The Gold Standard of Yarn Crafts | Reconstructing Starting Length from Stage-Specific Take-Up and Shrinkage

A notebook might contain several lengths for what appears to be the same piece: a starting reference span, a relaxed woven span and a finished span. These numbers can all be accurate while describing different conditions. The first task is to identify which measurement each percentage compares.

For our exercise, call the starting reference length A, the intermediate relaxed span B and the final span C. The fictional record explicitly supplies B/A and C/B. The record also states that the corresponding boundaries refer to the same intended section. Without that continuity, a comparison could mix different pieces or include an extra border.

The physical observations in real weaving can involve interlacement, tension release, material behaviour and finishing. Terminology varies with the recording method. We therefore define the stages in the exercise rather than assuming the word take-up always identifies one universal denominator.

Write the measurement condition beside every number: unit, direction, marked section and stage. A bare note saying ten per cent shrinkage leaves the essential comparison unresolved. A clear record makes the next calculation possible and lets another maker ask a useful question if the conditions do not match. It also prevents an apparently correct equation from comparing lengths that were never intended to be interchangeable.

Contents · Next chapter

Suppose the starting reference span A is 100 centimetres and the corresponding relaxed woven span B is 90 centimetres. The decrease is ten centimetres. Relative to A, that decrease is ten divided by one hundred, or ten per cent.

The retained proportion is B/A = 90/100 = 0.9. Therefore B = 0.9A. Both statements describe the same relationship: ten per cent of the starting span is lost from the measured span, and ninety per cent remains.

Use a strip of paper divided into one hundred equal positions as a simple representation. Mark ninety as the resulting span. The model makes the denominator visible before a learner uses a decimal factor. It does not claim that ten centimetres of yarn vanished; the measured span and the amount of material are different quantities.

That distinction matters in crafts. Yarn can follow a curved path through a structure while the fabric span is measured between endpoints. Relaxation can change dimensions without removing an equal physical length of yarn from the project. The exercise tracks the defined span. It should not be turned into a material-conservation claim about how much yarn was discarded, chemically changed or left on the loom.

Contents · Previous chapter · Next chapter

If B = 0.9A and the desired intermediate span B is ninety centimetres, divide by 0.9 to recover A. The answer is one hundred centimetres. Multiplying ninety by 1.1 gives ninety-nine, which would retain only 89.1 centimetres under the supplied relationship.

The failed shortcut uses a different base. Ten per cent of ninety is nine, while the original ten-centimetre difference was ten per cent of one hundred. An equal percentage decrease and increase do not generally undo one another when the bases differ.

For a ten per cent loss relative to the starting span, the required addition relative to the resulting span is one ninth, or approximately 11.11 per cent. Ninety plus ten equals one hundred, and ten is one ninth of ninety. Both percentage statements can be correct when their denominators are named.

A learner should be able to explain the units as well as the arithmetic. Dividing a length by a dimensionless retention factor produces a length. If the calculation instead mixes centimetres with a percentage entered as ten rather than 0.10, inspect the representation before judging the craft plan. The mathematical repair begins at the first unstable relationship.

Contents · Previous chapter · Next chapter

Now suppose the ninety-centimetre intermediate span retains ninety per cent through the specified finishing stage, giving C = 0.9B. The final span is eighty-one centimetres. The second ten per cent decrease is nine centimetres because its base is ninety, not one hundred.

The overall decrease is nineteen centimetres from the one-hundred-centimetre start. It is nineteen per cent, not twenty. The two stage percentages describe different bases and therefore should not simply be added as if they were both percentages of A.

This is an accounting of supplied dimensions, not a guarantee that two real material processes occur independently. The exercise supplies the two relationships in sequence. An actual project needs observations or reliable instructions that support their use together. If the second factor changes with the first stage’s conditions, the simple fixed-factor model may need revision.

Contents · Previous chapter · Next chapter

To obtain the fictional finished target of eighty-one centimetres, first reverse the second stage: eighty-one divided by 0.9 equals ninety. Then reverse the first: ninety divided by 0.9 equals one hundred. The sequence recovers the original record.

Equivalently, combine the factors: C = 0.9 × 0.9 × A = 0.81A. Therefore A = C/0.81. This compact expression gives the same one hundred centimetres, while the two-step calculation makes the meaning of each factor easier to inspect.

Compare the tempting addition of twenty per cent to eighty-one. It gives 97.2 centimetres. Passing that proposed starting span forward through both stages yields 97.2 × 0.81 = 78.732 centimetres. It fails to reproduce the target.

The forward check is the decisive test. We are not choosing the longer starting value merely to be cautious. We are finding the value that satisfies the supplied relationships. Extra provision for uncertain behaviour is a separate planning decision. Mixing that decision into the exact inverse would make it harder to distinguish the calculated requirement from a deliberately chosen margin.

Contents · Previous chapter · Next chapter

Two notes may both mention ten per cent while defining different relationships. One might say the final span is ten per cent shorter than the starting span. Another might say the starting span includes an addition equal to ten per cent of the final target. Those statements are not identical.

Let t be a loss fraction measured against the start, so final = start × (1 − t). Let a be the added allowance measured against the final target, so start = final × (1 + a). The same pair of lengths requires 1 + a = 1/(1 − t).

Thus a = t/(1 − t), and conversely t = a/(1 + a). A ten per cent loss gives an 11.11 per cent addition relative to the result. A ten per cent addition corresponds to a loss of approximately 9.09 per cent relative to the resulting larger start.

Neither convention is inherently dishonest or wrong. The problem is transferring a number from one convention into a formula using the other. Preserve the source’s definition, or reconstruct it from the stated lengths. If the definition is missing and the difference matters, seek clarification rather than choosing whichever interpretation produces the preferred supply estimate.

Contents · Previous chapter · Next chapter

A yarn can travel over and under other yarns rather than following a perfectly straight line between the ends of a fabric section. A path-length comparison and a span comparison therefore have different measured objects, even when both are recorded in centimetres.

For an original geometry illustration, suppose a particular marked yarn path is 110 centimetres while the corresponding projected fabric span is one hundred. The excess relative to the fabric span is ten per cent. The span’s shortfall relative to the yarn path is ten divided by 110, approximately 9.09 per cent.

Those are two descriptions of the same pair. They are not competing measurements. The denominators differ. Calling both “ten per cent take-up” without a definition would erase that distinction and make later inversion unreliable.

This illustration does not provide a laboratory crimp-testing method or infer a complete yarn path from the fabric’s appearance. Real measurement conditions, including tension and treatment, matter. Its educational purpose is to keep the measured quantities and bases distinct. A maker can then read a supplied method more accurately and recognise when a path-based percentage cannot simply be inserted into a span-based planning formula.

Contents · Previous chapter · Next chapter

A before-and-after comparison is useful only if the lengths refer to the corresponding section. A starting measurement that includes fringe cannot be directly compared with a finished body measurement that excludes fringe without adjusting the boundary.

Suppose an initial record gives 120 centimetres, including twenty centimetres reserved outside the modelled fabric body. The relevant body start is one hundred. Under the two-stage retention of 0.81, its finished body is eighty-one centimetres. Applying 0.81 to all 120 would give 97.2 and mix two different scopes.

The arithmetic has not failed; the input boundary has. Identify which lengths undergo the stated dimensional changes and which are fixed or treated separately. Preserve the meaning of any marked endpoints through each measurement.

The same discipline applies to hems, borders and sample areas. A method might legitimately include them, but the report should say so. A learner who can identify the unmatched boundary has found the practical cause of the discrepancy. Repeating the multiplication more carefully would not repair it. The useful next step is to reconstruct a consistent section ledger before choosing the numerical operation.

Contents · Previous chapter · Next chapter

Continue the fictional eighty-one-centimetre finished-body target. The two retention factors require a one-hundred-centimetre starting body span. Suppose the exercise separately requires twenty centimetres of fixed extra length per warp end, outside those modelled stages.

The total starting length per end is then 100 + 20 = 120 centimetres. The twenty is not another percentage of the finished target, and it should not be passed through the body-retention calculation. It is an independently supplied addition with its own purpose and boundary.

If the project has forty such ends, the simplified total is forty times 120 centimetres, or 4,800 centimetres, equal to forty-eight metres. This result does not include unlisted swatches, fringe rules, broken ends or purchasing units. Do not fill those gaps with a universal percentage.

The existing guide on budgeting usable yarn across colours, swatches and held strands owns that wider material ledger. This article’s contribution is recovering the body span before it enters such a budget. Keeping the two jobs separate prevents an accurate inventory from being supplied with an incorrectly inverted dimension.

Contents · Previous chapter · Next chapter

Length and width need not have the same retention factors. A record measured along the warp direction should not be applied across the fabric merely because both results are called shrinkage. Keep the direction attached to every observation.

For a fictional width model, let the first retained proportion be 0.85 and the second 0.95. Their product is 0.8075. A starting width of thirty centimetres gives an intermediate width of 25.5 and a final width of 24.225 centimetres.

To recover the starting width from 24.225, divide by 0.8075 and obtain thirty. Using the earlier length factor of 0.81 would instead give approximately 29.91. The closeness of those results in this invented case does not make the direction mismatch valid.

An actual design can also constrain the permitted width through end counts, sett, borders and structure. The continuous arithmetic supplies a dimensional relationship; it does not authorise every theoretical width or count in a pattern. Record the relevant construction constraints and use the appropriate technique guidance when moving from a modelled width to a workable project. The calculation is a step in the decision, not the whole design.

Contents · Previous chapter · Next chapter

If the simplified length and width factors both apply to a rectangular body, its area factor is their product. With length retention 0.81 and width retention 0.8075, area retention is 0.654075. The area decrease is therefore 34.5925 per cent of the starting rectangle.

Adding the two linear decreases, nineteen per cent and 19.25 per cent, would give 38.25 per cent. That overstates the area decrease because it counts the overlap between the directional changes twice. Multiplying the retained proportions preserves the relationship between the two dimensions.

A visible check uses a rectangle on squared paper. Reduce one dimension, then reduce the other on the already changed rectangle. The second operation acts on the remaining area rather than the original area alone. The representation makes the cross term easier to understand before writing a formula.

This area calculation assumes the body remains adequately represented by the stated rectangular model and corresponding directional spans. It does not establish uniform local deformation or a fabric’s functional properties. A numerical area result cannot decide whether drape, texture or fit still meets the design purpose. Those observations belong beside the dimensional calculation when the actual object is evaluated.

Contents · Previous chapter · Next chapter

A supplied pattern or sample estimate may already incorporate some dimensional allowance. Before adding another percentage, identify what the existing number represents. Finished size, length to weave and starting warp length are different stages in a planning chain.

Suppose a fictional instruction explicitly states that one hundred centimetres is the calculated starting body span for an eighty-one-centimetre finished body under the two factors. Dividing that one hundred by 0.81 again would apply the same inverse twice and produce approximately 123.46 centimetres for the body alone.

The larger number is not automatically a better plan. It comes from changing the interpretation of the supplied value. A cautious margin should be stated as a separate choice with a reason, not hidden inside a second application of the same transformation.

Write a stage label above each spreadsheet input and output. If two neighbouring cells both say required length, rename them to identify the conditions. This small change can prevent a large downstream discrepancy. A clear calculation allows another maker to see which factors were already included and which additions remain to be considered under the actual project instructions.

Contents · Previous chapter · Next chapter

A sample can supply a relationship between two measured conditions for a particular material, construction and method. Record the starting and resulting spans, their direction and how the corresponding section was identified. Follow the material’s appropriate care and finishing information.

If a fictional sample goes from 125 to 115 centimetres at the first stage, its retained proportion is 0.92 and its decrease is eight per cent of the starting span. If it then goes to 103.5 centimetres, the second retained proportion is 0.9. Overall retention is 0.828.

That record supports the sample’s calculated factors. It does not prove that every larger piece will behave identically or that the factors apply to a replacement yarn. Borders, tension, structure, measurement conditions and finishing differences can matter. Keep extrapolation conditional and test what is consequential for the actual project.

The Weaving guide provides the specialist construction route. The arithmetic here helps interpret a supplied sample record. It does not replace an appropriate weaving method or transform an unidentified textile into a reliable source of measurements simply because a percentage can be calculated from it.

Two samples can also challenge the choice of a proportional model. Suppose a separate fictional record gives a starting span of one hundred and an intermediate span of ninety, but a starting span of two hundred gives 185. The observed retention factors are 0.90 and 0.925. One unchanged multiplier does not reproduce both pairs exactly.

An affine teaching model B = 0.95A − 5 reproduces these two particular observations: ninety at a start of one hundred, and 185 at a start of two hundred. It contains a fixed five-centimetre term as well as a proportional term. Two observations alone do not validate that model or establish the physical cause of either term. They simply show why assuming one constant percentage would be a substantive choice rather than a consequence of the supplied measurements.

To reverse this alternative relationship for an intermediate target B, use A = (B + 5)/0.95. Dividing B by 0.90 would answer a different model. Keep this alternative distinct from the fixed-retention examples elsewhere in the article. A learner should first ask which relationship the record supports, then reverse that relationship. Choosing a formula because it worked in the preceding exercise is not enough.

This is another reason to retain raw measurements. A notebook containing only an average percentage could conceal the disagreement between the two samples. The original pairs let the maker examine proportionality, measurement uncertainty and changed conditions before treating a summary as a dependable project rule.

Contents · Previous chapter · Next chapter

Sometimes a record supplies a range rather than one fixed retention. Let the first stage retain between 0.88 and 0.92, and the second between 0.91 and 0.95. Suppose the exercise permits every combination of those factors and all values are positive.

The smallest combined retention is 0.88 × 0.91 = 0.8008. The largest is 0.92 × 0.95 = 0.874. For a fixed eighty-centimetre final target, the corresponding required starting spans run from 80/0.874, approximately 91.53 centimetres, to 80/0.8008, approximately 99.90 centimetres.

The larger required start belongs to the smaller retention. That inverse direction is an important check. A worksheet that pairs the least shrinkage with the largest required starting span may have reversed the relationship.

The positive-factor condition is essential. If a purely mathematical variant sets one retention factor to zero, the product model predicts a zero final span for every starting value. It cannot produce a positive target by division, and a zero target would not identify one unique start. This boundary is not a proposed physical finishing process. It shows why the inverse requires a nonzero relationship before a calculator is used. Likewise, a missing factor is not the same as a factor of zero. Keep unknown values unresolved until the record supplies a defensible value, range or alternative model that answers the actual dimensional question.

This is a conditional range from supplied bounds, not a confidence interval or a promise that an actual project will fall inside it. If the two factors cannot vary independently across the full rectangle of allowed values, use the compatible combinations instead. The result should preserve the structure of the uncertainty rather than inventing combinations that the evidence does not permit.

Contents · Previous chapter · Next chapter

A wide required-length range should lead to a specific question. Which stage contributes the most unresolved variation? Are the measurements from the same direction and section? Is one factor copied from a different material or finishing condition? A more precise answer may come from improving the relevant evidence rather than adding an arbitrary margin.

For the supplied range, one could hold the second factor fixed while examining the first, then reverse that comparison. This isolates which assumption changes the result. The exercise is a sensitivity analysis, not a claim that the physical processes can always be independently controlled.

Preserve the practical consequence. A small decorative sample may tolerate a size difference that a fitted object cannot. The appropriate next step depends on purpose, consequences and suitable technique guidance. Mathematics identifies the relationship and range; the maker decides what evidence is needed for the actual project.

Do not turn uncertainty into a judgement about ability. A missing condition in a notebook can make an experienced person’s measurement difficult to reuse. A better record can restore clarity. The useful outcome is knowing which observation or definition would narrow the decision, rather than treating every mismatch as a mysterious failure of craft skill.

Contents · Previous chapter · Next chapter

Carry enough precision through the stage calculations to avoid accumulating unnecessary error. Then round the final planning quantity according to the actual decision and its constraints. Intermediate factors and a permitted whole-number count may require different treatment.

For example, a computed body start of 99.9001 centimetres should not be rounded down at every stage before later multiplication. Conversely, displaying many decimal places does not establish that a physical measurement was that precise. The reporting precision should reflect the evidence and purpose.

If a construction permits only certain end counts or repeat sizes, calculate the candidate dimensions from those allowed counts. Do not round a theoretical count casually and assume the structure survives. The Yarn Crafts root already demonstrates the distinction between a width target and permitted repeats; this guide preserves it after the stage inverse has been calculated.

An explanation should say which quantity is theoretical, which is constrained by the design and which is a chosen provision. Keeping those roles visible makes the plan easier to review. It also helps a learner see why two makers can use the same percentage relationship while making different legitimate design decisions under different size tolerances and construction rules.

Contents · Previous chapter · Next chapter

A fictional learner, Tessa, reads that a span loses ten per cent and adds ten per cent to the desired result. Her mistake is not simply insufficient care with multiplication. She has treated percentages with different bases as if they were equal lengths.

Start with the one-hundred-to-ninety paper strip. Ask her to identify the ten-centimetre difference as a fraction of each length. Ten divided by one hundred is ten per cent; ten divided by ninety is approximately 11.11 per cent. Both statements can now be explained visibly.

Then ask Tessa to recover a sixty-three-centimetre result after one stage retaining ninety per cent. She should obtain seventy and check that seventy times 0.9 gives sixty-three. Only after that relationship is stable should the second stage be introduced.

Finish with a record whose two percentages use different written conventions: a loss relative to the start and an addition relative to the target. Tessa must name the denominator before choosing an operation. Independent success means she can explain the changed case without being told to divide twice. That is a stronger sign of learning than completing a page of identically labelled percentage questions.

Contents · Previous chapter · Next chapter

Use a new fictional record with a starting body span of 144 centimetres, an intermediate span of 132 and a final span of 118.8. The measurements concern the same marked body section and direction under the exercise’s declared conditions.

Calculate each stage’s retained proportion and loss percentage, and calculate the overall retained proportion. Then find the starting body span needed for a final target of ninety-nine centimetres under the same fixed relationships. Do not begin by adding the two loss percentages to the final target.

The exercise also supplies fifteen centimetres of fixed extra length per end, outside the body model, and thirty-two ends. Determine the total simplified starting length. State which part underwent the stage inverse and which part was added afterwards.

Finally, compare a note that calls the first stage’s difference a percentage of the intermediate span rather than of the start. Calculate that percentage and explain why it differs from the first-stage loss percentage without indicating a disagreement about the measured lengths. Write a short conclusion identifying the target, assumptions and omitted project requirements. Keep the modelled body, fixed addition and total material quantity separately labelled throughout your working.

Contents · Previous chapter · Next chapter

The first retained proportion is 132/144 = 11/12. Its loss is 1/12, approximately 8.33 per cent of the start. The second retained proportion is 118.8/132 = 0.9. Their product is 0.825, so the overall loss is 17.5 per cent of the starting body span.

Check forward: 120 × 11/12 gives 110; 110 × 0.9 gives ninety-nine. The inverse therefore reproduces the target under the supplied model. The fixed fifteen centimetres never entered that body-retention chain.

For the alternative percentage base, the first difference is twelve centimetres divided by 132, or 1/11, approximately 9.09 per cent of the intermediate span. The loss percentage relative to the start was approximately 8.33 per cent. Both describe the same twelve-centimetre difference. The denominator, rather than a contradictory measurement, explains the change in percentage.

Contents · Previous chapter · Next chapter

A useful record begins with project identity, yarn identity, construction, direction, corresponding section and measurement condition. Add the raw lengths before storing only percentages. Another reader can then reconstruct the chosen convention instead of guessing what a ten per cent note meant.

Separate observed sample factors from assumed project factors. If you choose to reuse a sample relationship for a larger object, state that assumption. If a pattern supplies an allowance on another base, retain its definition and show any conversion explicitly.

List fixed additions separately from the modelled body span. Record any requirements not yet included, such as a retained sample or a construction-specific finishing provision. The goal is not an elaborate form for its own sake; it is preventing one ambiguous number from silently doing several jobs.

For review, ask another learner to explain the plan using only the record. Can they identify which lengths were measured, which were inferred and which were chosen? If they apply a percentage to the fixed addition or cannot tell whether a factor is already included, revise the labels. A plan becomes dependable when its reasoning can travel with the numbers rather than remaining only in the original maker’s memory.

Contents · Previous chapter · Next chapter

Is adding ten per cent always wrong?

No. It can be the correct stated allowance when the addition is explicitly measured against the target. It is not the exact inverse of a ten per cent loss measured against the start. Name the base.

Can I combine two stage percentages?

Combine their retained proportions when the supplied sequential model applies. Two ten per cent losses retain 0.9 × 0.9, or 0.81. Adding the loss percentages ignores their changing bases.

Do these factors apply to every cotton or wool project?

No. All numerical factors here are fictional. Use the actual material, structure, measurement conditions, care information and appropriate project guidance. A fibre label alone does not supply a universal factor.

Does the shorter fabric contain less yarn by the same amount?

Not necessarily. Fabric span and yarn path are different quantities. The stage model tracks corresponding measured spans; it does not by itself establish how much material was removed or consumed.

Why retain the raw sample lengths?

They preserve the denominator and let another maker recalculate or compare conventions. A percentage without its stages, direction and conditions can be difficult to interpret accurately later.

Contents · Previous chapter · Next chapter

Use Yarn Crafts for purpose, material and sample decisions, and Weaving for the construction route. For the material ledger after dimensional planning, follow usable yarn across colours, swatches and held strands.

For different pattern questions, reconstructing weave interlacements explains threading and lift plans. Translating colour charts explains viewing side, working direction and repeats. Neither grid operation supplies a dimensional retention factor by itself.

The primary craft-context routes are Gist Yarn’s take-up and shrinkage explanation and Creative Fibre’s project-planning guidance, checked on 22 October 2026. They support careful sampling and the distinction between project quantities. This article’s stage definitions, dimensions, ranges and exercises are original teaching assumptions, not universal allowances copied from those sources.

Continue into learning that transfers into life under How X Works. Keep each percentage attached to its base, reverse the appropriate relationship, and pass the proposed start forward to see whether it reproduces the intended result.

Contents · Previous chapter · Return to Yarn Crafts