Textile Structure
how cloth is put together
The collection

Every essay

One idea per essay, ordered so that the earlier ones set up the later ones — but nothing here depends on being read in sequence.

What cloth is Weaves Setting and geometry Knits and other structures Mechanics and drape Pattern and colour Compound and figured cloths After the loom Cloth doing a job

FieldsSeriesThreadsWeave indexRefutationsConceptsSearch

Cloth doing a job

What happens when a fabric has to meet a specification rather than merely be described. A reinforcement that must hold fibre and still admit resin, a filter that must retain the soil and still pass water, a seam that must grip harder than the load without perforating the cloth, a pattern cut for a shape it does not yet have. Every requirement here is a pair of inequalities on quantities the earlier fields compute — and the useful answer is sometimes that no fabric satisfies both.

How much fibre a woven reinforcement holds. A unit cell of a woven reinforcement in section: flat tows, the thickness they add up to, and the wave the warp makes to cross them. The bar below is the fibre volume fraction against the same tows laid flat in two plies and against the packing factor of a tow, which is the most any cloth of them could be. The vertical scale is exaggerated so the interlacing is legible; the horizontal scale is the cloth's own.

The crimp is the price of being cloth

A woven reinforcement is bought for stiffness along its fibres, and the usual explanation of what it gives up — fibre content, lost to the crimp — has the sign wrong. At a given thickness a woven fabric holds slightly more fibre than two flat plies of the same tows. What the crimp costs is stiffness, and the float length is the knob.

7 figures · Preform A fabric to fill and a fabric to load. Fill time against the sett of a woven reinforcement, with the two limits a part imposes. Stiffness wants fibre, which wants a close sett; the resin has to arrive before it gels, and the channels between the tows — which is what carries the flow — close as the sett rises. The interval between the two is where a fabric can exist, and it narrows with the size of the part.

A fabric to fill and a fabric to load

A reinforcement has to hold as much fibre as possible and still let a resin through it, and the two demands are the same decision pulling opposite ways. Both bounds are computable, the interval between them narrows as the square of the part, and past a certain size it is empty — which is the most useful thing the arithmetic says.

6 figures · Preform Two sections, one yarn. The same yarn given a circular cross-section and a racetrack one of equal area, both jammed. The spacing the two models allow is nearly the same; the cover and the cloth thickness they predict are not.

A tow is not a yarn

Every geometric model on this site assumes a thread with a diameter. A reinforcement tow has a width and a thickness instead, and the two are eight to one — which moves the jammed sett, the crimp, the cover and the cloth's thickness at once, in four different directions, from one change.

6 figures · Sett A preform through its thickness, and whether it is one piece. An orthogonal three-dimensional preform in section, with the binder's path, beside the digraph the integrity criterion consumes: one node per level, an arrow from each level to the one above it, and the binder's own contacts. The warp and weft here are straight and do not interlace at all, so the whole of the connectivity is the binder — and the count of separable pieces is what the criterion returns, unchanged from the weave it was written for.

The third index is not a repeat

The fancy weaves left three-dimensional weaving open as a possible fifth escape from the binary matrix. It is not one. The criterion that decides whether a plain weave is one cloth decides a five-layer preform unchanged — and what a third dimension actually takes away is periodicity, because a thickness has a top and a bottom and a repeat does not.

6 figures · What cloth is The hole between four threads. Three millimetres of a woven filter cloth in plan, with one clear opening dimensioned and a grain of the soil it must retain drawn at the same scale. Every hole in the repeat is this size — a woven cloth's pore distribution is a single value, which is why it is specified by an opening size while a nonwoven needs a curve.

The hole between four threads

A woven cloth's holes are all the same size. That is not an approximation — it is what a repeat means — and it is the whole reason a woven filter is specified by one number while a nonwoven needs a curve. The number is the spacing less the diameter, which this site has been computing since its first questions about setting.

6 figures · Cover A filter cloth has two jobs. Opening size against sett for a woven filter cloth, with the two limits that specify one: the soil is retained below the retention line, and the water passes above the open-area floor. They pull opposite ways, so the answer is an interval in the sett — and for a fine enough soil there is no interval at all.

A filter cloth has two jobs

Hold the soil back and pass the water. The first wants a close sett and the second wants an open one, so a filter cloth is not a fabric but an interval — one that is sometimes a single sett wide, and for a fine enough soil is empty. The empty case is the useful one, because it says the answer is not a woven cloth at all.

7 figures · Openings What a balanced cloth wastes under pressure. The fraction of a balanced fabric's fibre that is along for the ride, in a stress field of each ratio. A closed cylinder is exactly two to one — the ratio of the two areas the pressure acts on — so a balanced cloth reaches its limit around the circumference with the axial system at half its capacity, and a quarter of the fibre is doing nothing.

An inflated cylinder wants an unbalanced cloth

Balance is a virtue in almost every other cloth. Under pressure it is a defect with a size — a closed cylinder carries exactly twice the stress around its circumference as along its axis, so a balanced fabric reaches its limit in one direction with a quarter of its fibre doing nothing at all.

6 figures · Membrane The angle a pressurised hose wants. A fixed length of yarn wound on a cylinder at a stated angle, with the radius and length the geometry gives, beside the volume it encloses as the angle varies. The maximum is at arctan √2 — 54.74° — with no material constant anywhere in it, and the same angle comes out of balancing the hoop and axial stresses, which is a different calculation with the same answer.

The angle a hose wants

A braided hose has one angle at which pressure neither lengthens it nor shortens it, and the angle is arctan √2 — 54.74° — with no friction coefficient, no modulus and no fitted constant in it. Two arguments that share no algebra arrive at the same number, and which side of it a hose was braided on decides which way it moves.

6 figures · Bias A membrane is cut smaller than it is. Warp and fill compensation against how far the crossings flatten under prestress. With the thread lengths and the cloth's thickness both fixed there is no strain available at all — the closure condition determines the state — so the whole of the compensation is the crossings squeezing and the crimp coming out. The direction that arrives with less crimp gives back less, which is why a warp compensates less than a fill.

A membrane is cut smaller than it is

A tensioned fabric roof is cut to a pattern smaller than the shape it will take, because stressing it makes it grow. At a fixed cloth thickness a prestress can only interchange crimp — one direction grows and the other shrinks — so everything that makes both directions grow is the crossings flattening, which is the one quantity here this site cannot compute.

6 figures · Crimp Where a bias cut's waste actually is. A bolt of cloth with panels placed at a stated angle, the ones that fit drawn and the rest of the cloth left shaded. Identical panels at a single angle tile the plane exactly, so the interior of the bolt loses nothing at all and the whole of the waste is at the two selvedges. The inset is the single-panel bounding box, which is the picture the usual account of bias cutting draws.

The bias cut and the selvedge

A bias-cut square costs exactly twice its own area, and every other shape costs more. That is an exact result about one panel — and it is not where a cutting room's waste comes from, because identical panels at one angle tile the plane. The loss is at the two selvedges, so it falls as the cloth gets wider, which no account in terms of the diagonal can explain.

6 figures · Cutting How much of the curvature a cloth can take without being cut. The dart angle a spherical cap still demands after the fabric's own shear has absorbed what it can, against how closely the cloth is set. The total the cap demands is fixed by Gauss–Bonnet and is the same for all of them; what changes is how much of it the trellis can supply before its threads jam. An open cloth drapes a hemisphere with no dart at all; a closely set one has to be cut from the start.

A hemisphere costs one full turn

The total angle a pattern must remove to cover a hemisphere is exactly 360 degrees, and it is the same for a hat and for a stadium dome. What changes with the fabric is how much of that the cloth can supply by shearing instead of by being cut — and that is a property of the sett, computed from the angle at which its threads jam.

6 figures · Cutting What holds a thread in a seam. A cloth thread inside a seam allowance, drawn past the crossings that grip it, with the capstan's factor at each. The tension falls by that factor at every crossing, so the grip is exponential in the crossing count and the count is what the weave decides. Beyond the thread's own strength the thread breaks rather than slides, and the rest of the allowance holds nothing.

What holds a thread in a seam

A seam fails in two quite different ways, and the one the trade worries about is not the stitches breaking. It is the cloth's own threads sliding out of the weave beside the seam — and what resists that is friction at the crossings, accumulating multiplicatively, so a satin gives a seam a thousandth of the grip a plain weave gives it.

6 figures · Seams The stitch density that makes the strongest seam. Seam strength against stitch density, as the two limits that decide it: the sewing thread crossing the seam, which rises with the stitches, and the fabric the needle perforates, which falls. The seam is the lower of the two, so the optimum is where they cross — and whether the fabric line falls at all is decided by the clear gap between threads against the width of the needle.

The stitch that weakens the seam

More stitches per centimetre put more sewing thread across a seam and more holes through the cloth beside it, so seam strength rises, crosses and falls. Whether the fabric line falls at all is decided by the clear gap between two threads against the width of the needle — the same arithmetic a filter cloth is specified by, doing a different job.

6 figures · Seams A cloth does not mind a hole. Strength left after a hole, against the width of the hole, for a woven cloth and for a film of the same width. The cloth's threads carry their own load, so it loses exactly the threads the hole removes and the loss is linear. The film is a continuum, so a hole of any size at all costs it the stress concentration factor — quoted from Kirsch's elasticity solution, not derived here — and the size of the hole does not enter.

A cloth does not mind a hole

A hole in a film costs it two thirds of its strength whatever the hole's size, because a continuum concentrates stress at an edge. A cloth's threads carry their own load and hand almost nothing to their neighbours, so a hole costs exactly the threads it removes — a loss that is linear in the hole, independent of the sett, and zero for a slit along the load.

7 figures · Damage A tear breaks threads or pulls them out. The grip a cloth has on a thread at the tip of a tear, against the sett, with the thread's own strength drawn across. Below the line the thread slides and the yarns group, which is the trade's explanation of why a loose weave tears well; above it the thread breaks where it is and grouping never happens. The essay that asked it could not compute this: it needed a friction, and the fancy weaves supplied one.

A tear stops where the grip is

This collection asked whether a loose weave tears better and had to answer that the geometry could not say — it supplies a grip count and a slack, and neither is a force. The fancy weaves brought a friction. With it the question has an answer — a cloth's threads slide up to a computable sett and break above it — and a ripstop grid has a bound with no free parameter in it.

6 figures · Damage Two pore systems in one cloth — 24 threads per centimetre. A plain weave of 20 tex cotton in section, at 24 threads per centimetre, so the yarn is 167 µm across and the clear hole between two picks is 250 µm. That hole's hydraulic radius is 124.8 µm. Inside the yarn, fibres 14 µm across packed at 0.6 leave spaces of hydraulic radius 2.33 µm — 53 times finer, and by Jurin's law 53 times higher: 6.37 m against 119 mm. The yarn's interior is magnified 6 times and the two discs at the foot are the only part drawn at one scale.

How high a cloth wicks

A woven cloth has two capillary systems and they are a factor of twenty to a hundred apart. The one every diagram draws — the hole between four threads — lifts 119 mm. The one nobody draws, inside the yarn, lifts 6.37 m.

6 figures · Wicking A front on a thread with 8 per cent crimp. Three rows at one scale. The top row is the warp end laid out straight, with the wetted front marked at four equal quarters of its own length — which is where Washburn's law puts it at four times whose square roots are evenly spaced. The middle row is the same thread crimped at 8 per cent, so it covers 92.6 per cent of the paper the straight one did. The bottom row is the cloth, and the four fronts on it are the four above pulled back by 1.08. A coefficient is a length squared over a time, so it comes down by 1.1664 — exactly (1 + c)², with no property of the liquid or the fibre in it. The thread's thickness is not drawn and neither is the liquid: a meniscus in a 2.33 µm pore is finer than any line on this canvas.

Wicking is slower along a crimped thread

A front travelling up a warp end travels the thread's path, which is longer than the cloth by exactly the crimp. So the wicking coefficient measured on the fabric is the yarn's own divided by (1 + c)² — 14.3 per cent lost at eight per cent crimp, whatever the liquid.

6 figures · Wicking A coating over the hole it has to bridge. Four warp ends of 40 tex seen end-on at 20 per centimetre, with a 120 micrometre PVC film over them. The film is supported everywhere it lies on a thread and unsupported over the 252 micrometre clear span between them, which is the same span at every hole in the repeat. The film's thickness is drawn to scale against that span; the bulge is exaggerated. At 20 MPa the film will hold at most 381 bar, which is a bound and not a prediction.

A coated cloth fails at its holes

Spread a film over a fabric and it is supported everywhere it lies on a thread and unsupported over every hole. A woven cloth's holes are all the same size — exactly, because a repeat is a repeat — so the film has one span to bridge and the fabric has a burst pressure rather than a distribution. A nonwoven gives it a distribution, and a membrane fails at the largest hole it meets.

6 figures · Coating Which seams slip and which break. For each cloth, the grip a 10 mm seam allowance offers divided by the thread's own breaking load. Above one the fabric or the thread gives first and the seam holds until it does; below one the threads slide out and the seam opens with the cloth intact. voile, batiste, poplin, sheeting break; cheesecloth, muslin, duck, filter slip. The allowance that would save every cloth in the table is 74 mm, which is set by the openest of them alone. What the chart cannot show is the stitching itself, which has its own strength and its own way of cutting the threads it passes through.

A seam slips before it breaks

A sewn seam fails in one of two ways and the trade names them separately: the threads pull out of the cloth beside the stitching, or something breaks. Which one a given cloth does is decided by whether its seam allowance clears the length at which grip beats strength — and at the ordinary ten millimetres, the table divides four against four.

6 figures · Seams The size of the capstan correction. The ratio of the capstan crossover to the crossover a sum of independent contacts gives, for every cloth in the table at a friction coefficient of 0.30. It is exactly ln(1 + z)/z, where z is the thread's breaking load times the wrap angle, over the contact force — a quantity with no friction coefficient in it at all. That is why the earlier result that μ·L* is exactly constant survives this correction to twelve figures: μ was only ever in the factor outside the logarithm. The correction is largest for the duck, whose coarse strong yarn makes z large, and smallest for the batiste. What the rows cannot show is that a real cut edge frays at a friction nobody measured on that particular cloth.

How far a cut edge frays

A seam allowance, a fray width and a tuft's bound length are the same number wearing three hats, and the earlier estimate of it was four times too long. Correcting it moves the whole table across the boundary an ordinary allowance sits on — from four cloths holding and four slipping, to all eight holding — and turns a specification argument into a different one.

6 figures · Seams Shrinkage from wetting, cloth by cloth. The change in each cloth's relaxed construction when its threads swell by 20% across and 1.2% along. Both states are least-energy states of their own constant-thread-length locus, which is the only comparison the locus admits: its minimum is a state at its own sett, so it may be compared with another minimum and not with a quoted construction. Positive is smaller. Every cloth that has a wet state closes up except the cheesecloth, which opens by half a per cent because the fibre's extra hundredth of length outruns the extra crimp a thicker partner costs. What the bars cannot show is the three cloths that are missing, which have no wet state at all.

A garment is cut dry and worn wet

A cutting allowance is one number and the thing it allows for is two, in two directions. Worse: when the two directions differ, a panel cut on the bias does not merely shrink — it rotates, by half a degree for an ordinary poplin, which is nine and a half millimetres of skew across a metre and is invisible to a tape measure.

6 figures · Cutting The shear a dome demands, against how far round it the cloth reaches. A flat sheet of inextensible threads takes a double curvature only by shearing, and the shear it needs depends on how far round the dome it has to reach rather than on how big the dome is — a knee, a shoulder and a beach ball demand exactly the same at the same fraction of their own radius. The horizontal line is the locking angle for a sheeting, where the threads are touching side by side and the mechanism has nowhere left to go. Reaching one radius round takes the cloth to 85% of that, and reaching 1.2 radii passes it. What the plot cannot show is the frictional part: a shear well inside the locking angle is still a shear at crossings friction is holding, so a knee that is domed a thousand times keeps a little of each one.

A knee is a dome imposed a thousand times

A flat sheet of inextensible threads takes a double curvature only by shearing, and how much shear it needs depends on how far round the dome it has to reach — not on how big the dome is. So a knee and a beach ball demand the same, and a trouser knee covered to its own equator is at 85 per cent of the angle at which the threads touch side by side.

7 figures · Damage Seven weaves at one construction, and what each of them really passes. Every weave here is drawn at the same filter — the same yarn, the same sett, the same cover and therefore the same open area, equal to twelve decimal places. Each one shows an opening of 263.7 µm to anybody looking straight through it, and that is the number a specification quotes. The bar is what each will actually let past, which is the narrowest section anywhere along the channel rather than the narrowest view down it. The plain weave passes exactly what it shows and it is the only weave here that does: its ends transit at every gap, so the four threads round every hole are level in pairs and the waist is at the middle. A float leaves two ends together at the top of the cloth, their neighbours at the bottom, and a passage between them that is wider than its own mouth — up to 12.8 per cent over the rating, for the basket.

A filter is rated by the hole it does not show

A woven filter cloth is sold on two numbers pulling opposite ways: an opening small enough to hold the soil and an open area large enough to pass the water. Both are computed from the same holes, and they are not computed from the same statistic of them. One reads the maximum and the other reads the mean, so a change that improves either can worsen the other without moving a single measurable property of the cloth.

6 figures · Filtration What a muslin passes, against how closely it is set. A muslin's air permeability at 100 Pa as the sett is closed from 6.9 to 34.2 threads per centimetre, with the two paths separated. The channels between the threads carry 8285 mm/s at the open end and 1330 at the close one, and they go to zero when the cloth jams. The threads themselves carry 1.7 to 6.4 mm/s and never go to zero at all, because a thread is sixty per cent fibre whatever the sett is. Extended to a cloth with no channel left, that path is 8.1 mm/s — a floor set by the yarn rather than by the construction, and a specification asking for less has asked for a fabric that cannot be woven from this yarn at any sett.

A cloth stops having holes before it stops passing air

Close a woven cloth up and the channels between its threads shut. What passes through it does not go to zero, because a thread is sixty per cent fibre and forty per cent air and stays that way whatever is done to the construction. So a fabric's permeability has a floor, the floor belongs to the yarn rather than to the weaver, and no sett on any loom reaches it.

6 figures · Permeability