Spirality of knitted fabric is obtained when the wale is not perpendicular to the course, forming an angle of spirality with vertical direction of the fabric. It affects particularly single jersey fabrics and presents a serious problem during garment confection and use. The T-shirt production, for example, suffers from many quality problems linked to fabric spirality such as mismatched patterns, sewing difficulties, displacements of side seams to the back and front of the body, and garment distortion. Spirality has an evident influence on garment aesthetics.

The spirality phenomenon concerns essentially unbalanced structures such as single jersey fabrics. The symmetry of rib structures reduces considerably the spirality. In literature[1,2,3], it has been demonstrated that spirality is due to the relaxation of torsional stresses in the yarn. Unset yarns under low tensile loads have a tendency to return to their untwisted state.

Mainly two methods for determining the spirality of knitted fabric are available in literature: The manual method [3] and the theoretical method [1]. The manual method consists to measure manually the spirality angle on a real fabric by using a protractor. This method presents some difficulties such as wales and courses deformation during measurement and depends on human precision. The theoretical method permits to calculate fabric spirality from fabric and machine parameters (number of feeders on the knitting machine, loop’s length and number of courses and number of wales per fabric unit length). The large number of measured parameters increases the number of error sources and affects the reproducibility of this method.

In literature, several studies aimed to measure weft and warp yarn directions of woven fabrics by using image processing techniques [4,5]. All the studies focused on the measurement of several woven fabric properties such as weft and warp yarn placement, fabric count, weave pattern, woven fabric skewness, etc. Although several methods have been used to determine skewness on woven fabrics, studies which specifically determine the spirality of knitted fabrics are extremely rare. In general, image analysis of knitted fabrics involve difficulties due to the loop structures and yarn hairiness, compared to woven fabrics consisting of neat warp and weft yarns.

Some research works investigated the effect of yarn on spirality of single jersey fabric. Jiang et al [3] observed a linear relation between the spirality of single jersey fabrics produced with a single feeder laboratory knitting machine and the yarn twist factor. Lord et al [6] demonstrated that magnitude of spirality in plain knitted fabrics produced from cotton/polyester yarn increased with increasing the percentage of polyester in the blend yarn. De Araujo and Smith [2] investigated the spirality on single jersey fabrics. They compared spirality of 100% cotton and 50/50 cotton/polyester blend yarns produced with different spinning techniques such as ring, rotor, friction and jet air and observed that yarn structure has an obvious effect on the spirality behaviour of knitted fabrics.

In a previous study [1], the authors demonstrated theoretically that the number of feeders in a circular knitting machine influences the spirality angle. The analysis was based on simulations using theoretical formulas for spirality angle calculations. These observations were not validated by an experimental study. Other parameters such as yarn tension during knitting can be suspected to have an influence on fabric spirality. They were not evoked in literature.

Several studies [1,3,7] investigated the effect of fabric relaxation on spirality. Relaxation was based on simple dry relaxation in conditioned atmosphere laboratory and washing combined with tumble drying. Curiously, all relaxation treatments were performed in laboratory conditions and never in real finishing conditions, although different studies have emphasised the importance of working in commercially produced and finished specimens for reliably predicting the distortion and dimensional properties of knitted fabric [8,9].

We propose to develop a novel test method permitting to measure spirality angle of cotton plain knitted fabric by using image processing technique. The effect of yarn, fabric and machine parameters such as yarn twist factor, yarn tension, loop length and number of feeders of the knitting machine on fabric spirality was studied. The influence of the finishing process on spirality behaviour of commercially produced fabrics was also investigated.

Materials & methods

For the measurement of spirality angle, plain knitted fabric samples having 10x10 cm dimensions were prepared. Digital photographs were taken by a digital optical microscope using a software (Motic images plus 2.0) permitting to acquire and save images with 10 to 40 times magnifying. These images were taken from the back side of the plain knitted fabric because stitch edges from this side were easier to distinguish than those from right side (Figure 1).

Right side Back side

Figure 1. Aspect of right and back side of plain knitted fabric.

The images were then treated by Microsoft Visual Basic VB 6.0 software. The image processing consists to improve the image quality (brightness, contrast) and to draw two lines by clicking on four points belonging to a wale edges and calculate the mean straight line equation (Figure 2).

Figure 2. Microsoft Visual Basic programme for spirality angle measurement.

The same procedure is applied to a fabric course. The two mean straight lines equations allow the determination of the spirality angle ?. All measurements were performed under standard textile testing conditions of 21°C ± 1°C, and 65% ± 2% relative humidity. No tension was applied to samples under microscope. Seven specimen of each sample were tested and the mean spirality angle and the corresponding CV% were calculated. We produced a series of 16 cotton plain knitted fabric (100% combed cotton yarn) commonly used in the clothing industry by using an industrial single jersey circular knitting machine. (Diameter = 23 inch, gauge = 24, total number of feeders = 74).

The influence of four parameters linked to yarn, fabric structure and machine were studied. The knitted specimen covered a large range of yarn twists, yarn tension, loop’s length and feeders density. When studying one parameter the three other ones were kept constant. Yarn tension was measured by using an electronic tensiometer and loop length was obtained by using a yarn debimeter. Feeder density corresponds to the number of feeders per inch of machine’s diameter. For the variation of this parameter, we performed specific machine setting by using interchangeable miss cams in order to be able to cancel some feeders and to vary progressively the number of working feeders on the single jersey machine.

Results

Average spirality angles obtained in the different knitting conditions are summarised in Table 1.

Table 1. Knitting conditions and corresponding spirality angles.

Figure 3 shows plot of averages spirality angle versus yarn twist. It appears that spirality increases linearly with the yarn twist over the range tested. The correlation coefficient for linear association between fabric spirality and yarn twist is very high (correlation coefficient R = 0.99). This shows the existence of a strong linear dependence between fabric spirality and yarn twist when other variables such as stitch length, yarn tension and number of feeders are held constant.

Figure 3. Variation of spirality angle with Yarn twist.

shows plot of averages spirality angle versus stitch length. The degree of fabric spirality increases linearly with stitch length. The relationship between fabric spirality and stitch length is again strong (correlation coefficient R = 0.98).

Variation of spirality angle with stitch length. Relationship between fabric spirality and yarn tension during knitting is shown in Figure 5. Fabric spirality decreases linearly with yarn tension with a quite strong (correlation coefficient R = 0.97).

Variation of spirality angle with yarn tension. shows the influence of the number of feeders and fabric spirality. Linear correlation cannot be tested in this case since the number of feeders is not a continuous variable, but spirality increases strongly when increasing the number of working feeders on the machine.

Variation of spirality angle with number of feeders. A very common finishing process, typically used for fine gauge cotton knitted structures was applied to the fabric presented in the first line of Table 1. First, the plain knitted fabric was washed and dyed. After squeezing, the fabric was dried and relaxed by using a tumbler drier. Finally, the fabric was stabilised and ironed by using a tubular compactor. Figures 7 show the fabric aspect before and after finishing. Corresponding average spirality angle are presented in Table 2. We can easily observe that finishing reduces fabric spirality. The fabric shrank and stitch wales were straightened.

Before finishing After finishing

 Fabric aspect before and after finishing.

Table 2. Spirality angles for finished and unfinished fabrics.

Discussion

The primary purpose of this study was to develop a new method for spirality measurement based on image processing technique. The developed method permitted to measure spirality angle with reasonable CV%. During measurement, the fabric is not handled and then not deformed. The manual method evoked in literature leads to very high CV% that can reach 32% [3], probably because of wales and courses deformation during measurement. Results obtained with the theoretical method developed by De Araujo and Smith [1] has unfortunately not been compared to any experimental results.

The second purpose of this study was to analyse the influence of some fabric constructional parameters as well as the impact of finishing process on spirality behaviour of commercially produced plain knitted fabrics. The strong linear dependence obtained between fabric spirality and yarn twist show that the main source of spirality is yarn twist. When a twisted yarn is knitted into a loop, it will have a tendency to rotate inside the fabric in order to release its torsional strain during relaxation. Similar observations have been formulated by Jiang et al [3] and De Araujo and Smith [1] but they concerned respectively fabrics made on very low diameter laboratory knitting machine having a single feeder.

Stitch length expresses the tightness of knitting construction. The fabric is as tight as stitch length is low. The observed proportionality between fabric spirality and stitch length can be explained by the fact that compared to tight fabrics, slack fabrics have higher stitch length and then the yarn composing the loop has a higher tendency to rotate inside the fabric after relaxation. In literature [3] this phenomenon was explained in terms of the ease of freedom of the loop movement in knitted fabric construction. In a more tightly knitted fabric, the movement of a knitted loop is restricted, and thus spirality is reduced.

The influence of yarn tension during knitting on fabric spirality has not been studied in literature. The observed linear dependence between these two parameters is linked to yarn deformation. During knitting, yarn undergoes an important tension. At high tensions, the viscoelastic nature of the yarn causes yarn fibres to slip inside the structure. This slippage straightens fibres and reduce yarn twist and then yarn tendency to rotate inside the fabric after relaxation. This explains why at high yarn tensions, the fabric spirality is reduced.

The observed increase of fabric spirality with the number of knitting feeders at a constant machine diameter is due to the nature of weft circular knitting. A fabric course knitted in a given feeder has to be inclined with a certain angle in order to permit the knock over of the row of stitches knitted in the following feeder. This angle depends on the number of feeders per machine diameter as can be seen in Figure 8. This confirms the results obtained with simulations using theoretical formulas for spirality angle calculations described in literature [1]. The increase of the feeder density in circular knitting machines is the subject of high competition between machines manufacturers because of its impact on machines productivity. Mayer & Cie holds the record in this matter with the single jersey machine Relanit 4.0 which has 4 feeders per inch of machine diameter [10]. These technological advances will certainly increase the importance of fabric quality problems linked to spirality.

Effect of the number of feeders on fabric spirality. The observed reduction of fabric spirality after a typical cotton finishing process is due two main obligatory operations: Squeezing and compacting. During wet treatments (ie, washing and dyeing) fabric is relaxed and fabric spirality increases, but during squeezing, stitch wales are straightened thanks to the air injection device equipping the squeezing machine. Fabric compacting contribute also to the reduction of spirality by correcting mechanically wale direction and fixing this correction with a thermal treatment. Finishing reduces fabric spirality but a residual spirality angle always remains. Generally a spirality angle under 4° is tolerated before garment confection. It is then important to reduce fabric spirality from knitting process in order to make spirality correction during finishing possible.

Conclusion

In the present work, the authors developed a new method for fabric spirality measurement based on image processing. The interest of this method is the reduction of operator fabric handling during measurement which is generally an important source of error.

The paper presents also an experimental investigation of the effect of different parameters linked to yarn, structures and machine on the tendency of a cotton plain knitted fabric to spiral. The study has revealed that yarn twist, fabric tightness, yarn tension and feeder density contribute to fabric spirality.

The authors demonstrated that typical finishing process of knitted cotton fabric in industrial conditions reduces spirality. Further work will focus on the quantification of the contribution of each finishing step to fabric spirality.This would permit to determine the finishing process aptitude to correct fabric spirality.

References

1. De Araujo M D and Smith G W: Spirality of Knitted Fabrics, Part I : The Nature of Spirality, Textile Res J 59, 247-256 (1989).

2. De Araujo M D and Smith G W: Spirality of Knitted Fabrics, Part II : The Effect of Yarn Spinning Technology on Spirality, Textile Res J 59, 350-355 (1989).

3. Jiang T, Dhingra R C, Chan C K and Abbas M S: Effect of Yarn and Fabric Construction on Spirality of Cotton single Jersey Fabrics, Textile Res J, 67, 57-68 (1997).

4. Kang T J, Choi S H, Kim S M and Oh K W: Automatic Structure Analysis and Objective Evaluation of Woven Fabric Using Image Analysis, Textile Res J, 71, 261-270 (2001).

5. Ravandi S A H and Toriumi K: Fourier Transform Analysis of Plain Weave Appearance, Textile Res J, 65, 676-683 (1995).

6. Lord P R, Mohamed M H and Ajgaonkar D B: The Performance of Open-End, Twistless, and Ring Yarns in Weft Knitted fabrics, Textile Res J, 44, 405-414 (1974).

7. Anand S C, Brown K S M, Higgins L G, Holmes D A, Hall M E and Conrad D: Effect of Laundering on the Dimensional Distortion of knitted Fabrics, AUTEX Res J, 2, 85-100 (2002).

8. Heap S A, Greenwood P F, Leah R D, Eaton J T, Stevens J C and Keher P: Prediction of Finished Weight and Shrinkage of Cotton Knits – The Starfish Project, Part I: Introduction and General Overview, Textile Res J, 53, 109-119 (1983).

9. Heap S A, Greenwood P F, Leah R D Eaton J T, Stevens J C and Keher P: Prediction of Finished Weight and Shrinkage of Cotton Knits – The Starfish Project, Part II: Shrinkage and the Reference State. Textile Res J, 55, 211-222 (1985).

10. Mowbray J: Machinery Trends at ITMA. Knitting International 110, 36-40 (2003).

Note: For detailed version of this article please refer the print version of The Indian Textile Journal November 2008 issue.

Saber Ben Abdessalem,
Technology High School of Ksar Hellal,
Textile Research Unit, Tunisia.

Saber Elmarzougui,
Technology High School of Ksar Hellal,
Textile Research Unit, Tunisia.

Sofiene Mokhtar
Technology High School of Ksar Hellal,
Textile Research Unit, Tunisia.

Heni Riadh
National Engineering School of Monastir,
Textile Department, Tunisia.