free web hosting | website hosting | Business Hosting Services | Free Website Submission | shopping cart | php hosting

For any technical consultancy in Textiles and Nonwovens Email N.Balasubramanian or phone 9869716298




Articles by Dr. N.Balasubramanian


ba1ja2@yahoo.co.uk

SPINNING
NONWOVENS
Count variation in yarnYarn IrregularityYarn HairinessIntimacy of Mixing and Blend VariationDevelopments in DraftingTips to get full benefits from Modernisation
Polyester vsPolypropyleneDifferential Hooking By FibrographEffect of Feeding ConditiQns at the Ringframe Upon Yarn Quality and Spinning PerformanceTranslation of strengthStrength and elongation of polyester cotton and polyester cotton blends at different stages of processingMaturity by Micronaire
Drawing Sliver irregularityB L curveMedium and Long term variations of Rotor yarnsInfluence of rotor speed and diameter on properties of Rotor yarnsTop roller Weighting apron spacing and top roller settingUpgrading by Apron drafting
Core spinning AbstractEnergy SavingCharacteristics of Imperfections
Technical consultancy in Spinning NonwovensEnd breakage rateTesting of man made fibres and filamentsArticleYarn diameter specific volume and packing density
Importance of count variability in yarn hardly needs any emphasis. Higher count variability invariably leads to higher strength variability. The weak patches in the yarn lead to frequent end break in further processing, which often reaches annoying levels leading to rejection of bobbins and cones. In latest autoconers, which have settings for rejecting bobbins with count of yarn exceeding beyond certain limits of the nominal, processing of yarns with even slightly high count variations becomes extremely difficult. Winding efficiency reaches unacceptably low levels with such yarns. Higher count variability especially of medium to long length range results in moire like appearance in fabric and increases warp way streaks and weft bars. Ring cuts and soiled ring packages is another problem with higher count CV. To overcome this, wider clearance is kept between ring diameter and full package leading to lower doff weights. With higher count variability, percentage of bobbins exceeding tolerance limits of nominal count increases, leading to sales rejections and market complaints. In shuttleless looms, problem of weft tear is encountered during weaving when count of weft changes abruptly beyond certain limits at the time of pirn change. High count variations in weft are also a cause of warp way fabric creases in processed fabrics like dyed poplins1. Dependence on Length Variability of count depends upon the length of the yarn used for estimating count. Though 120 yd. or lea is normally used for estimating yarn count, sometimes half leas are used especially in coarse polyester blend counts to keep strength measurement within the capacity of strength tester. In very fine counts like 120s, two leas are weighed together to estimate count to achieve better accuracy in weighment. It is well known that CV of count decreases with increase in length but the rate of reduction decreases with increase in length. CV of half lea will be1.414 (i.e.; �� 2) of full lea, if there is no serial correlation between adjoining leas. But there is usually a positive serial correlation because of long term variations. So CV of half lea will be 1.2 to 1.3 times the CV of full lea. The same rule holds good in the case of slivers and rovings. CV of 5yd wrapping of roving is much lower than 1.732 (i.e.; ��3) times CV of ��15 yd. wrapping in inter. The former is usually around 1.4 to 1.5 times that of latter. Tracing the Source of Count Variation Location of source of count variability will be greatly facilitated if wrappings and estimate of CV from the same are based on corresponding wrapping lengths of material at different stages. Thus wrappings and CV of wrappings may be based on 5yd instead of the traditional 15 yd length at Inter. 5 yd length at Inter after drafting will be closer to 120 yd length in yarn than 15 yd length and CV estimates based on the former will be more helpful to show if ringframe is contributing to additional variation. At drawframes, CV of wrappings based on 0.5 yd length will be more useful from the same consideration. Estimation of CV of such lengths can be obtained from modern Evenness testers like Uster tester 3. Norms for CV It is beneficial to set norms for CV not only at Ringframes but also at different stages of processing. Two sets of Norms, one with and another without autoleveller at Drawframe are given in Table 1
Table 1: Norms for CV at different stages
Material Wrapping Length Yd Without Autoleveller With Autoleveller
Yarn 120 2.5 - 3 1.5 - 2
Roving 15 1.5 - 1.8 0.9 - 1.2
Roving 5 2.2 - 2.7 1.3 - 1.8
Drawframe 5 0.7 - 0.8 0.4 - 0.5
Drawframe 1.0 - 1.2 0.5 - 0.7
Drawframe 1 1.5 - 1.6 0.9 - 1.1
Drawframe 0.5 1.9 - 2.1 1.3 - 1.5
Contribution from various processes Ring frame Contribution to count variation from ringframe comes from Variation in mechanical draft between frames Slippage of top roller Stretch of material in creel Variation in Mechanical Draft Variations in mechanical draft come from the use of different change pinions on frames of the same make and drafting system. Some common causes for this defective practice are Lack of sufficient stock of change pinions Using change pinions differing by one tooth on the two sides of a frame, ostensibly to achieve an average count close to nominal. This should be discouraged as it increases variability in count between frame sides. Frames of different makes and drafting systems are used for spinning the same count but the same mechanical draft is not kept on them, Slippage of strand under top rollers arises because of inadequate weighting or improper grip. Count variation is therefore reduced upon conversion of older version of top arms to later versions with higher pressures2, 3. Higher frequency for cot buffing, higher starting diameter for cot up to 30 cm are therefore helpful to reduce count variation4, 5. Disturbances to weighting also comes from worn out springs, leakage of air and improper seating of plunger on rib in pneumatic drafting, leading to count variation. One of the reasons for stretch of strand in the creel is low roving twist. The level of creel breaks can assess this. Misalignment of creel roving bobbin in relation to the creel roving guide is another contributory factor to stretch. Improper location of creel guide rod in relation to the bobbin can also cause stretch. If located too high or too low, stretch takes place when roving unwinds from top most or bottom most portion of roving bobbin. Group Control of Count Group control of count should invariably be preferred to individual frame control as a basis for changing the change pinion except under special circumstances. For this purpose, average count from wrappings taken from all ringframes working on a count should be determined and if the count exceeds the preset limits even after a repeat wrapping, pinion change should be done on all frames. The only exception for this is when frames or drafting systems differ widely in age. Under such conditions, slippage of strand occurs under drafting rollers with older frames and drafting systems leading to a lower actual draft at the same mechanical draft5. In one study2 the mechanical draft had to be reduced from 36.5 to 32.6 to get the same yarn count with the same back material when drafting system was changed from SKF PK211E to SKF PK225. Speedframe Most of the problems enumerated above like variations in mechanical draft between frames and slippage of strand under top roller are sources of count variation at speedframe. In addition, the variation in stretch from start to full position is an added source of variability. Stretch variation during build can be estimated accurately, only if a correct method of estimating wrapping is followed. Traditionally roving bobbin is placed on the top of self-weighted top roller driven by wrapping block, with a spindle inside while preparing wrapping. This method has the drawback of slippage of roving between the self- weighted top roller and wrapping drum. The amount of slippage increases from full bobbin to empty bobbin, often to the extent of 5 - 6 %, leading to errors in estimation of wrapping variation from start to full. A better and more accurate method, free from such error, is to suspend the bobbin from a bobbin holder located at the back of wrap block. The roving drawn from the bobbin is passed through the nip of self-weighted top roller and wrapping drum. Actual studies in the mill showed that wrapping estimated by this method is 5 - 10 % coarser than the one obtained by the traditional method. Draw Frame Major contribution to yarn count variation comes from drawframe. Primarily there are two sources of variability 1. Medium term variations in the sliver and 2. Long term variations in the average hank between frames and between shifts. Medium Term Variations Variations in the length region between 0.2 to 0.7 m, depending on the yarn count and hank organisation influence directly count CV as it corresponds to length of lea divided by draft between drawframe and ring frame. Contribution to this variability comes from irregularities in breaker sliver, which do not get reduced by doubling in finisher. Long Term Variations Long term variability in the hank of drawing sliver arises from variation in average weight of lap weight in lap feed systems and weight of feed sheet to card in chute feed systems, that occur from shift to shift. The piano feed regulating motion in scutcher and pressure switch control in chute feed to card operate on the basis of volumetric control and the degree of openness of the material therefore affects the extent of control. In piano feed belt shift would occur when the sheet becomes bulkier (because of better openness) even though weight per unit length remains the same. Likewise, the pressure developed in the chute would be higher for the same weight with well-opened material. Thus it is necessary to standardise the evenness roller to inclined lattice setting in scutcher and speed of beaters and pressure switch setting in the chute, based on factors like type of cotton, whether the feed is directly from bales or from a sandwich mixing from pre-opened material. This will minimise variations caused by openness of material. The variation in average hank of card sliver from shift to shift have less chance of getting reduced from doubling at draw frame, as slivers made in one shift do not get doubled with those in the next shift. But the variability due to this can be kept down by increasing storage capacity of laps and card sliver cans so that material made at longer intervals of time get doubled. But most of the mills do not have such storage space. Moreover material stored for long lengths run the risk of contamination with dust and fly. Autoleveller Modern draw frames are equipped with autoleveller and sliver monitor to reduce both type of variability mentioned above. The autoleveller brings down the variability in sliver lengths beyond 0.3 - 0.5 m. Further, the rate of reduction in CV of wrapping with increase in wrapping length is steeper with autolevelled product. This will be clear from Fig.1, where the variance length curves of Finisher drawing slivers made on the same drawframe with and without autoleveller are compared. The amount reduction in CV with autoleveller increases as wrapping length increases from 0.5 to 3 m. Variance length curve of Breaker drawing sliver (without autoleveller) fed to this draw frame is also shown plotted in the same Fig. The figure shows that doubling in the draw frame brings down the CV with length 0.5 m and about and the order of reduction increases with length. So the variance length curves become steeper from Breaker to Finisher. The short term irregularity in sliver is however not reduced by doubling or autolevelling. Norms for CV of wrapping of drawing sliver at various lengths with and without Autoleveller are given in Table 1. Functioning of autoleveller gets affected by zero setting, obstruction to free movement of scanning roller, improper functioning of servo motors, drives, solenoids and related parts. Leveling action of autoleveller has therefore to be checked at regular intervals. A method usually followed in the mills is to feed 9 ends and 7 ends of sliver and check the wrapping of outgoing sliver. The extent of deviation of hank from nominal under such circumstances is taken as leveling sensitivity of autoleveller. In normal practice, however, such wide variations in input hank seldom occur and the effectiveness of autoleveller in leveling lower order of variations would be of greater interest. The best method of judging this is by comparing CV of 0.5, 1 and 3 m lengths of material with and without autoleveller. Sliver Monitor Sliver monitor consists of a sensor fitted to the calendar rollers of draw frame to check the hank of the outgoing sliver If the actual hank deviates from the nominal by more than the preset limits the drawframe will be stopped. The limits of deviations from nominal hank at which drawframe will be stopped can be varied. It is normally advisable to set it at +/- 2 %. In some makes, sliver monitor can also be set to stop drawframe if CV of wrapping goes beyond certain limits. If the drawframe stoppage due count exceeding preset limits is high then the hank of breaker sliver should be checked to find if it is deviates considerably from normal. If this is not the case, then the functioning of autoleveller has to be checked. Thus a step by step approach should be followed. Mechanical Condition of Draw Frame Poor mechanical condition of draw frame, coupled with inadequate weighting for top rollers have sometimes been traced as a source of count variation. When older versions of Whitin drawframes were replaced by later versions of Laxmi Rieter drawframes, among other noticeable benefits found by the mills, was a significant reduction in count CV of yarn6 as shown in Table 2.
Table 2: Effect of mechanical condition of Drawframe on Count and Strength CV of yarn (30s)
Drawframe Type Age Mechanical Condition CV of
Count%		 CV of
Strength%
Mixing1 Mixing2 MIxing1 Mixing2
Whitin J5 Old Unsatisfactory 4.9 3.8 11.4 8.2
Laxmi Rieter DO2S New Good 4.4 3.0 8.1 4.7
Card Sliver Though card sliver is characterised by high variability of wrapping, its contribution to yarn count CV is not that significant because doublings in drawframe reduce the variations substantially. Autolevelling at card is therefore not found very beneficial in bringing down count CV of yarn. This will be clear from Table 3 where card slivers made with and without autoleveller were processed up to ringframe and variability at different stages assessed.
Table 3: Effect of Autolevelling at Card on Yarn Count CV % (20s)
U % of Card sliver CV % of 1 yd of Card sliver CV % of 1yd of Finisher Drawing sliver Count CV % of Yarn
With Autolevelling 3.8 1.2 1.3 2.6
Without Autolevelling 4.1 7.0 1.6 3.2
Though autolevelling at card brings down CV of 1 yd wrapping of card sliver from 7 to 1.2, CV of 1 yd of finisher drawing sliver comes down only marginally from 1.6 to 1.3 and count CV of yarn drops marginally from 3.2 to 2.6. U % of card sliver also does not have much effect on count CV of yarn. This will be clear from a study where cards giving low U % and wrapping CV % and those giving high U % and wrapping CV % were separately channelised up to yarn. The results (Table 4) show that even with large differences in U % and wrapping CV % of card sliver, there is no significant difference in yarn count CV %.
Table 4: Effect of card sliver irregularity on Yarn count CV % (20s)
Property Cards with low sliver variability Cards with high sliver variability
Card sliver U % 4.3 7.2
CV of 6 yd wrapping of card sliver 6.1 7.3
CV of Yarn count % 2.8 2.9
Comber Sliver High comber sliver U % arising from piecing wave can affect yarn count CV, particularly if single post comber drawframe passage is used. This will be clear from the following study where combers giving low and high U % were channelised up to ring frame separately and checked for yarn count CV. The results given in Table 5 show that yarn count and strength variations are higher with yarns made from combers with higher variability.
Table 5: Effect of comber sliver U % on variability of count and strength of Yarn (40s)
U % of comber sliver CV of yarn count % CV of yarn lea strength %
3.7 1.75 4.12
6.9 2.78 7.96
Medium Term Variations in Yarn Count variation is traditionally used to indicate variability in 120 yd lengths. Variability in shorter lengths are also equally important though they do not form part of quality monitoring activity in most of the mills. Variability of weights in lengths in the region of 0.5 to 10 m affect the appearance of fabric, lead to moire like defects and contribute to weft bars and warp way streaks. U % of drawing sliver has a direct influence on the variations in these lengths. This will be clear from the results of a study in a mill given in Table 6.
Table 6. Contribution of drawing sliver U % on medium term variations in yarn (20s)
Draw Frame
settings,
mm Fr/Back		 U % of Drawing sliver CSP U % CV of 1 m % CV of 3 m % CV of 10 m % CV of 120 yd %
49 - 56 6 2044 15.6 8.0 6.5 4.6 2.5
44 - 53 4.6 2116 13.8 6.2 5.2 4.0 2.6
Drawing sliver U % was brought down substantially in the mills by optimising roller settings. When this material was channelised up to ring frame, medium term variations in the yarn as indicated by CV of 1 to 10 m lengths came down markedly though CV of 120 yd length showed little improvement. The fabric appearance also improved markedly. This shows the importance of monitoring CV of yarn count in length regions 1 to 10 m. This is conveniently obtained in later model Uster Evenness testers like UT3. References: 1. Warpway Creases in fabrics N.Balasubramanian and C.Chatterjee, Indian Textile J., 1996 Oct, p66 2. Benefits from modernisation of ring frames from second generation Top arms M.Balakrishnan and N.Balasubramanian, BTRA Scan 1980 Sept., XI, p4 3 Yarn quality improvements from second generation Top arms at Ring Frame G.Janakiraman and N.Balasubramanian, BTRA Scan, 1987 Dec, XVIII, p 9 4.Cot buffing at Speed Frame - A useful measure to reduce count variation S.K.Sett, G.V.Aras and N.Balasubramanian, BTRA Scan, 1984 March, XV, p 4 5.Contribution of Ring Frame drafting condition to yarn count variability in fine counts N.Balasubramanian and G.Janakiraman, Indian J. of Fibre & Textile Research, 1990, 15, p 198 6. Influence of mixing characteristics and Draw Frame condition on yarn quality G.V. Aras and N.Balasubramanian, BTRA Scan, 1982 June, XIII, p 10
  • Yarn Irregularity - Concept and measurement
    N.Balasubramanian
    Retired Joint Director & Consultant,
    Importance of Irregularity
    Irregularity is the most important quality characteristic of yarn, its importance arising from the following factors ● Irregularity has a profound influence on appearance of yarn and fabric. More regular the yarn, better will be the appearance and aesthetic value of the product. As a result, better sale value can be achieved. With advent of Uster Evenness tester and Uster standards, importance of irregularity has become even higher. Buyers insist on yarns meeting certain Uster standards (such as 25% or 5%). In export market, especially in developed countries, Yarns have to meet Uster 5 or 10% standards. Mills who produce consistently more regular yarn get a premium in the selling price
    ● Regularity contributes to a smoother feel. In apparel and most of other textiles, smoothness is most desired characteristic. Sale value of fabric is dependent, among other things, on smoothness.
    ● Regular yarns will have fewer weak places and, as yarn breaks at weakest place, will have a better strength. Better strength realisation from fibre can be achieved if regularity of yarn is improved. It is for this reason good mills, which produce consistently more regular yarn, are able to produce a yarn of much higher strength from the same cotton
    ● Because of the lower incidence of weak places, fewer end breaks are encountered with regular yarns in weaving preparatory, weaving and knitting. Efficiency in these processes is improved leading to higher productivity. Importance of regularity will be appreciated from the better performance achieved by rotor spun yarns in weaving processes compared to ring spun yarns. Rotor yarns have a lower strength than ring yarns but have a better regularity (short, medium and long term). Though mean strength is lower, strength of weak place is higher in rotor yarn than ring yarns and so it performs better.
    ● Fabric defects and rejections are critically influenced by irregularity of yarns. Periodic and quasi periodic irregularities in yarn result in warp way streaks and weft bars in woven fabrics leading to fabric rejections1,2. Yarn defects like slubs, crackers, long thick places and long thin places downgrade the fabric and cause considerable value loss. Mills which produce more regular yarns therefore get better realisation and contribution and as a result higher profitability
    Types of Irregularity
    ○ Weight per unit length
    Variation in weight per unit length is the basic irregularity in yarn. All other irregularities are dependent on it. This is because weight per unit length is proportional to fibre number i.e; number of fibres crossing a section of yarn. Variations in fibre number is the factor influenced by drafting. So any improvement in drafting or spinning will first reflect in improvement in variability of weight per unit length. Count variation in yarn has been found to have profound e ffects on weaving and finishing. High count variation has been a cause of weft way tear in Sulzer looms. Weft count variation has been found to cause warp way creases in finished fabric3.
    ○ Diameter
    Variability in diameter is important because of its profound influence on appearance of yarn. Variations in diameter are more easily perceived by eye. Latest models of evenness testers have therefore a module for determining diameter variability. Diameter variability is however caused by weight variability. As twist has tendency to run into thin place, variability in weight gets exaggerated in diameter variability.
    ○ Twist
    Twist variation is important because of its influence on performance of yarn and fabric dyeability and defects. Soft ends are a major cause of breaks in weaving preparatory and loom shed, arising from twist variations. Soft twisted yarns take more dye and so uneven dyeing is caused by high twist variation. Weft bars and bands are also caused by low twisted yarns. Twist variations come from slack spindle tapes, jammed spindles. A certain amount of twist variation is also present along the chase of cop.
    ○ Strength
    Importance of strength variation is easy to appreciate. Yarn breaks at the weakest element and so yarns with high strength variability will result in high breakages in further processes. Strength variability is partly dependent upon count variability and partly upon spinning conditions and mechanical defects.
    ○ Hairiness
    High variation in hairiness leads to streaky warp way appearance and weft bars in fabric. More light will be reflected from portions of weft where hairiness is more and this leads weft bands. High hairiness disturb warp shed movement in weaving and result in breaks, stitches and floats. Among other factors, worn out rings and travellers, vibrating spindles, excessive ballooning and variation in humidity in spinning room cause variations in hairiness from bobbin to bobbin
    ○ Colour
    Variations in colour of yarn cause batch to batch variation in fabric colour, which leads to rejects. This is particularly critical in cloth marketed to garment units. Variations in colour of yarn and fabric are caused by variations in colour of cottons used in mixing. Larger lot sizes made from a large number of bales help to mitigate this problem. Checking of cotton and mixing for colour will also minimise large variations in colour. HVI testing has therefore a module for checking colour in cotton and Uster�s latest models have a module to detect colour variations in yarn.
    ○ Fibre ends per unit length
    Irregularity in thickness or weight/unit length arises primarily from short term variations in fibre end density in the yarn or sliver. We will first consider the case where all fibres are of the same length, which is the case with man made fibres and fibres are straight and parallel to the axis of sliver/yarn. Consider two sections of sliver A and B separated by a distance l.
    Fig 1:Arrangement of fibres in a yarn

    All fibres that have their left hand ends in AB will cross B and no other fibres will cross B. Number of fibres crossing section B (which is equal to thickness or weight/unit length) is given by
    n l
    Where n = number of fibre ends per unit length or fibre end density and l is fibre length.
    .Variability in thickness of yarn is therefore dependent of short term variations in fibre end density. This formula is a particular case of more complicated relation connecting fibre end density and thickness.
    Variable fibre length
    Number of fibres of length l crossing section B, N , is the number of fibres whose left hand ends lie in AB. This is given by
    N = n l f(l)dl
    where
    n = fibre end density and f(l) = frequency of fibres of length l.
    The total number of fibres is given by
    =∫0lmaxn f(l) dl
    where
    Irregularities in fibre end density with wavelength less than fibre length get masked by the fibre length and do not show up or show with reduced amplitude in thickness irregularity4,5.. Further, with variable fibre length, short length periodicity in left hand ends gets considerably reduced in amplitude with right hand ends. Let wavelength of fibre end density in left hands be = λ Mean fibre length = L
    Amplitude of periodicity in left hand fibre ends = r1
    Amplitude of periodicity in thickness = R
    Amplitude of periodicity in right hand ends =r2
    Fig shows how the amplitude of periodicity in fibre end density in left hand ends gets reduced in the amplitude of thickness and of fibre right hand ends,when wavelength is lower than fibre length.
    amplitude reduction Fig 2. Reduction in amplitude of periodicity of left hand ends, in thickness and rigth hand ends

    When λ is less than fibre length L, amplitude of periodicity in thickness reduces to 0.15 and that of right ends to .01 of the amplitude of periodicity in left hand ends. This shows the advantage of variable fibre length in curbing short term periodicities as there is a reversal in direction of drafting from one process to next and fibre front ends gets fed as rear ends. This advantage is not there with man made fibres with constant staple length. And it is for this reason we encounter weft bars in man made fibre material due to short term periodicities in fibre ends introduced at speed frame2.
    Range
    Measurements of weights or diameter are classified in number of rows, each column containing about 4, 5, 6,or 7 items as shown below
    x11x12x13x14
    x21x22 ......
    ............
    ............
    xn1 ......xn4
    .
    Range, which is the difference between maximum and minimum value is determined in each row. Mean range,rm for all the rows is then calculated. Range% = (rm/xm)100. Higher the range%, higher the variability. If the irregularity is random, there is a relationship between range and standard deviation, σ. Let mean range be rm. Then σ = rm/a. The value of a is dependent upon up on the number of units in a row and its values are shown in Table 1 below.
    TABLE 1
    Number in a row235 61012
    A1.128112.0592.3262.5323.0783.25
    To check the accuracy of this method, S.D was calculated from range for the 3 sets of date given in the paper by Burkhalter7. The data was rearranged in 7 rows with 6 weights in each row. The estimated and actual S.D. (σ) are given in Table 2 below
    Table 2 Comparison of SD estimated from Range and actual
    Set NoRangeestimated from rangeActual
    10.240.0950.1
    262.372.87
    30.4560.1800.165
    The agreement is reasonably good for mill work.
    � Mean Deviation%
    Let the individual weights of specimen be as given in Table 3.
    Table 3
    Weight of specimen DeviationSquare of Deviation
    x1x1 -xm(x1 -xm)2
    x2x2 -xm (x2 -xm)2
    .....
    .....
    .....
    .....
    xnxn - xm(xn -xm)2
    . Deviation of each specimen weight from mean is determined, ignoring the sign. Let mean be xm
    Mean Deviation %,
    MD% =1xm×(∑n1|x - xm|/n) ×100 Mean deviation % is measured by linear integrator of Uster venness tester. If a recorder chart of weight variations is available, Mean deviation is given by the area enclosed by the chart/ average value. The area can be determined by planimeter6. ● Standard Deviation Deviation is squared as shown in 3rd column of table and the average of square of deviation is determined. Its square root is determined as shown below.
    σ = (1/(n-1)×√∑n1(xi - xm)2
    Burkhalter7 has suggested a quick method for estimating standard deviation. Since 67% of data normally fall within � σ , he suggests that standard deviation can be estimated with good approximation by determining half the difference between values within which 67% of data lie. However this approximation will hold good only for normal or near normal distribution.
    ● Coefficient of Variation, CV% Coefficient of Variation, CV% = . CV is a better measure of irregularity than MD as it gives greater weightage to portions moved far away from mean. However MD is much easier to calculate and there is a fairly good correlation between CV and MD. Later models of Evenness testers are equipped with linear and quadratic integrator and give both U% and CV%.
    In Evenness testers like, Uster, continuous measurement of weight/unit length/mass is carried out where
    x = instantaneous value of mass
    = Mean value
    In such a case
    MD% = 1xmT∑0T|xi - xm|
    CV% = 1x(√∫0T(xi - xm)2 dt ×100
    U% or CV% measured by Uster Evenness tester is a weighted running average of variability of specimen, the weighting being more for lengths of specimen which have just passed through the tester( i.e; weighting is inversely proportional to the time that has elapsed after the passage of material through the measuring head) ● If irregularities are random and follow normal distribution as shown in Fig 3 below) then CV = 1.25 MD.
    normal
    Fig 3 : Normal distribution

    � If frequency distribution is triangular as shown in Fig 4 below, CV/MD is lower than 1.25
    Triangular
    Fig 4: Triangular distribution

    ● If frequency distribution is skewed as shown in Fig 5 , then CV/MD is greater than 1.25
    skewed Fig 5: Skewed distribution
    ● If irregularity distribution is wholly periodic(Fig 6 ) then CV/MD = 1.117
    periodic
    Fig 6 : Periodic distribution

    ● As normal yarns contain extra irregularities due to drafting waves and mechanical faults, CV = 1.27 to 1.4 times of MD. Since the type of frequency distribution does not vary much from yarn to yarn, it is safe to estimate CV by multiplying MD by 1.3 Typical frequency distribution of a normal yarn in relation to ideal yarn with random distribution is shown in Fig 7.
    Idealand normal Fig 7 Comparison of normal yarn with ideal yarn

    ● Original models of Uster integrators were determining MD (U%) and CV with a certain amount of approximation. Later with improvements in electronics, later models were designed to determine CV precisely. But U% is still estimated with a certain amount of approximation. So the relationship between CV and U depends upon the model of Uster Evenness tester and type of irregularity in yarn as shown in Table 3 below.
    Table 4 Type of irregularity Model of Evenness tester CV/U Symmetric with a single peak GGP (very old model), Uster1, 2 and 3 ≈1.25 Symmetric with 2 or more peaks GGP ≈1.25 Uster 1, 2, 3 and 4 > 1.25 Asymmetric GGp, Uster 1, 2, 3, and 4 >1.25 with a higher conversion factor for Uster 1, 2, 3 and 4
    ● Exceeding Frequency
    This is defined as the frequency f with which a particular linier density is exceeded in a unit length of yarn and is given by
    f = A/L
    where A= number of times a specified value of CV% is exceeded in a given length L of yarn. Number of end breaks in subsequent processes is related to exceeding frequency.
    ● Fraction Defective/Deviation rate
    This estimates the percentage of material that lies beyond certain limits away from the mean on either side. Weights removed far away from the mean are mainly responsible for end breaks and rejections. Deviation Rate % = Where li is the length of ith piece which falls beyond prescribed limits and L is total length of yarn under study
    One of the conditions stipulated in purchase of yarn is that not more than 5% of bobbins will have a count beyond mean � specified percentage. Supplies that do not meet this specification are liable to rejection. The following examples show that yarns with a lower variability will be able to meet more stringent requirements.
    ● Example
    Count = 30s. The supplier�s requirement is 30s � 1.5. ● Mill A - CV of count = 2.5%
    Standard Deviation, σ = 0.75
    % bobbins falling outside 30 � 1.5 = 5 (From Table 5 below) Mill B - CV of Count = 5%
    SD σ = 1.5
    % of bobbins with count outside 30 � 1.5 = 31.7.(Table 5) Mill A which has a lower CV% also has fewer % bobbins falling beyond � 1.5 of mean. This shows the utility of checking the % of bobbins falling outside certain limits on either side of mean.
    5:Proportion of values beyond a limit with Normal distribution
    Number of SD on each side of mean Proportion of values lying outside limits%
    0100
    0.524460
    131.7
    1.64510
    1.965
    24.56
    32.7
    ● Irregularity Testers
    ○ Visual Assessment
    An estimate of yarn irregularity is obtained by winding the yarn on a black board and comparing it with ASTM standards. A good correlation is found between yarn appearance grade and irregularity as measured by Uster Evenness tester8 ○ Cut and weigh Method
    Yarn is wound on a drum and pieces of known length are cut with the help of a template. The cut pieces are weighed on a sensitive balance and CV% is estimated. This is a time consuming method and is adopted only in special research program. Moreover accuracy has to be maintained in cutting and weighing. Only advantage is it is a direct method and does not involve any assumptions.
    ○ Mechanical Type tester
    Tongue and Groove tester
    This is used mainly for slivers and rovings. This consists of two rollers one with a tongue and another with a groove, in between which the sliver is passed. Tongue roller moves up and down depending upon the weight of sliver. The roller movement is amplified recorded by pen recorder and integrated to find the CV%. Groove width ranges from � in for coarse, � in for medium and 1/16 in for fine slivers. This is also time consuming. Further errors are introduced because of variation in compactness of sliver. A similar type of instrument was converted to a microprocessor based one by De9. The displacement of the lever carrying the tongue roller is picked up by transducer to convert displacement into DC voltage. This is converted to a 8 bit digital word by a converter and digitally processed to estimate CV. The equipment is designed to determine variance length curve of a sliver. Good agreement with Uster Evenness tester is reported.
    � Shoe type tester
    This is mainly for yarns. Yarn is passed between a hard steel shoe and a a half round ball fitted on a light lever suitably fulcrumed in the middle. The diameter variations in yarn cause the lever to move up and down which is amplified by means of a light falling on a mirror fitted on the lever. This is amplified and CV% is estimated by an integrator
    � Photo Electric tester
    A light source is made to fall on yarn, running in front of the slit. The image of yarn is made to fall on a photo cell. The amount of light falling on photo cell varies depending upon diameter of yarn. The current generated by photocell is amplified and integrated.
    � Capacitance type tester
    The material is passed between two plates of a condenser at a known speed. Capacity of condenser varies depending upon weight variations in the material. The voltage generated is amplified and a continuous record of variations is recorded on a recorder. An integrator is used to determine MD%(U%) or CV%. Uster and Fielden walker testers are the most commonly used equipments based on this principle. Merits of this system are that it attempts to measure a quantity proportional to weight per unit length of material. Condenser slots of different sizes are used for slivers, rovings and yarns.
    The results of yarn irregularity measurement are affected by ● Fibre inclination � Fibre inclination in yarn may vary from place to place. This will affect the cross-sectional area for the same number of fibres in cross-section
    ● Fringe Effect - The fields of force between the plates are curved at the edges and as a result the material is detected before it enters the space between the condenser plates. This affects linearity of mass voltage relationship leading to errors in measurement. This can be overcome by incorporating a guard plate prior to normal condenser.
    ● Filling proportion (λ) - This is the ratio between material and air space and is given by λ = (d/D)100. This should be kept much lower than 1. Otherwise relation between change in capacitance and mass of material becomes non linear. Locher10 showed that the change in capacity of a condenser as a result of material of dielectric constant ε is given by
    1/C =1/C1 + 1/C2
    where C1 = C0 ε D/(D-d), C2 = C0 ε D/d. Since ε0 = 1 and δ C = C - C0 δ C/C0 = (ε - 1)/(( ε ((1/λ) - 1) + 1). When ε is much lower than 1, δ C/C0 is close to λ .Relation between change in Capacitance and dielectric constant is shown in Fig 8 below
    Fig 8 : Relation between change in capacitance and dielectric constant

    At relatively high dielectric constants (above10) and low filling proportion, capacitance of condenser (and instrument reading) is primarily affected by weight per unit length variation of material and is not much affected by variations in dielectric constant, caused by variations in blend proportion and moisture content. Textile materials usually have a dielectric constant above 12 and therefore variations in dielectric constant have only a small effect on measured value. However Hearle and Walker11 showed that only cotton and viscose have dielectric constants near 10 and wool has a value between 6-7 and synthetics less than 5. Therefore for these materials variations in RH and blend proportion are likely to affect instrument results. ● Inadequate conditioning and variations in relative humidity during testing.
    Foster12 showed that the error due to uneven moisture content can be kept down if following precautions are taken
    1. If testing room humidity is below 67%, its variations should not exceed 16% during conditioning and if it is above 67%, the variation should be below 8%.
    2. Yarn and roving bobbins should be conditioned in testing room with a free circulation of air from all sides of each bobbin. Conditioning for yarn may be about an hour, for roving about 2 - 3 hours. Slivers should be tested immediately without conditioning, after discarding a few layers on the top. Relative humidity and temperature vary between production area and testing room. Therefore irregularity increases with conditioning in test room, because of uneven penetration of moisture into inner sliver layers.
    A typical study13 showed sliver irregularity increases with conditioning time(TABLE 6) . Conditioning is therefore not recommended while testing slivers.
    Table 6: Increase in U% of finisher drawing sliver with conditioning time
    . Conditioning timeU% CV(1m)%
    Without conditioning2.51.31
    1hr2.991.42
    4hr3.271.51
    8hr3.851.49
    24hr3.881.39
    ● Shape factor of material - Fibre assembly should be as close to cylindrical as possible and should not be asymmetric. A slight twist is given to multi filament yarns prior to testing for this purpose. ● Material should be kept away from both plates or kept in touch with one of the plates.
    Principle of operation of Uster Integrator
    Grosberg and Palmer14 showed that in simple terms, the operation of Uster integrator is based on the use of resistance - Capacitance circuits. The integrator consists of two such circuits as shown in Fig 9 below
    Fig 9 : Integrator circuit of Uster Evenness tester
    A voltage ft is developed upon insertion of material in between measuring plates of the condenser in the main instrument. This is impressed upon R1C1 circuit. R1 and C1 are very large and as a result, condenser C1 charges slowly to the mean value of fluctuating voltage ft. The voltage across R1 at any time is equal to the deviation of ft from the mean value. The voltages across R1 are impressed across R21 without taking sign into account in the case of linear integrator. In the case of Quadratic integrator, the deviations are squared and impressed across R2'. The voltage across C2 charges slowly to the mean value of deviation(in linear integrator) and mean value of square of deviation (in case of Quadratic integrator) and is displayed as irregularity value. Thus the equipment measures CV(8mm, kv/α1) where v is material speed α1 is time constant of R1C1 circuit and k is a constant.
    Improvements in Uster Evenness Tester
    Over the years Uster has brought out several improved models like Uster 3 and Uster 4 and lately Uster 5. The important additional features in these are
    1. Maximum testing speed has been increased up to 800m/min
    2. Variance length curve is automatically determined and the resolution is increased. Variance length from cut lengths of 2 cm to 600 m can be measured in latest models
    3. Spectrogram range is vastly increased (from 1/5 of measured length to a maximum of 2200 m) and spectrograms of different bobbins can be placed side by side for comparison in a 3D plot
    4. Frequency distribution
    5. Module for hairiness measurement is incorporated. A homogenous ray of parallel light falls on the yarn and rays of scattered light from protruding hairs reach the detector and provide a measure of hairiness. Yarn hairiness is expressed in terms of a single value
    6. Module for determination of diameter variation by use of an opto electronic sensor is incorporated. A parallel infra red light beam falls on the yarn and the image is evaluated by line scan camera and corresponding algorithms. The measuring sensor determines absolute mean diameter (over a measuring field of .3mm) and diameter variations CV, variance length curve and spectrogram.. The roundness of the yarn and density of yarn are also measured
    7. Facility to measure various characteristics of slub yarns made on a fancy slub yarn ring frame. Slub frequency, spectrograms of base and slub yarn and simulation of yarn in woven and knitted fabric and on yarn board are useful features
    8. Presence of dust, trash and foreign fibres is estimated
    9. Manual involvement in testing substantially reduced. Automatic transfer of yarn takes place from the package changer and insertion of yarn into the desired slot is also done automatically.
    Prototypes
    QQM3 is a mobile Evenness measuring equipment15 which can measure yarn evenness at the production stage. It has 2 optical sensors of 2 mm width and uses infra red light for projection of yarn on the sensor. and measuring speed is limited to 300m/min.Comparison of evenness by optical and capacitance testers has been done by Sparavigna et al;16 A good agreement is found between measurements of diameter by Uster 4 and QQM and image analyser. Determination of diameter from Uster 4 data shows that it follows bimodal distribution as against unimodal obtained by the instrument. Further CV estimated from extracted data from Uster4 is higher than that given by the instrument.Carvalho17 et al; employ a mass parametirzation system using1 mm capacitance sensors. New parameters like integral deviation ratio and signal processing techniques were used to characterize irregularity apart from U%, CV%, spectrogram and faults. Ewald and Landstreet18 use beta radiation to measure uniformity of slivers, rovings and yarns. Thalium204 is used as Beta ray source, ionized airfilled chamber is used as detector and an electronic integrator is used to determine CV.
    � Pneumatic Method
    Yarn is passed through a narrow tube into which a stream of air is forced. Air flow rate is then measured. Air flow rate is affected by the mass of yarn in the tube. Variation in air flow rate is therefore a measure of irregularity. � Acoustic Method
    Yarn is moved through a sound field between a sound generator and a pick up device. Time taken for sound waves to move across the gap is measured electronically. Transit time of sound is dependent upon the weight of yarn in the gap.
    � Conductometric method
    Yarn is passed through a bath of electrically conducting fluid. When a DC voltage is applied against a section of yarn, current is produced proportional to the conductivity and therefore to weight per unit length of the section. The variability in current is therefore used as a measure of irregularity.
    While CV and MD are the common measures of irregularity, a complete picture of irregularity is obtained by 1 Coorelogram 2 Varaince Length Curve and 3 Spectrogram
    ● Correlogram
    A number of papers have been published on using correlogram for characterizing irregularity19,20,21.
    If r(x) is the Autocorrelation coefficient of thickness( or weight per unit length)of yarn at points 'x' apart, then the plot of r(x) against x is Correlogram.
    For determining auto correlation coefficient, thickness or weight/unit length is measured at specified intervals, say �s,� along the length of yarn. The results are rearranged as shown below
    Table 9
    ● t1●t1+s
    ● t2●t2+s
    ● t3●t3+s
    ● ..●..
    ● ..●..
    ● ..●..
    ● ..●..
    ● ..●..
    ● ti●ti+s
    ● ..●..
    ● ..●..
    ● ..●..
    ● ..●..
    ● ..●..
    ● ..●..
    ● ..●..
    ● tn●tn+s
    Auto correlation coefficient (rs) of thickness between points separated by distance s is given by rs =
    n(Σti×ti+s - Σt i×Σti+s)
    divided by
    [(n×Σti2 - (Σti)2)×(n×Σti+s2 - (Σti+s)2)]0.5
    A plot of rs against s for various values of s gives the correlogram.
    For lengths below fibre length, correlogram is determined by the length distribution of fibre. Since part of the fibres are common to both crossections, there will be a positive auto correlation coefficient as indicated by principle of common elements.
    Fig 10 : Arrangement of fibres in yarn
    Consider two sections of sliver, A and B, along the length of sliver distance x apart. Let AC= l. All fibres of length l that have their right hand ends lying in BC will cross both sections A and B. Their number is = (l - x ) n f(l)dl where
    n = fibre end density
    f(l) dl = proportion of fibres of length l
    Total number of fibres is given by
    Total number of fibres crossing A is given by
    ∫xlm(l - x)n f(l) dl Correlation coefficient between points x apart (rx), is given by
    l max is maximum fibre length and f(l) is frequency of length l.
    The shape of correlogram in the region below fibre length is shown in Fig 11
    Fig 11 : Correlogram in the region below fibre length
    However, normal yarns contain drafting waves and other irregularities. Drafting wave is a damped wave with successive amplitudes diminishing. The correlogram of a normal yarn is more likely to be as shown in Fig 12
    Fig 12 : Correlogram of normal yarn
    Correlogram can be determined by measuring thickness or weight per unit length of yarn at successive intervals of preset distance. Recorder chart can be used for this purpose. The date is rearranged to determine serial correlation. A computer program can be used to determine the correlogram from the data. Uster Evenness tester can also used for determining correlogram22. The yarn after passing through the slot is taken round a pulley placed by the side and passed through the slot again. By altering the distance of pulley from the slot the distance between two successive points can be varied.
    ● Variance Length curve
    Variance length curve is a graph relating variance of weight of length of yarn and the length. This is a very useful tool in characterising the various types of irregularities in yarn and will assist in locating processes which require improvement. A number of papers have appeared on Variance length curve23,24,25,26,27,28 Irregularity in a yarn is given by Variance(A L). Here L denotes the total length of yarn taken for study and �A� denotes the length of pieces cut from the yarn and weighed. IfA is varied keeping L constant, variance length curve obtained is known as �Between length� curve or B-L curve. On the other hand, if A is kept constant and L is varied, the curve obtained is known as �Within length� curve or V-L curve. Total variance of yarn, B(0) or V(∞) is given by
    B(0) = V(∞) = B(L) + V(L).
    With increase in length, B(L) reduces initially slowly, then rapidly and afterwards slowly to reach a value asymptotically. V(L is a mirror image of B(L). In Uster tester, within length is given by kv/α1. Therefore U% increases with speed as shown in Fig 13
    Fig 13 : Effect of material speed on U%
    At lengths shorter than longest fibre length, B-L curve is influenced by fibre length distribution of fibre, as some fibres will be common to both sections. Variance length curve is related to correlogram of yarn as indicated below B(a) = B(0)×(2/l2)×∫(a - x) rxdx
    where denotes auto correlation coefficient of thickness between points x distance apart, B(a) = Variance of weight of lengths �a� long. When �a� is than lt, where lt is half the length biassed mean fibre length of fibre,lc = length biassed mean length, 2lt and If there are no other irregularities( no extra auto correlation coefficients) . B(a) = B(0)
    where is mean fibre length. For lengths longer than longest fibre length,
    B(a) = B(0 � ])
    Where Vt = square of coefficient of variation of fibre length in tuft (Fibrograph) diagram This enables one to find out how much the actual variance length curve deviates from the ideal.
    Fig 14 shows the variance length curve for random yarn. determined from the above equations
    Fig 14 : Variance length curve of random yarn
    Variance length curve shows the amount of short, medium and long term variations present in a yarn. While short term variations come from ring frame, medium term variations arise from roving frame and long term variations from drawframe. Grosberg29 showed that B-L curve of worsted yarn can be predicted from the total variance of slivers at different stages of processing. Malatinszky and Grosberg30 found good agreement between B-L curve predicted by the above method and actual for worsted yarns. Balasubramanian31 applied this method to cotton yarns and found it to give too large a value compared to actual value. In a later article an explanation32 is given why this method is not applicable to cotton yarns . The assumption that fibre ends are arranged at random at very short lengths (of the order of L/D) does not hold good for cotton slivers and hence the prediction of B-L curve of yarn from variance of intermediate products is not possible.
    Measurement
    Cut and Weigh method
    Cut and weigh method is the classical method for determining variance length curve. The material is wound under constant tension on a drum of known circumference and templates of different sizes 1.e; .01, .02, .05, 0.1 0.2 m are used to cut the yarn into required lengths and the pieces are weighed on a sensitive balance. To minimize errors due to any periodicity coinciding with diameter of drum, pieces were collected from random rows and weighed. For longer lengths, the material is wound on a drum of 1 m circumference and single cut is made across the width to get pieces of 1 m length and cuts were made after 5 or 10 wraps to get pieces of 5 and 10 m lengths. A special arithmetic procedure was suggested by Grosberg and Palmer26 for getting an unbiased estimate of variance. Bandyopadyay33 suggested a further improvement taking into account variation of local means. As per this method B(x,y) = - W) Van Zwet28 has proposed a quicker method of estimation of B-L curve from the the points read from recorder chart. Nienhuis suggested that CB(G) curve enables easier comparisons between material from different stages of processing than CB(L) curve, where G is the weight of yarn of length L. O�coneel34 et al modified the Pasfic Evenness tester to record the measured signals in a magnetic tape and feeding the tape to a digital computer to determine weight per unit length and its variance. The method was used to determine variance length curve of yarn. Moderate agreement is found between the measurements of the instrument and cut and weigh method for medium length range
    Cut and weigh method is highly time consuming and so evenness testers were equipped with facility to determine variance length. The old Uster Evenness testers were fitted with �inert� setting which is used to determine variance length curve. Grosberg and Palmer14 examined the integrator circuit and showed that an additional resistance R0 and condenser C0 are introduced in the circuit ( as shown by dotted lines in Fig ) when service selector is changed to �inert� test. The voltage across C0 is average voltage over a time of kv/α0. The average of this is the voltage developed across C1, and the deviations from average are the voltage across R1. The mean value of the deviations (or square of deviations) are estimated by the voltage across C2. By varying the material speed from 1 to 100 m per min, �Between lengths� can be varied and variance length curve estimated. Grignet and Monfort further improved upon this method27. A fairly close agreement is found between B-L curves estimated by Uster tester and cut and weigh methods.
    ● Spectrogram
    Spectrogram is a fourier analysis of variations present in the material. The amplitude of the variations are sorted as per their wavelength and plotted as a amplitude vs wavelength curve in one diagram.
    Constant Staple length
    With random fibre arrangement the amplitude vs wavelength is given by equation below
    Where
    ( λ) = amplitude
    l = fibre length
    λ = Wavelength
    The spectrogram of yarn of constant staple length shows two peaks, one at 2.7 l and another smaller one at l which is a lower harmonic.
    Spectrogram of staple fibre Yarn Fig 15 : Spectrogram of staple fibre yarn (with random fibre arrangement)
    The spectrogram of a yarn with variable fibre length due to random fibre arrangement has a �hump� whose maximum wavelength lies in the region of 2.5 to 3 times fibre length.
    Spectrogram Fig 16 : Spectrogram of cotton and woolen yarn
    On the top of it, waves introduced by drafting waves is superimposed. Wavelength of drafting wave is also 2.5 to 3 times fibre length and so in normal yarn the "hump" is more pronounced depending upon the amplitude of drafting wave. Thus spectrogram of cotton yarns has a hump at 6 to 9cm, of woolen yarn at about 20 -25 cm. OE rotor yarns have a peak at a slightly lower length than ring yarns because fibres are curled with hooks leading to a lower projected length on yarn axis. Spectrogram has a hump at 3.5 times the fibre length with cotton roving and 4 times the fibre length with cotton drawing sliver. When periodic variation is present, spectrogram will show a sharp peak at the point corresponding to wavelength of periodicity. The defect should occur at least 25 times during material passae to be shown as a peak. Serious types of periodic faults are distinguished by the amount by which the peak exceeds the base spectrogram. Anything in excess of 50% is likely to show as fabric defect. Intensity of the defect, correspondence between its wavelength and fabric width determine the occurrence of weft bars. Old Uster tester GGP has 35 filters with each filter encompassing a certain wavelength. Spectrogram is therefore a stepped curve with 35 steps Later models have a higher number of filters, 54-55 in Uster 1 and 2 Latest model Uster testers have increased filters up to 80 and the wavelength determination up to 2560 m. Further, 3 dimensional plots of spectrograms from 10 bobbins can be displayed side by side which helps in identifying the defective bobbins. The amplitude in the spectrogram chart is a function of the variance calculated from the amplitude of the wave. Diana Germanova- Krasteva35 have developed a method for calculating the amplitude and wavelength from the CV% values and spectrogram chart of Uster 3 tester. So spectrogram is a useful tool for detecting the periodicities in the material. It also gives their wavelength which can be used to trace the cause of periodicity and rectify it.
    Fast fourier transforms36 are used to determine spectrograms to detect the presence of sinisoidal waves caused by eccentricity in drafting roller. Fast Walsh Hadamond transform is used for determining rectangular patterns such as flattened top roller. Fast impulse frequency transform is used to detect impulse type of defects such as those caused by presence of trash in rotor groove or broken draft gear tooth. Sinusoidal irregularity produces only a single peak. Rectangular, triangular and impulse type of irregularity produce not only fundamental but also harmonics. Jun and Korobov37 suggested suggested a new method for determining spectrogram based on multiresolution analyser that has greater precision and higher speed than Fourier transform. Matsumoto38 et al; examined irregularity of sliver by estimating bispectra. The bispectrum of ideal sliver and two pseudo ideal slivers made of fibre clusters were calculated using computer simulation program. Tallant and Patureau39 suggested an improved method of calibrating Uster spectrograph by using a tape with square wave instead of sinusoidal wave provided by manufacturer. The advantages of this method are that it can evaluate the performance of more channels and can detect harmonics from fundamental even when wavelength of fundamental is beyond the range of spectrograph. Sharieff40 et. Al; used spectral analysis to separate yarn irregularity into periodic, semi periodic and random components. Periodic component increases non linearly with the extent of eccentricity in roller and draft and quasi periods also an show a increase .
    Imperfections
    These represent outlayers of variations in yarn which have a profound influence on appearance of yarn and fabric and performance of yarn41. Uster imperfection tester measures three types of imperfections viz; Thin places, Thick places, Neps. Nep is differentiated from a thick place by the length of the defect. The instrument evaluates nep as a thick place whose length is shorter than 4mm and longer thick places are evaluated as thick. Even so some imperfections are found to get counted as both thick place and neps. Gupta42 concluded that the instrument overestimates imperfections as a result of double counting() However Gupte43 showed after careful examination of imperfections that thick places and neps occur together in many instances. This is because the presence of foreign matter and fibre cluster (that get counted as nep) causes disturbance to smooth fibre movement in drafting and results in thick places and therefore there is no overestimation.. 4 classes of each of these imperfections are measured as indicated in Table below.
    Table 10
    Thin(%)Thick (%)Nep (%
    -30+35+140
    -40+50+200
    -50+70+280
    -60+100+400
    However -50% setting for thin places, +50% setting for thick places and +200% for neps are customarily used settings for ring yarns. For Open end yarns +280% setting for neps is recommended. Thin places are influenced by fibre properties, combing and drafting conditions. Thick places are influenced by fibre properties, blow room, carding and combing and drafting conditions. Neps are influenced by fibre properties and blow room, carding and combing conditions. Gupte and Balasubramanian44 found that CV% by quadratic integrator is better correlated with thin and thick places than U% by linear integrator. Further CV% has a better correlation with thin places than thick places. Gupte and Balasubramanian45 found that while actual diameter of thin place agrees with calculated, that of thick place is much higher than calculated. Twist at thin place is 150% and that at thick place is 70% of normal place. Tenacity of thin place is lower while that of thick place is same as normal portion
    Faults
    Table11 below shows the difference between imperfections and faults.
    Table 11
    ImperfectionsFaults
    imperfections are frequent occurring defects ( 100 � 5000 per 1000 metres)Faults are seldom occurring defects (200 -5000 per lakh metre)
    Imperfections affect the appearance of yarn and fabric
    Faults affect appearance of and sale value of fabric and lead to end breaks in winding and weaving and knitting
    Imperfections are too small to necessitate clearing in winding.Clearing of objectionable faults is desirable in winding
    Yarn fault classification system like Uster Classimat has facility to detect faults as per their size and length. ● Short Length faults
    16 categories of short length faults are measured by Classimat.
    Table12
    Short Length Faults/TD> /TD> /TD> /TD>
    Size of Fault/TD>A - 0.1 to 1cm/TD>B - 1 to 2cm/TD>C � 2 to 3cm/TD>D - above 4cm/TD>
    +100 to +150%/TD>A1/TD>B1/TD>C1/TD>D1/TD>
    +150 to +250%/TD>A2/TD>B2/TD>C2/TD>D2/TD>
    +250 to +400%/TD>A3/TD>B3/TD>C3/TD>D3/TD>
    above +400%/TD>A4/TD>B4/TD>C4/TD>D4
    ● Very short thick places are caused by the presence of seed coats, broken seeds, trash in the case of cotton and cut fibres in synthetics. Medium size short thick places are caused by embedded fluff, spun-in loose lint among others. Faults of 4cm and above are caused by slubs, undrafted ends, bad piecings etc.
    ● Long Thick Faults
    5 catergories of long thick faults are measured by classimat.
    E - Above 8cm length and +100% and above crosssection size.
    Table 13
    Size of fault8 - 32cmabove 32cm
    +45 - +100%FG
    Long thick places of E category are known as 'Spinner's double' and are caused by lashing of an end with adjoining end at roving or ring frame and over end piecing at ring frame. F and G faults are due to drafting defects at earlier stages, sliver splitting in creel at roving and long overlap of sliver at the time of sliver break at draw frame.
    ● Long Thin Faults
    4 types of thin places are measured
    Table 14
    Size of Fault8 - 32cmabove 32cm
    -30 to -45%H1I1
    -45 to -75%H2I2
    Long thin faults are due to raw material defects, drafting faults in roving and drawing, split sliver in creel of roving, mal functioning of stop motion at drawframe. Longer and thicker size faults are easily perceivable in fabric. So A4, B4, C3, C4, D3, D4 are termed as objectionable faults. Quality conscious buyers insist that these faults are below 1 for one lac metre. German and Japanese buyers consider even C2 and D2faults as objectionable. E, H2, I1, I2 faults are also objectionable as they are easily perceivable in woven and knitted fabrics, lead to fabric rejections and high end breaks during subsequent processing. Clearer settings in winding should be set to remove these faults. Keisokki classifault trichord flex has an extended range of fault measuring capability.
    Neps and slubs - 25 classes
    Thick places � 15 classes
    Thin places � 15 classes
    Foreign fibre faults � 25 classes
    Class limits for fault limits can be freely chosen by user. Histogram analysis of faults is an added feature that helps the mills to locate the problem areas. Both capacitance and optical sensors are used.
    Analysis of characteristics of faults will give useful clues as to their source of origin. This is facilitated by the provision to set the classimat to cut the yarn at the occurrence of preset fault. Mukhopadhyay and Balasubramanian46 found count variation to be higher in the fault portion. Twist level is higher in thin faults like H1, H2, I1, I2 and higher at thick faults like E and F. This is because of tendency of twist to flow to thin place. Tenacity is lower at E fault. Though thin faults are weaker, their tenacity is comparable to normal portion. Agarwal47 et al found that C3, C4 and D faults have a lower strength even after sizing and are therefore a major source of warp breaks in weaving.. Mukhopadhyay48 et. Al; found that while the diameter of thin faults lies close to that prescribed in Classimat, thick fault diameter exceeds that prescribed limits.
    Online measurement
    Online measuring equipments are increasingly used by quality conscious and export oriented mills. Sensors fitted on winding units measure faults, imperfections and evenness of yarns. Continuous variation charts of yarn mass/diameter are also displayed. Such systems have the merit that the entire production is monitored and enable rapid corrective action. Loepfe offers yarnmaster, based on optical measurement, which detects off standards with respect to CV%, imperfections, count and hairiness and stops winding machine. It has facilities for detection and removal of long thick and thin faults, neps, coloured fibres, transparent filaments and online simulation of fault appearance in fabric. Uster Quantum Classifault can operate on production winding machine as well as laboratory winding machine. Apart from the 23 thick faults and 4 thin faults by capacitance principle, it has an optical sensor for measuring 27 classes of foreign material faults caused by contamination. Further yarn count is monitored and if count beyond certain pre-set limits, the bobbin is rejected thereby ensuring no rejections due to count beyond limits. Optimum clearing curve is provided based on scatter plot of faults. Faults remaining in the package after clearing is also estimated.
    A comparison of evenness and imperfections by Uster tester and Corolab fitted on rotor machine showed good correspondence between the two measurements in the case of CV%49. The former is based on capacitance while the latter is based on optical measurement. But imperfections measured by corolab was found to be higher than that by Uster. Yarn type, weaving or knitting influences the relationship between thick places measured on the two equipments.
    Random Fibre Arrangement
    No discussion on concept of irregularity will be complete without a reference to irregularity due to random fibre arrangement. In order to produce a yarn without any irregularity, there should be a mechanism to put a new fibre as soon as one fibre ends. But there is no such mechanism in currently available spinning systems. The best that can be done is to arrange fibres randomly. In Blow room and Carding, fibres are randomly placed and so the best possible arrangement in card sliver will be one with random arrangement. Even if it were possible to produce a card sliver with fibres arranged one behind another, this arrangement will be destroyed by doubling in drawframes, as slivers are fed side by side without regard to position of fibre ends. In other words, doubling at drawframe randomises fibre arrangement further. A certain amount of minimum irregularity is likely to be present due to random fibre arrangement as shown by Martindale50. Since fibre length is very small compared to yarn length, the probability that a particular fibre is present in a certain cross-section is very small and the probability of its non occurrence in the cross-section is very large. Under such conditions, the number of fibres in cross-section will vary as per Poisson distribution. For a Poisson distribution, standard deviation, σ is equal to square root of Mean.
    If n = mean number of fibres in a cross-section
    σ = √n
    CV%= (σ/n)×100
    =100/√n
    A certain amount of irregularity is also caused by variations in weight per unit length along the length of a fibre and between fibres. If CVf is the irregularity due fibre weight variations then
    CV %= √((100/√n)2+ (CVf/√n)2).
    A good deal of variation exists between material to material in variability of fibre weight per unit length but roughly CVf may be taken as 30% for cotton, 50% for wool and 10% for man made fibres. Upon putting these values in the above equation
    CV% =106/√n for cotton
    112/√n for wool and
    102/√n for man made fibres. This means that that irregularity due to random fibre arrangement will increase as n decreases. If C= Yarn Count(Ne) and M= Micronaire value of fibre, D= Denier of fibre
    n=15000/(C×M )
    =5314.5/(C×D )
    CV due to random arrangement will therefore be very low with Drawing sliver and increases slightly at roving and becomes noticeable in yarn. In fine counts CV% due to random fibre arrangement will be a substantial part of the yarn irregularity. This also means that CV% of yarn can be brought down by using finer fibres in place of coarser fibres. Van den Abelle51 pointed out that in actual processing arrangement of fibres in a yarn is not exactly random because of certain limitations for formation of extreme thick places. An alternative model where fibres are arranged in m lanes is prposed so that yarn thickness no where exceeds m fibres. Variance for such a yarn is given by V2 =1002/n(1 - (n/m) ) A similar reasoning is also given by Picard52
    The assumption is more of academic in nature and will not affect significantly the value for random irregularity.
    Index of Irregularity
    Index of irregularity53(I) is the measure, proposed by Huberty, to find out the extent to which actual irregularity deviates from that due to random. Index of Irregularity I =CVa/CVr where CVa is actual measured irregularity and CVf is random irregularity. A higher value means that there is more scope for improving the processes.
    Linhart54 has proposed an alternative index to characterize irregularity of yarn. Bias and standard error of Huberty�s coefficient and new index have been calculated.
    References
    1 , Weft bars and Weft way defects, N.Balasubramanian and S.Sekar Indian Textile J, 1982 Dec,p101,
    2. Weft bars in man- made fibre blends due to torsional vibration of back rollers at speed frame ,S.K.Sett and N.Balasubramanian, J. Textile Association,1985,Nov, p 183
    3. Influence of weft count variation on warp way creases in fabrics, N.Balasubramanian and C.Chatterjee, I. Textile Journal, 1996 Oct p 66
    4.Relaation between fibre end density and thickness, N.Balasubramanian, J. Textile Institute, 1965, 56, T342,
    5. Some calculations relating to arrangement of fibres in slivers and rovings, G.A.R. Foster, J.Gregory and J.R., Womersley, J. Textile Institute, 1945, 36, T311
    6. Planimeter provides easy way for evaluation of evenness tests, Textile Industr, 1949, 113, Jan, p 92
    7. A rapid approximation to Stanadard deviation and Coefficient of variation, J.H. Burkhalter Textile Research J., 1958, 28, p 91
    8. A comparative study of visual assessment of yarn irregularity with Uster Evenness results, N.Balasubramanian, Indian Cotton Growing Review, 1964, March
    ,9. Microprocessor based sliver unevenness analyser, S.K. De, Textile Research J, 1981, 51, p426
    10. H.Locher, J. Textile Institute, 1953, 44, p648
    11.Testing of the irregularity of blended yarns using apparatus of dielectric-capacitance type, J.W.S. Hearle and P.H. Walker, Testing of irregularity of blended yarns using apparatus of dielectric- capacity type, J. Textile Institute, 1953, p811
    12. G.A.R.Foster, J. Textile Institute, 1957, 48, T109
    13.Effect of conditioning on U% of sliver, N. Balasubramanian and G. Janakiraman, I, Textile Journal, 1994, June, p 67
    14. The use of the Zellweger irregularity tester for finding the variance length curve of worsted yarns, P.Grosberg and R.C. Palmer, J. Textile Institute, 1954, 45, T275
    15. Characterisation of yarn irregularity and variation in yarn diameter, J. Militky, S. Ibrahim, D. Kremenakova and R. Mishra, 2010 Beltwide Cotton conferences, New Orliance, 4-10 Jan 2010
    16. Beyond capacitative systems with optical measurements for yarn evenness evaluation, A. Sparavigna, E. Broglia and S.Lugli, Mechatronics, 14, 10, 2004, p1163
    17. Yarn Evenness parameters- A new approach, V. Carvalho, J.L. Monteiro, F.O. Soares and R.M. Vasconcelos, Textile Research J, 78, p119
    18. Uniformity Analysis of Yarns, Rovings, and Slivers Using Beta Radiation P.R. Ewald and C.B. Landstreet, Textile Research J, 1957, 27, p49
    19. Use of correlograms for measuring yarn irregularity, D.R. Cox and M.W.Townend, J. Textile Institute, 1951, 41, p145,
    20. J.L.Spencersmith and H.A.C. Todd, Suppl. J. royal Statist. Soc, 1941, 7, p131, 21. G.A.R. Foster, Suppl, J. Roy, Stat., Soc., 1946, 8, p42
    22. A simple method of plotting correlograms of worsted yarns, W.C.Onions and Selwood, , J. Textile Institute, 1956, 46, T, 127 and T 603
    23. M.W. Townsend, J. Textile Institute, 1949,40, p566 24. Calculation of variance length curve for ideal sliver, H. Olerup, J. Textile Inst., 1952, 43, p290,
    25. Calculation of variance length curve from length distribution of fibres Part I & II H.Breny, J. Textile Inst., 1953, 44, p1 and p10,
    26. On the determination of the B(L) curve by cutting and weighing, P. Grosberg and R.C. Palmer, J.Textile Inst., 1954, 45, T291,
    27. Modification to the calculation of B(L) curvd by the inert test method. J.Grignet and F. Monfort, J. Text., Inst., 1958, 49, T706,
    28. C.J. Vanzwet, A method for calculation of CB(L) curve, J. Textile., Inst, 1955, 46, p794
    29. Medium and long term variations of a yarn I, P.Grosberg, J. of Textile Institute, 1955, 46, T301
    30. Medium and long term variation of yarn II, J. Textile Institute, P.D.Malatinszky and P.Grosberg 1955, 46, T310
    31. Contribution to the study of B-L curve of cotton yarns, N.Balasubramanian, Textile Research J, 1963, 33, p697
    32. B-L curve of Cotton yarns, N.Balasubramanian, Textile Research J, 1965, 35
    33.S.B. Bandyopadhyay, J. Textile Institute, 1955, 46, T63
    34 Unevenness (Length Variance) of a Worsted Yarn by Cutting-and-Weighing and the Use of the Pacific Tester, R. A. O�conell, F. J. AhRrens, and R. J. Martsch, Textiles Research J, 1975, 45, p 596
    35. Spectral Analysis of the Mass Irregularity of Slivers using Uster Tester, Diana Germanova-Krasteva, Fibres and Textiles Eastern Europe, July/Sept. 2003, 41,p42
    36. Yarn irregularity parametrisation using optical sensors . V.Carvalho, P. Cardoso, M.Belsley and R.M.Vasconsilos and F.O.Soares, FIBRES & TEXTILES in Eastern Europe,
    January/March 2009, Vol. 17, No. 1 (72) pp. 26-32 37. Calculational method for spectrogram and variance length curve of yarn based on multiresolution analysis, M.Jun and N.A.Korobov, 6th international conf on parallel and distributed technology computing applications and technologies (PDCATOS), 2005, p993
    38. Bispectra of Sliver Irregularities, Y. Matsumoto, K. Toriumi, I. Tsuchiya and K. Harakawa, Textile Research Journal,,1991,61, p334
    39. Harmonic Response of the Uster Spectrograph, J.D. Tallant and M.A. Patureau, Textile Research J, 1968, 38, p 208
    40. Spectral Analysis of Yarn irregularity and Its Relationship to Other Yarn Characteristics, I.Sharieff , S. G. Vinzanekar and T. Narasimhan, Textile Research Journal, 1983, 53, p 606
    41. Improving the yarn appearance and imperfections in the mills, N.Balasubramanian and K.Viswanathan, J. Textile Association, 1977, June, p67
    42.An interesting observation oncounting of neps and thick places by Uster imperfection indicator, A.K.Gupta, J.T.I, 1986, 77, p354
    43. A.A.Gupte, Ph.D. Thesis, Bombay University, 1993
    44. Influence of imperfections on U% and CV% of yarn, A.A.Gupte and N.Balasubramanian, BTRA Scan, 1991, Septr., 22, p4
    45. Characteristics of yarn imperfections in cotton and blended yarns, A.A. Gupte and N.Balasubramanian, Indian J. Fibre and Textile Research1988, 13, p192
    46. Optical Analysis of Classimat Faults, D.Mukhopadhyay and N.Balasubramanian, BTRA Scan, 1993, 24, No 2, p1
    47. Evaluation of Classimat Faults for Their Performance in weaving S. K. Aggarwal, P. K. Hari and T.A. Subramanian, Textile Research J, 1987, 57, p735
    48. A.Mukhopadhyay, I. C. Sharma, and K. Dasgupta, Textile Research J, 2002, 72, p 178
    49. A Comparison of Yarn Evenness and Imperfection Data, J. B. Price, T. A. Clamari and W.R. Meredith, Textile Research J, 2002, 72, p810
    50. Theory of random fibre arrangements in worsted yarns, J.G. Martindale, J. Textile Inst., 1945, 36, T33
    51. A.M Van den Abellee., J. Textile Inst, 1951, 42, P162
    52. M.C. Picard, 1951, 42, T503
    53. A. Huberty, Proc. I. W.T.O. Techn. Comm. 1, 55, 194,
    54. On Huberty�s coefficient of irregularity of yarns and its estimators, H.Linhart, Textile Research J, 1965, 35, p1055
  • Yarn Hairiness

    Hairiness of Yarns

    N.Balasubramanian
    Retd. Joint Director(BTRA) and Consultant, Tel No 91 22 25280767, Mobile 9869716298
    Protruding fibres, loops from the surface of yarn and loosely wrapped wild fibres constitute hairiness. Hairiness is a unique feature of staple fibre yarns that distinguishes it from filament yarns. Hairiness is generally regarded as undesirable because of the following factors
    1. 1. It adversely affects the appearance of yarns and fabrics. Hairiness is one of the factors that go to determine the appearance grade of the yarn. Higher hairiness downgrades the appearance. Hairiness in yarns leads to fuzzy and hazy. According to Uster 15% of fabric defects and quality problems stem from hairiness. appearance of fabric. Warp way streaks and weft bars are caused by high hairiness and variation in hairiness. Periodic variation in hairiness has been traced to be a cause for alternate thin and thick bands in fabrics
    2. It affects performance of yarn in subsequent stages. Adjoining warp threads cling together in the loom shed because of long hairs in yarn, thereby resisting separation of sheet during shedding. This leads to more warp breaks and fabric defects like stitches and floats.
    3. Excessive lint droppings in sizing, loom shed and during knitting are encountered with hairy yarns because of shedding of hairs and broken hairs.
    4. In printed goods, prints will be hazy and lack sharpness if yarn is hairy.
    5. In sewing breakages will be high with hairy yarns and removal of hairiness by singeing is invariably practised.
    6. Pilling tendency will be more with higher hairiness. Is hairiness undesirable
    In spite of these drawbacks, hairiness has some beneficial effects. It adds to the textile character of the fabric and contributes to comfort , liveliness, skin friendliness and warmth. This will be apparent from a comparison of fabrics made from filament yarn and staple fibre yarn of the same type of fibre and count.Fabric made from filament yarn will have �plastic� feel. Warmth found in woolen and flannel fabrics is to some extent due to hairiness. Thus a certain amount of hairiness is desirable but beyond that it causes problems enumerated above. Measurement of Hairiness Hairiness consists of protruding fibres, looped fibres and loosely wrapped wild fibres. Subjective Methods Yarns can be graded for hairiness by comparison of appearance. Relative levels of two yarns can be easily done by comparison of full bobbins. Grading for hairiness can also be done by wrapping the yarn on a black board and comparing them. Uster has developed yarn hairiness grade standard boards. This will assist in grading of yarns. Paired comparisons by a number of unbiased observers can determine statistically significant differences in hairiness by estimating coefficient of consistency. Microscopic Methods Before instruments were developed, hairiness was measured by viewing the yarn under a microscope. Barela1 was one of the earliest to use this method. Image of yarn is projected on a screen and number of protruding hairs and loops in a known length are counted. Length of protruding hairs is also measured with the help of micrometer eye piece scale. From this total length of hairs per unit length is determined. Onions and Yates2 used photographic standards to grade the hairiness of projected yarn image. Pillay3 projected the yarn on Projectina microscope and counted the number of protruding ends and loops in 10 mm length. Length of protruding fibres is measured by a curvimeter on a tracing of yarn image. Jedryka4 took photographs of yarn image under microscope with a magnification of 50x. The boundary levels of yarn were marked. Four zones parallel to the boundary line were drawn on either side. The lines were equidistant with space between adjoining lines being kept as half the diameter of yarn. Hairiness is determined by the number of intersections of fibre with the lines marking the zone on either side. This method gives the hairiness as per the length of the hairs. About 50 to 100 yarn samples were examined from which average hairiness is determined. Major difficulty in microscopic methods is in indetifying the boundary of yarn. Looped fibres, wild fibres, low twisted portions, variation in yrn diameter and cross-section smudge the boundary. High variation in hairiness is found both within and between bobbins and as a result large number of yarn specimens have to be examined to get a fairly reliable estimate. This makes the method laborious and time consuming.
    Photelectric method
    Several instruments are available for measurement of hairiness based on photoelectric method.
    Shirley- Atlas Hairiness Tester
    A measuring head consisting of a photocell placed close to the yarn counts the number of interruptions made by the protruding hairs to an LCD beam. The measuring head is infinitely adjustable from 0 to 10mm from the surface of yarn. This enables measurement of hairiness as per the length of hairs. Yarn is driven by nip rollers at 50 to 300 m/min by an electronic variable speed drive. Latest version is operated by a pc. Continuous chart of hairiness can also be obtained through a recorder or printer.
    Zweigle Hairiness Tester
    This also uses a measuring head with a photocell and a laser light source. The instrument measures hairiness of 9 length zones from 1 to 15mm in a single run of the yarn. The equipment is controlled by a pc which carries out statistical analysis of the results. An automatic bobbin changer up to 24 bobbins is available which makes the instrument fully automatic.
    Meiners Dell Hairiness Tester
    The instrument measures simultaneously hairiness of hair lengths from 1 to 10 mm in steps of 1mm. A single run of the yarn gives hairiness for all lengths at a selected speed. A portable model is also available which enables on line measurement of hairiness. This equipment will be useful in spotting out spindles higher hairiness thereby enabling prompt corrective action.
    Changling Hairiness Tester
    This uses a laser light source and a sensitive integrated photocell for measuring number of projecting hairs. Measurements of hair number for lengths from 1 to 9mm is possible.
    Uster tester
    Uster Evenness tester has a hairiness attachment The measuring field consists of homogenous rays of parallel light from a infra red light source. Yarn is placed in this field.. Scattered light from the protruding hairs of yarn reach an optical sensor which converts it into an electronic signal. The body of yarn itself is dark as it is not transparent. Hairiness thus measured is an estimate of total length of protruding fibres in a cm length and is termed as Hairiness index. Hairiness index of 4 means that the total protruding length of hairs in 1cm length is 4cm. While this method has the merit that it gives a single index to characterize the hairiness, it has the drawback that it does not provide information on long length and short length hairs separately. Thus two yarns may have the same hairiness index but one may have more long hairs and fewer short hairs than other. Since long hairs are more objectionable than short hairs, information on the level of hairs as per their length will be more useful. Even so there is a good correlation is observed between hairiness by Shirley hairiness tester and Uster tester5. Uster tester further gives coefficient of hairiness over measured lengths 1cm(normal) 10, 100, 300, 10000, 5000 cm or in other words variance length curve of hairiness. Presence of periodicity in hairiness can also be determined by spectrogram of hairiness.
    Premier Electronic tester
    Premier qulaicenter which is similar to Uster tester has an attachment to measure hairiness by hair count as well as Hairiness index method. ASTM standard D5647-01(1995) gives a standard method for measuring hairiness with photoelectric instruments. This will be helpful to minimize variations from laboratory to laboratory.
    Weighing Technique
    Difference in the mass of yarn before and after singeing6 is used as a measure of hairiness. Flaw in this method is that a large amount of yarn has to be singed to get an accurate estimate. Moreover singeing does not fully remove fully projecting hairs particularly the shorter length hairs.
    Factors influencing Hairiness
    Raw Material
    Fibre length, short fibre content, fineness and rigidity are the most important properties of fibre that influence hairiness7. Number of fibre ends per unit length increases as fibre length reduces and as each fibre end is a potential source of hairiness, yarns from shorter fibres are more hairy.. As a result any process from picking to ginning to opening of cotton that results in fibre breakages will increase hairiness in yarns. This is supported by the study of Mclister and Rogers8 who found yarns spun from spindle harvested cotton had less hairiness than that from stripper harvested cotton. Spindle harvesting, being gentler than stripper harvesting, results in fewer fibre breakages thereby leading to less hairy yarn. Hairiness increases with coarseness of fibre because of higher resistance to twisting . For the same reason yarns from fibres with higher flexural and torsional rigidity have higher hairiness. With wool fibres with higher curvature result in less hairines. Xungai Wang, Lungi Cheng and Bruce McGregor9 found hairiness increases with increase in cashmere in wool/cashmere blends because of this reason. Jackowski10 found blended yarns have higher hairiness than yarns made from component fibres. On the contrary Barella and Vigo11 found that hairiness was more in polyester than cotton yarn and 50/50 blend of these fibres has an intermediate value. The discrepancy between these findings may be because of difference in fibre fineness and length of the blended fibres. Fibres prone to static generation generally result in more hairy yarns because of repulsion of fibres. This is the reason why yarns from polyester and other synthetic fibres have higher hairiness than those from natural fibres.
    Yarn Parameters
    Count and twist have considerable influence on hairiness. Coarser yarns have more hairiness than finer yarns because of higher number of fibres in cross-section in the former. Yarn hairiness chart has therefore close correspondence with irregularity chart with coarser regions having more hairiness than finer portions. Hairiness reduces with increase in twist12 because shorter spinning triangle and more effective twisting in of surface fibres into yarn. As a result hairiness is more in hosiery yarns.
    Process Parameters
    Preparatory
    Pillay3 found fibre parallelisation has significant influence on hairiness. Hairiness reduces with increase in number of drawframe passages. With more draw frame passages, fibre orientation is increased and fibre hooks are reduced. As a result fibre extent along the length of strand is increased which is the reason for reduction in hairiness. For the same reason combed yarns have less hairiness than carded yarns. Further with combing short fibre content is reduced which is another reason why hairiness is reduced. A compact roving by use of front zone condenser will bring down hairiness as this will strand width at ring frame
    Ring Frame
    Strand width at the front roller nip has the maximum influence on hairiness. This is because twist will closer to front roller nip and spinning triangle will be smaller and fibres in selvedge will be integrated better into yarn. As will be discussed later compact spinning was developed based on recognition of this. A coarser roving hank and higher ring frame draft will therefore increase hairiness. This is confirmed by studies by Pillay3.
    Spindle speed
    Higher spindle speed is generally found to increase hairiness13,14. This is bcause of the larger balloon at higher speed. With a larger balloon, traveler tilt will be more and this will reduce the space available for traveler movement. Moreover yarn will dash against separator with bigger balloon.
    Doff position
    Doff position and chase position have a significant influence on hairiness. Krishnaswamy, Paradkar and Balasubramanian15 compared the hairiness of yarns at different doff positions in mills and found hairiness was higher at cop bottom position as shown in Fig1. This is because of larger balloon found at cop bottom which increases traveler tilt and causes dashing of yarn against separator. In some cases an increase in hairiness is found towards the end of doff.
    Spindle speed
    Higher spindle speed is generally found to increase hairiness[33,34]. This is because of the larger balloon at higher speed. With a larger balloon, traveller tilt will be more and this will reduce the space available for yarn passage and there will be chafing and abrasion of yarn. Twist flow at lappet will also be reduced. Moreover yarn will dash against separator with bigger balloon thereby generating hairiness. Chaudhuri[35] however found that hairiness is not affected by spindle speed in acrylic yarns.
    Doff position
    Doff position and chase positions have a significant influence on hairiness. Krishnaswamy, Paradkar and Balasubramanian[36,37] compared the hairiness of yarns at different doff positions in mills and found hairiness to be higher at cop bottom position as shown in Fig1. This is because of larger balloon found at cop bottom, which increases traveller tilt and causes dashing of yarn against separator. In some cases, an increase in hairiness is found towards the end of doff.
    Hairiness at different doff positions
    Fig 1: Effect of doff position on hairiness

    Chase position
    Hairiness is more at the shoulder and reduces progressively towards the nose of the chase as shown in Fig2. The balloon is bigger at shoulder and traveller tilt is more. Yarn also dashes against separator. Both these factors increase hairiness. The periodic variation in hairiness in the chase, thus caued, is sometimes a source of hairiness[38,39]
    Hairiness at different chase psitions
    Fig 2 : Effect of chase positon on hairiness

    A mill was experiencing high fabric rejections due to weft bar. The weft bar consisted of alternate thick and thin bands with varying amplitude and periodicity. Thick and thin band portions did not show any difference in pick spacing, count, twist or diameter of yarn. But the hairiness of yarn was found to be markedly higher in thick band portion compared to thin band portion as shown in Table I
    Table 1 Hairiness in thick and thin bands
    Number of fibres beyondThick BandThin band
    4d36.926.8
    8d13.28.5

    The yarn from thick band is found to be more hairy. As a result, hairs in this portion cover interspaces between yarns and more light gets reflected leading to a denser appearance. Length of yarn in one period of thick and thin band was found to be close the length of yarn in one chase of ring bobbin. The cloth was woven on Sulzer loom where 2 splits were made side by side. Cloth width in each split was close to one half of the length of yarn wound during chase movement of ring rail. As a result, yarn from shoulder regions goes to one split and that from nose goes to the next split. But as the yarn made during one half of chase movement is slightly longer than cloth width in one split, a gradual shift in poisoning of yarns from shoulder to the nose region takes place. Yarns from shoulder and nose regions group together alternately in the fabric leading to formation of thick and thin bands.
    Traveller
    Weight, profile and type of cross-section of traveller have critical influence on hairiness.
    Weight
    Heavier traveller up to a limit reduces hairiness[40] because of improved flow of twist to front roller nip. As a result pilling of knitted material reduces. Higher tension associated with heavier traveller will also help to firmly twist the surface fibres into yarn.
    Profile
    Elliptical traveller has a low bow size and as a result limited space is available for passage of yarn. Chafing of yarn will therefore be more resulting in increased hairiness. C shape traveller has a high bow size, which provides ample space for passage of yarn. Hairiness will be least with this traveller. But as centre of gravity is higher with C traveller it results in unstable flight and traveller fly especially at high speeds. Further traveller profile does not match with profile of anti wedge rings, which leads to unsteady traveller flight and rapid wear. As a compromise, Clip and EM1 and EM2 travellers were developed. While having an elliptical profile these travellers have a higher bow size than elliptical. Hairiness will therefore be lower with these travellers compared to elliptical without compromising on speed. Bow size becomes more critical when rings are worn out.
    In one mill hairiness was high on 44s warp yarn, due to worn out condition of rings. The rings were No1 flange antiwedge and traveller used was HRW clip. As change of ring will take time, and since spindle speeds were not high, C type traveller was used in place of clip. A marked reduction of hairiness was found[41] with �C� as will be seen from Fig 3. As C type traveller wears out fast, traveller replacement cycle has to be accelerated.

    C traveller

    Fig 3 : Comparison of hairiness with clip and 'C' type travellers, 1 - 'Hairy' bobbin with clip traveller 2 - Good bobbin with 'C' traveller

    Cross-section
    Round wire or half round wire cross-section will give less hairiness than flat wire. This is because of reduced frictional resistance to yarn movement by the former.
    Applicant of lubricant to traveller
    Application of specially developed lubricant to the traveler[36,37,42] has been found helpful in reducing hairiness by 20 � 30%. The reduction is more prominent immediately after application of lubricant and gradually reduces with passage of time. Application of lubricant once in 6-9 days is therefore necessary to good full benefits. It is important to ensure while choosing the lubricant that it does not stain the yarn. BTRA has developed lubricants meeting this requirement.
    Coated Travellers Travellers with coatings, such as silver and ceramic coating and chromium plating, are available for reducing traveller wear and for extending traveller-changing frequency. Because of their smooth finish, friction between yarn and traveller is reduced, which brings down hairiness. Usta and Canoglu40 found that with heavier travellers, silver coating brings down hairiness.

    Traveller Changing Frequency
    Hairiness is found to increase over the traveller replacement cycle because of traveller wear. Rate of increase is initially slow but after a point of time becomes rapid as shown in Fig For hosiery, sewing threads and p/c blends, where low hairiness is desired, traveler replacement frequency has to be kept low.Hairiness increases with traveler wear because of unstable traveler flight and flutter which accompanies it.
    Traveller changing cycle
    Fig 4 : Variation of hairiness over traveller changing cycle

    New Traveller Design
    Newer Ring and traveler designs like Spicon of BRT and Orbit by Rieter have been developed to reduce traveler wear.With such designs the traveler in its running position lies in a plane close to resultant of all forces acting on it and as a result traveler tilt is minimum. As a result hairiness is lower with such ring/traveler combinations. Moreover round wire cross-section could be used without compromise on speeds.

    Ring
    Flange No, type and wear influence hairiness considerably
    Flange No
    Higher flange number gives more space for passage of yarn and will reduce hairiness. But taveller wear will be more and higher speeds cannot be achieved in finer counts. Normally No 2 flange should be used up to 20s count and No1 flange should be used for counts 30s and above. For bringing down hairiness No 2 flange may be used in counts of border range.

    Wear and tear
    Worn out ring is a major cause of hairiness and variation in hairiness in mills[43]. When wear is pronounced, the bobbins are highly hairy and exhibit whisker like defects. Rings where not changed for 9 years in a mill and were extremely worn out. Upon replacing the rings a substantial reduction of hairiness is seen[43]
    When rings are more than 3 years old, hairiness starts increasing. Replacement of rings will bring significant reduction in hairiness. Some typical results are given in Table2 to show the effect of ring life on hairiness.

    Table3:Effect of ring life on hairiness
    MillCountAge of ring yearsHairiness Hairs/m
    A31s P/C426.4
    New21.7
    40s P/V47.7
    New4.6
    B60s HT T/C511.0
    12.9
    Yarns spun on pilot plant ring frame give lower hairiness[36] and higher strength[44] than those on mill�s ring frame though same roving bobbins were used as feed material. This is because rings in pilot plant ring frame are worn out to a lesser extent than mills ring frame because of less running.
    Lappet
    Abrasion against lappet is a source of hairiness. This gets aggravated when lappet is grooved or is worn out. Some manufacturers have come out with glass finish lappet, which minimizes friction and thereby reduces hairiness. Height of lappet above the ring bobbin has to be optimised to reduce not only end breaks but also hairiness. If lappet to bobbin tip distance is high, balloon will be longer. This will reduce twist flow and also increase area of contact between yarn and lappet. As a result hairiness will be higher. Controlled studies[36] have shown a lower hairiness with reduction in lappet height. Care should however be taken to ensure that yarn does not touch bobbin tip while lowering lappet height.
    Disturbed Spindle Centering
    Disturbed spindle centering is one of the major causes for the spindle-to-spindle variation in hairiness. On spindles where cantering is disturbed, hairiness is found to be higher and upon accurate centering hairiness comes down significantly[37]. When spindle is not centered traveller movement is not smooth because of peak tensions in yarn. Traveller tilts and flutter also increases leading to higher hairiness.
    Separator
    Plastic separator will increase hairiness because of static generation. Disturbed, slanting and bent separators generate hairiness because of excessive dashing of balloon on separator.
    Spindle and bobbin vibrations
    Vibration of spindle arises because of worn out spindle tip and bearing. Bobbin vibrations arise not only from spindle vibration but also from eccentricity in bobbin and improper fit. When bobbin vibrates hairiness increases because of uneven traveller flight.
    Plastic Bobbin
    Plastic bobbins generally give more hairiness than wooden bobbins especially with polyester blend yarns. This is because of static generation.
    Relative Humidity
    Recommended humidity in ring frame department is 55 � 60%. At higher humidity levels, fibres tend to stick to drafting rollers resulting in protruding hairs and loops. At low humidity levels static generation causes repulsion of fibres, particularly with p/v and p/c blends, leading to more hairiness.
    Winding
    Hairiness increases in winding[45, 46,47,48,49]. This is because of abrasion of yarn against tension disc, guide eyes, balloon breakers and winding drum. Extent of increase varies from 50 to 150%. Extent of increase in hairiness increases with winding speed46. Lang et al[48, 49] showed, through a theoretical analysis, that hairiness increase takes place mainly at tension discs because of frictional resistance offered by disc surface to projecting hairs. As the yarn moves forward, these fibres get pulled out of yarn. Loosely bound surface fibres may also become projecting hairs because of rubbing action. The authors used a parameter K and a critical length to estimate the effect of winding on hairiness increase. Friction coefficient between yarn and friction disc has the maximum influence on K. Increase in initial tension of yarn will reduce generation of hairiness.
    A very interesting finding of practical significance is that initial level of hairiness in ring yarn has considerable influence on the extent of increase in hairiness in winding[50]. Ring bobbins judged to be more hairy and less hairy were selected from a ring frame in a mill spinning 60s. The yarns were separately wound on Autoconer. Hairiness of ring yarns and wound yarns are given in Table3.

    Table3 : Increase in hairiness upon winding with yarns of different levels of hairiness
    Type of YarnLess hairy yarnMore hairy Yarn
    3mm5mm6mm7mm3mm5mm6mm7mm
    Ring Yarn3.0.18.06.04201.7.33.10
    Wound Yarn13.6.9.22.1218.91.71.48.19

    Short length hairs increase by 4-4.5 times with winding with �less hairy� yarns. But with �more hairy� yarns, short length hairs do not increase with winding. Long length hairs however show an increase with winding with both �less hairy� and �more hairy� bobbins.
    Longer hairs being on the surface of yarn are more likely to come in contact with tension disc and so get pulled out because of frictional resistance. This is the reason why they increase with both type of yarns. With � more hairy yarns� the surface of yarn body and short length hairs are well buried under long length hairs and therefore do not come into contact with the tension disc. There is therefore no generation of short length hairs at tension disc and short length hairs therefore do not show an increase with winding with such yarns. Moreover, with �hairy� yarns, most of the fibres whose ends are loosely anchored on the surface have already developed into hairs in ring frame itself because of abrasion at traveller/ring junction. Therefore there are fewer such fibres that can develop into in to hairs during winding. This is not the case with �less hairy� yarns as there are many loosely anchored fibres in the yarns. The work of Dash et al[51] supports this winding. The authors found that hairiness increase in winding is more in compact yarns (which have low hairiness) than normal yarns.
    Modifications to reduce hairiness
    Compact Spinning
    As pointed out earlier, major cause for hairiness is the spinning triangle at front roller delivery point, which restricts flow of twist up to nip. In compact spinning, a condensing zone is introduced after normal drafting zone, as shown in Fig 5. As a result, the strand width becomes closer to yarn diameter and the size of spinning triangle is considerably reduced. Selvedge fibres get fully integrated into yarn and projecting fibres are markedly reduced.
    Fig 5 : Comparison of Compact and Normal spinning

    Several manufacturers have developed compact spinning based on different versions of condensing system. Zinser Aircom Tex Condensing zone has a pair of rollers in front of regular drafting. The top roller drives an apron, which has a set of perforations in the middle. The drafted strand is guided underneath the apron. The perforations consist of elliptical and circular pores. The apron runs over a profile tube, which has a slot in the region S (Fig 6) through which suction is applied. The strand follows the perforation track of the apron thereby getting condensed.
    Zinzer Air com
    Fig 6 :Zinser Air Com Tex Compact spinning

    Suessen EliTe
    The condensing zone consists of a lattice apron, located at the bottom and driven by the delivery top roller as shown in Fig 7. The apron runs over a profile tube, which has a slot S, at the middle. Suction is applied through the slot. Front top roller of drafting system drives the delivery top roller through a gear. The diameter of delivery roller is slightly higher than front roller of drafting system, due to which the fibres in the strand are delivered in a straightened condition. Air drawn through inclined slot causes rotation of fibres around their axis, which contributes to better integration of short fibres into strand.
    Suessen Compact
    Fig 7 : Seussenelite Compact Spinning

    Rieter comforspin system
    A perforated drum replaces the front bottom roller of drafting (fig 8). A stationery insert, I, with a specially designed suction slot, in the middle over a length S, is located inside the drum. Apart from the normal top roller a second nip roller, with weighting, is also placed on the drum. Condensation of strand takes place in the zone between the top roller and nip roller As a result of suction inside the drum, the fibres follow the suction slot and get condensed. An air guide element ensures that suction operates in the slot area. The system is suitable only cottons beyond 1.07 inch length and is therefore applicable for finer counts.
    Reiter comforspin
    Fig 8 : Rieter ComforSpin Compact Spinning

    Toyota RX240NewEST
    Condensing unit is similar to Suessen and consists of a pair of delivery rollers and a perforated apron running over a suction tube with a suction slot. Condensing takes place as the perforations span a narrow width. Delivery rollers, driven by gear, drive the apron. Each condensing unit covers 4 spindles and can be conveniently dismantled. Retrofitting to Toyota normal ring frameRX240New is possible.
    Rotorcraft
    Instead of suction, Rotorcraft and Lakshmi machine works make use of a magnetic compacting system to condense the strand. Front bottom roller supports two top rollers in between which a magnetic compactor is placed. The compactor is pressed against the bottom roller by permanent magnets. The shape of the compactor enables condensation of strand. The main merit of this system is that it can be retrofitted into an existing ring frame, which should bring down the cost. Further power required to produce air suction and costs associated with it are reduced.
    Cognotex
    Compact spinning machine is similar to Rieter Comforspin and is designed for long staple fibres like wool. Angled balloon rollers are used as front rollers in the compacting zone to accommodate longer fibres.
    Officine Gaudino
    This is also for long fibres. Instead of suction, mechanical compacting is done. A smooth bottom front roller and angled top roller are located in front of drafting zone. The axle of top roller is in a slanting position in relation to axle of bottom roller. These rollers run at a slightly slower speed than the front roller of drafting. The negative draft, thus created, together with offset top roller create a false twisting action, which condenses the strand. The system can be retrofitted to an existing ring frame. A noticeable feature of the system is the much lower cost (only 20% higher than normal ring frame) compared to other compacting systems (where costs are 200-250%higher than normal).
    Considerable reduction in hairiness by compact spinning has been reported by many authors[52,53,54,55,56,57,58]. The extent of reduction varies with the type of raw material, current levels of hairiness, type of compact spinning, count and twist factor. On an average reduction in hairiness is about 10 � 30 % in Uster Hairiness index and about 50 � 80 % in S3 values. Reduction in hairiness is more with short staple cottons and those with high short fibre content than with long staple cottons and those with low short fibre content[59,60]. As mentioned earlier there is a significant negative correlation between hairiness and fibre length and fibre length uniformity in normal yarns. But the correlation reduces with compact spinning because of higher order of reduction in hairiness with short staple and non - uniform cottons[60]. With long staple cottons with high uniformity ratio, there is not much reduction in hairiness with compact spinning.
    Fibre length distribution of projecting hairs follows an exponential distribution[61,62,24]. While Barella[61,62] found that data falls in 2 or 3 different segments, Wang et al[24] found that in the case of worsted yarns, the data fits in one single curve. Hairiness is plotted against hair length based on the data of Celik et al[56] and Nikolic et al[54] in Figs 9 and 10.
    Hairiness length variation
    Fig 9 Variation of hairiness with hair length in woolen yarn
    Hairiness length variation cotton
    Fig 10 Variation of hairiness with hair length in cotton yarn
    Hairiness reduction with compact spinning becomes more marked above 2 mm length. Slope of hairiness versus hair length on semi logarithmic scale is not linear but appears to be curvilinear especially in the case of cotton yarns. It could be considered as made up of several linear segments with varying slopes as proposed by Barella and Manich[62]. In the case of woolen yarns, the departure from linearity is only slight.
    Extent of reduction in hairiness obtained with compact spinning with different hair length is given in Fig 11. Hairiness reduction with compact
    Fig 11 : Reduction of hairiness with compact spinning for various hair lengths
    Reduction in hairiness increases with hair length in all the cases and the increase is more prominent beyond 2mm length. This means that long length hairs are more effectively reduced in compact spinning. Long length hairs are generally more objectionable from the point of view of appearance and breakages and performance in subsequent stages. Compact spinning is therefore preferable with high quality apparel material and modern weaving units with shuttleless looms.
    Strength of yarn improves by 5 to15% and elongation by 5-8% in compact spinning. Strength improvement is more prominent at low twists and in 100% cotton yarns[63,64] In normal yarns projecting fibres do not fully contribute to yarn strength. When these fibres are fully integrated into yarn as in compact spinning their contribution to yarn strength and elongation improves. This is the reason for the increase in strength and elongation of yarn in compact spinning. Strength improvement is also more with short staple cottons and those with high short fibre content because of higher order of reduction in hairiness. Basal and Oxenham[63] examined the structure of compact and normal yarn using a tracer fibre technique and image analysis application version 3.0.Compact yarns have a high rate and amplitude of migration, which is likely to improve inter fibre friction. This may be another reason for the higher strength of these yarns. Yarns from compact spinning, with a lower twist factor, have a lower hairiness and comparable strength with normal yarn with normal twist factor[56,64]. This means that higher productivity can be achieved in compact spinning which should enable rapid payback of investment. Carded compact yarns have a lower hairiness and broadly comparable strength with combed normal yarn[64]. An interesting feature is that carded compact yarns have a higher strength than combed normal yarns with cottons having low short fibre content. But with cottons having high short fibre content, combed normal yarn has higher strength than carded compact yarns. Overall, there appears to be scope for reducing comber noil with compact spinning. This provides another avenue for getting rapid return on investment. However, it must be noted that normal combed yarns have a lower Uster U%, imperfections and better appearance than carded compact yarns This is because improvements from combing arise not only from removal of short fibres but also from removal of neps and parellelisation. Except in some isolated cases, there is no reduction in Uster evenness and imperfections in compact spinning. This is because drafting conditions determine irregularity and condensing the strand has little effect on it.
    Comparison of Compact Spinning systems
    Of the three systems, condensing zone is used after normal drafting zone in Zinser and Suessen. In Rieter, condensing is done on the front bottom roller of drafting system itself. A bigger diameter front bottom roller (drum) is used for this purpose and this increases the setting length in front zone. This restricts the use of this system for long staple cottons. Between Zinser and Suessen, condensation of strand takes place right up to delivery point in the Suessen system. In zinser, strand traverses a small distance after release from condensing zone before it reaches delivery nip. This can result in partial loss of condensation. But the suction slot is much wider and bigger in Zinser than in Suessen and Rieter systems. This improves the condensation and minimizes choke up of perforations. Further, suction level is also more in Zinser. Hairiness reduction is found to be better with Suessen system compared to Zinser in the work of Nikolic et al[53] (shown in Fig 10). This could be because condensation takes place right up to delivery nip in the Suessen system. However, contrary results were found by Goktepe et al64 who have compared the three compacting systems*. They found the system with apron at the top (Zinser), had the lowest hairiness value. Compact spinning system with apron at the bottom (Suessen) has the highest hairiness value and the system where front bottom roller is replaced by drum (Rieter) has an intermediate hairiness value in coarse to medium counts (21s and 31s). This may be because of bigger suction area and suction pressure in Zinser system. High variations in hairiness and higher irregularity were also observed in the system with apron at the bottom due to choke up of fibre and dust under perforations. Rieter system however gives the lowest hairiness value in medium to fine yarn (41s) Apart from hairiness reduction and strength improvement, compact spinning offers some other benefits which may help to pay back the higher cost of investment.
    Compact spinning has however the following limitations
    Solo Spinning
    Woolmark, CSIRO and WRONZ have developed Solospun system for reducing hairiness in wool and wool blend yarns. This is a clip on attachment to front top roller of drafting, which holds solospun rollers, shown in Fig 12. Solospun roller is placed just below the front top roller and rests on bottom roller and loaded by means of a spring. The surface of solospun roller is made up of 4 segments separated by land, which is flush with the roller surface. There are a series of slots that are offset in each segment. Twist flow to front roller is intermittently blocked by this arrangement and as a result drafted strand is split in to a number of sub strands. Twist passing through the solospun roller twists the sub strands. Upon emerging from solospun roller, the sub strands are twisted into yarn by final twist. As a result the protruding fibres are entrapped into the yarn and final yarn has fewer protruding fibres. Twisting principle of solospun yarn has been studied by Cheng et al[65]. Solospun worsted yarns have significantly lower number of S3 hairs and hairiness index[66]. Hair length distribution of solospun yarn follows exponential distribution.
    Solo spinning
    Fig 12: Solospun spinning system

    Wrapped yarn
    Filament wrapping is another means to reduce hairiness by binding the protruding fibres on the yarn body. The filament is fed, through a guide fitted above front top roller, to the strand emerging from front roller. Twist introduced during spinning will wrap the filament round the yarn body. Stop motion has to be fitted to each spindle to stop the production in the event of filament/roving break or exhaust.
    False twisting in winding
    Though compact spinning reduces hairiness, the benefits obtained are partially lost in winding as hairiness increases in winding as shown in Fig 13. The extent of increase is also more with compact yarns.
    False twist in wining
    Fig 13: Increase in hairiness with winding

    By passing the yarn through an air jet nozzle unit or any other false twisting arrangement prior to winding, hairiness can be reduced and several manufacturers have come out with such systems. Fig 14 shows an outline of an air jet nozzle unit in winding.

    Fig 14: Air jet nozzle in winding

    The swirling air in the air jet unit wraps the protruding fibres around the yarn body. The merits of such attachments compared to compact spinning are
    Muratec has developed an air jet device known as Perla A, which is fitted on their Process coner 21C. Savio has developed a �hairless device� consisting of air jet nozzle. The geometry and shape of the air jet has to be designed carefully to achieve entangling and wrapping effect of protruding fibres. Muratec has also developed another unit Perla D where instead of air nozzle, a disc driven by servomotor, is used to give false twist to yarn. Winding speed up to 1800 and 1200 m/m can be achieved with Perla A and Perla D respectively. The benefits obtained in compact spinning can be maintained by using such a false twisting unit in winding, as shown in Fig 13. Hairiness increases in normal winding but with an air jet suction unit hairiness in wound yarn is maintained close to that in ring bobbin. Several authors have reported significant reduction in hairiness by air jet nozzle in winding[67,68,69]. Air pressure and type and size of air nozzle have influence over the extent of reduction in hairiness. Higher air pressure and lower orifice angle give greater reduction in hairiness[68,69,70]. Strength is unaffected and a marginal deterioration in evenness of yarn is found with air jet unit. Computer fluidic dynamics was used to simulated airflow pattern inside the jets[71]. Air velocity in the core of the jet leads to wrapping of protruding fibres over the body of yarn. Two-phase model used to simulate wrapping of protruding hairs on yarn was attempted by Zeng et al[70]. This showed that yarn hairiness reduction occurs because of wrapping of hairs on yarn. Optimum nozzle pressure of 1.5 - 2.5 bar and a jet orifice angle of 40 � 500 were found by this model. Too high a nozzle pressure and very low nozzle angle increase air recirculation zone and are therefore not advisable. It is not clear if the wrapped hairs will stay wrapped or will get released upon abrasion at next stage.
    Air jet unit at Ring Frame
    Studies have also been made to incorporate an air jet unit between lappet and front roller of ring frame[72,73,74,75,76]. Yarn from front roller nip passes through the nozzle and compressed air is forced through an orifice into the nozzle. The air vortex produced by the air current winds the protruding fibres on the yarn body due to false twisting. After the yarn emerges from the air jet unit, real twist binds the wrapped fibres firmly into yarn. Starter yarn is fed to the lower end of the air jet unit and airflow sucks the yarn through it to facilitate piecing. Air pressure, orifice size and angle have considerable influence on hairiness reduction[70]. Higher air pressure and smaller orifice size reduce hairiness to a greater extent but such nozzles are not preferable from working point of view. Too high nozzle pressure and low orifice angle increase air recirculation zone and so affect wrapping action. An air pressure of o.5kgf/cm2 is adequate to reduce hairiness[75]. Nozzle distance of 10cm from front roller nip and nozzle angle of 450 gives best results in terms of yarn quality[75,76]. Nozzle which produces air current in the same direction as that of yarn twist gives greater reduction in hairiness than that where air vortex is in opposite direction to yarn twist[73]. On the other hand, airflow movement against the direction of yarn movement gives more reduction in hairiness[75]. Yarn evenness deteriorates significantly with the use of air jet in ring spinning. Fabrics made from yarns with air jet attachment as weft show a lower pilling tendency, which arises from lower hairiness. Use of air jet in ring spinning is still at experimental stage and most of the studies reported are on laboratory model spin tester. The effect of air jet unit on end breakage rates and piecing time, have to be still studied. No commercial versions of ring frame with air jet unit have so far been developed.
    Relative merits of Compact Spinning and Air Jet in Winding
    It would useful to consider the relative merits of compact spinning and Air jet unit in winding. Compact spinning involves much higher investment and retrofitting is also not possible in most of the systems. It should therefore be considered in products where hairiness plays a critical role like hosiery, high value apparels, sewing threads.
  • At the same time, compact spinning has the merit that it improves strength and elongation of yarn apart from hairiness. Twist in the yarn can be reduced enabling higher ring frame productivity. Higher spindle speed is also possible, as end breakages will be reduced.
  • Power cost will be higher with compact spinning
  • The benefits from compact spinning in terms of reduced hairiness is partially lost in winding as hairiness increases in winding
  • Air jet unit in winding involves lower cost, as retrofitting of the unit to an existing winding machine is possible.
  • Strength and elongation do not improve with air jet unit. Slight deterioration in evenness and imperfections is also seen with some yarns.
  • Thus to get full benefits from compact spinning, air jet unit should be incorporated in winding Hairiness in different Spinning systems
    Hairiness is higher in ring spinning than rotor, air jet and air vortex spinning systems Reasons for high hairiness in ring spinning have been discussed earlier.
    Rotor Spinning
    Fibres are well controlled in rotor and get attached to sweeping tail of yarn. Number of protruding ends is therefore lower but protruding loops are higher[77]. Fibres are more densely packed near the centre and mean fibre position is closer to the center[78] in rotor yarn. Wrapper fibres bind any fibre protruding outside. As a result, hairiness is lower in rotor yarn than ring yarn[19]. Rotor yarns have about 20% lower hairiness than ring yarns[79]. Higher twist and higher rotor speed reduce hairiness[15]. Plain naval with smooth surface reduces hairiness. Spiral shaped ceramic naval reduces hairiness[80]. Naval with groove which is used for introducing false twist, increases hairiness because of rough surface.
    Air Jet Spinning
    In Murata air vortex yarns, core fibres have no twist and are wrapped by surface fibres. More than half of yarn cross-section is made of wrapper fibres. Uniformity of wrapping of wrapper fibres and higher number of wrapper fibres results in low hairiness in air vortex yarns. Soe et al81 found lowest hairiness, for hair length 3mm and above, with air vortex yarns followed by rotor yarns and ring yarns. As a result weight loss in enzymatic treatment is highest in ring spun yarns followed by open end and air jet yarns[82]. Higher delivery speed in air vortex spinning[83,84] increases hairiness. Higher nozzle pressure reduces hairiness[83,84] but adversely affects evenness of yarn. Higher nozzle angle and smaller spindle diameter reduce hairiness in air vortex spinning[83].
    Air jet spinning is the first version of jet spinning and air vortex is the second improved version by Murata. In air jet spinning, edge fibres on the selvedge of drafted strand form wrapper fibres. In air vortex spinning, fibres in the selvedge region of emerging strand (which is about half of the total number of fibres) are thoroughly separated leading to higher number of wrapper fibres. Hairiness is therefore lower in air vortex spinning compared to air jet spinning[85]. While hairiness increases with cotton content in air jet, no such trend is found in air vortex spinning. So air vortex is more suitable for cotton rich blends.
    References
    1. Z.Tang, X.Wang, L.Wang and W.B. Fraser, The effect of yarn hairiness on air drag in ring spinning, Textile Research J., 2006, 76, p559
    2. L.Chang, Z.X.Tang, and X. Wang, Effect of yarn hairiness on energy consumption in rotating a ring yarn package, Textile Research J., 2003, 73, 949
    3. S.Adanur and S.Bakhtiyarav, Characterisation of air-yarn interface in air jet weaving, NTC Project F99-AE10(formerly Z99-A10), Auburn University
    4. N.G.Celab, Analysis of yarn flight characteristics in air jet weaving machine, Honors Thesis, West Virginia University, 2006
    5. N.Ozdil, E.Ozdogan and T.Oktem, Effect of enzymatic treatment on various spun yarn fabrics, Fibres and Textiles in Eastern Europe, 2003,Oct/Dec, p43
    6. J.Pere, N.Puolakka, P.Nousiainen and J.Burchart, Action of purified Trichoderma reesel cellulases on cotton fibres and yarn, J. Biotechnol, 2001, Aug, 89, p247
    7. A.Kumar, C.Purtell and Lapolahttp, Enzymatic treatment of man-made cellulosic fabrics, Textile Chem Color, 1994,29, 10, p25
    8. A. Barella , J. Textile Institute, 1956, 47, p210
    9. A. Barella, Yarn Hairiness, Textile Progress, 1981, 13(1)
    10. A. Barella, The hairiness of yarns, Textile Progress, 1992, 24(3)
    11. W.J.Onions and Yates, J. Textile Institute, 1954, 45, T783
    12. K.P.R.Pillay, A study of yarn hairiness in cotton yarns- PartI Effect of fibre and yarn factors, Textile Research J.,1964, 34, p663
    13. T.Jedriyka, Textile Research J, 1963, 33, p663 14. V.K.Kothari, S.M.Ishtiaque, and V.G.Ogale, Hairiness properties of polyester-cotton blended fabrics Indian J. of Fibre and Textile Research, 2004, 29, p30
    15. G.K.Tyagi, Hairiness of viscose OE rotor spun yarns in relation to test speed and process parameters, Indian J. of Fibre and Textile Research, 2004, 29, p35
    16. X.Wang and L.Chang, An experimental study on effect of test speed on yarn hairiness, Textile Research J., 1999, 69, p25
    17. H.R..Boswell and P.P.Townend, J. Textile Institute, 1957, 48, T135
    18. S.Atlas and H.Kadoglu, Determining fibre properties and linear density effect on cotton yarn hairiness in ring spinning, Fibres and Textiles in Eastern Europe, 2006, 14, July/Septr., p48
    19. A.Barella and A.M.Manich, J., Textile Institute, 1988, 79, p189
    20. D.D.Mclister and C.D.Rogers, Effect of harvesting procedures on fibre and yarn quality of Ultra-Narrow- raw cotton, J. of Cotton Science, 2005, 9, p15
    21. S.Gorden, Effect of short fibre and nep level on MVS yarn quality, Australian Cotton Grower, 2002, 23, No1, p28
    22. I.Ahmad, M.Nawaz and M.Tayyab, Interaction study of staple length and fineness of cotton with ultimate regularity and hairiness, J., of applied sciences, 2004, 4, (1), p48
    23. R.Y.Zhu, M.D.Ethridge, Predicting hairiness of ring and rotor spun yarns and analyzing impact of fibre properties, Textile Research J., 1997, 67, p694
    24. X.Wang, L.Cheng and B.McGregor, The hairiness of worsted wool and cashmere yarns and the impact of fibre curvature on hairiness, Textile Research J, 2006, 76, p281
    25. T.Jackowski, Textol Text Industr., USSR, 1961, 5, p28
    26.A.Barella and J.P.Vigo, Annal., Sci., Text., Belges, 1975, 23, p133
    27.D.Das, S.M.Ishtiaque, T.Mac, D.Viet and T.Gries, A quick reliable and economic method for evaluating the properties of rotor spun yarns, Autex Research J., 2004, 4,No3, p123
    28.A.Barella, Influence of twist, J., Textile Institute, 1957, 48, p268
    29. W.E.Morton, The arrangement of fibres in single yarns, Textile Research J., 1956, 26, p325
    30. X.Wang, W.Huang, and X.Huang, A study on the formation of yarn hairiness, J., Textile Institute, 1999, 90, Part1, No 4, p555
    31. X.Wang and L.Chang, Reducing yarn hairiness with a modified path in worsted spinning, Textile Research J., 2003, 73, p327
    32.K.P.R.Pillay, A study of yarn hairiness in cotton yarns- Part II Effect of processing factors, Textile Research J., 1964, 34, p783
    33.M.S.Parthasarathy, Factors affecting hairiness of Terene cotton yarns, Proceedings of 8th Jt. Tech., Conference, 1966, ATIRA, Ahmedabad, p28
    34. B.C.Goswamy, Hairiness of yarn, An improvement over existing microscopic technique, Textile Research J., 1969, 39, p324
    35.A.Chaudhari,Effect of spindle speed on properties of ring spun yarn, IE(I)Journal, 2003 Aug, 84, p10
    36. R.Krishnaswamy, T.L.Paradkar and N.Balasubramanian, Some factors affecting hairiness of polyester blend yarns, BTRA Technical Report No 04.2.8, 1989 June
    37.R.Krishnaswamy, T.L.Paradkar and N.Balasubramanian, Some maintenance measures to control hairiness of polyester blend yarn, J., Textile Association 1990 March, p297
    38.W.Nutter and B.Whitehead, Ring spinning as a source of weft bars in cloth, Textile Recorder, 1961, Oct.,79, p 102
    39.N.Balasubramanian and S.Sekhar, Weft bars and Weft way defects, Indian Textile J., 1982,Dec, p101
    40. F.Usta and S.Canoglu, Influence of traveler weight and coating on hairiness of acrylic yarns, Fibres & Textiles of Eastern Europe, 2002 Oct/Dec, p20
    41. B.N.Bhanot, M.Text Theisis, Bombay University,1974
    42. G.V.Kumar and J.Zacharia,Study on ring yarn hairiness with special reference to the effect of BTRA ring cleaner-cum lubricant, BTRA Scan, 1997, 28, (3), p6
    43. N.Balasubramanian, G.K.Trivedi and B.N.Bhanot, Replace wornout rings to get better yarn quality, BTRA Scan,1975, 6, p6
    44. N.Balasubramanian, A critical study of yarn quality and spinning performance in the mills PartI- Evaluation of mill�s processes, BTRA Research Project Report 18, 1973 Dec
    45.A.Barella and J.P.Vigo, Effect of repeated winding on hairiness of open end and conventional cotton and viscose rayon yarns, J., Textile Institute, 1974, 65, p607
    46.S.Peykamian and J.P.Rust, Yarn hairiness and the process of winding, Textile Research J., 1992, 62, p685
    47.A.Barella, The hairiness of yarns, Textile Research J., 1993, 63, p431
    48.J.Lang, S.Zhu, and N.Pan, Change of yarn hairiness during winding process- Analysis of protruding ends, Textile Research J., 2006, 76, p71
    49. J.Lang, S.Zhu, and N.Pan, Change of yarn hairiness during winding process � Analysis of trailing end, Textile Research J., 2004, 74, p905
    50.R.Krishnawamy, T.L.Paradkar and N.Balasubramanian, Influence of winding on hairiness:Some interesting findings, BTRA Scan 1990 June, p8
    50. J.R.Dash, S.M.Ishtiaque and R.Alaguruswamy, Indian J., Fibre and Textile Research, 2002, 27, p362 51. P.Artz, ITB � Yarn and Fabric forming, 1997, No2, p41
    52. K.P.S.Chang and C.Yu, A study of compact yarns, Textile Research J., 2003, 73, p345
    53. M.Nikolic, Z.Stjepanavic, L.Lesjak and A.Skritof, Compact spinning for improved quality of ring-spun yarns, Fibres & Textiles in Eastern Europe, 2003, 11, Oct/Dec, p43
    54.M.M.Ahmad, Compact spinning and advantages of EliTe yarns, Pakistan Textile J., 2004, Feb
    55. P.Celik and H.Kadoglu, A research on the compact apinning for long staple yarns, Fibres & Textiles in Eastern Europe, 2000, 12, No4,Oct/Dec, p27
    56. H.Stalder, Ring spinning advance, Meliand Textilberishte international, 2000, 6(1), p22
    57. M.Krifa, E.Hequet and D.Ethridge, Compact spinning- New potential for short staple cottons,, Textile Topics, 2002-2, p2
    58.M.Krifa, E.Hequet and M.D.Ethridge, Food and Fibre Commission funded project Report, Tex Tech University, 2002, Spring
    59.M.Krifa and E.Hequet, Compact Spinning- Effect on cotton yarn quality Interaction with fibre characteristics, Textile Research J., 2006, 76, p398
    60.A.Barella and A.M.Manich, Yarn hairiness update, Textile Progress, 26, (4)
    61. A.Barella and A.M.Manich, Hair length distribution of yarns measured by means of Zweigle hairiness tester, J., Textile Institute, 1993, 84, (3) p326
    62.G.Basel and W.Oxenham, Comparison of properties and structures of compact and conventional spun yarns, Textile Research J., 2006, 76, p567
    63. M.Krifa and D.Ethridge, Compact ring spun yarn � An examination of productivity issues, Textile Topics, 2, p2
    64.F.Goktepe, D.Yilmaz and O.Goktepe, A comparison of compact yarn properties produced on different systems, Textile Research J., 2006, 76, p226
    65. L.Cheng and X.Wang, Relationship between hairiness and twisting principles of solospun and ring spun yarns, Textile Research J., 2004, 74, sept.
    66. L.Chang and X.Wang, Comparing hairiness of solo spun and ring spun worsted yarns, Textile Research J., 2003, 73, p640
    67.M.Miao and X.Wang, Reducing yarn hairiness with an air-jet arrangement in winding, Textile Research J., 1997, 67, p481
    68.K.P.Chellamani, D.Chattopadhyay and K.Kumaraswamy, Yarn quality improvement with an air jet arrangement in cone winding, Indian J., of Fibre and Textile Research, 2000, 25, p289
    69.S.S.Salem and M.Azam, Impact of air jet nozzle pressures and winding speed at autocone on imperfections and hairiness of 20s cotton yarn, Pakistan Textile J., 2004, March
    70.Y.Zeng and C.W.Yu, Numerical and experimental study on reducing yarn hairiness with jet ring and jet wind, Textile Research J., 2004,74, March, p222
    71.R.S.Rengaswamy, V.R.Kothari, A.Patnaik, A.Ghosh and H.Punekar, Reducing yarn hairiness in winding by means of jet: optimization of jet parameters, yarn linear density and winding speed, Autex Textile J., 2005, 5, Septr., p128
    72.K.P.S.Cheng and C.H.L.Li, Jet Ring spinning and its influence on yarn hairiness, Textile Research J., 2002, 72, p1079
    73.K.Ramachandran and B.S.Dasaradan, Design and fabrication of air jet nozzles for air vortex ring spinning system to reduce hairiness of yarn, IS(I) JournalTX, 2003, 84, Aug, p6
    74. X.Wang, M.Miao and Y.L.How, Studies in jet ring spinning Part I Reducing hairiness with jet ring, Textile Research J., 1997, 67, p253
    75.R.S.Rengaswamy, V.K.Kothari, A.Patnaik, and H.Punekar, Airflow simulation in nozzle for hairiness reduction in ring spun yarns Part I Influence of airflow direction, nozzle distance and air pressure, J., of Textile Institute, 2006, 97, Issue 1, p89
    76.A.Patnaik, R.S.Rengaswamy, V.K.Kothari, and H.Punekar, Airflow simulation in nozzle for hairiness reduction in ring spun yarns Part II Influence of nozzle parameters, J., of Textile Institute, 2006, 97, Issue 1, p97
    77.J.Lunnenschloss and E.Hummel, Textil Praxis, 1968, 23, p591
    78.Y.Huh, Y.R.Kim and W.Oxenham, Analysing structural and physical properties of ring, rotor and friction spun yarns, Textile Research J., 2002, 72, Feb, p56
    79. G.M.E.Kog and T.Ogulata, Textil Teknik, 1999 Septr., p88
    80.Anon, Fibres and Textiles in Eastern Europe, 2006, July/Septr., 14, p357
    81.A.K.Soe, M.Takahose, M.Nakajima and T.Matsuo, Structure and property of MVS yarns in comparison to ring and rotor yarns Textile Research J., 2004, 74, p819
    82.A.Khoddomi, M.Siavashi, S.A.H.Ravandi and M.Morshed, Enzymatic hydrolysis of cotton fabrics with weft yarns produced by different spinning systems, Iranian Polymer J., 2002, 11, No2, p99
    83.G.Basal and W.Oxenham, Effect of some process parameters on the structure and properties of vortex spun yarns, Textile Research J., 2006, 76, p492
    84.H.G.Ortlec and S.Ulku, Effect of some variables on properties of 100%cotton vortex spun yarns, Textile Research J., 2005, 75, p458
    85.G.Basal and W.Oxenham, Vortex spun yarn vs air-jet spun yarns, Autex Research J., 2003, 3 No3

    Batt manufacture
    The polymer chips laid in hopper are melted inside an extruder. The molten material is filtered and is pumped by a spinning pump through a spinneret to form a bunch of filaments. As the fabric is wide two or three spinnerets are laid side by side to increase the number of filaments. The emerging filaments are quenched by a stream of cold air (Fig 22) and are attenuated mechanically or pneumatically to orient the molecules so as o increase the strength. There are two methods of batt formation
    1.Partial orientation
    2.Full orientation
    The strength improvement by partial orientation is adequate for many products like coverstock for diapers and hygiene materials. With partial orientation higher production rates can be achieved. Partial orientation is achieved pneumatically by an air generation unit. However for products like geotextiles, carpet backing, roofings and industrial products, full orientation is achieved by drawing the filaments over heated godets, with draw ratio of 1 : 3 or 4 followed by pneumatic acceleration. The filaments are then passed through a pneumatic air gun where high velocity air is forced through a constricted area of low pressure. This helps to achieve high uniformity and cover. Electrostatic charge is applied to keep the filaments separate. Batt formation takes place by random and uniform deposition of filaments on a moving perforated conveyor. The batt is then taken forward for bonding. Lay down with preferential direction in cross or machine direction is also possible. Suction is provided under the conveyor to improve randomisation. Fig 22 outlines batt formation in spunbonding.
             Spunbonding
          Fig :22 Spunbonding
    Recofil system developed by Reifenhauser is the most widely used system of spunbonding. Latest Reicofil model 4 can produce light weight material for hygienic products at speeds of 800 meters/min. Energy consumption of 1 � 1.2 kwh/kg for melting and conveying the raw materials. Other systems are �Docan system� and �Lutrafil sysem�. Reiter spunbond system claims a production capacity of 5000 tons/year/beam and energy consumption less than 1kwh/kg Hills and Fre specialise in spunbonding equipment from bi and multi-component material and claim speeds over 5000 metres/min, In the latest developments, special emphasis is laid on energy conservation and reuse of waste made in production.Ratio between he components can be varied from 50 : 50 to 85 : 15. Spunbonding from bi component and multi-component fibres provides products with unique properties. The different components are melted in separate extruders and passed through separate spin assembly and afterwards combined to pass through special spinneret orifices to form bi or multi-component filaments Special spinneret orifices are available to form core/sheath or side by side and segmented pie profile bi-component structures.
    The relative merits of spun bonding and staple fibre bonding are given in Table 5 below.
    Table 5 : Relative merits of filament bonded and staple fibre bonded Nonwovens
    Spun BondingStaple fibre bonding
    Higher strengthLower Strength
    Lower ElongationHigher Elongation
    Higher Uniformity in thickness and gsmLower uniformity in thickness and gsm
    Lacking in textile character and feelHas good textile character and feel
    Higher tear strengthLower tear strength
    Products are normally of low to medium gsm (20 � 250)Wide range of products from low to medium to high gsm are available (20 � 1500)
    Less flexibility in regard to raw material. Generally line is suitable for either polypropylene or polyester or Nylon or bicomponent fibre. Some latest models however claim flexibility in regard to fibre All types of raw material can be processed in the same lineAll types of fibres, manmade and natural can be made in the same line
    High Plant capacity in terms of productionLow to medium capacity in terms of production
    High capital investmentLow to Medium capital investment
    Major Applications
    1.Coverstock for diapers and hygiene products
    2.Surgical materials
    3.Carpet backing
    4.Bedding and furniture
    5.Geotextiles
    6.Roof Materials and other construction material
    7.Filters
    8.Industrial products
    Meltblown Nonwoven
    Melblown nonwoven is batt formed of ultra fine filaments deposited on a screen. Polymer is extruded through a die consisting of several hundred holes.(Fig 23) Streams of hot air at temp of 220 to 3400C, passing from an outer channels of the die, rapidly attenuate the extruded polymer at 0.5 o 0.7 times the speed of sound, to form extremely fine micro filaments. The filaments are quenched by cool air flowing from the contours of the die and blown on to collector screen thus forming a self bonded nonwoven batt. The fibres are bonded together because of interlacing action of deposition and thermal bonding by the hot air. The meltblown material has low strength and is often used in combination with other nonwoven material. Because of the ultra fine size of the filaments and increase in specific surface area, meltblown material has a high filtration efficiency for gas and liquids and particles.
    Barrier material is formed by combining meltblown material with spunbonded. Absorbents are made by combining cellulose or wood pulp with meltblown material Meltblown material sandiwiched between two layers of spunbonded materials, known as sms nonwovens are widely used as hospital gowns, hygiene, and filtration products. SMMS are also sometimes made.
    Meltblown
            Fig 23 Meltblown Process
    Process Checkpoints
    1. Viscosity of polymer chips should be checked in viscosity meter.
    2.The spinnerets and other parts of the spinning system must be thoroughly cleaned of polymer build-up at periodic intervals and at the time pack change. When polymer build up contaminates the parts, production has to be brought down to maintain satisfactory processing. The quality of product also gets adversely affected by presence of defects like stripes, bars and doglegs. Traditional cleaning consists of 1 burning the hardened polymer by placing the spinneret in a furnace 2. Cleaning in an ultrasonic bath of caustic soda and 3. spray bath cleaning. This method is time consuming and there is risk of spinneret damage. The vacuum pyrolysis process of cleaning is much more rapid and ensures defect free thorough cleaning
    3. Spinnerets should be examined under magnification for wear and tear. Preferably, an automated spinneret examination system should be used to inspect the spinneret at the time of purchase and during use to ensure that the design specifications are maintained as this speeds up the process. If the holes of the spinneret show signs of wear they should be refurbished.
    References
    1. Nonwovens � Technoogy of manufacture and properties
    A.K. Rakshit, A.N. Desai and N. Balasubramanian, Proceedings of Symposium on Nonwovens, BTRA, 1987, Feb
    2. A comparison of woven and nonwoven geotextiles,
    J. Perfetti, Meliand Textilber (Eng. Edn.), 1985, March, p207
    3.Woven and Nonwoven medical/surgical materials
    D.V.Parikh, T.A.Calamari, A.P.S.Sawhney, N.D.Sachinwala, W.R.Goynes, J.M.Hemstreet, and T. Van Hoven, International Nonwovens J, 1999 Spring
    4. Development of a geocomposite canal liner � BTRA�S experience
    A.N.Desai and N.Balasubramanian, J. Textile Association, 1991 Jan, p187
    5. http://www.inda.org/wet-laid.html
    6. Relative merits of polypropylene and polyester
    N.Balasubramanian, Indian Textile J, 2004, 32, p32
    7. Influence of processing conditions on functional properties of high loft structures
    A.N.Desai and N.Balasubramanian, Indian J of Fibre and Textile Research, 1990, 15, p169
    8. Opening size and water permeability of nonwoven geotextile
    A.K.Rakshit and N.Balasubramanian, Indian Textile J., 1991 June, p 26
    9. Indian Textile J, 2009, July, p 71
    10. http://www.texdata.com/content/0082e.pdf