EP1262581A1 - Method for controlling the manufacturing quality of textile slivers - Google Patents

Abstract

A comparison in line or individual measurements/values taken simultaneously from emerging tapes (2a, 2b, 2c) or entering webs on the same stretching or gill machine located before preparation finishers and equipped with a rack (3), feed cylinder assembly (4a, 4b, 4c), fibre monitoring zone (5a, 5b, 5c), assembly of stretching cylinders (6a, 6b, 6c) and measurement detectors (9a, 9b, 9c). <??>The procedure consists of making a comparison in line or individual measurements or values taken simultaneously from at least two emerging tapes (2a, 2b, 2c) or at least two entering webs on the same stretching or gill machine located before preparation finishers and equipped with a rack (3), at least one feed cylinder assembly (4a, 4b, 4c), at least one fibre monitoring zone (5a, 5b, 5c), at least one assembly of stretching cylinders (6a, 6b, 6c) and measurement detectors (9a, 9b, 9c). The detectors are positioned between the stretching cylinders and storage units (8a, 8b, 8c), and the data from them is used to produce graph curves based on progressive or sliding values. The values are selected so as to filter out any very short term variations and show only medium or long term variations, and each curve is compared with a progressive mean. When faults are detected a light or sound alerting signal is given, and the machine halted when the size of the fault is unacceptable.

Description

The present invention relates to the field of the textile machines, in particular drawing machines commonly called Gills, and relates to a quality control process for manufacture of textile fiber ribbons.

Currently, there are many control devices in line of manufacturing quality on drawing machines. These devices, whose measurement principle differs depending on the model, aims to control the variation of the titer in g / m or in Ktex of a fiber ribbon textiles leaving or entering a drawing machine.

For this purpose, we know, from CH-A-441 067, a device for measure capable of determining long-term variations in the title of a tape by weighing the quantity of ribbon deposited in a jar per unit of length of ribbon. The main disadvantage of this device is that the measurement is distorted when part of the tape is removed from the jar, i.e. for a periodic check, either because of an incident. In addition, such a device does not not allow to control short term variations.

WO-A-98/18985 and FR-A-2 768 437 respectively describe, moreover, the control of the volume occupied by the fibers which make up the tape using a feeler roller or a feeler. These devices are bulky and high inertia, so they are unsuitable for detection short-term variations, and ill-suited to high speeds when are placed at the exit of the stretching machine where they create windings of fibers.

In addition, the measuring principle requires compressing the tape. output, which deteriorates its quality.

FR-A-1 473 942 describes a deformation measurement method elastic, in which the fiber ribbon passes through a funnel reduced size, which slows down the passage of the tape and deformation elasticity of a material is monitored using a sensor, which can be at strain gauge, inductive, capacitive or optical. The size of the funnel must exactly match the title of the outgoing ribbon and should be change each time the title of the outgoing ribbon is changed. In addition, the measuring principle deteriorates the quality of the outgoing tape.

We also know a measurement method by radiation, in which the ribbon moves in front of a Beta radiation source by example (FR-A-1 209 535). The title of the ribbon is proportional to absorption of Beta particles. This kind of device presents the disadvantage of irradiating the textile material and poses the problem of recycling ray sources.

There is also an optical measurement method using operates a device as known, for example, by FR-A-1 375 439 and which consists in circulating the ribbon in front of a light source emitting on a wavelength of the visible range and to detect variations of light using an electric photo cell. Textile material absorbs part of the electromagnetic waves and variations in light are proportional to the title of the ribbon.

The devices used for this purpose have the disadvantage major to be very sensitive to the color of the fibers and to display results aberrant when the textile fibers are dyed.

It has also been proposed, by FR-A-2 520 119, a method of sonic measure consisting in moving the tape in a room, in which we emit a sound wave. The textile material absorbs and slows down the propagation of the sound wave according to the title of the ribbon. This disadvantage of being very sensitive to deposits of dust and fiber in the measuring chamber, so that we observe gradually a drift in the quality of the measurement.

A pneumatic measurement method has been described in BE-A-873 050 and proposes to pass the fiber ribbon through a funnel, in which the air pressure is measured, the value of which is proportional to the titer ribbon. However, this method has the disadvantage of being sensitive to the linear speed of the ribbon.

We also know a capacitive measurement method, according to DE-A-1 043 886, which consists in passing the tape between the electrodes of an electric capacitor, of which it modifies the dielectric capacity. The variations in dielectric capacity are a function of the title of the ribbon. The sensor implemented, although it is non-contact, very high sensitivity and very fast, however has the disadvantage of being sensitive to variations in the environment, including temperature and humidity of the ambient air, so that we observe drifts of the measured.

The object of the present invention is to overcome the drawbacks known devices and methods by proposing a new method of quality control of the manufacture of textile fiber ribbons allowing an online check.

To this end, the method is characterized in that it consists of compare, online, individual measurements performed simultaneously on at least two outgoing ribbons or on at least two incoming layers of a same drawing machine, placed upstream of preparation finishers, at an average reference itself determined from these measurements individual.

La figure 1 représente une machine d'étirage mettant en oeuvre le procédé conforme à l'invention, et

les figures 2a à 2f sont des représentations graphiques des signaux de mesure individuelle délivrés par le dispositif suivant la figure 1.

The invention will be better understood from the description below, which relates to a preferred embodiment, given by way of nonlimiting example, and explained with reference to the appended schematic drawings, in which:

FIG. 1 represents a drawing machine implementing the method according to the invention, and

FIGS. 2a to 2f are graphic representations of the individual measurement signals delivered by the device according to FIG. 1.

Figure 1 of the accompanying drawings shows, for example, a drawing machine intended to produce three ribbons. Such a machine drawing 1, such as a Gills, making it possible to simultaneously produce at least two ribbons 2a, 2b, 2c is provided with a rack 3, at least one set of food cylinders 4a, 4b, 4c, of at least one zone for controlling the fibers 5a, 5b, 5c, of at least one set of stretching cylinders 6a, 6b, 6c, funnels 7a, 7b, 7c and at least one storage device 8a, 8b, 8c. This machine is also equipped with measurement sensors 9a, 9b, 9c and / or 10a, 10b, 10c. Thus, measurement sensors 9a, 9b, 9c are advantageously placed between the stretching cylinders 6a, 6b, 6c and the device 8a, 8b, 8c.

According to the invention, this machine 1 implements a quality control process for manufacturing textile fiber ribbons which consists of comparing the individual measurements made or values individual read simultaneously on at least two outgoing ribbons 2a, 2b, 2c or on at least two incoming sheets from the same machine 1, placed upstream of preparation finishers, to a reference average itself determined from these individual measurements.

A graphic example of the results obtained is given by Figures 2a to 2c, where A is the amplitude of the signal delivered by the sensor 9a, B the amplitude of the signal delivered by the sensor 9b, and C the amplitude of the signal supplied by sensor 9c.

According to a characteristic of the invention, the method consists in producing for each sensor 9a, 9b, 9c, etc. a curve AT , B , VS corresponding to a progressive or sliding average of "m" individual values provided by said sensors.

According to another characteristic of the invention, the method consists in then performing a curve ABC progressive average or sliding average of the arithmetic mean of the individual amplitudes of "n" values A, B and C, so as to smooth the evolution curve. This progressive average forming the curve A, B, C is shown in Figure 2d.

The values "n" and "m" are chosen so that the very short term variations are filtered out and only the medium and long term variations.

Each individual curve AT , B , VS is compared, according to another characteristic of the invention, with the curve A, B, C representing the progressive average and any exceedance of a predetermined difference threshold results in the transmission of either detected fault information or a machine stop signal due to fault of unacceptable size. To this end, the emission of an alert signal will be given in a light or sound manner and supplemented by information in text mode describing the type of fault detected.

Defects are classified according to different types which can be detected according to freely chosen thresholds, both in duration and amplitude, i.e. thresholds for detecting a fault T4, T5, warning thresholds for faults of different sizes S1 and S2 and a stop threshold S3 for unacceptable size defect. So, for example, the duration threshold T4 may be of relatively short duration, while the T5 threshold is more bigger than T4. The alert thresholds S1 and S2 can be chosen so as to signal to the operator different sizes of faults that one accepts to produce, while the stop threshold S3 is chosen so as to command machine shutdown due to size defect deemed unacceptable. Other thresholds can obviously be chosen to replace or complement.

Figure 2f shows the magnitude of a disturbance Z such as, for example, increasing the relative humidity of the ambient air. From t1, there is an increase in the relative humidity of the air surrounding the stretching machine 1. This disturbance creates a drift of the signals delivered by the sensors 9a, 9b, 9c, which is represented by an increase in the amplitude of signals A, B and C.

From t2 to t3 we note the absence of a ribbon of the drawing machine 1. This absence is detected by the sensor 9b.

By comparison of curve B with curve A, B, C, there appears between t2 and t3 a difference which exceeds in duration and in amplitude both the duration thresholds T4 or T5, the alert thresholds S1 and S2 and the stop threshold S3. Thus, for example, the alert will be given in a luminous or audible manner and supplemented by information in text mode describing the type of fault detected and a command to stop the machine will be issued.

Faults can be counted according to their type, as shown, for example, below: Standard fault duration Amplitude Feature 1 T 4 S 2 Short, high amplitude 2 T 5 S 1 Long, low amplitude 3 T 5 S 2 Long, high amplitude Etc ...

According to an alternative embodiment of the invention, it is also possible to determine the curve ABC starting by performing the progressive average of A, B and C over "m" values so as to obtain the curves AT , B and VS then by doing the arithmetic mean of AT , B and VS before calculating the progressive average ABC. on "n" values.

Thanks to the invention it becomes possible to get rid of disadvantages linked to indirect contactless measurements of the optical type, sonic, pneumatic or capacitive. Indeed, faced with a disturbance the inevitable measurement drifts of these types of sensors are supposed to be identical between two sensors of the same principle.

Preferably the quality control of the manufacturing of ribbons textile fibers is placed on the last stretching machine making up a preparation line, which is located upstream of the paver, such as the wiper or the spindle bench or upstream of the spinning machine, in the case of a preparation line for direct spinning without finisher.

In this case, generally this last machine of preparation simultaneously produces at least two ribbons from the same batch and more commonly four ribbons. This machine feeds the finishers producing wick coils which are difficult to recycle in the event of default, (respectively the spools of thread in the case of direct spinning). Of this fact, it is better to detect faults at the end of the last drawing machine producing ribbons which are easily recyclable if necessary.

The example according to figure 2f represents the evolution of a disturbance Z as a function of time "t", such as, for example, a rapid increase in the relative humidity of the air surrounding the machine. The amplitude of this variation is relatively greater than its consequence measured by the sensors 9a, 9b, and 9c, which is shown in Figures 2a, 2b and 2c.

In the same way we observe that the appearance of the disturbance Z does not reproduce directly on curves AT , B and VS . Also, identifying a fault would be difficult if it were carried out, for example an operation AT - Z .

In accordance with another alternative embodiment of the invention, measurement sensors 10a, 10b and 10c can be placed between the rack 3 and the food cylinders 4a, 4b, and 4c.

These sensors 10a, 10b and 10c can also be placed on each individual ribbon 12 supplying the machine, upstream of the cylinders food 4a, 4b, and 4c.

In the case of individual sensors on ribbons, said sensors may, alternatively, be able to control only the presence or the absence of a power ribbon and care should be taken to ensure that each drawing head of the machine is well fed by the number adequate ribbons.

It is also possible to place sensors 11a to 11f of the surrounding environment in addition to sensors 9a, 9b, 9c or 10a, 10b and 10c.

The sensor 11a is an empty textile material sensor, of the same type as those, 9 or 10, used on the machine 1, while the sensors 11b, 11c and 11d respectively measure the temperature of the ambient air, ambient air humidity and machine speed and the 11th sensor records the machine usage parameters, such as the type of material, title, etc ...

By correlating information from environmental sensors 11a to 11f with the values measured by the sensors 9a, 9b, 9c or 10a, 10b and 10c, the accuracy of the control process can be improved the quality of manufacture of textile fiber ribbons, object of the invention.

Of course, the invention is not limited to the mode of embodiment described and shown in the accompanying drawings. Modifications remain possible, particularly from the point of view of the constitution of different elements or by substitution of technical equivalents, without leaving provided the scope of protection of the invention.

Claims (12)

Method for controlling the quality of manufacturing textile fiber ribbons, characterized in that it consists in comparing, on line, the individual measurements (A, B, C) made or individual values recorded simultaneously on at least two outgoing ribbons (2a , 2b, 2c) or on at least two incoming sheets of the same drawing machine (1), placed upstream of preparation finishers and provided with a rack (3), of at least one set of food cylinders (4a, 4b, 4c), at least one fiber control zone (5a, 5b, 5c) and at least one set of stretching cylinders (6a, 6b, 6c), by means of measurement sensors ( 9a, 9b, 9c and / or 10a, 10b, 10c), to an average reference which is itself determined from these individual measurements. Method according to claim 1, characterized in that measuring sensors (9a, 9b, 9c) are placed between the stretching cylinders (6a, 6b, 6c) and a storage device (8a, 8b, 8c). Method according to either of Claims 1 and 2, characterized in that it consists in producing for each sensor (9a, 9b, 9c, etc.) a curve ( AT ), ( B ), ( VS ) corresponding to a progressive or sliding average of "m" individual values provided by said sensors. Method according to any one of Claims 1 to 3, characterized in that it consists in carrying out a curve ( ABC ) of progressive average or sliding average of the arithmetic average of the individual amplitudes of "n" values (A), (B) and (C), so as to smooth the evolution curve. Method according to any one of Claims 1 to 4, characterized in that the values "n" and "m" are chosen so that the very short-term variations are filtered and that only the medium-term variations appear and long term. Method according to any one of Claims 1 to 4, characterized in that each individual curve ( AT ), ( B ), ( VS ) is compared with the curve ( ABC ) representing the progressive average and any exceedance of a predetermined difference threshold results in the transmission of either detected fault information or a machine stop signal due to fault of unacceptable size. Method according to claim 6, characterized in that the emission of an alert signal is given in a light or sound manner and supplemented by information in text mode describing the type of fault detected. Method according to claim 6, characterized in that the faults are classified according to different types which can be detected as a function of freely chosen thresholds, both in duration and in amplitude, namely thresholds for duration of detection of a fault (T4 , T5), alert thresholds for faults of different sizes (S1 and S2) and a stop threshold (S3) for faults of unacceptable size. Method according to claim 1, characterized in that it consists in determining a curve ( ABC ) starting with the progressive average of (A), (B) and (C) over "m" values so as to obtain the curves ( AT , B and VS ) then by doing the arithmetic mean of ( AT ), ( B ) and ( VS ) before calculating the progressive average ( ABC ) on "n" values. Method according to claim 1, characterized in that measuring sensors (10a, 10b and 10c) are placed between the rack 3 and the food cylinders (4a, 4b, and 4c). Method according to claim 10, characterized in that the sensors (10a, 10b and 10c) are placed on each individual strip (12) supplying the machine, upstream of the food cylinders (4a, 4b, and 4c). Method according to any one of claims 1, 3, 10 and 11, characterized in that sensors (11a to 11f) of the surrounding medium are placed in addition to the sensors (9a, 9b, 9c or 10a, 10b and 10c) .

Images