What Causes Barré in Knitted Fabric?

Barré is unintended banding that runs in the course direction of a knitted fabric, usually across its width. It occurs when successive courses do not reflect light, absorb dye, or form loops consistently. The source may be yarn linear density, twist, blend, fibre maturity, package or lot; feeder tension, stitch length, needles, sinkers, cams, positive feed, or take-down; or variation introduced or revealed during wet processing and finishing. Barré is therefore an appearance symptom, not one single defect with one universal correction.

Barré should be separated from an intentional stripe

AATCC terminology describes barré as an unintended, generally repetitive series of bars parallel to the courses of circular knitted fabric. In production language the term is also applied more broadly to comparable coursewise banding in other weft knits. A designed stripe changes yarn, colour, stitch, or feed according to an approved repeat. Barré is an uncontrolled difference inside an area intended to read as uniform.

The band may be tonal, glossy, dull, dense, open, thick, thin, or texturally uneven. It may be obvious in grey fabric, appear only under oblique light, or become visible after scouring and dyeing. These differences are diagnostic. A band visible before wet processing points strongly toward material or loop geometry. A band that appears only after dyeing may still originate in fibre or yarn variation; dyeing can reveal a difference in affinity that was optically quiet in the grey state.

Barré is not the same as spirality, bow, or skew. Those faults displace structural lines; barré changes appearance along them. The note on garment twist and fabric skew addresses rotational distortion rather than banding.

Yarn variation can alter mass, cover, lustre, and dye uptake

Mixed yarn lots are an immediate risk because nominally similar cones may differ in fibre source, blend proportion, colour, finish, twist, hairiness, or actual count. A change in yarn linear density changes the material entering a loop. Variation in twist changes compactness and reflected light. Unevenness at wavelengths that repeat visibly in the fabric can form a regular shadow pattern even when average count remains within tolerance.

For cotton, fibre maturity and micronaire distribution influence dye response. For blends, a shift in component ratio or mixing uniformity changes the amount of each dye-receptive fibre presented at the surface. Synthetic filaments can vary in denier, texturing, heat history, spin finish, or dye affinity. Woollen and worsted yarns can differ in fibre distribution, bulk, oil, moisture condition, and setting. A cone change is therefore not a neutral event simply because the ticket shows the same supplier code and nominal count.

Package build and storage also affect delivery. Hardness, winding tension, cone geometry, wax or lubricant, damage, and conditioning alter unwinding tension. If packages from different positions or histories are mixed across a creel, feeder-to-feeder input can become inconsistent. Cotton Incorporated's yarn-management guidance treats identification, storage, staging, and creeling as part of knitted-fabric reproducibility rather than warehouse administration.

Machine and loop-formation differences can create structural barré

Each feeder on a circular weft-knitting machine contributes courses to the fabric. If one feeder delivers a different yarn length, input tension, or material, its courses can differ from those formed at the other feeds. A positive-feed setting, pulley or belt condition, yarn path, guide, stop motion, ceramic, tension device, or package can be responsible. The effect may repeat after every complete feeder cycle and become a clearly spaced band.

Stitch length is central because it controls the amount of yarn in the loop. A shorter loop tends to create a tighter, denser, more reflective band; a longer loop can appear more open or slack. Cam setting, needle and sinker wear, needle-butt variation, dial-to-cylinder relation, timing, lubrication, take-down, speed, and thermal condition can disturb loop formation. In plated or multi-feed constructions, yarn position and feeder geometry add another visible variable.

A machine fault does not always create a sharp stripe. Slow drift in yarn input or take-down can produce broad shading. A damaged element may create a narrow line. Periodic mechanical variation can generate a repeating pattern whose wavelength relates to a rotating part. Random thick and thin places may produce cloudiness rather than classical barré. The description should record band width, contrast, repeat interval, direction, and whether the edges are sharp or diffuse.

Dyeing may reveal the defect without being its root cause

Preparation removes oils, waxes, knitting lubricants, and soils; dyeing makes chemical and physical differences visible through colour. Uneven scouring, circulation, temperature, pH, electrolyte, alkali, dye addition, fixation, rinsing, or loading can create processing variation. Yet a dyed band should not automatically be assigned to the dyehouse. If alternating feeders contain fibre with different maturity or affinity, a uniform bath can produce alternating shades.

The grey-to-finished comparison is therefore essential. If the same band can be detected by transmitted or low-angle light before dyeing, loop geometry or yarn mass is involved. If the grey fabric appears physically uniform but a band follows a yarn lot or feeder map after dyeing, fibre or yarn dye affinity remains a strong suspect. If bands follow folds, liquor flow, machine loading, or heat exposure rather than courses, wet processing may be primary.

Finishing can reduce some visual differences by relaxation, compaction, brushing, shearing, or surface levelling, but it cannot reliably erase a chemical or structural mismatch. Aggressive correction may change mass, width, handle, shrinkage, or surface while leaving the underlying periodicity. An acceptable reprocess route must be validated against the full specification, not judged only under one light immediately after finishing.

A useful diagnosis preserves the fabric's production history

Begin by quarantining representative rolls and retaining the exact yarn packages, creel map, machine number, diameter and gauge, feeder plan, style program, stitch-length records, production time, operator record, and wet-process batch. Mark the wale and course direction. Examine the fabric flat and relaxed under diffuse, oblique, and transmitted light; compare face and reverse. Do not cut away the repeat before its spacing is measured.

Next compare grey, scoured, dyed, and finished samples from the same sequence. Measure the band interval in courses and physical distance. Relate that interval to feeder count, machine revolution, package changes, shift events, and rotating components. Knit controlled comparison tubes by moving suspect packages between feeders, replacing one variable at a time, and preserving an unaffected control. If the band follows the package, the material path is implicated; if it remains at the feeder, inspect the machine path and loop setting.

Yarn tests may include actual count, evenness and periodic mass variation, twist, hairiness, tensile properties, friction, moisture, composition, fibre maturity, filament denier, and dye response as appropriate. Fabric checks include course and wale density, loop or stitch length, mass per area, width, colour difference, and microscopic loop comparison. Instrumental colour readings are useful, but a spectrophotometer cannot fully describe texture, lustre, or directional reflection; controlled visual assessment remains necessary.

For luxury knitwear, the same logic applies at a smaller production scale. Flat-knitted panels from separate cones, machine beds, or times can shade differently even when the silhouette and stitch program are correct. The production file should identify yarn lot and cone allocation by component, machine and program version, knitting sequence, relaxation time, finish, and assembly pairing. This keeps sleeves, fronts, backs, collars, and trims within a controlled visual family.

Prevention is a chain of material and machine controls

Use one approved yarn lot for a visually continuous area wherever possible. Segregate packages, preserve labels, condition them consistently, and record their creel positions. Approve actual count and critical evenness data rather than only the nominal ticket. Establish changeover rules so a replacement cone is not introduced casually into the middle of a sensitive body.

At the machine, calibrate positive feed and input tension, verify stitch length by feeder, inspect guides, needles, sinkers, cams, and take-down, and maintain a repeatable thermal and lubrication state. Produce a first-off length long enough to reveal the full suspected periodicity. Inspect both grey and trial-dyed fabric before authorising bulk, because some material barré remains almost invisible until colour development.

At finishing, keep lots and roll sequences traceable, use controlled preparation and dyeing, and retain standard lighting and approved references. The sample route should reproduce bulk yarn lot, machine settings, finishing, and rest. These controls belong in the sample-to-production handoff, not in an informal correction made after inspection.

Common questions about barré in knitted fabric

Barré control depends on traceability, not visual repair

The fastest route to a cause is an unbroken record from fibre and yarn through creeling, knitting, preparation, dyeing, finishing, cutting, and assembly. Without that history, several unrelated variations can produce the same visible band. With it, the repeat can be mapped to a physical event and confirmed by a controlled trial.

Material and surface studies appear in Work. The Process page outlines the relationship between stitch development, technical documentation, sampling, fitting, and factory collaboration. Production-development enquiries can be sent through the contact form.