A bouclé is set by its loops and weighed by its count

A bouclé yarn is made by feeding an effect thread faster than the core it is wrapped on, and its surplus goes into loops that stand off the core and carry no load. That account of the yarn ended on what it had not computed: nothing in it was about the cloth. A bouclé weft, it said, behaves as a much thicker yarn than its count and sets far more openly than the count allows.

How much more openly is a question with an exact answer, and the answer is not a number but a range. A bouclé has two diameters. One is the diameter its count implies — the circle its fibres would fill at an ordinary packing — and for an 89 tex bouclé that is 0.35 millimetres. The other is the outline its loops occupy, the core plus a loop on either side, and for the same yarn at an eighty per cent overfeed with loops every three millimetres that is 4.43 millimetres.

A cloth jams where its yarns touch. A bouclé’s yarns can touch at their loops, or they can press their loops aside and touch much closer in, and every cloth that could be woven lies between those two. The range is twelve and a half times wide. And the cloth’s weight does not care where in that range it is set: it goes with the count, so the open end of the range weighs a twelfth of the close end.

A bouclé has two diameters an order of magnitude apart #

An ordinary yarn’s count and its diameter move together. A count is a mass per length, a diameter follows from the mass at the packing a twisted yarn reaches, and the count systems are usable as a measure of thickness precisely because that packing varies over a narrow range.

A folded yarn already strains the relation. Two singles twisted together occupy the smallest circle holding two touching circles, while their mass implies a circle √2 times smaller, so a two-fold yarn has a diameter bracket forty-one per cent wide, and every cover factor computed from it inherits the choice.

A loop yarn breaks the relation deliberately. Its loops are sized by the overfeed and the spacing: a semicircular loop of radius r at one loop every s millimetres consumes r(π − 2) of surplus, so r is the overfeed times the spacing over π − 2. At an overfeed of 0.8 and a spacing of three millimetres the loops stand 2.10 millimetres off a core whose own diameter is 0.22. The envelope is 4.43 millimetres and the count’s diameter is 0.352 — a ratio of 12.6. The semicircle is the simplest loop and not the real one: solved as the elastica a loop held along its core is, this loop stands 1.95 millimetres rather than 2.10 and the ratio is 11.7. Everything below holds at either number, and the places where the difference grows large are marked.

A cloth jams where its yarns touch, and a bouclé can touch twice #

A plain weave jams when each end, together with the pick passing between it and the next, fills its share of the width: two diameters for every end. So the closest a plain weave can be set is one end in every two diameters, and the diameter in that rule is whichever one touches.

If the loops meet and do not give way, the diameter is the envelope, and the 89 tex bouclé jams at 1.13 ends a centimetre. If the loops are pressed aside until the yarns meet at the count’s diameter, it jams at 14.2. A loop yarn in a cloth is always somewhere between: its loops interleave with their neighbours’ to a degree the finishing, the beat-up and the loop’s own stiffness decide, and nothing about the count or the envelope alone says how far.

Wherever it jams, the cloth weighs its count #

Whatever the sett, a centimetre of width carries that many ends, and each weighs its count. A plain cloth’s weight is therefore twice the sett times the count, crimp aside, and the loops enter it only through the sett they allow.

So at the open end of the bracket the cloth weighs 20 grams a square metre, and at the close end 252. The ratio is exactly the count’s diameter over the envelope, because the sett goes as one over the diameter and the weight as the sett. The cloth jammed at its loops covers its face with loops and puts four per cent of its area under yarn by count; the cloth jammed at its count puts fifty.

That is the bouclé’s property stated as a cloth rather than a yarn: an outline set by one number and a mass set by another. A bouclé cloth near the open end of its bracket is almost all air by weight and entirely loops to the eye, which is why such cloths are bulky, soft and light, and why they are warm for what they weigh — a thickness of trapped air that the count does not pay for.

A folded yarn’s bracket is √2, a bouclé’s is twelve #

The comparison with a folded yarn is the useful scale. A two-fold’s two diameters are a fixed √2 apart whatever its count, because the envelope of two touching circles is always twice a single’s and the mass always √2. The resulting uncertainty in a jamming sett is forty-one per cent — already enough to matter, since a warp jams where its threads are thickest and a diameter is what decides where that is.

A bouclé’s bracket is not fixed. It is set by how far the loops stand off the core compared with how thick the count is, and at an eighty per cent overfeed on three-millimetre loops that comparison is 12.6. The same jamming arithmetic that a folded yarn’s √2 disturbs is, for a bouclé, not a correction but a choice of which cloth to make.

The bracket widens with the overfeed and the spacing #

The envelope is the core plus two loop radii, and a loop’s radius is the overfeed times the spacing over π − 2, so the envelope grows in a straight line with each. The count’s diameter grows too, because a larger overfeed puts more effect thread into every metre — but only as the square root of the count, which is slow. The bracket therefore widens nearly in proportion to the overfeed.

At three-millimetre loops it is 5.6 at an overfeed of 0.3, 9.9 at 0.6, 12.6 at 0.8 and 21.7 at 1.6. Spacing the loops further apart has the same effect at a fixed overfeed: 8.6 at two millimetres, 20.5 at five. With semicircular loops, doubling the overfeed nearly doubles the range of cloths a yarn can make, and cuts the weight at the open end by about a third. This is where the semicircle misleads most. A real loop past a certain surplus spends its extra thread fattening a bulb rather than gaining height, so with the loops solved as elasticas the doubling from 0.8 to 1.6 widens the bracket by 37 per cent rather than 73, and lightens the open end by eighteen per cent.

Small loops are the gentle end. At an overfeed of 0.3 the loops stand 0.79 millimetres off the core, the envelope is 1.80 millimetres, and the bracket runs from 2.78 to 15.6 ends a centimetre, from 41 to 230 grams a square metre. That yarn weaves a cloth recognisably like an ordinary one at the close end and an open, loopy one at the other, with nine per cent of its area under yarn at the open end. The semicircle errs the other way here: a shallow loop rises as the square root of its surplus, so the real thirty-per-cent loop stands 1.10 millimetres rather than 0.79, its bracket is 7.6 wide rather than 5.6, and its open end is 2.06 ends a centimetre at 31 grams a square metre.

A woven bouclé’s weight says where in the bracket it sits #

The two ends of the bracket are computed. Where a real cloth sits between them is measured, and the measurement is one a mill already makes: the cloth’s weight. Given the yarn’s count, a plain cloth’s weight gives its sett, and a jammed cloth’s sett gives the diameter its yarns are actually touching at.

An 89 tex bouclé plain cloth of 100 grams a square metre has 5.6 ends a centimetre, so its yarns are touching at 0.89 millimetres — two and a half times the count’s diameter and a fifth of the loop envelope. On a logarithmic scale from the count end to the envelope end, that is 37 per cent of the way up. The loops that the yarn stood 2.10 millimetres off the core now stand a third of a millimetre off it between neighbours: the cloth has pressed its loops to about a sixth of their free height, and that fraction is the one number that describes how the loops and the weave have met.

At 50 grams a square metre the same yarn touches at 1.8 millimetres, 64 per cent of the way up, with loops at under two fifths of their free height; at 200, at 0.45 millimetres, near the count end, with the loops almost entirely pressed away. A cloth heavier than 252 grams in this yarn is not a jammed plain weave at all.

The weight prices the cloth and the loops sell it #

A cloth is bought by its weight and its width, and for an ordinary cloth those two numbers say a great deal about what arrives. For a bouclé they say much less. Two plain bouclé cloths of 100 grams a square metre, one in the eighty-per-cent yarn and one in a yarn of the same count with smaller loops, have the same sett and the same yarn in every square metre — and the first has its loops pressed to a sixth of their free height while the second may barely have pressed its loops at all. The label is the same and the cloths are not.

That is the same complaint a designed slub makes about a coefficient of variation, arriving from the cloth instead of the yarn: a summary number that works for a family stops working when somebody builds outside the family on purpose. A specification for a loop-yarn cloth that gives its weight and not the yarn’s overfeed and loop spacing has specified its cost and left its appearance free.

It is also why a loop yarn resembles a chenille more than it resembles the yarn its count names. Both carry an outline far larger than their mass implies, in material that stands off a load-bearing core, and both make cloths whose surface is built out of what the count leaves out. The difference is that a chenille’s pile is cut and a bouclé’s loops are closed, so a bouclé’s outline can be pressed flat by the weave and a chenille’s cannot be pressed below its tufts.

The loops are where the cloth wears #

Where a bouclé cloth sits in its bracket also says where it will wear. The loops carry no load — the effect thread is slack at every extension below its own surplus — so abrading them does not weaken the cloth’s strength. It changes what the cloth looks like, which is what the loops were bought for.

At the open end of the bracket the loops are the whole surface and stand at their full height, so every rub meets loop and nothing else. A float that stands proud of a surface takes the wear first, and a loop is the extreme case: a length of thread held only at its two ends, at the top of the cloth. At the close end the loops are pressed nearly flat between their neighbours, so a rub meets a surface of pressed loops and cores together and the loops are shielded by the structure around them.

The reading from weight puts numbers on that. The eighty-per-cent yarn at 50 grams a square metre keeps its loops at 37 per cent of their free height; at 100 grams, 16 per cent; at 200, five. A lighter cloth in the same yarn is bulkier, warmer and softer, and exposes more of its loops to wear, in exactly the proportion that it is lighter. That is the same kind of trade a raised surface makes: the part of a cloth that gives it its handle is the part standing out of the load path, and standing out is what exposes it.

What was counted, and how #

The loop radius is the overfed loop’s own geometry: a semicircular loop of radius r consumes r(π − 2) of the effect thread’s surplus, so r is the overfeed times the spacing over π − 2. The core’s diameter is taken from the core and binder together and the count’s diameter from the resultant count, both through volume arithmetic at a packing of 0.6 in cotton. The envelope is the core’s diameter plus two loop radii.

A plain weave’s jamming sett is the rule the setting essays use — each thread takes a diameter and each interlacing another — which for plain weave is one end in two diameters. The weight is twice the sett times the resultant count, with crimp left out. The weight ratio between the two ends was confirmed equal to the count’s diameter over the envelope to twelve figures across fifteen yarns; the envelope was confirmed to be the core plus two loop radii exactly; the bracket was confirmed to widen with the overfeed at each spacing and to exceed √2 everywhere; and reading a weight back through the jamming sett was confirmed to return each end of the bracket exactly.

Where the model stops #

The loops are semicircles, and a loop is not one. A loop held along its core at two binder points is an elastica, and solved as one it is taller than the semicircle below an overfeed of 66 per cent and shorter above it — 1.10 millimetres rather than 0.79 at thirty per cent, 1.95 rather than 2.10 at eighty. So the envelope end of the bracket computed here is too open at large overfeeds and not open enough at small ones; at eighty per cent the bracket is 11.7 rather than 12.6, and the reading of a cloth’s position from its weight moves by under a percentage point.

The cloth is taken as jammed. A plain weave set more openly than its jamming sett at either diameter is not touching at all, and its weight says only its sett; the reading of loop compression from weight applies to a cloth that is known to be tight.

The weave is plain, and the bracket does not care. A twill or a satin jams at a closer sett than a plain weave, because fewer of its threads have a crossing pick to make room for, and so both ends of a bouclé’s bracket move closer together in sett when the same yarn is woven as a twill. They move by the same factor, though: every weave’s jamming sett goes as one over the diameter that touches, so the ratio between the loop end and the count end is the ratio of the two diameters in any weave, and the open end’s weight is the same fraction of the close end’s. The weave chooses where the bracket sits; the yarn alone chooses how wide it is.

Crimp is left out of the weight, and a bouclé’s crimp is hard to state, because the yarn’s outline and its load-bearing core bend around the crossing pick at different radii. The weights are therefore lower bounds by the few per cent crimp would add.

And the loops are treated as either standing or pressed. A loop half-pressed on the face and fully pressed where two yarns cross is the real state, and a single effective diameter is an average over it.

Still open: where woven bouclés actually sit #

The reading from weight turns any bouclé cloth whose yarn is specified into a point in its bracket, and a table of commercial bouclés — yarn count, overfeed, loop spacing, cloth sett and weight — would show whether real cloths cluster near the count end, near the envelope end, or across the range. The expectation from the bulk these cloths are bought for is the middle, with the loops pressed to a fraction of their free height, but that is a prediction and not a reading.

The measurement is cheap: the counts and the weight are on every specification sheet, the overfeed is a machine setting, and the loop spacing can be read off a length of yarn. What it would establish is whether the bracket is a design space designers use or a range most cloths sit at one end of.

Who worked it out #

Loop and bouclé yarns are old, and the observation that a fancy yarn’s count misstates its thickness is as old as the yarns. The jamming sett as a rule of diameters is Ashenhurst’s and Peirce’s, from the late nineteenth and early twentieth centuries. The two-diameter bracket for a loop yarn, its exact weight ratio, and the reading of loop compression from a cloth’s weight were computed directly.