TL;DR: A COA that lists peel adhesion without specifying test substrate, dwell time, and peel angle is functionally useless for incoming inspection — demand the full test condition matrix before accepting any lot.
For a related specification, compare Double-Sided Flexographic Pressure-Sensitive Labels: Production Methods, Substrate Requirements, and Supplier… before finalising the packaging brief.
TL;DR: In our incoming inspection protocol, we reject PSL facestock rolls where caliper variation exceeds ±4% across the web width, because that tolerance directly triggers print pressure inconsistency on our flexo lines.
When a Supplier’s COA Passes — And the Labels Still Fail in the Field #
A brand partner running a cold-brew beverage line came to us after switching PSL suppliers mid-season. The new supplier’s COA looked clean: peel adhesion reported at 28 N/25mm, liner release force within spec, facestock caliper at 80 µm nominal. Three weeks after launch, labels were lifting at the leading edge on refrigerated glass bottles. Not peeling cleanly — lifting selectively, inconsistently, in a way that screamed adhesive cohesive failure under cold-humid cycling rather than simple low tack.
The root cause wasn’t the adhesion number. The adhesion had been tested at 23°C on stainless steel, 20-minute dwell, 180° peel. The application was 4°C HDPE with a 6-second machine dwell and 90° in-line peel. Same supplier, same COA, completely different performance outcome. This is the gap between a compliant COA and a qualified supplier.
Supplier qualification for pressure-sensitive labels isn’t a document review exercise. It’s a structured process of verifying that your supplier’s test conditions map to your actual application conditions — and that their internal controls are tight enough to hold those conditions lot to lot. Without that, a COA is just paper.
The COA Fields That Actually Predict Application Performance #
A complete PSL COA should cover at minimum: peel adhesion, tack, shear resistance, liner release force, facestock caliper, facestock tensile strength, coat weight, and moisture content. Each of those needs test conditions stated explicitly, not assumed.
Peel adhesion per PSTC-1 or ASTM D3330 must specify substrate (stainless steel, HDPE, glass, PP — these give different readings), dwell time (20 minutes is standard; shorter dwell reads lower), and peel angle (180° vs 90° can differ by 15–25% on the same construction). A COA that says “28 N/25mm” with no substrate listed tells you almost nothing.
Initial tack (often PSTC-6 loop tack or ASTM D2979 probe tack) matters most for high-speed automatic labelling lines above 200 labels/min, where dwell before the label reaches the product is under 2 seconds. If your supplier’s COA only reports 20-minute peel and omits tack entirely, that’s a red flag for machine-applied labels.
Liner release force is commonly overlooked. Per TLMI/PSTC-4, acceptable range for most matrix-strip applications is 0.05–0.25 N/25mm. Below that, liners pre-release during die-cutting. Above 0.30 N/25mm and you start seeing flagging on high-speed dispensing equipment.
Facestock caliper should be stated with tolerance. A nominal 80 µm BOPP facestock is acceptable with ±5% tolerance (76–84 µm) for most applications. For labels running through high-precision thermal transfer printers, the tolerance tightens to ±3%. Caliper variation across the web (not just at the roll edge) is the parameter that matters for print consistency — and it’s the one most COAs omit.
Coat weight for the adhesive layer typically runs 18–25 g/m² for permanent acrylic PSL constructions. Deviations below 16 g/m² usually show up as adhesion failure under stress; above 28 g/m² you get adhesive squeeze-out at die-cut edges, which contaminates the dispensing path.
| COA Parameter | Minimum Acceptable Detail | Common Supplier Shortcut | Consequence of Gap |
|---|---|---|---|
| Peel adhesion | Substrate + dwell + peel angle | Substrate omitted | Field performance unpredictable |
| Liner release force | Force value + test method | Method omitted | Matrix waste, flagging on dispenser |
| Facestock caliper | Nominal + cross-web tolerance | Edge measurement only | Print pressure inconsistency |
| Adhesive coat weight | g/m² + tolerance range | Nominal only | Squeeze-out or adhesion failure |
| Tack (loop or probe) | Method + value | Omitted entirely | Machine-apply failure at high speed |
The most commonly overlooked parameter in our qualification reviews is cross-web caliper uniformity. Suppliers measure caliper at roll edges. We measure at 5 points across the web width on every incoming lot, because that’s where flexo print impression variation shows up first.
Incoming Inspection Protocol — Pass/Fail Thresholds #
Our incoming inspection for PSL rolls follows what we internally call the IQC-L9 checkpoint sequence. It covers three layers: document verification, dimensional spot-check, and functional test sampling.
Document verification happens before the rolls enter our facility. We require a lot-traceable COA, an FSC chain-of-custody certificate where paper facestocks are specified, and a REACH compliance declaration for the adhesive system (confirming absence of SVHC substances above 0.1% w/w per REACH Article 57). For food-adjacent labels, we additionally require confirmation against EU Regulation No 10/2011 migration limits or FDA 21 CFR 175.105 as applicable. Lots arriving without complete documentation are quarantined pending resolution — not accepted provisionally.
Dimensional spot-check covers roll width (±0.5mm tolerance), core ID (76mm ±0.5mm for our dispensing equipment), caliper at 5 web positions (reject if range exceeds ±4% of nominal), and label pitch where die-cut rolls are supplied (±0.3mm tolerance for our automatic labelling lines).
Functional testing is sampled at AQL 2.5, Level II per ISO 2859-1. For a typical lot size of 50,000 labels, that means a sample of 200 labels minimum. We test peel adhesion on the same substrate as the end application — not stainless steel, unless the end substrate is stainless steel. We test at application temperature, not lab temperature, for any cold-chain or hot-fill product.
If a supplier consistently passes document review but generates functional test failures on 2 or more consecutive lots, they move to our Category B monitoring status — 100% functional testing on every incoming lot until 5 consecutive lots pass clean. Most suppliers self-correct before reaching that point, but we’ve maintained a handful of suppliers on Category B for over 6 months when they couldn’t stabilise coat weight.
Decision Framework for Qualification Depth #
If your label application is ambient temperature on a standard substrate (glass, rigid plastic, metal can) with conventional dispensing speeds below 150 labels/min, a thorough COA review plus AQL 2.5 incoming inspection is sufficient. Standard peel adhesion, tack, and caliper checks cover the primary failure modes.
If your application involves temperature cycling — cold-chain, hot-fill above 60°C, or outdoor UV exposure — you need qualification testing beyond the standard COA. Specifically: thermal stress cycling per ASTM D1183, minimum 5 cycles between -20°C and 40°C, and UV resistance per ISO 4892-2 (Xenon arc, 500 hours minimum for outdoor applications). Not every PSL supplier runs these in-house; ask for third-party test reports. If they don’t have them, that’s not a disqualifier by itself — but you need to fund the testing before approving the supplier, not after.
If your substrate is flexible or low-surface-energy (polyethylene bags, matte PP pouches, silicone-treated surfaces), the qualification requirement changes substantially. Low-surface-energy substrates read peel adhesion values 30–50% below stainless steel baseline on the same adhesive construction. A supplier who only tests on steel cannot give you a reliable performance number for these applications. We’d always request a sample run on the actual substrate, minimum 50 labels, before committing to a qualification decision.
One non-obvious recommendation: for any supplier producing both paper and film PSL on the same coating line, request coating line changeover records. Cross-contamination between adhesive batches at changeover is a real cause of lot-to-lot tack variation, and it’s rarely captured in a standard COA. Ask whether they run QC pulls after changeover — and what the acceptance threshold is.
Specification Notes for Brand Partners #
When you brief us on a pressure-sensitive label project, the information that moves qualification fastest is: substrate material and surface treatment, application temperature range (not just ambient — think storage, transit, and retail environment), dispensing method (manual, semi-automatic, or fully automatic with line speed), and any regulatory requirement on the adhesive system (food contact, cosmetics, REACH SVHC-free).
The gap we see most often in incoming briefs is the substrate surface energy. A brand will specify “clear PP pouch” without noting whether it’s corona-treated or untreated. On untreated PP, standard permanent acrylic adhesive at 22 g/m² coat weight may read as low as 12 N/25mm peel — well below the 18 N/25mm minimum we target for reliable field performance. A quick dyne pen test (target: surface energy above 38 mN/m for reliable PSL adhesion) would catch this before sampling begins.
Our standard sampling timeline for PSL qualification is 15–18 working days from receipt of confirmed substrate samples and complete brief. That timeline extends by 5–7 working days if third-party cold-chain or UV cycling tests are required. Rush sampling is possible but compresses functional test dwell times, which can produce optimistic peel results — factor that in.
Frequently Asked Questions
What should I do if a supplier refuses to share their full COA test conditions?
Treat it as a qualification risk, not an administrative inconvenience. Test conditions are not proprietary — they are standard method references (PSTC, ASTM, ISO). A supplier who won’t disclose them is either testing informally or has results that don’t look good under scrutiny. Require full method disclosure as a precondition of approval.
Our label volumes are small — is AQL 2.5 incoming inspection worth running for a 10,000-label lot?
At 10,000 labels, AQL 2.5 Level II under ISO 2859-1 gives you a sample size of 125 labels — roughly 30 minutes of lab time. The cost of a field failure on a consumer product (recall, repackaging, brand damage) is orders of magnitude higher. The sample size is small enough that there’s no practical reason to skip it for qualified lots.
How often should a qualified supplier be re-evaluated?
Our practice is annual re-qualification for active suppliers in stable supply, based on 12 months of incoming lot data. If a supplier changes adhesive formulation, coating equipment, or facestock source mid-year, we treat that as a requalification trigger regardless of the annual cycle. Ask specifically about formulation change notification — a good supplier will have a formal change control procedure and will notify you at least 60 days in advance.
Does FSC certification on facestock affect adhesive qualification?
FSC covers the facestock fibre chain of custody under FSC-STD-40-004, not adhesive chemistry. An FSC-certified paper PSL can still use an adhesive system that fails REACH compliance. The two certifications address completely different aspects of the construction. Treat them as independent qualification tracks and verify each separately.
Can I rely on a supplier’s self-reported tack value, or do I need to test it myself?
It depends on your dispensing speed. For manual application, self-reported loop tack above 8 N/25mm is generally adequate and low-risk to accept on the COA alone. For fully automatic lines running above 200 labels/min, we’d always run our own probe tack test on the incoming lot, because the dynamic tack behaviour at machine speed doesn’t correlate reliably with static loop tack values.
Planning a packaging project? Contact our team to request a complimentary specification review and sample quote.
