a blank zone entry signal from the first start signal generating means, ignores position information from any of the other position detection units from the time when it receives a blank zone entry signal from a further start signal generating means upstream of (not necessarily, but preferably adjacent to) the position detection unit from the time when it receives a blank zone entry signal from a further start signal generating means upstream of (not necessarily, but preferably adjacent to) the position detection unit from the time when it receives a blank zone entry (used / interpreted blank zone departure) signal from the second start signal generating means from the time when it receives a blank zone entry (used / interpreted blank zone departure) signal from a further start signal generating means downstream of (not necessarily, but preferably adjacent to) the position detection unit from the time when it receives a blank zone entry (used / interpreted blank zone departure) signal from a further start signal generating means downstream of (not necessarily, but preferably adjacent to) the position detection unit from the time when it reconsiders position information from any of the other position detection units.

[0051] In any of the conveyor embodiments of the present invention in which the blank zone entry signal and / or blank zone exit signal are generated by the position sensing unit itself (such as in embodiments A1 and B1), the blank zone entry signal and / or blank zone exit signal can be derived from the phase (i.e., sum of squares) of the sine and cosine signals of the position sensing unit, or from the square root of the sum of squares. The sum of squares is also constant within experimental error, and its square root is equal to the predetermined maximum amplitude of the sine and cosine signals within experimental error. However, if one of the two magnetoresistive sensors of the position sensing unit enters the blank zone of the belt and generates a signal with reduced amplitude, the sum of squares or its square root of the two signals will be significantly lower than the predetermined amplitude. A significant drop in intensity, typically 0.4 to 0.9 times the predetermined amplitude, indicates that the position sensing unit has entered the blank zone and triggers the generation of the blank zone entry signal. On the other hand, if the intensity of the sum of the squares of the two signals or the intensity of its square root rises again to more than 0.9 times or substantially 1.0 times the predetermined intensity, this indicates that both magnetoresistive sensors of the position detection unit are again outside the blank zone, triggering the generation of the blank zone exit signal.

[0052] Also, in any embodiment of the conveyor of the present invention in which the generation of the blank zone entry signal and the blank zone exit signal is performed by additional means such as Hall sensors, and each position detection unit is sandwiched between two such additional means (as in embodiments B2, B3, B31, and B32), the phase may be obtained from the signals of the two additional means sandwiching the position detection unit. The evaluation of the blank zone entry signal and the blank zone exit signal from the decrease and re-increase in phase, respectively, is as described above. The pinch start signal generating means and end signal generating means (embodiment B2), the pinch start / end signal generating means (embodiment B31), or the pinch start signal generating means (embodiment B32) are spaced from each other by a distance LH given by the following equation:

[0053]

number

[0054] Here, N is a positive integer at least equal to 0, and DP is as defined above. However, N is large enough to accommodate the position detection unit to be pinched. The intensities of the signals of the pinch start signal generating means and end signal generating means (embodiment B2), the pinch start / end signal generating means (embodiment B31), or the pinch start signal generating means (embodiment B32) are sine wave signals and corresponding cosine wave signals, respectively, that enable the phase (the sum of the squares of their intensities) to be formed therefrom, taking into account the defined interval LH. With such phase-derived blank zone entry signals and blank zone departure signals, the control unit ignores position information of any position detection unit from the time it receives a phase-derived blank zone entry signal from the phase generating additional means that pinches the position detection unit, and reconsiders the position information of the position detection unit from the time it receives a phase-derived blank zone departure signal from the phase generating additional means that pinches the position detection unit.

[0055] In all embodiments of the conveyor of the present invention, all position sensing elements and, if present, all additional means capable of generating a blank zone entry signal or a blank zone exit signal are preferably located near, on or below that part of the periphery of the conveyor belt that is not folded over or in contact with the pulleys.

[0056] A position detection unit suitable for the most preferred embodiment B2 (whether or not it uses phase-derived blank zone entry and exit signals) comprises a start signal generating means and an end signal generating means as a combined unit, optionally arranged in one and the same housing, as shown in the left or right part of Fig. 1. The combined unit shown in the left part of Fig. 1 comprises a start signal generating means 511 and an end signal generating means 512 in the form of Hall sensors, and two Wheatstone bridges 513, 514 (whose legs are made up of resistors of AMR sensors). The units have a spatial arrangement such that the start signal generating means 511 (e.g., a Hall sensor) is located outside the first Wheatstone bridge 513, and the end signal generating means 512 (e.g., a Hall sensor) is located outside the second Wheatstone bridge 514. The start signal generating means 511 and the end signal generating means 512 sandwich Wheatstone bridges 513, 514 between them, forming a start signal generating means 511 (e.g., Hall sensors) / first Wheatstone bridge 513 / second Wheatstone bridge 514 / end signal generating means 512 (e.g., Hall sensors) column arrangement. The two Wheatstone bridges 513, 514 are shown in a vertically staggered arrangement to allow for spacing, to scale, of DP / 4 apart to enable the generation of sine and cosine signals. This staggered arrangement symbolically represents the spatial intertwining of the legs of the first Wheatstone bridge 513 and the legs of the second Wheatstone bridge, which is advantageous for generating sine and cosine signals therefrom. When the combination unit having the row arrangement is aligned so that the rows are parallel to the belt movement direction, the belt moves in the movement direction, and the start signal generating means is located upstream of the belt movement direction, the start signal generating means 511 becomes the first of four sensors that detect the start of the belt end joining zone, and generates a blank zone entry signal before one of the two Wheatstone bridges 513, 514 enters the blank zone.On the other hand, the end signal generating means 512, located downstream in the belt movement direction, is the last of the four sensors that detect the end of the belt's blank zone, and generates a blank zone exit signal only after each of the two Wheatstone bridges 513, 514 has exited its blank zone. The blank zone entry signal is generated directly from the decreasing signal of the start signal generating means entering the blank zone, or from a phase generated from the sum of the squares of the signal intensities of the start signal generating means 511, 512 or the square root of that sum. The blank zone entry signal becomes a sine wave or cosine wave when the distance LH between them shown in the left part of FIG. 1 is calculated using the above formula. Similarly, the blank zone exit signal is generated directly from the re-increasing signal of the end signal generating means 512 leaving the blank zone, or from a re-increasing phase generated from the signals of the start and end signal generating means 511, 512. The blank zone exit signal becomes a sine wave or cosine wave when the distance LH shown in the left part of FIG. 1 is calculated using the above formula. In either case, it is ensured that the sine and cosine signals from the two magnetoresistive sensors 513, 514 forming the position sensing unit, either directly or through phase formation, are taken into account by the control unit only if they represent useful position information.

[0057] A further position detection unit suitable for the most preferred embodiment B2 (whether or not it uses phase-derived blank zone entry and exit signals) comprises a start signal generating means and an end signal generating means as a combined unit, optionally provided in one and the same housing, as shown in the right part of Fig. 1. However, the combined unit of this embodiment comprises a first pair of Wheatstone bridges 513, 514 and a second pair of Wheatstone bridges 515, 516 (where the legs are again composed of resistors, which are AMR sensors), in the position detection unit 51, the two Wheatstone bridges of each pair are again shown in a vertically offset arrangement (see the corresponding description in the left part of Fig. 1). The first and second pairs of Wheatstone bridges are separated from each other by a complete DP, and one of the pair of Wheatstone bridges (513, 514) has a power supply whose polarity (VCC vs. GND) is inverted relative to the polarity (GND vs. VCC) of the power supply of the other pair of Wheatstone bridges (515, 516). The purpose of such a complete DP spacing between the Wheatstone bridge pair and the reversed power supply polarity is as described above in the general definition of the position sensing unit. The purpose and meaning of the spacing LH between the start signal generating means 511, such as a Hall sensor, and the end signal generating means 512, and the remaining explanation are as described above for the left portion of Figure 1.

[0058] The conveyor of the unit preferably includes an analog or digital circuit for generating the phase. In the analog circuit, the square of the two squares may typically be formed by two conventional analog multipliers, and the sum of the two squares may typically be formed by a conventional sum operational amplifier. The analog circuit may include a comparator for comparing the sum of the squares or its square root with a threshold voltage, typically 0.4 to 0.9 times the constant magnitude, and for determining whether to generate a blank zone entry signal or a blank zone exit signal. A digital circuit for performing the phase calculation and comparison with the threshold may form part of the control unit and be programmed by software.

[0059] The conveyor of the present invention preferably comprises at least three position detectors, more preferably 3 to 16 position detectors, and even more preferably 4 to 8 or 4 to 16 position detectors, and preferably relies on position information obtained from at least two position detectors for error correction. In this case, at any time during belt movement, there are at least two position detectors that are not in the blank zone of the belt. To ensure this, the at least three position detectors are spaced apart, and each position detector detects the maximum L of the blank zone present on the conveyor belt in the direction of belt movement. j That is, if any of these at least three position sensing units is denoted by index i and any one of the remaining position sensing units is denoted by index k, where i and k range from 1 to N and are different from each other, then the associated peripheral distance between them is L ik and the magnitude is

number

[0060] (where t i , t j are the times at which the i-th and j-th position detection units (or the associated i-th and j-th start signal generation means) generate blank zone entry signals, respectively, and v is as defined above), and the maximum L j is greater than L j is measured as described above.

[0061] FIG. 2 shows the embodiment A1 or B1 of the conveyor 1 of the present invention. It comprises an endless conveyor belt 2 having a periphery consisting of a portion 21 bent over a pulley and a portion 22 not bent over a pulley. The belt may be straight (as shown) or may be convexly bent, for example, using a suitably bent support. The belt movement (loop) direction is clockwise in this example. In this embodiment, no additional means are used to detect the entry of the blank zone 4, but the signal information provided by the position detectors 51, 52, and 53 themselves is used as a means for generating a blank zone entry signal. The generation of the corresponding blank zone exit signal can consist of a suitable timer (see the explanation for embodiment A1 above), or the position detectors 51, 52, and 53 themselves can generate the corresponding blank zone exit signal (see the explanation for embodiment B1 above). In the latter case, the corresponding blank zone exit signal can be generated, for example, from the sine and cosine wave signals thereof by the phase calculation described above. The alternating, preferably essentially periodic, pattern of magnetic markers 3 runs along the entire periphery of belt 2 (the portions curved over pulley 21 and the portions not curved over pulley 22) except for blank zone 4, which may be the end joining zone of conveyor belt 2. In this blank zone, the markers generate a magnetic field that is less than a predetermined maximum strength (as generated by magnetic markers 3 outside the blank zone) or are completely erased, as shown in the figure. The three position sensing units are designated by numerals 51, 52, and 53. A pair of spacing distances L between any two of these three spaced units is 12 , L 13 , L 23 is the length of each only blank zone L j By meeting this requirement, it does not matter which of the three position sensing units is in the blank zone 4 and is therefore unable to generate a useful position signal. Since there are two other position sensing units that are not in the blank zone 4, the control unit 6 can always rely on position information received from at least two of the position sensing units.

[0062] Belt blank zone length L j " is defined for purposes of the present invention as follows when the belt is moving in the belt travel direction:

number

[0063] where t1-t2 is the time period between a blank zone entry signal and a corresponding blank zone departure signal, and where applicable, both the blank zone entry signal and the corresponding blank zone departure signal are generated by one and the same position detection unit as described above, or the blank zone entry signal is generated by a start signal generating means as described above and the corresponding blank zone departure signal is generated by an end signal generating means as described above, or the blank zone entry signal is generated by one upstream start signal generating means and the corresponding blank zone departure signal is generated by an adjacent downstream start signal generating means, but it is the blank zone entry signal that is used (interpreted) as the blank zone departure signal, where "upstream" and "downstream" refer to the direction of belt movement, the blank zone entry signal and the blank zone departure signal are derived from the phase as described above, and v is the belt speed.

[0064] FIG. 3 shows another embodiment of the conveyor 1 of the present invention, which is the most preferred embodiment B2. This conveyor 1 includes two position detectors 54 and 55, a start signal generating means 511 and an end signal generating means 512, such as a Hall sensor, for detecting entry into and exit from the first position detector 54 of the blank zone 4, and a start signal generating means 521 and an end signal generating means 522, such as a Hall sensor, for detecting entry into and exit from the second position detector 55 of the blank zone 4. The belt travels (loops) clockwise in this example as well. The start signal generating means 511 and 521 generate blank zone entry signals to the first and second position detectors 54 and 55, respectively, when entering the blank zone 4, and the end signal generating means 512 and 522 generate blank zone exit signals to the first and second position detectors 54 and 55, respectively, when leaving the blank zone 4. Alternatively, the start and end signal generating means 511, 512 may generate both blank zone entry and exit signals to the first position detecting unit 54 using the phase calculation described above. Also, the start and end signal generating means 521, 522 may generate both blank zone entry and exit signals to the second position detecting unit 55 using the phase calculation described above. In this alternative phase calculation, the interval LH between the start and end signal generating means 511, 512 and between the start and end signal generating means 521, 522 satisfies the above formula for calculating LH. The end signal generating means 512 of the upstream position detecting unit 54 receives the length L of the only blank zone 4 present on the conveyor belt 2 from the start signal generating means 521 of the downstream position detecting unit 55. j The upstream position sensing units 54 and downstream position sensing units 55 are themselves spaced apart by a distance L greater than the distance L. The remaining parts and associated numerals are as shown and described with respect to FIG. 2. FIG. 4 shows a further embodiment of the inventive conveyor 1 according to embodiment B32. The direction of movement (loop) of the conveyor belt 2 is also clockwise in this example. In this further embodiment, the upstream position sensing units 54 and downstream position sensing units 55 are spaced apart by a distance L greater than the distance L. jThere are first start signal generating means 531 such as a Hall sensor capable of generating a blank zone entry signal to a downstream (not necessarily, but preferably adjacent) position detector 56 when entering the blank zone 4 of the first start signal generating means 531; second start signal generating means 534 such as a Hall sensor located downstream of all the position detectors and capable of generating a blank zone entry (leaving a used / unused blank zone) signal to an upstream (not necessarily, but preferably adjacent) position detector 57 when entering the blank zone 4; and two further start signal generating means 532, 533 such as Hall sensors capable of generating a blank zone entry signal to each of the downstream and adjacent position detectors 57, 58 when entering the blank zone 4. This blank zone entry signal is simultaneously used (interpreted) as a blank zone exit signal for each of the upstream and adjacent position detectors 56, 57. Therefore, the number of such further start signal generating means 532, 533 is two, which is one less than the number of position detectors 56, 57, 58. Each of the further start signal generating means 532, 533 is sandwiched between the upstream (and adjacent) position detector 56, 57 and the respective downstream (and adjacent) position detector 57, 58. The distance L by which either of the further start signal generating means 531, 532 is spaced from the upstream position detector 56, 57 is m is the length L of the only blank zone 4 present on the conveyor belt 2. j Furthermore, the distance L by which the second start signal generating means 534 is spaced from the adjacent upstream position detecting unit 58 is larger than i Also, the above L jWhen the first additional means 531 thereof enters the blank zone 4, the control unit 6 ignores the position information from the most upstream position detection unit 56 from the time when it receives the blank zone entry signal from the first start signal generation means 531, ignores the position information from the other position detection units 57, 58 from the time when it receives the blank zone entry signal from the adjacent upstream start signal generation means 532, 533, reconsiders the signal from the most downstream position detection unit 57 from the time when it receives the blank zone entry (used / interpreted blank zone departure) signal, and reconsiders the position information from any of the other position detection units 56, 57 from the time when it receives the blank zone entry (used / interpreted blank zone departure) signal from the further adjacent downstream start signal generation means 532, 533. The remaining parts and related numerals are as shown and explained in Figures 2 and 3.

[0065] For the purposes of the present invention, a length including an additional means as one of its endpoints is measured from the geometric center of the additional means, in particular, if such additional means is a Hall sensor, from the geometric center of the semiconductor layer included therein. Similarly, a length including a magnetoresistive sensor as one of its endpoints is measured from the center of the bridge in the Wheatstone bridge included therein, in relation to its endpoint.

[0066] As described above, the conveyor of the present invention includes at least two, preferably at least three, and more preferably multiple position sensing units. The output voltage of any position sensing unit is preferably converted to digital information, typically 8-16 bits deep, using a conventional analog-to-digital converter (ADC). This conversion occurs regardless of whether the position sensing unit generates a sine wave / sine wave voltage signal of a predetermined intensity (as long as the position sensing unit is near a readable magnetic marker) or a sine wave / sine wave voltage signal of less than the predetermined intensity (as long as the position sensing unit is located in one of the belt's blank zones). It is the task of the control unit to decide whether to take into account the digital information thus generated based on the blank zone entry and blank zone exit signals received from each position sensing unit.

[0067] The position detector, optional start and end signal generating means, and optional start / end signal generating means may be positioned closer to the conveyor belt's conveying surface than to the conveyor belt's pulley-facing surface, or closer to the conveyor belt's pulley-facing surface than to the conveyor belt's conveying surface. In either of these two variations, the magnetic markers may be on the conveying surface (not covered by a cover layer), near the conveying surface (e.g., covered by an upper cover layer), embedded in the belt, on the pulley-facing surface (not covered by a lower cover layer), or near the pulley-facing surface (e.g., covered by a lower cover layer).

[0068] The output signal of the position detector may optionally be filtered with a low pass filter to remove high frequency noise before being converted to a digital signal by the ADC.

[0069] The output of each ADC is typically an unsigned byte (if the ADC is 8-bit deep). In this case, the maximum positive value of the sine and cosine signals, assuming ideal, will be near 255 (e.g., 238). The zero point will be 128. The minimum negative value of the sine and cosine signals will be near zero, such as 18 (e.g., 18). In this case, the amplitude of the digitized cosine and sine signals will be 110. If the ADC is 16-bit deep, its output is typically an unsigned short integer, and the maximum positive value of the sine and cosine signals, assuming ideal, will be near 65535 (e.g., 62768). The zero point will be 32768. The minimum negative value of the sine and cosine signals will be near zero (e.g., 2768). In this case, the amplitude of the digitized cosine and sine signals will be 30000. The amplitudes of the analog input sine and cosine signals are appropriately scaled, for example using two op-amp instrumentation amplifiers, before being fed to the ADC so that the ADC's 256 (8-bit depth) or 65538 (16-bit depth) operating range is used as much as possible without saturating the ADC's output. This optimized digital output amplitude of the ADC (110 for an 8-bit depth ADC, 30000 for a 16-bit depth ADC) is referred to below as the "nominal amplitude A" of the ADC's digital signal output.

[0070] The control unit essentially performs all subsequent tasks using the digital signals received from the ADCs of all the position sensing units. The control unit may be implemented in software by a suitably programmed computer or in hardware by a field programmable gate array FPGA combined with a microcontroller.

[0071] The required and optional functions of the control unit are described below. In this specification, "ωt" is the angular velocity multiplied by time and the position angle (radian). If the position detection unit includes an AMR sensor, the position angle is within a full period of 2π corresponding to the pole pitch DP, i.e., half the distance between two adjacent north poles (or two adjacent south poles) of the magnetic marker pattern. If the position detection unit includes a GMR or TMR sensor, the angular position (radian) corresponds to a full period of 2π corresponding to 2DP, i.e., the full distance between two adjacent north poles (or two adjacent south poles) of the magnetic marker pattern. The same assumptions are made for the calculated position angle φ, phase shifts α, and Δα used in the following description.

[0072] A first essential task of the controller is to ignore any digital signal received from any position detector within the time window between the blank zone entry signal and the blank zone exit signal from that position detector. To this end, the controller may configure a corresponding flag for each position detector that the controller sets to a value indicating "don't care" (e.g., "zero") when the controller receives a blank zone entry signal from that position detector, and that the controller sets to a value indicating "care" (e.g., "1") when the controller receives a blank zone exit signal from that position detector.

[0073] A second (optional but preferred) task of the control unit is to process the digital signals from any ADC (i.e., digital signals from any associated position sensing unit) based on the first task described above, to remove phase, offset, and amplitude (POA) errors from the digital sine and cosine signals from any considered ADC. Two error-loaded input signals cos from each of the considered ADCs are input ,sin input can be written as follows, even if digitized, within experimental error:

[0074]

number

[0075] where C and S (dimensionless) indicate the offset; ΔAc and ΔAs (integers in ADC units) indicate the amplitude deviation from the nominal amplitude A; α (radians) indicates the additional phase shift of a sine wave relative to a cosine signal beyond the normal phase shift of -π / 2; and Z (integers in ADC units) indicates the zero point of the ADC. If the output of the ADC is an unsigned integer, the zero point Z is usually half the resolution of the ADC, i.e., 2 for an 8-bit ADC. 7 =128, for a 16-bit ADC, 2 15 = 32768. If the output of the ADC is a signed integer, Z is usually zero.

[0076] The purpose of this second (optional but preferred) task is to derive a digitally error corrected cosine signal cos from the input signal. corr and sine wave signal sin corr In the present invention, these digital error correction signals are assumed and defined, within experimental error, as follows:

[0077]

number

[0078]

number

[0079] where G C and G S (dimensionless) is the scaling gain relative to the nominal amplitude A. ΔC and ΔS (dimensionless) are digital correction offsets. Δα (radians) is the correction phase shift. After applying these corrections,

number

number

[0080] Furthermore, in (3b) and (4b), we assume that Δα is small and α / 2 is small (from the first term of the Taylor series, respectively).

[0081]

number

number

[0082] When (3c) and Δα and α / 2 are reduced, the above-mentioned (3b) is further corrected as follows:

[0083]

number

[0084] 1-α 2 and

number

[0085]

number

[0086] Furthermore, if we arbitrarily assume that (A+ΔAc)=(A+ΔAs), then equation (3e) can be further simplified to:

[0087]

number

[0088] If we further assume arbitrarily that αΔαC is zero and 1-α is essentially equal to 1, then (3f) can be further simplified to

[0089]

number

[0090] In the simplest embodiment, assuming that (A+ΔAc) is A, (3g) becomes:

[0091]

number

[0092] Similarly to (4c), when Δα and α / 2 are reduced, the above (4b) is further corrected as follows:

[0093]

number

[0094] Again, 1-α 2 and

number

number

[0095] Furthermore, if we arbitrarily assume that (A+ΔAc)=(A+ΔAs), then equation (4e) can be further simplified to:

[0096]

number

[0097] Furthermore, if we arbitrarily assume that αΔαS is zero and that 1-α is essentially equal to 1, then (4f) can be further simplified to

number

[0098] In the simplest embodiment, assuming that A is (A+ΔAs), (4g) becomes:

number

[0099] Formulas (3f), (3g) or (3h), and (4f), (4g) or (4h) are represented by sin corr ,cos corr is a simple preferred formula for calculating

[0100] For the second (optional but preferred) task, three exemplary variations are: input and cosine wave input signal cos input Assuming that meets the most preferred definition above, the following is an example:

[0101] In the first modification, the control unit calculates cos input and sin input It stores the value of the stored cos input The maximum value (assuming φ=0) and the minimum value (assuming φ=π) of the value are scanned, and these are respectively C max and C min When used in (1) above, the result is as follows:

number

[0102] Similarly, the control section is the stored sin inputThe maximum value (assuming θ=π / 2) and the minimum value (assuming θ=3π / 2) of the value are scanned, and these are respectively S max ,S min Then, when these are used in (2) above, the result is as follows:

number

[0103] The control unit uses Gc and Gs as the gains.

number

[0104] If the above (6) is used in the above requirement C-ΔC=0, the following is obtained.

number

[0105] When the requirement S-ΔS=0 is used in (8), the following results:

number

[0106] Furthermore, the control unit simultaneously input1 -Z=cos input1 For the first embodiment, where Z is the stored cos input value and sin input Scan the value pair and set it to CS eq1 and at the same time sin input2 -Z=-cos input2 For the second embodiment, where Z is the stored cos input value and sin input Scan the value pair and set it to CS eq2 Let sin input and cos input If there is no additional phase shift between eq1 becomes φ=π / 4, and CS eq2becomes φ=3π / 4. sin input and cos input The additional phase shift α between input sin for the cosine term of input (see (1) and (2)). input cosine to sine term of input It is also the inverse phase shift -α of the cosine term of CS. eq1 For , it is as follows:

[0107]

number

[0108] Using (7) and (8) here, we get:

[0109]

number

[0110] CS eq2 For , it is as follows:

[0111]

number

[0112] Using (5) and (6) here, we get the following:

[0113]

number

[0114] From (13) and (14), the desired phase shift α / 2, and the correction phase Δα which must be equal to it according to (3) and (4), can be found as follows:

[0115]

number

[0116] To perform the first variation of the error correction, the control unit calculates cos input value and sin input (5), (7), (9), (10), (11), (12), and (15), respectively, are stored as input value and sin input Then, for one or more subsequent time periods, equations (3d), (3e), or (3f) and (4d), (4e), or (4f) are used to calculate the cos(ω) of any input signal value from the ADC being considered. input / sin input The obtained values of ΔAc, ΔAs, Gc, Gs, ΔC, ΔS, and Δα are used to simultaneously correct the offset, amplitude, and phase errors of the error-corrected signal cos corr ,sin corr One iteration is required to obtain cos(x,y) and no further processing is required. The control unit preferably performs the same operation as the stored cos(x,y) from a given period after one or two subsequent periods have been observed. input value and sin input Replace the value pairs with the corresponding updated values from the subsequent period.

[0117] This first variation does not use feedback of the simulated position angle and / or error correction to combine it with the measured position angle. The omission of such feedback avoids potential stability issues associated with high-frequency noise in the measured input position angle. High-frequency noise can cause oscillations by phase-shifting the fed-back calculated position angle relative to the original measured position angle, switching from negative feedback to positive feedback. This problem can also arise when the calculation rate of the simulated position angle becomes comparable to the sampling rate of the digital measured position angle if the feedback is digital.

[0118] A second variant for error correction is the pair of corrected values (sin corr |cos corr In this second variant, each pair of values (sin corr |cos corr ), the magnitude M and phase φ between the two signals can be derived as follows:

[0119]

number

[0120] φ can be evaluated using equation (17) or in a "vectorized" mode using the known CPRDIC algorithm. An exemplary outline of such a "vectorized" CORDIC algorithm is shown in steps i) through v) below.

[0121] i) Assume a negative approximation of the phase φ, φ' (i.e., -φ) (the initial value of φ' is zero), and calculate the incremental correction angle φ i Assume an array of tangent values of atn(2 -N ) is chosen so that it is no larger than the allowable residual error for φ, and the value of the i-th tangent is tan(φ i )=2 -i is given as the initial value sin corr If is 0 or more, it is +1, the initial value is sin corr Assume a sign variable σ that is −1 if is less than 0.

[0122] ii) sin corr , cos corr Calculate the new iteration value of as follows:

number

number

[0123] iii) (sin corr ) i+1 But (sin corr ) i If it has the same sign as φ' to σ×atn(2 -i ) values. set(sin corr ) i =(sin corr ) i+1 set(cos corr ) i =(cos corr ) i+1

[0124] iv) Increment i by 1.

[0125] v)(sin corr ) i+1 Repeat steps ii) to iv) until i is equal to zero or until i reaches N.

[0126] v) The magnitude M to be obtained is used as an input in the last iteration of ii)+iii) (cos corr ) i and the phase φ to be obtained is φ' obtained in the last iteration of ii)+iii).

[0127] atn(2 -i ) and

number

[0128] The pair of values (M|φ) obtained from one or two periods of the input signal is needed to derive the values of ΔC, ΔS, Δα, Gc, and Gs, which are explained as follows (for the magnitude M at the phase angle φ, "M φ " is used, e.g., the magnitude M at zero radians is written as "M0".) Each value φ is the sum of its cos input value and sin input It is assumed that the value of ωt±α / 2 in

[0129] Next, from the equation (3g) for the correction gain G and the difference (C-ΔC), set Z to zero (cos input and sin input (subtracted from, see above) and

number

number

number

number

[0130] If two pairs of values (M|φ) are collected for two periods, the correction gain Gc and the difference (C-ΔC) are given by:

number

number

[0131] The collected correction gain Gs and difference (S-ΔS) are obtained from equation (4d):

number

number

number

number

[0132] If two pairs of values (M|φ) for two periods are collected, then:

number

number

[0133] Correction of the additional phase shift Δα

[0134]

number

[0135]

number

[0136]

number

[0137] Δα is

number

[0138] Here, equation (18a) can be used for the correction gain Gc.

[0139] If a pair of values (M|φ) is collected from two periods, then:

[0140]

number

[0141] Here, equation (18b) can be used for the correction gain Gc.

[0142] To perform the second variation of the error correction, the control unit proceeds as follows.

[0143] a) one- or two-period cos input and sin input For each pair of input values, calculate the magnitude M and phase φ using equations (16) and (17), respectively. Here, the values of ΔC, ΔS, and Δα are set to zero, and the values of Gc and Gs are set to 1. Therefore, cos input and sin input Each pair of input values of is converted to a corresponding pair of corrected values cos corr and sin corr Therefore, equations (16) and (17) are the same as cos input value and sin input The control unit stores all the magnitude M and phase φ value pairs obtained in this way for further evaluation. The control unit preferably evaluates equations (16) and (17) using the CORDIC algorithm described above in "vectorized" mode.

[0144] b) From the stored pair of values (M|φ), calculate the value of Gc using equation (18a) and the value of C-ΔC using equation (19a) assuming (A+ΔAc) equals A (if one period was stored), or from the stored pair of values (M|φ), calculate the value of C-ΔC using equation (19b) assuming (A+ΔAc) equals A (if two periods were stored). From the stored pair of values (M|φ), calculate the value of Gs using equation (20a) and the value of S-ΔS using equation (21a) assuming (A+ΔAs) equals A (if one period was stored), or from the stored pair of values (M|φ), calculate the value of Gs using equation (20b) and the value of S-ΔS using equation (21b) assuming (A+ΔAs) equals A (if two periods were stored). Calculate the value of Δα using equation (22a) (if one period has been stored) or equation (22b) (if two periods have been stored).

[0145] c) Add the value of C-ΔC obtained in step b) or d1) to the value of ΔC, as appropriate, and add the value of S-ΔS obtained in step b) or d1) to the value of ΔS, as appropriate.

[0146] d) cos for at least one further period input and sin input For any pair of values of , d1) Assume Z is zero, and (A + ΔAc) and (A + ΔAs) are equal to A. Using equations (3d) and (4d), find cos corr and sin corr The values of Gc, Gs, and Δα are obtained by step b) or step d1), and the values of ΔC and ΔS are obtained by step b) or step d1), and step c) as needed. corr value and sin corrFrom the value pairs, calculate the value pairs of magnitude M and phase φ using equations (16) and (17), preferably using the CORDIC algorithm in "vectoring" mode. From the value pairs of magnitude M and phase φ, calculate new values of Gc, C-ΔC, Gs, S-ΔS, and Δα obtained by the procedure described in step b).

[0147] d2) If Gc and / or Gs obtained in step d1) deviate from 1 by more than an allowable difference, and / or if C-ΔC, S-ΔS, and / or Δα obtained in step d1) deviate from zero by more than an allowable difference, proceed to step c) and the subsequent steps, and calculate cos input and sin input Otherwise, the pair of values of the magnitude M and the phase φ obtained in step d1) is used as the final magnitude M and the final phase φ, and the cos of said at least one next period is calculated by performing step d1) and the subsequent steps. input and sin input The next pair of values of is processed.

[0148] e) The control unit calculates the cos( input and sin input After a further 1 to 10 periods of value pairs, preferably 4 periods, steps a) and b) are performed again.

[0149] If the sign and preferably approximate magnitude of α / 2 is known, a third variant for error correction may be used. This variant is based on a pair of corrected values (sin corr |cos corr ) again and use the identity

number

[0150]

number

[0151]

number

[0152] Assuming (A+ΔAc) and (A+ΔAs) to be equal to magnitude M, as appropriate, we obtain (3j) and (4j) in turn.

[0153]

number

[0154]

number

[0155] On the other hand, assuming that Z is zero, and using equations (3a) and (4a) in equation (16a), we obtain the following:

[0156]

number

[0157] In the above equation, ωt is assumed to be approximately equal to φ, and Δα and α / 2 can be assumed to be approximately zero.

[0158]

number

[0159]

number

[0160] Similarly using equations (3a) and (4a) in equation (16b) and assuming Z is zero, we get:

[0161]

number

[0162] In the above equation, we can assume that ωt is approximately equal to φ, and that Δα and α / 2 are approximately zero.

[0163]

number

[0164]

number

[0165] M φ Since and φ are independent of each other, equation (23a) becomes, using the identity,

[0166]

number

[0167] Similarly, M φ Since and φ are independent of each other, equation (23b) becomes, using the identity,

[0168]

number

[0169] Based on (24a) and (24b), the values of (A + ΔAs) / Gs and (A + ΔAc) / Gc are always the same. In practice, both must converge to the nominal amplitude A.

[0170] The third variant is recursive and forms an independent object of the present invention. Using equations (3j) and (4j), we find sin input value and cos input Sine from value corr value and cos corrCalculate the value of sin corr value and cos corr Using the values, M is calculated using equations (16) and (17) or the CORDIC algorithm. φ and φ are calculated, and cos(φ) and sin(φ) are calculated by equations (16a) and (16b), respectively, and M obtained by equations (24b), (24a), (25b), and (25a) is calculated. φ , cos(φ), and sin(φ) are used to calculate (A+ΔAs) / Gs, (A+ΔAc) / Gc, (S-ΔS), and (C-ΔC), respectively, and the latter are used again in equations (3j) and (4j). In these equations, the terms

number

number

[0171] The input signal cos according to the third modification input and sin input The iteration phase adjustment for a given pair of values of is as follows:

[0172] a) To start the iteration, the initial values of Δα, ΔC, and ΔS (defined for equations (3a) and (4a)) are assumed to be zero, the initial values of k and σ are chosen as above, and the magnitude M φis assumed to be reasonably close to the nominal amplitude A, for example in the range of 0.5 to 2 times the nominal amplitude (e.g., 1.25 times). φ is the input signal cos input and sin input where ω at ωt is the nominal angular velocity which can be derived as πv / DP', where v is the nominal belt travel speed and DP' is the pole pitch as described above, representing the nominal (unstretched and unshortened pole pitch), or ω at ωt is derived from a predetermined position angle φ and forms its first derivative with respect to time. t at ωt is derived from a timer or clock.

[0173] b) using σ from step a), k from step a), or k from step g), as appropriate;

number

[0174] c) Using Δα, ΔC, and ΔS from step a) or step f2) as needed,

number

[0175] d) cos of step c) corrand sin corr , the new associated magnitude M φ and a new associated φ, and then using equations (16a) and (16b) to calculate new associated cos(φ) and sin(φ) values, and optionally, if equation (17) or numerical calculations using the CORDIC algorithm provide φ bounded between -π / 2 and +π / 2, then further: i) sin corr <0 and cos corr >0, add 2π to the limited φ, and ii) sin corr >0 and cos corr <0, add π to the limited φ, and iii) sin corr <0 and cos corr If <0, add π to the limited φ.

[0176] e) M obtained in step d) φ Using the values of Δα in step (a) or step f2) and Δα in step b) as needed

number

[0177] f) If the new value of (A+ΔAs) / Gs = (A+ΔAc) / Gc is closer to A than the corresponding previous value of (A+ΔAs) / Gs = (A+ΔAc) / Gc in step b), f1) setting the new value of (A+ΔAs) / Gs=(A+ΔAc) / Gc in step e) as the corresponding previous value of (A+ΔAs) / Gs=(A+ΔAc) / Gc; f2) Step b)

number

[0178] g) Increase k by 1.

[0179] h) Repeat steps b) through g) until k is the other maximum index.

[0180] i) M obtained in step d) of the last iteration φ and φ, and / or the final cos(φ) and final sin(φ) are used to determine the position.

[0181] Computer simulation of this third variant for error correction shows that cos input and sin input It has been found that while this third variant can largely or almost completely eliminate the additional phase shift α between sin and sin, it is not very suitable for eliminating the amplitude deviations ΔAc, ΔAs and the offsets C and S. Therefore, this third variant is a sin input and cos input The offsets C and S and the amplitude deviations ΔAc and ΔAs are previously averaged, for example by using a position detector in accordance with the right-hand part of FIG. 1 and the associated description. In this third variant, the phase error α is preferably a positive value, for example in the range of 0 to +0.06 radians, more preferably in the range of 0 to +0.03 radians. Computer simulations have further shown that the phase error α is input and cos input It has been shown that the additional phase shift α can also be largely or nearly eliminated if the signal is loaded with a small amount of random noise.

[0182] 8a and 8b show the results of phase error correction according to the third modification of the present invention. Equations (1) and (2) include the terms -α / 2 and +α / 2 as the phase errors, respectively. If α / 2 itself is greater than zero, then cos inputis delayed (shifted to the right on the time or angle scale), whereas sin input is advanced (shifted to the left on the time or angle scale). As a result, sin input is a phase shift of α sin input becomes closer to cos input From sin input The phase shift to is π / 2-α, rather than the usual π / 2. This situation typically occurs when a magnetically marked belt stretches slightly during use (when the pole pitch increases from the normal unstretched DP' to stretched DP'), without a corresponding increase in the distance between the sensors generating the sine and cosine signals in the position sensing section. Figure 8a shows the situation in radians over the entire period when α / 2 is +0.03 radians. This corresponds to the belt stretch rate, i.e., the rate of increase of DP relative to DP', of 1 + 0.06 / 2π = 1.0095, or approximately 1% belt stretch rate. sin input and cos input has a nominal digitized amplitude (in ADC units) of 55000. input If the signal is delayed in time or angle, it is indicated by a dashed line, and sin input The signal is shown as a dotted line if it is advanced in time or angle. After iterating from an initial value of k=6 to a maximum value of 31, the initial magnitude M is 1.25 times the nominal amplitude A. φ The phase error corrected cos corr and sin corr The value of cos is plotted against the corresponding phase error corrected angle φ and is shown as a solid line. corr is essentially sin corr shows a normal phase shift of π / 2 for , and the additional error is negligible. The respective corrected Δα, ΔC, and ΔS values obtained after the last iteration are plotted in Fig. 8b against the corresponding corrected phase angle φ over the entire period in radians, and are shown as solid, dashed, and dotted lines, respectively.

[0183] 5 shows an exemplary schematic diagram of a control unit 6 suitable for implementing the second variant according to equations (3f) and (4f) and the third variant according to equations (3j) and (4j). This control unit comprises an FPGA 7, for example a Cyclone 5 (Intel). First, the FPGA 7 calculates the ADC signal A(cos input ) and B(sin input ) to perform the calculations of equations (3f) and (4f). Typically, two successive Altera lpt_Mult megafunctions are used for each of the two equations. The first of equations (3f) and (3j) has reference number 71 and is a function of sin with any necessary coefficients. input The first of equations (4f) and (4j) has reference numeral 72 and is multiplied by cos input The second of equations (3f) and (3j) has reference numeral 73 and performs a final scaling multiplication by (A+ΔAc) / Gc. The second of equations (4f) and (4j) has reference numeral 74 and performs a final scaling multiplication by (A+ΔAs) / Gs. Further logic gates 75, 76, 77, and 78 perform the addition / subtraction operations appearing in equations (3f), (3j), (4f), and (4j). Furthermore, FPGA 7 may comprise two AND gates 81 and 82 for determining whether to consider the pair of input signals A / B based on the signal state of signal line E. The state of signal line E may be derived directly from the respective position detectors as a result of Hall sensors representing the blank zone entry signal and the blank zone exit signal. Alternatively, the state of this signal may be derived from the state of the above-mentioned "consider / don't consider" flag set by the blank zone entry signal and the blank zone exit signal. Furthermore, the FPGA 7 outputs the error-corrected output signal A'(cos corr ) and B'(sin corr ) according to the CORDIC algorithm, φand the phase angle φ. The control unit 6 further comprises a microcontroller 10 for calculating the error correction values Gc, Gs, (C-ΔC), (S-ΔS), and Δα using the equations (18a) / (18b), (20a) / (20b), (19a) / (19b), (21a) / (21b), and (22a) / (22b), respectively (second variant), or for calculating the initial values sin(φ) and cos(φ) and the error correction values (A+ΔAs) / Gs, (A+ΔAc) / Gc, (C-ΔC), and (S-ΔS) using the equations (24b), (24a), (25a), and (25b), respectively, and adding the sum of the increments of (α / 2-Δα) to Δα (third variant). When used in the second variant for error correction, the control unit 6 further determines the magnitude M of the pair of values of one or two periods. φ and the phase φ. If the control unit 6 is used in the third modification, such a read / write storage device 11 is not required. The output of the microcontroller 10 serves as the input of the FPGA 7. On the other hand, φ The final output values of and are used for further processing as described below.

[0184] The operational paths shown in Figure 5 consider one ADC input signal pair A / B and one signal E. In practice, the FPGA 7 will have a corresponding operational path for each ADC input signal received from the position sensing unit (whether considered or not).

[0185] The control unit performs, as a third essential task, the calculation of the incremental position angle φ, which is calculated by multiplying each sin input and cos input and calculating the arctangent from the quotient thus obtained, or may be the final phase φ obtained when said second, but preferred, error correction is not performed, or after performing one of said second, but preferred, error corrections.

[0186] The control unit performs a fourth optional task, either simultaneously with the third task or sequentially after the third task, of calculating the first derivative in time of the incremental position angle φ to obtain ω as defined above, which is the raw sin ω obtained from the different considered ADCs. input and cos input It does not depend on the (unknown) phase shift β between the signals. If the fourth task runs simultaneously with the third task, it is executed directly as follows:

[0187]

number

[0188] Such an unknown phase shift β can be calculated, for example, by the distance L between the two position sensing elements to be considered. jk is a non-integer multiple of 2*DP (DP is defined above), and / or the length L of the end joining zone j This may be due to the fact that ω is a non-integer multiple of 2*DP. The controller can only approximately calculate ω at discrete times, each time point separated from the next by a finite, constant time interval Δt, as follows:

[0189]

number

[0190] where Δt is small enough to constitute a reasonable approximation of the above formula.

[0191] The control unit may assume that ω is constant over time or may assume that ω is variable over time. In particular, in the case of the approximate calculation, the control unit may calculate the approximate ω obtained in this way by dividing the approximate ω by a pair of discrete values t n / ω(t n ) where t n are n time points, and ω(t n ) is the value of ω at that point in time. The controller may store the sequence in an internal memory and calculate an approximate angular acceleration or deceleration value therefrom.

[0192] The control unit performs incremental position determination as a fifth required task. This can be done by counting at a frequency (constant or variable) proportional to ω when the fourth optional task is performed. An "increment" is an arbitrarily selected increment length ΔL, chosen small enough to achieve the required fineness of position determination. ΔL can typically range from 1 millimeter to 1 nanometer, depending on the required fineness. The spatial increment length ΔL may be stored in the control unit in an erasable / overwritable format or in a read-only format. To count such ΔL increments, the control unit typically includes an integer register that can be incremented in one step. Each such increment of 1 added to the integer register corresponds to one spatial increment of the length ΔL relative to the absolute position. The control unit increments the count of the integer register at a frequency f according to the following formula:

[0193]

number

[0194] where all symbols are as defined above, and if the fourth optional task is not performed, the incremental position determination is performed directly from the incremental position angle φ obtained in the third required task, according to the methods and calculations disclosed in WO 2016 / 146463.

[0195] As a sixth (optional but preferred) task, the control unit may perform a reset to zero for each of the integer registers mentioned above upon receiving a blank zone entry signal from each magnetic position sensing unit or a blank zone exit signal from that position sensing unit, so that the integer registers will be updated to indicate the absolute position P of each i-th position sensing unit with respect to the start of the end junction zone as a reference point (if the control unit uses the blank zone entry signal for resetting) or with respect to the end of the end junction zone (if the control unit uses the blank zone exit signal for resetting). i An integer p representingi Therefore, the relation P i ≡ΔL×p i is defined.

[0196] As a seventh (optional but preferred) task, the control unit uses the position information contained in its integer registers to perform one or more of the evaluations outlined below.

[0197] The first type of evaluation is that each pi in the ith integer register corresponding to the ith position detector being considered (i.e., when the associated "consider / not consider" flag is set to "consider") may deviate at any time by a certain offset from the theoretically correct integer value representing the absolute position of each position detector along the belt periphery at that time relative to the edge of the edge joining zone as a reference. To correct this first type of error, the control unit typically performs a blank zone exit signal when a position detector with a predetermined fixed index (typically index 0) generates a blank zone exit signal and the p in its associated integer register is i is reset to zero (see above), the contents of any integer register for which the flag is set to "consider" are compared with respective comparison values stored in the look-up table, and the contents of each integer register that deviates from the corresponding comparison value are replaced with the comparison value. Alternatively, the control unit may detect when a position detection unit with a predetermined fixed index (typically index 0) generates a blank zone exit signal and p contained in the associated integer register i is reset to zero (see above), the contents of all other integer registers may simply be replaced by the corresponding comparison value from the look-up table. The comparison value stored in the look-up table for each position sensing element is the quotient obtained by dividing (by integer division) its absolute position along the belt periphery relative to the end of the belt's end juncture zone by ΔL as defined above. Assuming the belt is positioned such that a position sensing element with said predetermined fixed index, such as 0, is located exactly at the end of the belt's end juncture zone, the comparison value obtained by said integer division will be 0.

[0198] The second type of evaluation that the control unit can correct is belt position deformation due to uneven belt acceleration and deceleration. In the case of a multi-pass inkjet printer, which is one of the preferred applications of the conveyor of the present invention, it should be noted that the belt typically undergoes acceleration and deceleration steps by a servo motor when it is supposed to move from one printed line to the next, but the belt does not move at all during the printing of the next line. To correct this second type of error, the control unit may perform a comparison after a certain period of time.

[0199] The controller may issue its final position information on one or more channels, for example as a conventional quadrature signal using channels A and B. When the conveyor of the present invention is used in a single pass or multi-pass inkjet printer, particularly for printing on textiles, the final position information may be supplied to a conventional printer head driver for controlling printing.

[0200] A suitable conveyance device of the present invention suitable for effectively performing the phase error correction (third modification) of the present invention preferably comprises the following configuration.

[0201] (a) An endless conveyor belt having a belt periphery, an outer conveying surface, and an inner pulley-facing surface, wherein magnetic markers are periodically arranged in the direction of belt movement so that their north or south poles alternately face the outer conveying surface, adjacent north and south poles are spaced apart by an elongated pole pitch DP that generates an alternating magnetic field, and the elongated pole pitch DP is greater than, equal to, or less than the non-elongated, non-reduced pole pitch DP'.

[0202] (b) at least one position sensing unit arranged in spatial proximity to the magnetic marker along the belt periphery, capable of detecting from the magnetic field of the magnetic marker 1) a cosine wave voltage signal, 2) a corresponding phase-shifted sinusoidal voltage signal shifted by a phase angle α-π / 2 (α is greater than zero) relative to cosine wave signal 1) when the conveyor belt is traveling in the direction of travel and DP is greater than or less than DP', and capable of detecting from the magnetic field of the magnetic marker 3) a cosine wave voltage signal corresponding to the angular position of the position sensing unit relative to the magnetic marker pattern when the conveyor belt is traveling in the direction of travel and DP is equal to DP', and 4) a sinusoidal voltage signal shifted by a phase angle -π / 2 relative to cosine wave signal 3) and corresponding to the same angular position of the position sensing unit relative to the magnetic marker pattern.

[0203] (d) a controller for deriving position information from the digitized sine wave voltage signals and cosine wave voltage signals of the at least one position sensing unit, the digitized signals having a nominal amplitude A, the controller further comprising: input ) and cosine wave voltage value (cos input ) using the third variant using equations (3j) and (4j) to perform iterative phase error correction as follows:

[0204] a) To start the iterations, assume initial values of Δα, ΔC, and ΔS as set forth in equations (3j) and (4j) and defined in equations (3a) and (4a) to be zero, assume a value σ that is +1 if α / 2 is positive or zero, or −1 if α / 2 is negative, assume an initial integer value k in the range of 4 to 8, preferably 6, or if the magnitude of α / 2 is known,

number

[0205] b) Using σ from step a) and k from step a) or step g), as appropriate,

number

[0206] c) Δα, ΔC, and ΔS in step a) or step f2) as needed, and

number

[0207] d) cos obtained in step c) corr and sin corr Using the values, a new associated magnitude M can be calculated using equations (16) and (17), or preferably using the CORDIC algorithm described above in "vectorized" mode. φand the new associated φ value, and using equations (16a) and (16b) calculate new associated cos(φ) and sin(φ) values, and optionally, if calculation according to equation (17) or calculation according to the CORDIC algorithm provides φ bounded between -π / 2 and +π / 2, do the following:

[0208] i) sin corr <0, cos corr If >0, add 2π to the limited φ. ii) sin corr >0, cos corr If <0, add π to the restricted φ. iii) sin corr <0, cos corr If <0, add π to the restricted φ.

[0209] e) M obtained in step d) φ the value of Δα in step (a) or step f2) as required, and the value of Δα in step b)

number

[0210] f) Determine whether the new value of (A+ΔAs) / Gs=(A+ΔAc) / Gc is closer to A than the corresponding previous value of (A+ΔAs) / Gs=(A+ΔAc) / Gc in step b), and if so, f1) setting the new value of (A+ΔAs) / Gs=(A+ΔAc) / Gc in step e) as the corresponding previous value of (A+ΔAs) / Gs=(A+ΔAc) / Gc; f2) Step b)

number

[0211] g) Increase k by 1.

[0212] h) repeating steps b) through g) until k reaches a maximum index greater than or equal to said initial value, for example in the range of 10 to 64, for example 10, 15, or 31;

[0213] i) M obtained in step d) of the last iteration φ and φ and / or cos(φ) and finally sin(φ) are used for position determination.

[0214] Thereby, the control unit is configured not to use the read / write storage means in which the digitized sine and / or cosine voltage signals are stored while performing said iterations.

[0215] In the conveyor of the present invention, the phase error α is preferably in the range of 0 to +0.06 radians, and the ratio DP / DP' (i.e., belt elongation) corresponds to a range of 1 to 1.00955. More preferably, the phase error α is in the range of 0 to +0.03 radians, and the ratio DP / DP' (i.e., belt elongation) corresponds to a range of 1 to 1.00477.

[0216] Furthermore, the conveyor of the present invention is designed to reduce the phase error load cos input ,sin input It is more preferable that the signal offsets C and S appearing in equations (1) and (2) are zero, and the ADC zero point Z appearing in equations (1) and (2) is also zero or corrected to zero.

[0217] Preferably, in the conveyor of the present invention, the control unit comprises an FPGA and / or a microcontroller.

[0218] In the conveyor of the present invention, the sensors generating the sine and cosine voltage signals may be Hall sensors generating signals from the intensity of the magnetic flux density B, or may be Wheatstone bridges generating signals from the direction of the magnetic flux density B, as described above. In the latter case, each sensor may comprise one or two Wheatstone bridges. Here, more preferred embodiments of this conveyor are as set out in claim 3 (one Wheatstone bridge for each sensor) and claim 4 (two Wheatstone bridges for each sensor).

[0219] For the purposes of the present invention, the absolute position of the position sensing element is understood as the absolute position of a point along the belt periphery in the middle of the interval between two magnetoresistive sensors included therein, relative to the end (or start) of the belt end junction zone, where the length of said interval including the two magnetoresistive sensors as end points is calculated as defined above.

[0220] The magnetic marker may typically be in the form of a matrix of elastomer, thermoplastic elastomer, or thermoplastic material containing particles of ferromagnetic material. The ferromagnetic material for the magnetic marker is not particularly limited as long as it satisfies the magnetic parameters described above. Preferably, the ferromagnetic material may be selected from the group consisting of alkaline earth metal ferrites (e.g., barium ferrite, strontium ferrite), alnico type, ferromagnetic chromium (IV) oxide type, and iron oxide type. Preferably, the ferromagnetic material is anisotropic.

[0221] The matrix material in which the particles of ferromagnetic material are embedded is either an elastomer, a thermoplastic elastomer or a thermoplastic. The most preferred materials for the matrix are elastomers, especially NBR or EPDM, or thermoplastic elastomers, especially TPU.

[0222] The filling rate of the ferromagnetic particles in the matrix is usually 60 to 90% by volume, preferably 60 to 80% by volume, and more preferably 65 to 75% by volume, based on the matrix in which the ferromagnetic particles are embedded.

[0223] The geometric thickness of the marker thus introduced is preferably 30 to 70% of the overall thickness of the belt, and more preferably 30 to 60%.

[0224] Such a magnetic marker generates a raw magnetic signal that can be detected by a magnetoresistive sensor when it is positioned such that there is a certain distance A (or air gap) between the surface of the magnetic marker and the center of the bridge of the Wheatstone bridge included in the magnetoresistive sensor (see Figure 1). The air gap A typically depends on, and is preferably approximately proportional to, the pole pitch DP. For example, the air gap A is typically 0.2 to 1.0 times the pole pitch DP.

[0225] The periodic pattern of magnetic markers may be provided in the form of a preformed strip of ferromagnetic particle markers embedded in a matrix of elastomeric material, optionally with a suitable backing material. Preformed strips of this type are commercially available in open-ended (linear) form for use in static (non-conveyor belt) applications. Examples at the time of filing this application include Plastiform® strips from Arnold Magnetic Technologies, Inc., Norfolk, Newark, USA, and Tromaflex® strips from Max Baermann Holding AG, Bergisch Gladbach, Germany. Preferably, the strip, having an elastomer or thermoplastic elastomer matrix containing particles of ferromagnetic material, is incorporated into the cover layer of the belt. More specifically, it can be embedded in grooves milled into the cover layer that correspond to the longitudinal grooves of the belt. After being incorporated into the belt, the strip may be welded with an adhesive or thermoplastic adhesive. Preferably, the cover layer and the strip of matrix material having the ferromagnetic material embedded therein are further coated with an overlayer to protect the strip from environmental dust and dirt.

[0226] The belts included in the conveyor of the present invention are conventional. They may be any type of single-layer or multi-layer fabric-reinforced or cord-reinforced belt suitable for inserting a periodic pattern of magnetic markers and suitable for end splicing from a double-ended to an endless belt. Splicing of the belt ends to make the belt endless is preferably accomplished using the so-called "finger end" method, in which both ends of an open-end belt are saw-tooth cut so that each tooth on one end can seamlessly engage a corresponding recess between two teeth on the other end, and vice versa. Once the teeth on one end of the belt are seamlessly engaged with the teeth on the other end, the two belt ends are welded together using heat and pressure, optionally with a hot-melt adhesive. This results in a zigzag belt splice across the entire width of the endless belt.

[0227] The conveyor of the present invention does not use additional fiducial markers to detect the blank zone, but relies solely on the magnetic markers themselves to detect the blank zone.

[0228] The conveyor of the present invention can be used in any field where belt positioning, especially high precision positioning, is required. The conveyor of the present invention may also be configured and used, for example, as a timing belt, a positive drive belt, a flat belt, a positive drive belt, or a power transmission belt. A preferred application for the conveyor of the present invention having alternating markers, especially periodic marker patterns, is in single-pass or multi-pass inkjet printers, especially for printing on textiles. Such printers are actually conveying devices that typically have at least a drive pulley and an idler pulley, and an endless conveyor belt looping around them, but also have the ability to simultaneously print on the substrate being conveyed.

[0229] 5 illustrates an exemplary embodiment of an inkjet printer incorporating the conveyor of the present invention. The printer (which may be a single-pass or multi-pass inkjet printer) includes four inkjet printheads 101, 102, 103, and 204, corresponding to the colors C, M, Y, and K. There is also a controller 6 that evaluates angle information received from position sensors 51, 52, 53, ... 5N (for a total of eight position sensors). In this embodiment, two position sensors are associated within controller 6 for each inkjet printhead. For example, two position sensors 101 and 102 (considered to have indexes 1 and 2) are associated within controller 6 with first printhead 101 (considered to have index 1), and two position sensors 53 and 54 (considered to have indexes 3 and 4) are associated within controller 6 with second printhead 102 (considered to have index 2). The diagram also shows the absolute position Pm of the mth print head (here m=4) relative to the end of the end join zone, assuming the belt moves in a clockwise rotation around the pulley. The control unit 6 generates position information for each print head from the two position sensors associated with that print head, and the positions are sent to that print head including a printer driver for timing the print heads 101, 102. Generating such timing based on belt position information and the image to be printed is well known in the art of inkjet printers, particularly in textile printing machines.

[0230] The present invention will now be further described with reference to examples.

[0231] Example 1: Synchronization error test with the conveyor of the present invention

[0232] A conveyor similar to that shown in Figure 4 is installed. The periodic pattern of magnetic markers 3 is in the form of a commercially available magnetic strip without magnetic markers, which has a spatial frequency ξ m is 200m -1 The belt has a length of L jis made endless to form an endless belt of 4 m length with an end joining (blank) zone of about 100 mm and attached to the conveyor 1. Then, the magnetic strip is fitted with a magnetic marker having a spatial frequency ξ m is 200m -1 The magnetic markers are provided on the entire circumference of the endless belt, except for a blank zone of approximately 100 mm, which is the belt end joining zone. A set of identical small, periodically repeating test images of approximately rectangular shape are then burned by a laser all around the longitudinal circumference of the conveying surface 23 of the belt 2. The number of test images Ni is 800, and they are spaced 5 mm apart from each other in the direction of belt movement. The spatial frequency ξ i is 200m -1 and is the same as the above-mentioned ξm.

[0233] Two position detection units (one on the upstream side and one on the downstream side based on the belt movement (loop) direction, and 1 / ξ) are provided on the upper part of the peripheral edge of the belt 2 that is not folded around the pulley 22, according to a preferred embodiment shown in the right part of FIG. 1 and the related description (AL780AMA, Sensitec). m an integer multiple of , i.e., 1 / ξ i the precisely known distance L, which is the same integer multiple of 12 Leave space between each other.

[0234] To detect the end joint (blank) zone of the conveyor belt 2, four Hall sensors (EQ-731L, manufactured by Asahi Kasei Microdevices Corporation) are arranged on the unbent portion of the outer periphery of the belt 2 on the pulley 22, so that each of the two position detection units is sandwiched between two Hall sensors. The two inner Hall sensors, corresponding to the end signal generating means of the upstream position detection unit and the start signal generating means of the downstream position detection unit, are spaced apart by a distance greater than the length of the blank zone, and accordingly, the length L12 is made even greater than the length of the blank zone. This corresponds to the most preferred embodiment B2.

[0235] The raw sine and cosine outputs of the two position sensing elements are each low-pass filtered, scaled by two op-amp instrumentation amplifiers, and converted to digital signals by two 18-bit ADCs to provide digitized sine and cosine output signals in the form of signed integers (Z=0) with a nominal amplitude A of 50,000 ADC units. These are processed by a control unit 6 similar to that outlined in FIG. 4 and the associated description, using error correction according to the second variant of the general specification and edge-join (blank) zone detection as also described in the general specification. The control unit 6 defaults to 400,000 m from the upstream position sensing element signal unless its output is ignored (see above). -1 In this case, the control unit 6 generates the rectangular wave signal from the signal of the downstream position detection unit. The spatial resolution of the rectangular wave signal is determined by the spatial resolutions 1 / ξm and 1 / ξ i The spatial resolution is 2000 times that of the position detection units, which is an integer multiple of these two spatial resolutions. Furthermore, a photographic camera focused on the conveying surface 23 of the belt 2 is attached near each of the two position detection units, and the two cameras are positioned at the same distance L as the position detection units. 12 Each camera is controlled by an individual synchronization station to take one photograph of the belt surface every 2000 pulses of the square wave signal, resulting in a spatial frequency of 400,000 m. -1 Taking into account this, a photo is taken every 5mm of belt travel, so each synchronized camera will take exactly one photo for each small test image that passes by the camera.

[0236] When the small test image pattern passes the first upstream position detector and camera, the associated synchronization station starts generating a trigger signal for the upstream camera every 2000 pulses of the square wave signal. The photos generated by the upstream camera are corrected for vignetting effects and brightness and stored in a computer. After the small test image pattern has completely passed the upstream camera and position detector, the photo-taking process by the upstream camera is stopped.

[0237] However, the synchronization station of the upstream camera also sends its first trigger signal to the synchronization station of the downstream camera. After receiving the trigger signal, the synchronization station of the downstream camera waits a predefined integer number of pulses from the square wave signal of the control unit 6. The predefined number is L 12 to 400,000m -1 After reaching the predefined number of pulses, the synchronization station of the downstream camera starts to generate a trigger signal for the downstream camera for every 2000 pulses of the square wave signal from the control unit 6. The photos generated by the downstream camera are also corrected for vignetting effect and brightness and saved in a computer. After the small test image pattern has completely passed through the downstream synchronization station, the photo acquisition process by the downstream camera also stops.

[0238] Therefore, theoretically, each photograph of the nth small test image taken by the upstream camera with a count of n × 2000 pulses of square wave signals is calculated by counting (n + L 12 ×400,000m -1 )×2000 pulses should produce an exactly corresponding (identical) photograph of the same test image taken by the downstream camera.

[0239] The conveyor 1 of the present invention thus prepared and installed has a nominal belt travel speed of 0.25 ms -1 This speed is the maximum number of photos the camera can take, i.e. 60 photos per second. -1 If a photo is taken every 5 mm of belt travel (see above), the time is 0.25 ms -150 photos are taken per second. The belt is rotated three times. Corresponding photos from the upstream and downstream cameras of a given small test image with index n (described in the previous paragraph) are compared by computer to determine whether the centers of the test images in the two corresponding photos are offset from each other by a distance Δx in the horizontal direction of the photos (corresponding to the direction of belt movement during conveyance). This distance Δx, typically measured in micrometers, is considered the synchronization error in the longitudinal direction (direction of belt movement) of the nth test image. All synchronization errors thus obtained are classified into groups with a width of 2 micrometers, ranging from -40 micrometers to +40 micrometers (i.e., the first group includes the number of synchronization errors observed in the range of -40 micrometers to -38 micrometers, and the second group includes the number of synchronization errors observed in the range of -38 micrometers to -36 micrometers). The number of synchronization errors observed in each group is divided by the total number of synchronization errors to determine the probability that the synchronization error belongs to a given group. The resulting probability groups are best fitted to a first Gaussian curve. The resulting group is shown as a hatched bar in Figure 7, with the first best-fit Gaussian curve superimposed on it. The X-axis of the graph in Figure 7 is the synchronization error (micrometers), and the Y-axis is the probability unit (0 to 1). The mean of the first best-fit Gaussian curve is -5.33 micrometers, and the standard deviation is 6.28 micrometers.

[0240] Comparative Example 2: Synchronization error test with conveyor using a conventional linear encoder

[0241] Instead of two position detection units and a control unit 6, the same spatial frequency 400000m -1 The setup of Example 1 is repeated, except that a commercially available linear encoder (Lika) is used to generate a square wave signal of . This commercially available linear encoder can measure the length L of the conveyor belt without losing position information. jThe edge junction zones of the linear encoder can be crossed. However, the exact structure and operating principle of this commercially available linear encoder are unknown. Similar to this comparative setup, the inventive setup of Example 1 also exhibits the same synchronization errors, which are again classified into probability groups as described in Example 1. The resulting probability groups are shown as white bars in Figure 7, with a best-fit second Gaussian curve superimposed on them. This second best-fit Gaussian curve has a mean value of +9.27 micrometers and a standard deviation of 10.96 micrometers, indicating that the synchronization errors are larger on average and more unevenly distributed than those of the inventive setup.

Claims

1. a) Belt peripheral portion (21, 22), outer conveying surface (23), inner pulley opposing surface (24), length L ｊ An endless conveyor belt (2) comprising magnetic markers (3) along the entire belt periphery (21, 22) except for one or more blank zones (4), wherein j is the index of the blank zone, and the magnetic markers (3) are arranged in the belt movement direction such that the N poles (31) or S poles (32) of the magnetic markers (3) alternately face the outer conveying surface (23), and adjacent N poles and S poles are spaced apart from each other by a pole pitch DP to generate an alternating magnetic field of a predetermined maximum intensity, and in each of the blank zones (4), the magnetic markers either do not generate a magnetic field or generate a magnetic field of a maximum intensity lower than the predetermined maximum intensity, b) At least two position detection units (51, 52, 53, 54, 55, 56, 57, 58) positioned spatially near the magnetic markers (3) along the belt periphery (21, 22), wherein the position detection units detect a magnetic field from magnetic markers (3) that are not within the blank zone (4) when the conveyor belt (2) is moving in the direction of movement, and 1) a sine corresponding to the angular position of the position detection unit with respect to the pattern with respect to the direction of the magnetic flux density B detected from the magnetic markers 1) Generates a wave voltage signal, and 2) generates a cosine wave voltage signal corresponding to the angular position of the position detection unit with respect to the pattern, and of the at least two position detection units (51, 52, 53, 54, 55, 56, 57, 58), there are at least two spaced position detection units (51, 52, 53, 54, 55, 56, 57, 58), and one of the two spaced position detection units determines the length L of an arbitrary blank zone (4) from any one of the remaining spaced position detection units. ｊ At least two position detection units (51, 52, 53, 54, 55, 56, 57, 58) are arranged at intervals equal to a peripheral distance in the belt movement direction that is greater than the belt movement distance, c) Means (51, 52, 53, 511, 521, 531, 532, 533, 534) capable of generating a blank zone entry signal when the maximum magnetic field strength detected by one position detection unit decreases or is about to decrease to a threshold strength lower than the predetermined maximum strength, and means (512, 522) capable of generating a blank zone exit signal to the same position detection unit when the maximum magnetic field strength detected by the same position detection unit increases or is about to increase to the threshold strength, d) A control unit (6) that derives belt position information from the sinusoidal voltage signals and cosine voltage signals of all position detection units (51, 52, 53, 54, 55, 56, 57, 58) when the conveyor belt (2) is traveling in the direction of movement, wherein the control unit (6) ignores the sinusoidal voltage signals and cosine voltage signals from any position detection unit from the time it receives the blank zone entry signal to that position detection unit, and reconsiders the sinusoidal voltage signals and cosine voltage signals from that position detection unit from the time it receives the blank zone departure signal to that position detection unit, A conveyor (1) equipped with [a specific feature].

2. There are at least three position detection units (51, 52, 53, 54, 55, 56, 57, 58), and at least three of these position detection units (51, 52, 53, 54, 55, 56, 57, 58) are spaced apart, and one of the at least three spaced-apart position detection units is connected to any one of the remaining spaced-apart position detection units, L ｊ The conveyor (1) according to claim 1, which is arranged with a gap equal to a peripheral distance in the direction of belt movement that is greater than the distance of the belt movement.

3. Each position detection unit (51, 52, 53, 54, 55, 56, 57, 58) i) Two Wheatstone bridges (513, 514), each having an upper leg and a lower leg, respectively, of magnetoresistive sensors operated by anisotropic magnetoresistive (AMR), arranged apart from each other by a spacing distance DP / 4, wherein the position sensing units (51, 52, 53, 54, 55, 56, 57, 58) are positioned in the spatial vicinity of the magnetic markers (3) such that the direction of movement of the conveyor belt (2), which is the direction of the pattern of the magnetic markers, is parallel to the spacing distance, and the conveyor belt (2) travels in the direction of movement, one of the two Wheatstone bridges (513, 514) generates the sinusoidal voltage signal from the angle between the direction of the magnetic flux density B and the direction of the pattern of the magnetic markers, and the other of the two Wheatstone bridges (513, 514) generates the corresponding cosine voltage signal, The conveyor (1) according to claim 1, comprising:

4. Each position detection unit (51, 52, 53, 54, 55, 56, 57, 58) A first pair of Wheatstone bridges (513, 514) are spaced apart by DP / 4 from each other so as to generate a sine wave signal and a corresponding cosine wave signal from an alternating pattern of north and south poles, A second pair of Wheatstone bridges (515, 516), spaced apart by DP / 4, are arranged to generate a sine wave signal and a corresponding cosine wave signal from the alternating pattern of north and south poles. Equipped with, The first pair and the second pair of Wheatstone bridges (513, 514, or 515, 516) are arranged with a gap of DP between them. One of the first pair and the second pair of Wheatstone bridges (513, 514, or 515, 516) has a power supply polarity that is inverted with respect to the power supply polarity of the other of the first pair and the second pair of Wheatstone bridges (515, 516, or 513, 514). The first Wheatstone bridge (515) of the second pair of Wheatstone bridges (515, 516) detects the north pole when it is on the south pole (32), and simultaneously the first Wheatstone bridge (513) of the first pair of Wheatstone bridges (513, 514) detects the north pole when it is on the north pole (31). The first Wheatstone bridge (515) of the second pair of Wheatstone bridges (515, 516) detects the south pole (32) when it is on the north pole (31), and simultaneously the first Wheatstone bridge (513) of the first pair of Wheatstone bridges (513, 514) detects the south pole when it is on the south pole (32). The signal from the first Wheatstone bridge (513) of the first pair of Wheatstone bridges (513, 514) and the signal from the first Wheatstone bridge (515) of the second pair of Wheatstone bridges (515, 516) are added together and optionally averaged. The signals from the second Wheatstone bridge (514) of the first pair of Wheatstone bridges (513, 514) and the signals from the second Wheatstone bridge (516) of the second pair (515, 516) are added together and optionally averaged. The conveyor (1) according to claim 1.

5. Each position detection unit (51, 52, 53, 54, 55, 56, 57, 58) generates a blank zone entry signal when the maximum strength of the detected magnetic field falls below a threshold strength lower than the predetermined maximum strength, and generates a blank zone exit signal when the maximum strength of the detected magnetic field rises above the threshold strength. The conveyor according to claim 1.

6. It is possible to form the sum of the squares or the square root thereof of the intensities of the sine wave signal and cosine wave signal detected by each position detection unit (51, 52, 53, 54, 55, 56, 57, 58), and when the sum of squares or the square root from the position detection unit falls significantly below the predetermined intensity, a blank zone entry signal can be generated to the position detection unit, and when the sum of squares or the square root from the position detection unit rises above the predetermined intensity, a blank zone exit signal can be generated to the position detection unit. The conveyor according to claim 5.

7. Each position detection unit (54, 55) A start signal generating means (511, 521) is capable of generating a blank zone entry signal when the maximum intensity of the magnetic field detected by the start signal generating means (511, 521) falls below a threshold intensity lower than the specified maximum intensity, Termination signal generating means (512, 522), which is capable of generating a blank zone exit signal when the maximum intensity of the magnetic field detected by the termination signal generating means (512, 522) rises to or above the threshold intensity, Equipped with, The start signal generating means and the end signal generating means are positioned so that they straddle the corresponding position detection unit, with each start signal generating means (511, 521) located upstream of the corresponding position detection unit (54, 55), and each end signal generating means (512, 522) located downstream of the corresponding position detection unit (54, 55). In the at least two spaced position detection units (54, 55) described above, any pair of spaced upstream and downstream position detection units (54, 55) has an end signal generating means (512) for the upstream position detection unit (54) of the pair, which is spaced at a distance L longer than the maximum length Lj of any blank zone (4) present on the conveyor belt (2), from the start signal generating means (521) of the downstream position detection unit (55) of the pair. The terms "upstream side" and "downstream side" refer to the direction of movement of the conveyor belt. The conveyor according to claim 1.

8. Start signal generating means (511, 521) corresponding to specific position detection units (54, 55) and end signal generating means (512, 522) corresponding to the specific position detection units (54, 55) are arranged at a distance LH from each other given by the following equation, 【Number 1】 N is a positive integer at least zero, and is large enough to be sandwiched between the start signal generating means (511, 521) and the end signal generating means (512, 522). The conveyor (1) is capable of forming the sum of the squares of the signal intensities detected by a pair of start signal generating means and end signal generating means (511, 512, or 521, 522) arranged at intervals, and when the sum of the squares of the signal intensities falls below the predetermined intensity, it is capable of generating the blank zone entry signal to the sandwiched position detection units (54, 55), and when the sum of the squares of the signal intensities rises to or above the threshold intensity, it is capable of generating the blank zone departure signal to the sandwiched position detection units (54, 55). The conveyor according to claim 7.

9. A first start signal generating means (531) is located upstream of all position detection units (56, 57, 58) and is capable of generating a blank zone entry signal to the upstreammost (but preferably adjacent) position detection unit (56) when entering the blank zone (4), A second start signal generating means (534) located downstream of all position detection units (56, 57, 58) and capable of generating a blank zone entry signal which is used or interpreted as a blank zone entry signal to the furthest downstream (but preferably adjacent) position detection unit (58) when entering the blank zone (4), A further number of start signal generating means (532, 533) that is one less than the number of position detection units (56, 57, 58), wherein a pair of position detection units (56, 57, or 57, 58) is sandwiched between them, Equipped with, The further start signal generating means (532, 533) are, When entering the blank zone (4), a blank zone entry signal can be generated for the position detection units (57, 58) sandwiched between the downstream (but preferably adjacent) of the further start signal generating means, and this signal is used or interpreted as a blank zone departure signal for the position detection units (56, 57) sandwiched between the upstream (but preferably adjacent) of the further start signal generating means, and any further start signal generating means (532, 533) is positioned at a distance L from the corresponding upstream sandwiched position detection unit (56, 57). ｍ L is the maximum length of any blank zone (4) present on the conveyor belt (2). ｊ A distance L greater than that, at which the second start signal generating means (534) is positioned apart from the downstream position detection unit (58). ｎ L is the maximum length of any blank zone (4) present on the conveyor belt (2). ｊ Larger than that, "upstream" and "downstream" refer to the direction of movement of the belt. The conveyor according to claim 1.

10. a) The first start signal generating means (531) and one further start signal generating means (532) are arranged with the upstream position detection unit (56) in between them, and are spaced apart from each other by a distance LH. b) The second start signal generating means (534) and one further start signal generating means (533) are arranged with the downstream position detection unit (58) in between them, and are spaced apart from each other by a distance LH. c) Another pair of further start signal generating means (532, 533) are arranged with another position detection unit (57) in between them, spaced apart from each other by a distance LH. The distance LH can be calculated using the following formula: [Math 2] N is a positive integer at least zero, and is large enough to accommodate the position detection unit sandwiched between them. The conveyor (1) is i) It is possible to form the sum of the squares or the square root thereof of the signal intensities detected by the first start signal generating means (531) and the one further start signal generating means (532) arranged at a distance from each other, and when the sum of the squares of the signal intensities detected by the first start signal generating means (531) and the one further start signal generating means (532) arranged at a distance from each other falls significantly below the predetermined intensity, it is possible to generate a blank zone entry signal to the uppermost position detection unit (56) sandwiched between them, and when the sum of the squares of the signal intensities rises above the threshold intensity, it is possible to generate a blank zone exit signal to the uppermost position detection unit (56) sandwiched between them, ii) It is possible to form the sum of the squares of the signal intensities detected by the second start signal generating means (534) and the further start signal generating means (533) arranged at intervals, or the square root of said sum of the squares, and when the sum of the squares of the signal intensities detected by the second start signal generating means (534) and the further start signal generating means (533) arranged at intervals, or the square root thereof, falls significantly below the predetermined intensity, a blank zone entry signal can be generated to the furthest downstream position detection unit (58) that is sandwiched, and when the sum of the squares of the signal intensities or the square root thereof rises above the threshold intensity, a blank zone departure signal can be generated to the furthest downstream position detection unit (58) that is sandwiched, iii) It is possible to form the sum of the squares or the square root thereof of the intensities of signals detected by any pair of further start signal generating means (532, 533), and when the sum of the squares or the square root thereof of the intensities of signals detected by a pair of further start signal generating means (532, 533) arranged at intervals falls significantly below the predetermined intensity, a blank zone entry signal can be generated for the position detection unit (57) sandwiched between them, and when the sum of the squares or the square root thereof of the intensities of the signals rises above the threshold intensity, a blank zone exit signal can be generated for the position detection unit (57) sandwiched between them. "Upstream" and "downstream" refer to the direction of movement of the belt. The conveyor according to claim 9.

11. The conveyor (1) according to claim 7, wherein the first start signal generating means (531) and / or the second start signal generating means (534) are Hall sensors.

12. The conveyor (1) according to claim 9, wherein the first start signal generating means (531), the second start signal generating means (534), and / or the further start signal generating means (532, 533) are Hall sensors.

13. The conveyor (1) according to any one of claims 1 to 12, wherein there are no other markers for detecting the blank zone (4) other than the magnetic marker (3).

14. The conveyor (1) according to claim 1, comprising 3 to 16 position detection units (51, 52, 53, 54, 55, 56, 57, 58).

15. The conveyor (1) according to claim 1, wherein the position detection units (51, 52, 53, 55, 56, 57, 58) are located near, above, or below, the portion (22) of the conveyor belt (2) that is not bent on the pulley.

16. The conveyor (1) according to claim 1, wherein the DP is in the range of about 0.5 mm to about 10 mm, preferably about 0.5 mm, about 1 mm, about 2 mm, about 2.5 mm, about 5 mm, or about 10 mm.

17. The conveyor (1) according to claim 1, wherein the magnetic markers form an essentially periodic pattern in the direction of belt movement.

18. A single-pass or multi-pass inkjet printer comprising the conveyor (1) described in claim 1.

Images