Mill-Side Brand-Buyer Ribbon Slit-Width Tolerance & Selvedge-Weave Stability Interlock Architecture — Edge Parallelism, Bow Symmetry and Repeat-Order Width Drift 2026
A mill-side brand-buyer guide to ribbon slit-width tolerance and selvedge-weave stability. Covers why a nominal 10 mm ribbon measures 9.4 mm at the selvedge and 10.6 mm at the centre, how loom-width tolerance and slittability interact with bow loop symmetry, the width-versus-weave-construction trade-off, how to set a measurement protocol that survives a change of roll width, and why width drift shows up first in pre-tied bows rather than on the roll.
The width you order is not the width you receive
Almost every brand buyer working with ribbon has had the same experience. The specification says 10 mm. The supplier confirms 10 mm. The roll arrives and the tape measure says 10 mm in the middle and something else — noticeably less — about four millimetres in from each edge. Nothing is wrong with the roll in any legal sense. Nothing is wrong with the mill either. What has happened is that a single nominal number has been asked to stand in for a physical quantity that varies across the width of the material and from roll to roll within a programme.
This matters far more for ribbon than it does for most textiles, because ribbon is not usually consumed as a length of material. It is consumed as a shape. A ribbon becomes a bow, and a bow becomes a product that a customer sees from a distance and compares against a photograph. Width error that would be irrelevant in a metre of lining fabric becomes the single most visible attribute of a gift bow. The buyer who specifies width as one number without a measurement protocol is buying a shape whose dimensions are unknown.
This guide sets out how slit width and selvedge weave stability actually interact, how the two contribute to bow symmetry defects, and how to write a width specification that a mill and a QA laboratory can both execute without ambiguity.
Why width varies across the width of the ribbon
Width variation in ribbon has three separate physical sources, and they need to be understood separately because they have different remedies and different tolerances.
1. Loom-width tolerance. The ribbon is woven on a loom at a nominal reed width. The reed does not deliver an absolutely uniform spacing of warp ends across its full width, because warp tension is not perfectly uniform, because the reed dents flex slightly under load, and because the fabric is under tension as it leaves the reed and relaxes as it moves through the rest of the loom. The practical result is that the fabric as woven has a width that is close to nominal at the centre and marginally narrower toward each selvedge. This is the dominant source of width error in narrow woven ribbon, and it is why an 8 mm ribbon typically behaves differently from a 38 mm ribbon: the absolute tolerance in millimetres that the mill can hold does not scale linearly with the nominal width.
2. Slitting tolerance. A wide ribbon fabric is produced and then slit into narrower widths for the slitted product. The slitting knife has to be set to the required width, and the setting is a physical position that is repeated on a rotating or traversing cutter. Setting accuracy, blade deflection under load and roll-mounting concentricity all contribute. Slitting is the source of the roll-level width error: a roll that is uniformly 9.6 mm when it should be 10.0 mm.
3. Relaxation and finishing shrinkage. Woven ribbon relaxes when it comes off the loom and again when it is heat-set or coated. If the width specification is written against the loom state rather than the finished state, the finished product will not meet it. This is the most common technical dispute in ribbon procurement and it is entirely avoidable by stating which state is being measured.
The important consequence is that a buyer who measures only at the centre of the ribbon will systematically over-read the width and will miss the edge behaviour that determines whether the bow hangs correctly.
The width-versus-weave-construction trade-off
Narrow widths impose real technical constraints that buyers rarely appreciate when they write the specification. A 6 mm ribbon and a 38 mm ribbon are not the same product with different numbers on the label.
- Selvedge structure. Below roughly 10 mm, a plain-weave construction often has no room for a dedicated selvedge, and the edge is simply the termination of the warp and weft. That edge will fray under handling, and it will not support a wired edge or a turned hem. Above 15 mm, a mill can run a woven selvedge or a reinforced edge treatment that changes the character of the material entirely.
- Pick density per unit width. To hold a consistent weight per metre across a range of widths, the pick density per centimetre must increase as the width decreases. This raises tension, raises the risk of a snagging weft, and can make the ribbon noticeably stiffer or more board-like at narrow widths.
- Dimensional stability. A narrow ribbon has less transverse resistance to distortion. Once a bow is tied, or a ribbon is folded, the narrow width more readily takes a permanent set.
- Print registration tolerance. On very narrow widths, a printed repeat cannot be centred with the same absolute precision, and edge waste becomes proportionally larger. Buyers specifying narrow printed ribbon should expect the mill to quote a wider registration tolerance than for a 25 mm product.
The point is not that narrow ribbon is problematic. It is that a width specification which ignores construction is not a specification the mill can quote accurately, and a mill forced to quote it will either absorb a cost it should have flagged or guess and get it wrong.
Selvedge weave stability and edge parallelism
Selvedge weave stability describes whether the edge holds its geometry under the stresses the ribbon will actually meet: winding tension during slitting and rewinding, tension during conversion, tension during tying into a bow, and handling friction during packing and shipping.
Three measurable properties define it.
Edge parallelism is the deviation of the edge from a straight line across the width of the roll. It is measured by taking the roll and checking whether the edge is parallel to the roll axis over a stated length. A ribbon with a non-parallel edge will produce a bow whose two loops are different lengths even when the ribbon itself is nominally the same width throughout — because a bow is made from a finite segment of material and the width available for each loop varies along that segment.
Edge straightness over the roll length is related but separate. A ribbon can be parallel and still wander, or straight and not parallel. Both cause loop asymmetry, and both cause a visible difference in how the ribbon lies in a folded stack.
Selvedge fray resistance is the ability of the edge to survive handling without threads pulling out. This is where construction matters most: a plain selvedge on a narrow ribbon will fray under the tension of a pre-tied bow assembly line, and the fray will be visible on the finished product. A mill that has tested its edge under bow-assembly tension will be able to tell you how it performs; one that has only tested it under a static pull test has not answered the question you are asking.
The buyer-facing implication is that the width specification and the edge specification should be written together and should be tested together. A ribbon that measures perfectly at the centre but frays at the edge will produce a defective bow that no incoming inspection based on centre measurement will catch.
How width defects become bow defects
Translating a mill measurement into a product outcome is what turns a tolerance into a specification. For a pre-tied bow made from a fixed length of ribbon, the relationship between width error and the finished bow is direct.
Loop symmetry. A bow's visual balance depends on the two loops being the same length and the same width. If the ribbon width varies along the segment used for the bow, or if the segment spans a region where the edge is non-parallel, the two loops will differ. A buyer who has received returns described as "off-centre bows" should look first at width tolerance and edge parallelism, not at the bow-making labour.
Loop volume and drape. Ribbon width controls how much material is available to form a loop. At a constant length, a narrower ribbon produces a flatter, tighter loop; a wider ribbon produces a fuller loop with more body. Where a product photography shot or a previous product line defines the intended look, the width tolerance is doing aesthetic work and should be specified more tightly than a purely functional tolerance would require.
Height consistency. If a bow is assembled by folding the ribbon at a fixed distance from the cut end, and the ribbon is wider than the fold allows, the fold encroaches on the loop. If the ribbon is narrower than the fold pattern expects, the bow looks under-filled. The assembly tolerance and the width tolerance are therefore coupled, and the tighter of the two governs the achievable product consistency.
None of this is a reason to specify an impossibly tight tolerance. It is a reason to specify width at the level that the end product actually needs, and to know which end-product attribute each increment of tolerance buys.
A width measurement protocol that survives reorders
The single most valuable thing a brand buyer can put in a ribbon specification is a statement of exactly where and how the width is measured. The following protocol has the advantage that it can be executed identically by the mill, by an incoming inspection laboratory and by a third-party inspector, and it survives a change of roll width without needing to be rewritten.
- State the measurement position. Measure at the centre, and separately at 4 mm in from each selvedge. Report all three. For ribbons narrower than 12 mm, use 2 mm in from each edge instead, so that the measurement point is not inside the selvedge structure itself.
- State the state of the material. Measure on the finished, conditioned ribbon — after heat setting and after any reasonable period of conditioning in the standard atmosphere — not on the loom or immediately off the slitter.
- State the conditioning condition. Temperature and relative humidity for the conditioning period, and the conditioning period itself. Width is a physical measurement and it is not independent of the atmosphere.
- State the sampling rule. How many points along the roll, measured from which end, and whether the roll is measured before or after rewinding onto the shipping core.
- State the instrument. A calibrated micrometer or a digital caliper of stated resolution, applied without compression — a micrometer's anvil compresses soft ribbon and a caliper's jaws do the same, which is why ribbon width measurements should specify the closure force or, more practically, the measurement method that the mill and the inspector both already use consistently.
- State the acceptance rule. Whether the requirement is on the mean width, on every individual measurement, or on a stated proportion of measurements within tolerance. This is the point at which most ribbon disputes actually originate.
Two further details save arguments later. First, specify the tolerance relative where the width is narrow and absolute where it is wide: a ±0.3 mm tolerance on a 6 mm ribbon is a five percent variation, while the same tolerance on a 38 mm ribbon is under one percent and is tighter than most mills can hold consistently. Second, always state the tolerance as it applies to the width of the supplied product, not to the width of the loom product before slitting.
Managing width drift across a reorder programme
Width drift across reorder cycles has the same root cause as colour drift, expressed differently: the specification described a target rather than a controlled process. Four practices prevent it.
Lock the slittability envelope, not just the target width. The mill's ability to slit a given fabric to a given width depends on the fabric. The buyer should specify the fabric construction — yarn counts, weave, pick density, finish — because it is the construction that constrains the width tolerance. A mill that changes yarn count to manage cost will change the width behaviour, and the buyer should know that this is a specification-controlled event, not a purchasing decision.
Retain the production roll reference. Where a programme has an approved width at a specific roll construction, that construction should be named in the purchase order so that a reorder is checked against it rather than against a bare number.
Require a width profile, not a width number. The three-point measurement protocol should be attached to the first production lot approval, so that the actual profile of the material is on record rather than inferred from a single centre measurement.
Treat a loom-width change as a qualification event. If the programme moves to a different loom width or a different weaving programme, the slittability of the new fabric must be re-established before it is relied upon for the established widths. This is a short test and it prevents a systematic narrowing that would otherwise appear gradually across a year of orders.
Buyers who adopt this discipline find that width complaints largely stop, not because the tolerance was tightened but because the tolerance finally meant one thing to both parties.
Frequently Asked Questions
FAQ 1: What is the standard width tolerance for ribbon?
There is no universal figure, and any mill that quotes one without asking about the width, the construction and the end use is guessing. In practice, a narrow plain-weave ribbon below 10 mm behaves in the region of plus or minus 0.2 to 0.3 mm at the centre, while a wider satin or grosgrain above 25 mm may be held to plus or minus 0.5 mm. The more useful approach is to specify relative tolerance for narrow widths and absolute tolerance for wide ones, and to state where on the ribbon the measurement is taken. If you need the edge value as well as the centre value, say so explicitly, because the centre reading alone will not predict bow loop symmetry.
FAQ 2: Should I measure ribbon width at the centre or at the edge?
Measure at both and record three values: the centre and two points set in from each selvedge. The centre reading tells you whether the roll is the right width. The edge readings tell you whether the material is stable across its width, which is what determines whether a cut segment will produce two identical bow loops. A ribbon can pass a centre-only inspection and still produce asymmetric bows, because the inspection never looked at the property that causes the defect.
FAQ 3: How does ribbon width affect pre-tied bow quality?
Width affects bow quality through three channels: loop symmetry, where a width that varies along the segment makes one loop fuller than the other; loop volume and drape, where a narrower ribbon gives a flatter, tighter loop and a wider one gives a fuller loop with more body; and height consistency, where a width that does not match the fold pattern produces either an encroaching fold or an under-filled bow. If a buyer is receiving returns for off-centre or uneven bows, width tolerance and edge parallelism should be investigated before the assembly operation is blamed.
FAQ 4: Can you slit ribbon narrower than the loom width?
Yes, and every narrow ribbon in production is slit from a wider fabric. The limit is not the loom but the fabric. Slitting tolerance depends on how well the woven construction supports the cut, and on whether the selvedge has enough structure to survive it. Below roughly 10 mm, a plain-weave construction often has no true selvedge, so the cut edge is an unfinished edge and will fray under tension. A mill should tell you whether your target width can be cut from your chosen construction without edge treatment, and what the resulting edge will survive.
FAQ 5: What happens to width when the mill changes loom width or yarn count?
It moves, usually gradually, and it usually moves in the direction of the new construction rather than randomly. Changing yarn count or pick density to manage cost changes the transverse behaviour of the fabric, which changes both the loom-width tolerance and the slittability. This should be treated as a specification-controlled event: the buyer approves the new construction, the mill re-establishes the width profile against the three-point protocol, and the first production lot is released only after the profile has been confirmed. A qualification test is inexpensive and prevents a year of drift.
FAQ 6: Why does ribbon width vary between rolls within the same order?
Within-roll and between-roll variation come from different sources. Between-roll variation is typically the slitter setting and the roll-mounting concentricity, and it is the larger of the two for a slitted product. Within-roll variation comes from the loom-width behaviour and from relaxation, and it tends to be larger at the edge than at the centre. Because both are real and both are controllable to a degree, the specification should carry a sampling rule that states how many measurement points along the roll and how many rolls from the order are checked. A single measurement per roll cannot detect either failure mode.