Processing

Please wait...

Settings

Settings

Goto Application

1. WO2020115649 - METHOD AND SENSOR FOR ROTATION MEASUREMENT OF AN ELONGATED OBJECT

Note: Text based on automatic Optical Character Recognition processes. Please use the PDF version for legal matters

[ EN ]

METHOD AND SENSOR FOR ROTATION MEASUREMENT OF AN ELONGATED OBJECT

TECHNICAL FIELD

The invention relates to a method and device for contactless measurement of (axial) rotation and/or rotational speed of an (electrically) non-conductive or poorly conductive elongated object, and in a preferred embodiment yarn during a twisting process, and optimisation of the twisting process on yarns, through improved insight into the real time parameters of the twisted and to be twisted yarn during the process. It goes without saying that the further aspects of optimisation and improved insight can also be applied to processes on elongated objects other than yarns.

PRIOR ART

The present invention relates to a method and detection device for determining rotation of elongated objects, in particular at high rotational speeds and possibly also with longitudinal movement of the objects.

In particular, the invention applies to the twisting of yarns, and methods and systems or devices for twisting them.

Existing systems for producing twisted yarn are all based on an identical principle. Imposing a torsion or twist on multiple yarns in a device for this purpose, so as to produce twisted yarns. In a next phase, these yarns can untwist and thus twist into one another (i.e. twining or cabling).

One of the problems in applying the torsion is that little or no in-line control of the twist level or rotation of the yarns is possible during the process itself, making adjustment thereof virtually impossible, and then only with a serious delay (since this is based on an evaluation of the end product).

In addition, and more importantly, it is hitherto impossible to quantify without contact the undergone rotation of a yarn, or in particular, the torsion present over segments of the yarn (typically over a very short length, all the more so as to avoid too much variation in local rotation), during the twisting process itself. However, it is crucial to finally reach a certain torsion level on the segments of the yarn, and thus the (degree of) manipulation of the yarn must be adjusted to the condition (already present torsion) of the yarn itself. If the yarn is already partially provided with varying torsions over different segments, this must be taken into account in the further manipulation of the yarn, so as to ultimately obtain a desired torsion over the segments. This adaptation of the manipulation (i.e. the application of an air current to obtain torsion) thus requires knowledge of the initial state (or intermediate state) of the yarn, which is currently impossible to measure without contact.

Systems and methods in the prior art are found, for example, in US 4,349,784, EP 0,610,147, US 4,399,648 and EP 1,033,579.

The applicant intends to solve this problem by, on the one hand, providing a system capable of measuring the rotation or rotational speed of the yarns directly and without contact, and on the basis of which it is possible to determine local twist levels on the yarn, and/or over a length thereof. With this data the torsion devices can then be adjusted in terms of the torsion exerted on the yarns, so as to obtain the desired twist level on end and intermediate products. The applicant notes that there is a lack of equipment for contactless measurement of these matters, but also that as a result of this lack, consideration has not yet been given to adjusting torsional processes on the basis of data relating to the (almost) real time rotation or rotational speed of the yarns.

SUMMARY OF THE INVENTION

The invention relates to an improved method for contactless measurement of local rotation of an (rotating) elongated object, preferably a yarn, the method comprising the following steps:

a. guiding the object along a longitudinal axis through at least one detection arrangement comprising a plurality of electric charge detecting components arranged in a first substantially circular pattern around the longitudinal axis at equal mutual distances, and comprising a plurality of grounding electrodes equal to the plurality on electric charge detecting components, which grounding electrodes are arranged in a second substantially circular pattern around the longitudinal axis at equal mutual distances, the second substantially circular pattern staggered with respect to the first substantially circular pattern, the grounding electrodes functioning as a screen between the successive charge detecting components and thus act as a spatial filter;

b. detecting one or more electrical charges on the object by a plurality of the charge detecting components, and processing the detection into an output signal;

c. combining the output signals of each of the charge detecting components (according to an identical time schedule, in other words with the same to);

d. deriving one or more, preferably one, dominant frequencies from the combined output signals.

In a second aspect, the invention relates to a device for contactless measurement of local rotation of an elongated object extending substantially along one dimension, preferably a yarn, comprising one or more detection arrangements, the detection arrangement each comprising a plurality of electric charge detecting components arranged in a first substantially circular pattern about a longitudinal axis and comprising a plurality of grounding electrodes equal to the plurality of electric charge detecting components, which grounding electrodes are arranged in a second substantially circular pattern around the longitudinal axis, the second substantially circular pattern staggered with respect to the first substantially circular pattern.

In a third aspect, the invention relates to a method for contactless determination of undergone rotation of a yarn extending substantially along a longitudinal axis at a predetermined longitudinal point on a yarn during the twisting of said yarn as described in the claims.

In a fourth aspect, the invention relates to a method for applying a (preferably alternating S and Z) torsion to a yarn via a substantially tangential air current at a fixed location relative to (possibly longitudinally moving) yarn as described in the claims.

In a fifth aspect the invention relates to an improved system for manufacturing an (S/Z) twisted yarn from a (substantially torsion-free) yarn as described in the claims.

In a sixth aspect, the invention relates to the use of the aforementioned methods and/or devices and/or systems according to the invention in controlling the twisting process.

DESCRIPTION OF THE DRAWINGS

Figures 1A-B show a perspective view of an embodiment for a device for determining local rotation on an elongated object.

Figure 2 shows an unrolled representation of the embodiment of Figure 1.

Figure 3 shows a longitudinal section of the embodiment of Figure 1.

4A-B shows a schematic representation of an antenna arrangement for detecting electric charges, and the resulting signal.

5A-B shows a schematic representation of an antenna arrangement for detecting electric charges, and the resulting signal.

6A-B shows a schematic representation of an antenna arrangement for detecting electric charges, and the resulting signals.

7A-B shows a schematic representation of an antenna arrangement for detecting electric charges, and the resulting signals.

DETAILED DESCRIPTION

The invention relates in a first aspect to a method and device for contactless measurement of local rotation of an (electrically) non-conductive or poorly conductive rotating, elongated object, preferably a yarn, according to the claims. The method is particularly suitable for measuring the (undergone) rotation of yarns that are twisted at very high speeds, both in terms of longitudinal movement of the yarn and of rotational movement. In a situation such as this it is impossible to interact with the yarn due to the high speeds (yarn could be damaged, untwisted, etc.), all the more so since this measurement should preferably take place during the twisting process. In addition, each contact would also influence the measurement process itself, and thus invalidate the results.

From this improvement, the applicant has continued to optimise the twisting process of yarns, and has developed further methods and systems for this purpose, using the new information obtained by the above method and device for contactless measurement of local rotation and/or rotational speeds.

The applicant first noted that in the current prior art there is no sensor capable of measuring the torsion level of yarns moving longitudinally at high speed (in approximate real-time). In view of the speeds at which yarns typically move during the twisting (order of magnitude of a few hundred meters per minute), the (approximate) measurement must be able to take place almost instantaneously. Given this problem, the applicant decided to approach the problem in a different way, namely by measuring the rotational speed or local rotation of yarns. Based on this, calculations can then be made back to an approximate torsion level of the yarn.

On the basis of this knowledge, the entire twisting process can then be optimised, in order to obtain a product that is twisted as uniformly and as accurately as possible.

Unless otherwise defined, all terms used in the description of the invention, including technical and scientific terms, have the meaning as commonly understood by a person skilled in the art to which the invention pertains. For a better understanding of the description of the invention, the following terms are explained explicitly.

In this document, 'a' and 'the' refer to both the singular and the plural, unless the context presupposes otherwise. For example, 'a segment' means one or more segments.

When the term 'around' or 'about' is used in this document with a measurable quantity, a parameter, a duration or moment, and the like, then variations are meant of approx. 20% or less, preferably approx. 10% or less, more preferably approx. 5% or less, even more preferably approx. 1% or less, and even more preferably approx. 0.1% or less than and of the quoted value, insofar as such variations are applicable in the described invention. However, it must be understood that the value of a quantity used where the term 'about' or 'around' is used, is itself specifically disclosed.

The terms 'comprise', 'comprising', 'consist of', 'consisting of', 'provided with', 'have', 'having', 'include', 'including', 'contain', 'containing' are synonyms and are inclusive or open terms that indicate the presence of what follows, and which do not exclude or prevent the presence of other components, characteristics, elements, members, steps, as known from or disclosed in the prior art.

The term 'yarn' refers to a spun yarn, in this case comprising several filaments, or BCF (bulked continuous filament) yarn. The individual yarns typically have a diameter comprised between 0.2 mm to 2 mm, the already twisted yarns have a larger diameter, comprised between 0.5 mm to 5 mm, depending on the circumstances. It is also noted here that BCF yarn is compressible and that, therefore, yarn numbers preferably indicate the diameter or thickness of the yarn, as the ratio of the mass and length of a piece of yarn. In practical terms, this comes down to a range of 250 dtex to 4000 dtex for individual yarns, and for twisted yarns from 2000 dtex to 20000 dtex. Narrower ranges are possible here, for example 600 dtex to 2000 dtex for individual yarns, and 2000 dtex to 10000 dtex for twisted yarns, but this in no way limits the applicability of the invention.

The terms 'twine' and 'twined' refer to the procedure, or a characteristic of the product thereof, in which one or more yarns twist around each other or around each other with another set of one or more yarns.

The term 'twist' and 'twisted' refers to the procedure, or a characteristic of the product thereof, in which torsion is applied to a yarn, leading to a deformation whereby the energy of the torsion is stored on the yarn, and visually leads to a twisted yarn.

The term 'entangling' or 'interlacing' refers to the joining of several separate yarns, or several separate, twisted yarns, wherein the yarns comprise multiple filaments. In entangling, the yarns are joined by entangling some of these filaments with each other over a limited length, for example, by closely bringing the individual yarns together and then supplying an air current pulse, thus performing the filament entanglement via an air vortex.

The term 'cabled' refers not only to the traditional process of laying yarns around each other, but also to a product obtained by twining two or more yarns that have already been twisted.

The term 'connected alternating S/Z twined yarn' refers to a yarn made by the antiphase combination of alternating S/Z twined yarns, and connecting them in the short zones without torsion. In this case no self-twisting occurs because the connected yarns have an opposite torsion. The opposing torsions counteract each other and prevent the yarn from untwisting.

The term 'alternating S/Z twisted' and 'alternating S and Z twisted' refers to the condition of a yarn on which a spatially alternating torsion is applied.

The terms 'alternating S and Z twined' and 'alternating S/Z twined' refers to yarns twisted around each other as a result of applying an alternating S/Z torsion to the yarns and the subsequent self-twisting of the yarns together.

The term 'electric charge detecting components' refers to an instrument that quantifies the proximity of one or more electric charges, in particular by detecting the field generated by the electric charge.

The term 'grounding electrodes' refers to electrodes that are electrically connected to the 'common' or are grounded. Because they are grounded, the grounding electrodes

draw the lines of force of the generated field towards themselves through the charges on the yarn/elongated object, and shield the charge detecting components. However, this effect depends on the distance between the charges and the grounding electrodes, whereby the grounding electrodes can only/mainly shield the charge detecting components from charges that are close to the grounding electrodes, and further from the charge detecting components. With charges that are close to or 'in front of' the charge detecting components, this shielding effect is much less and insufficient to conduct the lines of force, so that the charge detecting components can only effectively perceive the charges when they are at the opening between two consecutive grounding electrodes, and thus right in front of the charge detecting component.

The term 'staggered' with respect to the two sets of components (charge detecting and grounding electrodes) arranged in concentric, circular patterns, refers to a situation where, when projecting the components of the sets onto a third concentric, circular pattern, the components of the two sets alternate (ABABABA ...), as can also be clearly seen in the Figures. This can further also be considered as that the set of charge detecting components are suitable as a first, substantially regular n-angle around the longitudinal axis as a centre point. The set of grounding electrodes is also suitable as a second, substantially regular n-angle around the longitudinal axis as the centre point, wherein a ray from the centre point through each angular point of the n-angle cuts a separate side of the second n-angle, preferably substantially around the centre of said side.

The term 'local rotation' refers to the rotation around an axis (internal to the object). This is particularly clear for yarns. Where the prior art deals with rotation around an external axis (virtual axis along which the yarn is guided), this is actually the so-called ballooning that is monitored, which does not pose any major technical difficulties due to the clearly observable physical effects as well as the lower speeds. In addition, however, the yarn also turns itself around an internal axis, which is what the invention wishes to monitor, and which creates a technically much more challenging problem.

Quoting numerical intervals by endpoints includes all integers, fractions and/or real numbers between the endpoints, these endpoints included.

The text refers to the determination of one or more dominant frequencies. However, it is to be expected that only one dominant frequency will occur, namely that associated with the rotational speed of the object.

In what follows, specific reference will often be made to yarns. However, it should be taken into account that the methods, devices and systems are applicable to all elongated objects that undergo rotation, and that the term 'yarns' also partly serves for simplicity of notation, as well as for a better representation of the inventions. Thus, in most situations, the term can be replaced by 'elongated object', preferably threadlike, while the invention, in its most preferred embodiments, is aimed at yarns.

It is further to be understood that the invention applies to non-or poorly conductive objects, where charges remain substantially static (relative to the speeds at which the objects themselves rotate).

In a first aspect, the invention relates to a method according to claim 1.

The applicant notes that object/yarn carries a multitude of electrical charges on the outside thereof (among other things by friction, similar to static electricity), the charges having a substantially fixed position on the yarn, and thus rotating at a certain speed around the axis of the rotating yarn. By providing a plurality of charge detecting components around the yarn, these charges can be detected and thus a rotational speed can this be derived from the yarn, which is impossible with existing techniques without interference with the yarn (which is absolutely avoidable).

The elongated objects, typically yarns, contain a very large multitude of electric charges on their surface, which are stochastically distributed over them. Because of this imperfect distribution, the charge distribution has a certain profile that acts as a fingerprint or pattern. If this imperfect distribution then moves along a spatial filter (see below), and is detected by it, a single, dominant frequency can be derived from the resulting output signals.

In particular, the applicant therefore uses a so-called 'spatial filter', whereby the individual charge detecting components or antennas can only substantially detect a passing charge for a very short time, because they are shielded by the earthing electrodes. In this way, the passage of an electric charge along an antenna produces a signal that only runs over a very short time (with relevant amplitude). Because the multiple antennas can detect the passage separately, it is possible to use the composite signal from all antennas (or on the basis of the signals separately assuming a shared to) the time can theoretically be determined between passages at the individual antennas, by means of the repetition of the 'profile' or 'fingerprint' of the charge distribution, and thus the rotational speed of the charge. Shielding by the grounding electrodes (spatial filter) has already been described somewhat above, and the general principle can also

be found, for example, in EP 1,033,579 where a longitudinal spatial filter is used to measure longitudinal velocities contactlessly via charge detection.

In view of the large amount of electric charges on the yarn, and the high rotational speed of the yarn, it is necessary to provide sufficient separate charge detecting components that are sufficiently sealed off from each other, so as to obtain, as it were, pulsed signals upon passage of a charge to a charge detecting component. Without the shielding of the grounding electrodes, which are grounded, a smeared signal would be obtained at each of the antennas, from which a large number of frequencies would follow if a DFT (Discrete Fourier Transform), preferably an FFT (Fast Fourier Transform) or DTFT (Discrete-Time Fourier Transform) would be already have been performed on them. Due to the grounding electrodes, only when a charge passes almost straight in front of the charge detecting component does the output signal become significant (when the charge is thus at a closest position to the charge detecting component), the amplitude furthermore decreasing very rapidly in the case of deviation relative to this closest position (by shielding by grounding electrodes). The shielding greatly limits the length/duration of a detection signal (or at least the length over which the amplitude is sufficiently high) by taking into account the signals from the various charge detecting components.

This further contributes (in addition to the spatial filter functionality of this arrangement) to the preference for combining the different output signals of the individual charge detecting components, so that the composite signal can be processed more easily, in particular to determine the periodicity/dominant frequencies.

On the basis of the known positions of the charge detecting components (angle with respect to the axis along which the object is guided, radius of the circular arrangement), the local rotation over a certain time can then be found at the level of the detection arrangement. In principle, a local rotational speed can thus also be determined on the basis of the number of charge detecting components and dominant frequency (or the period, inverse of the dominant frequency, between successive pulses).

The charge detecting components and the grounding electrodes, which are arranged in a circular pattern, thus together form a cylindrical envelope of the object over a certain length, as can also be seen in the Figures.

In a preferred embodiment, the dominant frequencies are derived by applying a Discrete Fourier Transform (DFT), preferably a Fast Fourier Transform (FFT), or DTFT, to the combined output signals over a certain period of time.

FFT is particularly suitable for determining the period of the periodic signal generated by the detection arrangement.

In a preferred embodiment, the dominant frequencies are derived from the merged output signals by measuring the number of peaks in the merged output signals over a given period of time.

The applicant notes that with the elaboration of the detection arrangement with the spatial filter by alternating charge detecting components and grounding electrodes, the merged signal can also be analysed more directly (possibly without DFT/FFT/DTFT), simply by 'counting' the peaks. However, it should be mentioned that DFT/FFT/DTFT will be more reliable in this case, and more practical, and therefore preferred. Preferably, such an embodiment will also use an arrangement with two groups of charge detecting components, as described below, and/or wherein the merged output signals are subjected to a small band filter around the dominant frequency after determining the dominant frequency. The resulting, purer merged output signals allow a much simpler further analysis.

In a preferred embodiment, the electric charge detecting components comprise two separate groups, the electric charge detecting components of the two groups being arranged in alternating succession in the first circular pattern at equal distances from each other, and the dominant frequencies being determined on the basis of the difference of the output signals (i.e., a differential signal, not to be confused with derivative) of the first group and the second group, preferably wherein the electric charge detecting components of the first group are connected in series, and wherein the electric charge detecting components of the second group are connected in series. Note that this leads to an arrangement in which a grounding electrode (R) is always present between the charge detecting components of the first (A) and second group (B), which leads to the following sequence: A R B R A R B R A R B R ...

Working with the difference of the output signals or differential signal leads to a signal that is less affected by electrical interference fields from outside the sensor, since these will be subtracted from each other. Preferably, the absolute value of the differential signal is then also taken. This is also represented in the Figures.

In a preferred embodiment, the local rotation is calculated on the basis of the dominant frequencies and the distance between successive charge detecting components and/or grounding electrodes.

On the basis of the dominance frequency, it is easy to deduce what the time difference is between the detection of a charge by successive charge detecting components. Knowledge of the positioning of the charge detecting components and/or grounding electrodes makes it possible to determine which distance or angle has been traversed

in this time difference, with which the local rotation can subsequently be determined from a point on the yarn during its passage through (i.e. over the length of) the detection arrangement.

In a preferred embodiment, the merged output signals are subjected to a small band filter, which small band filter is centred around the dominant frequency (e.g. derived by means of a DFT/FFT/DTFT of the merged output signal), and where local rotation of the elongated object is quantified on the basis of the filtered output signals. By performing this filtering, the resulting combined signal is purified (from DC components, from low-frequency signals, for example from lighting, surrounding electronics, etc.), and the number of angular rotations (i.e. through an angle between two consecutive charge detecting components) can be very clearly are counted (number of pulses of the signal) that the yarn has undergone during passage through the detection arrangement. It is in particular this further filtering that allows to accurately quantify the local rotation. Namely, the dominant frequency gives a view of the local rotational speed at which the yarn turns, but the purified signal makes it possible to specifically derive the effective local rotation, which is not possible with the still impure first signal. By purifying the signal via the small band filter, the individual pulses can be easily detected, and thus a certain point on the yarn itself, the amount of rotation (local undergone rotation) can be determined over the path along the sensor device.

In a second aspect the invention relates to a device for contactless measurement of local rotation of an elongated object extending substantially along one dimension, preferably a yarn, comprising one or more detection arrangements, the detection arrangements each comprising a plurality of electric charge detecting components arranged in a first substantially circular pattern about a longitudinal axis and comprising a plurality of grounding electrodes equal to the plurality of electric charge detecting components, which grounding electrodes are arranged in a second substantially circular pattern around the longitudinal axis, the second substantially circular pattern staggered with respect to the first substantially circular pattern, wherein the grounding electrodes act as a screen between the successive charge detecting components and thus act as a spatial filter, said device further comprising a processor coupled to the one or more detection arrangements, and said processor being adapted to determine deriving one or more, preferably one, dominant frequencies from output signals of the detection arrangements.

The advantages of such a sensor device follow from the method already discussed.

In a preferred embodiment, the first and second circular pattern have a substantially equal radius, and together form a single circular pattern.

In an alternative embodiment, the first circular pattern has a substantially greater radius than the second circular pattern.

The grounding electrodes will in both cases act as a screen between the successive charge detecting components, and thus act as a spatial filter. The applicant noticed that such a shielding effect already occurs when the components and the grounding electrodes are at an identical distance from the axis along which the elongated object is guided.

In a preferred embodiment, the electric charge detecting components comprise two separate groups, the electric charge detecting components of the two groups being arranged in alternating succession in the first circular pattern at equal distances from each other, and the dominant frequencies being determined on the basis of the difference of the output signals of the first group and the second group, preferably wherein the electric charge detecting components of the first group are connected in series, and wherein the electric charge detecting components of the second group are connected in series.

As already indicated in the method, preferably two separate groups of charge detecting components are used, which are alternately placed in the first circular pattern (see previously described). Because both groups of external sources will measure almost the same, the two signals can be subtracted from each other to almost eliminate this noise from external sources.

More preferably, the antennas of the first group extend from a first common conductor located at a first longitudinal end of the detection device, and (at least) partially extending around the periphery of the first circular pattern at said first longitudinal end. The antennas of the second group extend from a second common conductor, which is located at the second longitudinal end (opposite to the first) of the detection device, and (at least) partially runs around the circumference of the first circular pattern at said second longitudinal end. The first group and second group are pushed together in a staggered manner, as visible in the unrolled Figures.

The grounding electrodes are preferably one single antenna, which runs along and between the antennas of the first and second group. Note that said arrangement makes it necessary to give the antennas sufficient length (see below), since measurements at the first and second ends may contain errors due to the presence of the common conductors. The measurement values from the 'belly' between the two ends are much more reliable, and will allow the errors to be removed by filtering out the ends as noise.

In a preferred embodiment, the electric charge detecting components comprise antennas, which essentially extend parallel to each other perpendicular to the first circular pattern over a predetermined distance.

In a preferred embodiment, the antennas extend along a predetermined length along the longitudinal axis, which predetermined length is at least 2, preferably 4, more preferably 6, more preferably 8, 10, 15, 20 times larger than an approximate estimation of the longitudinal path of the object travelled per unit of time divided by an approximate estimate of the number of revolutions of the object per unit of time, and divided by the number of antennas of the detection arrangement. By making the length of antennas large enough, it is guaranteed that sufficient signals are obtained and, moreover, that a limited part of the signal is due to noise, among other things that each charge carrier is perceived by different antennas. At the above lower limit, a certain charge distribution would be detected at a specific point (fingerprint) e.g. at least twice (4, 6, 8, 10, 15, 20). By choosing the length of the antennas sufficiently large, it is ensured that the profile of the stochastic charge distribution is fully observed at least twice. As more accuracy is required, you can opt for a longer antenna and/or a larger number of antennas, and therefore a finer distribution.

Preferably, the predetermined length is at most 100, preferably 75, more preferably 50, 40, 30, 25 times greater than an approximate estimate of the longitudinal path of the object travelled per unit time divided by an approximate estimate of the number of revolutions of the object per unit time, and divided by the number of antennas of the detection arrangement. By not making the antennas too long, it is ensured that there is not too much variation in terms of rotation speed over the length of the object over which the antennas measure (not every point rotates at an identical rotation speed, but the rotation speed will be continuous, and without significant jumps), since a substantially constant rotation speed is assumed over the length of the antennas.

It is clear from the preceding sections that a middle ground must be found in order to obtain a clear enough signal, about which an approximately constant rotation speed is correct. Given that the parameters used depend on application to application, the result will also depend strongly on this. However, it should be noted that approximate estimates are typically certainly possible, since the order of magnitude of the

speeds/distance travelled is easy to estimate or known in most processes. The purpose of the measurement is therefore not to determine the order of magnitude of the rotational speed or rotation, but to perform a very specific, local determination thereof. The deviation from an estimated value will be fairly limited, and thus still lead to sufficient measurements on the one hand (and limited variation of rotational speed on the other).

In a further preferred embodiment, the antennas extend along a predetermined length along the longitudinal axis, which predetermined length is at least 3 mm, preferably at least 5 mm, and at most 100 mm, preferably at most 50 mm.

In a preferred embodiment the device comprises at least 1, 2, 3, 4, 5, 6, 7 or 8 charge detecting components, preferably at least 10, 12, 14, 16, 20, 32, 40, 50, 60, 70, 80 , 90, 100, 125, 150, 175, 200, 250, 300, 350, 400, 450, 500, 600, 700, 800, 900, 1000, 1250, 1500, 1750, 2000, 2500, or more. By providing more charge detecting components, more signals are generated, and the length of the sensor can be further reduced, in that sense the only limitation is the cost price, which will already be less relevant for some applications, and the available space. However, the physical limitations (size of the components, radius of circular pattern) when applied to yarns lead to a pragmatic choice between 8 and 32 charge detecting components, preferably a maximum of 16.

The arrangement with only one charge detecting component is somewhat more unusual, and will preferably comprise a grounding electrode on either side of the charge detecting component for shielding the charge detecting component. In that sense, the term 'a plurality of charge detecting components' can therefore also refer to a single charge detecting component, preferably flanked on both sides by a grounding electrode. The amount of detected pulses will be more limited than in situations with a higher number of charge detecting components (a charge will only provide one pulse per rotation), but will nevertheless be sufficient to determine the rotation undergone. Note that such a choice can further be compensated for in other ways, such as a longer charge detecting component.

In a preferred embodiment, the device comprises a digital signal processor (DSP) for processing output signals of the electrical charge detecting components, preferably comprising a low-pass filter (as anti-aliasing filter). Alternatively, the DSP can be separately provided with the measuring device, the measuring device being adapted to be coupled to a DSP, and to provide the output signals in a format suitable for the DSP.

In a preferred embodiment, the circular pattern has a minimum distance (radius of circular pattern minus radius of elongated object) with respect to the surface of the elongated object of 0.01 mm, preferably 0.05 mm, more preferably 0.1 mm, still more preferably 0.25 mm, still more preferably 0.5 mm, still more preferably 0.75 mm, still more preferably 1.0 mm, still more preferably 1.5 mm, still more preferably at least 2 mm, still more preferably at least 2.5 mm, still more preferably at least 3.75 mm, still more preferably at least 5 mm, still more preferably at least 10 mm, still more preferably at least 15 mm, still more preferably at least 25 mm, still more preferably at least 50 mm. As indicated, the lower limit is mainly a result of practical limitations in the structure of the array of antennas. Stronger miniaturisation leads to much higher costs, and typically much more sensitive components. The applicant therefore believes that reality will dictate a minimum distance of 0.01 mm from the object. In certain applications, a larger minimum distance may be required, for example with an object with transversal variations (in order not to touch the sensor, as well as to obtain a more stable signal strength). Note that the aforementioned dimensions also apply to the method of contactless determination of local rotation on an object, since the effective dimension of the device depends on the object and thus the area of application (for yarns, for example, there is a limited variation in thickness and a first radius can be used for the sensor, while for rigid objects, such as bars, a much wider sensor - in terms of radius - may be required).

In a preferred embodiment, the circular pattern has a minimum radius of 0.25 mm, preferably 0.3 mm, more preferably 0.4 mm, still more preferably 0.5 mm, still more preferably 0.75 mm, still more preferably at least 1 mm, still more preferably at least 1.25 mm or 1.5 mm, and even more preferably at least 2 mm. At present, a lower limit of less than 0.25 mm seems difficult to achieve, but it will become possible in the future to build the sensor in even smaller radii, and cost will typically be the biggest obstacle in this.

In a preferred embodiment, the circular pattern has a shortest distance (radius of circular pattern minus radius of elongated object) with respect to the surface of the elongated object of at most 100 mm, preferably at most 75 mm, more preferably at most 50 mm, and even more preferably a maximum of 25 mm or 15 mm. As indicated, the limiting factor here is the rapid decay of the electric field of the charges over a greater distance (as well as partly sensor quality), and noise will increasingly drown out the real signal as the distance to the object to be analysed increases. Note again that the aforementioned dimensions also apply to the method of contactless determination of local rotation on an object, since the effective dimension of the device depends on the object and thus the area of application (for yams, for example, there is a limited variation in thickness and a first radius can be used for the sensor, while for rigid objects such as bars, a much wider sensor - in terms of radius - will be required).

In a preferred embodiment, the circular pattern has a maximum radius of 100 mm, preferably 75 mm, more preferably at most 50 mm, still more preferably at most 25 mm or 15 mm.

Note that when determining a 'maximum' radius of the circular pattern, the diameter of the elongated object must be taken into account. In addition, the number of charge detecting components, an estimate of the rotation speed and the sampling speed of the DSP can possibly also be taken into account, such that the charges do not pass by successive antennas too fast.

In particular (or alternatively to the foregoing, depending on the diameter of the elongated object being analysed), the diameter of the circular pattern is at most 100 mm, 80 mm, 60 mm, 40 mm, preferably 30 mm, and more preferably 20 mm, larger than the diameter of the elongated object being analysed. The distance must be kept to a minimum in order to be able to clearly detect the charges and to generate a strong enough (and noise-free) signal. However, the size of the components must be taken into account, as well as possible influence on each other, which ensures that sufficient distance must nevertheless be maintained. To this end, the applicant noted that a maximum distance with respect to the object to be analysed of 50 mm is expedient for obtaining detection signals strong enough. The minimum distance depends more strongly on the diameter of the object to be analysed, but with very thin objects, such as yarns, it is opted for a minimum distance of 0.01 mm, preferably 0.05 mm, more preferably 0.1 mm, more preferably 0.25 mm, still more preferably 0.5 mm, still more preferably 0.75 mm, still more preferably 1.0 mm, still more preferably 1.5 mm, still more preferably at least 2 mm, still more preferably at least 2.5 mm, still more preferably at least 3.75 mm, still more preferably at least 5 mm, still more preferably at least 10 mm, still more preferably at least 15 mm, still more preferably at least 25 mm, to reduce the costs of the detection arrangement (miniaturisation is typically very expensive).

In a third aspect, the invention relates to a method for contactless determination of undergone rotation of a yarn extending substantially along a longitudinal axis at a predetermined longitudinal point on a yarn over a certain path along the longitudinal

axis or duration of time during the twisting of said yarn in an air jet device, comprising the following steps:

a. measuring local rotation at at least one, preferably at least 2, 3, 4, 5 or more, immobile, separated positions along the longitudinal axis of the yarn, said positions being upstream along the yarn relative to, and/or at the level of the location where the yarn is provided with torsion by an external actuator, preferably wherein at least one of the positions is at a distance of at most 10 cm from where the yarn is provided with torsion by an external actuator;

b. calculating, on the basis of the measured rotation, the total undergone rotation of the yarn at the predetermined longitudinal point of the yarn over said path.

It should be noted here that preferably the local rotation is measured in the air jet device itself. The applicant notes that the devices for measuring can be manufactured compact enough, and moreover can be positioned in the air jet device such that they have a minimal effect on the air current generated therein, and a negligible (if not even completely non-existent) effect on the air current that provides the torsion to the yarn.

The specific application of the previously described concepts to yarn results from the very high speeds, both longitudinal and rotational with which yarn is moved during processing (e.g., into a twisted yarn). In order to be able to accurately control/adjust the process, these speeds must therefore be measured as accurately as possible, for the rotational speed without making contact with the yarn. In addition, however, as indicated in the background, it is desirable to be able to approximate the rotation of the yarn undergone at any point. On the basis of the above methods, and sensors operating according to this principle, the rotation undergone can also be approximated at any point of the yarn, and thus the torsion on each segment of yarn, which allows the manipulation of the yarn to be coordinated with the existing torsion on the yarn.

In a preferred embodiment, the calculation of the rotation undergone is done on the basis of an interpolation, and optionally an extrapolation, of the measured local rotation, the total rotation of the said predetermined longitudinal point on the yarn being calculated on the basis of an integration or weighted summation of the measured and/or interpolated, and optionally extrapolated, rotation at longitudinal positions of said predetermined longitudinal point of the yarn at the times when said predetermined longitudinal point of the yarn is at said longitudinal positions, starting from a substantially known, or approximately derivable, longitudinal velocity profile of the yarn.

To accurately predict the rotation undergone on a yarn, sufficient detection arrangements must be provided that record the local rotation of the yarn at different points. On the basis of these different measuring points, a rotation profile can be defined over the entire path of the yarn, on the one hand by interpolation for points between the measuring points, as well as possibly extrapolation for outlying segments. This is based on a continuous course of the local rotation, which is also the case in practice. With sufficient measuring points, therefore, a very reliable picture can be obtained of the local rotation on the yarn in each (fixed) point along the longitudinal axis. On the basis of this, for a specific point on the yarn itself (thus moving along with it), the total rotation undergone can be determined over a certain range or over a certain period of time. Nevertheless, it should be noted that already with a single detection arrangement, the rotation profile can be defined, either on the basis of a theoretical or experimental model that extrapolates the profile from the one known point.

It is thus to be understood that the measuring devices measure the local rotation at certain (fixed) positions along the axis of the yarn, but not at a certain point of the yarn moving longitudinally along said axis. In other words, in order to determine the total rotation undergone for a given point of the yarn, the local rotation along the trajectory that this point has travelled must be integrated, or via a discrete weighted summation of the local rotation over a number of positions.

In a preferred embodiment, measuring the local rotation comprises the following steps: a. guiding the yarn along the longitudinal axis through at least one, preferably 2, 3, 4, 5 or more, detection arrangements at said positions along the longitudinal axis, each detection arrangement comprising a plurality of electric charge detecting components that are arranged in a first substantially circular pattern around the longitudinal axis at equal mutual distances, and comprising a plurality of grounding electrodes equal to the plurality of electric charge detecting components, which grounding electrodes are arranged in a second substantially circular pattern around the longitudinal axis at equal mutual distances, the second substantially circular pattern staggered with respect to the first substantially circular pattern, wherein the grounding electrodes act as a screen between the successive charge detecting components and thus act as a spatial filter;

b. detecting one or more electrical charges on the yarn by a plurality of the charge detecting components, and processing the detection into an output signal;

c. combining the output signals of each of the charge detecting components (according to an identical time schedule);

d. deriving one or more dominant frequencies from the combined output signals.

In a preferred embodiment, the undergone rotation is determined at one or more individual longitudinal points on the yarn, and then a torsion of the yarn between said one or more separate longitudinal points is determined by subtracting the undergone rotations from the two separate longitudinal points. Note that in case a measurement is only made at one point, a rotation of the yarn can still be determined along the entire line, and thus a torsion over segments, for example by defining a local zero rotation at a certain point (typically based on previous measurement, which leads to the correctness of this assumption in future situations). To support such an assumption, it may be desirable to perform a number of reference measurements in advance to determine the system characteristics for a given yarn. Once these are known, these assumptions can be used in the future, which offers great advantages as the systems typically process a limited number of types of yarn, and thus a kind of reference table can easily be created on the basis of these reference measurements.

As indicated earlier, in many cases it is crucial to know the torsion present across segments of the yarn, in order to determine how much torsion must still be added to obtain a desired torsion over the segment (and for this purpose a device for applying of a torsion on the yarn positioned after the measuring devices).

In addition, it is also possible, on the basis of measurements and/or theory, to draw up a model that the local rotation complies with, as it were, to establish a certain profile that invariably develops over a yarn. On the basis of a single measuring point, such a predetermined profile would allow to predict the local rotation, and therefore the torsion present, over the entire yarn.

In further embodiments, the method is adapted according to one or more of the further embodiments for the method for contactless measurement of local rotation of an elongated object, preferably a yarn, according to the first aspect of the invention.

In particular, a small band filter is preferably applied to the merged output signals centred around the dominant frequency.

Furthermore (whether or not in combination with the foregoing), use is preferably made of two groups of charge detecting components, the resulting output signals of the two groups being subtracted from each other, so as to result in a differential signal. We refer to previous passages that further highlight and clarify these specific versions. It is to be understood that all the aforementioned advantages also apply here.

In a fourth aspect, the invention relates to a method for applying a torsion, preferably an alternating S and Z torsion, to a yarn or elongated object via a substantially tangential air current at a fixed location relative to the (possibly longitudinally moving) yarn, wherein the air current is controlled on the basis of torsion of a section of the yarn of a predetermined length, which section is at said fixed location, and wherein the torsion is determined by a method according to the third aspect of the invention, in particular wherein the torsion is determined over a length of yarn on the basis of the rotation undergone at the start and end points of said length of yarn. Because previous methods and sensors allow measuring the local rotation that an elongated object undergoes, the determined value of the local rotation can be used to accurately approximate the total rotation undergone, on the basis of which the air current for applying torsion to the yarn can be adjusted in order to obtain a (uniform and correct) final torsion over the yarn.

In a fifth aspect, the invention relates to a system for manufacturing a twisted yarn from a, preferably substantially torsion-free, yarn that is presented along a longitudinal axis, the system comprising the following elements:

a. an air jet device for (possibly alternating S/Z) applying a torsion to the yarn, to obtain a twisted yarn;

b. at least one, preferably at least 5, devices for contactless measurement of local rotation of the yarn according to the invention, wherein the devices are positioned along the longitudinal axis separately from each other upstream in front of the yarn relative to and/or at the level of the air jet device, preferably wherein at least one of the devices is positioned at a maximum upstream distance of 10 cm with respect to the air jet device, and is preferably integrated in the air jet device.

As indicated earlier, it is advisable to provide sufficient measuring points where the local rotation is determined, in order to establish an accurate rotation profile. In principle this can already be approximated with a single measuring point (and therefore one device), but preferably more devices are used which are distributed along the longitudinal axis along which the yarn is fed.

Hereby it is especially recommended to provide a device as close as possible to the entrance of the air jet device, since here the local rotation can already start to deviate more strongly from the measured values further upstream, due to the influence of the air jet device. It is also advisable to place the individual measuring points close to the air jet device closer together, again in order to be able to approximate the rotation profile more correctly on the basis of the measured discrete values of the local rotation.

In a preferred embodiment, the system comprises a feed member for feeding the yarn along the longitudinal axis, to (and through) the air jet device.

In what follows, the invention is described by way of non-limiting examples illustrating the invention, and which are not intended to and should not be interpreted as limiting the scope of the invention.

EXAMPLES

EXAMPLE 1 : Measuring device for local rotation

Figures 1A-B, 2 and 3 show a possible arrangement of a sensor according to the invention, wherein Figure 2 represents the 'unrolled' sensor to clarify the structure of Figure IB, and Figure 3 shows the section in the longitudinal direction along which the elongated object would be passed through the sensor, as well as of Figure IB. The sensor herein comprises a number of charge detecting components or charge detecting antennas (1) and an (equal) number of grounding electrodes (2), or grounded antennas. The charge detecting antennas are divided into two separate groups (la, lb), with elements of the groups successively following each other. The charge detecting antennas (la, lb) of each group are connected in series on a first and second output line (3a, 3b) which provide a first and second output signal. By subtracting these signals from each other, (low-frequency) noise is largely eliminated, which already provides a much purer signal.

The grounded antennas (2) are positioned between successive charge detecting antennas (la, lb), which shield charges that are not substantially in front of the charge detecting antennas.

As visible in Figure 1A-B, the antennas (both grounded (2) and charge detecting (la, lb)) are at a substantially identical radial distance from the elongated object that is passed through it. Although not strictly necessary, the applicant noted that the signal was much purer in such an arrangement, and easier to process in post-analysis. Alternatively, the grounded antennas (2) can also be positioned at a smaller distance from the elongated object (4) than the charge detecting antennas (la, lb). In addition, the antennas are also at fixed distances/angles from each other, although this is not strictly necessary, as long as the positions are known.

In Figure 1A, the circular pattern is provided with three charge detecting components / antennas (la, lb) of each group, and with six grounding electrodes (2) between each of two adjacent antennas (la, lb). The grounding electrodes are all connected in this case, as can also be seen in Figure 2. In Figure IB, four antennas (la, lb) of each group are provided each time, and eight grounding electrodes (2) that are connected to each other.

EXAMPLE 2: Measuring principle

Figures 4-7 show the effect of shielding by the grounded antennas or grounding electrodes. Where Figure 4B shows what is perceived by a single charge detecting component with a passing electric charge, as in Figure 4A, Figures 5A-B and Figures 6A-B and Figures 7A-B show further embodiments.

Figure 4B shows that a passing charge (5) leads to a smeared signal with a simple antenna. Although this one signal can be used to approximate when the charge (5) is closest to the antenna, this will become impossible once there are several charges. Partly to prevent this smearing, the grounded electrodes are used, which provide a much more compact signal from the charge detecting antenna, which now approximates a pulse more, as in Figure 5A-B. Here, two shielding, grounded electrodes are placed on either side of the charge detecting component.

However, to derive meaningful information about speed or effective rotation of the charge from this, multiple charge detecting antennas must be used, each of which is also shielded by grounded electrodes, such as in Figure 6A-B. For a single passing charge, this leads to the first graph of 6B, from which the speed or rotation over a certain time can be deduced quite easily.

However, when yarns are worked with, there is a very large multitude of charges that are detected, the charges being stochastically distributed over the yarn. This results in a much more complex signal, where it is no longer possible to look per charge, so it is necessary to use precisely that stochastic distribution, which acts as a fingerprint for a particular charge distribution that is moving (note here that all charges move at a rotational speed that is substantially the same), whereby that fingerprint can be recognised in the further course (albeit not so much visually, but by signal analysis, e.g. via DFT/FFT/DTFT).

The summation of the signals by the passing of a stochastically distributed plurality of charges (å) is represented in the second graph of Figure 6B, which then leads to the determination of the dominant frequency via FFT (third graph Figure 6B).

Subsequently, this line of thought was applied to the arrangement of EXAMPLE 1, in which two groups of charge detecting antennas were used, see also Figure 7A. This leads to a set of signals for each charge detecting antenna that are combined into a single signal per group, which is visible as the two signals in the first graph of Figure 7A. However, this is applied to a plurality of stochastically distributed charges (å), leading to a total signal for all of these charges as shown in the second graph of Figure 7B. Finally, the signals from the two sets of charge detecting antennas are subtracted from each other ('Diff'), which removes the low-frequency or constant components, but also (high-frequency and low-frequency) external components, since the two sets of antennas will detect the same charges due to an external source or charge, regardless of its frequency. The result of the following FFT was no longer shown, but would lead to an even clearer dominant frequency than with previous versions in Figures 4-6.

Note that in Figures 4-7 the grounded antennas and charge detecting antennas are shown for clarity on the basis of a linear movement of the charge, and a linear positioning of the antennas, these being also not at an identical distance from the charge path.

The present invention should not be construed as being limited to the embodiments described above and certain modifications or changes may be added to the examples described without having to re-evaluate the appended claims. For example, the present invention has been described with reference to yarn, but it should be understood that the invention can be applied to any elongated object on which a twist, torsion or rotation is to be applied. These elongated objects can on the one hand be yarns, but on the other hand also rigid objects, for example cylindrical but also non-cylindrical objects, rolls, bars, etc. It may thus be possible for the invention to be applied even more efficiently to non-cylindrical objects, since the variation in distance between the surface of the object and the charge detecting component fluctuates, and the strength of the pulse will strongly depend on it. This will provide the stochastically distributed charges with a still clear fingerprint in terms of measured signals via the device/method of the invention, which makes the rotation undergone easier to deduce.