Title of Invention	A METHOD FOR THE CHARACTERIZATION OF FANCY YARN
Abstract	Abstract According to the method for characterizing fancy yam, at least one characteristic of the fancy yam is scanned along the longitudinal direction of the fancy yam. Values of the scanning are evaluated and the results of the evaluation are outputted. The results of the evaluation are the fancy yam parameters such as base yam mass, base yam diameter, slub distance, mass increase (AM) of a slub, slub diameter increase, slub diameter, slub length (Le) and/or slub total mass. At least one result of the evaluation is outputted in the form of a graphic representation, for example as a scatter diagram. Such a graphic representation provides a clear representation of the measurement results. The occurring data is reduced and may be divided up into different categories.

Title of Invention

A METHOD FOR THE CHARACTERIZATION OF FANCY YARN

Abstract

Abstract According to the method for characterizing fancy yam, at least one characteristic of the fancy yam is scanned along the longitudinal direction of the fancy yam. Values of the scanning are evaluated and the results of the evaluation are outputted. The results of the evaluation are the fancy yam parameters such as base yam mass, base yam diameter, slub distance, mass increase (AM) of a slub, slub diameter increase, slub diameter, slub length (Le) and/or slub total mass. At least one result of the evaluation is outputted in the form of a graphic representation, for example as a scatter diagram. Such a graphic representation provides a clear representation of the measurement results. The occurring data is reduced and may be divided up into different categories.

Full Text	A METHOD FOR THE CHARACTERIZATION OF FANCY YARN FIELD OF THE INVENTION The invention relates to a method for the characterization of fancy yam according to the preamble of the first patent claim. It may be applied in the textile laboratory (offline) as well as in textile production (online), e.g. in spinning works or winding works. STATE OF THE ART Fancy yam is yam whose structure or fiber composition differs from the normal smooth yam. It is applied in weaving mill products and knitting mill products as an enriching element. Fancy yam usually has a multitude of thick places or thin places - so-called slubs - whose diameter is significantly larger or smaller than the diameters of the yam sections lying between the slubs - the so-called base yam. Stmctured yam with deliberately produced thickness variations with which no base yam may be identified is however also counted among the fancy yams. The increasing popularity of fancy yam demands reliable and meaningful methods for its characterization. Variables such as base yam diameter, diameter increase at a slub, slub mass, slub length, slub distance, etc. are of particular interest. These variables may e.g. be used for the control of the quality of the present fancy yam or for determining the manufacturing parameters which are necessary for copying a given fancy yam. Methods and devices for characterization of yam are known. They are usually based on a capacitive and/or optical scanning of the yam moved in the longitudinal direction. The capacitive scanning principle provides a signal corresponding to the yam mass, whilst the optical scanning principle provides a signal proportional to the yam diameter. The scanning signal is evaluated in an analog or digital manner, and one or several results of the evaluation are outputted. Examples of such methods and devices for the characterization of yam are specified in the patent publications EP-0'578'975 Al or EP-0'249741 A2. Both relate to the yam testing system USTER®TESTER which is marketed worldwide by the proprietor of the present protective property right. It is also known to obtain information on the color, i.e. on the spectral reflection characteristics of yam. Thus e.g. according to WO-2004/044579, one applies a multi-colored hght source for illuminating the yam. The light reflected by the yam is detected separately in at least two different spectral ranges. The at least two detection signals permit information with f regard to the yam cx)lor. A simultaneous optical scanning of a textile material at different wavelengths is also known from CH 674379 or from DE-198'59'274. WO-2005/07 1150 Al, WO-2005/038'105 Al and WO-2005/037699 Al especially deal with fancy yam. The teaching of the latter document may be summarized as follows: determining the base yam diameter: Firstly, the arithmetic mean of the yam diameter is formed over a large yam length. This mean is subtracted from the individual values of the yam diameter. The arithmetic means of all negative values which have been measured adjacentiy to the other negative values is defined as the base yam diameter. determining the beginning, end and length of a slub: A slub begiiming is present if a limit diameter which lies above the base yam diameter is overshot, and the overshoot persists over a certain yam length. A slub end is present if the limit diameter is undershot, and the undershoot persists over a certain yam length. The slub length is defined as a distance between the slub begiiming and the slub end. determining the slub diameter. A plurality of the largest diameters is determined within a slub . The slub diameter is defined as the mean of these largest diameters. Although this teaching permits a useful characterization of fancy yam, it also has a few disadvantages. A large quantity of data occurs as a result, which is unclear and difficult to handle. Suitable possibilities for the clear representation of the readings are not specified. The described method, if anything, results in a base yam diameter which is too large and in a slub length which is too small. DESCRIPTION OF THE INVENTION It is an object of the present invention to specify a method for the characterization of fancy yam which provides a clear representation of the measurement results. The occurring data should be reduced compared to the teaching of WO-2005/037688 Al. Furthermore, the invention should permit the occurring data to be divided up into certain categories in a suitable manner, i.e. into data which concern the slubs, and data which concem imperfections of the yam. These and other objects are achieved by the method defined in patent claim 1. Advantageous embodiments are specified in the dependent patent claims. According to the method according to the invention, for the characterization of fancy yam which preferably has a sequence of slubs and base yam, at least one characteristic of the fancy yam is scaimed along the longitudmal direction of the fancy yam. Values of the scanning are evaluated and results of the evaluation are outputted. Results of the evaluation are outputted: it is preferably the case of fancy yam parameters such as base yam mass, base yam diameter, slub distance, mass increase of a slub, slub diameter increase, slub diameter, slub length and/or slub mass. At least one result of the evaluation is outputted in the form of a graphic representation. Such a graphic representation reduces the occurring data quantity. It permits a rapid visual perception of the results. By way of this, it is also possible to divide up the data into different categories. In a preferred embodiment, the graphic data is a scatter diagram, and the at least one scanned characteristic is a mass and/or a diameter of the fancy yam. LISTING OF THE DRAWINGS The invention is hereinafter described in more detail by way of the drawings. Figure 1 schematically shows a device for carrying out the method according to the invention. Figure 2 shows one example of a series of readings with regard to a fancy yam, specifically yam mass against length coordinate. Figure 3 shows a frequency distribution of the measured yam mass plotted in Figure 2. Figure 4 shows an enlarged section of Figure 2. Figures 5-15 show possible representation types according to the invention, for outputting the yam parameters. Figure 16 shows an example of a series of reading with regard to a two-step fancy yam, in the same representation as Fig.2. Figure 17 shows one possible representation type according to the invention, for outputting the characteristics of a multi-step yam. Figure 18 shows a spectrogram of a series of readings with regard to a fancy yam. Figure 19 shows a meastired and an idealized slub . Figure 20 shows a measurement curve with a thin place, analogous to Figure 4. Figure 21 shows the course of the area below a measurement curve, as a function of the length coordinate. Figure 22 shows the mass per length unit as a function of the length coordinate, (a) for the measurement curve, (b) for an idealized curve and (c) for a curve which arises by subtraction of the curve (b) from the curve (a). Figure 23 shows spectrograms (a) of the curve of Fig. 22(a), (b) of the curve of Fig. 22(b), and (c) of the curve of Fig. 22(c). Figure 24 show scatter diagrams (a) of the readings of Fig. 22(a), (b) of the slubs of Fig. 22(b), and (c) of the virtual base yam of Fig. 22(c). IMPLEMENTATION OF THE INVENTION A device 1 for carrying out the method according to the invention is shown schematically in Figure 1. It contains a scanning unit 2 for scanning a fancy yam 9 with base yam 91 and slubs 92, 92', which is moved in a longitudinal direction -x. Here, the sequential recording of a multitude of readings at different, preferably equidistant locations of the fancy yam 9 is to be understood under the term "scanning". Such scanning imits 2 are knovra per se, and do not need to be explained in more detail here. The scanning unit 2 may contain a capacitive, optical or other sensor; also several equal or different sensors may be arranged within the scanning unit 2. The scanning unit 2 may be provided with evaluation means for a preliminary evaluation of the readings. It outputs a preferably electrical output signal which is a measure for the mass, the thickness or other characteristics of the fancy yam 9, on a first data lead 23. The first data lead 23 runs into an evaluation unit 3 which is suitable for evaluating the output signal of the scanning unit 2. For this purpose, it contains suitable analog and/or digital evaluation means, e.g. a microprocessor. It may also contain further means such as memory means for storing data. The evaluation unit 13 is preferably a computer. Furthermore, the device 1 contains an output unit 4 for outputting measurement data and/or results of the evaluation. The output unit 4 is coimected to the evaluation unit 3 by way of a second data lead 34. It may e.g. be designed as a monitor and/or printer. The device 1 preferably also contains an input unit 5 for inputting data on the part of the user. The input unit 5 may e.g. be a keyboard, a mouse and/or a touch screen. Figure 2 shows one possible output signal 100 of the scanning unit 2. Thereby, a variable for example which is a measure for the mass M per unit length of the fancy yam 9 (see Figure 1), is plotted along the length coordinate x of the fancy yam 9. Such a measure signal M is typically provided by a capacitive yam sensor. A representation of the thickness of the fancy yam 9 along the length coordinate x would appear in a similar manner, wherein the scales on the abscissa and the ordinate as well as the resolution could be different; a thickness signal is typically provided by an optical yam sensor. The curve M(x) is not necessarily continuous, but may be composed of individual (not shown in Fig. 2) scanning points which on the fancy yam 9 are typically distanced from one another by a few millimeters; the distance depends on the scanning rate of the scanning xmit 2 and on the speed of the fancy yam 9. The signal M(x) has a noise floor 101 which corresponds to the base yam mass Ms or the base yam diameter. The peaks 102 corresponding to the slubs project significantly from the noise floor 101. A mean Mm of all readings M(x) over a large yam length, on account of the peaks 102, lies significantly above the noise floor 101 and is therefore not suitable as a measure for the base yam mass Ms. At this location, it should be noted that Figure 2 and the subsequent discussion only represents one possible, non-limiting example. With other types of fancy yam, it is no longer even possible to identify base yam. The base yam mass Ms may be determined according to the present invention preferably as follows. A frequency distribution H(M) of the measured masses M represented in Figure 2 is determined. Such a frequency distribution H(M) is schematically shown in Figure 3. The frequency distribution H(M) must be determined for a large yam length, under which here a yam length is to be understood as one which contains may slubs, for example at least 10 and preferably at least 100. The frequency distribution H(M) with an fancy yam 9 comprises at least two local maxima 121, 122: a first local maximum 121 for the base yam 91, and at least one second local maximum 122 for the slubs 92, 92', .... According to definition, amongst all local maxrnia 121, 122, it is the local maximum 121 belonging to the base yam 91 which has die smallest mass M. For this reason, the smallest mass M at which a local maximvmi 121 occurs in the frequency distribution H(M) is defined as the base yam mass Ms. The yam number of the yam body or of the base yam 91 may be computed from the base yam mass Ms. One may proceed in an analogous manner in order to determine the base yam diameter. According to an alternative embodiment of the method according to the invention for determining the base yam mss, one defines a mass interval Im (see Figure 3) which contains that smallest mass at which a local maximum 121 occurs in the frequency distribution H(M). The mass interval Im may, but need not be symmetrical with respect to the "smallest" mass, i.e. the "smallest" mass may, but need not he in the middle of the mass interval Im- The upper limit of the mass interval Im is preferably selected below the global mass mean Mm- The width and the position of the mass interval Im may be predefined in a fixed manner - for example ±5% of the "smallest" mass - automatically computed by the evaluation unit 3, or be inputted by a user. A mean is subsequently formed over all masses M measured in this mass interval Im- The base yam mass is defined as this mean. Figure 4 shows a more detailed view of a section of the measurement curve M(x) of Fig. 2. Hereinafter, it is explained by way of Figure 4 how further fancy yam parameters such as slub length Le, slub distance Ls, mass increase AM and slub total mass Me are evaluated. For ascertaining a beginning 103 and end 104 of a slub 92, one previously determines a threshold value Mt which is larger than the base yam mass Ms. The threshold value Mr may be predefined in a fixed mannei- - for example 110 % of the base yam mass Ms, automatically computed by the evaluation unit 3 or inputted by a user. The beginning 103 of a slub 92 is present if, proceeding fixim the noise floor 101, the threshold value Mj is overshot. In order to exclude artifacts due to outliers, one may additionally control as to whether the overshoot persists over a predefined yam length, i.e. whether a few further measurement points, which directly follow the measurement point exceeding the threshold value Mt, likewise he above the threshold value Mt. In order to determine the beginning 103 of the slub 92 in a more accurate manner, one goes so far back on the measurement curve M(x) until, for the first time, a reading is smaller or equal to the base yam mass Ms - or another predefined or computed value. This reading is defined as the beginning 103 of the slub 92. The end 104 of the slub 92 is ascertained mutatis mutandis in an analogous method: on xmdershooting the threshold value Mt proceeding fi-om the signal peak 102, one moves so far forwards on the measurement curve M(x) until, for the first time, a reading is smaller or equal to the base yam mass Ms. If required, one may use different threshold values for determining the slub beginning 103 and the slub end 104. The slub length Le according to the present invention is defined as the distance between the beginning 103 and the end 104 of the slub 92. The slub distance Ls is defined as the distance between the end 104 of a slub and the beginning 103' of a subsequent slub 92' to which a subsequent signal peak 102' belongs. The distance between two adjacent slubs 92, 92' is defined as the sum Le+Ls of slub length and slub distance. Typical slub lengths Le and slub distances Ls He in the range between 2 cm and a few meters. The mass increase AM which corresponds roughly to a diameter increase of the fancy yam 9 is defined in the simplest case as the difference between a local maximum 105 of the corresponding signal peak 102 and the base yam mass Ms. Refined methods are possible for determining the mass increase AM, which take account of the fluctuations of the scanning signal M(x) along the slub length. Thus for example - analogously to the evaluation of the base yam mass Ms described above - the most common value within the corresponding slub length may be selected. A mean fomiation of values on the slub ridge is also considered. The mass increase AM is preferably specified as a multiple of the base yam mass Ms, e.g. in percentage values, wherein the base yam mass Ms is preferably defined as 100 %. Typical mass increases AM lie in a range between 20 % and 500 %. A further parameter for characterizing a slub 92 is the so-called slub total mass Me. This is essentially the difference between (i) the integral of the measurement curve M(x) over the slub length Le and (ii) the mass Ms-Le of the yam body on this slub length Le. The slub total mass Me may be determined by calculation by the evaluation imit 3. The yam number of the slub 92 may be computed fi-om the fancy yam mass Me, wherein the computation may contain a division of the slub total mass Me by the slub length Le. The shape of a slub 92 may also be determined and outputted. Thereby, one may fall back on a comparison with simple geometric shapes such as a bar, triangle, step, trapezium, or bell; cf Fig. 18. The respective shape may e.g. be outputted on the output unit 4. Not only are "local" parameters of the individual slubs of interest, but also "global" parameters of a whole yam section, of a yam or of a group of several yams. Such a global fancy yam parameter is the average yam number (average yam mass), which may be computed for example by way of mean formation over all readings. The yam number of all base yam 91 may also be of interest. A further global fancy yam parameter is the average spatial frequency of the slubs, i.e. the average number of slubs per length unit. The fancy yam parameters mentioned above, and possible further ones are preferably determined dynamically during a running time of the scanning. The slub parameters are stored for the purpose of outputting. It is advantageous to additionally store a continuous running number allocated to the respective slub, in order not only to be able to provide information on the individual slubs, but also on their sequence. It may be advantageous to equip the scanning unit 2 (Fig. 1) with a capacitive as well as an optical yam sensor which may simultaneously measure one and the same yam. The output signals of the capacitive and of the optical sensor may be linked to one another in a suitable manner for the purpose of an improved or more accurate evaluation. Capacitive measurement has the advantage that it provides a signal with a good signal-to-noise ratio. The signal however is proportional to the mass per length unit, and thus does not correspond to the visual impression of the yam. This has a disadvantageous effect indeed with fancy yam which in the region of a slub 92 often has a different yam density than in the region of a base yam 91. Optical measurement has the advantage of better representing the visual impression of yam, because it measures essentially the visible yam diameter and it is therefore better suitable to fabric simulations. For this, the optical measurement signal has a greater noise than the capacitive measurement signal. It is possible by way of suitable linking of the two output signals, to profit from the advantages of both measurement types and to eliminate or at least weaken their respective disadvantages. Results of the evaluation may on the one hand be variables such as e.g. minima, maxima, arithmetic means and/or standard deviations of the above defined fancy yam parameters. The number of slubs 92 per yam length may be a further variable of interest. These variables may be outputted as alphanumeric signs. On the other hand, the results of the evaluation may be graphically represented and outputted on the output unit 4 in a suitable manner. Preferred presentation types are shown in the Figures 5-11. One possible representation type is a histogram, i.e. the graphic representation of the class frequencies of the classed yam parameters. Three examples in the form of bar charts are specified in Figure 5. The ordinate in each case is the frequency H. In Figure 5(a) the mass increase AM, in Figure 5(b) the slub length Le, and in Figure 5(C;) the slub distance Ls have been use as the abscissa. The abscissas may have a linear, logarithmic or other division. The division and/or scale of the axes may be automatically computed or may be selected by way of input on the part of an operating person. The same applies to the selection of the classification, i.e. to tiie width of the classes. Not all classes necessarily need to have the same with. The representation manner of Figure 6 is a so-called scatter diagram (aggregate of plots). In this, the mass increase AM is plotted against the slub length Le for all slubs 92 of an fancy yam 9 or a yam section. Each slub 92 is plotted as a point at the correct location. This representation simplifies the division of different slubs 92 into different classes. Thus for example, it is immediately evident from Figure 6(a), that the examined fancy yam 9 has three classes 111 -113 or populations of slubs 92: a first class 111 with short, low slubs a second class 112 with long, low slubs, and a third class 113 with long, high slubs. An outlier analysis is carried out by way of the scatter diagram of Fig. 6(a). For this purpose, one may provide a tool in order to define slub populations 111-113 and to the delineate them from one another as well as from outliers. Such a tool may e.g. permit part surfaces 111.1-113 .1 of the scatter diagram to be defined on a monitor with a mouse, as is represented in Figure 6(b). One slub population 111 -113 is allocated to each part svirface 111.1-113.1. The part surfaces may e.g. be shaped as a rectangle 111.1, as a circle 112.1 or a polygon 113.1. Points lying outside the part surfaces 111.1-113.1 are to be graded as outliers and with further evaluations and/or representations, may be characterized as such or not taken into account. If no part surfaces are defined, then all points of the scatter diagram are counted as slubs and are treated as such. Other scatter diagrams are possible, for example slub total mass Me against slub length Le, slub total mass Me against mass increase AM, etc. The scatter diagram may be issued in a colored manner, wherein different colors may indicate different measurements, different point densities and/or populations or outliers (cf Fig. 9). A three-dimensional representation is analogously possible, in which two coordinates correspond to those of Fig. 6, and the third coordinate corresponds to the point density; cf Fig. 8. The scatter diagram may be represented for an individual fancy yam sample or for several fancy yam samples. In the latter case, one may use different colors for the different fancy yam samples, in order to display possible differences between the samples. With several fancy yam samples, one may display the result of the entirety of all fancy yam samples additionally to the results of the individual fancy yam samples, preferably in a color which is individually allocated. Apart from the actual values, one may also represent nominal values or nominal regions for one or more classes of slubs on the scatter diagram. The nominal and actual values may also be compared in other representation types, or in a purely numeric manner. Such nominal-actual comparisons permit e.g. a control on the quality of a copy (actual value) or a predefined fancy yam (nominal value). The results may also be issued in the form of a table or a classification matrix - as shown in Figure 7, instead of a scatter diagram. The table axes correspond to the axes of the scatter diagram of Fig. 6. The numbers of respective slubs 92 are entered into the fields of the tables. Alternatively to the absolute number of slubs 92, one may also indicate their relative share; e.g. in percent or per thousand. Each table field thus represents a class of slubs 92, analogously to the classes 111-113 which are described on the occasion of Fig. 5. The selected size of the table fields is directed to the desired classification. With the selection of the size of the table fields, one should also take note that the resolution is sufficiently fine, but that the table still remains clear. In the example of Figure 7, relatively small table fields were selected, which permit a finer classification than the three classes 111-113 which were described on the occasion of Figure 6. The table fields may be filled with colors, patterns or steps of gray, which in turn are allocated to different numbers of slubs 92, for the purpose of an improved visualization. With the scatter diagram (Fig. 6) as well as the classification matrix (Fig. 7), the axes, independently of one another, may have a linear, a logarithmic or other division. The di\dsion and/or scale of the axes may be computed automatically, or may be selected by way of input on the part of the operating person. The same applies to the width and height of the table fields in the classification matrix of Figure 7, i.e. for the selection of the classification. Figure 8 shows a further preferred output possibility for the determined fancy yam parameters. It is the case of a table which is subdivided into five main columns for the five parameters of slub length Le, slub distance Ls, mass increase AM, diameter increase and number # of the slubs. The first four main columns for their part are subdivided in each case into three sub-columns for the minimum Min, the mean 0 and the maximvim Max of the respective parameter. The lines of the table may contain the values for the entire yam or for the entire examined yam section, as well as for the individual slub populations 111-113. The minima LF,n.m, AMmin, the means Le,0, AlvIo and the maxuna LE,max» AM^ax are indicated for the population 111 in Fig. 6(b). Of course, the table may also be formed in a different manner. In any case, such a tabular representation of the determined fancy yam parameters leads to a reduction of data. Particularities of the fancy yam concerned may be very quickly detected and different fancy yams may be easily compared by way of the table. A further manner of representation for the results of the evaluation of fancy yam parameters is shown in Figure 9. It is the case of a surface in three dimensions (3D). Two of the three dimensions, the two horizontal axes, correspond to the slub length Le and the mass increase AM, as in the scatter diagram of Fig. 6. The third, vertical dunension indicates the respective frequency H of the measured values, i.e. the point density in the scatter diagram of Fig. 6 or the numbers of Fig. 7. The 3D-surface which arises in this manner gives the impression of a mountain [range] which permits a rapid and memorable visual perception of the peculiarities of the respective fancy yam 9. It is particularly a synoptic comparison of two such "mountains" which very quickly shows whether the fancy yam concerned has similar or different characteristics, and in tiie latter case where the main differences he. It is to be noted that the example of Fig. 9 relates to a different fancy yam than the example of Fig. 6. Whilst the fancy yam of Fig. 6 has three classes 111-113 of slubs 92, the fancy yam of Fig. 9 only has two of them 114,115. The three-dimensional representation maimer of Fig. 9 may be reduced also to two dimensions. Figure 10 shows such a diagram which arises by way of the projection of the "mountain" of Fig. 9 into the plane spanned by the two horizontal axes Le, M. A "map" thus arises on which the "mountains" 114, 115 of Fig. 9 are repesented by way of "altitude contours", i.e. lines of the same fi-equency H. Instead of "altitude contours", one may apply colors or cross-hatchings for rendering the different fi^equencies H visible. The representation types of the Figures 5-10 do not take into account mformation on the sequence of the mdividual slubs. This information is completely present in the measurement series as is represented in Figure 2. It is possible from this to determine the respective yam parameter such as mass increase, slub length and slub distance as the measurement variables (number value x unit), and to list these measurement variables in the sequence of their occurrence, after one another. Such an alphanumeric listing of the reduced measurement data may be useful for certain cases, but is however less clear. Information on the sequence of the individual slubs is fully contained in the representation manner of Figure 11, wherein graphics have the advantage of an improved clarity and visual perception compared to an alphanumeric value table. A horizontal bar is drawn in for each slub 92 and an adjacent base yam 91. The bar is composed of two parts. The length of a first, left part indicates the respective slub length Le, the length of a second, right part indicates the respective slub distance Ls. The next bar lying therebelow characterizes the subsequent slub, etc. The horizontal bars of Fig. 11 may of course also be replaced by vertical columns. Measurement variables other than lengths Le, Ls may be plotted as bars, e.g. a slub mass and the associated base yam mass, which may be advantageous in particular with stmctured yam with different base yam masses. It is also possible for a bar or column to indicate more than two slub parameters, e.g. with multi-step slubs (see Fig. 16) the first slub length Le,i, the second slub length Le2 and the associated slub distance Ls. Figure 12 shows a manner of representation which at least represents a part of the information on the sequence of the individual slubs in a clear maimer. It is the case of a classification matrix which assumes a classification of the slubs as has been implemented with the classes 111-113, somewhat as in Figure 6. In each case, a pair of two adjacent slubs 92, 92' are considered, of which a first slub 92 is called a "leading slub" and a second slub 92' a "trailing slub". A corresponding entry into the classification matrix of Figure 12 whose horizontal axis indicates the leading slub 92 and whose vertical axis indicates the trailing slub 92' is effected for each pair 92, 92'. One may deduce fi-om the Active example of Figure 12, that in practice, two slubs of the first class 111 (cf. Fig. 6), and two slubs of the second class 112 are never successive, but that a slub of the first class 111 often follows a slub of the second class 112, and a slub of the third class 113 very often follows a slub of the first class 111. Various slubs of a fancy yam 9 may have different colors. For this reason, it may be desirable to obtain information on the color of the fancy yam 9. Sviitable scanning units 2 and evaluation units 3 (see Fig. 1) fi-om the state of the art mentioned above are known for this. One possible representation manner for the obtained color information is shown in Figure 13. Here it is the case of a circular chart which indicates the measured shares of the differently colored slubs. In the example of Figure 13, the fancy yam contains red (R) and blue (B) slubs which occurred with a fi-equency of 45 % and 55 % respectively. An enhancement to more than two colors is of course also possible. In the case that the output unit 4 (see Fig. 1) permits a colored The circular chart of Figure 13 contains no information on the sequence of the color slubs. This information is at least partly present in the representation manner of Figure 14. Analogously to Figure 12, in each case two successive slubs are considered, and the frequency of their colors R, B is plotted in the table, wherein the horizontal table axis indicates the color R, B of the leading slub, and the vertical axis indicates the color R, B of the trailing slub. One may deduce from the table of Figure 14 that a color change often occurs in this Active example, whereas two adjacent slubs having the same color is rather rare. The three-dimensional column charts of Figure 15, by way of Active examples, show how the color information may be combined with the geometric information in a single representation manner. Thereby, the geometric information hes in the plane of the drawing and corresponds to that of Fig. 5(a) and Fig. 5(b) respectively. The frequency of classes of the mass increase AM is plotted in the diagram of Figure 15(a), and the frequency of classes of the slub length Le is plotted in the diagram of Figure 15(b). The third dimension is used for the color information R, B. One may deduce from the diagram of Figure 15(a) that the red slubs R tend to have smaller mass increases AM than the blue slubs B. The diagram of Figure 15(b) indicates that the red slubs R tend to be longer than the blue slubs B. The Figures 5-15 discussed above only indicate a few example for representing the fancy yam parameters of mass increase AM (or diameter mcrease), slub length Le, slub distance Ls and color. One may of course also graphically represent further relations between these and further fancy yam parameters in a similar and two-dimensional or three-dimensional manner. Whilst single-step slubs have been discussed up to now, multi-step slubs are considered hereinafter. One example of a series of readings on a two-step fancy yam is specified in Figure 16, in the same representation as Fig. 4. Here one may differentiate between a first slub step with a first slub total mass Me,i per length unit, a first mass increase AMi and a first slub length Le,i, and a second slub step with a section slub total mass Me2 per length unit, a second mass increase AMi and a second slub length Le,2- The mentioned parameters may be determined in a manner which is analogous to that described for a single-step fancy yam. One possible manner of representation for the parameters of multi-step fancy yam is shown in Figure 17. Here it is the case of a table whose horizontal axis corresponds to the classed mass increase AM; cf Fig. 5(a). The lines of the table represent the different steps of the fancy yam. The respective frequencies are plotted in the fields of the table. The fancy yam is two-step in the fictive example of Figure 17, wherein the second step occurs in two variants: with a relatively small mass increase AMa on the one hand, and with a relatively large mass increase AM2 on the other hand. An analogous manner of representation is also possible for the slub length Le,i, Le2. A further parameter of fancy yam is the so-called pattern length. This is the length of the shortest sequence of slubs which are periodically repeated in the fancy yam. There is no periodicity whatsoever within this sequence, i.e. at least one slub parameter such as e.g. the slub distance Ls is random or pseudo-random. The pattern length may be obtained e.g. by way of correlation computation from readings, as they are represented for a short yam section in Figure 2. Such a correlation computation with all readings may be extensive with regard to computation. In order to reduce the computational effort, the measurement data may be previously reduced in that e.g. the respective yam parameters such as mass increase, slub length and slub distance are determined and the correlation computation is based on this reduced data. In an analogous manner, one may also obtain information on the presence of - mostly undesired - sub-pattems and their lengths. The pattern length and/or sub-pattem lengths are preferably issued in alphanumeric or graphic fomi. A spectrogram of the measurement signal M(x) of Figure 2 may also provide useful information on the fancy yam. The measurement signal M(x) is preferably subjected to a Fourier transformation for determining the spectrogram. One fictive example of a spectrogram \|F {M} \| or more precisely of the real part of the Fourier transform is shown in Figure 18, wherein as is customary, a period length L is selected as the abscissa, preferably in a logarithmic scale. Usually, the spectrogram \|F{M}\| .displays a relatively broad distribution 131 of the spatial frequencies or of the period lengths L. One may deduce an average distance of the slubs from the position of the maximum 132. A pronounced peak in the specfrogram \|F{M}\| would indicate a -mostly undesired - periodicity in the fancy yam. The periodicity on the one hand may relate to the individual slubs. With a fancy yam with a constant slub distance Le + Ls, within which the slub length Le and the slub distance Ls vary, the maximum 132 appears as a pronounced peak. In order to ascertam this, a yam length of at least ten, preferably one hundred and more slub distances should be measured. The periodicity on the other hand may relate to the pattems. With a sufficiently long measurement series - at least ten, but preferably one hundred and more pattern lengths - one may also read out the pattem length from the position of a respective peak 133 in the long-waved region of the spectrogram \|F{M} \|. A further graphic representation manner for the slubs is a smoothed or idealized representation of the readings, as is shown in Figure 19. On the one hand, the readings of Figure 4 are drawn in a dotted manner. On the other hand, an idealized course of the measurement curve is indicated with an unbroken line. As the man skilled in the art knows, there are many possibilities of obtaining an idealized curve in the manner of Figure 19 from a real measurement curve. In the application example of Figure 19, base yam 91, 91' have been approximated by horizontal straight lines which all lie at the height of the previously determined base yam mass Ms. One slub 92 is idealized in each case as a trapezium with flanks 93, 94 and a horizontal roof 95. Thereby, the trapezium does not necessarily need to be symmetrical, i.e. the flanks 93, 95 with regard to magnitude may also have different gradients. The approximation of the slub 92 by a trapezium may be effected according to methods and criteria known to the man skilled in the art. The flanks 93, 94 may roughly be straight lines whose positions have been determined by way of the method of least squares, or in another maimer. The height of the trapezium, i.e. the position of the roof 95, may be determined according to the criterion that the area of the trapezium is equal to the area below the real measurement curve. The slub length Le may for example be defined as the base length of the trapezium, i.e. as the distance between a slub beginning 103* and a slub end 104, wherein the slub beginning 103 and the slub end 104* are the intersection points of the horizontal representing the base yam mass Ms and the left flank 93 or the right flank 94. Alternatively, the slub length Le may be may be defined as the width of the trapezium at half the height, which is equal to half the sum of the base length and the roof length. The mass increase AM may be defined at the height of the trapezium, i.e. as the distance between the base and the roof Under certain circumstances, the trapezium may degenerate into flie special case of a triangle (trapezium with a roof length equal to zero) or of a rectangle (trapezium all with right angles). Other shapes for the idealized measurement curve are likewise possible. Such idealized courses of curves permit an unproved visual perception of the characteristics of the fancy yam. Parameters of the idealized curve, such as e.g. height, base length, roof length, the slub length Le or the flank gradients of the trapezium may be outputted as characteristic variables in a table for example, and be used for further evaluations. The output of such variables leads to a reduction in the data. An incomplete manufacturing process for the fancy yam may lead to a thin place 106 being present directly next to a slub 102, as indicated in Figure 20. Such undesired thin places 106 may be detected according to the present invention. Their number or share with regard to quantity at the slubs 102 may be outputted as a result. A thin place share of 50 % means that a thin place 106 was observed next to half of all slubs 106, which indicates a deficient manufacturing process. Spinning works which manufacture fancy yam have the need to differentiate between the following two phenomena: on the one hand, the virtual base yam manufacture, which may introduce imperfections, irregularities and faults such as thick places or thin places into the yam, and on the other hand, the slub manufacture which incorporates the desired slubs onto the virtual base yam, e.g. in the form of thickenings. These two phenomena are sometimes impossible or difficult to differentiate with conventional yam testing methods and apparatus. The exemplary curve of Fig. 4 was selected for didactic reasons, so that it is quite clear from this, what a slub 102 is and what a base yam 101 is. In practice however, the undesired thickness fluctuations on the base yam 101 may be so large, that they exceed the threshold value Mr and as a result are wrongly considered to be a slub on evaluation. The results of the evaluation are therefore adulterated. The results of this may lead to the wrong measures being taken in the manufacturing process. If e.g. long thick places or yam mass fluctuations are assumed to be small slubs, the part process for slub manufacture is changed such that larger slubs may be produced. This measure unnecessarily changes the slub stmcture without alleviating the actual cause of the fault - perhaps a defect in one path. Here, a preferred embodiment solves this problem by way of specifying the definition of a slub. The method described on the occasion of Fig. 4 may be disadvantageous since on setting the threshold value Mn, it only takes into account a one-dimensional mass increase. As described, although one may attempt to reduce this disadvantage by way of the demand that the excess must last for a certain length. This however also does not lead to an optimal recognition of the slubs. The observation of an area below the measurement curve - or what is considered here as being equivalent to the area, a certain integral of the measurement curve, has been shown to be more advantageous. Such an area may be computed at least approximately with one of the known numeric integration methods, like with the rectangle or trapezium method. Figure 21 schematically shows a typical course of the area A(x) computed in this manner, as a function of the length coordinate x, wherein the previously determined base yam mass Ms (see Fig. 4) serves as basis for determining the area. The values of A(x) fluctuate around the value zero in the region of the base yam 101. Any thick places, even if they have a large mass increase, do not cause any significant area changes, since they only extend in each case over a short length interval. A real slub 92 differs from a thick place in that it causes a significant rise of the curve A(x) which sets in at the beginning of a slub 102. A local maximum 105 of the slub 92 which may also be described as the position of the slub 92, is located at the turning point of the curve A(x). After the slub end 104, the curve A(x) again fluctuates around a constant value, which indicates the slub total mass. If one ascertains such a slub end 104, then the further course of the curve may be set back to the value zero by way of subtraction of the value Me. A further base yam follows etc. All slubs and their positions may be recognized in a reliable and stable manner and be differentiated from thick places by way of this. The ascertaining of a slub is preferably made dependent on the simultaneously fulfilhnent of several criteria, e.g. of the following three criteria: (i) exceeding a predefined threshold value for the area A(x), (ii) exceeding a predefined threshold value for the slub length Le, and (iii) exceeding a predefined threshold value for the mass increase AM. Only when these criteria are simultaneously fulfilled may one reliably assume that a proper slub and not a yam imperfection is present. The differentiation between the virtual base yam and slubs is simplified even greater by way of a further embodiment of the method according to the invention. According to this embodiment, the idealized course of the curve according to Fig. 19 is subtracted fi-om the real measurement curve. This is schematically represented in Figure 22. In this, Figure 22(a) shows the curve of the original readings, Figure 22(b) the idealized curve (cf Fig. 19), and Figure 22(c) the curve which arises when one subtracts the idealized curve from the original measurement curve. The curve of Fig. 22(b) shows only the (idealized) slubs, the curve of Fig. 22(c) only the virtual base yam without slubs. These representations alone may contribute to a deeper understanding of the stmcture of the examined fancy yam. The data obtained in this manner may however be evaluated even, further, as is discussed hereinafter. Figure 23 schematically shows results of an evaluation of the data of Figure 22 in a spectrogram, i.e. in a representation which corresponds to that of Fig. 18. Figure 23(a) shows the spectrogram of the original readings, i.e. the curve of Fig. 21(a). As already indicated above, it is difficult or even impossible to differentiate between slubs and virtual base yam only by way of this spectrogram. The representations of the Figures 23(b) and (c) alleviate this problem. Figure 23(b) shows the spectrogram of the slubs alone, i.e. of the curve of Fig. 22(b). This data related to the slub permits malfunctioning in the slub production to be localized and overcome in a targeted manner, or permits the slub production to be changed in a targeted manner. Peaks in the long-waved region may e.g. indicate undesired periodicities, which may be avoided with suitable measures. Figure 23(c) shows the spectrogram of the virtual base yam, i.e. of the curve of Fig. 22(c). Any peaks in this spectrogram give good hints as to certain faults in the spinning process, or in the process stages preceding the spinning process, such as eccentricities of certain rollers in the drawing arrangement. These faults may be localized by way of the respective wavelengths in the spectrogram of Fig. 23(c), and subsequently dealt with. The data of Fig. 22 may just as easily be represented in a scatter diagram, analogously to Fig. 6. This form of representation may also be very useful in order to differentiate between slubs and the virtual base yam. A schematic example is specified in Figure 24. Figure 24(a) shows a scatter diagram of the original readings of the curve of Fig. 22(a). Here, three phenomena intermingle: the lesser disturbing thick places occurring in each yam, slubs and • thick places wrongly evaluated as slubs. Figure 24(b) shows a scatter diagram of the slubs on their own, as they are represented in Fig. 22(b). In particular, when the robust method for slub recognition described on the occasion of Fig. 21 is used, this scatter diagram will have no points which undesirably originate from imperfections such as thick places. Rather it only contains real • slubs. Figure 24(c) shows a scatter diagram of the virtual base yam of Fig. 22(c). The points drawn therein do not represent slubs, but imperfections such as thick places. The scatter diagram of Fig. 24(c) may provide information on the applied yam manufacturing process. Other representation types such as e.g. the histograms of Fig. 5 may be used separately in an analogous manner for the data related on the one hand to the slub, and for the data related on the other hand to the virtual base yam. It may be advantageous in the method according to the invention to filter the readings as are represented somewhat in Figure 2 or 4, according to certain filter criteria. Such filter criteria may for example be the following: yam imperfections such as neps, thick- and/or thin places. Thus for example, it is known from CH-678'173 A5 or from US-5,537,811 A, to arrange possible yam errors in a table in the manner of a coordinate system, for setting the clearing limit of an electronic yam cleaner. The abscissa of the coordinate system represents the error mass and the ordinate represents the error length. An upper and a lower clearing limit are applied in this coordinate system. Neps and thick places above the upper clearing limit, and thin places below the lower clearing limit are automatically removed from the yam. One may proceed in an analogous manner also with the method according to the invention by way of defining at least one "clearing limit". Thick or thin places within this limit are filtered out and only slubs outside this limit are evaluated further and/or represented graphically. Slub characteristics. Thus e.g. slubs which fall short or exceed a certain slub length Le, which fall short or exceed a certain slub total mass Me, which have or do not have a certain slub shape (cf Fig. 15) etc. may be filtered out. Only the remaining slubs which are not filtered out are evaluated fiirther and/or graphically represented. It is also possible to provide several different filters, so that for example, with a first filter, only short slubs, and with a second filter, only long slubs are able to be evaluated and/or graphically represented. The method according to the invention preferably permits an interactive input of certain parameters on the part of an operating person. Such parameters to be inputted may be filter parameters for the filters discussed above. The basics for the evaluation may also be inputted as parameters, thus e.g. a defined base yam mass Ms. hi the method according to the invention, it may be advantageous to provide an interface for outputting data which has been obtained in the method. Such data may e.g. be the fancy yam parameters discussed above, which are transferred to simulation software. From this, the simulation software may be a simulation of the examined yam of a sheet formation woven or knitted fi-om the yam. Such a simulation based on evaluation data is comparatively quick and simple compared to a simulation based on measurement data. Of course, the present invention is not limited to the embodiments discussed above. The man skilled in the art, with the knowledge of the invention, is capable of deriving further variants which also belong to the subject-matter of the present invention. Individual features of the method according to the invention, which are described above, in particular the evaluation algorithms for the individual fancy yam parameters, may also be applied detached fi-om the graphic representation according to the invention. Although the present description is concentrated on the example of the capacitive measurement of the yam mass, the invention is not limited to this scanning principle. Indeed, other scanning principles - possibly with other measurement variables - may be applied with the method according to the invention, e.g. the optical measurement of the yam diameter. Combinations of different scanning principles are also possible. LIST OF REFERENCE NUMERALS 1 device scanning unit 23 first data lead evaluation unit 34 second data lead output unit input unit 9 fancy yam 91,91' base yam 92,92' slub 93,94 slub flanks 95 slub roof 100 measurement curve 101 noise floor 102,102' signal peak 103,103', 103* slub beginning 104,104* slub end local maximum of a peak thin place next to slub 111-115 classes of slubs, slub populations 111.1-113.1 part areas of the scatter diagram local maximum which belongs to the base yam local maximum which belongs to the slubs 131 distribution in the spectrogram 13 2 maximum of the distribution 121 133 peak in the spectrogram A area below me measmement curve B color blue H frequency of a reading 1m mass interval L period length Le slub length Ls slub distance Le +Ls slub distance M mass per length unit Me slub total mass Mm mean of the measured mass per length unit Ms base yam mass per length unit Mt threshold value R color red X length coordmate AM mass increase of a slub. AMENDED PATENT CLAIMS 1. A method for the characterization of fancy yam (9), wherein at least one characteristic of the fancy yam (9) is scanned along the longitudinal direction (x) of the fancy yam (9), values of the scanning are evaluated and results of the evaluation are outputted, characterized in that at least one result of the evaluation is outputted in the form of a graphic representation. The method according to claim 1, wherein the graphic representation is selected from the group of diagrams which comprises a recording of a scanned fancy yam characteristic with respect to the position (x) on a fancy yam (9) or with respect to time, a histogram, preferably in the form of a two- or three-dimensional column chart or bar chart, a scatter diagram, a classification matrix, a surface in a three-dimensional or two-dimensional representation, a column chart, a bar chart, a circular chart, a pie chart, a table and a spectrogram. The method according to claim 2, wherein the graphic representation is a scatter diagram on which a mass increase (AM) or a diameter increase of slubs (92) is plotted against a slub length (Le). The method according to claim 2, wherein the graphic representation is a classification matrix, whose horizontal axis is a characteristic of a leading slub (92) of a pair (92, 92') of adjacent slubs and whose vertical axis indicates the same characteristic of a trailing slub (92') of the pair (92,92'). The method according to claim 2, wherein the graphic representation is a classification matrix, whose one axis indicates a slub parameter and whose other axis indicates various steps of slubs. The method according to claim 2, wherein the graphic representation is a surface which lies in three-dimensional space, whose first dimension indicates a slub length (Le), whose second dimension indicates a mass increase (AM) or a diameter increase of slubs (92), and those third dimension indicates an observed frequency (H) of the respective slub characteristics (Le, AM), or wherein the graphic representation is a projection of such a surface into the plane spanned by a slub length axis and a mass increase axis or a diameter increase axis. The method according to claim 2, wherein the graphic representation is a column- or bar chart whose columns or bars are allocated to the slubs (92, 92') on the fancy yam (9), and in each case are composed of at least two parts, and wherein the length or area of a first part indicates a characteristic (Le) of the respective slub (92), and the length or area of a second part indicates a characteristic (Ls) of an adjacent base yam (91). The method according to one of the preceding claims, wherein different classes (Hill 3) of slubs (92) are identified by way of the graphic representation. The method according to claim 8, wherein the identified classes (111 -113) of slubs (92) are delimited fi-om one another and/or fi-om outUers The method according to one of the preceding claims, wherein on evaluation, a measurement curve (M(x)) is produced fi-om the values of the scanning, and an area (A(x)) below the measurement curve (M(x)) is computed. The method according to claim 10, wherein the area (A(x)) between the measurement curve (M(x)) and a previously detemined base yam mass (Ms) or a previously determined base yam diameter is computed. The method according to one of the preceding claims, wherein the at least one scanned characteristic is a mass (M) and/or a diameter of the fancy yam (9). The method according to one of the preceding claims, wherein results of the evaluation are selected fi-om the group of fancy yam parameters which includes a base yam mass (Ms), a base yam diameter, a slub distance (Ls), a mass increase (AM) of a slub (92), and slub diameter increase, a slub diameter, and slub length (Le), a slub total mass (Me), an average yam number, a number of slubs per length unit, a pattern length, a sub-pattem length, a shape and a color. The method according to of the claim 13, wherein a running number is allocated to each slub (92, 92'), and the running number is stored together with the parameters of the associated slub (92). The method according to claim 13 or 14, wherein results of the evaluation are minima, maxima, arithmetic means and/or standard deviations of the fancy yam parameters and/or the number of slubs (92) per yam length. The method according to one of the preceding claims, wherein the evaluation includes a smoothing or idealization of the scanning values. The method according to claim 16, wherein the evaluation includes a linking of the smoothed or idealized scanning values with the original scanning values. The method according to claim 17, wherein the linking is a difference formation or a quotient formation. 19. The method according to one of claims 16-18, wherein the smoothing or idealization contains the approximation by straight stretch sections and slubs (92,92') of the fancy yam (9) are approximated as trapeziums, triangles or rectangles. 20. The method according to one of the preceding claims, wherein the evaluation includes a filtering of the scanning values. 21. The method according to one of the claims 17-18 and one of the claims 1-9, wherein an individual graphic representation for the smoothed or idealized scanning values and/or an individual graphic representation for the data which has come from the linking is outputted.

Full Text

A METHOD FOR THE CHARACTERIZATION OF FANCY YARN
FIELD OF THE INVENTION
The invention relates to a method for the characterization of fancy yam according to the preamble of the first patent claim. It may be applied in the textile laboratory (offline) as well as in textile production (online), e.g. in spinning works or winding works.
STATE OF THE ART
Fancy yam is yam whose structure or fiber composition differs from the normal smooth yam. It is applied in weaving mill products and knitting mill products as an enriching element. Fancy yam usually has a multitude of thick places or thin places - so-called slubs - whose diameter is significantly larger or smaller than the diameters of the yam sections lying between the slubs - the so-called base yam. Stmctured yam with deliberately produced thickness variations with which no base yam may be identified is however also counted among the fancy yams. The increasing popularity of fancy yam demands reliable and meaningful methods for its characterization. Variables such as base yam diameter, diameter increase at a slub, slub mass, slub length, slub distance, etc. are of particular interest. These variables may e.g. be used for the control of the quality of the present fancy yam or for determining the manufacturing parameters which are necessary for copying a given fancy yam.
Methods and devices for characterization of yam are known. They are usually based on a capacitive and/or optical scanning of the yam moved in the longitudinal direction. The capacitive scanning principle provides a signal corresponding to the yam mass, whilst the optical scanning principle provides a signal proportional to the yam diameter. The scanning signal is evaluated in an analog or digital manner, and one or several results of the evaluation are outputted. Examples of such methods and devices for the characterization of yam are specified in the patent publications EP-0'578'975 Al or EP-0'249741 A2. Both relate to the yam testing system USTER®TESTER which is marketed worldwide by the proprietor of the present protective property right.
It is also known to obtain information on the color, i.e. on the spectral reflection characteristics of yam. Thus e.g. according to WO-2004/044579, one applies a multi-colored hght source for illuminating the yam. The light reflected by the yam is detected separately in at least two different spectral ranges. The at least two detection signals permit information with
f
regard to the yam cx)lor. A simultaneous optical scanning of a textile material at different wavelengths is also known from CH 674379 or from DE-198'59'274.
WO-2005/07 1150 Al, WO-2005/038'105 Al and WO-2005/037699 Al especially deal with fancy yam. The teaching of the latter document may be summarized as follows:
determining the base yam diameter: Firstly, the arithmetic mean of the yam diameter is formed over a large yam length. This mean is subtracted from the individual values of the yam diameter. The arithmetic means of all negative values which have been measured adjacentiy to the other negative values is defined as the base yam diameter.
determining the beginning, end and length of a slub: A slub begiiming is present if a limit diameter which lies above the base yam diameter is overshot, and the overshoot persists over a certain yam length. A slub end is present if the limit diameter is undershot, and the undershoot persists over a certain yam length. The slub length is defined as a distance between the slub begiiming and the slub end.
determining the slub diameter. A plurality of the largest diameters is determined within a slub . The slub diameter is defined as the mean of these largest diameters.
Although this teaching permits a useful characterization of fancy yam, it also has a few disadvantages. A large quantity of data occurs as a result, which is unclear and difficult to handle. Suitable possibilities for the clear representation of the readings are not specified. The described method, if anything, results in a base yam diameter which is too large and in a slub length which is too small.
DESCRIPTION OF THE INVENTION
It is an object of the present invention to specify a method for the characterization of fancy yam which provides a clear representation of the measurement results. The occurring data should be reduced compared to the teaching of WO-2005/037688 Al. Furthermore, the invention should permit the occurring data to be divided up into certain categories in a suitable manner, i.e. into data which concern the slubs, and data which concem imperfections of the yam.
These and other objects are achieved by the method defined in patent claim 1. Advantageous embodiments are specified in the dependent patent claims.
According to the method according to the invention, for the characterization of fancy yam which preferably has a sequence of slubs and base yam, at least one characteristic of the fancy yam is scaimed along the longitudmal direction of the fancy yam. Values of the scanning
are evaluated and results of the evaluation are outputted. Results of the evaluation are outputted: it is preferably the case of fancy yam parameters such as base yam mass, base yam diameter, slub distance, mass increase of a slub, slub diameter increase, slub diameter, slub length and/or slub mass. At least one result of the evaluation is outputted in the form of a graphic representation. Such a graphic representation reduces the occurring data quantity. It permits a rapid visual perception of the results. By way of this, it is also possible to divide up the data into different categories. In a preferred embodiment, the graphic data is a scatter diagram, and the at least one scanned characteristic is a mass and/or a diameter of the fancy yam.
LISTING OF THE DRAWINGS
The invention is hereinafter described in more detail by way of the drawings.
Figure 1 schematically shows a device for carrying out the method according to the
invention.
Figure 2 shows one example of a series of readings with regard to a fancy yam,
specifically yam mass against length coordinate.
Figure 3 shows a frequency distribution of the measured yam mass plotted in
Figure 2.
Figure 4 shows an enlarged section of Figure 2.
Figures 5-15 show possible representation types according to the invention, for
outputting the yam parameters.
Figure 16 shows an example of a series of reading with regard to a two-step fancy
yam, in the same representation as Fig.2.
Figure 17 shows one possible representation type according to the invention, for
outputting the characteristics of a multi-step yam.
Figure 18 shows a spectrogram of a series of readings with regard to a fancy yam.
Figure 19 shows a meastired and an idealized slub .
Figure 20 shows a measurement curve with a thin place, analogous to Figure 4.
Figure 21 shows the course of the area below a measurement curve, as a function of
the length coordinate.
Figure 22 shows the mass per length unit as a function of the length coordinate, (a)
for the measurement curve, (b) for an idealized curve and (c) for a curve which arises by subtraction of the curve (b) from the curve (a).
Figure 23 shows spectrograms (a) of the curve of Fig. 22(a), (b) of the curve of Fig.
22(b), and (c) of the curve of Fig. 22(c).
Figure 24 show scatter diagrams (a) of the readings of Fig. 22(a), (b) of the slubs of
Fig. 22(b), and (c) of the virtual base yam of Fig. 22(c).
IMPLEMENTATION OF THE INVENTION
A device 1 for carrying out the method according to the invention is shown schematically in Figure 1. It contains a scanning unit 2 for scanning a fancy yam 9 with base yam 91 and slubs 92, 92', which is moved in a longitudinal direction -x. Here, the sequential recording of a multitude of readings at different, preferably equidistant locations of the fancy yam 9 is to be understood under the term "scanning". Such scanning imits 2 are knovra per se, and do not need to be explained in more detail here. The scanning unit 2 may contain a capacitive, optical or other sensor; also several equal or different sensors may be arranged within the scanning unit 2. The scanning unit 2 may be provided with evaluation means for a preliminary evaluation of the readings. It outputs a preferably electrical output signal which is a measure for the mass, the thickness or other characteristics of the fancy yam 9, on a first data lead 23.
The first data lead 23 runs into an evaluation unit 3 which is suitable for evaluating the output signal of the scanning unit 2. For this purpose, it contains suitable analog and/or digital evaluation means, e.g. a microprocessor. It may also contain further means such as memory means for storing data. The evaluation unit 13 is preferably a computer.
Furthermore, the device 1 contains an output unit 4 for outputting measurement data and/or results of the evaluation. The output unit 4 is coimected to the evaluation unit 3 by way of a second data lead 34. It may e.g. be designed as a monitor and/or printer. The device 1 preferably also contains an input unit 5 for inputting data on the part of the user. The input unit 5 may e.g. be a keyboard, a mouse and/or a touch screen.
Figure 2 shows one possible output signal 100 of the scanning unit 2. Thereby, a variable for example which is a measure for the mass M per unit length of the fancy yam 9 (see Figure 1), is plotted along the length coordinate x of the fancy yam 9. Such a measure signal M is typically provided by a capacitive yam sensor. A representation of the thickness of the fancy yam 9 along the length coordinate x would appear in a similar manner, wherein the scales on the abscissa and the ordinate as well as the resolution could be different; a thickness signal is typically provided by an optical yam sensor. The curve M(x) is not necessarily continuous, but may be composed of individual (not shown in Fig. 2) scanning points which on the fancy yam 9 are typically distanced from one another by a few millimeters; the distance depends on the scanning rate of the scanning xmit 2 and on the speed of the fancy yam 9. The signal M(x) has a noise floor 101 which corresponds to the base yam mass Ms or the base yam diameter. The peaks 102 corresponding to the slubs project significantly from the noise floor 101. A mean Mm of all readings M(x) over a large yam length, on account of the peaks 102, lies significantly above the noise floor 101 and is therefore not suitable as a measure for the base yam mass Ms. At this location, it should be noted that Figure 2 and the subsequent discussion only represents one possible, non-limiting example. With other types of fancy yam, it is no longer even possible to identify base yam.
The base yam mass Ms may be determined according to the present invention preferably as follows. A frequency distribution H(M) of the measured masses M represented in Figure 2 is determined. Such a frequency distribution H(M) is schematically shown in Figure 3. The frequency distribution H(M) must be determined for a large yam length, under which here a yam length is to be understood as one which contains may slubs, for example at least 10 and preferably at least 100. The frequency distribution H(M) with an fancy yam 9 comprises at least two local maxima 121, 122: a first local maximum 121 for the base yam 91, and at least one second local maximum 122 for the slubs 92, 92', .... According to definition, amongst all local maxrnia 121, 122, it is the local maximum 121 belonging to the base yam 91 which has die smallest mass M. For this reason, the smallest mass M at which a local maximvmi 121 occurs in the frequency distribution H(M) is defined as the base yam mass Ms. The yam number of the yam body or of the base yam 91 may be computed from the base yam mass Ms. One may proceed in an analogous manner in order to determine the base yam diameter.
According to an alternative embodiment of the method according to the invention for determining the base yam mss, one defines a mass interval Im (see Figure 3) which contains that smallest mass at which a local maximum 121 occurs in the frequency distribution H(M). The mass interval Im may, but need not be symmetrical with respect to the "smallest" mass, i.e. the "smallest" mass may, but need not he in the middle of the mass interval Im- The upper limit of the mass interval Im is preferably selected below the global mass mean Mm- The width and the position of the mass interval Im may be predefined in a fixed manner - for example ±5% of the
"smallest" mass - automatically computed by the evaluation unit 3, or be inputted by a user. A mean is subsequently formed over all masses M measured in this mass interval Im- The base yam mass is defined as this mean.
Figure 4 shows a more detailed view of a section of the measurement curve M(x) of Fig. 2. Hereinafter, it is explained by way of Figure 4 how further fancy yam parameters such as slub length Le, slub distance Ls, mass increase AM and slub total mass Me are evaluated.
For ascertaining a beginning 103 and end 104 of a slub 92, one previously determines a threshold value Mt which is larger than the base yam mass Ms. The threshold value Mr may be predefined in a fixed mannei- - for example 110 % of the base yam mass Ms, automatically computed by the evaluation unit 3 or inputted by a user. The beginning 103 of a slub 92 is present if, proceeding fixim the noise floor 101, the threshold value Mj is overshot. In order to exclude artifacts due to outliers, one may additionally control as to whether the overshoot persists over a predefined yam length, i.e. whether a few further measurement points, which directly follow the measurement point exceeding the threshold value Mt, likewise he above the threshold value Mt. In order to determine the beginning 103 of the slub 92 in a more accurate manner, one goes so far back on the measurement curve M(x) until, for the first time, a reading is smaller or equal to the base yam mass Ms - or another predefined or computed value. This reading is defined as the beginning 103 of the slub 92. The end 104 of the slub 92 is ascertained mutatis mutandis in an analogous method: on xmdershooting the threshold value Mt proceeding fi-om the signal peak 102, one moves so far forwards on the measurement curve M(x) until, for the first time, a reading is smaller or equal to the base yam mass Ms. If required, one may use different threshold values for determining the slub beginning 103 and the slub end 104.
The slub length Le according to the present invention is defined as the distance between the beginning 103 and the end 104 of the slub 92. The slub distance Ls is defined as the distance between the end 104 of a slub and the beginning 103' of a subsequent slub 92' to which a subsequent signal peak 102' belongs. The distance between two adjacent slubs 92, 92' is defined as the sum Le+Ls of slub length and slub distance. Typical slub lengths Le and slub distances Ls He in the range between 2 cm and a few meters.
The mass increase AM which corresponds roughly to a diameter increase of the fancy yam 9 is defined in the simplest case as the difference between a local maximum 105 of the corresponding signal peak 102 and the base yam mass Ms. Refined methods are possible for determining the mass increase AM, which take account of the fluctuations of the scanning signal M(x) along the slub length. Thus for example - analogously to the evaluation of the base yam mass Ms described above - the most common value within the corresponding slub length may be selected. A mean fomiation of values on the slub ridge is also considered. The mass increase AM
is preferably specified as a multiple of the base yam mass Ms, e.g. in percentage values, wherein the base yam mass Ms is preferably defined as 100 %. Typical mass increases AM lie in a range between 20 % and 500 %.
A further parameter for characterizing a slub 92 is the so-called slub total mass Me. This is essentially the difference between (i) the integral of the measurement curve M(x) over the slub length Le and (ii) the mass Ms-Le of the yam body on this slub length Le. The slub total mass Me may be determined by calculation by the evaluation imit 3. The yam number of the slub 92 may be computed fi-om the fancy yam mass Me, wherein the computation may contain a division of the slub total mass Me by the slub length Le.
The shape of a slub 92 may also be determined and outputted. Thereby, one may fall back on a comparison with simple geometric shapes such as a bar, triangle, step, trapezium, or bell; cf Fig. 18. The respective shape may e.g. be outputted on the output unit 4.
Not only are "local" parameters of the individual slubs of interest, but also "global" parameters of a whole yam section, of a yam or of a group of several yams. Such a global fancy yam parameter is the average yam number (average yam mass), which may be computed for example by way of mean formation over all readings. The yam number of all base yam 91 may also be of interest. A further global fancy yam parameter is the average spatial frequency of the slubs, i.e. the average number of slubs per length unit.
The fancy yam parameters mentioned above, and possible further ones are preferably determined dynamically during a running time of the scanning. The slub parameters are stored for the purpose of outputting. It is advantageous to additionally store a continuous running number allocated to the respective slub, in order not only to be able to provide information on the individual slubs, but also on their sequence.
It may be advantageous to equip the scanning unit 2 (Fig. 1) with a capacitive as well as an optical yam sensor which may simultaneously measure one and the same yam. The output signals of the capacitive and of the optical sensor may be linked to one another in a suitable manner for the purpose of an improved or more accurate evaluation. Capacitive measurement has the advantage that it provides a signal with a good signal-to-noise ratio. The signal however is proportional to the mass per length unit, and thus does not correspond to the visual impression of the yam. This has a disadvantageous effect indeed with fancy yam which in the region of a slub 92 often has a different yam density than in the region of a base yam 91. Optical measurement has the advantage of better representing the visual impression of yam, because it measures essentially the visible yam diameter and it is therefore better suitable to fabric simulations. For this, the optical measurement signal has a greater noise than the capacitive
measurement signal. It is possible by way of suitable linking of the two output signals, to profit from the advantages of both measurement types and to eliminate or at least weaken their respective disadvantages.
Results of the evaluation may on the one hand be variables such as e.g. minima, maxima, arithmetic means and/or standard deviations of the above defined fancy yam parameters. The number of slubs 92 per yam length may be a further variable of interest. These variables may be outputted as alphanumeric signs. On the other hand, the results of the evaluation may be graphically represented and outputted on the output unit 4 in a suitable manner. Preferred presentation types are shown in the Figures 5-11.
One possible representation type is a histogram, i.e. the graphic representation of the class frequencies of the classed yam parameters. Three examples in the form of bar charts are specified in Figure 5. The ordinate in each case is the frequency H. In Figure 5(a) the mass increase AM, in Figure 5(b) the slub length Le, and in Figure 5(C;) the slub distance Ls have been use as the abscissa. The abscissas may have a linear, logarithmic or other division. The division and/or scale of the axes may be automatically computed or may be selected by way of input on the part of an operating person. The same applies to the selection of the classification, i.e. to tiie width of the classes. Not all classes necessarily need to have the same with.
The representation manner of Figure 6 is a so-called scatter diagram (aggregate of plots). In this, the mass increase AM is plotted against the slub length Le for all slubs 92 of an fancy yam 9 or a yam section. Each slub 92 is plotted as a point at the correct location. This representation simplifies the division of different slubs 92 into different classes. Thus for example, it is immediately evident from Figure 6(a), that the examined fancy yam 9 has three classes 111 -113 or populations of slubs 92:
a first class 111 with short, low slubs
a second class 112 with long, low slubs, and
a third class 113 with long, high slubs.
An outlier analysis is carried out by way of the scatter diagram of Fig. 6(a). For this purpose, one may provide a tool in order to define slub populations 111-113 and to the delineate them from one another as well as from outliers. Such a tool may e.g. permit part surfaces 111.1-113 .1 of the scatter diagram to be defined on a monitor with a mouse, as is represented in Figure 6(b). One slub population 111 -113 is allocated to each part svirface 111.1-113.1. The part surfaces may e.g. be shaped as a rectangle 111.1, as a circle 112.1 or a polygon 113.1. Points lying outside the part surfaces 111.1-113.1 are to be graded as outliers and with further evaluations and/or representations, may be characterized as such or not taken into account. If no
part surfaces are defined, then all points of the scatter diagram are counted as slubs and are treated as such.
Other scatter diagrams are possible, for example slub total mass Me against slub length Le, slub total mass Me against mass increase AM, etc. The scatter diagram may be issued in a colored manner, wherein different colors may indicate different measurements, different point densities and/or populations or outliers (cf Fig. 9). A three-dimensional representation is analogously possible, in which two coordinates correspond to those of Fig. 6, and the third coordinate corresponds to the point density; cf Fig. 8.
The scatter diagram may be represented for an individual fancy yam sample or for several fancy yam samples. In the latter case, one may use different colors for the different fancy yam samples, in order to display possible differences between the samples. With several fancy yam samples, one may display the result of the entirety of all fancy yam samples additionally to the results of the individual fancy yam samples, preferably in a color which is individually allocated.
Apart from the actual values, one may also represent nominal values or nominal regions for one or more classes of slubs on the scatter diagram. The nominal and actual values may also be compared in other representation types, or in a purely numeric manner. Such nominal-actual comparisons permit e.g. a control on the quality of a copy (actual value) or a predefined fancy yam (nominal value).
The results may also be issued in the form of a table or a classification matrix - as shown in Figure 7, instead of a scatter diagram. The table axes correspond to the axes of the scatter diagram of Fig. 6. The numbers of respective slubs 92 are entered into the fields of the tables. Alternatively to the absolute number of slubs 92, one may also indicate their relative share; e.g. in percent or per thousand. Each table field thus represents a class of slubs 92, analogously to the classes 111-113 which are described on the occasion of Fig. 5. The selected size of the table fields is directed to the desired classification. With the selection of the size of the table fields, one should also take note that the resolution is sufficiently fine, but that the table still remains clear. In the example of Figure 7, relatively small table fields were selected, which permit a finer classification than the three classes 111-113 which were described on the occasion of Figure 6. The table fields may be filled with colors, patterns or steps of gray, which in turn are allocated to different numbers of slubs 92, for the purpose of an improved visualization.
With the scatter diagram (Fig. 6) as well as the classification matrix (Fig. 7), the axes, independently of one another, may have a linear, a logarithmic or other division. The di\dsion and/or scale of the axes may be computed automatically, or may be selected by way of input on
the part of the operating person. The same applies to the width and height of the table fields in the classification matrix of Figure 7, i.e. for the selection of the classification. Figure 8 shows a further preferred output possibility for the determined fancy yam parameters. It is the case of a table which is subdivided into five main columns for the five parameters of slub length Le, slub distance Ls, mass increase AM, diameter increase and number # of the slubs. The first four main columns for their part are subdivided in each case into three sub-columns for the minimum Min, the mean 0 and the maximvim Max of the respective parameter. The lines of the table may contain the values for the entire yam or for the entire examined yam section, as well as for the individual slub populations 111-113. The minima LF,n.m, AMmin, the means Le,0, AlvIo and the maxuna LE,max» AM^ax are indicated for the population 111 in Fig. 6(b). Of course, the table may also be formed in a different manner. In any case, such a tabular representation of the determined fancy yam parameters leads to a reduction of data. Particularities of the fancy yam concerned may be very quickly detected and different fancy yams may be easily compared by way of the table.
A further manner of representation for the results of the evaluation of fancy yam parameters is shown in Figure 9. It is the case of a surface in three dimensions (3D). Two of the three dimensions, the two horizontal axes, correspond to the slub length Le and the mass increase AM, as in the scatter diagram of Fig. 6. The third, vertical dunension indicates the respective frequency H of the measured values, i.e. the point density in the scatter diagram of Fig. 6 or the numbers of Fig. 7. The 3D-surface which arises in this manner gives the impression of a mountain [range] which permits a rapid and memorable visual perception of the peculiarities of the respective fancy yam 9. It is particularly a synoptic comparison of two such "mountains" which very quickly shows whether the fancy yam concerned has similar or different characteristics, and in tiie latter case where the main differences he. It is to be noted that the example of Fig. 9 relates to a different fancy yam than the example of Fig. 6. Whilst the fancy yam of Fig. 6 has three classes 111-113 of slubs 92, the fancy yam of Fig. 9 only has two of them 114,115.
The three-dimensional representation maimer of Fig. 9 may be reduced also to two dimensions. Figure 10 shows such a diagram which arises by way of the projection of the "mountain" of Fig. 9 into the plane spanned by the two horizontal axes Le, M. A "map" thus arises on which the "mountains" 114, 115 of Fig. 9 are repesented by way of "altitude contours", i.e. lines of the same fi-equency H. Instead of "altitude contours", one may apply colors or cross-hatchings for rendering the different fi^equencies H visible.
The representation types of the Figures 5-10 do not take into account mformation on the sequence of the mdividual slubs. This information is completely present in the measurement series as is represented in Figure 2. It is possible from this to determine the respective yam
parameter such as mass increase, slub length and slub distance as the measurement variables (number value x unit), and to list these measurement variables in the sequence of their occurrence, after one another. Such an alphanumeric listing of the reduced measurement data may be useful for certain cases, but is however less clear. Information on the sequence of the individual slubs is fully contained in the representation manner of Figure 11, wherein graphics have the advantage of an improved clarity and visual perception compared to an alphanumeric value table. A horizontal bar is drawn in for each slub 92 and an adjacent base yam 91. The bar is composed of two parts. The length of a first, left part indicates the respective slub length Le, the length of a second, right part indicates the respective slub distance Ls. The next bar lying therebelow characterizes the subsequent slub, etc. The horizontal bars of Fig. 11 may of course also be replaced by vertical columns. Measurement variables other than lengths Le, Ls may be plotted as bars, e.g. a slub mass and the associated base yam mass, which may be advantageous in particular with stmctured yam with different base yam masses. It is also possible for a bar or column to indicate more than two slub parameters, e.g. with multi-step slubs (see Fig. 16) the first slub length Le,i, the second slub length Le2 and the associated slub distance Ls.
Figure 12 shows a manner of representation which at least represents a part of the information on the sequence of the individual slubs in a clear maimer. It is the case of a classification matrix which assumes a classification of the slubs as has been implemented with the classes 111-113, somewhat as in Figure 6. In each case, a pair of two adjacent slubs 92, 92' are considered, of which a first slub 92 is called a "leading slub" and a second slub 92' a "trailing slub". A corresponding entry into the classification matrix of Figure 12 whose horizontal axis indicates the leading slub 92 and whose vertical axis indicates the trailing slub 92' is effected for each pair 92, 92'. One may deduce fi-om the Active example of Figure 12, that in practice, two slubs of the first class 111 (cf. Fig. 6), and two slubs of the second class 112 are never successive, but that a slub of the first class 111 often follows a slub of the second class 112, and a slub of the third class 113 very often follows a slub of the first class 111.
Various slubs of a fancy yam 9 may have different colors. For this reason, it may be desirable to obtain information on the color of the fancy yam 9. Sviitable scanning units 2 and evaluation units 3 (see Fig. 1) fi-om the state of the art mentioned above are known for this. One possible representation manner for the obtained color information is shown in Figure 13. Here it is the case of a circular chart which indicates the measured shares of the differently colored slubs. In the example of Figure 13, the fancy yam contains red (R) and blue (B) slubs which occurred with a fi-equency of 45 % and 55 % respectively. An enhancement to more than two colors is of course also possible. In the case that the output unit 4 (see Fig. 1) permits a colored
The circular chart of Figure 13 contains no information on the sequence of the color slubs. This information is at least partly present in the representation manner of Figure 14. Analogously to Figure 12, in each case two successive slubs are considered, and the frequency of their colors R, B is plotted in the table, wherein the horizontal table axis indicates the color R, B of the leading slub, and the vertical axis indicates the color R, B of the trailing slub. One may deduce from the table of Figure 14 that a color change often occurs in this Active example, whereas two adjacent slubs having the same color is rather rare.
The three-dimensional column charts of Figure 15, by way of Active examples, show how the color information may be combined with the geometric information in a single representation manner. Thereby, the geometric information hes in the plane of the drawing and corresponds to that of Fig. 5(a) and Fig. 5(b) respectively. The frequency of classes of the mass increase AM is plotted in the diagram of Figure 15(a), and the frequency of classes of the slub length Le is plotted in the diagram of Figure 15(b). The third dimension is used for the color information R, B. One may deduce from the diagram of Figure 15(a) that the red slubs R tend to have smaller mass increases AM than the blue slubs B. The diagram of Figure 15(b) indicates that the red slubs R tend to be longer than the blue slubs B.
The Figures 5-15 discussed above only indicate a few example for representing the fancy yam parameters of mass increase AM (or diameter mcrease), slub length Le, slub distance Ls and color. One may of course also graphically represent further relations between these and further fancy yam parameters in a similar and two-dimensional or three-dimensional manner.
Whilst single-step slubs have been discussed up to now, multi-step slubs are considered hereinafter. One example of a series of readings on a two-step fancy yam is specified in Figure 16, in the same representation as Fig. 4. Here one may differentiate between a first slub step with a first slub total mass Me,i per length unit, a first mass increase AMi and a first slub length Le,i, and a second slub step with a section slub total mass Me2 per length unit, a second mass increase AMi and a second slub length Le,2- The mentioned parameters may be determined in a manner which is analogous to that described for a single-step fancy yam.
One possible manner of representation for the parameters of multi-step fancy yam is shown in Figure 17. Here it is the case of a table whose horizontal axis corresponds to the classed mass increase AM; cf Fig. 5(a). The lines of the table represent the different steps of the fancy yam. The respective frequencies are plotted in the fields of the table. The fancy yam is two-step in the fictive example of Figure 17, wherein the second step occurs in two variants: with a relatively small mass increase AMa on the one hand, and with a relatively large mass increase AM2 on the other hand. An analogous manner of representation is also possible for the slub length Le,i, Le2.
A further parameter of fancy yam is the so-called pattern length. This is the length of the shortest sequence of slubs which are periodically repeated in the fancy yam. There is no periodicity whatsoever within this sequence, i.e. at least one slub parameter such as e.g. the slub distance Ls is random or pseudo-random. The pattern length may be obtained e.g. by way of correlation computation from readings, as they are represented for a short yam section in Figure 2. Such a correlation computation with all readings may be extensive with regard to computation. In order to reduce the computational effort, the measurement data may be previously reduced in that e.g. the respective yam parameters such as mass increase, slub length and slub distance are determined and the correlation computation is based on this reduced data. In an analogous manner, one may also obtain information on the presence of - mostly undesired - sub-pattems and their lengths. The pattern length and/or sub-pattem lengths are preferably issued in alphanumeric or graphic fomi.
A spectrogram of the measurement signal M(x) of Figure 2 may also provide useful information on the fancy yam. The measurement signal M(x) is preferably subjected to a Fourier transformation for determining the spectrogram. One fictive example of a spectrogram |F {M} | or more precisely of the real part of the Fourier transform is shown in Figure 18, wherein as is customary, a period length L is selected as the abscissa, preferably in a logarithmic scale. Usually, the spectrogram |F{M}| .displays a relatively broad distribution 131 of the spatial frequencies or of the period lengths L. One may deduce an average distance of the slubs from the position of the maximum 132. A pronounced peak in the specfrogram |F{M}| would indicate a -mostly undesired - periodicity in the fancy yam. The periodicity on the one hand may relate to the individual slubs. With a fancy yam with a constant slub distance Le + Ls, within which the slub length Le and the slub distance Ls vary, the maximum 132 appears as a pronounced peak. In order to ascertam this, a yam length of at least ten, preferably one hundred and more slub distances should be measured. The periodicity on the other hand may relate to the pattems. With a sufficiently long measurement series - at least ten, but preferably one hundred and more pattern lengths - one may also read out the pattem length from the position of a respective peak 133 in the long-waved region of the spectrogram |F{M} |.
A further graphic representation manner for the slubs is a smoothed or idealized representation of the readings, as is shown in Figure 19. On the one hand, the readings of Figure 4 are drawn in a dotted manner. On the other hand, an idealized course of the measurement curve is indicated with an unbroken line. As the man skilled in the art knows, there are many possibilities of obtaining an idealized curve in the manner of Figure 19 from a real measurement curve. In the application example of Figure 19, base yam 91, 91' have been approximated by horizontal straight lines which all lie at the height of the previously determined base yam mass
Ms. One slub 92 is idealized in each case as a trapezium with flanks 93, 94 and a horizontal roof 95. Thereby, the trapezium does not necessarily need to be symmetrical, i.e. the flanks 93, 95 with regard to magnitude may also have different gradients. The approximation of the slub 92 by a trapezium may be effected according to methods and criteria known to the man skilled in the art. The flanks 93, 94 may roughly be straight lines whose positions have been determined by way of the method of least squares, or in another maimer. The height of the trapezium, i.e. the position of the roof 95, may be determined according to the criterion that the area of the trapezium is equal to the area below the real measurement curve. The slub length Le may for example be defined as the base length of the trapezium, i.e. as the distance between a slub beginning 103* and a slub end 104*, wherein the slub beginning 103* and the slub end 104* are the intersection points of the horizontal representing the base yam mass Ms and the left flank 93 or the right flank 94. Alternatively, the slub length Le may be may be defined as the width of the trapezium at half the height, which is equal to half the sum of the base length and the roof length. The mass increase AM may be defined at the height of the trapezium, i.e. as the distance between the base and the roof Under certain circumstances, the trapezium may degenerate into flie special case of a triangle (trapezium with a roof length equal to zero) or of a rectangle (trapezium all with right angles). Other shapes for the idealized measurement curve are likewise possible. Such idealized courses of curves permit an unproved visual perception of the characteristics of the fancy yam. Parameters of the idealized curve, such as e.g. height, base length, roof length, the slub length Le or the flank gradients of the trapezium may be outputted as characteristic variables in a table for example, and be used for further evaluations. The output of such variables leads to a reduction in the data.
An incomplete manufacturing process for the fancy yam may lead to a thin place 106 being present directly next to a slub 102, as indicated in Figure 20. Such undesired thin places 106 may be detected according to the present invention. Their number or share with regard to quantity at the slubs 102 may be outputted as a result. A thin place share of 50 % means that a thin place 106 was observed next to half of all slubs 106, which indicates a deficient manufacturing process.
Spinning works which manufacture fancy yam have the need to differentiate between the following two phenomena:
on the one hand, the virtual base yam manufacture, which may introduce imperfections, irregularities and faults such as thick places or thin places into the yam, and
on the other hand, the slub manufacture which incorporates the desired slubs onto the virtual base yam, e.g. in the form of thickenings.
These two phenomena are sometimes impossible or difficult to differentiate with conventional yam testing methods and apparatus. The exemplary curve of Fig. 4 was selected for didactic reasons, so that it is quite clear from this, what a slub 102 is and what a base yam 101 is. In practice however, the undesired thickness fluctuations on the base yam 101 may be so large, that they exceed the threshold value Mr and as a result are wrongly considered to be a slub on evaluation. The results of the evaluation are therefore adulterated. The results of this may lead to the wrong measures being taken in the manufacturing process. If e.g. long thick places or yam mass fluctuations are assumed to be small slubs, the part process for slub manufacture is changed such that larger slubs may be produced. This measure unnecessarily changes the slub stmcture without alleviating the actual cause of the fault - perhaps a defect in one path.
Here, a preferred embodiment solves this problem by way of specifying the definition of a slub. The method described on the occasion of Fig. 4 may be disadvantageous since on setting the threshold value Mn, it only takes into account a one-dimensional mass increase. As described, although one may attempt to reduce this disadvantage by way of the demand that the excess must last for a certain length. This however also does not lead to an optimal recognition of the slubs. The observation of an area below the measurement curve - or what is considered here as being equivalent to the area, a certain integral of the measurement curve, has been shown to be more advantageous. Such an area may be computed at least approximately with one of the known numeric integration methods, like with the rectangle or trapezium method. Figure 21 schematically shows a typical course of the area A(x) computed in this manner, as a function of the length coordinate x, wherein the previously determined base yam mass Ms (see Fig. 4) serves as basis for determining the area. The values of A(x) fluctuate around the value zero in the region of the base yam 101. Any thick places, even if they have a large mass increase, do not cause any significant area changes, since they only extend in each case over a short length interval. A real slub 92 differs from a thick place in that it causes a significant rise of the curve A(x) which sets in at the beginning of a slub 102. A local maximum 105 of the slub 92 which may also be described as the position of the slub 92, is located at the turning point of the curve A(x). After the slub end 104, the curve A(x) again fluctuates around a constant value, which indicates the slub total mass. If one ascertains such a slub end 104, then the further course of the curve may be set back to the value zero by way of subtraction of the value Me. A further base yam follows etc. All slubs and their positions may be recognized in a reliable and stable manner and be differentiated from thick places by way of this.
The ascertaining of a slub is preferably made dependent on the simultaneously fulfilhnent of several criteria, e.g. of the following three criteria:
(i) exceeding a predefined threshold value for the area A(x),
(ii) exceeding a predefined threshold value for the slub length Le, and
(iii) exceeding a predefined threshold value for the mass increase AM.
Only when these criteria are simultaneously fulfilled may one reliably assume that a proper slub and not a yam imperfection is present.
The differentiation between the virtual base yam and slubs is simplified even greater by way of a further embodiment of the method according to the invention. According to this embodiment, the idealized course of the curve according to Fig. 19 is subtracted fi-om the real measurement curve. This is schematically represented in Figure 22. In this, Figure 22(a) shows the curve of the original readings, Figure 22(b) the idealized curve (cf Fig. 19), and Figure 22(c) the curve which arises when one subtracts the idealized curve from the original measurement curve. The curve of Fig. 22(b) shows only the (idealized) slubs, the curve of Fig. 22(c) only the virtual base yam without slubs. These representations alone may contribute to a deeper understanding of the stmcture of the examined fancy yam. The data obtained in this manner may however be evaluated even, further, as is discussed hereinafter.
Figure 23 schematically shows results of an evaluation of the data of Figure 22 in a spectrogram, i.e. in a representation which corresponds to that of Fig. 18. Figure 23(a) shows the spectrogram of the original readings, i.e. the curve of Fig. 21(a). As already indicated above, it is difficult or even impossible to differentiate between slubs and virtual base yam only by way of this spectrogram. The representations of the Figures 23(b) and (c) alleviate this problem. Figure 23(b) shows the spectrogram of the slubs alone, i.e. of the curve of Fig. 22(b). This data related to the slub permits malfunctioning in the slub production to be localized and overcome in a targeted manner, or permits the slub production to be changed in a targeted manner. Peaks in the long-waved region may e.g. indicate undesired periodicities, which may be avoided with suitable measures. Figure 23(c) shows the spectrogram of the virtual base yam, i.e. of the curve of Fig. 22(c). Any peaks in this spectrogram give good hints as to certain faults in the spinning process, or in the process stages preceding the spinning process, such as eccentricities of certain rollers in the drawing arrangement. These faults may be localized by way of the respective wavelengths in the spectrogram of Fig. 23(c), and subsequently dealt with.
The data of Fig. 22 may just as easily be represented in a scatter diagram, analogously to Fig. 6. This form of representation may also be very useful in order to differentiate between slubs and the virtual base yam. A schematic example is specified in Figure 24. Figure 24(a) shows a scatter diagram of the original readings of the curve of Fig. 22(a). Here, three phenomena intermingle:
the lesser disturbing thick places occurring in each yam,
slubs and
• thick places wrongly evaluated as slubs.
Figure 24(b) shows a scatter diagram of the slubs on their own, as they are represented in Fig. 22(b). In particular, when the robust method for slub recognition described on the occasion of Fig. 21 is used, this scatter diagram will have no points which undesirably originate from imperfections such as thick places. Rather it only contains real • slubs. Figure 24(c) shows a scatter diagram of the virtual base yam of Fig. 22(c). The points drawn therein do not represent slubs, but imperfections such as thick places. The scatter diagram of Fig. 24(c) may provide information on the applied yam manufacturing process.
Other representation types such as e.g. the histograms of Fig. 5 may be used separately in an analogous manner for the data related on the one hand to the slub, and for the data related on the other hand to the virtual base yam.
It may be advantageous in the method according to the invention to filter the readings as are represented somewhat in Figure 2 or 4, according to certain filter criteria. Such filter criteria may for example be the following:
yam imperfections such as neps, thick- and/or thin places. Thus for example, it is known from CH-678'173 A5 or from US-5,537,811 A, to arrange possible yam errors in a table in the manner of a coordinate system, for setting the clearing limit of an electronic yam cleaner. The abscissa of the coordinate system represents the error mass and the ordinate represents the error length. An upper and a lower clearing limit are applied in this coordinate system. Neps and thick places above the upper clearing limit, and thin places below the lower clearing limit are automatically removed from the yam. One may proceed in an analogous manner also with the method according to the invention by way of defining at least one "clearing limit". Thick or thin places within this limit are filtered out and only slubs outside this limit are evaluated further and/or represented graphically.
Slub characteristics. Thus e.g. slubs which fall short or exceed a certain slub length Le, which fall short or exceed a certain slub total mass Me, which have or do not have a certain slub shape (cf Fig. 15) etc. may be filtered out. Only the remaining slubs which are not filtered out are evaluated fiirther and/or graphically represented. It is also possible to provide several different filters, so that for example, with a first filter, only short slubs, and with a second filter, only long slubs are able to be evaluated and/or graphically represented.
The method according to the invention preferably permits an interactive input of certain parameters on the part of an operating person. Such parameters to be inputted may be filter
parameters for the filters discussed above. The basics for the evaluation may also be inputted as parameters, thus e.g. a defined base yam mass Ms.
hi the method according to the invention, it may be advantageous to provide an interface for outputting data which has been obtained in the method. Such data may e.g. be the fancy yam parameters discussed above, which are transferred to simulation software. From this, the simulation software may be a simulation of the examined yam of a sheet formation woven or knitted fi-om the yam. Such a simulation based on evaluation data is comparatively quick and simple compared to a simulation based on measurement data.
Of course, the present invention is not limited to the embodiments discussed above. The man skilled in the art, with the knowledge of the invention, is capable of deriving further variants which also belong to the subject-matter of the present invention. Individual features of the method according to the invention, which are described above, in particular the evaluation algorithms for the individual fancy yam parameters, may also be applied detached fi-om the graphic representation according to the invention. Although the present description is concentrated on the example of the capacitive measurement of the yam mass, the invention is not limited to this scanning principle. Indeed, other scanning principles - possibly with other measurement variables - may be applied with the method according to the invention, e.g. the optical measurement of the yam diameter. Combinations of different scanning principles are also possible.
LIST OF REFERENCE NUMERALS
1 device
scanning unit 23 first data lead
evaluation unit 34 second data lead

output unit
input unit
9 fancy yam
91,91' base yam
92,92' slub
93,94 slub flanks
95 slub roof
100 measurement curve
101 noise floor
102,102' signal peak
103,103', 103* slub beginning
104,104* slub end
local maximum of a peak
thin place next to slub
111-115 classes of slubs, slub populations
111.1-113.1 part areas of the scatter diagram
local maximum which belongs to the base yam
local maximum which belongs to the slubs
131 distribution in the spectrogram
13 2 maximum of the distribution 121
133 peak in the spectrogram
A area below me measmement curve
B color blue
H frequency of a reading
1m mass interval
L period length
Le slub length
Ls slub distance
Le +Ls slub distance
M mass per length unit
Me slub total mass
Mm mean of the measured mass per length unit
Ms base yam mass per length unit
Mt threshold value
R color red
X length coordmate
AM mass increase of a slub.

AMENDED PATENT CLAIMS
1. A method for the characterization of fancy yam (9), wherein
at least one characteristic of the fancy yam (9) is scanned along the longitudinal direction (x) of
the fancy yam (9),
values of the scanning are evaluated and
results of the evaluation are outputted,
characterized in that
at least one result of the evaluation is outputted in the form of a graphic representation.
The method according to claim 1, wherein the graphic representation is selected from the group of diagrams which comprises a recording of a scanned fancy yam characteristic with respect to the position (x) on a fancy yam (9) or with respect to time, a histogram, preferably in the form of a two- or three-dimensional column chart or bar chart, a scatter diagram, a classification matrix, a surface in a three-dimensional or two-dimensional representation, a column chart, a bar chart, a circular chart, a pie chart, a table and a spectrogram.
The method according to claim 2, wherein the graphic representation is a scatter diagram on which a mass increase (AM) or a diameter increase of slubs (92) is plotted against a slub length (Le).
The method according to claim 2, wherein the graphic representation is a classification matrix, whose horizontal axis is a characteristic of a leading slub (92) of a pair (92, 92') of adjacent slubs and whose vertical axis indicates the same characteristic of a trailing slub (92') of the pair (92,92').
The method according to claim 2, wherein the graphic representation is a classification matrix, whose one axis indicates a slub parameter and whose other axis indicates various steps of slubs.
The method according to claim 2, wherein the graphic representation is a surface which lies in three-dimensional space, whose first dimension indicates a slub length (Le), whose second dimension indicates a mass increase (AM) or a diameter increase of slubs (92), and those third dimension indicates an observed frequency (H) of the respective slub characteristics (Le, AM), or wherein the graphic representation is a projection of such a surface into the plane spanned by a slub length axis and a mass increase axis or a diameter increase axis.
The method according to claim 2, wherein the graphic representation is a column- or bar chart whose columns or bars are allocated to the slubs (92, 92') on the fancy yam (9), and in each
case are composed of at least two parts, and wherein the length or area of a first part indicates a characteristic (Le) of the respective slub (92), and the length or area of a second part indicates a characteristic (Ls) of an adjacent base yam (91).
The method according to one of the preceding claims, wherein different classes (Hill 3) of slubs (92) are identified by way of the graphic representation.
The method according to claim 8, wherein the identified classes (111 -113) of slubs (92) are delimited fi-om one another and/or fi-om outUers
The method according to one of the preceding claims, wherein on evaluation, a measurement curve (M(x)) is produced fi-om the values of the scanning, and an area (A(x)) below the measurement curve (M(x)) is computed.
The method according to claim 10, wherein the area (A(x)) between the measurement curve (M(x)) and a previously detemined base yam mass (Ms) or a previously determined base yam diameter is computed.
The method according to one of the preceding claims, wherein the at least one scanned characteristic is a mass (M) and/or a diameter of the fancy yam (9).
The method according to one of the preceding claims, wherein results of the evaluation are selected fi-om the group of fancy yam parameters which includes a base yam mass (Ms), a base yam diameter, a slub distance (Ls), a mass increase (AM) of a slub (92), and slub diameter increase, a slub diameter, and slub length (Le), a slub total mass (Me), an average yam number, a number of slubs per length unit, a pattern length, a sub-pattem length, a shape and a color.
The method according to of the claim 13, wherein a running number is allocated to each slub (92, 92'), and the running number is stored together with the parameters of the associated slub (92).
The method according to claim 13 or 14, wherein results of the evaluation are minima, maxima, arithmetic means and/or standard deviations of the fancy yam parameters and/or the number of slubs (92) per yam length.
The method according to one of the preceding claims, wherein the evaluation includes a smoothing or idealization of the scanning values.
The method according to claim 16, wherein the evaluation includes a linking of the smoothed or idealized scanning values with the original scanning values.
The method according to claim 17, wherein the linking is a difference formation or a quotient formation.
19. The method according to one of claims 16-18, wherein the smoothing or
idealization contains the approximation by straight stretch sections and slubs (92,92') of
the fancy yam (9) are approximated as trapeziums, triangles or rectangles.
20. The method according to one of the preceding claims, wherein the evaluation
includes a filtering of the scanning values.
21. The method according to one of the claims 17-18 and one of the claims 1-9,
wherein an individual graphic representation for the smoothed or idealized scanning
values and/or an individual graphic representation for the data which has come from the
linking is outputted.

Documents:

http://ipindiaonline.gov.in/patentsearch/GrantedSearch/viewdoc.aspx?id=RfSpucO+SrU3by2M/M6gBg==&loc=egcICQiyoj82NGgGrC5ChA==

« Previous Patent

Next Patent »

Patent Number

277133

Indian Patent Application Number

3065/CHENP/2008

PG Journal Number

47/2016

Publication Date

11-Nov-2016

Grant Date

11-Nov-2016

Date of Filing

18-Jun-2008

Name of Patentee

USTER TECHNOLOGIES AG

Applicant Address

WILSTRASSE 11, CH-8610 USTER,

Inventors:

#	Inventor's Name	Inventor's Address
1	MEIXNER,CHRISTINE,	AM KANAL 17,CH-8636 WALD,
2	EDALAT-POUR, SANDRA,	DIENSBACH 60B CG-8340 HADLIKON,
3	PETERS, GABRIELA	STALDENBACHWEG 12, CH-8320 FEHRLATORF,

PCT International Classification Number

G01N33/36

PCT International Application Number

PCT/CH06/642

PCT International Filing date

2006-11-15

PCT Conventions:

#	PCT Application Number	Date of Convention	Priority Country
1	925/06	2006-06-08	Switzerland
2	1004/06	2006-06-21	Switzerland
3	2059/05	2005-12-23	Switzerland
4	1846/05	2005-11-18	Switzerland