Bending properties of wet-abraded woven fabrics

Transcription

1 Indian JournaJ of Fibre & Textile Research Vol. 19, December 1994, pp Bending properties of wet-abraded woven fabrics Joshua Osayande Ukponmwan Department of Polymer and Textile Technology. Federal University of Technology. Owerri. Nigeria Received II May 1993~ accepted 31 January 1994 The bending properties of unabraded and abraded woven fabrics have been studied under different wet conditions in terms of elastic and frictional resistance to bending deformation, bending recovery and frictional residual curvature using the Shirleycyclicbending tester to determine the effectof abrasion damage on these properties. The correlationships among the bending properties and also between the bending properties and the per cent weight loss have been studied using the regression analysis. technique and Spearman rank correlation coefficient respectively. The damp-abraded fabrics show higher percentage variation in bending properties than the wet-abraded fabrics. Large amount of water present in the sample and abradant during abrasion testing alters the mode of fabric abrasion significantly. Keywords: Abrasion resistance, Bending properties, Elastic resistance, Frictional resistance, Regression analysis, Shirleycyclicbending tester, Spearman rank correlation coefficient,woven fabrics 1 Introduction Theeffect of wear on the bending behaviour of woven cotton/polyester fabrics' and the changes in the bending properties of cotton fabric Ias a result of laboratory wear- have been investigated. These include the bending properties before and after the dry abrasion, the effect of abrasion damage on the bending behaviour and the changes in properties (expressed as percentage difference) after wear compared to the original value. A significant difference between the unabraded (original) and abraded fabrics was observed at 95% confidence interval with respect to bending properties as determined by the simple paired t-test. Recently, the effect of incremental wear on the bending properties of woven cotton fabrics" was compared with the changes in the bending properties caused by the increments of accelerator abrasion in dry condition. The present work was carried out using different woven fabrics, viz. cotton, cotton/polyester, wool, nylon and trevera mix, to confirm the previous findings. In addition to this, the effect of wear in wet conditions on the changes in bending properties of fabrics was studied using the Shirley cyclic bending tester before and after abrasion. The effect of abrasion damage on the bending properties was also studied. The changes in properties, expressed as percentage difference, after wear were compared with those of the unabraded fabrics. The correlations between the parameters derived from the bendinghysteresis test and other mechanical test such as wear were also examined using the Spearman rank correlation coefficient technique and regression analysis technique. Previous work" - 6 on the bending behaviour of woven fabrics emphasized the importance of the mechanical properties of the fibre and the frictional and geometrical restrains within and between the yarns in the fabric. It is, therefore, reasonable to expect that the effect of wear due to abrasion might be considerable. The bending-hysteresis test has been proved to be a powerful tool in the investigation of the properties of fabrics (both woven and knitted). With the help of this test, the stiffness and liveliness of fabrics can be readily assessed. The use of Shirley cyclic bending tester, its principle and applications have been descri bed elsewhere". The advantages of using the accelerator abrasion tester have been reported earlier'. 2 Materials and Methods Objective measurements on each fabric were carried out as per the standard procedure laid down in the British standard handbook? The definitions of the bending parameters have been given elsewhere". 2.1 Materials Fabrics selected represent the thickness and mass ranges commercially available within their class. Set I consisted of three 100% cotton fabrics (CI-C3), set 2 of five cotton/polyester fabrics (CP l-cp5), set 3 of

2 230. INDIAN J. FIBRE TEXT. RES., DECEMBER 1994 three 100% wool fabrics (WI-W3) and set 4 of two synthetic fabrics - nylon (S I) and Trevera mix (S2). The construction details of these fabrics are given in Table I. No specific information about the finishing treatments of these fabrics was available and normal finishing treatments (desizing, scouring, bleaching and mercerization with no crease-resistant finishing) was assumed for the intended end-uses. Five bending tests were made on each sample. The fabrics were conditioned and all the tests were carried out in an atmosphere of20±2 C and 65±2% R H. 2.2 Methods Measurement of Bending Properties The bending-hysteresis tests were performed on the Shirley cyclic bending tester+":" -10 designed to provide a more complete characterization offabric bending behaviour based on its ability to separate the elastic and frictional components of fabric stiffness. Briefly, a fabric sample of25 mrn x 8 mrn and bending length of 5 mm is bent to curvatures of ± 4 em - 1. The couple resisting bending is measured for each curvature. From the hysteresis loops obtained, the following parameters were calculated: low-curvature elastic flexural rigidity (Go); coercive couple (Co); bending recovery (BR), i.e. the ratio of recovered deformation to the maximum deformation given in bending; and frictional residual curvature (Co/Go),.a measure of the degree of recovery of the fabric from gentle crumpling which should be closely related to cloth liveliness. A lively fabric is expected to have a low value of Co/Go and vice versa. Bending length, as measured by the Shirley stiffness tester, was not considered in this experiment, since this technique does not differentiate between elastic and frictional components of bending. Five tests were made on each sample and the average of the five readings was taken Resistance to Abrasion (Wear Test) Fabrics were tested for wear using the accelerator tester in wet conditions. The details of the accelerator and the standard methods of using it have been reported elsewhere In this experiment, the accelerator tests were carried out under the following conditions: Table I-Fabric construction details Fabric Fabric Fabric Threads! Yarn count Cover factor Thickness Mass Density Specific code structure cm tex (K)a mm g/m 2 mg/cm:' volume cm 3 /g Warp Weft Warp Weft Warp Weft Batyb Instron" Set I Cotton CI Twill 2/ Cotton C2 Twill 1/ Cotton C3 Plain Set II Cotton/Polyester CPI Twill 2/ (20:80) Cotton/Polyester CP2 Plain (40:60) Cotton/Polyester CP3 Plain (30:70) Cotton/Polyester CP4 Twill 2/ (20:80) Cotton/Polyester CP5 Plain (30:70) Set III Wool WI Plain Wool W2 Plain Wool W3 Plain Set IV Nylon SI Plain Trevera Mix S2 Plain ~l.<: = (Treads/em) 100 x.jtex ; b Pressure, 3.5 knm- 2 ; and C Pressure, 1.9 knm- z

3 UKPONMWAN: BENDING PROPERTIES OF WOVEN FABRICS 231 Wet condition 20"C Water, 150 ml of I % detergent (teapol) Damp condition Specimen size No. of specimens: Rotor Rotor speed blade Type of collar Sample cleaning Wet Damp Duration Drying of test 20"C, 80% Moisture 13.4cm rpm Straight, Ribhed 41 in pitched metal Cold water rinse Vacuum 3 min Samples content were left in open air for 24 h The mean of the test results for five specimens per sample was taken. Short abrasion treatment period (3 min) was chosen as some of the fabrics used in the experiment proved difficult to abrade satisfactorily due to the bursting of threads prematurely from the specimen when subjected to longer treatment period at 3000 rpm. In addition, the fabric structure was disturbed, resulting in very high area shrinkage and increased compactness as in the case of dampabraded wool fabrics (Table 2). This led to change in the dimensions of the fabrics, which, in turn, caused changes in size, shape and stiffness of the abraded fabrics, making the samples unsuitable for bending test. 3 Results and Discussion Fig.l shows the typical bending-hysteresis curves for unabraded, dry-abraded and wet-abraded cotton/polyester woven fabric (CPl) samples. The loops are centred on the origin '0' and are symmetrical about it. Except for a small region following each reversal of curvature, the sides of the loops are straight and parallel, i.e. the bending behaviour is linear once the motion is occurring at all inter-fibre contact points. The elastic flexural rigidity (Go) is measured as the slope of the linear region of the hysteresis curve, and the coercive frictional couple (Co) as half the width of the hysteresis loop at zero deformation. This is illustrated in Fig.2. The loop for the wet-abraded 700 c Curvature. cm-1 A = Unabrcdctd sample B = Dry Accelerator Tt:5f{ obrooed) C = Wet Accwlrratlr jest (abraded) Fig. I-Typical bending-hysteresis curves for cotton/polyester (20:80) woven fabric Table 2-Loss in weight and area shrinkage of abraded fabrics Fabric Loss in weight, % Area Shrinkage code (damp abraded) Wet abraded Damp abraded % CI C C CPI Go= the slope of the straight portions QP and RS of the curvei.e CP Slope QP + Slope RS CP l.l CP Co= the intercept OP or OR on the CP couple ClJCis,i.e OP + OR 2 WI 0.9\ , BR('I,) = 100(C D - V W') W CD W3 0.9\ C /6 _ Curvature at the points V. W S\ o 0- where QP and SR (extrapolated) meet the curvature axis S Fig. 2-A typical bending-hysteresis curve for woven fabric

4 232 INDIAN J. FmRE lext. RES. DECEMBER 1994 accelerator test sample (C) is steeper and wider in damp abrasion for 3 min are shown in Table 3. curvature than that for unabraded (A) and dry- Essentially, the properties reflect the measurements abraded (B) accelerator test samples (Fig. I). This is of Co. Go, Co/Go and BR. The changes in properties because the presence of moisture and large amount of after wear, expressed as percentage differences, are water in the wet-abraded samples stiffened the shown in Table 4. samples and rendered them less pliable. In the wet Table 3 shows that as a result of slight wear (3 min state, the water, acting as a lubricant or damping accelerator abrasion test), the Co and Go of cotton agent, reduces the severity of abrasion conditions and fabrics decrease after the wet and damp abrasion. The hence the wear rate is reduced (Table 2). As a result of decrease for fabrics C I and C2 is more after damp dry wear, the flexural rigidity decreases and the abrasion than after wet abrasion, while for fabric C3, fabric samples become less stiff and lively. the decrease is less after damp abrasion than after wet abrasion. Co/Go and BR show inconsistent results 3. i Bending Properties with increase and decrease in some fabrics after both The bending properties before and after wet and wet and damp abrasion (Table 3). Table 3-Bending properties of unabraded and abraded fabrics Fabric Abrasion Coercive couple (C,) Flexural rigidity (Go) Residual curvature Bending recovery code condition mnmmjmm mnmm 2 /mm (Co/Go). mm- 1 0/0 Unabraded CI Wet Damp Unabraded C2 Wet Damp Unabraded C3 Wet J Damp Unabraded CPI Wet Damp Unabraded CP2 Wet Damp Unabraded CP3 Wet Damp Unabraded CP4 Wet Damp Unabraded CP5 Wet Damp Unabraded WI Wet Damp Unabraded W2 Wet Damp Unabraded W3 Wet Damp 0.90 \.to Unabraded SI Wet Damp Unabraded S2 Wet Damp

5 UKPONMWAN: BENDING PROPERTIES OF WOVEN FABRICS 233 Table 4-Bending properties of abraded fabrics - % Difference" Fabric Abrasion Co Go C"/G,, BR code condition mnmmjmm m'nmm-jmm mm- I % CI Wet D<lmp - R1.00 -R5.R C2 Wet Damp C3 Wet Damp CPI Wet Damp CP2 Wet Damp CP3 Wet Damp CP4 Wet Damp CP5 Wet Damp WI Wet Damp W2 Wet W3 Damp Wet Damp SI Wet Damp S2 Wet Damp Fabric property Fabric property after abrasion before abrasion a Difference (%) = x 100 Fabric property before abrasion For cotton/polyester fabrics, the values of Co, except for fabric CP2, and Co/Go increase after both wet and damp abrasion. The bending recovery decreases while Go shows inconsistent results with increase and decrease in some fabrics after both wet and damp abrasion. The increase and decrease in the bending properties are inconsistent when wet and damp accelerator abrasion tests are compared. For all the three wool fabrics, the values of Co, Go and Co/G~ increase and the bending recovery decreases after both wet and damp abrasion (Table 3). The increase in Co and Go for wool fabrics W I and W2 is higher after wet abrasion than after damp abrasion. The increase in frictional residual curvature is higher after wet abrasion for the wool fabric W Iand after damp abrasion for wool fabrics W2 and W3. The synthetic fabric S I shows increase in C, and Co/Go and decrease in Go and bending recovery, whereas the synthetic fabric S2 shows decrease in Co and Go and increase in CoJGo and bending recovery after wet and damp abrasion (Table 3). The increase in C.,/Go and the decrease in Go are higher after wet abrasion than damp abrasion. The changes in bending properties after wear, expressed as percentage differences, show similar trends as mentioned above for each set of fabrics (Table 4). The significance of the % changes has not been considered in greater detail but the differences of less then 20% were regarded as slight. The presence of moisture in cotton fabrics during wet abrasion might have decreased the stiffness of these fabrics, rendering them more pliable: Whereas the presence of moisture and a large amount of water in the wool fabrics during abrasion stiffens the fabrics and is believed to alter the mode of fabric abrasion significantly 15. This may ha ve increased the stiffness of the fabrics, rendering them less pliable. This is because in the wet state, water, acting as a lubricant or (mechanical) damping agent, reduces the severity of abrasion conditions and hence the wear rate is reduced (Table 2).

6 234 INDIAN J. FIBRE TEXT. RES., DECEMBER 1994 The general mobility of yarns within the fabric and of fibres within the yarn during tumbling in the accelerator may have increased in the case of cotton fabrics and decreased in the case of wool fabrics. The resultant swelling of fibres is likely to render the weave structure of fabric more compact and hence reduce the wear rate, as observed with damp accelerator abrasion test in the case of cotton fabrics and with wet accelerator abrasion test in the case of wool fabrics. It has been concluded that the stiffness characteristics of a fabric are functions of fabric thickness; the number, size and distribution of fibres in the cross-section, the coefficients of fibre-to-fibre friction, degree of entanglement, and the contact surface area of fibres!". In addition, the stiffness. of a fabric is affected by the constructional features such as weight, nature of fibres, compactness or density of the weave, and finishing treatments. It is observed from Table 2 that the cotton fabrics Cl and C2 show considerable abrasion since they are twill weaves which render the fabrics limpy and the yarns loosely bound, whereas the cotton fabric C3 shows much resistance to wet and damp wear since it is plain weave which renders the fabric more stiff and the yarns tightly bound. Similarly, it is observed that the three wool fabrics show better resistance to wear since they are plain weave. The synthetic fabrics S 1 and S2 and the cotton/polyester fabrics CP2, CP3 and CP5 being plain weave behave similar to the wool fabrics, while the cotton/polyester fabrics CPl and CP4, being twill weave, behave similar to the cotton fabrics C I and C2. Table I shows that among the three cotton fabrics, the fabric C3 is the heaviest, thickest and has the largest yam count. As a result, it is more compact and tightly woven since it is woven plain. This renders the fabric more stiff and show better resistance to wear. The five cotton/polyester fabrics differ among themselves in constructional details and physical and bending properties, but they do not appear to do so in any discernibly consistant manner. For example, while CP4 has a higher sett, it has finer yarn. CP2 and CP5 appear slightly stiffer, but their sett and count are similar to those ofcpi and CP3 respectively. Slight differences in the bending properties may be expected because of the differences in weave (plain/twill), heat setting and possibly in finishing processes. In the case of wool fabrics, taking wool fabric WI as the base for comparison, it is observed that the sett of wool fabric W2 is less weft-way but this is compensated by the use of a heavier yarn for both warp and weft way so that although the cover factors are similar, wool fabric W2 is heavier and thicker. Similarly, the use of coarser yarns results in wool fabric W3 being heavier, thicker, rougher, stiffer and softer than wool fabric WI. Also, the stiffness of a fabric is dependent to a high degree on the finishing treatment given. Since there was no information on the type of finishing given to the fabrics used in the experiment, the effect of this on the bending behaviour of the fabrics could not be ascertained. The loss in stiffness of the cotton fabrics may have been influenced by the direction of the threads in the fabrics and the effect of yarn crimp, but this effect was not considered because stiffness was measured without any reference to the direction of the threads. Finally, perhaps one can mention that the fibre content of the cotton/polyester fabrics is pertinent to the differences in their performance and bending behaviour, since it has been suggested that the abrasion resistance of fabrics is determined by the type offibre. The inclusion of the high proportion of polyester fibre in the blend might have improved their abrasion resistance. Fabrics CP3, CP5, CP4 and CPI performed better than fabric CP2. This confirms the findings of work on blended-fibre fabrics which shows that partial substitution by polyamide or polyester fibres improves the abrasion resistance of cotton 17 (Table 2). This also accounts for the low percentage area shrinkage in damp accelerator abrasion test of the cotton/polyester fabrics compared with the wool fabrics as less swelling takes place. 3.2 Property Characteristics Fabrics have been assigned a rank value for each bending property and percentage loss in weight. This ranking is based solely on the magnitude of increasing bending property and percentage loss in weight shown in Tables 2 and 3. Low rank values indicate a high value of the property, while higher rank values indicate a low value of the property. For instance, the unabraded fabric WI with the lowest coercive. couple is ranked I for that property and so on. The rankings of the fabrics are shown in Table 5 for comparison purpose. Not surprizingly, it is seen that the ranking positions alter with each property. A definite order would be expected if the fabrics are chosen for any systematic change of construction, e.g. increase in yarn sett and/or count or cover factor. In such circumstances, an agreement might occur from the highest, thinnest, least dense fabric to the heaviest, thickest and most dense fabric. But as stated earlier, this was not the reason for the choice of fabrics. There are probably several other reasons why consistant placings would not be expected. Among these couid be the error, chance, sample variation and method of scoring. In addition, some of the bending properties are known to be fibre dependent and others yarn dependent. Therefore, although the sets

7 UKPONMWAN: BENDING PROPERTIES OF WOVEN FABRICS 235 Table 5-Fabrics ranked in order Of bending properties Fabric Unabraded fabric Wet-abraded fabric Loss in Damp-abraded fabric Loss in code weight weight 0; C" G" BR.% C)G" C" Go BR,% C"/G,, '0 C,' G" BR.% C";G,, % CI I I 10 9" 6" C2 II 10 :I II " I" C3 13" II e " 4" 2 II I CPI 6 7 II e 6 9 CP2 9 9" 10" 4 :I 6 6" 2 9" 3 3 6" 8 12 CP II 7" 4" '} CP " 4" 7 6 II II CP5 8 9" " 4< 4 7 '} WI Ih I 10" 3 6 7" 7 5 5< 2\: 5 7 I 10 W I I I" W " 5 5" 6" 4" 5" 5 8 6" 2 I3 Sl " 1< 2 2" II I" I I I" I3 7 S I" I 21.: ," 2 8" 10 II " Low rank; b High rank; and ' Fabrics ranked the same of fabrics selected were all woven, they contained natural fibres (cotton and wool), blends of natural staple and filament fibres in different proportion and synthetic fibres, and some were plain weave while others twill weave. 3.3 Correlationships among Bending Properties and between Bending Properties and Percentage loss in Weight To determine the inter-relationship among the bending properties and between the bending properties and percentage loss In weight, the Spearman rank correlation coefficients were calculated using the fabric rank values of the unabraded, wet- and damp-abraded fabrics (Tables 5 and 6). Table 6 indicates significant correlations at hoth 5% and I % levels between Go and C, for the unabraded, wet- and damp-abraded fabrics. Correlations between Go and BR are also significant at both ~% and I% levelsfor the wet-abraded fabrics and at 1% level for the damp-abraded fabrics. Significant correlation at I % level between Co and C,/G o for the unabraded fabrics and between Co and BR for the wet-abraded fabrics is also observed. The correlations between the bending properties (Co,Go, Co/Go and BR) and the percentage loss in weight are poor for the wet- and damp-abraded fabrics. A linear regression 'analysis was carried out for the bending properties, yielding the linear relationships between the bending properties Co and Go and BR and Co/Go. Table 7 shows the results of the.linear regression analysis. It is observed that the linear regression correlations for Co vs Go are very high (0.953 and 0.901) for the unabraded and dampabraded fabrics and very low (0.082) for the wetabraded fabric. For BR vs Co/Go, a very high negative Table 6-~orrelation coefficient of the bending properties and % weight loss Bending property c, c" BR 'y',wl Unabraded fabric Go 0.8" Co/Go 0.6f, 0.3 BR Wet-abraded C" Go 0.8" Co/Go BR 0.6 h o.s fabric 'YoWL C" Go 0.9" Co/Go 0 BR 0.3 %WL -0.2 WL - Weight loss Damp-abraded b 0.1 fabric a Significant at 5% and I% level; and b Significant at I% level correlation ( ) for unabraded fabrics and fair negative correlations ( and ) for wetand damp-abraded fabrics respectively were obtained. The effect of bending recovery on Co/Go is shown in Fig.3. The points are scattered and do not show a particular trend. This is in agreement to the work reported earlier on cotton fabrics- but in contrast to the results 1 obtained in dry accelerator test.

8 236 INDIAN J. FmRE TEXT. RES., DECEMBER 1994 Table 7-Results of linear regression analysis for fabric bending properties Fabric sample Correlation coefficient Slope Intercept Best fitting straight line Co vs Go Unabraded Go = 0.88 Co Wet abraded Co = 0.06 Co Damp abraded Co = 1.01 Co BR vs Co/C" Unabraded Co/Co = BR Wet abraded Co./Go = BR Damp abraded CjG" = BR I ~ 1' E 3'5 E.: 3 0 0: ;:) t- 2'5 oc( > 0: ;:) u 2 0 & & Damp abrad~ 1. o o I 0 W~t abradtod FABRIC I.Cotton ocotton' pol y~st~r owool asyntht"tic E EĒ E z E w ~ Q. ~ ou w > u 3 Damp cbr ade d Ie 2 I- FABRI _.p)tton otton I Poldieste-r owo I ASynthetic o 2 I I.800 I- 0 Wet nbrcded - 0-0,..o. o I I 3~ ~ ~ oc( 1 5 ;:) 0 ~ 1 0 0: \0. a: w o u Un nbreded "5,,0 - (}S ~ 0 40 Unabrad~d A. 0' 0 I I I BENDING RECOVERY, '. Fig. 3-Effect of bending recovery on.residual curvature 1 - A " 0.A 0 0 I I FLEXURAL RIGIDITY mn mm2jm m Fig. 4-Coercive couple vs flexural rigidity

9 UKPONMWAN: BENDING PROPERTIES OF WOVEN FABRICS 237 Table 8-Statistical analysis of the experimental data of different fabrics Bending property Abrasion Mean Variance Standard Coefficient of condition deviation variation. % Coercive couple, mnmm/mm Unabraded Wet Damp Flexural rigidity. mnmm"(mm Unabraded 1.74 :U Wet Damp 0.90 O.IS Residual curvature. mm - I Unabraded Wet Damp Bending recovery, % Unabraded Wet Damp S Table 9-{;omparative presentation of statistical analysis of experimental data for different fabrics Wear conditions Mean" Variance Standard Coefficient Tabulated t-value Calculated t-value compared deviation of at 95% confidence at 95% confidence variation interval interval Coercive couple, mnmm/mm U and W U and D Wand D Flexural rigidity, mnmm2/mm U and W U and D I.S Wand D Frictional residual curvature. mm - I U and W U and D Wand D Bending recovery, % U and W U and D Wand a The mean of individual values of the differences between unabraded fabrics.u - Unabraded, W - Wet abraded, and D - Damp abraded and abraded fabrics and between wet abraded and damp abraded A graphical plot of the bending properties Co and Go for the unabraded, wet-abraded and dampabraded fabrics is shown in Fig.4. The points are widely scattered for the unabraded fabrics, much closer for the wet- and very close for the dampabraded fabrics. This is in agreement to the work reported 1 on cotton/polyester fabrics in dry accelerator test but in contrast to the work done on cot ton in dry accelerator test.'. Table 8 shows the statistical analysis of the experimental data given in Table 3, and the Table 9 shows a comparative statistical analysis of experimental data for the four different sets of fabrics. The details of the statistical methods have been reported elsewhere". Essentially, the tables reflect the mean, variance, standard deviation and a simple paired t-test to show whether the unabraded (original) and abraded test results are significantly different. Since the calculated t-values at 95% confidence interval are far less than the tabulated t-values at 95% confidence interval with n-i degrees of freedom (Table 9), it is concluded that there is a significant difference between unabraded and wet-abraded, unabraded and dampabraded. and wet-abraded and damp-abraded fabrics in Co, Go and BR. However, there is no

10 238 INDIAN J. FffiRE TEXT. RES., DECEMBER 1994 significant difference between unabraded and wet-abraded, and un abraded and damp-abraded fabrics in Co/Go but there is a significant difference between wet-abraded and damp-abraded fabrics in c.ic: 4 Conclusions 4.1 Large changes in yarn sett and/or count and fibre content produce. significant changes in bending properties which are more consistent when only one feature such as sett or count is varied. Changing both, as seen here, produce a see-saw effect, i.e. some bending properties increasing and some decreasing with fluctuating magnitude. 4.2 The presence of moisture in the abraded fabrics stiffens the fabrics. The presence of a large amount of water in the sample and the abradant during abrasion testing alters the mode of fabric abrasion significantly. The difference between the wetabraded and the damp-abraded fabrics indicates that the presence of water has a favourable effect on abrasion resistance. 4.3 As a result of slight wear, the coercive couple (Co) for cotton fabrics decreases, whereas for cotton/polyester, wool and synthetic (S I) fabrics it increases in both wet and damp abrasion tests. The flexural rigidity (Go) for cotton and synthetic fabrics decreases, for wool fabrics it increases and for cotton/polyester fabrics it shows inconsistent trend in both wet and damp abrasion tests. The frictional residual curvature (Co/Go) increases for cotton/polyester, wool and synthetic fabrics while for cotton fabrics it shows inconsistent trend in both wet and damp abrasion tests. The bending recovery (BR), however, decreases for cotton/polyester and wool fabrics while for cotton and synthetic fabrics, it shows inconsistent trend. 4.4 The damp-abraded fabrics show higher percentage variation in bending properties than the wet-abraded fabrics when compared with the unabraded fabrics. 4.5 Spearman rank correlation coefficients show significant correlations between coercive couple (Co) and elastic flexural rigidity (Go) for the unabraded, wet-abraded and damp-abraded fabrics. BR and Go for wet- and damp-abraded fabrics, Co/Go and Co for unabraded fabrics, and BR and Co for wet-abraded fabrics also show significant correlations. 4.6 The linear regression correlations are very high for Co vs Go (0.953 and for unabraded and damp-abraded fabrics respectively) and very low (0.082) for wet-abraded fabrics. Correlations between BR and Co/Go are negative: for unabraded, wet- and damp-abraded fabrics (-0.995, and respectively). 4.7 Significant differences are observed between unabraded and wet-abraded, unabraded and dampabraded, and wet-abraded and damp-abraded fabrics in coercive coupld (Co), elastic flexural rigidity (Go) and bending recovery (BR). The frictional residual curvature (Co/Go) shows significant difference between wet-abraded and damp-abraded fabrics. References I Ukponmwan J O. T/lxt Asia, 22(7) (1991) Ukponmwan J 0, J Test Evaluat, 18(6) (1990) Ukponmwan J 0, Colourage (Communicated). 4 Owen 1 D. J Text lnst. 59 (1968) Grosberg P & Swani N M. Text Res J, 36 (1966) 332, Popper P. Ph.D thesis, Massachusetts Institute of Technology, Methods oj"testfor textiles. B S Handbook No.ll. 4th edn (British Standards Institution, London), Livessy R G & Owen 1 D. J Textlnst, 55 (1,964) Owen 1 D, J Text Inst, 57 (1966) Elder H M, Hari P K & Steinhaner D B, J Text InsI,65 (1974) 519. II Elder H M & Ferguson A S. J Text lnst, 60 (1969) ASTM DOOOO ASTM D I I Hall H S & Kaswell E R, Text Res J, 15 (1945) Holme I & Pep pas A. J Text Inst, ) Press 1 J, J Text InSI, 43 (\952) Descheemaeker A, Text World, 104(5) (1954) Siegel S, Non-parametric statistics/or the behavioural sciences (McGraw Hill, New York), Ukponmwan 10. Text Res J, 57(8) (1987) 445.