BARRÉ: METHODS TO PREVENT

BARRÉ IN KNITTED FABRIC

Presented at 15th EFS Conference

by

Donald L. Bailey
Vice President,
Textile Research and Implementation
Cotton Incorporated
 BARRÉ: METHODS TO PREVENT BARRÉ IN KNITTED FABRIC

Donald L. Bailey, Cotton Incorporated

In textile production, one of the most common and perplexing quality issues is

“barré.” The factors that can cause or contribute to barré are varied and diverse.

For this reason, when a barré problem is detected, the skills of a sleuth may be

required to expose the problem and eliminate its cause. Once a cause is

identified, steps can be taken to minimize or eliminate the barré, and better

quality fabrics can be produced. This discussion will center on knitted fabrics.

DISCUSSIONS

The noun “barré” is defined by ASTM1 as an unintentional, repetitive visual

pattern of continuous bars and stripes usually parallel to the filling of woven fabric

or to the courses of circular knitted fabric. In a warp knit, barré normally runs in

the length direction, following the direction of yarn flow. Barré can be caused by

physical, optical, dye related differences in the yarn, geometric differences in the

fabric structure, or by any combination of these differences. A barré streak can

be one or several courses wide. A “typical” characteristic of barré is that it

generally consists of stripes that repeat. Isolated or intermittent defects may or

may not be barré.

1
ASTM D123-96a – Standard Terminology Relating to Textiles

1
Various aspects of barré are important for consideration and discussion. The

first is the identification of barré. Secondly, how can barré be analyzed? What

are the causes of barré, and finally, how can it be prevented?

IDENTIFICATION OF BARRÉ

The first step in a barré investigation is to observe and define the problem. Barré

can be the result of physical causes that can usually be detected, or it can be

caused by optical or dyeability differences that may be nearly impossible to

isolate in the fabric. Barré analysis methods that help to discriminate between

physical barré and barré caused by other reasons include Flat Table

Examinations, Light Source Observation, and the Atlas Streak Analyzer.

Flat Table Examination

For a visual barré analysis, the first step is to lay a full-width fabric sample out on a

table and view both sides from various angles. Generally, if the streaky lines run in

the yarn direction that is in the course direction, apparent color differences can be

seen by looking down at the fabric in a direct visual line with the yarn or course

direction, and the defect can be positively identified as a barré defect. Viewing the

fabric with a light source in the background will show if the barré is physical.

Light Source Observation

After completing an initial Flat Table Examination, a Light Source Examination may

provide further useful information. Full width fabric samples should be examined

2
under two surface lighting conditions, ultraviolet (UV) and fluorescent light.

Observations that should be made while viewing under lights are:

1. the frequency and width of the barré,

2. whether the streaks are dark or light, and

3. the total length of pattern repeat.

Ultraviolet light, commonly referred to as "black light," allows the presence of

mineral oils to be more easily detected, due to their radiant energy (glow). When

observed under UV light, fabrics with streaks that exhibit glow suggest improper or

insufficient preparation. A change in composition or content of oil/wax by the

spinner or knitter without appropriate adjustments in scouring can create this

problem. Fluorescent lighting simulates the mode of observation that is common to

most inspection tables in mills and will highlight whether the barré is perceivable in

an industry quality control setting.

Atlas Streak Analyzer

The function of the Atlas Streak Analyzer is to isolate barré caused by physical

differences. A fabric swatch is combined with polystyrene sheet film, and the Atlas

Streak Analyzer produces a plastic impression of a fabric surface by incorporating

specific conditions of pressure and heat. The absence of color on the plastic

impression ensures that only physical streak effects will be seen. The plastic

impression is examined to determine whether the streak alignment matches the

3
streaks observed on the fabric. However, impressions made from spun yarns such

as cotton can be difficult to read due to the inherent yarn variation characteristic of

spun yarns. Also, a too rapid cooling of the test specimen after making an

impression can produce a moiré pattern. From a valid plastic impression, the barré

source can be identified as:

1. physical with all streaks showing on the impression,

2. optical or dyeability variations where none of the color streaks are aligned on

the impression, and

3. a combination of physical and dyeability differences where some streaks align

with those on the impression, and some do not.

Fabrics with combination causes present the greatest challenge for analysis.

If the streak analyzer indicates the same barré pattern as seen in the fabric, then

the barré stripes are physical in nature. This can relate to several physical

causes such as yarn tensions, stitch length, yarn count, twist differences, etc.

Yarn tension causes can be found by raveling adjacent courses and measuring

the lengths of yarn removed from each course. If all the lengths raveled from the

fabric are the same, the pieces can be weighed to determine if the yarn counts

are the same.

4
If the plastic replica shows no stripes, then the barré is due to chemical causes or

to light reflectance differences. Chemical causes relate to improper preparation,

and light reflectance differences relate to non-uniform dyestuff penetration or

reflectance. The next step is to remove the color and evaluate the stripped

sample before re-dyeing to determine if the removal of dye was complete and if

the barré is still present. An uneven or incomplete stripping can indicate an

additional strip. If the color is stripped uniformly and the barré is gone, then the

sample should be re-dyed. If after stripping and over dyeing the fabric no longer

has barré, then the barré was caused by improper preparation. If the barré

remains, then the problem is related to optical or light reflectance problems.

PHYSICAL BARRÉ ANALYSIS

When the cause of barré is determined or presumed to be physical in nature,

physical fabric analysis should be done. Physical barré causes are generally

considered to be those which can be linked to yarn or machine differences.

Methods of physical barré analysis include fabric dissection, microscopy, and the

Roselon Knit Extension Tester.

Fabric Dissection

To perform accurate fabric dissection analysis, a fabric sample that contains

several barré repetitions is required. First, the barré streak boundaries are marked

by the placement of straight pins and/or felt markers. Individual yarns are removed

from light and dark streak sections, and twist level, twist direction, and cut length

5
weight determinations are made and recorded. For reliable mean values to be

established, data should be collected from at least two light/dark repeats. After

compilation of yarn information, the numbers can be compared individually to

adjacent yarns as well as by groupings of light and dark shades.

Microscopy

Microscopic examination is useful for verifying yarn-spinning systems. Yarns from

different spinning systems can have different light reflectance and dye absorption

properties resulting in barré when mixed. Ring-spinning produces yarn that is

smooth with all fibers twisted in a tight helix. Open-end spinning produces yarn

with wrapper fibers that form a belt around the diameter of the yarn at irregular

intervals. Air jet spinning produces yarn with more wrapper fibers that form a

continuous spiraling band around the inner fibers that are more parallel to the axis

of the yarn. Microscopy can also reveal a shift in loop formation in knitted fabrics

when twist direction (S and Z) differences are present.

Roselon Knit Extension Tester2

Barré produced by knitting machinery is relatively common, is the easiest to see

in the greige, and is the easiest to correct. Often uneven yarn tension during

knitting may be a cause. To test for uneven tension, the Roselon Knit Extension

Tester can be used. For this test, a fabric sample is cut and raveled to yield yarn

samples from light and dark streak areas. The yarn ends are taped and clamped

2
Source: Spinlon Industries Incorporated, 18 S. Fifth Street, Quakertown, Pennsylvania 18951

6
to the tester. As each yarn is stretched to the maximum extension point, the

points are plotted on graph paper. Comparisons are usually made visually rather

than mathematically.

CAUSES OF BARRÉ

The varied and diverse causes of barré can generally be summed up in one word -

INCONSISTENCY. An inconsistency that leads to barré can originate in one or

more of the following categories:

! fiber quality/raw material management,

! yarn formation/supply management,

! knitting processes, and

! preparation and dyeing techniques

Fiber Quality/Raw Material Management

1. Failure to control fiber diameter (micronaire or denier) from laydown to laydown.

2. Too high a C.V. of micronaire in the laydown for a given mill's opening line

blending efficiency.

3. Failure to control the fiber color in the mix (grayness Rd, yellowness +b).

4. Failure to control maturity/fineness in a laydown

5. Most, if not all, fiber barré can be controlled by the above four items; however,

under certain unusual circumstances, it may be beneficial to also select mixes

using ultraviolet reflectance information for each bale of cotton.

7
Micronaire

Average micronaire must be controlled within a laydown and from laydown to

laydown. Controlling average micronaire in the laydown may not be sufficient to

completely eliminate barré. Other micronaire related causes of barré are:

! > 0.2 difference in micronaire.

! > 0.1 change in mix-to-mix average micronaire.

! > 12.0 % CV of micronaire within the laydown.

It may be necessary to change the laydown averages periodically to make use of

all the bales in the warehouse. This must be done slowly with no more than a

0.1 change in mix-to-mix averages. Figure 1 shows acceptable changes in

micronaire from laydown to laydown.

No barré from micronaire differences should show in knitted fabrics when yarns

are mixed on the knitting machine from consecutive laydowns. However, if

laydowns vary more than 0.1 micronaire, for example laydowns #1 (4.3 mic.) and

#6 (4.6 mic.) in Figure 1, then barré is much more likely to occur.

8
 A c c e p t a b le C h a n g e in M ic r o n a ir e
F ro m L a y d o w n to L a y d o w n

4 .5
4 .4
4 .3
Max +/- 1.0 mic
4 .2

Micronarie
4 .1
4 .0
3 .9
3 .8
3 .7
3 .6
3 .5
1 2 3 4 5 6 7 8 9 10 11 12 13 14
La ydow n Num be r

Figure 1: Slowing Changing Micronaire Average in Laydown

Figure 2 shows a high CV% in the micronaire for bales within a laydown. The

laydown shown has 24 bales with an average of 4.1 micronaire and a CV% of

17.7. The change in micronaire from bale to bale is more than a 0.1 change.

Also, the CV% is more than 12.0% and would probably result in barré.

H igh M icronaire CV % in Laydow n

4.7 A vg . 4 .1 CV % 1 7.7 *
4.5
Micronaire

4.3
4.1
3.9
3.7
3.5
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

B ale N um ber

Figure 2: Results of Changing the Micronaire Average too Rapidly

9
Maturity and Fineness

Although micronaire gives an indication of fiber maturity, it does not directly

measure that property. Maturity and fineness related causes of barré include:

! Blending cotton from different growth areas or seed varieties.

! Immature fiber content such as white specks, neps, etc.

Yarn Formation/Supply

Even with proper fiber selection to reduce raw material effects on barré, techniques

in the opening, cleaning, blending, and carding areas can have an impact on barré.

These variations can occur at carding where different amounts of non-lint content

removal from card to card can be a problem. Poor blending of fiber from opening

through finisher drawing cannot be overlooked. In ring spinning, the running of

different types of spindle tapes and the use of different cots or new and used cots

mixed together on the same frame can create barré.

Faulty management in the spinning plant can result in the following yarn related

causes of barré:

1. Yarn count variations.

2. Yarn twist variation.

3. Wrong yarn – size, color, blend level, twist direction.

4. Mixed yarn lots.

5. Mixed shipment dates of same dye lot.

10
6. Uneven waxing or oiling of yarn

7. Improper conditioning.

Knitting Processes and Yarn Management

The set-up and operation of a knitting machine is complex and requires precision

settings to produce first quality goods. Machine related causes of barré include:

1. Different stitch settings (stitch lengths).

2. Improper tension at a feed.

3. Faulty cylinder or dial cam settings.

4. Malfunctioning of storage or tape feeders.

5. Improper threading of yarn.

6. Variations at take-down or spreader system.

7. Machine vibration.

8. Dirt, lint, and/or yarn fragments in the camming system, tricks, needles, or

sinkers.

9. Variation in oil content.

10. Worn needles, which generally produce length direction streaks.

11. Uneven cylinder height needles (wavy barré).

12. Worn cylinder and/or dial.

Even with a properly set-up machine, barré can still occur at knitting due to poor

yarn management. Examples of poor yarn management at knitting include:

11
1. Mixing yarns of different counts.

2. Mixing yarns from different spinning systems.

3. Mixing yarns with different blend levels.

4. Mixing yarns from different suppliers.

5. Mixing yarns with different twist level/twist direction.

6. Mixing yarns with different degrees of hairiness.

7. Mixing yarns with different amounts of wax.

8. Mercerization differences.

9. Excessive backwinding or abrasion during this process.

10. If yarns are conditioned, then each lot must be uniformly conditioned.

Preparation and Dyeing Techniques

Normally, dyehouses have standard preparation procedures and formulas that

work very well in terms of removal of non-fibrous contaminants and natural fiber

pigmentation. However, some fabrics can exhibit barré that is caused by either

optical or dyeability issues. Often, the barré is not created in dyeing, but can be

highlighted. In this scenario, the dyeing is not the cause but the “messenger” of the

problem.

Typically, barré is not a problem in whites, only in dyed fabrics. Therefore, if a

fabric is known to have barré, then that roll should be processed into a white.

12
Preparation can help “mask” barré, but cannot and will not eliminate it. The cause

of the barré must be eliminated for future rolls to be free of the defect.

The scour and/or bleach procedure can be intensified to mask the barré. This can

be done by combinations of using a higher temperature or longer processing times.

Usually the chemistry will remain the same; however, in some cases, it may have to

be increased or changed.

If any changes are made in the preparation procedure, the fabric and resultant

shade will possibly change in whiteness, strength, and appearance. In fact, the

whiteness will almost always change. If whiter, the old dye formula will need to be

adjusted. The strength of the fabric may be affected to the point where it becomes

a problem. Finally, the surface of the fabric may change so that it may not match

previous lots. As a result of more aggressive preparation to cover barré, it is

important to evaluate such changes on small samples in the lab or a sample

machine.

Tension can play an integral role in not only the formation of the fabric, but also

during preparation, dyeing, and finishing. As with all variables, it must be controlled

to known parameters and be consistent. Any inconsistencies in tension could

change the morphology of the cotton (i.e. during mercerization), level of dye pick-

13
up, or surface appearance (i.e. during mechanical finishing such as brushing or

sanding).

PREVENTION OF BARRÉ

As discussed in the text, barré is caused by inconsistencies in materials,

equipment, or processing. To prevent barré from occurring, consistency must be

maintained through all phases of textile production. Stock yarns should be properly

and carefully labeled to avoid mix-ups. Fugitive tints and/or marked cones can be

useful for accurate yarn segregation. Inventory should be controlled on a first

in/first out basis. All equipment should be properly maintained and periodically

checked. In spinning and knitting before beginning full-scale production, sample

dyeings can be done to check for barré. Knit machine operators should be trained

to look for barré as it occurs in the greige. If spotted, the machine should be

stopped until the cause is eliminated.

Salvaging a fabric lot with a barré problem may be possible through careful dye

selection. Color differences can be masked by using shades with very low light

reflectance (navy blue, black) or high light reflectance (light yellow, orange, or

finished white). Dye suppliers should be able to offer assistance in this area. Also,

if the cause of the barré is an uneven distribution of oil or wax, a more thorough

preparation of the fabric before dyeing may result in more uniform dye coverage.

14
With close cooperation between production and quality control personnel, barré

problems can be successfully analyzed and solved. Recommendations to

minimize barré include:

! Knit an entire dye lot from the same knitting machine.

! Use only yarn from the same spinning lot.

! Use only from the same shipment date if possible.

! If yarn shipment dates must be mixed, then use consecutive shipment dates.

! Determine through laboratory analysis and experience if the preparation

procedures are sufficient or can they be modified to eliminate the problem.

! Determine if some shades and dyestuffs are less susceptible to showing

barré, and apply those to problem fabrics.

! Make use of yarn/fabric analysis systems such as CYROS®.

! Identify those rolls within a dye lot that have mixed yarn shipment dates so

that it is known when the fabrics go through the dyehouse.

! Identify dye lots that have rolls from different yarn shipments before dyeing.

CASE STUDY IN BARRÉ

A 100% cotton, ring-spun single jersey style T-shirt is knitting in production with a

yarn that does not have a barré problem. This style has been knitting for

numerous days and no more of the yarn is available for creeling on the machine

as the yarns are knitting out. Some yarn packages are “skinners” and others are

full five-pound packages. The machine is 24-inch diameter with 80 feeds.

15
When the new yarn is creeled onto one of the 80 positions, the mixing of yarn

shipment dates begins to take place. For this discussion, we will assume that

this new yarn will cause barré when mixed with the old yarn. Once all positions

are knitting the new yarn, there will be no barré. How many rolls of fabric will

have barré when both these yarns are present in the rolls?

If the knitter is making 50 pound rolls, then each yarn package will supply 1/80th

of the 50-pound roll or 0.625 pounds of yarn. Therefore, the new yarn will make

8.0 rolls of fabric with barré (5-pound package divided by 0.625 pounds per

package for a 50 pound roll). It is important to realize that while this new

package is knitting, other old yarn positions on the creel are also being replaced

by new yarns. By the time the first new package has knitted out, all feeds will be

using the new yarn. When the last old package is replaced and all positions

have new yarn, the barré will disappear.

Often, many machines will be knitting the same style. If ten machines are

knitting this style and are using the old yarn and new yarns are placed on these

80 feed machines, then the total number of rolls knitted with barré will be 80. A

total of 4,000 pounds of fabric will be made with barré. If the style in question

weighs 5.8 ounces per linear yarn, then a total of 11,035 yards of fabric will be

made. Further, if it can be assumed that each yard of fabric can make 1.2 shirts,

then a total of 13,240 defective T-shirts will be made. This is 1,103 dozen

16
garments. If the cost per shirt is $2.05 per unit, then a loss of $27,142 plus is

realized.

If open-end yarn was used for this case study instead of ring-spun yarns, then 8-

pound packages of yarn could be used. This would result in 128 rolls containing

6,400 pounds and 17,655 yards of fabric. From this, 21,186 T-shirts (1766

dozens) would have been made. Based on a cost per shirt of $1.86, the loss

would be $39,405. The result is that bigger packages mean bigger losses.

The best methods to reduce the possibility of this catastrophe are listed below:

1. When mixing yarn shipment dates of only one week, try to reduce the number

of machines using the old and new yarns.

2. As more machines are expending the old yarn, consolidate the old packages

to fewer machines. This means removing the yarns from one creel and using

them on other machines that are still knitting the old yarn.

3. Knit dye lots from a single machine if possible.

4. When a roll is known to have mixed yarns, a laboratory dyeing should be

done on a swatch from the roll to determine if barré is present. If not, then

proceed as normal. If barré is present, then the roll should be processed in

17
 shades that are not known to be barré sensitive or prepared with more

aggressive chemistry.

CONCLUSIONS

It is evident that barré is a problem that results from inconsistencies and is a

result of poor management of fiber, yarn, and/or related knitting processes. The

spinner, the knitter, and the dyer must communicate and work as a team to

reduce the potential for barré to occur. A well planned and executed system of

monitoring the spinning, knitting, dyeing, and finishing systems in the mill can

provide for defect free fabrics.

18