Give the warp protection that survives the loom
A sizing record can look stable while the loom tells a different story. The bath concentration may be on target, yet one edge sheds more size than the centre. A stronger yarn may also have lost too much stretch. Adding more material will not explain either problem.
Warp sizing applies a temporary protective treatment before weaving. The yarn must withstand rubbing and repeated loads. Its ends should also separate cleanly and unwind well. The treatment must suit the later removal and finishing route.
This guide follows the measurements behind that result. Its examples are invented to explain the calculations. They do not recommend settings for cotton or any other fibre. A mill must set its own working range for the yarn, recipe, machine and fabric.
Ask three separate questions during a trial. How much size was added? Where does it sit in the yarn? How does the treated warp perform? Each needs evidence; a single percentage cannot answer all three.
Put a denominator beside every percentage
CottonWorks defines size add-on as the increase in yarn weight caused by sizing. For process calculations, state the moisture basis too. Otherwise, water retained in the sample can be mistaken for dry size.
Use the following definitions for the worked examples:
Bath solids: dry, nonvolatile material as a percentage of total liquor mass.
Retained liquor pickup: liquor carried by the yarn, expressed against the dry mass of unsized yarn.
Dry size add-on: dry sizing material retained, expressed against the dry mass of unsized yarn.
Moisture regain: water mass expressed against the dry mass of the material being assessed.
The retained-liquor definition matters. A wet sample may contain both incoming yarn moisture and water from the bath. Those sources need to be separated before a simple pickup calculation is used.
Let C be bath solids percent and W be retained liquor pickup percent. If the retained liquor has that solids content and there are no solid losses:
Dry size add-on (%) = W × C ÷ 100.
Both percentages use mass. Convert a recipe stated in grams per litre to a mass basis before using this equation. The conversion needs the liquor's density. Do not assume that every liquor has the density of water.
These definitions also distinguish application from purchasing. A drum labelled “50% solids” contains product and water. Its full mass is not the dry size applied to the warp.
Build one complete solids and water balance
Consider 500 kg of dry unsized warp. For the first calculation, assume it enters the balance without moisture. It carries 60% of its dry mass as liquor after application, and the retained liquor contains 12% solids.
The retained liquor is 500 × 0.60 = 300 kg. Of that, 300 × 0.12 = 36 kg is dry size and 264 kg is water.
Dry size add-on is therefore 36 ÷ 500 × 100 = 7.2%. The dry sized warp contains 500 + 36 = 536 kg of solids. Immediately before drying, its total mass is 500 + 300 = 800 kg.
Suppose the illustrative exit condition is 7% moisture regain on the dry sized mass. The remaining water is 536 × 0.07 = 37.52 kg. Exit mass is then 573.52 kg, and the dryer removes 264 − 37.52 = 226.48 kg of water.
The balance closes: 800 − 226.48 = 573.52 kg. This check is useful because it exposes a common denominator error. Applying 7% only to the original 500 kg would leave 35 kg of water. That calculation would use a different stated basis.
The 7% condition is chosen only to explain the arithmetic. It is not a storage limit or a universal setting. Nor is 226.48 kg an estimate of steam use. Heating the yarn and machinery also takes energy. Heat losses and the drying system affect the total demand.
Account for incoming yarn moisture
Now change one assumption. The same 500 kg of dry yarn brings 40 kg of water into application. Assume all that water remains with the sheet. In this simple balance, it adds to the same 300 kg of retained bath liquor.
Wet mass is now 840 kg, while dry solids remain 536 kg. Water entering the dryer is 40 + 264 = 304 kg. At the same illustrative exit condition, evaporation becomes 304 − 37.52 = 266.48 kg.
A wet-mass shortcut would give (840 − 500) ÷ 500 × 100 = 68%. That figure includes incoming water. Treating it all as 12% bath liquor would falsely predict 8.16% add-on, instead of 7.2%.
Water can exchange with the yarn in a real size box. Dilution and carryover may also affect the balance. Use the measured flows and sample basis. Define “wet pickup” clearly before two departments compare their numbers.
Check the dry measurement before changing the recipe
Sekisui's technical guide to warp sizing defines add-on against dry unsized yarn. It also distinguishes the amount of size from its position around and within the yarn. The lab removal method must suit both the fibre and the sizing material.
For an ideal matched specimen, suppose dry unsized material weighs 20.00 g. After sizing and drying on the same mass basis, it weighs 21.44 g. Assume no fibre was lost.
Added size is 1.44 g, so add-on is 1.44 ÷ 20.00 × 100 = 7.2%. Dividing by 21.44 instead gives 6.72%, the size fraction of the final dry sized mass. That can be a valid separate measure, but it is not the same percentage.
An extraction test often works backwards from sized yarn. A laboratory removes the size and weighs the remaining dry yarn. The result needs a validated method for that actual recipe. Incomplete removal leaves a high residue mass and understates the apparent add-on. Loss of fibre or other unsized-yarn material can push the estimate the other way.
Keep an appropriate unsized control. The treatment may also remove natural matter or other added material from that yarn. If so, the sized sample's whole mass loss cannot be labelled as size. Check the blank loss and method correction before accepting the result.
Equal cut lengths are not automatically equal original yarn masses. Stretch during sizing can change the amount of yarn in a fixed finished length. A comparison between separate samples also needs to allow for yarn-count variation. ApparelDiary's yarn-count conversion guide explains the mass-per-length basis.
Label the result clearly: dry add-on, conditioned mass gain or extractable content. Keep tare weights and raw masses so another person can check the calculation. Follow the chosen method for repeat tests. Rounding the answer cannot repair a weak sample comparison.
Control the bath as a changing mixture
A recipe states the starting mix. Water then evaporates, fresh liquor enters and the warp carries material away. Those flows can change the liquor at the application point. A tank entry alone does not prove the solids content reaching the sheet.
Suppose a well-mixed 500 kg bath contains 60 kg of dry solids. Its concentration is 12% by mass. If 20 kg of water evaporates without any other flow, the bath becomes 480 kg and its solids rise to 12.5%.
At an unchanged 60% retained-liquor pickup, the simple balance would predict 7.5% add-on. The result depends on that assumption. Viscosity and pickup may also change, so measure the actual output.
In a separate case, add 50 kg of water to the original bath. Its 60 kg of solids are now in 550 kg of liquor, or 10.91% solids. At the same assumed pickup, predicted add-on falls to 6.55%. These cases show opposite effects that a general instruction to “top up the box” could hide.
Check concentration and viscosity at defined conditions. Identify the cup, temperature and procedure when reporting a flow-cup time. Seconds from different cups are not interchangeable viscosity readings. Check any refractometer conversion against the actual recipe too. Its optical scale may not directly measure that blend's solids.
Keep the sample point consistent. Liquor in a poorly mixed corner may differ from liquor entering the nip. Record any water or fresh liquor added near the sampling time. Allow the agreed mixing and settling period before judging a change.
Locate the size as well as weighing it
Two warps can have the same dry add-on and different protection. One may have a fairly continuous surface film. The other may carry uneven deposits, leaving exposed fibres or joining nearby ends.
Surface coverage and penetration describe different features. Coverage concerns the outside of the yarn. Penetration concerns how far the material enters the fibre bundle. Increasing total mass does not specify how that mass is distributed.
Cross-sections or suitable surface examination can help investigate this difference. Use sample preparation and contrast suitable for the particular size. An image of one unusually neat section should not represent the whole beam.
Compare observations with an accepted warp of the same material and construction. Look for repeated exposed zones or large surface deposits. Check whether coverage differs between ends or changes after yarn separation. Relate these patterns to performance; an attractive image alone cannot earn a pass.
A 2026 study by Maatoug and Abu Nab assessed cotton warp treated with CMC made from wheat straw. It examined add-on, tensile behaviour, hairiness and abrasion under different machine settings. Those are distinct responses. The study does not provide a universal recipe beyond its material and test range.
If abrasion improves while shedding gets worse, retain both observations. The difference may help direct the next trial. Combining them into one undocumented “quality score” removes useful information.
Read machine settings in their physical context
A nip-pressure display describes the machine's control system. It may show hydraulic or pneumatic pressure. Another display may report total load or load per width. The machine's design and calibration determine how these quantities relate.
Do not compare a value in bar with one in newtons per metre as though they were the same setting. Check the manual, roller width, loading system and actual nip condition. A nominally unchanged display cannot rule out wear or unequal loading.
Speed also changes more than output. It affects time in the application zone and dryer. A change may alter retained liquor, drying and tension at the same time. Interpret the resulting warp with those linked changes in view.
KARL MAYER's PROSIZE description lists controlled wet transport and separate drying drives. These features are intended to manage stretch on that machine. The design description does not prove that a chosen setting will suit a particular mill's warp.
Record which zone changed. A beam-unwinding tension, wet-zone tension and winding tension act at different places. If a trial changes all three, a later result cannot be assigned to one of them without further evidence.
Use the approved machine procedure for adjustment and sampling. Retrieve a safely obtained, traceable sample. Do not take yarn by hand from a moving sheet to obtain a test result.
Inspect the beam that will reach weaving
The applied size does not correct a missing end or an incorrect draw-in order. CottonWorks' warp-preparation guide calls for equal tension and separated ends. It also identifies missing-end control as a preparation task.
Before release, check the beam identity and end count against the intended construction. Record repairs and their positions. Use the mill's approved inspection procedure to find crossed ends and bonded groups. Check for trapped waste and build defects too.
A beam that looks firm at its outer surface can still contain a problem from earlier in the run. Mark the length position of an interruption or repair. The loom team can then recognise the affected material. Keep the different size-box groups identifiable where the process allows.
Watch for yarns that need excessive force to separate, or film that breaks away during separation. Compare those observations with local add-on and coating results. Smooth beam winding does not prove that the ends will release cleanly in the shed.
Separate process stretch from lost breaking elongation
Process stretch compares the yarn's length before and after a stage. Breaking elongation measures how far a test specimen extends before it breaks. Their percentages answer different questions.
Suppose 1,000 m entering a stage becomes 1,012 m. Both lengths are checked under defined, comparable reference conditions. Process stretch is (1,012 − 1,000) ÷ 1,000 × 100 = 1.2%.
Now suppose matched tensile tests give an unsized breaking elongation of 5.5% and a sized result of 4.4%. The drop is 1.1 percentage points. Relative to the unsized value, the loss is 1.1 ÷ 5.5 × 100 = 20%.
The 20% is not a measured 20% extension through the machine. Nor does a 1.2% length change by itself predict the final breaking elongation.
ISO 2062:2009 covers single-end breaking force and elongation. It uses a constant-rate-of-extension tester and defines method choices and exclusions. Use the applicable full method and report sample condition. The public scope does not supply an acceptance limit for sized yarn.
A 2023 experiment by Abay and Ayele varied wet-zone tension, squeeze loading and speed. It measured several tensile responses in starch-sized rotor-spun polyester/cotton yarn. Such responses need to be assessed together. The experiment does not establish pressure settings that transfer to other setups.
Report breaking force and the linear-density basis used for tenacity. Added size changes mass, while process stretch can change mass per length. Keep the count basis clear. Do not silently switch between unsized and sized counts in a comparison.
Sample across the sheet and along the run
A centre sample taken at normal speed describes a different condition from an edge sampled as the line slows. Keep location and process state visible when comparing results.
The following invented readings come from five locations across one sheet. Suppose the trial team has chosen a local add-on range of 6.8–7.4% for this demonstration. It is an internal example, not an industry limit.
Sample location | Dry add-on | Within the example range? |
|---|---|---|
Left | 6.4% | No |
Left-centre | 7.0% | Yes |
Centre | 7.8% | No |
Right-centre | 7.2% | Yes |
Right | 6.6% | No |
The arithmetic mean is 7.0%, which sits inside the chosen range. Yet three local readings fall outside it, and the observed range is 1.4 percentage points. Reporting only the mean would conceal the reason to investigate.
Those five points are not a complete map of the sheet. Their mean is not automatically a mass-weighted beam average. Sample layout, end distribution and repeat results determine what can reasonably be inferred.
Keep start-up and restart material identifiable. Its application and drying history may differ from steady running. A setting change also takes time to reach the sampling point. Allow for the yarn path from the size box.
For a separate example, a 30 m path at a steady 60 m/min gives 30 seconds of travel. That does not include mixing delay in the box or settling of a control system. Use the actual line state before assigning a sample to the new condition.
Compare loom results on equal exposure
The final check is how the warp performs in the intended construction. Record confirmed warp breaks separately from weft stops, mechanical stops and planned checks. A total stop count cannot show whether sizing caused an improvement.
Consider two invented trials. They have the same warp end count and construction, with comparable loom conditions:
Measure | Trial A | Trial B |
|---|---|---|
Inserted picks observed | 1,800,000 | 2,000,000 |
Confirmed warp breaks | 36 | 30 |
Warp breaks per million picks | 20 | 15 |
Mean time per warp-break stop | 45 seconds | 60 seconds |
Warp-stop minutes per million picks | 15 | 15 |
Trial B has a 25% lower observed break rate: (20 − 15) ÷ 20 × 100. Its repairs take longer on average, however. Trial A loses 36 × 45 = 1,620 seconds, or 27 minutes. Trial B loses 1,800 seconds, or 30 minutes.
Divide each time loss by its pick exposure. Both lose 15 minutes per million picks to these stops. The break-count improvement alone has not reduced that measured time loss.
These figures do not describe all loom losses or prove a causal effect. Check the distribution of stop times and reasons. A few tangled breaks may need a different response from isolated clean end breaks. Fabric damage, shedding and restart marks also need separate assessment.
Account for warp exposure when end counts differ. More ends create more potential break locations. State the denominator and make the physical scope of the comparison clear.
Design the next trial around the unresolved difference
Choose a small set of plausible changes and define the response before starting. To test recipe concentration, retain the agreed yarn, machine route and test conditions. Measure pickup too. An unintended change in pickup can affect the result.
NIST's guide to randomized designs distinguishes treatment levels from repeated experimental runs. Randomizing a suitable run order helps avoid linking a treatment to time alone. Where different looms or yarn lots matter, plan comparisons within suitable groups.
Ten specimens from one application run provide information about that run. They do not create ten independent sizing trials. Repeat the application under the planned conditions so the result includes between-run variation.
A first trial may reveal that the measurement itself is unstable. Resolve that before chasing a small recipe effect. Otherwise, changing the process can create a persuasive pattern out of sampling noise.
Retain unsuccessful trials with their reasons. A recipe that gives acceptable add-on but poor yarn separation is useful evidence. It narrows the next question more than a pass label based on mass alone.
Include removal and downstream cost in the decision
The size must suit the finishing route as well as the loom. Confirm the actual ingredients with the supplier and agree how to assess removal. Starch, PVOH and blends need a suitable treatment. Do not assume that one treatment works for them all.
CottonWorks' dyeing-preparation guidance stresses identifying impurities and verifying their removal. Test the proposed sized fabric under a suitable trial. Examine residual material, absorbency and the later dyeing result. A generic article cannot supply a sound chemical recipe for that work.
Include wash-off demand and disposal or recovery requirements in the comparison. A lower application price can be offset by a difficult removal stage. Extra chemical cost may also be worthwhile. A controlled trial needs to show whether it reduces loss enough to justify that cost.
For example, increasing add-on from 7.2% to 7.8% on the same 500 kg dry warp adds 3 kg of dry size. At an invented CU 2.40 per dry kilogram, that is CU 7.20 extra material cost. The difference is 0.6 percentage points, or 8.33% more size relative to the first dose.
That calculation excludes water, heat, labour, recovery and cloth losses. It gives a clear starting amount for the business case. Measure the change in accepted cloth and total processing cost, then compare those results with the extra dose cost.
Read the product information before a chemical trial. ApparelDiary's guide to reading a safety data sheet explains how to use that document. Check the actual product and approved process, even when the generic name is familiar.
A useful final decision connects the dry amount and its distribution with yarn behaviour, loom loss and removal. Retain the recipe with the conditions that made it work. The next production team then has a tested starting point. Those records also help it recognise when the warp has changed.
Discussion0
Loading the discussion…