BACKGROUND OF THE INVENTION

The present invention relates to a method and apparatus for controlling a jet loom, according to the preamble of the independent claims.

DESCRIPTION OF THE PRIOR ART

So-called yarn clearers are used in spinning or cone winding machines for ensuring the yarn quality. Such an apparatus is known for example from EP-0'439'767 A2. It contains a measuring head with at least one sensor which scans the moving yarn. Frequently used sensor principles are the capacitive one (see EP-0'924'513 Al for example) or the optical one (see WO-93/13407 Al for example).

It has been proposed to scan the yarn with different sensors and to combine the different sensor signals for evaluation. For instance US-6,798,506 B2 teaches to arrange two sensors one after the other along the yarn path. A first of the sensors measures the optical reflection from the yarn; a second of the sensors measures capacitively or optically the mass or the diameter, respectively, of the yarn. The output signals of the two sensors are evaluated according to certain evaluation criteria. Based on the evaluation, at least two kinds of foreign matters are distinguished from each other.

In accordance with US-4,559,976 A, a yarn sensor is provided upstream of the weft storage unit of a loom. The weaving operation of the loom is interrupted when the sensor detects a yarn defect. The flawed yarn section is deflected from its normal passage.

WO-2006/133833 Al discloses a similar arrangement of a yarn sensor on a loom. When the yarn sensor detects any unevenness in the yarn, the weft yarn is inserted at a lower speed into the shed than the normal speed for the respective yarn. Thus, the risk of breaking of the weft thread is minimized. In contrast to the two publications mentioned above, ΕΡ-2Ί 75Ό58 Al discloses a loom in which a yarn sensor is arranged between the weft feed and the shed. The yarn sensor arranged in such a way can also detect yarn defects which originate only in the loom. Better quality monitoring is thus achieved.

US-2008/0185066 Al discloses at least one yarn sensor in an air-jet loom which measures the speed and/or the covered path of the weft yarn during the weft insertion. The measured values of the at least one yarn sensor are used for controlling the weft insertion.

US-7,654,290 B2 teaches controlling a jet of an air-jet loom as a function of the axial velocity of the weft yarn. For determining the velocity, an electrode array registers in a contactless manner electrostatic induction charges generated by the natural thread charges arranged irregularly on the weft yarn. The changing total charge on the electrode array is determined as a narrow-band frequency spectrum concentrated around a main component. Said main component's frequency is proportional to the velocity of the weft yarn. By integrating the velocity it is furthermore possible to determine the position of the tip of the weft yarn at a later instant, and the weft-yarn length can also be determined.

US-4,450,876 Al aims at compensating the influence of a yarn-package change on the weft time interval in a jet loom. A detection device detects the yarn-package change. As a reaction to the detection signal a temporary change of the feed pressure of the blowing nozzle is caused. The feed pressure depends on the yarn to be used.

In accordance with US-5,1 15,840 A, the flying characteristic of a weft yarn for each picking is monitored by an arrival angle detector arranged downstream of the shed. At least a lower limit value of jet pressure is automatically set on the basis thereof to thereby realize a continuation of stable picking operation without the occurrence of picking defects.

EP-0'573'656 Al proposes a neural network for controlling an air-jet loom. Input parameters of the neural network include weft-yarn parameters such as the kind (staple or filament yarn) and the count (average linear mass density, measured in denier) of the weft yarn. The weft-yarn parameters are input manually, but a sensor can be used for measuring the yarn count. Based on the output of the neural network, weaving conditions are changed or a warning signal is output. In accordance with WO-89/12122 Al, the relay nozzles of an air-jet loom which are arranged behind one another are actuated successively in a pulse-like manner. The pulse length of the actuating is controlled depending on the air effectiveness of the respectively processed yarn. The ratio of the carrying area of the yarn in air in comparison to the yarn mass is defined as the parameter of air effectiveness. The relevant yarn parameters such as yarn thickness, yarn hairiness or yarn mass can be detected with respective yarn sensors. The consideration of the air effectiveness of the yarn allows processing very different yarns with high quality in successive work cycles. A method and apparatus for controlling a loom, e.g. a jet loom, are known from

EP-2' 157'218 Al . A weft yarn is drawn off a cone and a first sensor is used for detecting a quality feature of the drawn weft yarn, e.g. a change in diameter. A loading variable of the weft yarn, e.g. the speed as a function of time, is detected during the weft insertion several times by means of a second sensor. The speed of the loom drive is controlled on the basis of the detected loading variable and as a result of the detected quality feature of the weft yarn. The number of weft yarn breakages can be reduced with a suitable configuration of the control system.

In accordance with US-4,463,783 A, the thickness of a weft yarn to be inserted through a warp shed in a jet loom is measured with a yarn-thickness detector prior to insertion. The electrical signals fed from the yarn-thickness detector are averaged over the length of the weft yarn readied for insertion. The pressure of ejection of a fluid, e.g., air, is controlled based on the average thickness so as to be optimum for inserting the weft yarn. By such pressure control dependent on the thickness of the weft yarn to be inserted, weft insertion failures are prevented.

SUMMARY OF THE INVENTION It is an object of the present invention to provide a method and apparatus for controlling a jet loom, with which the energy and fluid consumption of the jet loom is reduced and the productivity of the jet loom is increased. These and other objects are achieved by the method and the jet loom in accordance with the invention, as defined in the independent claims. Advantageous embodiments are disclosed in the dependent claims. The invention proposes to dynamically control the jet loom based on at least two different individual, intrinsic characteristics within the weft-yarn section currently conveyed through the jet loom. For this purpose, at least two different intrinsic characteristics of each individual weft-yarn section are determined before its insertion into a fluid feed conduit of the jet loom. Thus the control of the jet loom is continuously optimized for each individual weft- yarn section. Experiments have shown that the above-identified objects are better achieved when not only one, but several yarn parameters are considered in the jet-loom control. This experimental finding is confirmed by theory, according to which the conveyance of the weft yarn by the fluid flow through the fluid feed conduit depends on several intrinsic yarn parameters. The consideration of several intrinsic yarn characteristics thus allows a better adaptation of the jet-loom control to the individual characteristics of the weft-yarn section to be inserted.

An "individual" characteristic of a yarn section is understood to be a characteristic specific to the corresponding yarn section, which section is a small part (e.g., 1 m to 5 m) of an entirety of a yarn (e.g., 50 km to 200 km). The individual characteristic generally differs from the characteristic of the entirety of the yarn and from the characteristic of other yarn sections. Since the individual characteristics change with a position along the axis of the yarn, they may also be called "dynamic" characteristics. Thus, an individual characteristic of a yarn section has to be separately and specifically determined for the corresponding yarn section. A characteristic related to the entirety of the yarn, such as the kind (staple or filament yarn) or the count (average linear mass density, measured in denier) is not an individual characteristic of any section of said yarn, but can rather be called a "global" or "static" characteristic. An "intrinsic" yarn characteristic is understood to be such a yarn characteristic which is given by the build-up or the structure of the yarn itself. Examples of intrinsic yarn characteristics are the mass per length unit, diameter, density, surface structure, hairiness, material composition, presence of foreign matter, etc. In contrast, an extrinsic characteristic would be such a yarn characteristic which depends on the external influences or reference systems. Examples of extrinsic yarn characteristics are a yarn velocity or acceleration relative to a certain reference system, a position of a yarn tip in a certain reference system, or a mechanical yarn tension dependent on an external tensile force.

In the method for controlling a jet loom according to the invention, which jet loom contains a plurality of relay nozzles arranged along a fluid feed conduit, a weft yarn is introduced into the fluid feed conduit. The relay nozzles are actuated so as to eject time- staggered fluid pulses which produce a fluid flow in the fluid feed conduit. The weft yarn is conveyed by the fluid flow through the fluid feed conduit. At least two different individual, intrinsic characteristics of a weft yarn section to be introduced into the fluid feed conduit are determined. The relay nozzles are actuated based on the previously determined at least two different individual, intrinsic characteristics of the respectively conveyed weft yarn section.

In a preferred embodiment, values of at least two different intrinsic yarn parameter of the weft yarn are measured along the length of the weft yarn before the insertion into the fluid feed conduit. The measured intrinsic yarn parameter values are assigned to the respective locations on the weft yarn. The relay nozzles are actuated based on the intrinsic yarn parameter values associated with the respectively conveyed weft yam section.

At least one further nozzle for conveying the weft yam can be provided upstream with respect to the fluid feed conduit and is also actuated based on the previously determined at least two different individual, intrinsic characteristics of the respectively conveyed weft yam section. The relay nozzles are preferably combined into several groups of nozzles and all relay nozzles combined into a group of nozzles are respectively actuated together.

Preferably at least one control parameter is calculated in the actuating of the nozzles, which control parameter is chosen from the following group: initial time of a nozzle opening, end time of a nozzle opening, duration of a nozzle opening and fluid pressure in a nozzle. The at least two individual, intrinsic characteristics can be at least two different intrinsic yam parameters chosen from the following group: yam mass per length unit, yam diameter, yarn density, yarn-surface structure, yarn hairiness, yarn-material composition, presence of foreign matter in the yarn.

In the actuating of the nozzles the weft yarn section to be inserted into the fluid feed conduit is preferably modeled as consisting of an integer number of homogeneous subsections with equal lengths, each subsection being assigned a location on said yarn section and values of the at least two different intrinsic yarn parameters averaged over the respective subsection. The integer number lies for instance between 1 and 50 and preferably between 20 and 30.

The measurement of the at least two different intrinsic yarn parameters advantageously occurs upstream with respect to a weft storage unit, from which the weft yarn is inserted into the fluid feed conduit. The actuating of the nozzles preferably occurs in such a way that fluid consumption is minimized. Also the weaving process should not be interrupted and an arrival time of the weft yarn at the end of the fluid feed conduit should lie in a predetermined range. The nozzles can be actuated based on additional intrinsic parameters of the processed yarn and/or parameters of the jet loom.

The fluid is preferably air or water, and the jet loom is preferably an air-jet loom or a water-jet loom. It is particularly advantageous to apply the invention together with electronic shedding. Such a combination allows an even better adaptation of the weaving process to the individual characteristics of the weft-yarn section currently conveyed through the j et loom .

At least one of the at least two different individual, intrinsic characteristics can be used for assessing the quality of the weft yarn by assessing detected yarn faults according to predefined quality criteria. A weft yarn section of insufficient quality is then deflected from its regular path such that it is not inserted into the air feed conduit, and/or an alarm is given upon detection of such a weft yarn section. Thus, the quality of the fabric can be enhanced. The relay nozzles are preferably actuated based on a previously determined local distribution of the at least two different individual, intrinsic characteristics within the respectively conveyed weft yarn section. The invention also relates to an apparatus for controlling a jet loom which contains a plurality of relay nozzles arranged along a fluid feed conduit. The apparatus comprises a control unit for actuating the relay nozzles so as to eject time-staggered fluid pulses which generate a fluid flow in the fluid feed conduit, by means of which a weft yarn is conveyable through the fluid feed conduit. Upstream with respect to the fluid feed conduit at least two yarn sensors for determining at least two different intrinsic characteristics of the weft yarn are provided. The control unit is configured to actuate the relay nozzles based on the previously determined at least two different individual, intrinsic characteristics of the respectively conveyed weft yarn section. In a preferred embodiment the at least two yarn sensors are configured to measure values of at least two different intrinsic yarn parameters of the weft yarn along the length of the weft yarn. The control unit is configured to assign the measured intrinsic yarn parameter values to the respective locations on the weft yarn, and to actuate the relay nozzles based on the intrinsic yarn parameter values associated with the respectively conveyed weft yarn section.

At least one further nozzle is preferably provided upstream with respect to the fluid feed conduit for conveying the weft yarn and the control unit is configured to actuate the at least one further nozzle also based on the previously determined individual, intrinsic

characteristics of the respectively conveyed weft yarn section. The relay nozzles can be combined into several groups of nozzles and all relay nozzles combined into a group of nozzles are jointly actuatable. The at least two yarn sensors are preferably provided upstream with respect to a weft storage unit, from which the weft yarn is insertable into the fluid feed conduit. The control unit can be configured to actuate the nozzles based on additional parameters of the processed yarn and/or parameters of the jet loom. The at least two yarn sensors can be part of a yarn clearer. Such yarn clearers are well known from the prior art. They are usually mounted on spinning or winding machines for monitoring the yarn quality. In the preferred embodiment of an air-jet loom, each nozzle or group of nozzles can be associated with a control valve, e.g. an electromagnetic two-way valve, for the supply with compressed air, and the control valve is actuatable by the control unit.

BRIEF DESCRIPTION OF THE DRAWINGS

A preferred embodiment of the invention will be explained below by reference to the drawings.

Figure 1 schematically shows a jet loom with a control apparatus in accordance with the invention.

Figure 2 shows a block diagram of an embodiment of the control method in accordance with the invention.

Figure 3 schematically shows (a) a weft yarn section and (b, c) two models of it.

Figure 4 shows a diagram with the nozzle opening times for the control method in

accordance with the invention.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT Figure 1 schematically shows a jet loom 1 with a control apparatus in accordance with the invention. In this embodiment, the jet loom 1 is designed as an air-jet loom. Yarn 91 destined as weft yarn is provided on a cone 21. The yarn 91 is transferred from the cone 21 to a weft storage unit 22 which can be designed as a drum storage unit. Weft yarn 92 that is drawn off from the weft storage unit 22 is accelerated by means of one or several acceleration nozzles 31 , 32 (also called main nozzles) and supplied to an air feed conduit 8 which is disposed in a shed formed by means of a shed forming apparatus (not shown). The specific weft yarn section 93 inserted into the air feed conduit 8 along its longitudinal direction x is called pick 93. The pick 93 is conveyed by a plurality of relay nozzles 33 through the air feed conduit 8. The relay nozzles 33 are preferably grouped into groups 34.1, ... , 34.n of nozzles. In the embodiment of Figure 1 , there are n = 6 groups 34.1, ... , 34.6 of nozzles with four relay nozzles 33 each. The acceleration nozzles 31 , 32 and the groups 34.1 , 34.n of nozzles are respectively supplied with compressed air via a control valve 41 , 42, 44.1 , ... , 44.n such as an electromagnetic two-way valve. The supply of compressed air to the control valves 41 , 42, 44.1 , ... , 44.n is not shown in Figure 1 for the sake of simplicity, as also further elements of the jet loom 1.

A sensor unit 5 for the continuous detection of at least two different intrinsic yarn parameter of the weft yarn 92 is arranged upstream with respect to the air feed conduit 8, and preferably between the cone 21 and the weft storage unit 22. The sensor unit 5 can be designed substantially as a yarn clearer, as has been used until now for online monitoring of the yarn quality on spinning or winding machines and as is known from the state of the art. It comprises at least two yarn sensors 51 , 52. The yarn sensors 51 , 52 preferably work according to different measurement principles. A first yarn sensor 51 can, for instance, be a capacitive sensor, whereas a second yarn sensor 52 can be an optical sensor. Other measurement principles such as the triboelectric principle are possible as well. At least two different intrinsic yarn parameters are measured by the sensors 51, 52. The intrinsic yarn parameters can be the yarn mass per length unit, yarn diameter, yarn density, yarn-surface structure, yarn hairiness, yarn-material composition, presence of foreign matter in the yarn, or any other intrinsic yarn parameter. A first weft break stop motion 61 for detecting the take-off of a pick 93 is disposed downstream with respect to the weft storage unit 22. A second weft break stop motion 62 for detecting the arrival of the pick 93 is further arranged at the exit of the air feed conduit 8.

The control valves 41, 42, 44.1, 44.n are controlled by a control unit 7. The control unit 7 can be an independent module or a part of a loom control unit. It receives signals and/or data from the sensor unit 5, from the first 61 and/or second weft break stop motion 62 and possibly from further sensors and/or from the jet loom 1 itself. The control unit 7 calculates the optimal nozzle opening times for every single weft insertion on the basis of the yarn parameters detected by the yarn sensor 7. The air effectiveness of the pick 93 can be included in the calculation. The air effectiveness of a yarn is defined in WO-89/12122 Al as the ratio of the carrying surface of the yarn in air in comparison to the yarn mass. The block diagram of Figure 2 illustrates an embodiment of the control method in accordance with the invention. The control unit 7, which is indicated here with a rectangle shown with a dashed line, can serve two control loops. A first control loop with a first controller 71 corresponds to the loom control loops known from the state of the art, whereas a second control loop with a second controller 72 is essential for the method according to the invention. The representation in Figure 2, showing two separate blocks for the two controllers 71, 72, was chosen for didactic reasons. In practice, the two controllers 71, 72 may be physically separated, but alternatively may be incorporated in one single unit.

A first, closed control loop is slow or quasi-static in comparison with the rotational period of the jet loom 1. It controls the speed of the loom drive. Input parameters of the first control loop are on the one hand the starting values 201 for global or static parameters 202 of the used weft yarn 91 such as a mean yarn count, a mean yarn diameter, a mean yarn density, statistically determined values for yarn unevenness with regard to yarn mass or yarn diameter and/or a distribution of yarn hairiness length. The starting values 201 can originate for example from statistical examinations that were performed on the respective yarn 91 beforehand in a textile laboratory by means of a yarn testing apparatus, e.g. of type USTER ^® TESTER 5 of the applicant, or simply received from the yarn supplier. On the other hand, further input parameters are loom data 203. The actual control variable in the first control loop is the weft arrival time, as detected by the second weft break stop motion 62. The first controller 71 calculates from the above-mentioned and other input parameters at least one control parameter 204 for the jet loom 1, as an actuating variable. Examples of control parameters 204 are a rotational speed of the loom drive and/or an air pressure supplied to the nozzles 31 -33. The air pressure influences the weft arrival time and/or the air consumption 205.

A second, open control loop is fast, i.e. its reaction time is of the magnitude of the rotation period of the jet loom 1 or less than that. Its input parameters are the individual or dynamic yarn parameters 206 of the pick 93 provided for the next weft insertion as determined by the sensor unit 5. The second controller 72 calculates by means of an algorithm from these current individual, dynamic yarn parameters 206 at least one control parameter 204 for the jet loom 1. Examples of control parameters 204 are individual opening times (see Figure 4) of the nozzles 31-33 for the next weft insertion. The second control loop aims at minimizing the air consumption 205, yet keeping the arrival time 205 in an acceptable, predetermined range. An individual weft insertion control is thus realized by the second control loop.

The control parameters 204 calculated by the first controller 71 and second controller 72 must be adjusted to one another and/or linked in a suitable manner with each other. The quasi-static control parameters 204 calculated by the first controller 71 are used as a basis for the calculation of the dynamic control parameters 204 of the second controller 72. If the first controller 71 and the second controller 72 calculate different control parameters 204, e.g. a nozzle air pressure or nozzle opening times, the control parameters 204 of the first controller 71 enter the algorithm running in the second controller 72 as predetermined parameters. If the first controller 71 and the second controller 72 calculate the same control parameters 204, the control parameters 204 of the second controller 72 are used as corrective values for the control parameters 204 of the first controller 71 , which are used as basic values.

The individual or dynamic yarn parameters 206 determined by the sensor unit 5 can optionally also influence the starting values 201 of the weft yarn 91 which are used as input parameters for the first control loop. This has no effects on the next weft insertion because the first control loop reacts slowly. It rather concerns a correction of the starting values 201 which were previously entered in the control unit 7; such a correction has a later and long-term effect. Synchronization 207 between the sensor unit 5 and the jet loom 1 is necessary for satisfactory functioning of the second control loop. In particular, the synchronization 207 is needed for assigning yarn parameter values 206 measured by the sensor unit 5 to the respective locations on the pick 93. As is shown in Figure 1 , the sensor unit 5 is preferably arranged upstream with respect to the weft storage unit 22. This leads to the advantage that the second controller 72 has sufficient time in order to calculate the control parameters 204 before the respective pick 93 is inserted. The control parameters 204 thus calculated in advance are stored in a storage unit of the second controller 72 and retrieved from there when the associated pick 93 is ready for insertion. Yarn 92 is stored in the weft storage unit 22 with a length which corresponds to several, e.g., four, picks 93. The synchronization 207 is used for associating the respective dynamic yarn parameters 206 determined by the sensor unit 5 with each location along the stored yarn 92. Signals from the weft break stop motions 61 , 62 or from the weft storage unit 22 can be used for synchronization 207.

Figure 3 relates to a modeling of the pick 93 for the reason of simplification of calculations in the second controller 72. Figure 3(a) shows a longitudinal section through the pick 93. The longitudinal direction of the pick 93 is designated with x, the radial direction with r. The length of the pick 93 in the longitudinal direction x is L. The yarn sensors 51 , 52 scan the yarn parameters 206 with a high local resolution along the longitudinal direction x. The pick 93 typically has a position-dependent mass that can be measured as a yarn parameter 206 by a first yarn sensor 51 of the sensor unit 5, and a position-dependent diameter D(x) that can be measured as a yarn parameter 206 by a second yarn sensor 52 of the sensor unit 5. In addition to the mass and diameter D(x), one or several other yarn parameters 206 such as the yarn density or hairiness can be measured by the sensor unit 5, and used for the model described in the following.

The calculation of the control parameters 204 in the second controller 72 can consider the location-dependent position of the yarn parameters 206 on the pick 93. It can be performed with the measured signal D(x). However, such a calculation involves a large amount of data. It can be simplified, without any noteworthy drawbacks, by modeling the pick 93 as shown in Figure 3(b). For this purpose, the pick 93 is virtually divided into an integer number m > 1 of idealized subsections 94.1 , 94.m with equal lengths L/m. The integer m may lie between 2 and 50 and preferably between 20 and 30; the simple example of Figure 3(b) uses m = 10. Each subsection 94. i (i = 1 , 2, m) is assigned a location Xj on the pick 93, wherein for instance

*, = (i - l)- ·

m Moreover, each subsection 94.i (i = 1 , 2, ... , m) is assigned a homogeneous diameter Dj, which is preferably an average diameter of this subsection 94.i: τη r

D, =— j D(x)dx

The averaging or sampling can be done by a processing unit associated with the sensor unit 5 or by the second controller 72. For this purpose, the sensor unit 5 or the controller 72 receive appropriate synchronization signals 207 from the weft break stop motions 61, 62, from the weft storage unit 22 or from another component. This model reduces the amount of data and thus simplifies the calculations performed by the second controller 72.

The special case of the model with m = 1 is shown in Figure 3(c). Here the whole pick 93 is assumed to be homogeneous with an average diameter Do given by It is to be noted that even this simplest special case is consistent with the basic idea of the invention, which is the dynamic control of the jet loom 1 based on at least two different measured parameters 206 of an individual pick 93. However, a model with a plurality of subsections 94.1, 94.m, i.e., with m > 1, is preferred, since the consideration of the local distribution of the yarn parameters 206 allows a better adaptation to the individual characteristics of the pick 93.

The diagram of Figure 4 schematically shows the nozzle opening times for the nozzles 31 - 33 of the jet loom 1 (see Figure 1). The angle of rotation of the jet loom 1, which substantially corresponds to the time, is drawn on the horizontal axis, and the various nozzles 31 , 32 and groups 44.1 , ... , 44.n of nozzles are drawn in the vertical direction. The horizontal bars 431 , 432, 434.1 , ... , 434.n indicate when the respective nozzle 31 , 32 or group 34.1, 34.n of nozzles is active. In this example, n = 6 has been chosen, though without loss of generality. Some bars have white and black parts. The white parts symbolize the conditions in a jet loom according to the state of the art, whereas the black parts relate to a jet loom 1 according to the present invention. The bars that are filled completely in black indicate that the nozzle opening times have not changed in relation to the state of the art.

Two control parameters 204 were changed in comparison with the state of the art in the embodiment of Figure 4, which are the starting time of the nozzle opening and the duration of the nozzle opening (or the time of the nozzle closing). At least one of these control parameters 204 is influenced by the parameters 206 of the respective pick 93 which were determined by the sensor unit 5. The state-of the-art conditions can be used as initial values and are then optimized according to the present invention. An air-effective pick 93, i.e. a pick 93 of large diameter, many long hairs and small mass, requires shorter nozzle opening periods than a pick 93 with lower air effectiveness.

Efforts are made by the control method in accordance with the invention to optimize the nozzle opening times 204 to achieve the lowest possible air consumption 205. An important boundary condition that needs to be observed in the optimization is that every pick 93 arrives and the weaving process is not interrupted. Moreover, the weft yarn arrival time 205 as measured by the second weft break stop motion 62 shall lie within a predetermined range. On the one hand, the nozzle opening times 204 can be varied for optimization. The control unit 7 calculates the expected arrival time and compares it with the measured arrival time 205. When the expected arrival time is shorter than the measured one, the nozzle opening times can be reduced in order to reduce air consumption 205. In the opposite case, the nozzle opening times are slightly extended in order to prevent the likelihood of a standstill of the loom 1. On the other hand, the initial nozzle opening times can also be varied because they influence the weft insertion. Optimal initial nozzle opening times allow even shorter nozzle opening times and thus lower air consumption 205. As is illustrated in Figure 4, the air consumption can be reduced relevantly by the method in accordance with the invention.

The yarn parameters measured by the sensor unit 5 can be used not only for actuating the nozzles 31-33, but also for assessing the quality of the weft yarn 91 , in a similar way as a yarn clearer on a spinning or winding machine. Detected yarn faults such as thick places, thin places or foreign matter in the weft yarn are assessed according to certain quality criteria, which can be defined by means of a so-called clearing limit. When the sensor unit 5 detects an intolerable weft-yarn fault, the faulty pick 93 can be deflected from its regular path such that it is not inweaved into the fabric. In looms in which at least two weft-yarn cones supply a fluid feed conduit, the weft- yarn supply from the cone from which the faulty yarn comes can be stopped. Alternatively or additionally, an alarm can be given if the detected weft-yam quality is insufficient.

It is understood that the present invention is not limited to the embodiments as discussed above. The person skilled in the art will be able to derive further variants with knowledge of the invention which also belong to the subject matter of the present invention.

LIST OF REFERENCE NUMERALS

1 Jet loom

21 Cone

22 Weft storage unit

31, 32 Acceleration/main nozzles

33 Relay nozzles

34.1, .. ., 34.n Groups of nozzles

41, 42 Control valves

44.1, .. ., 44.n Control valves

5 Sensor unit

51 , 52 Yarn sensors

61, 62 Weft break stop motions

7 Control unit

71, 72 Controllers

8 Air feed conduit

91, 92 Weft yarn

93 Pick

94.1, .. ., 94.m Subsections of the pick

201 Starting values

202 Global or static yarn parameters

203 Other loom data

204 Loom control parameters

205 Arrival time, air consumption

206 Individual or dynamic yarn parameters

207 Synchronization

431 , 432 Bars indicating nozzle activity

434.1 , ... , 434.n Bars indicating nozzle activity