Casual Chic Linenwear Custom Manufacturing
What Shrinkage Test Should I Require on a Linen Sample?
Sample Testing
What Shrinkage Test Should I Require on a Linen Sample?
Require a dimensional change test run on washed specimens, measured to a named standard: ISO 6330 for the wash and dry procedure with results measured to ISO 5077, or AATCC 135 for fabrics and AATCC 150 for garments if your market works in AATCC methods. Specify three specimens minimum, 500 mm gauge marks, the exact water temperature, detergent, and drying method, and a pass band per direction. For pre-shrunk 100% linen, a commercial band of ±3% after one 40°C wash is a realistic requirement; for an unshrunk or garment-washed program the band is usually wider, in the 3-7% range, and should be agreed before the sample is cut.
The test itself takes a day. What takes longer is deciding what number you will accept, and writing it somewhere the factory can be held to.
The Test You Should Name in Your Purchase Order
Shrinkage is not a single test. It is a family of procedures, and two labs can produce different numbers on the same fabric if they use different standards, temperatures, or drying methods. That is why "please test shrinkage" is not a specification. The clause has to name a method, a procedure, and a drying condition.
Which method you name should follow the route the garment is on. Pre-shrunk fabric, wash before cut, and garment wash each produce a different set of expected figures, and the route decision itself is covered in our guide to controlling linen shrinkage before cutting.
Four standards cover most commercial linen programs. ISO 6330 defines the washing and drying procedures, and ISO 5077 defines how dimensional change is measured after those procedures. The AATCC pair splits by product type: AATCC 135 for fabric, AATCC 150 for garments. If you are buying for the European market, ISO is the usual reference; for the US market, AATCC is more commonly cited by retail compliance teams.
| Standard | What it governs | When to require it |
|---|---|---|
| ISO 6330 | Domestic washing and drying procedures for textile testing | Base procedure for EU-bound programs |
| ISO 5077 | Determination of dimensional change after washing and drying | Measurement method paired with ISO 6330 |
| ISO 3759 | Preparation, marking and measuring of specimens and garments | When you need the marking method defined too |
| AATCC 135 | Dimensional change of woven and knitted fabrics after home laundering | US market, fabric-stage approval |
| AATCC 150 | Dimensional change of garments after home laundering | US market, finished-garment approval |
Name one fabric method and, where relevant, one garment method. Requiring both is common on washed programs, because a fabric that passes at panel stage can still fail once seams and trims are added.
Two procedure details carry more weight than buyers expect. The first is water temperature: 40°C is the usual commercial reference for linen apparel, and a 30°C result is not interchangeable with it. The second is drying. Tumble drying produces more shrinkage than line drying, and a factory that tests by line drying while the care label instructs tumble drying has tested the wrong condition. Both details belong in the clause, not in a side conversation.
How to Run the Test on a Linen Sample
The procedure is simple enough that most factories can run it in-house, which is exactly why you should specify it precisely. A lab report is only as useful as the conditions recorded on it.
1
Take the specimen from the bulk lot, not the development sample. Cut three specimens at least 500 mm apart from the roll that will actually be cut. Record the lot number on the test record. Development-sample fabric and bulk fabric are frequently different lots and can behave differently in water.
2
Condition and mark. Condition the specimens flat for at least four hours, then mark a 500 mm gauge length in both warp and weft with a permanent marker or stitching. Measure to the nearest 1 mm. Marking accuracy limits the whole test: a 5 mm error on a 500 mm gauge is a 1% error in the result.
3
Wash to the named procedure. Run the ISO 6330 or AATCC 135 cycle at the specified temperature with the specified reference detergent and the specified load. A single specimen washed alone behaves differently from a specimen washed inside a normal load, so the ballast matters.
4
Dry by the method written in the clause. Tumble dry or line dry, and if tumble, state the setting. Do not let the lab choose, because labs choose the faster option and the faster option usually shrinks more.
5
Re-condition, re-measure, and calculate. Lay the specimen flat until it reaches equilibrium, then re-measure both gauge directions. Calculate change as final minus original, divided by original, expressed as a percentage. A negative figure is shrinkage.
6
Report per specimen, not only the average. Three specimens that read 2.1%, 2.4% and 2.2% describe a stable lot. Three that read 1.2%, 2.8% and 4.9% describe a consistency problem that an average of 3.0% would hide completely.
For a washed program, repeat the measurement on a sewn sample as well. Panels tested before sewing tell you what the fabric does; a finished garment tells you what the customer receives, including the effect of seams, interlining, and trims. Our guide to linen fabric testing sets this alongside colorfastness, pilling, and strength testing, which are usually run in the same lab session.
Reading the Numbers: What Counts as a Pass
A percentage on its own tells you nothing about whether to accept the sample. The number has to be read against the route the garment is on. Unshrunk linen in the 3-7% band is normal behavior, not a defect. The same 5% figure on a program where you specified pre-shrunk fabric is a failure of the specification, not of the garment.
| Program type | Typical measured change | Reasonable requirement | What a result outside it means |
|---|---|---|---|
| Pre-shrunk fabric, structured style | 1-3% | ±3% per direction | Fabric did not receive the finishing route quoted |
| Unshrunk 100% linen | 3-7% | Accept and allow in pattern | Outside 7% suggests an unstable lot |
| Wash before cut | 2-4% after wash | Consumption-based, not pattern-based | Check marker and width loss, not just the percentage |
| Garment wash, finished piece | 3-6% | ±5%, with size grading re-based | Grading drift; check largest and smallest sizes |
| Linen blend with 30%+ synthetic | 1-4% | ±3% per direction | Blend ratio or finishing may differ from approved |
Bands are commercial starting points for 100% linen apparel, not universal limits. Confirm the band against a tested specimen before writing it into a contract.
There is a second number worth reading: the spread. A lot that averages 3% with a range from 1% to 5% will produce garments that measure inconsistently across a size run even if the average sits inside tolerance. When the spread exceeds roughly two percentage points, ask for a second test from a different part of the roll before approving bulk. This is the single most useful check a buyer can add to a standard report, and it costs nothing extra because the specimens already exist.
Directionality matters as much as magnitude. Linen typically moves more in warp than weft, or the reverse depending on how it was woven and finished, and a garment cut on the bias can behave differently again. A report that gives only one combined figure is incomplete. Require warp and weft separately, and for bias-cut styles ask for a note on how the panels were oriented. Three things are worth checking on every report you receive:
- Both directions reported. A single combined percentage hides the difference between a garment that shortens and one that narrows.
- Per-specimen figures, not just the mean. The spread is what predicts consistency across a size run.
- The lot number. Without it, you cannot tell whether the result describes the fabric you will actually cut.
Our answer on how much linen shrinks and whether it can be prevented covers the fiber behavior behind these figures.
Match the test to the care label. If the label says machine wash warm and tumble dry low, the test must use those conditions. A sample tested at 30°C and line dried, then labeled for 40°C and tumble drying, has been approved under conditions the customer will never reproduce.
What Moves the Result, and What Does Not
Buyers often treat a shrinkage figure as a property of the fabric. It is closer to a property of the fabric plus the test conditions. Knowing which variables move the number prevents arguments about results that are both correct.
Water temperature
Higher temperature relaxes the fiber more. A 40°C cycle will usually produce a larger change than 30°C on the same cloth, and the gap widens on unshrunk fabric.
Drying method
Tumble drying with mechanical action removes more length than line drying. This is the variable most often left unspecified, and the one that changes results most.
Detergent
Reference detergent is specified in the standard for a reason. A laboratory using a different detergent is running a different test, even at the same temperature.
Number of cycles
Most of the movement happens in the first wash. A three-cycle test tells you about stability; a one-cycle test tells you about first-wash behavior. State which one you want.
Fabric construction
GSM, yarn count, and weave density all affect how much the cloth can move. Lightweight linen in the 100-160 GSM range generally moves more readily than 200-260 GSM cloth.
Finishing route
Compacted, washed, and steam-relaxed fabric behave differently. This is the variable the factory controls, and the one your specification should pin down.
What does not move the result, despite common belief, is the color of the fabric or the season of production. Shade and shrinkage are separate variables. A buyer who rejects a lot on shrinkage because the shade was also off is making two claims, and they should be recorded as two claims, because the remedies differ.
What a Sample Test Proves, and What It Does Not
A shrinkage test on the right sample lot proves that the fabric, washed by that method, moves by that amount. That is genuinely useful, and it is enough to set a pattern allowance and a tolerance.
It does not prove that the bulk will match. Three specimens from one roll describe one roll. A ten-roll bulk order may include fabric from more than one loom or finishing pass, and the second lot can behave differently from the first. This is why a lot-level check before cutting is standard practice on repeat programs: the cost is a day and a few meters of fabric, and the alternative is discovering the difference after 600 garments have been sewn.
It also does not transfer automatically from fabric to garment. Seams restrict movement in some directions and concentrate it in others. Fusible interlining can pull a front panel out of shape. Elastic and ribbing recover differently from woven linen. A fabric result of 3% can become a garment result of 4.5% once construction is involved, which is why garment-wash programs should always be validated on a sewn sample rather than a flat panel.
Finally, a sample test does not survive a change of route. If you approve a pre-shrunk fabric and later switch the style to garment wash for hand-feel, the earlier report is obsolete. The wash is a new process and needs its own validation. Buyers who treat a shrinkage report as a permanent property of a style rather than a property of a route are the ones who get surprised on the second order.
Checklist: Writing the Shrinkage Clause
A clause that survives contact with production contains seven elements. Each one closes a specific gap that otherwise turns into a dispute.
Standard named. ISO 6330 with ISO 5077, or AATCC 135 / 150. Not "standard test".
Specimen count and gauge. Three specimens minimum, 500 mm marks, measured to 1 mm.
Wash conditions. Temperature, detergent, load, and number of cycles written out.
Drying condition. Tumble or line, with the setting. Matched to the care label.
Pass band per direction. Separate warp and weft limits, not one combined figure.
Sampling point. Bulk lot fabric and a sewn garment where the route includes garment wash.
Report contents. Lot number, per-specimen results, average, and a pass or fail statement signed by the tester.
Those seven lines fit into a short paragraph in the technical pack, and they belong there rather than in a separate quality agreement, because the tech pack is the document the sample room actually reads. Our guide to tech packs for linen clothing shows the full structure and where the test clause sits relative to the measurement chart.
Do not accept a verbal result. "It shrinks about 3%" is not a test report. Without the lot number, the method, and the per-specimen figures, you cannot tell whether the number came from a 500 mm gauge at 40°C with tumble drying or from a rough observation at the sample table. When a claim is later disputed, the verbal figure is unenforceable.
If the Sample Fails
A failed result is not automatically a failed supplier. The first question is whether the sample came from the bulk lot and whether the method matched the clause. A result that fails because the lab tested at 60°C when the clause said 40°C is a testing error, and it should be re-run before anyone discusses remedies.
If the method was correct and the result is outside band, the next step depends on how far outside. A marginal fail, roughly within one percentage point of the limit, is usually handled by adjusting the pattern allowance and proceeding, with the test repeated on the first bulk cut. A clear fail, two points or more outside, points to the fabric or the finishing route, and the sensible response is to re-test a second sample from a different part of the roll before changing anything.
- Within one point of the limit: adjust the allowance, record the decision, and re-test on the first bulk cut.
- Two points or more outside: re-test from a different part of the roll before changing the fabric or the pattern.
- Confirmed fail on re-test: the lot does not match the specification; the discussion moves to replacement or re-finishing, not to a concession.
If the re-test confirms the first result, the fabric does not match the specification and the conversation moves to replacement or re-finishing.
Inspection and measurement are related but separate checks. Shrinkage is verified by the wash test described here; visual consequences such as skewed side seams, puckered seams, and distorted panels are caught at AQL inspection, which at Linenwind runs at AQL 2.5 with measurement results recorded separately from visual defects.
Test turnaroundOne to two days in-house including wash, dry, and re-conditioning
Specimens neededThree minimum; five if you want a usable spread figure
Fabric requiredRoughly 1.5-2 meters of bulk-lot cloth for fabric and garment checks
Where it sitsAfter pre-production sample approval, before bulk cutting is released
For context on what the sample itself costs and how long it takes to produce, see our answers on sample lead time and sampling cost and refundability. At Linenwind, sampling runs 7-12 days with a $40-150 fee that is fully refundable against bulk, and the shrinkage check sits inside that window rather than after it.
Want your shrinkage clause reviewed before the sample is cut?
Send your technical pack and target measurements. We will confirm the test method, the pass band, and the allowance the pattern needs, then quote against it.
MOQ 60 pieces per style and color · Sampling 7-12 days · Bulk 25-35 days · Reply within 48 hours
Linenwind is a Dongguan-based B2B linen garment factory with more than 20 years of OEM, ODM, and private label production for European and North American brands. Facilities are ISO 9001, OEKO-TEX Standard 100, and SEDEX audited; inspection runs at AQL 2.5; stock fabric covers 50+ specifications in the 100-260 GSM range. Shrinkage figures quoted here describe typical 100% linen behavior and should be confirmed on a tested specimen for each program. Related reading in the same series covers pattern allowance, cost impact, and where the shrinkage check sits in the production calendar.
