Material Development

B2B Fabric Spec Sheets: The 3 Things Good Suppliers Never Hide (And Why It Costs You When They Do)

If a fabric spec sheet doesn't tell you the test method, it's not a specification. It's a sales pitch.

I've been reviewing fabric specs for seven years now. Carbon fiber, glass fiber, coated textiles, the whole range. In our Q1 2024 audit alone, I went through 200+ supplier datasheets across three product lines. And I can tell you with certainty: most spec sheets are designed to get past your purchasing department, not to tell you what you're actually buying.

That sounds harsh. But here's the thing—it's rarely malicious. It's industry laziness, institutional habit, and the comfortable assumption that "everyone knows what MPa means." Everyone doesn't. And the gap between what's printed on the sheet and what actually ships has cost me and my team more money than any price negotiation ever saved.

Let me give you a concrete example. Last year we received a batch of glass fiber fabric listed at 3,400 MPa tensile strength. Great number. We built our specs around it. When we tested incoming material against ASTMD 2343, we got 3,070 MPa. Nearly 10% lower. The vendor said, "Well, that's typical value, not minimum." Right. And it passed their acceptance criteria because 3,070 was "within standard deviation." That single gap cost us about $15,000 in rework and delayed a customer shipment by nine days.

If that spec sheet had disclosed three simple things, I would've caught it before the PO went out. I'm sharing them here so you don't learn the same way I did.

1. Test conditions. Every time. Without exception.

3,400 MPa sounds like an absolute. It isn't. That number only means something when you know the test standard, the specimen preparation method, fiber volume fraction, strain rate, and environmental conditions. Change any one of those, and the number moves.

I've seen suppliers quote "600 MPa tensile strength" for aramid fabric without specifying whether that's from a single-filament test (ASTM D3379), an impregnated strand test (ASTM D2343), or a laminated composite test (ASTM D3039). All three are "real" numbers for the same fiber. They just tell you completely different things about how it'll perform in your application.

Early in my first year on this job, I made the classic rookie mistake: I approved a carbon fiber spec sheet that just said "3,000 MPa." No test method. No tolerance. I assumed 3,000 MPa was 3,000 MPa. I learned the hard way that the vendor's 3,000 MPa came from single-filament testing, while our composite design used the laminated standard—which for the same fiber runs closer to 2,100 MPa typical. We had to scrap the layup schedule and start over. Cost us about $600 in wasted prepreg and a very uncomfortable conversation with our engineering lead.

Now I won't sign off on a spec unless it reads something like: "Tensile strength: 3,400 MPa typical value per ASTM D2343 at 23°C and 50% RH, tested at 10 mm/min crosshead speed." That's a spec. Everything else is marketing. And if a supplier can't produce that level of detail, they either don't know their own product or they're hoping you won't ask. Either way, that's a problem for your production floor, not theirs.

2. "Typical" vs. "Minimum"—the gap that eats your margin

This is the one most buyers miss. "Typical" means "this is what our best batches usually hit." "Minimum" means "this is the contractual floor we will not ship below." In practice, the gap runs 5-15% on carbon fiber and often 8-20% on coated or waterproof fabrics, depending on the resin system and the coater's process control.

That gap is everything when you're engineering to a performance floor. You design to 2,100 MPa laminated. Spec sheet says "typical 2,100." Supplier ships at "minimum 1,890." Your safety factor just quietly disappeared. On a 50,000-unit automotive run, that's not a rounding error—that's a recall waiting to happen.

Here's where I got lucky, and I want to be clear it was luck. I'd asked for both numbers on our last prepreg order because a colleague insisted on it. The supplier's datasheet listed 2,350 MPa typical and 2,100 MPa minimum. We designed to 2,100. First shipment came in at 2,120. Passed. Second shipment at 2,095. Failed. If we'd designed to the typical number like we did in early days, we'd have shipped that second batch straight into a structural application without knowing it was borderline. That was the one time the "it'll never matter" assumption didn't bite us. It's mattered every other time someone on my team skipped that check.

For glass fiber fabric OEM orders and waterproof/coated textiles, this is even more critical. Moisture management performance, hydrostatic head pressure, and tensile retention after weathering all have much wider production spreads than the spec sheet's headline number suggests. I insist on typical, minimum, and the sample size behind those figures. If a supplier presents "typical" as a single guaranteed value, I push back before the contract is signed.

3. Certification scope—the hidden cost nobody mentions

"Aerospace-grade carbon fiber" is a phrase that has cost more projects their launch dates than almost anything else I've seen.

Here's what happens: you buy "aerospace-certified" fabric. Sounds great. Datasheet says so. But when you trace the actual certification documentation, it covers the fiber in prepreg form only. You're receiving dry fabric. Or the cert covers a specific sizing that your supplier no longer applies. Or—this one happened to us—the aerospace classification applies to the fiber tow, not the woven fabric made from it. Different traceability requirements. Different testing protocol. Suddenly you're paying for additional qualification testing that nobody budgeted for.

I've seen this add roughly $18,000 to a project that was originally quoted at $95,000. And it didn't surface until month three, when our quality team asked for the certificate number and the supplier sent back a document with more asterisks than a legal disclaimer. When we pushed, the answer was, "Well, generally this is considered aerospace-equivalent." That "generally" cost us eight weeks and a customer escalation.

My rule now: I ask for certification scope in writing before the RFQ goes out. Specifically: what's covered, what's excluded, what testing has been performed against which standard, and whether that testing applies to the exact form of material being purchased. If the answer involves the word "about" or "typically" or "in most cases," I treat it as uncertified. That's not cynicism. That's just arithmetic.

"But isn't asking for all this too demanding? Wouldn't it slow down procurement?"

I hear this pushback all the time, mostly from people who haven't been on the receiving end of a failed shipment.

Let me be direct: requesting test conditions, typical/minimum values, and certification scope is not vendor abuse. It's basic due diligence. The suppliers who complain about it are usually the ones whose data sheets are thin for a reason. And here's the counterintuitive part—the suppliers who push back hardest on transparency are almost never the cheapest ones. They're just the ones who've never been asked before.

Toray and other major fiber producers publish detailed, standard-referenced specifications because their products are engineered for specific applications and they know that half-specs create more support headaches than upfront clarity. That's not unique to them, but it's a good benchmark: if a supplier's response to "what's the test standard?" is anything other than a clean, specific answer, that tells you something about how they'll handle a quality dispute next quarter.

Bottom line: a performance number without a test condition is just a number

I've reviewed enough failed batches and enough rework reports to know that the most expensive phrase in fabric sourcing isn't "premium price." It's "we assumed it met the same standard as last time."

Transparency in fabric specifications isn't about asking suppliers to reveal trade secrets. It's about asking them to put their own numbers in context—test method, statistical floor, certification boundaries. That's not a favor. That's a specification.

So before your next fabric order, ask one question: Where are the test conditions? If the supplier can answer that cleanly, you're probably dealing with someone who knows their process. If they can't, you've just identified a risk before it reached your production line. That's the best kind of quality control—the kind that happens before you've paid the invoice.

Peter Wekesa
Peter Wekesa

Peter Wekesa is a textile-reinforcement and prepreg analyst covering carbon, glass, aramid and basalt fabrics, woven roving, unidirectional tapes, multiaxial reinforcements, meshes, and resin-impregnated materials. He applies ISO 4602 yarn-count methods while measuring fibre areal weight, tow size, warp and weft density, orientation, crimp, resin content, tack, out-time, storage temperature, cured thickness, and defect rate. His engineering guides help composite fabricators, construction specialists, and mobility suppliers coordinate layup design, handling limits, traceability, process windows, and incoming-material acceptance.