Evaluation of real yarn diameter Measuring yarn diameter, yarn compressibility and their effect on fabric quality is important

to both the fabric designer and the textile technologist, say Mounir Jaouadi, Slah Msehli and Faouzi Sakli.

In a yarn structure, fibres represent the main component. The other component is air pockets created by the technology forming the structure. The knowledge of the real fibres volume within the yarn allows to better estimate various textile fabrics properties. Yarn diameter is used to predict fabric structural parameters such as width, cover factor, porosity, fabric comfort and so on. Since thousands of ends or picks are presented side-by-side in the woven or the knit fabrics, a slight change in yarn diameter can result in a substantial change in the overall fabric cover factor. Measuring yarn diameter and compressibility and their effect on fabric quality is important to both the fabric designer and the textile technologist. In fact, the dimensional and mechanical characteristics of fabrics are dependent on yarn diameter, thread spacing of warp and weft in woven fabrics, courses and wales per unit length and stitch length in knitted goods. Many researchers studied the yarn diameter and they estimated it using empirical formula. One of the most commonly used expressions for estimating yarn diameter is that developed by Peirce in 1937 [7]. In this expression, yarn density was assumed to be 1.1 g/cm3. In a recent study, Elmoghazy et al in 1993 [5] developed empirical expressions for estimating the diameters of ring-spun, rotor-spun and MJS air-jet spun yarns. These expressions (Table 1) were developed by an extensive microscopic testing of actual yarn thickness for a wide range of yarn count and twist levels. In 2003 A Basu et al (SITRA) [6] measured the rotor spun cotton yarn diameter using the image analysis method. The used yarns count range from 6 Ne to 20 Ne and they found that the actual yarn diameter can be predicted by using the formula shown in Table 1. The formulae shown in Table 1 indicate that yarns made by different spinning systems and of equal nominal count will exhibit different values of yarn diameter. This is a result of the difference in fibre arrangement and fibre compactness of different yarn types. Table 1 Empirical formulae of yarn diameter Yarn type Ring Spun Expression Units inch Source Peirce (1937)

Ring Spun Rotor Spun MJS air jet yarn Rotor Spun

mm mm mm mm

El Mogahzy (1993) El Mogahzy (1993) El Moghazy (1993) SITRA (2003)

We should point out that the formula for ring spun yarn developed by El Moghazy in 1993 [5] tends to produce a value of yarn diameter that is slightly higher than that estimated by Peirce equation. The main reason for the difference was due to discrepancy in the value of yarn density, particularly in the coarse to medium range of yarn count. Using a combination of capacitive and optical measures of different yarns, the density of cotton ring-spun yarns can range from 0.85 to 1.2 g/cm3 depending on the spinning process, fibre characteristics, and structural parameters (count and twist). The majority of the studies were interested in the determination of the apparent yarn diameter However, the knowledge of the real fibres volume within the yarn allows to determine the real yarn diameter. Thus, the real diameter is defined as the measurement of the yarn dimension without air. Indeed, the determination of the real diameter influences so much the fabric geometrical and mechanical properties. In this paper, the authors discuss the development of an experimental device allowing the determination of the real yarn diameter and study the effect of yarn and fibre properties on the yarn packing fraction. Material and method Sixty one yarns of different material (polyester, cotton, acrylic, viscose and wool) and of fineness (Nm) ranging from 10 and 68 were tested. These yarns were randomly collected and were manufactured by two different processes: 33 Open-end yarns and 28 ring spun yarns. The yarns and fibres properties are summarised in Table 2. Table 2 Yarns and fibres properties.

Yarns and fibers properties

min max

Fiber micronnair (g/inch) 3 ,29 9,87 Fibers length (mm) 26 65 Yarn count (Nm) 10 68 Yarn irregularity CV (%) 8,65 13 ,42 Yarn twist T (tr/m) 200 936

From this brief review, the authors noticed that all researchers into the evaluation of yarn diameter have mainly accentuated the importance of yarn count. They also noted that the apparent yarn is in inverse proportion to square root of yarn count. The twist variation is proportional to the yarn count one, so they can conclude that there is a relationship between the yarn diameter variation and the twist one. In order to evaluate this variation, they developed an experimental device made up of a twist tester, a camera and a computer with image processing software. Figure 1 represents the used experimental device. The yarn sample undergoes a uniform tension while mounting on the grips. The tension of 1 cN/ tex was applied to the yarn sample according to the French Standard NF G07-079 [3] for removal of crimp. The first yarn sample edge was fixed on one grip and they applied the pretension and they fixed the second edge on the other grip. 500 mm test length yarn is maintained right between the grips of the twist tester and subjected to twist increase (a step of 50 turns). For each twist step, a yarn photo is captured and examined by a Motic microscope with 40x magnification. For analysis of the yarn images, "Image-Pro-Plus 2.0" software was used. The image analysis was performed using the sequence: process, image, acquise, segmentation, processing and measurement. With regard to the external structure, the yarn can be divided into two basic elements: the yarn core and hairiness. The core is defined as a part of the yarn that forms a compact agglomeration of fibres. It has a cylindrical form and a variable diameter. On its surface the fibres are laid along the curves characterising the twist. The rest of the yarn, consisting outlying fibres or their agglomerations, constitutes hairiness. So the yarn diameter can evaluated by determined the edges of the yarn core. For determination of them, the images were processed (Figure 2) and the yarn diameter was measured. Average of 10 measurements were taken in 10 different positions throughout the yarn length. The initial study of variation it was decided to take 100 readings. So the average of yarn diameter was calculated by the following equation [1].

(1) where: `d : average diameter in mm Li : diameter of yarn at position i in mm A statistical study was carried out in order to determine the more influent parameters on yarn packing fraction. This study allowed us, using Principal Components Analysis (PCA) [2] and contribution analysis (Path method) [4], to limit the number of factors. The principal component analysis (PCA) is an extraction method of the principal factors based on a quantitative analysis of the correlations [2]. Its goal is to study and to reduce the survey space of variables in order to simplify the raw data, to find out (graphically) some links and to identify some macro features

(principal components). This method consists on subjective grouping of the correlated variables, which were represented graphically by the coordinates corresponding to two centred and reduced principal variables. Theoretical study In the spinning process, the yarn generally undergoes a twist, which consists of rolling fibres in propellers so that the initially parallel fibres are helicoidally placed around the yarn axis. Twist has a significant role in the yarn density variation. Indeed, the yarn volume is occupied by the fibres volume and the air one. The expression indicates that the value of yarn diameter mainly depends on the linear density and the volumetric density of yarn. As indicated earlier, volumetric density describes the degree of compactness of fibres in the yarn structure. This means that yarn twist will have a significant effect on yarn diameter. More the twist is important more the radial forces exert a compressibility effect on fibres and the air volume will be increasingly negligible and in consequence, the yarn density "?yarn" tends to the fibres density "?fibre" and the apparent diameter tends to a limit noted real diameter (Figures 3 and 4). In order to validate the experimental device results, the authors compare them with the calculated ones drc by using the following theoretical model:

(2) Where: : theoretical real diameter : Fibre density (g/cm3) - Nm:Yarn metric count

This formula was established by supposing that the yarn is cylindrical. Indeed, the twist increase generates an increase in the radial forces applied to fibres with the same intensity in all directions resulting a circular section of the tested yarn. The knowledge of the real yarn diameter allows estimating its packing fraction "F", defined by the following equation:

(3)

with: -V f iber: fibers volume in yarn : yarn density - dm: yarn real diameter - dyarn : yarn apparent diameter. - Vyarn: yarn volume (fibers + air) - : fiber density

Results and discussion The yarn diameter variations versus the twist increase are shown in Figures 5 and 6. For different yarn counts (case of ring yarn cotton), the apparent yarn diameter decreases with the twist increase (Figure 5). The apparent diameter decreases until reaching a limit value known as real yarn diameter. This limit can be explained by the fact that from a certain twist value, the fibres approach the ones to the others in a maximum interference configuration, resulting in free air volume reduction. Thus, the measured diameter corresponds to the fibres volume, therefore to the real yarn diameter. This phenomenon was noticed for the other yarns having same count, Nm 10, with different materials (Figure 6) and for different spinning processes: Rotor spun and ring spun (Figure 7). Effect of spinning process on yarn diameter Figure 7 shows that for yarns with same yarn linear density and material, yarn diameter was greater for rotor spun process than ring spun one. This phenomena was observed to a limit twist value from wich the two yarn diameters tend to the same value since the yarn density tends to the fibre one. Different spinning processes produce different degrees of yarn density as a result of the different patterns of fibre compactness imposed by yarn twisting and spinning process. For instance, a ring-spun yarn will typically exhibit higher degree of compactness than a comparable rotor-spun yarn due to the true twist with the high tension used in ring spinning. The higher value of rotor-spun yarn diameter indicates that it is bulkier than the ring-spun yarn. Comparison of experimental and theoretical methods Figures 8 and 9 present a comparison of the real yarn diameter given by the microscopic method noted (dm) to the theoretical one (drc). For the two ring and OE processes, the measured error % (= [experimental value - theoretical value] / theoretical value %) is small, it is respectively about 5.33 % and 6.1%. The experimental measurements are within the confidence interval of the theoretical values. For all tested yarns, the average real diameter error determined by the microscopic method compared to the theoretical model is 4.82 %. Yarn packing fraction The packing fraction is an indication of the air pockets enclosed by the fibres. Most of spun cotton yarns have packing fraction above 0.5 [5]. This result has been confirmed by the authors study and for the other tested materials, packing fractions are summarised in Table 3. The importance of packing fraction lies in its powerful effects on many yarn and fabric properties. It is one of the major design parameters of textile fabrics. For a given fibre material, a yarn of very high packing fraction is likely to be stiff and probably weak. And for a yarn with

very low packing fraction, it is likely to lose the bulk and the surface integrity required to hold the yarn structure during processing. Statistical analysis of packing fraction variation The application of the principal component analysis method allows obtaining the graph shown in Figure 10. The two principal vectors (Figure 10) have an important weight with a ratio of 69.4 %. The correlated parameters are grouped in small groups: - The twist, the yarn count and the yarn irregularity are correlated. The yarn count is proportional to the twist. The yarn irregularity is important for the thin yarn. - The material and the fibre properties are positively correlated. - The spinning process has not any correlation with the others parameters But, this method doesn't allow judging the internal dispersion of each group. Thus, it is necessary to analyse the database by the method of contribution of the different factors. The statistical determination of the contribution coefficients of the different factors allows studying the internal dispersion of each group determined by the principal component analysis. Besides, the aim of this method is to reduce the number of factors and to eliminate the low influence parameters [8,9]. In order to determine the contribution of each factor on the studied properties, the analysed data had to be centred and reduced. Then, the standardised regression coefficients have been calculated in accordance with the Path analysis [4], which is the multiple regression application. The contribution of the variables (Xi) according to the yarn packing fraction (Y), is based on a contribution law obtained under the following shape: Y = a1 X1 + + aj Xj + + aq Xq (4) Where: Y : Yarn packing fraction ai : Regression coefficient of the normalised factor Xi Xi : Centred and reduced variables The contribution coefficient Ci of factor Xi is calculated by the following formula: (5) Where: R2(adjusted): Correlation coefficient of the normalised regression Y versus Xi So, the authors obtained the contribution factors corresponding to the different parameters, which are calculated for the yarn packing fraction, as shown in Figure 11. This method confirms the PCA results. Indeed, the most influent parameters on yarn packing fraction are the count and the twist yarn. In fact, they contribute with 65% in yarn packing fraction prevision, whereas, the fibre properties, the material and the yarn irregularity have a

lower influence. So, the effect of fibre and yarn properties can be summarised in the Table 4: In the context of fibre to yarn engineering, yarn diameter is a major design criterion. Factors affecting real yarn diameter are essentially those that affect yarn density or fibre compactness. As indicated earlier, fibre properties that are expected to influence fibre compactness include: fibre fineness, fibre stiffness, fibre length and fibre crimp. In general, coarse and stiff fibre will result in bulkier or thicker yarn than fine and flexible fibres. In other words, as the fibre becomes coarser (higher denier), yarn density becomes smaller, leading to an increase in a real yarn diameter, although the count of yarn remains unchanged. On the same ground, fibre length also affects yarn density and consequently yarn diameter. For a given yarn count and at the same twist factor, the larger the fibre length, the higher the yarn density, and the smaller the yarn diameter. So, these parameters (the count and the twist yarn) will be used to estimate the yarn packing fraction. This prevision is based on the determination of the apparent diameter by microscopical method and the real one by using the following linear model. dm = 0,320 - 8 10-4 T - 0,00265 Nm ; R2 = 78,6 % (6) where : dm: real yarn diameter (mm) T: yant twist (tr/m) Nm: yarn count (m/g) In order to test the significance of the model, the authors used the analysis of variance method. The analysis of variance in Table V shows that the regression model is highly meaningful at 5% significance level (high Fisher constant F value). Conclusion An enumeration of various research revealed that all the studies were interested in the determination of the apparent diameter and not of the real diameter. The objective of this work is to determine the real diameter and the packing fraction of textile yarns with determining the more influential intrinsic yarn parameters. In this study, the authors developed an experimental device measuring the real yarn diameter according to the twist variation. The twist increase results in a free air volume reduction to a null value. So the apparent diameter tends to a limit value, the real diameter. The experimental method validation results gave low error values, which can be due to the tested yarns irregularities. A statistical study allowed the authors, by using the principal components analysis and the contribution method (Path method) to retain the most influential factors in the yarn packing fraction prevision: the yarn count and the twist yarn. References 1. A SEYAM and Aly El-SHIEKH A: Mechanics of Woven Fabrics, Part III: Critical

2. 3. 4. 5. 6. 7. 8. 9.

Review of Weavability Limit Studies, Textile Research Journal, Volume 63 (7), pp 371-378, 1993. Lagarde P: Initiation l'analyse des donnes, Dunod, 1995. La norme Franaise : Mthode de mesure de la torsion du fil NFG07-079. M Sergent, D Mathieu, R Phan-Tan-Luu et G Drava, Correct and Incorrect Use of Multi-linear Regression, Chem and Intel Laboratory Systems, 1995, 27, pp 153-162. Yehia E.Elmoghazy: Understanding the Fibre-to-yarn System: Yarn Characteristics, 1993, pp 89-116. A Basu, I Doraiswamy and R L Gotipamul: Meaurement of Yarn Diameter and Twist by Image Analysis, Journal of Textile Institute 2003, Volume 94, Part 1, pp 47-58 Peirce F T: The Geometry of Cloth Structure, Journal of Textile Institute 1937, 28 T 45 Jambu M: Mthode de base de l'analyse des donnes, Edition 1999, pp 431. Legendre P: l'Analyse Statistiques Multivarie dans Cartes et Donnes, dcembre 1999.

Note: For detailed version of this article please refer the print version of The Indian Textile Journal September 2007 issue. Mounir Jaouadi, Slah Msehli and Faouzi Sakli. Textile Research Unit of ISET Ksar Hellal, B P 68 Ksar Hellal 5070, Tunisia. Tel: (+216) 73 475 900. E-mail: [email protected].