Wash testing a garment before bulk production
A wash test answers a specific question: does this garment, made this way, still meet its specification and still look approved after the treatment you want to put on the label. It is a designed test with a defined procedure, not a garment put through a machine to see what happens.
Contents
The test has two outputs and both matter. One is numeric: how much the garment moved, in each direction, measured against benchmarks applied before the treatment. The other is visual: what the garment looks like afterwards. A garment can hold its measurements perfectly and still fail, because the print has cracked or the colour has moved.
This page covers how that is done and how the result is used. Why measurements move in the first place is a different question, covered in measurement change after washing.
Decide what you are testing before you test it
A test is only as useful as its brief, and four things have to be settled before a garment goes anywhere near water.
- The treatment. The instruction you intend to recommend, expressed as a procedure rather than as a label. A test proves a treatment; it does not discover one.
- The number of cycles. A single cycle shows the initial relaxation, which is usually the largest movement. Repeated cycles show whether it continues, whether trims keep recovering and whether decoration survives use rather than survives once.
- What is measured. The points of measure from the specification, so the result can be compared with the tolerance that was agreed.
- What counts as a pass. This comes from the buyer's specification or an agreed standard. There is no universal figure, and a supplier who offers one is describing a house convention rather than a rule.
Preparing and marking the specimen
The measurement is a comparison, so the before state has to be captured in a way that survives the treatment. That means physical benchmarks applied to the garment, in both directions, at known separations, and a set of measurements recorded against them.
Benchmarks, before and after
Benchmark pairs are applied in both directions and measured before and after treatment. Change is rarely the same lengthwise and widthwise, which is why both are recorded.
Direction matters as much as magnitude. A garment that loses length and gains width has changed shape, not just size, and a single overall percentage would hide that. On knitted garments the two directions routinely behave differently, which is why the result is reported per direction and per measurement rather than as one number.
Standard methods exist for exactly this. ISO 3759 covers preparing, marking and measuring specimens for dimensional change testing, and ISO 5077 covers how the change is then determined from the before and after measurements. Their role is to make the result comparable between laboratories rather than to define what an acceptable result is.
The treatment
For the result to mean anything to anyone else, the laundering itself has to be a defined procedure rather than a domestic wash. ISO 6330, currently in its 2021 edition, specifies domestic washing and drying procedures for textile testing, including reference detergents and ballast so that a wash is repeatable.
Two details in that are worth understanding as a buyer. Reference detergent is used instead of a consumer product because detergents differ and a test has to be reproducible. Ballast is added so the machine is loaded consistently, because a garment washed alone behaves differently from a garment washed in a full drum.
The drying step is part of the procedure and not an afterthought. Line drying and tumble drying produce different dimensional results on the same garment, so a test that tumble dries is not evidence for a line-dry instruction or the reverse. Whichever the label will recommend is the one that has to be tested.
After treatment the specimen is conditioned before being remeasured, because textiles take up and release moisture from the surrounding air and a damp garment measures differently from a dry one. Measuring too early is one of the more common reasons two parties disagree about a result on the same garment.
Reading the numbers
Compare the remeasured garment against the specification, not against the pre-test garment. Those are different comparisons and only the first one answers a commercial question.
| Question | Why it is asked separately |
|---|---|
| Is the garment still within tolerance? | This is the pass or fail. A garment that started at the top of tolerance and shrank is a different case from one that started at the bottom |
| Is length or width moving more? | Unequal change alters the shape and can change how the garment fits even when both measurements pass |
| Did the fit change? | Some measurements matter more than others. Body length and chest are usually felt; a hem circumference often is not |
| Did the trims recover? | Rib necklines, cuffs and waistbands can relax rather than shrink, which is the opposite failure and is not caught by looking only for shrinkage |
| Did anything twist? | Rotation is a change in the garment that measurements taken flat may not reveal at all |
Where the pattern can be corrected for a predictable, repeatable change, that is often a better answer than a more cautious label. Where the change is not repeatable, the problem is upstream in the fabric rather than in the care instruction, and incoming fabric inspection is where it should have been caught.
Reviewing the appearance
This is the half of the test that is often skipped, and it is where care instructions are often decided. A washed garment should be examined deliberately, against the approved sample, in consistent lighting.
What to look at on a washed garment
- Seam appearance, particularly puckering that was not there before, as covered in seam puckering
- Whether the body has rotated, which is the mechanism described in twisting and spirality
- Rib necklines, cuffs and hems: have they recovered, or are they slack, as in rib collar and cuff shape loss
- Prints: cracking, lifting at the edges, loss of density, change in hand feel
- Embroidery: distortion around the stitching, and whether the panel has puckered
- Colour: overall shade change against the approved sample
- Colour transfer, both onto the ballast or a test cloth and onto lighter parts of the same garment
- Surface: pilling, fuzzing, or a change in the fabric's hand
- Layered and coated products: any separation, blistering or change in the layers
- Trims and closures: corrosion, distortion, or colour bleeding from a trim into the fabric
Colour transfer within a single garment deserves particular attention on anything combining a dark and a light colour, because it is invisible until the first wash and is exactly the sort of thing a warning on the label exists for.
What the result decides
A test produces one of four outcomes, and only the first is a simple pass.
| Result | What follows |
|---|---|
| Everything within tolerance and appearance acceptable | The proposed instruction is supported. Record it against the approved sample |
| Measurements out of tolerance, predictably and repeatably | Consider correcting the pattern rather than the label, then retest |
| A specific component failed | Change the component, or make the instruction more cautious. Which of the two is a cost and product decision |
| Something a consumer might reasonably do would cause damage | A warning is needed as well as an instruction, which is a requirement in its own right in the United States |
The fourth outcome is often overlooked. A garment that is safe when washed as directed can still need a warning against something else a reasonable consumer would try, and identifying that is part of reading the test rather than a separate exercise.
Where testing sits in the schedule
Care testing competes with label lead times, which is the practical difficulty. Woven and printed labels are ordered ahead of bulk, so the instruction is often needed before the pre-production sample exists.
The workable compromise is to test early on the most representative sample available, using the actual fabric, colour and decoration wherever those are settled, and then confirm on the pre-production sample. What that avoids is the situation where labels have been printed and the confirmation test then says something different.
Where formal testing to a named standard is required by a customer or a market, it is specified as part of the development programme and carried out by a laboratory to a written brief. That brief, rather than the label, is where the treatment, the cycles and the pass criteria are fixed.
Questions this raises
How many wash cycles should a garment be tested through?
It depends on what you need to know and it should be agreed rather than assumed. A single cycle shows the initial relaxation, which is usually the largest movement. Repeated cycles show whether change continues, whether rib trims keep recovering, and whether decoration survives ongoing use rather than surviving once. Customer standards often specify a number.
What dimensional change is acceptable?
There is no universal figure. Acceptable change comes from the buyer's measurement specification and its tolerances, or from a standard the parties have agreed to work to, and it can legitimately differ between measurements on the same garment. Agree the criteria before testing, alongside the measurement tolerances.
Why is a garment conditioned before it is remeasured?
Textiles take up and release moisture from the air around them, and a garment that is still damp measures differently from a dry one. Conditioning before remeasurement removes that variable and is a common reason two parties get different numbers from the same garment.
Can I test the fabric instead of the garment?
Fabric testing is useful and is not sufficient on its own. It tells you how the material behaves, but not how the assembled product behaves with its decoration, trims, linings and seams. Where a garment has several components, the evidence has to support the instruction for the finished garment.
Does the drying method have to match the label?
Yes. Line drying and tumble drying produce different results on the same garment, so a test that tumble dried is not evidence for a line-dry recommendation or the other way round. Whatever the label will say is what has to be tested.
Agree the test before the sample
Tell us the treatment you want to recommend and the tolerances the garment has to hold. We will produce the sample and report both the measurements and the appearance against your approved reference.
Start a project