A yarn package can look sound, weigh correctly and still cause trouble as it empties. The first layers may unwind well, while the last layers catch, shed loops or give sharp tension changes. An average density value cannot explain all of those faults.

Package winding therefore has to be judged at both ends of the route. The winder must build a stable package without harming the yarn. The next machine must withdraw that yarn under the conditions the product needs. For a dye package, liquor must also reach the yarn throughout the package.

This guide follows those requirements through measurement and a practical trial. All quantities and trial observations below are invented examples. They are not factory results, recommended settings or acceptance limits.

Start with the package's next job

Write down where the yarn will go after winding. A package feeding a warping creel has a different route from one entering a dye vessel. A weft supply package and a knitting supply package may also face different withdrawal speeds, guides and stop-start cycles.

Ask whether an extra winding pass is needed at all. Some spinning systems already build cross-wound packages. For example, Rieter's R 70 brochure describes an integrated winding system with separate package and yarn-guide drives. It allows different package densities and measured yarn lengths. This does not prove that every resulting package suits every next process. It does show why the route should start from the package actually supplied.

Rewinding may be justified by a different tube, package shape, yarn length or further processing need. State that purpose before adding the pass. More handling is not evidence of better preparation.

Also keep winding separate from warping, sizing and lubrication. They serve different purposes. Winding cannot replace a required warp treatment merely by making the package firmer. Nor should a knitting yarn receive a finish simply because another yarn used one. Establish the actual yarn and machine requirements.

Keep three package properties separate

Tension, density and unwinding performance are linked, but they are not interchangeable measurements.

Property

What it describes

What it cannot establish alone

Winding tension

Force in the running yarn at a stated point

Density at every place in the finished package

Average package density

Net yarn mass divided by its occupied package volume

Local bands, edge build or withdrawal behaviour

Unwinding performance

Yarn withdrawal under the tested downstream conditions

Performance on a different guide path or machine

Surface firmness adds another observation. A hand squeeze can help find an obvious soft area, but it does not measure the mass inside a known volume. If a hardness instrument is used, state its method and measurement positions. Do not rename its reading “density” without a valid relationship for that package.

A firm outer layer can hide a different inner build. The whole-package average can also conceal a dense band beside a softer zone. Inspect the shape and winding pattern as well as weighing the package.

This separation makes fault reports more useful. “The package is hard” is an observation needing a method. “The winding tension was high” is a claim about a force during production. Neither statement identifies why a loop caught at the next machine.

Understand which controls build the package

The yarn arrives under tension, the traverse lays it across the package, and the package grows as winding continues. Depending on the machine, contact pressure, yarn delivery and package drive also influence the build. These controls act together; changing one setting may alter more than one result.

Rieter's Preci FX brochure identifies crossing angle, yarn tension, winding speed and cradle pressure as package-design parameters. It distinguishes dense packages for further processing from lower-density dye packages with rounded flanks. These are supplier examples of different design aims, not universal density limits.

Cradle pressure and yarn tension are different things. One concerns contact at the package support or drive; the other concerns force along the yarn. A numbered machine setting may also be a control scale rather than a measured force. Read the actual machine definition before transferring a value to another head.

Feedback control does not remove the need for output checks. SSM's explanation of preciforce describes adjustment of backpressure to regulate package density. It illustrates the difference between watching a result and acting on it. It does not establish the density achieved by an untested recipe in another mill.

For a trial, retain the actual winding recipe and any changes made during the build. If speed was reduced near the end, record that fact. Two packages with the same nominal recipe may have followed different paths to completion.

Calculate density from net yarn mass

Use the volume occupied by the wound yarn, including the spaces between its strands. Exclude the tube or core. This gives an average bulk package density; it is not the density of the fibre substance itself.

Consider an ideal cylindrical package with straight ends and no bulging. Its outside diameter is 17 cm, its tube outside diameter is 7 cm, and its wound length along the axis is 15 cm. The tube runs through that full wound length.

For this geometry:

Yarn-envelope volume = π × wound length × (outside diameter² − tube outside diameter²) ÷ 4

Volume = π × 15 × (17² − 7²) ÷ 4

Volume = 900π ≈ 2,827.43 cm³

Suppose the package's gross mass is 1,595 g and the empty tube weighs 145 g. The net yarn mass is 1,595 − 145 = 1,450 g.

Average package density = 1,450 ÷ 2,827.43 ≈ 0.5128 g/cm³

Using gross mass would give about 0.5641 g/cm³. That is 10% higher, solely because the tube was counted as yarn. The winding process has not changed; the calculation has.

A second package with the same ideal dimensions contains 1,650 g of yarn. Its average density is about 0.5836 g/cm³, with 13.79% more yarn in the same envelope. That comparison says nothing yet about whether it dyes or unwinds better.

Use full precision during the calculation and round only the displayed result. Record the weighing basis too. Packages compared at different moisture states can differ in mass without having the same difference in fibre packing.

Use a shape that matches the package

The cylinder formula is unsuitable for a conical package or a barrel with rounded shoulders. Measure the actual profile and use a justified volume method. A single diameter measured at the widest point can overstate the volume occupied elsewhere.

Keep the tube's outside dimensions distinct from its bore. The yarn begins at the outer tube surface. Subtracting only the central hole would leave the tube wall inside the supposed yarn volume.

For a soft package, avoid squeezing it during measurement. Agree where dimensions are taken and how much contact force the method uses. Retain the raw mass and dimensions so that a later reviewer can check the calculation. A density figure with no geometry is difficult to reproduce.

Look for repeated bands and weak edges

Package build is also about where each new wrap lands. Repeated placement over the same tracks can create ribboning or other unwanted patterns. A useful inspection follows the whole barrel and both ends, rather than selecting one smooth-looking area.

Koranne and Mhedu's 2016 study defines traverse ratio as the number of coils laid in a double traverse. It examines how selected ratios and their changes affect step-precision package build. The study also reports difficult transition regions in its trials. Its ratio tables should not be treated as settings for every yarn or winder.

Check what a supplier means by winding angle or crossing angle. Different labels may use different reference directions. Copying a number without its definition can produce a different lay.

Mark the position of a band before dismantling a package. Record its distance from an end and its diameter at that stage. If a fault occurs during withdrawal, this gives a way to compare the event with the build. A photograph taken only after the affected layers have gone may lose the clue.

Loose turns beyond an edge, damaged tube ends and slipped groups of wraps need separate descriptions. They need not share one cause. Identify which feature changed before deciding whether tension, traverse, handling or another part of the route needs attention.

Follow the yarn from package to guide

During over-end unwinding, the package can remain stationary while the yarn moves over its surface and lifts into a rotating balloon. The guide position, package shape and yarn path are therefore part of the test.

Kong, Rahn and Goswami's 1999 study examined both the sliding yarn on a cylindrical package and the balloon beyond it. The work combined models with measured motion and tension. It helps explain why a force measured at a guide is not simply the tension that was applied when the yarn was wound.

Record which end of the package faces the guide, its alignment and the withdrawal direction. Include any balloon-control device and later tensioner. If one package is tested through an extra guide, the comparison includes that extra contact.

The guide distance deserves care. Celik and Eren's 2019 experiments studied balloon length, package diameter, yarn type, count and withdrawal speed. The effects differed with the tested conditions. Their findings support checking the actual combination rather than declaring one guide distance best for every yarn.

Do not hand-pull a short length and call the package approved for high-speed use. Such a check may reveal an obvious snag, but its speed and yarn path differ from production. Retain it as a limited inspection, then test the intended withdrawal route.

Measure tension where the decision is made

First decide whether the question concerns winding tension or downstream withdrawal tension. Name the measurement point. Include the instrument, range, calibration status, yarn path and running speed in the trial record.

Install and use the sensor according to its instructions. Its guides become part of the yarn path during the measurement. A changed wrap path or a dirty contact can alter the result. Keep people and tools clear of moving parts; use the machine's approved test procedure.

Ali, Ahmed and Amer's 2021 polyester soft-winding study used measured tension as feedback to a control system. Its tests concerned specified polyester yarns and equipment, and its reported comparison used a moderate supply-package diameter. That boundary matters: the paper does not establish full-to-empty performance for every package. Its proposed tension value is not a universal recipe.

Record more than an average. Tension varies over time, and a brief event may matter even when the mean changes little. State the recording interval and any filtering. The largest recorded reading is only the largest value the measurement captured.

Two short records with the same mean

Imagine two sets of ten equally spaced readings at the same downstream guide. All values are in cN, and both records cover the same duration:

  • Record A: 8, 10, 9, 11, 10, 8, 9, 11, 12, 12.

  • Record B: 6, 6, 7, 7, 7, 8, 8, 9, 12, 30.

Each set totals 100 cN across ten readings, so each mean is 10 cN. Yet the recorded maximum is 12 cN for A and 30 cN for B. The recorded ranges are 4 cN and 24 cN.

These tiny examples do not define a sampling plan or prove that either yarn will break. They show why equal means need not describe equal behaviour. The record may also miss a faster event between samples. Select equipment and recording conditions that can capture the event being investigated.

Keep units explicit. A force in cN and a force divided by yarn linear density are different quantities. A normalised value may help compare records, but it does not replace a yarn-strength distribution or an agreed running trial.

Include the last layers in the trial

A package's appearance changes as it empties, but diameter is not proportional to the amount of yarn left. The core occupies part of every measured cross-section.

Return to the ideal package with a 17 cm full diameter and 7 cm tube diameter. Assume its bulk density and wound length remain constant during withdrawal. At an outside diameter of 12 cm, the fraction of yarn volume left is:

(12² − 7²) ÷ (17² − 7²) = 95 ÷ 240 ≈ 39.58%

At an outside diameter of 8 cm, the fraction is:

(8² − 7²) ÷ (17² − 7²) = 15 ÷ 240 = 6.25%

Those percentages describe this ideal geometry only. A changing density profile or shape would change the mass relation. The calculation still shows why “half the diameter” is a poor substitute for measured yarn remaining.

Choose observation points that include full, intermediate and near-empty states. Retain their measured diameters or residual masses. Do not stop a successful trial while the inner layers remain untested, then describe it as a full-package result.

Also inspect starts and restarts. A package may run steadily but behave differently when withdrawal resumes. Keep restart events separate so that a steady-running average does not hide them.

Relate a late fault to the amount left

Extend the ideal geometry example using a constant 20 tex yarn. Its 1,450 g net mass represents 72,500 m. Suppose a fault begins when 10,000 m remains. That remaining length weighs 200 g, or about 13.79% of the original yarn mass.

Under the same constant-density assumptions, the remaining outside diameter follows from:

Remaining diameter² = 7² + (200 ÷ 1,450) × (17² − 7²)

Remaining diameter ≈ 9.06 cm

This calculation connects a length record with a package position. It does not locate an individual faulty wrap or prove why it failed. An actual tapered profile or changing density would require a better model.

For a normal package withdrawn from its outside, the outer material leaves before the inner material. A disturbance early in package building may therefore appear late in withdrawal. Match the build record to the material's position, not just to the clock time of the downstream fault.

If the package was rewound again, use the record for that later pass. The original layer order no longer describes the package being tested. Also allow for removed yarn and joins when comparing length records. An uncorrected counter offset can send the investigation to the wrong layer.

Compare packages on equal exposure

Suppose a mill tests two package formats on the intended downstream route. Both use the same verified 20 tex yarn. The lots, machine conditions and inspection rules are matched as far as practical. Trial order alternates to reduce the chance that one format always runs during a different period.

This is a separate invented trial. The package labels below do not refer to the earlier density examples. All stated yarn masses are actually withdrawn, with no unmeasured residual yarn or waste. The common count and mass basis are assumed exact for the arithmetic.

Since 20 tex means 20 g per 1,000 m, a 1,200 g yarn mass represents 60,000 m. A 1,600 g mass represents 80,000 m. ApparelDiary's yarn-count guide explains the unit basis.

Trial format

Packages tested

Yarn withdrawn per package

Total length withdrawn

Unplanned stop events

P

8

1,200 g

480,000 m

12

Q

6

1,600 g

480,000 m

8

Each format supplies 9,600 g over 480,000 m. Their observed stop rates are:

P = 12 ÷ 480,000 × 1,000,000 = 25 stops per million metres

Q = 8 ÷ 480,000 × 1,000,000 ≈ 16.67 stops per million metres

Q has 33.33% fewer stops per length in this trial. That is a descriptive result, not a forecast or proof that one winding setting caused it. The formats also differ in package content. The trial compares the complete formats unless a separate experiment isolates one factor.

Classify the events before choosing a format

Suppose P's twelve events include five slipped-wrap events, three caught tails and four events whose causes remain unknown. Q has one slipped-wrap event, one caught tail and six unknown events. The category totals reconcile, but the smaller total for Q does not explain its unknown cases.

Record actual lost time as well. Suppose P loses 38 minutes and Q loses 52 minutes. Over the same 480 km, that is about 7.92 minutes per 100 km for P and 10.83 minutes per 100 km for Q.

Fewer stops have not meant less lost time here. Inspect the long interruptions before selecting Q on event count alone. Planned package changes are excluded from those unplanned-event figures and should have their own time record.

Where possible, use measured delivered length and reconcile it with net mass and count. A nominal package label is not proof of exposure. If yarn remains on a tube, weigh or measure it and exclude it from the withdrawn quantity.

After this screening, repeat promising conditions over a wider range of packages and positions. Preserve results by package so that a few troublesome units are visible. A pooled total can hide them just as a mean tension can hide a short rise.

Give dye packages a separate process check

A winding trial for dyeing must include the dye route. Lower average density does not, by itself, prove even liquor flow or level shade. Package shape, tube suitability and loading on the carrier also belong in that check.

Thies' iCone and carrier-system brochure describes liquor circulation through packages and highlights suitable carriers and flow through each package. This supports judging the loaded arrangement rather than an isolated package alone. It does not provide a universal winding-density target.

In a comparison, identify samples from outer, middle and inner layers and from relevant positions in the load. Keep the agreed colour assessment method and conditioning consistent. A matching outer sample cannot prove that an inner layer received the same treatment.

If package density and the dye cycle both change, treat the result as a combined-route trial. Do not assign a shade improvement or water saving to winding alone. Measure any claimed resource change across the same process boundary and accepted output.

Check withdrawal after dyeing and drying too. The package that enters the vessel and the package supplied to knitting may be in different states. If rewinding is part of the approved route, include its effects and losses. If it is omitted, test the package that will actually reach the customer.

Approve a usable package, not a single number

An approval should identify the yarn, tube, package format and winding recipe. Attach the density method, full-to-empty withdrawal evidence and any dye-route results. Keep actual observations distinct from suspected causes.

ApparelDiary's statistical process control guide can help organise repeated measurements once the measurement basis is stable. Compare like with like: a revised tube, finish or guide path may create a new test condition.

The package is ready when its measured build and its next-process behaviour support the intended use. If those disagree, return to the affected stage and location. A higher density, a lower tension average or a faster winder is useful only when the resulting yarn can complete its route.

References