A meniscus reads how thick the ends are, not how wide
Worth reading first: A meniscus closes the gap only at two-thirds cover · What a flattened yarn does to its cover · A calender's best cloth is the one its nip fills.
A meniscus closes the gap only at two-thirds cover found the condition under which a liquid can hold across the gap between two neighbouring round ends, and found it out of reach. A liquid that wets perfectly needs a warp cover of 2/π, 0.637, and the closest of the eight named cloths, the sheeting, stops at 0.52. That essay ended by asking whether a flattened yarn, which covers more at the same sett, gets there sooner.
It does. Writing the condition out for ends with a flat top and rounded edges shows that the liquid never sees how wide the ends are, only how thick, and that changes what flattening does. Pressing an end flat widens it, which narrows the gap, and thins it, which lowers the gap the liquid can hold across. The first effect is the larger, so a flattened warp holds liquid across its gaps at a cover a round one cannot. But it gets there only near the end of the flattening each cloth has room for, and for every named cloth that end is further than a calender presses.
What the liquid touches
The earlier account set out the energy for two round threads. A full column of liquid between them, running along their length, has a nearly flat surface above and below. Each surface meets a thread where the thread’s tangent makes the contact angle θ with it, so the column wets an arc of each thread and pays for two free surfaces spanning the gap. Princen’s balance between the two is the whole condition.
A flattened end is a racetrack, which is how the cover arithmetic has always drawn one: a flat run across the top and bottom with a half-round edge on each side, of radius half the thickness t. Between two such ends the liquid meets only the two facing edges. Their flat tops are above and below the column, out of its way, and the free surfaces stretch from one rounded edge to the other across the clear gap g. Setting the wetted arcs against the free surfaces gives
The end’s width is gone. What remains is the clear gap against the thickness, and at θ = 0 the bracket is π/2 − 1, so a liquid that wets perfectly holds across any gap narrower than 0.571 of the ends’ thickness. A round thread is the case where width and thickness are the same, and the condition reduces exactly to the 2/π cover again, which is the check that nothing was lost in generalising it.
The bracket falls with the contact angle and reaches zero at ninety degrees, where wetting a thread earns nothing and no column pays for itself. At thirty degrees the gap must be under 0.407 of the thickness; at sixty, under an eighth of it.
Flattening narrows the gap faster than it thins the end
Take the sheeting’s warp. Its ends are 187 µm across when round and sit 357 µm apart, so the clear gap is 170 µm, while a perfect wetter could span 0.571 of 187, or 107 µm. The gap is the larger by sixty per cent and nothing holds.
Now press the ends flatter at constant area, which is what a calender nip does to a thread and what the racetrack keeps track of. At twice as wide as thick an end is 248 µm wide and 124 µm thick. The gap has fallen to 109 µm; the allowance, 0.571 of the new thickness, to 71. Both moved, and the gap is still a half again too wide.
The two shrink at different rates because they are different kinds of quantity. The allowance is a fixed share of the thickness, and the thickness falls slowly: at constant area it goes roughly as one over the square root of the flattening, so quartering it takes a flattening of sixteen. The gap is a difference, the spacing less the width, and a difference reaches zero. The width needs only to grow to the spacing, which it does at a finite flattening, while the thickness is still a good fraction of what it was. One curve falls to zero and the other only eases down towards it, so they must cross, and they cross before the ends touch.
For the sheeting they cross at ×3.08, with the gap down to 56 µm. The round-thread cover the curve stands for has fallen from 0.637 to 0.523, which is the sheeting’s own cover, and that is the first figure’s point of crossing. Across all flattenings the cover a perfect wetter needs runs 0.637 round, 0.586 at ×2, 0.527 at ×3 and 0.443 at ×5. Flattening lowers the limit, which is the answer to the earlier essay’s question.
How flat each named cloth must be pressed
The same crossing can be found for every named cloth at its own sett, and the flattening it falls at is a measure of how far each cloth is from holding liquid across its gaps.
For a perfect wetter the sheeting needs ×3.08, the filter cloth ×4.19, the poplin and the duck ×4.44, since their warp covers happen to be the same, the muslin ×6.46, the batiste ×7.46, the voile ×11.78 and the cheesecloth ×29.02. A liquid at thirty degrees needs a little more on each, ×3.51 on the sheeting; at sixty, ×4.15.
A calender’s best cloth is the one its nip fills took its nips from ×1.25 to ×3, the heaviest a calender is used at, and a calender works where nothing has been measured warned that even ×3 is a shape strain no fibre’s recovery has been measured at. On that range not one named cloth closes. The sheeting misses by eight hundredths. Every other cloth needs a third as much flattening again or more, a figure no finishing range gives a woven cotton.
The weft is further off still. The same crossing in the other direction, between neighbouring picks, comes at ×3.87 on the sheeting, whose picks are set a little more openly than its ends, at ×4.83 on the filter cloth and at ×7.99 on the poplin, whose 22 picks a centimetre are a long way from its 32 ends. A cloth that held liquid across its gaps in both directions would have to be pressed past the larger of its two figures, and on a warp-faced shirting that is nearly double the warp’s. A meniscus that spans a cloth’s gaps one way and not the other is the ordinary case in the narrow band where anything spans at all, so the closer system decides whether a cloth can hold a web, and the sett of the other decides nothing until it is pressed much further.
So the earlier essay’s conclusion stands and is sharper than it was. For a round thread it read the limit as too high a cover to reach by weaving closer. With flattening allowed, the same cloths would need to be pressed beyond anything a calender does, and on all but the sheeting well beyond. A matrix census’s prediction that a twill’s crowns carry a film from the first gram stays wrong for every cloth a finisher handles.
The last stretch before the ends touch
Each bar in the fourth figure ends where its cloth’s ends would meet their neighbours, and every dot sits near the end of its bar. The sheeting closes at seven tenths of the flattening it has room for, the poplin at three quarters, the voile at nine tenths and the cheesecloth at ninety-six hundredths. A meniscus closes a gap only when the gap is nearly gone.
That follows from the allowance being a fraction of the thickness. A gap a liquid can span is at most 0.571 of a thread that flattening has already thinned, and the gap has to fall under that before anything happens, so the closing always comes late. The more openly set the cloth, the further its ends have to spread to cover the spacing, the thinner they are by the time they get near, and the closer to touching the crossing has to wait for.
The calender arithmetic met the same wall from the other side. A covered cloth cannot be calendered, because a thread pressed as wide as its own spacing has no room left to widen into and so no plateau to shine with. The meniscus needs the ends within half their thickness of touching; the lustre needs them short of touching. Between the two is a strip of flattening inside which a cloth would hold liquid across its gaps and still have room to widen, and it is remarkably constant: from 1.16 units of flattening wide on the cheesecloth to 1.35 on the sheeting. No named cloth’s strip begins inside a calender’s range.
What a first web holds
Once a gap holds liquid, the least that spans it is a web whose two surfaces bow inward until they meet at the edges’ waist. Between flattened ends only the rounded edges hold it, so the web is the round pair’s web drawn on the edges’ own radius.
The sheeting’s first web, with every gap in its warp holding one, takes 1.69 grams of liquid a square metre; the poplin’s 0.79. The round-thread poplin past its own limit had needed several times as much. A flattened end is thinner, so its edges are smaller and the web between them is too. The webs a flattening makes are the thinnest webs a cloth could hold, and for a finish that has to bridge a gap they are the cheapest. They are also unstable to anything that takes liquid from them, since a web that thin has almost nothing in reserve before its two surfaces meet.
Where flat threads are ordinary
The cloths that never get there are the ones a calender handles. There is a class of fabric that is flat by construction.
A tow is not a yarn took a 400 tex glass reinforcement tow at eight times as wide as thick: 1.39 mm across and 0.174 mm thick. Its allowance for a perfect wetter is 0.571 of 0.174, about a tenth of a millimetre. A woven reinforcement whose tows lie within a tenth of a millimetre of each other — a cover, in tow widths, of 0.93 — therefore holds any well-wetting liquid across its gaps, and a reinforcement woven close for fibre content, which is what it is bought for, can sit there without being pressed at all.
The resin in such a cloth does not run only along the channels between tows; it can span them. That connects to the problem a fabric to fill and a fabric to load described, of a reinforcement that is a fast medium between its tows and a slow one inside them. A web across a gap is liquid held above the channel rather than in it, and whether that helps the channel fill or traps air beneath it is a question of the order in which the two arrive. The arithmetic here says only that the bridging is available to a flat tow and not to a round yarn.
The practical reading
For a woven garment cloth, a meniscus across the warp gaps is not available at any cover a weaver sets or any flattening a calender gives, so every finish behaves as the coating arithmetic found: it fills from the bottom of the relief, and a twill’s ridge is no easier to coat than a hopsack’s crowns. A thinner liquid or a better-wetting one does not change that; it moves the curve by less than the distance to the cloth.
For a cloth that is to hold liquid across its gaps, the thing to change is the thread’s thickness, not its width and not the sett. A finer yarn set closer has thinner ends and a smaller allowance; a flat filament tape or a spread tow has its allowance set by a thickness that has nothing to do with its width. That is why a flat-tape fabric can hold liquid where a cloth of the same cover in round yarn cannot.
And a cloth flattened in use — a seat fabric pressed under load, a ribbon wound tight — moves toward the limit with every increment, and holds liquid across its gaps late and suddenly if it gets there at all.
How the numbers were reached
The condition is Princen’s energy balance for a full column between two parallel surfaces, applied to the rounded edges of a racetrack section, the shape Peirce’s geometry set beside the round thread and that every flattened calculation here uses. The flattening is applied at constant area, so a thread pressed to f times as wide as thick has thickness κ(f)·d, with and d the round diameter. The crossing flattening is found by bisection on the inequality, separately for each cloth at its own sett and each contact angle, and the touching flattening by a second bisection on the width reaching the spacing.
The webs are the least webs between the two facing edges, each edge’s radius half the flattened thickness, found as the bowed pair of equal arcs whose surfaces meet at the edges’ waist, with the solid inside the edges subtracted. A web’s add-on is its section divided by the end spacing, times a density of 1.2.
Where the edges stop
The edges are half-round. A racetrack is a convenience and a pressed yarn’s edges may be blunter or sharper. A sharper edge has a smaller radius and spans less; a blunter one, more. The allowance is proportional to the edge radius rather than strictly to the thickness, and the two are the same only for this shape.
The pick under the gap is left out, as before. At a step of a twill’s ridge the pick lies below both ends, and a column bridging between them has air beneath it. Flattening thins the warp, which brings the pick’s crown closer to the warp’s underside, and that should help a web form. How much is not computed.
Every end is flattened alike. A real nip presses the ends that stand highest hardest, and a cloth with a spread of yarn diameters has a spread of flattenings. The gaps between the flattest pairs close first, and partial closure of that kind is not computed.
Equilibrium only. Whether a finish has time to form a web before it is absorbed or dried is a separate question; the thinness of the webs found here makes it a pressing one.
Hairs, pores and a yarn’s real contact angle are not modelled. All three reduce a yarn’s apparent contact angle, which would move each dot in the fourth figure left — by too little, at θ = 0 already, to bring any cloth inside a calender’s range.
Who found which part
The two-cylinder condition is Princen’s, from his 1969 and 1970 papers on capillary phenomena in assemblies of parallel cylinders, and is used here unchanged. The racetrack section is Kemp’s, from 1958, and flattening at constant area is the convention every account of yarn flattening uses. The observation that the cylinders’ spacing enters only as a clear gap against an edge radius, so that a flattened thread’s width drops out, is the step added here, along with the consequence that flattening lowers the cover limit but only near the touching point, and the calculation for each named cloth. That flat tows and tapes wet differently from round yarns is known to the composites and coated-fabric trades as practice; the threshold in thickness is not known to have been written down in this form.
Still open: whether a pressed edge is round
Every number here rests on the edge being a half-circle of radius half the thickness, and the allowance scales with that radius. A calender nip presses a thread between flat rolls, and there is no reason the edges it leaves are round: a yarn is a bundle, and a bundle pressed flat may bulge at its edges or square them off. A micrograph of a calendered cloth’s section would give the edge radius directly, and the dots in the fourth figure would move in proportion. If a pressed edge is blunter than half-round, the sheeting’s eight hundredths might be made up and the closest calendered cloths would hold a web; if sharper, the limit recedes further. The radius of a pressed edge, measured against the thickness of the end it belongs to, is the one number that decides whether this essay’s negative is true of real cloth or only of racetracks.
What links here
Computed from the collection rather than written here: the essays that point at this one.
Shares its objects with
Essays naming at least two of the same things, that neither author linked.
- A calender spends the compression for good — both name calendering, cover factor, racetrack
- Calendering is the cloth arriving at the other model — both name calendering, cover, racetrack
- A coated cloth fails at its holes — both name coating, cover
- A hole with nothing crossing — both name cover, cover factor
- A print is as sharp as the hairs are long — both name coating, wicking
- A windproof cloth is at its yarn's limit — both name coating, cover factor
Named objects
A flat tag is an object no other essay names yet.
CalenderingCoatingContact angleCoverCover factorRacetrackWicking