WO2024218677A1

WO2024218677A1 - Method for adjusting the sensor sensitivity of a motion sensor for detecting movement of a pile yarn in a tufting machine and adjusting system

Info

Publication number: WO2024218677A1
Application number: PCT/IB2024/053744
Authority: WO
Inventors: Geert Debuf; Giacomo DALLE NOGARE; Marco ARLONE; Carlo MILANACCIO
Original assignee: Vandewiele NV
Current assignee: Vandewiele NV
Priority date: 2023-04-17
Filing date: 2024-04-17
Publication date: 2024-10-24
Anticipated expiration: 2025-10-17
Also published as: CN120513326A

Abstract

Method and associated adjusting system (24) for adjusting the sensor sensitivity of a motion sensor (16, 17) for detecting movement of a pile yam (3) in a tufting machine (1), provided with several tufting needles (12), wherein this pile yarn (3) is incorporated into a fabric (7), comprising determining current process data, providing reference data for the sensor sensitivity of the motion sensor (16, 17), consisting of one or more sets, wherein each set comprises a value of a reference sensor sensitivity and corresponding process data, determining a connection between the current process data and the process data of the provided reference data, determining a value for the sensor sensitivity to be adjusted on the basis of this connection and the one or more values of the reference sensor sensitivity of the provided reference data, adjusting the value for the sensor sensitivity to be adjusted in the motion sensor (16, 17).

Description

METHOD FOR ADJUSTING THE SENSOR SENSITIVITY OF A MOTION SENSOR FOR DETECTING MOVEMENT OF A PILE YARN IN A TUFTING MACHINE AND ADJUSTING SYSTEM

The present invention relates to a method for adjusting the sensor sensitivity of a motion sensor for detecting movement of a pile yam in a tufting machine, provided with several tufting needles, wherein this pile yarn is incorporated in a fabric. The present invention also relates to a method for monitoring the tension of a pile yam in a tufting machine, provided with several tufting needles, wherein this pile yarn is incorporated into a fabric, comprising generating measurement signals by means of a motion sensor which are an indication of the pile consumption of this pile yam, determining a moving average of the measurement signals over a specific time period and determining whether this moving average exceeds a first limit value and/or drops below a second limit value.

In addition, the present invention relates to an adjusting system for adjusting the sensor sensitivity of a motion sensor for detecting movement of a pile yam in a tufting machine and a monitoring system for monitoring the tension of a pile yarn.

In a tufting machine, a pile yam may or may not be inserted in the fabric (backing or substrate) by means of a tufting needle in various machine cycles in order to form or not form pile by means of the pile yarn. Such a tufting machine is provided with several tufting needles, and several corresponding pile yams are supplied to the tufting machine in order to form a tufted fabric therewith. To this end, the tufting needles are typically arranged on a needle bar which is moved up and down during such a machine cycle. In this case, the tufting needles may be selected individually (individual needle selection) in order to follow or not follow the movement of the needle bar so as to insert or not insert the former into the fabric.

Systems and methods for monitoring the tension of a pile yam are in this case intended to be able to stop the tufting machine as quickly as possible in case of an increase in the tension and/or in case of breakage of a pile yarn. In case of an increase in tension, it is desirable to be able to stop the tufting machine before the pile yarn breaks. In addition, the tension of the pile yarns to be monitored has a significant effect on the quality of a fabric tufted using the former, such as for example a tufted carpet.

W02017/006226A1 describes a detection system and an associated method for monitoring the tension of a pile yam in a textile machine, such as for example a tufting machine, using the individual actuators which are each provided for supplying a corresponding pile yarn. However, with the monitoring system described therein, it is only possible to intervene in the case of deviating tensions after a considerable number of machine cycles have been performed. With tufting machines comprising motors for simultaneously supplying different pile yarns, the solution from W02017/006226A1 can therefore not be used.

GB 2 113 404 A describes a detection system and an associated method for monitoring the tension of a yarn in a textile machine, and more specifically a tufting machine, using sensors which generate measurement values which are a measure for the tension of the yarn, more specifically the pile yarn. At least one signal is generated per machine cycle. Measurement values of at least one machine cycle determined in this way are compared to the same reference values (such as a high and a low reference value), and an error signal is generated when the difference between these measurement values and the reference values deviates from a specific value. Error signals are usually only generated after several machine cycles.

EP 3 165 490 Al describes a detection system and an associated method for monitoring the tension of a yam in a sewing machine using a piezoelectric sensor in the form of a rectangular plate which also generates measurement values which are a measure for the tension of the yarn. This system could possibly be used with tufting machines, but in practice this is a rather laborious undertaking. Tufting machines require a very large number of pile yams. This system is not compact, takes up a considerable amount of space, and requires various components for it to operate. After all, the yarn tension is being converted into a mechanical force on an object (e.g. a cylinder), and this object in turn transmits the mechanical force onto a force-receiving component, following which a tension is generated in the piezoelectric sensor.

The Applicant of EP 3 165 490 Al, Eltex has introduced a detection system Eltex Eye onto the market which is specifically adapted to tufting machines, using another type of piezoelectric sensor which detects movement. In this case as well, action can only be taken after a considerable number of machine cycles. In addition, sensor control only starts when the machine is at operating speed. There is thus no detection while the machine accelerates or decelerates.

US 2020/0087103 Al describes a detection system and an associated method for monitoring the tension of a yarn in a textile machine, such as for example a tufting machine, using optical sensors which measure the speed of the yarn. Here too, the machine only stops when the sensor indicates the absence of a yarn movement for a sufficiently long time. Here too, sensor control only starts when the machine is at operating speed.

With the current systems, the machine also does not stop every time a pile yarn becomes dislodged from a tufting needle either. The sensor, which is situated before the tufting needle in the yarn feeding process, still detects a yarn movement, since the feeding system still supplies pile yam at a correct speed. This pile yarn passes through the sensor, as a result of which no error is detected.

There is therefore a need for an improved detection system for tufting machines which can be used with different kinds of tufting machines and by means of which it is possible to act more quickly when an error is detected.

This object of the invention is achieved, on the one hand, by providing a method for adjusting the sensor sensitivity of a motion sensor for detecting movement of a pile yarn in a tufting machine, provided with several tufting needles, wherein this pile yarn is incorporated into a fabric, wherein this method comprises the following steps: a) determining the current process data; b) providing reference data for the sensor sensitivity of the motion sensor, consisting of one or more sets, wherein each set comprises a value of a reference sensor sensitivity and corresponding process data; c) determining a connection between the current process data and the process data of the provided reference data; d) determining a value for the sensor sensitivity to be adjusted on the basis of this connection and the one or more values of the reference sensor sensitivity of the provided reference data; e) adjusting the value for the sensor sensitivity to be adjusted in the motion sensor. A motion sensor generates measurement signals at a specific sensor sensitivity, so that the resultant measurement signals are influenced in order to obtain, depending on boundary conditions, such as machine speed or type of pile yarn used, relevant measurement signals. With a relatively slow machine, for example, a motion sensor detects less yarn movement than is the case with a relatively quick machine during the same time interval. By using a different sensor sensitivity in both situations, in each case similar relevant measurement signals are obtained which correctly indicate whether or not there is yarn movement in both situations. In this way, the generated measurement signals are a clear and unambiguous indication for the pile yarn consumption and may be evaluated without taking into account the influence of boundary conditions and ambient effects.

The sensor sensitivity is an indication of the degree to which the motion sensor is responsive to movement. The sensor sensitivity thus determines how suitable the sensor-generated measurement signals are to be used to indicate movement (or pile yam consumption). The movement to be detected is the movement of the pile yarn which is an indication of the pile consumption of the pile yarn.

The sensor sensitivity consequently is a scale factor by which raw data, measured by the sensor, are multiplied in order thus to obtain measurement signals which can be evaluated without taking into account the influence of boundary conditions and ambient effects, such as vibrations.

This sensor sensitivity is individually adjustable for each motion sensor.

In the prior art, the sensor sensitivity may be adjustable, but this is then simultaneously the case for either all motion sensors which are used in a tufting machine for monitoring the tension of a pile yarn or for a group of motion sensors which are used in a specific zone of the tufting machine. If, in existing systems, there are too many false detections, for example when two greatly differing pile yarns are being incorporated into a fabric next to each other, the detection has to be supplemented with a visual inspection by people.

With existing tufting machine, adjusting the sensor sensitivity is also a laborious process which may take up considerable time when starting up a tufting machine, during which procedure significant measurement signals from as many motion sensors as possible are desired. By now making the sensor sensitivity individually adjustable for each motion sensor, it is also possible to take into account the influence of boundary conditions and ambient effects for each motion sensor individually, so that the sensor sensitivity can also be adjusted, for example, in the case of motion sensors which are to be used to detect movement of greatly differing pile yams.

In order to make it possible to adjust the sensor sensitivity in a simple and quick way, reference data are provided. By adjusting the sensor sensitivity, starting from predetermined reference data, it is possible to respond to changing boundary conditions and ambient effects more quickly for each motion sensor.

The reference data consist of one or more sets, with each set comprising a value of a reference sensor sensitivity and corresponding process data.

The process data comprise one or more characterizing features or properties of the process which incorporates the pile yam into a fabric.

By determining a connection between the current process data and the process data of the one or more sets of reference data and determining a value for the sensor sensitivity to be adjusted on the basis of the reference sensitivities of the one or more sets of reference data sets by means of this connection, it is possible to adjust a value of the sensor sensitivity in the motion sensor which is as optimal as possible.

The amount of movement of the pile yam per unit time to be detected by the motion sensor is proportional to the machine speed of the tufting machine. If the machine speed is halved, the movement of the pile yam to be detected also halves. Preferably, the process data comprise this machine speed.

In addition, the process data may also comprise one or more other parameters which have an effect on the amount of movement to be detected by the motion sensor. Thus, these parameters may, inter alia, comprise one or more characterizing features of the yams, such as the yarn type or the yarn thickness, the machine type of the tufting machine, the machine acceleration of the tufting machine (which may also include deceleration), the type of detection or also pattern information of the fabric to be produced. The expression pattern information is intended to mean not just the desired pile height of the tufted pile, but also the selection data of the needles of the tufting machine, the position data of the needle bar, to take into account the lateral position of the needles, the feeding velocity of the feeding devices of the pile yarn, etc. The measurement signals generated by a motion sensor may also be used to monitor the tension of a pile yarn. The expression type of detection is intended to mean, for example, a detection of a tension of the pile yam which is too low (BED or “Broken End Detection”) or a tension of the pile yam which is too high (TED or “Tight End Detection”).

Preferably, the one or more sets of reference data define a connection between the values of the reference sensor sensitivity and the corresponding process data expressed in a look-up table and/or a mathematical function.

After the connection between the current process data and the process data of the provided reference data has been determined, the sensor sensitivity to be set can easily be calculated by applying the mathematical function and/or by interpolation or extrapolation of the reference sensor sensitivities.

More specifically, the method is further characterized by determining specific reference data in step b) from the provided reference data for the sensor sensitivity of the motion sensor, based on the current process data. In this way, only the most relevant sets of reference data can be selected, with the process data of these selected reference data being as similar as possible to the current process data. It is for example possible to only take into account sets of reference data wherein the machine type in the process data is identical to the machine type of the current process data, or wherein the machine speed comes close to the machine speed of the current process data. Subsequently, a mathematical relation is determined between the process data of the specific reference data and the current process data in step c), in order to determine the connection between the two. This may be done, for example, by interpolation or extrapolation. Then, the sensor sensitivity is determined in step d) by applying said mathematical relation to the values of the reference sensor sensitivities of the specific reference data.

In a preferred method according to the present invention, providing reference data for the sensor sensitivity of the motion sensor in step b) comprises determining a set of reference data in a learning cycle, consisting of a value of a reference sensor sensitivity and corresponding process data, this learning cycle comprising the following steps: determining an expected percentage detection value per unit time; setting a starting value for the sensor sensitivity of the motion sensor; detecting an amount of movement per unit time by means of the motion sensor; comparing this amount to the specific expected percentage detection value per unit time; adjusting the sensor sensitivity set in the motion sensor until the amount of detected movement per unit time corresponds to the specific expected percentage detection value per unit time; adding to the reference data the set consisting of the value of the sensor sensitivity, set in the motion sensor, when the amount of detected movement per unit time corresponds to the specific expected percentage detection value per unit time, as a value of the reference sensor sensitivity and the current process data.

Based on the theoretical knowledge of the needle cycle of a tufting needle in a tufting machine, for example, the amount of movement of the pile yam to be expected can be determined in one needle cycle of the tufting needle to this end. When selecting a tufting needle, a needle cycle equals a machine cycle. From the combination of this amount of movement to be expected per needle cycle, the machine speed of the tufting machine (the number of machine cycles per unit time) and the pattern information of the fabric to be tufted (the ratio between the number of needle cycles and the machine cycles), it is possible to determine an expected percentage detection value per unit time for the amount of movement to be detected per unit time. A starting value for the sensor sensitivity can then be set in the motion sensor which may be determined, for example, from the reference data.

The amount of movement detected by the motion sensor can then be compared to this expected percentage detection value per unit time. If this amount of movement does not correspond to the expected percentage detection value per unit time, i.e. if this does not fall within specific predetermined boundaries of the percentage detection value, e.g. within 20%, 10%, 5% of the percentage detection value, the sensor sensitivity, set in the motion sensor, can then be adjusted until the detected amount of movement does correspond to the expected percentage detection value per unit time. Thereafter, the value of the sensor sensitivity set in the motion sensor can be added in the reference data as a value of the reference sensor sensitivity, together with the current process data.

The object of the invention is achieved, on the other hand, by providing a method for monitoring the tension of a pile yarn in a tufting machine, provided with several tufting needles, wherein this pile yam is incorporated into a fabric, comprising generating measurement signals by means of a motion sensor which are an indication of the pile consumption of this pile yarn, determining a moving average of the measurement signals over a specific time period, and determining whether this moving average exceeds a first limit value, wherein a sensor sensitivity is set in the motion sensor which is set according to a method for adjusting a sensor sensitivity of a motion sensor according to the invention. An increase in the moving average is typically a precursor to yarn breakage. Additionally (or alternatively), it is possible to determine whether the moving average drops below a second limit value.

In this case, the moving average may be a so-called simple moving average. Alternatively, this moving average may be, for example, a centred (cumulative) moving average, or a weighted moving average, or an exponential moving average.

In this case, a said limit value is preferably determined as a percentage deviation. Thus, the first limit value may, for example, be a deviation of 10%, in which case it is then determined whether the moving average has increased by 10%. Using a percentage deviation makes it simple to take into account differences, such as for example a difference in pile yarn. Alternatively, but less preferred, an absolute number may be chosen as limit value.

By evaluating this moving average in comparison with the limit values, it is possible to verify when the tension becomes excessively high with the risk of yarn breakage or loss of quality or when this tension becomes excessively low with the risk of loss of quality.

An increasing yam tension results in an increase in this moving average. A decreasing yam tension results in a decrease in this moving average.

In this way, it is also possible to detect increased and reduced tensions of the pile yam with an accuracy similar to that with which yarn breakage can be detected.

If the moving average exceeds the first limit value and/or drops below the second limit value, an alarm may be generated and/or the tufting machine may be stopped.

If desired, several limit values may be provided in case of an increase in the moving average and several limit values in case of a decrease in the moving average. Thus, it is possible, for example, for an alarm to be generated when the moving average increases above the first limit value and for the tufting machine to be stopped above a third limit value, which is higher than the first limit value. Analogously, it is possible to generate an alarm if the moving average drops below the second limit value, and for the tufting machine to be stopped below a fourth limit value, which is lower than the second limit value.

In addition, various possible limit values may also be provided for various possible detections.

By means of such a method according to the present invention for monitoring the tension of a pile yarn in a tufting machine, it is not only possible to detect yarn breakage (BED) quickly and correctly. With a solution according to the present invention, the term yam breakage (BED) not only covers the detection of actual broken pile yam, but also detections where the tension of the pile yam is much too low. Increases in the tension of the pile yarn (TED) can also be detected before yarn breakage occurs. By detecting undesired changes in the yarn tension in a timely manner, it is also possible to ensure a more constant quality of the tufted fabric.

The specific time is preferably adjustable. If desired, several specific times may be provided for various possible detections. Thus, for example, a first specific time may be provided for a BED detection and a second specific time for a TED detection, in which case a first moving average is determined for the BED detection over the first specific time period and a second moving average is determined for the TED detection over the second specific time period.

The one or more limit values are preferably also adjustable.

In a preferred embodiment of a method according to the present invention, the motion sensor measures the pile consumption of the pile yam between a yam storage system and a feeding device for supplying this pile yam from the yam storage system in the tufting machine. At this location, the tension already gradually starts to increase before a yam breakage.

Alternatively, it is also possible to measure the pile consumption of the pile yam by means of the motion sensor between a feeding device for supplying this pile yarn in the tufting machine and a tufting needle with which this pile yarn is incorporated into the fabric. Measurement data of a motion sensor which is built into the tufting machine at this location are in this case preferably additionally evaluated on the basis of machine position data which are an indication of one or more machine positions per cycle.

A machine cycle is a cyclic sequence of machine positions.

Traditionally, a machine cycle is divided into 360 degrees, analogous with the angular position of the shaft used to drive the tufting machine. Alternatively, it is possible to opt for another division between predefined limit values.

In a method according to the present invention, the motion sensor may take various forms. Thus, it is for example possible, analogous to US 2020/0087103 Al, to choose an optical sensor as said motion sensor. Alternatively, it is for example possible, analogous to an Eltex Eye, to choose a piezoelectric sensor as said motion sensor.

In order to improve or enhance the set sensor sensitivity, the sensor sensitivity may be adjusted in a learning cycle, striving for a specific acceptable percentage detection value of the movement. To this end, an acceptable percentage detection value is first determined, e.g. as described above. By means of the motion sensor, an amount of movement per unit time is then determined. The detected amount of movement per unit time is subsequently compared to the acceptable percentage detection value per unit time.

More specifically, it is possible to verify if the amount of detected movement per unit time corresponds to the acceptable percentage detection value per unit time, that is to say that in this case a check is performed to determine whether it falls within specific predetermined limits of the percentage detection value, e.g. within 25%, 20%, 15%, 10%, 5% of the percentage detection value.

If the detected amount of movement per unit time does not correspond to the acceptable percentage detection value per unit time, then the sensor sensitivity set in the motion sensor can be adjusted until the detected amount of movement does correspond to the acceptable percentage detection value per unit time.

When the detected amount of movement per unit time corresponds to the acceptable percentage detection value per unit time, the sensor sensitivity set in the motion sensor can then also be added in the reference data as a value of the reference sensor sensitivity, together with the current process data. Alternatively, the specific acceptable percentage detection value which is aimed for is adjustable.

In a said learning cycle, it is possible to (automatically) determine which is the optimum sensor sensitivity for each individual pile yarn. When completing such a learning cycle, the sensor sensitivity can quickly be adjusted individually for each pile yarn, depending on the yarn type and/or depending on the machine speed and/or depending on the desired type of detections, such as for example BED or TED.

With a method according to the invention, the tension of one or more additional pile yarns is monitored preferably analogously by generating corresponding additional measurement signals by means of one or more corresponding additional motion sensors. Preferably, a moving average of the corresponding measurement signals over a specific time period is determined for each monitored pile yarn and it is preferably determined whether this moving average exceeds a first limit value and/or drops below a second limit value. In this case, the first limit value and/or the second limit value may, if desired, be selected differently for each motion sensor or differently for each group of motion sensors, for example to take into account various types of pile yarn which are incorporated into the same fabric. Analogous to that which has been described above, it is optionally also possible to provide several limit values in case of an increase in the moving average and several limit values in case of a decrease in the moving average. Analogously, the necessary alarms may also be generated and/or the tufting machine may be stopped.

The object of the present invention is also achieved by providing an adjusting system for adjusting the sensor sensitivity of a motion sensor for detecting movement of a pile yarn in a tufting machine, provided with several tufting needles, wherein this pile yarn is incorporated into a fabric, this adjusting system comprising: a data unit for reading in current process data, a storage unit for storing reference data for the sensor sensitivity of the motion sensor consisting of one or more sets, wherein each set comprises a value of a reference sensor sensitivity and corresponding process data, and a calculation unit for determining a connection between the current process data and the process data of the provided reference data and determining a value for the sensor sensitivity to be adjusted on the basis of this connection and the one or more values of the reference sensor sensitivity of the provided reference data, wherein the adjusting system is provided for adjusting the value for the sensor sensitivity to be adjusted in the motion sensor.

More specifically, this adjusting system is in this case preferably provided for adjusting the sensor sensitivity according to an above-described method according to the invention.

In addition, the object of the invention is also achieved by providing a monitoring system for monitoring the tension of a pile yam in a tufting machine, wherein this pile yarn is incorporated into a fabric, comprising a motion sensor for generating measurement signals which are an indication of the pile consumption of this pile yarn, and an evaluation system for determining a moving average of the measurement signals over a specific time period and for determining whether this moving average exceeds a first limit value and/or drops below a second limit value, wherein this monitoring system comprises such an adjusting system.

More specifically, this monitoring system is in this case preferably provided for monitoring the tension of a pile yam in a tufting machine according to an abovedescribed method according to the invention.

This monitoring system furthermore preferably also comprises an adjusting unit for adjusting the specific time and/or for adjusting the limit value and/or for adjusting the sensor sensitivity.

The object of the present invention is furthermore also achieved by providing a tufting machine, comprising a said adjusting system according to the present invention. More specifically, this tufting machine may in this case comprise a said monitoring system according to the present invention.

The present invention will now be explained in more detail by means of the following detailed description of some embodiments of tufting machines, monitoring systems and methods according to the present invention. The sole aim of this description is to give illustrative examples and to indicate further advantages and particulars of the present invention, and can therefore not be interpreted as a limitation of the area of application of the invention or of the patent rights defined in the claims.

In this detailed description, reference numerals are to refer to the attached drawings, in which:

- Figure 1 diagrammatically shows a tufting machine according to the present invention; and

- Figure 2 diagrammatically shows a monitoring system according to the present invention.

In the tufting machine (1) illustrated in Figure 1, pile yams (3) are supplied from a yarn storage system ( creel) (2) (not shown) to the tufting machine (1) by means of a feeding device (4). To this end, the feeding device (4) comprises several yarn-feed modules (5) which are provided with an individual supply, for example, for each pile yam (3) by providing a drive roller driven by an actuator and a guide roller for each pile yarn (3). In addition, puller rolls (6) are also provided.

By means of the yarn-feed modules (5) and puller rolls (6), the pile yarns (3) are supplied to corresponding tufting needles (12).

The puller rolls(6) consist of a pair of rods between the feeding device (4) and the tufting needles (12) through which all pile yams (3) pass. These puller rolls (6) are arranged in such a manner that they lightly touch each of the pile yarns (3), so that the tension of the pile yarns (3) in the tufting machine (1) is equalized, as the pile yarns (3) are being supplied from different heights and at different speeds.

The tufting needles (12) are arranged on a needle bar (14) which is movable up and down in the tufting machine (1) by means of one or more connecting rods (13). By moving the tufting needles (12) up and down, the corresponding pile yams (3) are introduced into a fabric (backing or substrate) (7) in order thus to produce a tufted fabric (8).

To this end, the fabric (7) is passed from unwinders (10) under the tufting needles (12) by means of cloth feed rollers (9) and rolled back up onto winders (11). To this end, one or more cloth feed rollers (9) are driven rollers, while the other cloth feed rollers (9) are designed as guide rollers. The fabric (7) is clamped at the location of the tufting needles (12) by means of a presser foot (15). Furthermore, bed plate mechanisms (18) are present which may comprise grippers for forming loop piles and optionally knives for forming cut piles. This construction of tufting machines (1) is known and may be configured in various ways and in various variants, so that this will not be discussed in any more detail in the context of the present patent application. In the case of tufting machines (1) with individual pile delivery, for example, the puller rolls (6) will not be present.

According to the invention, each pile yarn (3) of such a tufting machine (1) is now provided with a corresponding motion sensor (16, 17). This motion sensor (16, 17) may be fitted at various positions in the line of the movement of the corresponding pile yarn (3). In a first illustrated position, the motion sensor (16) is arranged between the feeding device (4) and the tufting needle (12). In a second illustrated position, the motion sensor (17) is arranged between the yarn storage system (2) and the feeding device (4). Several such motion sensors (16, 17) may be fitted at each said position in the same housing in order to install these more easily in the tufting machine (1) as a group. Thus, for example, a housing comprising 16 of such sensors (16, 17) may be provided.

In the installed position in the tufting machine (1), these motion sensors (16, 17) are provided in order to generate measurement signals (D_v) which are an indication of the pile yam consumption for each supplied pile yarn (3).

Various kinds of motion sensors (16, 17) may be taken into consideration for this purpose, such as for example an optical sensor, analogous to that in US 2020/0087103 Al or a piezoelectric sensor, analogous to that in an Eltex Eye. In the specific embodiments described below, use was made of piezoelectric sensors. These examples also apply mutatis mutandis to other types of motion sensors.

The monitoring system (20) according to the present invention illustrated in Figure 2 comprises the motion sensors (16, 17) for installation in a tufting machine (1) as illustrated in Figure 1. A control unit (19) is provided for controlling this monitoring system (20) and to this end comprises, for example, a microprocessor. In order to determine a moving average (Dma) of the measurement signals (D_v) for each motion sensor (16, 17) and to determine whether this moving average (Dma) exceeds a first limit value (or possibly exceeds one or more limit values) and/or drops below a second limit value (or possibly drops below one or more limit values), an evaluation system (22) is provided. This evaluation system (22) will typically be distributed across the various motion sensors (16, 17) which are each separately or per group (for example per group of 2, 4, 8 or 16) provided with a local part of the evaluation system (22) for determining the moving average (Dma) of the measurement signals (D_v) and, if desired, comparing this moving average (Dma) with said one or more limit values. In housings comprising 16 of said motion sensors (16, 17), these motion sensors (16, 17) may, for example, be controlled all together or divided up into various blocks (of 2, 4, 8 or 16) by a local control unit which is in turn controlled by means of the control unit (19). The various motion sensors (16, 17) in one block may in this case be scanned one by one in each case and the resulting measurement signals (D_v) may be compared to the value on a comparator in the local part of the evaluation system (22). The sensor sensitivity corresponding to the motion sensor (16, 17) to be scanned (and optionally also corresponding to the detection zone, if a difference is made in sensitivity in two different detection zones of one machine cycle) may in this case be filled in.

If desired, the control unit (19) (for example designed as a microprocessor) may additionally be provided with a central part of the evaluation system (22) (implemented in the microprocessor), for example if the control unit (19) further compares, based on the moving average (Dma), to said one or more limit values or if the motion sensors (16, 17) generate a different signal (S) for each transgression of a respective limit value and the control unit (19) determines, on the basis of this signal (S), whether and which alarm should be generated and whether the tufting machine (1) is possibly stopped. Alternatively, it would also be possible to have the measurement signals (D_v) be read in by the control unit (19) and for the evaluation system (22) to completely form part of the control unit (19).

By distributing the evaluation system (22) across local parts for one or more motion sensors (16, 17), only limited information has to be exchanged between these motion sensors (16, 17) and the control unit (19) (microprocessor), as the measurement signals (D_v) themselves do not have to be forwarded to the control unit (19). If the measurement signals (D_v) were to be forwarded to the control unit (19), more complex evaluations could be implemented in the central part of the evaluation system (22) and/or further statistical processing of measurement signals (D_v) could be implemented for a longer period of time and/or of measurement signals (D_v) of various motion sensors (16, 17) with respect to each other.

Each of the motion sensors (16, 17) is assigned a separate identification signal which is sent together with the information of this motion sensor (16, 17) to be forwarded, so that it is possible to record where any errors occur.

The monitoring system (20) furthermore comprises an adjusting unit (21) (for example a touchscreen) for adjusting said limit values and/or a specific time during which a moving average (Dma) is to be determined and/or a sensor sensitivity for generating the measurement signals (D_v) and/or the type of pile yarn and/or the type of detection, etc.

In addition, the monitoring system (20) may comprise a reading unit (23) for reading in data from the tufting machine (1), such as for example the pile pattern and/or the machine speed at which the tufting machine (1) is driven, etc. In order to read in the data, use may optionally be made of a conventional fieldbus or of a separate position channel. Optionally, but less preferred, the data may also be forwarded wirelessly.

The adjusting unit (21) and/or the reading unit (23) may for example form part of the control unit (19), as is illustrated in Figure 2.

In this case, the control unit (19) of the monitoring system (20) may be integrated in an existing control unit of the tufting machine (1) which is additionally configured to control the monitoring system (20), both with completely new tufting machines (1) according to the present invention and with any existing tufting machines (1) which are modified to become tufting machines (1) according to the present invention. The motion sensors (16, 17) are then installed on such a tufting machine (1), and the control unit of the tufting machine (1) is coupled to the motion sensors (16, 17) in order to control these motion sensors (16, 17) and to read in signals generated by the motion sensors (16, 17).

Alternatively, it is also possible to configure this control unit (19) completely separately from an existing control unit of a tufting machine (1), as a component of a monitoring system (20) according to the present invention, so that a monitoring system (20) according to the present invention may also be provided as a separate unit, as a result of which an existing tufting machine (1) can easily be upgraded. It is then for example possible to couple this control unit (19) of the monitoring system (20) to a control unit which is already present in the existing tufting machine (1), for example in order to read in the machine speed (V_m) in order to be able to adjust the sensor sensitivity on the basis thereof or to pass on alarms in order to stop the tufting machine (1) on the basis thereof. The control unit (19) of the monitoring system (20) may, for example, also be coupled to a control unit which is already present in the existing tufting machine (1) in order to read in other data by means of a said reading unit (23) so as to take these data into account when adjusting the sensor sensitivity and/or evaluating the measurement signals (D_v). Thus, the former may be configured to read in a pile pattern in order to determine machine position data (D_m) on the basis thereof for evaluating the measurement signals (D_v) based on these machine position data (D_m). The movement sensors (16, 17) are then installed on this tufting machine (1), and the control unit (19) of the monitoring system (20) is optionally coupled to the control unit of the tufting machine (1).

Measurement signals (D_v) are generated by means of one or more motion sensors (16, 17). The evaluation system (22) determines a moving average (Dma) of these measurement signals (D_v) over a specific time period for each motion sensor (16, 17) and determines whether this moving average (Dma) exceeds a first limit value and/or drops below a second limit value.

Several limit values may be provided in case of an increase in the moving average (Dma) and several limit values may be provided in case of a decrease in the moving average (Dma). Thus, it is possible, for example, for an alarm to be generated when the moving average (Dma) increases above the first limit value and for the tufting machine to be stopped above a third limit value, which is higher than the first limit value. Analogously, it is possible to generate an alarm if the moving average (Dma) drops below the second limit value, and for the tufting machine (1) to be stopped below a fourth limit value, which is lower than the second limit value.

Furthermore, various possible limit values may also be provided for various possible detections.

In this case, the limit value(s) is/are adjustable by means of the adjusting unit (21). The alarms to be generated and/or the optional stopping of the tufting machine (1) may also be provided to be adjustable via the adjusting unit (21). The specific time is also adjustable by means of the adjusting unit (21). In this case, this specific time may be chosen, for example, in function of the desired detection. Thus, for a BED detection, it is for example possible to choose 1 machine cycle as the specific time, or a few machine cycles as the specific time. More specifically, in this case, for example approximately 10 machine cycles may be chosen as the specific time. For a TED detection, it is possible to choose, for example, a longer time or a few tens of cycles. More specifically, in this case, for example approximately 100 machine cycles may be chosen as the specific time.

At 2000 revolutions per minute, this means, for example, a specific time of 0.3 s for a BED detection and 3 s for a TED detection. At 1500 revolutions per minute, 0.4 s and 4 s, respectively, and at 600 revolutions per minute, 1 s and 10 s, respectively.

The measurement signals (D_v) are generated at a specific sensor sensitivity. This sensor sensitivity is also adjustable by means of the adjusting unit (21).

If the sensor sensitivity is too high, movements are detected at machine positions in which the needle is stationary. If the sensitivity is too low, the probability of a movement being detected at a high needle speed is too low (e.g. < 0.8). If the sensitivity is too high, such a piezoelectric sensor may miss missing pile yarn, and if the sensitivity is too low, the piezoelectric sensor may emit incorrect reports of yarn breakage.

In order to determine a value for the sensor sensitivity to be adjusted, the monitoring system (20) may comprise an adjusting system (24) for adjusting the sensor sensitivity of a motion sensor (16, 17). This adjusting system (24) may form part, for example, of the control unit (19).

In order to read in the process data of the tufting machine (1), such as the machine speed of the tufting machine (1), the machine acceleration, the characterizing features of the pile yarn (3), the type of detection or pattern information, the adjusting system

(24) comprises a data unit (25). In this case, use may optionally be made of a conventional fieldbus or of a separate position channel or, less preferred, the data may also be forwarded wirelessly.

More specifically, the aforementioned reading unit (21) may also serve as a data unit

(25). The adjusting system (24) furthermore comprises a storage unit (26) in which reference data for the sensor sensitivity of the motion sensor (16, 17) are stored, consisting of one or more sets in which each set comprises a value of a reference sensor sensitivity and corresponding process data. The one or more sets of reference data define a connection between the value of the reference sensor sensitivity and the corresponding process data. This connection may be expressed in a look-up table or in a mathematical function.

The adjusting system (24) furthermore comprises a calculation unit (27) in which the connection between the current process data and the process data of the reference data is determined and in which a value for the sensor sensitivity to be adjusted is determined on the basis of this connection and the values of the reference sensor sensitivity of the reference data.

In the calculation unit (27), either all reference data are retained or the most relevant reference data, based on the current values of the process data, are retained. Only sets of reference data with the same machine type or with a machine speed in a specific interval may, for example, be selected.

Thereafter, the calculation unit (27) can determine a mathematical relation between the current process data and the process data of the retained reference data, and determine a value for the sensor sensitivity to be adjusted by applying this relation to the reference sensor sensitivities of the retained reference data.

By using reference data, it is possible to set a sensor sensitivity which is as accurate as possible on the basis of the current process data.

An additional set of reference data, consisting of a value of a reference sensor sensitivity and corresponding process data, can be determined during a learning cycle. To this end, an expected percentage detection value is first determined.

The percentage detection value is the percentage of the measurement signals (D_v) which indicates a movement.

The aim is for the motion sensor (16, 17) to detect the yam movement and not to falsely detect the non-yam movement. Based on the physical knowledge of the tufting process, in which the pile yam (3) is incorporated into the fabric (7), it is known for how much time of every machine cycle the pile yam (3) moves. In each machine cycle, there is a zone in which there is movement in any case which has to be detected, and there is a zone in which there is no movement and where consequently none should be detected. On the basis thereof, it is also possible to determine what percentage of the measurement signals (D_v) should be permitted to indicate a movement. This percentage is preferably chosen as the specific expected percentage detection value which is strived for in said learning cycle.

In the motion sensor (16, 17), a starting value for the sensor sensitivity may then be set, for example determined from the already existing reference data.

The amount of movement detected by the motion sensor (16, 17) can then be compared to this expected percentage detection value per unit time. If this amount of movement does not correspond to the expected percentage detection value per unit time, that is to say it is not within specific predetermined limits of the percentage detection value, e.g. within 20%, 10%, 5% of the percentage detection value, the sensor sensitivity, set in the motion sensor (16, 17), can then be adjusted until the detected amount of movement does correspond to the expected percentage detection value per unit time. Subsequently, the value of the sensor sensitivity set in the motion sensor (16, 17) can be added in the reference data as a value of the reference sensor sensitivity, together with the current process data.

In this way, a large amount of reference data can be determined, so that a reference sensor sensitivity is available for as many different process data as possible.

In order to further improve or refine the value of the set sensor sensitivity, a learning cycle may additionally be completed during monitoring of the tension of a pile yarn (3). In this case, the sensor sensitivity may be adjusted, while aiming for a specific acceptable percentage detection value of the movement. The acceptable percentage detection value can be determined in the same way as the expected percentage detection value, it being possible for the expected percentage detection value and the acceptable percentage detection value to have different predetermined limits within which the detected amount of movement has to fall in order to correspond. This acceptable percentage detection value may also be provided in the adjusting unit (21) so as to be adjustable. This percentage is preferably chosen as the specific percentage detection value which is strived for in said learning cycle. This may be, for example, 30% as specific acceptable percentage detection value. If, with a set sensor sensitivity, the detected percentage detection value deviates greatly from this specific acceptable percentage detection value, for example 70% to 80%, when the specific acceptable percentage detection value has been adjusted to 30%, then it is clear that this sensor sensitivity has not been adjusted correctly. The sensor sensitivity in the learning cycle is then adjusted until the detected percentage detection value corresponds to the specific acceptable percentage detection value.

Depending on the type of tufting machine (1) and/or the desired detections, such a sensor sensitivity may be determined in this way for various types of pile yarn (3) and/or for various types of detections and/or for various pile deliveries and/or pile heights and/or on the basis of needle selection data, etc. Below, some specific examples are discussed in more detail. Further sensor sensitivities may be determined, for example via interpolation, and/or may be worked out more precisely with a selflearning system.

This sensor sensitivity is preferably configured to be adjustable and preferably individually adjustable for each motion sensor (16, 17).

The sensor sensitivity may, for example, be configured to be automatically adjustable on the basis of a desired detection, such as for example a TED detection or a BED detection.

In existing tufting machines (1), the sensor sensitivity is optimized for an operating speed of the tufting machine (1). At other machine speeds, only less accurate measurements are possible as the sensor sensitivity is optimized for a different operating speed. By adjusting the sensor sensitivity on the basis of the machine speed (revolutions per minute), more accurate detections at different machine speeds become possible. In order to adjust the sensor sensitivity on the basis of this machine speed, the optimum sensor sensitivity may be determined at 2 or more machine speeds in an above-described learning cycle. Via interpolation, sensor sensitivities to be set for other machine speeds can then be determined. If the tufting machine (1) accelerates or decelerates, the sensor sensitivity may in this case also be adjusted on the basis of this acceleration or this deceleration of the machine speed. Thus, errors can be detected as early as possible under all circumstances.

The optimum set sensor sensitivity can then also be added to the reference data as reference sensor sensitivity, together with the current process data.

If the motion sensor (16) is arranged between the feeding device (4) and the tufting needle (12), a statistical distribution of these measurement signals may be determined for each machine position in order to determine the sensor sensitivity of the motion sensor (16) to be set on the basis of measurement signals of this motion sensor (16) over several machine cycles as an alternative for the expected percentage detection value or the acceptable percentage detection value.

In a learning cycle over various machine cycles, the sensor sensitivity of this motion sensor (16) may then be adjusted until the specific statistical distribution virtually corresponds to the typical up and down needle movement over a machine cycle, such as for example described in patent application BE 2023/5287.

By means of this statistical distribution, the needle cycle of a tufting needle (12) with a specific pile yam (3) arranged therein can be determined specifically by means of the measurement signals using the corresponding motion sensor (16). The probability with which the motion sensor (16) determines a yarn movement normally virtually corresponds to a typical needle movement. A typical needle movement means that little to no yam movement is detected when the tufting needle (12) is at its highest point or when the tufting needle (12) is at its lowest point and maximum movement is detected halfway between both points.

Claims

1. Method for adjusting the sensor sensitivity of a motion sensor (16, 17) for detecting movement of a pile yam (3) in a tufting machine (1), provided with several tufting needles (12), wherein this pile yarn (3) is incorporated into a fabric (7), characterized in that this method comprises the following steps: a) determining current process data; b) providing reference data for the sensor sensitivity of the motion sensor (16, 17), consisting of one or more sets, wherein each set comprises a value of a reference sensor sensitivity and corresponding process data; c) determining a connection between the current process data and the process data of the provided reference data; d) determining a value for the sensor sensitivity to be adjusted on the basis of this connection and the one or more values of the reference sensor sensitivity of the provided reference data; e) adjusting the value for the sensor sensitivity to be adjusted in the motion sensor (16, 17).

2. Method according to Claim 1, wherein the process data comprise the machine speed of the tufting machine (1) in order to thereby control the tufting machine (1).

3. Method according to Claim 2, wherein the process data in addition comprise one or more of the following parameters: machine acceleration of the tufting machine (1), pattern information of the fabric (7) to be produced, yam thickness of the pile yam (3), yarn type of the pile yarn (3), machine type of the tufting machine (1), desired type of detection.

4. Method according to Claim 2 or Claim 3, wherein the one or more sets of reference data define a connection between the values of the reference sensor sensitivity and the corresponding process data expressed in a look-up table and/or a mathematical function.

5. Method according to one of the preceding claims, wherein step b) is further characterized by determining specific reference data from the provided reference data for the sensor sensitivity of the motion sensor (16, 17), based on the current process data and wherein, in step c), a mathematical relation is determined between the process data of the specific reference data and the current process data in order to determine said connection and wherein, in step d), the sensor sensitivity is determined by applying the mathematical relation to the values of the reference sensor sensitivities of the specific reference data.

6. Method according to one of the preceding claims, wherein providing reference data for the sensor sensitivity of the motion sensor (16, 17) in step b) comprises determining a set of reference data in a learning cycle, consisting of a value of a reference sensor sensitivity and corresponding process data, wherein this learning cycle comprises the following steps: determining an expected percentage detection value per unit time; setting a starting value for the sensor sensitivity of the motion sensor (16, 17); detecting an amount of movement per unit time by means of the motion sensor (16, 17); comparing this amount to the specific expected percentage detection value per unit time; adjusting the sensor sensitivity set in the motion sensor (16, 17) until the amount of detected movement per unit time corresponds to the specific expected percentage detection value per unit time; adding to the reference data the set consisting of the value of the sensor sensitivity, set in the motion sensor (16, 17), when the amount of detected movement per unit time corresponds to the specific expected percentage detection value per unit time, as a value of the reference sensor sensitivity and the current process data.

7. Method for monitoring the tension of a pile yarn (3) in a tufting machine (1) provided with several tufting needles (12), wherein this pile yarn (3) is incorporated into a fabric (7), comprising : a. generating measurement signals (D_v) by means of a motion sensor (16, 17) which are an indication of the pile consumption of this pile yarn (3); b. determining a moving average (Dma) of the measurement signals (D_v) over a specific time period; and c. determining whether this moving average (Dma) exceeds a first limit value and/or drops below a second limit value; characterized in that a sensor sensitivity is adjusted in the motion sensor (16, 17) which is adjusted according to a method according to one of Claims 1 to 6.

8. Method according to Claim 7, characterized in that this method furthermore comprises: a. determining an acceptable percentage detection value per unit time; b. detecting an amount of movement per unit time by means of the motion sensor (16, 17); c. comparing the amount of detected movement to the acceptable percentage detection value.

9. Method according to Claim 8, characterized in that this method furthermore comprises adjusting the sensor sensitivity set in the motion sensor (16, 17) until the amount of detected movement per unit time corresponds to the acceptable percentage detection value per unit time.

10. Method according to Claims 8 or 9, characterized in that this method furthermore comprises adding to the reference data the set consisting of the sensor sensitivity set in the motion sensor (16, 17) as the reference sensor sensitivity and the current process data when the amount of detected movement corresponds to the acceptable percentage detection value.

RECTIFIED SHEET (RULE 91) ISA/EP

11. Method according to Claim 8 to 10, characterized in that the acceptable percentage detection value is adjustable.

12. Method according to one of Claims 7 to 11, characterized in that the specific time is adjustable.

13. Method according to one of Claims 7 to 12, characterized in that the limit value is adjustable.

14. Method according to one of Claims 7 to 13, characterized in that the pile consumption of the pile yam (3) is measured by means of the motion sensor (17) between a yam storage system (2) and a feeding device (4) for supplying this pile yarn (3) from the yam storage system (2) in the tufting machine (1).

15. Method according to one of Claims 7 to 14, characterized in that the measurement signals (D_v) are generated by means of an optical sensor as said motion sensor (16, 17).

16. Method according to one of Claims 7 to 14, characterized in that the measurement signals (D_v) are generated by means of a piezoelectrical sensor as said motion sensor (16, 17).

17. Method according to one of Claims 7 to 16, characterized in that the tension of one or more additional pile yams (3) is monitored by generating corresponding additional measurement signals (D_v) by means of one or more corresponding additional motion sensors (16, 17), and in that for each monitored pile yam (3) a moving average (Dma) of the corresponding measurement signals (D_v) over a specific time period is determined and it is determined whether this moving average (Dma) exceeds a first limit value and/or drops below a second limit value.

18. Adjusting system (24) for adjusting the sensor sensitivity of a motion sensor (16, 17) for detecting movement of a pile yarn (3) in a tufting machine (1), provided with several tufting needles (12), wherein this pile yarn (3) is incorporated into a fabric (7), characterized in that this adjusting system (24) comprises: a. a data unit (25) for reading in current process data; b. a storage unit (26) for storing reference data for the sensor sensitivity of the motion sensor consisting of one or more sets, wherein each set comprises a value of a reference sensor sensitivity and corresponding process data; and c. a calculation unit (27) for:

- determining a connection between the current process data and the process data of the provided reference data;

- determining a value for the sensor sensitivity to be adjusted on the basis of this connection and the one or more values of the reference sensor sensitivity of the provided reference data; wherein the adjusting system (24) is provided for adjusting the value for the sensor sensitivity to be adjusted in the motion sensor (16, 17).

19. Adjusting system (24) according to Claim 18, characterized in that this adjusting system (24) is provided for adjusting the sensor sensitivity according to a method according to one of Claims 1 to 6.

20. Monitoring system (20) for monitoring the tension of a pile yam (3) in a tufting machine (1), wherein this pile yam (3) is incorporated into a fabric (7), comprising a motion sensor (16, 17) for generating measurement signals (D_v) which are an indication of the pile consumption of this pile yarn (3), and an evaluation system (22) for determining a moving average (Dma) of the measurement signals (D_v) over a specific time and for determining whether this moving average (Dma) exceeds a first limit value and/or drops below a second limit value, characterized in that this monitoring system (20) comprises an adjusting system (24) according to Claim 18 of 19.

21. Monitoring system (20) according to Claim 20, characterized in that this monitoring system (20) comprises an adjusting unit (21) for adjusting the specific time and/or for adjusting the limit value.

22. Tufting machine (1), comprising an adjusting system (24) according to Claim

18 or 19, or a monitoring system (20) according to Claim 20 or 21.