What Measurement Tolerance Is Normal in Linen Production?
- Share
- Issue Time
- Nov 13,2026
For bulk linen garments, a normal measurement tolerance is ±1.0 cm on main body points such as chest, waist, hip, and body length, ±0.5 cm on fit-critical points such as shoulder, sleeve length, armhole, and neck, and ±1.5 cm on forgiving points such as hem width on flared or oversized shapes. Cuff height and button spacing are usually held to ±0.3 cm. These bands assume the garment is measured flat, closed, and relaxed for at least 24 hours after pressing, seam to seam. Larger tolerances are not a sign of a relaxed factory; they are a sign that nobody defined the measurement method.
The Three Bands, and Why Linen Needs All Three
Shoulder width, sleeve length, armhole, neck width, button spacing. Errors here propagate into construction and cannot be corrected after sewing. A 0.8 cm shoulder deviation is small in absolute terms but throws the armhole and sleeve setting out on every unit cut from the same block.
Chest, waist, hip, body length, pocket placement. These drive how the garment fits and reads on the body. Linen relaxes and recovers, so a band tighter than ±0.8 cm on a woven linen shirt is difficult to hold across a bulk run without producing rejections that do not correspond to any real fit problem.
Hem width on A-line, boxy, or oversized shapes; sweep; some cuff widths on relaxed silhouettes. Applying a fitted-garment band to an oversized garment creates false failures. The band must match the silhouette, not the buyer's habit.
The bands are asymmetric in strictness because percentage deviation matters more than absolute deviation. A 0.5 cm error on a 14 cm cuff is 3.6% and reads as wrong on the body. The same 0.5 cm on a 110 cm body length is under half a percent and reads as normal. Setting one blanket tolerance for the whole sheet is the most common specification error we see from new brands, and it reliably produces disputes at final inspection. If you are at sample stage, the 12-point fit testing checklist shows how to capture the actual drift of your fabric before you commit to bands.
Reference Tolerance Table by Measurement Point
| Measurement point | Band | Class if exceeded |
|---|---|---|
| Chest / bust width | ±1.0 cm | Major |
| Waist width | ±1.0 cm | Major |
| Hip width | ±1.0 cm | Major |
| Body length (HSP to hem) | ±1.0 cm | Major |
| Shoulder width | ±0.5 cm | Critical |
| Sleeve length | ±0.5 cm | Critical |
| Armhole (straight) | ±0.5 cm | Critical |
| Neck width / collar | ±0.5 cm | Critical |
| Pocket placement | ±0.5 cm | Major |
| Button / placket spacing | ±0.3 cm | Critical |
| Cuff height | ±0.3 cm | Minor |
| Hem width (flared / oversized) | ±1.5 cm | Minor |
Bands apply to a garment measured flat, closed, relaxed 24 hours minimum after pressing, seam to seam, rounded to the nearest millimetre. Adjust the numbers to your construction, but keep the three-tier logic: a band without a class cannot be counted in an inspection verdict.
Where the Millimetres Actually Go in Linen
Before you argue about a band, it helps to know which process step produced the deviation, because the remedy is different in each case. In bulk linen production, dimensional drift almost always traces back to one of five places, and four of them are decided before the garment ever reaches the sewing floor.
- Relaxation after cutting. Fabric is wound under tension and recovers once cut. A panel cut at 60 cm can settle to 59.3 cm within 24 hours. Measuring immediately after cutting means measuring a stressed state, not the final state, which is why the relaxation window belongs in the spec.
- First-wash shrinkage. Linen commonly moves 3% to 7% on the first wash depending on weave, yarn count, and finish. A garment-washed order has already absorbed that movement; a piece-dyed order has not. The two cannot share a band.
- Seam take-up. A 1 cm seam allowance consumed slightly unevenly across a long shoulder seam produces a 3 to 5 mm shift at the armhole end. On lightweight linen, seam grin compounds the effect.
- Pressing and steaming. Final pressing can add or remove 5 to 8 mm on body length depending on how much steam and stretch the operator applies. It is the least documented step in the chain and the most variable.
- The measurement method itself. Flat versus hanging, closed versus open, tape tension, and seam line versus seam edge. Two inspectors measuring the same garment can differ by 5 mm purely on method, without either being wrong.
The last point is free to fix and is routinely skipped. A tolerance table without a stated method is only half a specification. Write down flat, closed, relaxed, seam to seam, nearest millimetre, and half of the marginal disputes at final inspection disappear before they can start. The full method by measurement point is set out in measurement tolerance in bulk linen.
How Tolerance Is Set in Practice
What Happens When a Measurement Falls Outside Tolerance
An out-of-band reading is not automatically a rejection. The disposition depends on the class of the point, how many units are affected, and whether the cause is correctable. There are four practical responses, and they should be agreed in writing before the inspection rather than negotiated during it.
| Disposition | When it applies | Commercial effect |
|---|---|---|
| Rework to spec | Seam allowance or hem turn-back permits correction; defect is localised to one point | Cost and time to factory; no price change |
| 100% sort and replace | Defect clusters on one pattern point across the lot | Factory replaces failed units; delivery may slip |
| Concession with written approval | Deviation is minor or invisible in wear; buyer formally accepts it | Price adjustment per unit, agreed in advance |
| Reject units | Critical point failure, or deviation that changes fit or function | Units deducted from invoice; lot re-inspected |
A concession is a commercial decision, never an inspection decision. The inspector records the deviation; the buyer decides whether to accept it. Mixing the two roles is how inspection reports become unenforceable.
Measurement defects cluster by pattern point, while visual defects scatter. If six sampled units fail and four of them fail on chest width, the cause is a pattern block or a cutting marker, not operator variance, and one targeted correction fixes the lot. If the six failures are spread across six different points, you have a systemic control problem instead. Read the pattern of failures before choosing the remedy.
What Counts as Normal, and What Does Not
Conditions That Change the Numbers
A garment-washed order has already paid the shrinkage cost at inspection. A piece-dyed order has not. The two do not share a tolerance band, and comparing them produces false failures.
Measuring within hours of pressing catches the fabric mid-recovery. Readings taken the same day can sit 5 mm away from the settled value. Twenty-four hours is the practical minimum.
Linen yarn count varies along a single roll. Panels from different parts of the roll behave differently under the same pattern. The mitigation is marker planning, not tighter sewing tolerance.
A boxy or oversized shape justifies ±1.5 cm on chest and hem. A fitted shape does not. The band follows the design intent.
What Belongs on the Measurement Page
A tolerance table is usable only if the conditions under which it applies are on the same page. Six items, each one line, and the page becomes an enforceable document rather than a reference drawing.
- Measurement state. Flat, closed, relaxed at least 24 hours after pressing, seam to seam.
- Nominal values with one unit system. Centimetres or inches, not both, and state which side of the seam the tape sits on.
- A band for every point. Not one blanket figure for the whole sheet. A single tolerance across twelve points is the most common specification error we review.
- A class for every point. Critical, major, or minor, so the inspector knows what to count and the buyer knows what remedy applies.
- A rounding rule. Nearest millimetre, round half up. This single line prevents the "it was only 1 mm over" conversation entirely.
- The shrinkage basis. Pre-wash or post-wash nominals, plus the accepted first-wash range. For linen that range is normally 3% to 7%; readings outside it are a fabric conversation, not a sewing conversation.
Two of these six items, the class and the rounding rule, cost nothing to add and resolve the majority of disputes we see. The remaining four require you to have measured the approved sample before writing the sheet, which is the step most often compressed when a launch date is tight. If you would rather have the page reviewed before cutting starts, send it to our team with the target market and wash type.
How Tolerance Feeds the AQL 2.5 Verdict
The two numbers do different jobs and they meet at the inspection sheet. AQL 2.5 supplies a sample size and an acceptance number for the lot; your tolerance table supplies the definition of a defective unit. At a lot size of 501 to 1,200 units, the standard calls for 80 pieces inspected, with 5 defective units accepted and 6 rejected. There is no conditional pass in between.
Run a worked case. A 900-unit linen shirt order, 80 pieces inspected against the table above. Three units fail chest width beyond ±1.0 cm, one fails sleeve length at 0.7 cm short, two fail pocket placement by 0.6 cm. That is six defective units against an acceptance number of five, so the lot is rejected at AQL 2.5 even though the overall defect rate is 7.5% and no single deviation is dramatic. Because four of the six failures sit on chest and sleeve, the cause is almost certainly one pattern block or one cutting marker. A targeted 100% sort on those two points resolves the lot far faster than a full re-inspection.
This is the practical value of separating the two systems. The tolerance table tells you what went wrong and where; the AQL arithmetic tells you whether the lot ships. Buyers who conflate them end up arguing about severity when they should be reading the pattern of failures and issuing a scoped correction.
Limitations Worth Stating Plainly
A tolerance table cannot tell you whether a garment feels right. It tells you whether the garment matches the approved sample within a defined band. Fit judgement happens at sample approval, which is the single most consequential decision in the order. Sampling runs 7 to 12 days, and compressing that window to save a week is how tolerance disputes get created months later.
Tighter is not better. Asking a linen factory to hold ±0.3 cm on a long dimension in a fabric that relaxes and shrinks produces either false rejections or falsified inspection sheets. Match the band to the dimension and to the fibre.
AQL 2.5 is not a tolerance. It is an acceptance number expressed as a percentage of defective units. A ±2.5% dimensional tolerance is a separate commercial decision. The digits coincide and the confusion is routine; the inspection and AQL sampling guide separates the two mechanics in more detail.
Send your tech pack, target market, and whether the order is garment-washed or piece-dyed. Those three details shift the recommended bands materially, and we will tell you which points need tightening and which are already stricter than the fibre can hold.
Send your measurement page Read the size grading guideLinenwind is a Dongguan-based B2B linen garment manufacturer working with fashion brands on OEM, ODM, and private label programmes. MOQ 60 pieces per style per colour, sampling 7 to 12 days, bulk production 25 to 35 days, first order 35 to 60 days. Fabric held in 50+ in-stock specifications at 100 to 260 GSM, inspected to AQL 2.5 under ISO 9001, OEKO-TEX Standard 100, and SEDEX.