Upholstery Fabric Specifications Buyers Misread

Upholstery fabric specifications fail in a particular way: the number is correct and the order still goes wrong. Your spec sheet says 450 g/m. The buyer's lab reports 310 g/m². Both are right. Nobody wrote down the width.

That is the whole problem with fabric specs. Four of them — weight, width, abrasion, repeat — look like single numbers and are not. Each one is either two different quantities wearing the same name, or a value that means nothing without a second value attached. Buyers who have been burned know this and ask. Buyers who haven't will accept your spec sheet, order, and then reject the shipment.

Here is what each number actually is, and what has to sit next to it.

Fabric Weight: Two Different Numbers, One Standard

Fabric weight comes in two forms that are not interchangeable:

Mass per unit area — grams per square metre (g/m², "GSM") or ounces per square yard (oz/yd²). A property of the cloth itself.
Mass per unit length — grams per linear/running metre, or ounces per linear yard. A property of the cloth at a specific width.

This isn't a folk distinction. The governing standard is titled to cover both: ISO 3801 is "Textiles; Woven fabrics; Determination of mass per unit length and mass per unit area" — two quantities, one document. The ASTM equivalent, D3776 "Standard Test Methods for Mass Per Unit Area (Weight) of Fabric", covers area only.

The conversion is trivial once you have the width:

grams per linear metre = g/m² × width in metres

A real mill spec makes it concrete. Camira's Xtreme publishes "Width: 140 cm minimum" and "Weight: 310 g/m² ±5%". At that width, 310 g/m² is 434 grams per linear metre. Quote either number alone and your buyer cannot cost the order.

Regulators write it the same way. California's furniture flammability bulletin specifies its own reference cover fabric as "WEIGHT/LINEAL YD.: 14.5 oz., (54 inches)" — the width is printed inside the same field as the weight, because the state's own lab could not use one without the other.

One more trap for anyone quoting in both systems: 1 oz/yd² = 33.906 g/m². And heavier is not automatically tougher. That same Xtreme is a fairly light 310 g/m², and its own spec sheet claims it is "independently certified to ≥100,000 Martindale cycles." GSM measures mass, not durability — they are separate specs, and buyers who treat weight as a proxy for wear are reading the wrong column.

Fabric Width: "140 cm" Is a Floor, Not a Fact

There is a standard for measuring width — ISO 22198:2006, "Textiles. Fabrics. Determination of width and length", which measures fabric "in a tension-free relaxed state." There is no standard that mandates a width value. 140 cm is convention, not regulation.

Notice how the mills actually publish it: Camira lists Xtreme at "140 cm minimum" and its Synergy Quilt Channel at "165 cm minimum". Minimum, because the roll varies and they will not be held to an exact figure.

Two consequences your buyer cares about:

Full width is not cuttable width. The measured width includes selvedge. How much of it your buyer can actually lay a pattern on is a separate question — and one that no standard answers for you. If you don't state usable width, you have told them nothing about yield.

54" and 140 cm are not the same width. 54 inches is 137.16 cm. 140 cm is 55.1 inches. Spec sheets routinely print "140cm (54")" as if the two were one number. That is a 2.8 cm error running the entire length of every roll — real yield loss, and an easy thing for a buyer's cutting room to catch after the container lands.

Abrasion Resistance: Martindale and Wyzenbeek Do Not Convert

This is the most expensive misunderstanding in the category, and it is worth being blunt about.

Two abrasion tests dominate. Wyzenbeek (ASTM D4157) reports double rubs. Martindale (ASTM D4966) reports cycles. Buyers in North America usually quote the first. Buyers in Europe usually quote the second.

There is no conversion between them. Not an imprecise one — none. The Association for Contract Textiles, which writes the performance guidelines the contract trade runs on, states it in four words: "There is no correlation between Wyzenbeek and Martindale results." A Camira technical paper on the subject puts it at more length: "There is absolutely no correlation between the tests, so it is not possible to estimate the abrasion results that would be achieved on one test if the results of the other test are already known."

Any conversion chart you find online mapping double rubs to cycles was invented by whoever published it.

It gets less precise still. ACT's own studies note that repeat abrasion tests on the same woven fabric "may vary significantly – as much as 60 percent or more." Camira tested one of its own fabrics 19 times and got results ranging from 32,000 to 78,000 cycles. An abrasion number is a band, not a point.

ACT's published thresholds — and note the tier names, because they are widely misquoted as "residential" and "commercial":

Application	Wyzenbeek (ASTM D4157)	Martindale (ASTM D4966, 12 kPa)
Low Traffic / Private Spaces, woven upholstery	15,000 double rubs	20,000 cycles
High Traffic / Public Spaces, woven upholstery	30,000 double rubs	40,000 cycles
High Traffic / Public Spaces, coated upholstery	50,000 double rubs	—

ACT has no residential tier at all. Both tiers sit inside contract textiles.

And there is a ceiling on the whole exercise. ACT requires licensees publishing results above 100,000 double rubs to carry this disclosure: "Wyzenbeek results above 100,000 double rubs have not been shown to be a reliable indicator of increased fabric lifespan." If a competitor is advertising 200,000 double rubs as a selling point, that claim is outside the range the test body itself will stand behind.

What to put on the spec sheet: the test method, the exact result, and the loading. "40,000 cycles, ASTM D4966, 12 kPa" is a specification. "40,000 rubs" is a number your buyer's lab cannot reproduce or check.

Pattern Repeat: A Published Spec, Not a Standardised One

Vertical and horizontal repeat drive fabric consumption on patterned goods — the cutting room has to align the pattern across panels, and every alignment costs yardage. On a plain fabric, consumption is close to a straight area calculation. On a large-repeat fabric, it isn't.

Be honest about the status of this one: no ISO or ASTM standard defines or measures pattern repeat. It is a per-product spec that mills publish with their own tolerance — Camira's Synergy Quilt Channel lists "Stitch Repeat 5 cm ±5%".

That's fine. It just means the number carries no external authority, so it needs to be stated precisely and with a tolerance, the way the mills state it. Give both directions, and say whether the fabric is railroaded — because a buyer who assumes the wrong direction has mis-costed every metre.

The Rest of the Spec Sheet

Beyond the four confusable numbers, these are the specs export buyers ask for by name, with the test methods they'll expect:

Spec	Method	Common requirement
Flammability (US)	California TB 117-2013, Section 1	Pass
Flammability (UK)	Furniture and Furnishings (Fire) (Safety) Regulations 1988	Cigarette and match tests
Flammability (EU)	EN 1021-1 / EN 1021-2	Smouldering cigarette / match flame
Crocking (colour transfer)	AATCC 8	Dry grade 4 min; wet grade 3 min (woven)
Colourfastness to rubbing	ISO 105-X12	Grade stated wet and dry
Lightfastness	AATCC 16, Option 1 or 3	Grade 4 min at 40 hours
Pilling	ASTM D3511 or D4970	Class 3 min
Breaking strength	ASTM D5034 (grab test)	50 lbs min, warp and weft
Seam slippage	ASTM D4034	25 lbs min, warp and weft

Two footnotes worth carrying over verbatim from ACT: on lightfastness, "there is no direct correlation between the numbers of testing hours and hours of service in the field." And on flammability, a cover fabric that fails TB 117-2013 can still be used "if a barrier (interliner) material that passes this test method is used" — which is why a failed cover test is a conversation, not the end of the order.

Get the Numbers Onto the Product Image

A complete spec sheet still loses orders, for a mundane reason: it arrives as a separate attachment, and the buyer is looking at your product photo. The photo is where the decision happens; the specification is in a file nobody opened.

The fix is to put the load-bearing numbers where the eye already is — the width and its direction marked across the actual roll or panel, the repeat marked at true scale, the abrasion rating and its test method as a label on the swatch image. That means measuring on the image rather than estimating: endpoints snapped to the real edge of the material so the marked width is the width, then exported at whatever size the buyer's platform or catalogue wants.

The distinction matters more here than in most categories. An AI-generated product image will happily render a plausible-looking fabric and a plausible-looking dimension callout, and the number will be invented. For a spec diagram, an invented number is worse than no number — it is a rejected shipment with your own document as the evidence. Deterministic measurement pins the real value; that's the only kind that survives a buyer's incoming inspection.

If wrong specs are already costing you, the return cost calculator will size what a rejection round-trip actually costs against the margin on the order. For a worked example of dimension labels on furniture listings, the furniture size-label case study shows the before and after.

Next Steps

Audit your current fabric spec sheet for bare numbers — any weight without a width, any abrasion figure without a test method, any width without "minimum" or a tolerance.
Pick one system per field and state it, then give the conversion rather than substituting it. Weight in g/m² and g/linear m at your stated width.
Add the test method next to every performance figure. "ASTM D4966, 12 kPa" is checkable; "40,000" is not.
Move the four decision numbers onto the swatch and product images, measured rather than estimated, so they arrive with the photo instead of behind it. The same discipline applies to the piece itself — see how product spec sheet fields map to what buyers actually read, and where spec sheet vs technical drawing draws the line between a buyer document and an engineering one.

FAQ

How do I convert GSM to grams per linear metre?

Multiply the GSM by the width in metres. A 310 g/m² fabric at 1.40 m wide is 434 grams per linear metre. The conversion is impossible without the width, which is exactly why a weight quoted alone is not a specification. The reverse is g/m² = grams per linear metre ÷ width in metres.

Can I convert Martindale cycles to Wyzenbeek double rubs?

No. The Association for Contract Textiles states plainly that "there is no correlation between Wyzenbeek and Martindale results," and Camira's technical paper confirms it is "not possible to estimate the abrasion results that would be achieved on one test if the results of the other test are already known." If a buyer asks for the test you didn't run, you have to run it — there is no valid conversion chart, and any you find online was made up.

What abrasion rating does upholstery fabric actually need?

Under ACT's guidelines, woven upholstery for low-traffic private spaces needs 15,000 double rubs (Wyzenbeek) or 20,000 cycles (Martindale, 12 kPa); high-traffic public spaces need 30,000 double rubs or 40,000 cycles. Above 100,000 double rubs, ACT requires a disclosure that higher results have not been shown to indicate longer fabric life — so extremely high advertised numbers carry less meaning than they appear to.

Is 140 cm the standard upholstery fabric width?

It is the common convention, not a standard — no standard mandates a width value. Mills publish it as a minimum (140 cm minimum, 165 cm minimum) because the roll varies. Also note that 54 inches equals 137.16 cm, not 140 cm, so writing "140cm (54")" as an equivalence introduces a 2.8 cm error along the whole roll.

Why does pattern repeat change my fabric consumption?

Because panels have to align, and alignment wastes yardage. A large vertical repeat forces the cutter to skip material between pieces to keep the pattern matching across a seam. Repeat has no defining ISO or ASTM standard — it is a published product spec — so state both vertical and horizontal repeat with a tolerance, and say whether the fabric is railroaded.

Upholstery Fabric Specifications: 4 Numbers Buyers Misread