Analyzing the Effect of Various Sizing Machine Settings on Abrasion Resistance and Size Pick-Up of Polyester/Cotton Blend Sized Yarn Using Box-Behnken Design

ABSTRACT The main aim of this study relies on analyzing multiple sizing machine setting on sized yarn pick-up and abrasion resistance using response surface methodology. To perform the experiment, three independent variables each at three levels (wet zone yarn tension (A) (340N, 380N, 420N), squeezing pressure (B) ((13N, 15N, 17N), and sizing machine speed (C) 30 m/min, 49 m/min and 68 m/min)) have been taken. Fifteen experimental runs have been performed using design expert 11 software (with Box-Behnken design). It has been observed from the Analysis of variance (NOVA) results of the experiment that wet zone yarn tension, squeezing pressure, and speed of the sizing machine have a significant effect on size pick-up% and abrasion resistance of the yarn. The interaction effect of BC with (p-value = .0053) and the interaction effect of AC with (p-value = .0143) have also a significant effect on abrasion resistance. Moreover, the interaction effects (AC and BC) have a significant effect on size pick-up. Regression analysis shows that wet-zone yarn tension and squeezing pressure have a negative relation with size pick-up and the speed of the machine has a positive relation. Whereas machine speed and wet-zone yarn tension have a negative relation with abrasion resistance and squeezing pressure has a positive relation.


Introduction
Weaving is the most common method for production of textile materials, and woven fabrics have a far greater market share than non-woven and knit fabrics (Zhao, Helan, and Yang 2017). However, in the last four decades, the nonwoven and knitting segments compute the weaving sector. To improve the performance of weaving one of the most essential preparatory processes involved is sizing of warp yarn.
Sizing is one of the most important processes in weaving preparatory, and it is the process of application of thin films of adhesive on the yarn surface so as to withstand the stress and strain applied on the warp yarn during weaving. Sizing is frequently called the "heart of weaving" because it is not only imparted strength to the yarn but also improves the abrasion resistance of the yarn and without sizing weaving in modern loom is almost impossible (Devare et al. 2016;Turukmane, Gulhane, and Patil 2019). The size material shall be removed easily in the desizing process as the size material is unwanted after weaving process (Sabır and Sarpkaya 2016).
To provide a good quality sized yarn to weaving, many sizing parameters are responsible. The yarn tension, squeezing pressure, hardness of rubber coated roller, speed of sizing machine, viscosity of size material, sow box temperature, depth of immersion roller, and so on are some of the main parameters in the sizing process that determine the quality of sized yarn (Patil, Turukmane, and Raichurkar 2017;Salama et al. 2021;Turukmane, Gulhane, and Patil 2019).
The performance of warp yarns on loom during weaving is affected by several factors as it is exposed to complex deformation including abrasion, impact loading, and cyclic bending together with tension. Controlling yarn's structural characteristics and examining the level of mechanical parameters together with evaluation of yarn's weaving-ability are essential. As clearly known, one of the main objectives of sizing is to enable sized yarn withstand abrasion and friction in loom during shedding and beating of reed. Abrasion resistance of the yarn is the ability of the yarn material to withstand surface wear due to rubbing contact with different machine parts during weaving on the loom. Therefore, abrasion resistance and its measurement for sized yarn can help in the decision of yarn's weaving-ability (Krupincová and Hatipoglu 2013;Walker and Olmstead 1945).
The other factor that affects the weaving performance is the size pick-up. It can be affected by several factors such as sizing pressure, viscosity of the size pate, speed of sizing machine, tension of the yarn, the level of size paste in the size box, depth of the emersion roller, yarn twist, yarn count and types of fiber used. Among these factors, yarn tension, squeezing pressure, and sizing machine speed are considered in this study. The squeezing pressure determines the degree of penetration of the size paste between the fibers in the yarn. It also removes excess amount of size paste and hence determines the level of the size pick-up. In case of higher yarn tension during sizing the set of warp yarns faces a stretch of relatively higher tension and thus the yarn increases in length. If this increase is too high then the elongation property of the yarn will be decreased. So, the yarn will face relatively higher warp yarn breakage in subsequent processes (Mazharul 2013).
Various research has been done regarding the effect of various sizing machine setting on size pickup. However, the methods used in the previous research work didn't consider a multiple factor effect at the same time (interaction effects of two factors on size pick-up) and didn't not show the empirical relationship between the factor and the response. The effect of sizing machine setting on abrasion resistance of warp yarn has not been studied so far with this level with the best of our knowledge.
In the current research the effect of wet zone yarn tension, squeezing pressure, and sizing machine speed on size pick-up and abrasion resistance of cotton yarn is examined using response surface methodology (Box-Behnken design). To investigate the effect of these factors on size pick-up, ANOVA, regression analysis, interaction effect using 3D plot, and each factor's effect on the size pickup and abrasion resistance have been dealt with in detail.

Materials
In this research work, 30/70 polyester/cotton blend yarn produced by a Rotor spinning machine (R923) in Bahir Dar Textile Share Company was used as input material. The properties of un sized yarn are illustrated in Table 1.

Methods
This research work is designed to study the effect of sizing machine settings on the 30/70 polyester/ cotton blended yarn. Karl Mayer-Rotal sizing machine, which is found at Bahir Dar Textile Share Company, was used for production of sized yarn samples. The wet zone yarn tension, machine speed, and squeeze roller pressure were varied in the current research. The other sizing machine variables which could influence the properties of yarns like size liquor concentration, viscosity, temperature, and R.F% have been kept constant.

Experimental design
Design expert 11 software with Response surface methodology and Box-Behnken design has been used to design the experiment, and analyze the results of the experiment. Three factors (wet zone yarn tension, machine speed, and squeezing pressure) each at three levels have been selected and two response variables (size pick-up, and abrasion resistance of yarn) have been analyzed. The selected factors and levels of the factors are shown in Table 2. Fifteen experimental runs have been developed by the software randomly (Table 3) and 15 samples of sized yarn have been produced.
The results of the experiments for each response variable were analyzed by ANOVA at 95% significant level (Iwundu 2015;Montgomery 2017). Size pick-up and abrasion resistance of the sized yarn were analyzed by ANOVA, regression analysis, one factor at a time and 3D plots.

Characterization of dependent variables/responses
Yarn size pick-up. There are typically two ways to determine size pick-up percentage: the lab test technique and the continuous determination method. In lab tests also two methods are available for determination of size pick-up%, i.e., gravimetric test and desizing test method. In the current study gravimetric method was used to determine size pick-up%. The procedure followed to determine the size pick-up% is illustrated as follows.
Dry weight of sized and unsized yarn. The dry weight of the sized and un-sized yarn was determined by using the ETADRY (moisture testing oven) machine according to ISO 6741. 100 cm of yarn from each experiment was first measured and weighed for its mass. The weighed samples were placed in the opened canister. Drying chamber lids was closed and oven switched on for approximately 20 minutes at 105 ± 5ºc. Then the oven was Switched and the sample was re-weighted. Weighing of sample was done repeatedly at an interval of 5 minutes until there was no progressive weight change of the sample greater than 0.05%. The last weight record was taken as the dry weight of the sample. Then the amount of size pick-up% of the yarn was calculated by using the formula given in Equation (1) (Goswami, Anandjiwala, and Hall 2004). Five replicates have been taken for each sample, and the average value was used for analysis.
Size pick À up% ¼ Sized yarn dry weight À Unsized yarn dry weight Unsized yarn dry weight

Yarn abrasion resistance.
Abrasion is wearing of any part of the material by rubbing against another surface. There is no standard for the determination of yarn abrasion resistance, the manufacturer's instructions are used (Kovačević and Gordoš 2009;Kovačević and Penava 2004). In this research, the abrasion resistance of the sized yarn was determined by using the Shirley yarn abrasion tester SDL model. The instrument consists of two reciprocating bars: one is made of hardened steel, and the other is covered with the standard abradant. Yarns are threaded from the fixed holders and clipped onto the flexible holders. The initial tension exerted on each yarn was 0.5N. When a yam breaks, the flexible holder falls, and the number of rubs for that particular yarn is recorded in terms of a number of cycles. Five replicates have been taken for each sample, and the average value was used for analysis.

Result and discussion
To investigate the variation in the sized yarn properties, which could happen from the effect of the three independent variables (squeezing pressure, wet zone tension, and sizing machine speed) on the response variables, 95% confidence level has been used, i.e., if P-value less than 0.05, the factor has a significant impact on the response, while if P-value is higher than 0.05 the factor has no significant effect on the response (Iwundu 2015;Montgomery 2017).

Size pick-up percentage
Size take-up percentage is the amount of size materials added on the warp yarn surface (Kovačević, Schwarz, and Brnada 2008). The key factors affecting the size pick-up studied in this research are the wet zone yarn tension, squeezing pressure, and speed of sizing machine. To investigate the effect of these factors on size pick-up, ANOVA, regression analysis, interaction effect, and each factor's effect on the size pick-up have been dealt with in detail in the following paragraphs. Table 5 shows the model is significant and leaner. The coefficient of determination with R 2 of 0.96 suggests that the analyzed factor explains 96% of the size pick-up for sized yarn. Therefore, a good agreement is achieved between the predicted and actual value of the size pick-up of the yarn.
As shown Table 4, the values of adjusted R 2 and predicted R 2 for yarn size pick-up are 0.94 and 0.86 respectively. The difference between the adjusted R 2 and predicted R 2 is less than 0.2. This indicates a high degree of correlation between the actual and predicted values. Figure 1 also shows the relation between actual and predicted values of size pick-up%. The results shows that it is approximately a straight-line curve obtained by the analysis. It indicates that the actual and predicted values are nearly equal and the error is very less. The lack of fit value is insignificant with the p-value of 0.5983, which points the fact that pure errors such as the experimental error were at their minimum value. Table 5 demonstrates that the linear term of squeezing pressure, sizing machine speed, and wet zone yarn tension all (with p-value <.0001) have a significant effect on the size pick-up%. Moreover, the interaction effects (wet zone yarn tension with sizing machine speed, and squeezing pressure with sizing machine speed) have a significant effect on size pick-up with a p-value of 0.0056 and 0.029 respectively. But the interaction effect of wet zone yarn tension and squeezing pressure has no significant influence on the size pick-up%.
The regression model Equation (2) shows that the wet zone yarn tension, squeezing pressure, and the interaction of squeezing pressure with sizing machine speed have a negative correlation with size  pick-up%. But the linear term of sizing machine speed and the interactions between wet zone yarn tension and sizing machine speed own a positive correlation with the size pick-up of the sized yarn.
Sizepick À up % The effects of the selected sizing machine parameters (wet zone yarn tension, squeezing pressure, and sizing machine speed) on the of size pick-up% are shown in Figure 2. As shown in Figure 2(a), the amount of size pick-up% is influenced by wet zone yarn tension. As the graph indicates while increasing the wet zone yarn tension, size pick-up% decreases. The reason behind this is when the wet zone yarn tension is increased, there is an increase in the length of yarn, which reduces the amount of size material per unit length. Figure 2(b) shows that there is an inverse correlation between the size pick-up of the yarn and the squeezing pressure because high squeezing pressure increases the penetration of the sizing material into the yarn core structure. The penetration of size material into the interstices of the yarn causes the yarn became at saturation level with a minimal amount of size material and reduces the absorptive capacity of the yarn. In addition, high squeezing pressure removes excessive size material from the yarn and this is also in agreement with Kovačević and Gordoš (2009 Figure 2(c), there is a direct correlation between the size pick-up of the yarn and the sizing machine speed. When the sizing machine speed increases, the time for squeezing decreases which removes less size material from the yarn or fast squeezing effect, and the size pick-up% increases which is supported by the former study (Fernando and Jayawardana 2015;Turukmane, Gulhane, and Patil 2019). It can be concluded that low squeezing time has a dominant effect on size pick-up% than low waiting time of the yarn in a size box.

As shown in
The 3D plot shown in Figure 3 illustrates the interaction effects of an independent variable on the amount of size pick-up%. Figure 3(a) shows that the interaction of sizing machine speed and wet zone yarn tension on the size pick-up is significantly more than the others interaction, with the highest p-value (0.0056). The interaction impact of wet zone yarn tension and sizing machine speed shows a negative ratio. Which means to obtain the optimum pick-up, one factor shall be increased while the other factor decreases. When wet zone yarn tension is at 340N and speed of sizing machine is 68 m/ min, with constant squeezing pressure the size pick-up reaches the maximum level. Figure 3(b) and Table 5 show that the interaction influence of squeezing pressure and sizing machine speed on the size pick-up is significant with a p-value of 0.029. Similar to the previous interaction effect it has also a negative ratio effect on size pick-up. The maximum amount size pick-up is obtained at a high speed of sizing machine and lowest level of squeezing pressure in the given range. In this case, the penetration is low and the coating rate is high which is also in agreement with Fernando and Jayawardana 2015;Jia and Zhang (2010). At a high speed of sizing machine, the time for squeezing is low which results in low removal of the size material from the yarn and more size material remain on the yarn. At 13KN squeezing pressure and 68 m/min sizing machine speed and at a constant wet zone tension of 380N, the size pick-up reaches a maximum value.

Yarn abrasion resistance
Abrasion resistance of the yarn is the ability of the yarn material to withstand surface wear due to rubbing contact with different machine parts during weaving on the loom (Friedman et al. 1989;Özdil, Özçelik Kayseri, and Süpüren Mengüç 2012). The end breakage of warp yarn is greatly dependent on the abrasion resistance properties of the sized yarn (Krupincová and Hatipoglu 2013). In the current research, the effect of wet zone yarn tension, squeezing pressure, and speed of sizing machine on abrasion resistance of sized yarn are analyzed. As observed from Table 6, the model is significant with p-value of 0.0039. The lack of fit value was insignificant (p = .4018), which points to the fact that pure errors such as the experimental error are at their minimum values. Fit statistic in Table 4 indicates that the values of adjusted R 2 and predicted R 2 for yarn abrasion resistance are found to be 0.90 and 0.74 respectively. This indicates a high degree of correlation between the actual and predicted values. And also, Figure 4 shows the relation between actual and predicted values of abrasion resistance of the sized yarn. The results show that all the actual and predicted values of abrasion resistance are fitted to the straight line, showing that the error is very less. Table 5 shows the linear term of wet zone yarn tension with (p-value = .0006) has the largest effect on the yarn abrasion resistance, followed by the interaction effect of squeezing pressure and sizing machine speed (p-value = .0053), quadratic term of wet zone yarn tension (p-value = .0086), the interaction effect of wet zone yarn tension and sizing machine speed (p-value = .0143), and linear term of squeezing pressure and sizing machine speed with p-values of 0.0132 and 0.0209 respectively. The quadratic regression model illustrated in Equation (3) shows that the wet zone yarn tension, sizing machine speed, quadratic term of wet zone yarn tension, and the interaction of squeezing pressure and sizing machine speed have a negative correlation with the yarn abrasion resistance. But the squeezing pressure has a positive correlation with the sized yarn abrasion resistance.
Based on the coefficient of the independent variables of regression equation the following can be claimed: (i) For each unit increase in the number of wet zone yarn tension, with other independent variables being constant, the yarn abrasion resistance is reduced by 6.63%. (ii) For each unit force increase in the squeezing pressure, with other independent variables being constant, the yarn abrasion  resistance is increased by 3.25%. (iii) For each unit of sizing machine speed increase in the sizing process, with other independent variables being constant, the yarn abrasion resistance is reduced by 2.88% and the other relation can be explained in similar manner from Equation (3).
The individual effects of the selected sizing machine parameters (wet zone yarn tension, squeezing pressure, and sizing machine speed) on the abrasion resistance are shown in Figure 5. As shown in Figure 5(a), the abrasion resistance of the sized yarn is influenced by wet zone yarn tension. As the graph indicates that while the wet zone yarn tension increases the abrasion resistance decreases. This is because when increasing the wet zone yarn tension, warp imposes comparatively higher stretch, thus increasing yarn length. Due to this the sized material moves from inside to outside due to stretching and compactness of the yarn (Turukmane, Gulhane, and Patil 2019). This moves the size material at the outer surface of the yarn which can be removed easily by abrader within a minimum number of cycles which results in yarn breaking. This will cause high-end breakage in weaving as it has a low resistance to friction during weaving. Figure 5(b) shows there is a direct correlation between the abrasion resistance of sized yarn and the squeezing pressure. When a squeezing pressure increases, the penetration of the sizing material into the yarn increases (Hari et al. 1989). Size penetration has a dual role: it provides inter-fiber binding and an anchorage for the size coating (Maatoug, Ladhari, and Sakli 2007), which improves the abrasion resistance ability of the sized yarn. Figure 5(c) confirms that when increasing sizing machine speed, the abrasion resistance of the sized yarn is decreased and vice versa. During a sizing process when increasing the sizing machine speed, the amount of size material picked increases. The reason for this is as the speed of the machine increases the squeezing time becomes shorter and excessive amount of size material remains on the yarn surface. However, the penetration of size material into the yarn core reduces, leaving the size material on the surface of the yarn which is also in agreement with (Fernando and Jayawardana 2015) which leads the yarn to exhibit low resistance of abrasion.
NB: AR=Abrasion resistance The interaction effects of the three independent variables (wet zone yarn tension, squeezing pressure, and sizing machine speed) on sized yarn abrasion resistance are shown in Figure 6. Figure 6(a) shows the interaction effect of sizing machine speed and wet zone yarn tension which has a significant effect on yarn abrasion resistance. The influence of wet zone tension on yarn abrasion resistance property decreases when increasing wet zone tension which is also supported by previous work (Turukmane, Gulhane, and Patil 2019), while the sizing machine speed is relatively flat. When wet zone yarn tension is 34N and sizing machine speed is 30 m/min, with a constant squeezing pressure of 15KN, the maximum number of yarn abrasion resistance cycles is obtained. Figure 6(b) approves that the interaction effect of squeezing pressure and sizing machine speed has a significant effect on yarn abrasion resistance. It can be observed that yarn abrasion resistance increases with decreasing sizing machine speed and an increase of squeezing pressure. The reason for this is at low speed of sizing machine the time for squeezing increases and excess amount of size material is removed. At the same time, at higher squeezing pressure of sizing machine the penetration of size material into the yarn increases and the size material at the outer surface is removed. The simultaneous effect of sizing machine speed and squeezing pressure affects the abrasion resistance of the warp yarn positively (i.e., improves the abrasion resistance of the yarn). When squeezing pressure is 17KN and sizing machine speed is 30 m/min, at a constant wet zone yarn tension of 380N, the yarn abrasion resistance reaches a maximum number of the cycle and vise versa.

Conclusion
The sizing process determines the sized yarn quality and to attain the maximum weaving efficiency, excellent sizing process is required. The result of this study shows that as the wet zone yarn tension increased, size pick-up and abrasion resistance of the yarn decreased. Another finding observed is when squeezing pressure increases, the abrasion resistance increased significantly, whereas the size pick-up% reduced. Furthermore, at higher level of sizing machine speed (68 m/min) the size pick-up increased significantly and the abrasion resistance decreased.
Besides the single factor effect, the interaction effects also show a significant effect on the sized yarn quality. The analysis also revealed that the interaction effect of squeezing pressure and sizing machine speed highly influence the abrasion resistance of sized yarn.
The regression model equations also reveals that wet zone yarn tension and squeezing pressure have a negative correlation with size pick-up% and the sizing machine speed has a positive correlation with the size pick-up of the sized yarn. On the other hand, wet zone yarn tension and sizing machine speed have a negative correlation with the yarn abrasion resistance, whereas the squeezing pressure has a positive correlation with the sized yarn abrasion resistance.
The maximum value of size pick-up is attained at 13KN squeezing pressure and 68 m/min sizing machine speed with a constant wet zone tension of 380N and vice versa. At 340N wet zone yarn tension and 30 m/min sizing machine speed with a constant squeezing pressure of 15KN, the maximum number of yarn abrasion resistance cycles is obtained.

Highlights
• The interaction effects of the factors have also a significant effect on the size pick-up and abrasion resistance of the sized yarn. • It has been observed from the NOVA results of the experiment that wet zone yarn tension, squeezing pressure, and speed of the sizing machine have a significant effect on size pick-up and abrasion resistance of the yarn. • The regression model equations revealed that wet zone yarn tension and squeezing pressure have a negative correlation with size pick-up% and the sizing machine speed has a positive correlation with the size pick-up of the sized yarn. • On the other hand, wet zone yarn tension and sizing machine speed have a negative correlation with the yarn abrasion resistance, whereas the squeezing pressure has a positive correlation with the sized yarn abrasion resistance.