**7. Results and discussion**

Tightly woven fabrics produced from microfilament yarns have a very compact structure due to small pore dimensions between the fibers inside the yarns and between yarns themselves. These fabrics provide very good resistance against wind for different end uses such as parachutes, sails, wind-proof clothes, tents while serving light weight and high durability properties (Kaynak & Babaarslan, 2011). Wind resistance is usually assessed by measuring air permeability. Air permeability is the rate of air flow per unit area of fabric at a standard pressure difference across the faces of the fabric (Horrocks & Anand, 2000). The passage of air is of importance for a number of fabric end uses such as industrial filters, tents, sail cloths, parachutes, raincoat materials, shirting, down proof fabrics and airbags (Saville, 2002). Airflow through textiles is mainly affected by the pore characteristics of fabrics (Bivainyte & Mikucioniene, 2011). As fabric interstices increase in number and size, air permeability increases. In other words as fabric porosity increases, air permeability increases (Collier & Epps, 1999).

Fig. 3. Air permeability of plain weave type samples

sample was replicated twice and ten repeated measurements were done for each replication.

Design-Expert statistical software package was used to interpret the experimental data and to compose the regression models. Regression models were formed to define the relationship between independent variables (filament fineness and weft sett) and response variable (air permeability) for plain, twill and satin weave types. General Factorial Design was selected to compose regression models. The air permeability test results of samples were used to analyze the general factorial design. The analysis of variance, lack of fit tests and residual analysis were performed to select the proper model for the air permeability.

Tightly woven fabrics produced from microfilament yarns have a very compact structure due to small pore dimensions between the fibers inside the yarns and between yarns themselves. These fabrics provide very good resistance against wind for different end uses such as parachutes, sails, wind-proof clothes, tents while serving light weight and high durability properties (Kaynak & Babaarslan, 2011). Wind resistance is usually assessed by measuring air permeability. Air permeability is the rate of air flow per unit area of fabric at a standard pressure difference across the faces of the fabric (Horrocks & Anand, 2000). The passage of air is of importance for a number of fabric end uses such as industrial filters, tents, sail cloths, parachutes, raincoat materials, shirting, down proof fabrics and airbags (Saville, 2002). Airflow through textiles is mainly affected by the pore characteristics of fabrics (Bivainyte & Mikucioniene, 2011). As fabric interstices increase in number and size, air permeability increases. In other words as fabric porosity increases, air permeability

0 0,5 1 1,5 2 2,5 3 3,5

30 wefts/cm 32 wefts/cm 34 wefts/cm 36 wefts/cm

**Filament fineness, dtex** 

The mean values of the test results were used in graphical representation.

**7. Results and discussion** 

increases (Collier & Epps, 1999).

**Air Permeability, mm/sec** 

Fig. 3. Air permeability of plain weave type samples

Figure 3 exhibits the air permeability of plain weave samples for different filament fineness and weft sett. The highest air permeability value is 71.2 mm/sec and the lowest value is 10.2 mm/sec. As seen from Figure 3, increasing weft sett values cause a decrease of air permeability values for plain weave samples. Higher weft sett values provide the gaps between the yarns which the air pass through to become smaller thus lead to lower air permeability. The literature survey shows that some former studies on this topic (Fatahi & Yazdi, 2010; Çay &Tarakçoğlu, 2008; Çay &Tarakçoğlu, 2007) are agree with our work. Nevertheless, it must be considered that the effect of weft sett on air permeability is more obvious for coarser filaments. In other words the influence of weft sett on air permeability decreases as the filament fineness decreases. On the other hand lower filament finenesses cause lower air permeability. Because, lower filament fineness results in higher number of filament in yarn cross section. Thus, dimensions of gaps between the filaments within the yarns decreases. This is an expected result, since micro voids between the fibers become smaller as the fiber diameter decreases, thus the air permeability decreases as pointed out in an earlier study (Varshney, 2010).

The statistical analyses show that the best fitting model is the cubic model for plain weave type. ANOVA results for air permeability of plain weave type samples is given in Table 5.


Table 5. ANOVA results for air permeability of plain weave type samples

ANOVA results in Table 5 show that the effect of filament fineness (A) and weft sett (B) on air permeability is significant for plain weave samples from a statistical approach.

The regression equation of the cubic model for plain weave type is as follows:

$$\begin{array}{l}\text{Air Permeability (mm/sec)} = 36.02279 + 450.27578 \text{ A} + 1.132469 \text{ B} + 47.20693 \text{ A}^2 \\ \text{ } + 0.053972 \text{ B} \text{ } - 29.40924 \text{ AB } - 1.35053 \text{ A} \text{ } \text{B} + 0.47562 \text{ AB}^2 \end{array} \tag{1}$$

In this equation (1); A and B are the filament fineness (dtex) and weft sett (wefts/cm) independent variables respectively. The air permeability of plain weave polyester

Figure 4 illustrates the air permeability versus filament fineness of twill weave samples for different weft sett values. The highest air permeability value is 69.9 mm/sec while the lowest value is 14.5 mm/sec. It must be emphasized that reducing the air permeability from 69.9 mm/sec to 14.5 mm/sec is an important result showing the effect of filament fineness and weft sett on the air permeability. It is clear from Figure 4 that decreasing the filament fineness has a decreasing effect on the air permeability. A direct linear relationship between filament fineness and air permeability is also obvious. This relationship is similar for all weft

Furthermore, increasing weft sett causes a decrease of air permeability for twill woven fabrics. But the magnitude of this effect is decreased as the filament fineness decreased. The difference of air permeability value between the highest and the lowest weft sett are; 30.1 mm/sec, 12.3 mm/sec, 6.1 mm/sec, 6.1 mm/sec and 3.3 mm/sec for 3.05, 1.14, 0.76, 0.57 and 0.33 dtex filament finenesses respectively. These findings suggest that, total void volume has the vital importance with respect to air permeability and total void volume which the air flow pass through is affected by filament fineness much more than

The statistical analyses show that the best fitting model is the cubic model for twill weave type. ANOVA results for air permeability of twill weave type samples are given in Table 7.

> **Mean Square**

**Model** 9399.30 6 1566.55 3897.76 < 0.0001 Significant **A** 190.33 1 190.33 473.56 < 0.0001 Significant **B** 1232.42 1 1232.42 3066.40 < 0.0001 Significant **A2** 11.58 1 11.58 28.80 < 0.0001 Significant **B2** 10.10 1 10.10 25.13 < 0.0001 Significant **AB** 526.39 1 526.39 1309.71 < 0.0001 Significant **A3** 34.22 1 34.22 85.15 < 0.0001 Significant

**Lack of Fit** 8.28 13 0.64 2.55 0.0289 Significant

ANOVA results in Table 7 show that the effect of filament fineness (A) and weft sett (B) on

Air Permeability (mm/sec) = 261.98458 +55.06852 A - 11.05995 B + 27.54636 A2 + 0.12562 B2 - 1.66115 AB -5.87156A3 (2) In this equation (2); A and B are the filament fineness (dtex) and weft set (wefts/cm) independent variables respectively. The air permeability of polyester microfilament woven fabrics can be predicted by this equation. Mean Square Error (MSE), Mean Absolute Error (MAE), Mean Absolute Percent Error (MAPE%), R-square predicted (R2predicted) and R-

**F** 

**value P value Significance** 

**Degree of freedom** 

Table 7. ANOVA results for air permeability of twill weave type samples

air permeability is significant for twill weave samples from a statistical approach.

The regression equation of the cubic model for twill weave type is as follows:

sett values.

weft sett.

**Source Sum of** 

**squares** 

**Residual** 13.26 33 0.40

**Pure Error** 4.98 20 0.25

**Cor. Total** 9412.56 39

microfilament woven fabrics can be predicted by this equation. Mean Square Error (MSE), Mean Absolute Error (MAE), Mean Absolute Percent Error (MAPE%), R-square predicted (R2predicted) and R-squared (R2) values which contribute the performance of the statistical model for plain weave type samples are seen in Table 6.

According to model performance values, the correlation coefficient between predicted and observed air permeability values is 0.9854 indicating a strong predictive capability of the regression model for plain weave types. Also, this regression model can predict the air permeability with 95.47% accuracy. So it can be said that the regression equation gave satisfactory results.


Table 6. Performance of the model for plain weave type

Fig. 4. Air permeability of twill weave type samples

microfilament woven fabrics can be predicted by this equation. Mean Square Error (MSE), Mean Absolute Error (MAE), Mean Absolute Percent Error (MAPE%), R-square predicted (R2predicted) and R-squared (R2) values which contribute the performance of the statistical

According to model performance values, the correlation coefficient between predicted and observed air permeability values is 0.9854 indicating a strong predictive capability of the regression model for plain weave types. Also, this regression model can predict the air permeability with 95.47% accuracy. So it can be said that the regression equation gave

> **Performance parameter Value**  R2 0.9936 R2predicted 0.9854 MSE 1.35 MAE 0.89 MAPE, % 4.53

0 0,5 1 1,5 2 2,5 3 3,5

**Filament fineness, dtex** 

41 wefts/cm 43 wefts/ cm 45 wefts/cm 47 wefts/cm

model for plain weave type samples are seen in Table 6.

Table 6. Performance of the model for plain weave type

Fig. 4. Air permeability of twill weave type samples

satisfactory results.

0

10

20

30

40

50

60

70

80

**Air permeaility, mm/sec** 

Figure 4 illustrates the air permeability versus filament fineness of twill weave samples for different weft sett values. The highest air permeability value is 69.9 mm/sec while the lowest value is 14.5 mm/sec. It must be emphasized that reducing the air permeability from 69.9 mm/sec to 14.5 mm/sec is an important result showing the effect of filament fineness and weft sett on the air permeability. It is clear from Figure 4 that decreasing the filament fineness has a decreasing effect on the air permeability. A direct linear relationship between filament fineness and air permeability is also obvious. This relationship is similar for all weft sett values.

Furthermore, increasing weft sett causes a decrease of air permeability for twill woven fabrics. But the magnitude of this effect is decreased as the filament fineness decreased. The difference of air permeability value between the highest and the lowest weft sett are; 30.1 mm/sec, 12.3 mm/sec, 6.1 mm/sec, 6.1 mm/sec and 3.3 mm/sec for 3.05, 1.14, 0.76, 0.57 and 0.33 dtex filament finenesses respectively. These findings suggest that, total void volume has the vital importance with respect to air permeability and total void volume which the air flow pass through is affected by filament fineness much more than weft sett.

The statistical analyses show that the best fitting model is the cubic model for twill weave type. ANOVA results for air permeability of twill weave type samples are given in Table 7.


Table 7. ANOVA results for air permeability of twill weave type samples

ANOVA results in Table 7 show that the effect of filament fineness (A) and weft sett (B) on air permeability is significant for twill weave samples from a statistical approach.

The regression equation of the cubic model for twill weave type is as follows:

Air Permeability (mm/sec) = 261.98458 +55.06852 A - 11.05995 B + 27.54636 A2 + 0.12562 B2 - 1.66115 AB -5.87156A3 (2)

In this equation (2); A and B are the filament fineness (dtex) and weft set (wefts/cm) independent variables respectively. The air permeability of polyester microfilament woven fabrics can be predicted by this equation. Mean Square Error (MSE), Mean Absolute Error (MAE), Mean Absolute Percent Error (MAPE%), R-square predicted (R2predicted) and R-

76.9 mm/sec as the difference between these values is considerable. This value (76.9 mm/sec) is higher than the value (61 mm/sec) obtained for plain weave samples and the value (55.4 mm/sec) obtained for twill weave samples. Based on this result, it is clear that the effects of filament fineness and weft sett on satin weave type samples are higher than those of plain and twill weave type samples. Decreasing filament fineness causes a decrease on air permeability for satin weave samples. This is the result of reducing the void volume of the fabric due to decreasing filament fineness as seen before in plain and twill weave samples. A direct linear relationship between filament fineness and air permeability is also

Besides, higher weft sett values resulted in lower air permeability for satin weave type samples. Similar with plain and twill weave samples, the influence of decreasing weft sett on air permeability decreases as the filaments become finer. The difference of the air permeability value between the highest and the lowest weft sett are; 45.7 mm/sec, 12.3 mm/sec, 6.4 mm/sec, 5.6 mm/sec and 5.5 mm/sec for 3.05, 1.14, 0.76, 0.57 and 0.33 dtex filament fineness respectively. This common tendency for both all weave types explains the effect of filament fineness on total void volume of fabrics. It should be noted that, for satin weave type there is a direct proportion between filament fineness and air permeability and a reverse proportion between weft sett and air permeability as seen in earlier studies (Fatahi &

Yazdi, 2010; Çay &Tarakçoğlu, 2008; Çay &Tarakçoğlu, 2007; Varshney, 2010).

**Degree of freedom** 

ANOVA results for air permeability of satin weave type samples are given in Table 9.

**Mean Square** 

**Model** 14798.49 5 2959.70 778.77 < 0.0001 Significant

**Lack of Fit** 76.20 14 5.44 2.05 0.0691 Not Significant

ANOVA results in Table 6 show that the effect of filament fineness (A) and weft sett (B) on

Table 9. ANOVA results for air permeability of satin weave type samples

air permeability is statistically significant for satin weave samples.

**A** 3831.95 1 3831.95 1008.28 < 0.0001 Significant **B** 2267.35 1 2267.35 596.60 < 0.0001 Significant **B2** 95.39 1 95.39 25.10 < 0.0001 Significant **AB** 1259.27 1 1259.27 331.35 < 0.0001 Significant **AB2** 81.93 1 81.93 21.56 < 0.0001 Significant

**F** 

**value P value Significance** 

available for satin weave samples.

**Source Sum of** 

**squares** 

**Residual** 129.22 34 3.80

**Pure Error** 53.01 20 2.65

**Cor Total** 14927.70 39

squared (R2) values which contribute the performance of the statistical model are seen in Table 8.

According to model performance values, the correlation coefficient between predicted and observed air permeability values is 0.9979 indicating a strong predictive capability of the regression model for twill weave types. Also, this regression model can predict the air permeability with 98.4% accuracy. So it can be said that the regression equation gave satisfactory results.


Table 8. Performance of the model for twill weave type

Fig. 5. Air permeability of satin weave type samples

The air permeability of satin weave type samples are shown in Figure 5. The highest and the lowest air permeability values are 89.3 mm/sec and 12.4 mm/sec respectively. The value of

squared (R2) values which contribute the performance of the statistical model are seen in

According to model performance values, the correlation coefficient between predicted and observed air permeability values is 0.9979 indicating a strong predictive capability of the regression model for twill weave types. Also, this regression model can predict the air permeability with 98.4% accuracy. So it can be said that the regression equation gave

> **Performance parameter Value**  R2 0.9986 R2 predicted 0.9979 MSE 0.33 MAE 0.41 MAPE % 1.60

0 0,5 1 1,5 2 2,5 3 3,5

43 wefts/cm 45 wefts/ cm 47 wefts/cm 49 wefts/cm

**Filament fineness, dtex**

The air permeability of satin weave type samples are shown in Figure 5. The highest and the lowest air permeability values are 89.3 mm/sec and 12.4 mm/sec respectively. The value of

Table 8. Performance of the model for twill weave type

Fig. 5. Air permeability of satin weave type samples

Table 8.

satisfactory results.

**Air Permeability, mm/sec** 

76.9 mm/sec as the difference between these values is considerable. This value (76.9 mm/sec) is higher than the value (61 mm/sec) obtained for plain weave samples and the value (55.4 mm/sec) obtained for twill weave samples. Based on this result, it is clear that the effects of filament fineness and weft sett on satin weave type samples are higher than those of plain and twill weave type samples. Decreasing filament fineness causes a decrease on air permeability for satin weave samples. This is the result of reducing the void volume of the fabric due to decreasing filament fineness as seen before in plain and twill weave samples. A direct linear relationship between filament fineness and air permeability is also available for satin weave samples.

Besides, higher weft sett values resulted in lower air permeability for satin weave type samples. Similar with plain and twill weave samples, the influence of decreasing weft sett on air permeability decreases as the filaments become finer. The difference of the air permeability value between the highest and the lowest weft sett are; 45.7 mm/sec, 12.3 mm/sec, 6.4 mm/sec, 5.6 mm/sec and 5.5 mm/sec for 3.05, 1.14, 0.76, 0.57 and 0.33 dtex filament fineness respectively. This common tendency for both all weave types explains the effect of filament fineness on total void volume of fabrics. It should be noted that, for satin weave type there is a direct proportion between filament fineness and air permeability and a reverse proportion between weft sett and air permeability as seen in earlier studies (Fatahi & Yazdi, 2010; Çay &Tarakçoğlu, 2008; Çay &Tarakçoğlu, 2007; Varshney, 2010).


ANOVA results for air permeability of satin weave type samples are given in Table 9.

Table 9. ANOVA results for air permeability of satin weave type samples

ANOVA results in Table 6 show that the effect of filament fineness (A) and weft sett (B) on air permeability is statistically significant for satin weave samples.

It is seen from Figure 6 that for all filament fineness and weft sett values, satin woven fabric samples show higher air permeability than twill fabric samples. The number of interlacing is lower in satin weave type than twill weave type. So the yarn mobility is higher for satin weave types. Yarn mobility is higher, thus the pore dimensions become larger because of the deformation during air flow (Çay & Tarakçoğlu, 2008). So enlarging the pore dimensions cause the air flow to pass through the fabric more easily and air permeability of the fabrics increase. On the other hand, the effect of weave type on air permeability is considerable for

Wind resistance was achieved by coating fabrics formerly. But, it is already known that obtaining the wind proof fabrics with a better breathability is possible by weaving microfilament yarns with high densities. These types of fabrics provide a good thermal insulation in windy conditions in addition to submitting a comfortable wear by transporting the sweat vapor more easily than other counterparts. So, it is widely important to know how the woven fabric parameters affect wind resistance and to determine the proper values of these parameters for particular end uses. Consequently, air permeability of polyester microfilament and conventional filament fabrics is presented here. It is already known that a good wind resistance can be achieved by ensuring lower gaps in fabric structure with finer filaments and higher densities. A considerable effect of filament fineness on air permeability is seen, for all weave types used in this study. The experimental results showed that decreasing the filament fineness have a decreasing effect on fabric air permeability. This is not surprising, since the air gaps between the filaments within the yarns become smaller as the filament fineness decreases. Thus, air flow through the filaments is prevented. Furthermore, higher weft sett values provided lower air permeability values because of obtaining smaller air gaps between the yarns in fabric structure. This situation causes the air flowing through the fabric more difficultly and fabric air permeability decreases. It should be noted that our study was investigated the effect of filament fineness and fabric density by differentiating the parameters of weft direction solely. It may be concluded that by changing the parameters in the warp direction, lower air permeability can be achieved. From the point of weave type, it is observed that weave types with higher number of float or lower number of interlacing have higher air permeability values. Because this type of weaves provides better mobility for yarns in their structure and gaps between these yarns become larger with air flow

As mentioned earlier, very good resistance against wind can be achieved by tightly woven microfilament fabrics for different end uses. The most convenient fabric construction can be determined for a particular end use such as wind proof cloth, tent, e.t.c. with the aid of the results and regression analysis obtained from this experimental study. This study also lends assistance to decide the structural parameters for barrier fabrics in specialized applications such as surgical gowns which will be on horizon in near future. For further studies, fabric properties can be developed by using different fiber blends, applying fabrics mechanical surface treatments and special finishes. In the near future, it is expected to see nanofiber yarns for producing woven and knitted fabrics as well as

3.05 dtex filament fineness.

more easily than others.

**8. Conclusion** 

The regression equation of the cubic model for satin weave type is as follows:

Air Permeability (mm/sec) = -400.27938 +909.42821 A +17.24904 B - 0.18175 B2 - 36.27354 AB + 0.36635 AB2 (3)

In this equation (3); A and B are the filament fineness (dtex) and weft sett (wefts/cm) independent variables respectively. The air permeability of polyester microfilament woven fabrics can be predicted by this equation. Mean Square Error (MSE), Mean Absolute Error (MAE), Mean Absolute Percent Error (MAPE%), R-square predicted (R2predicted) and Rsquared (R2) values which contribute the performance of the statistical model are seen in Table 10.

According to model performance values, the correlation coefficient between predicted and observed air permeability values is 0.9815 indicating a strong predictive capability of the regression model for twill weave types. Also, this regression model can predict the air permeability with 95.37% accuracy. So it can be said that the regression equation gave satisfactory results.


0 10 20 30 40 50 60 70 80 90 100 43 w/cm 45 w/cm 47 w/cm 43 w/cm 45 w/cm 47 w/cm 43 w/cm 45 w/cm 47 w/cm 43 w/cm 45 w/cm 47 w/cm 43 w/cm 45 w/cm 47 w/cm 0.33 dtex 0.57 dtex 0.76 dtex 1.14 dtex 3.05 dtex **Air Permeability, mm/sec**  Twill Satin

Table 10. Performance of the model for satin weave type

Fig. 6. The effect of weave type on air permeability

It is seen from Figure 6 that for all filament fineness and weft sett values, satin woven fabric samples show higher air permeability than twill fabric samples. The number of interlacing is lower in satin weave type than twill weave type. So the yarn mobility is higher for satin weave types. Yarn mobility is higher, thus the pore dimensions become larger because of the deformation during air flow (Çay & Tarakçoğlu, 2008). So enlarging the pore dimensions cause the air flow to pass through the fabric more easily and air permeability of the fabrics increase. On the other hand, the effect of weave type on air permeability is considerable for 3.05 dtex filament fineness.
