TL;DR: A flexographic printing supplier’s COA is only as useful as the pass/fail thresholds written into your incoming inspection protocol — without both, you’re auditing paperwork, not quality.
For a related specification, compare Digital Printing — Supplier Qualification Guide before finalising the packaging brief.
TL;DR: In our incoming inspection protocol, anilox cell volume deviation beyond ±0.3 BCM from the specified value triggers an automatic hold and re-measurement before any job approval.
What Failure Looks Like Before You’ve Placed a Single Order #
Three symptoms show up repeatedly when a brand has chosen the wrong flexo supplier — and none of them appear on a capabilities brochure.
The first is color drift across a production run. You see it on the finished reel: density shifts of 0.10–0.15 ΔE (CIE L*a*b*) between roll 1 and roll 8 of the same job, with no documented corrective action in the press log. The root cause maps to two places — either inconsistent ink viscosity control (typically ±5 cP vs. a required ±2 cP window), or worn anilox rolls with cell volume losses that were never measured against the qualified baseline.
The second symptom is adhesion failure on laminated flexible packaging. You peel a pouch and the ink layer separates cleanly from the film substrate. This maps to either incorrect corona treatment specification (surface energy below 38 dynes/cm at press time), wrong ink chemistry for the substrate, or no incoming surface energy verification in the supplier’s protocol at all.
The third is register inconsistency on multi-color work. Misregister above 0.25 mm is detectable by the human eye at normal reading distance and unacceptable for fine-print label copy or process-color brand imagery. When this happens repeatedly, the diagnostic usually lands on press mechanical condition or sleeve/mandrel concentricity — not operator error, which is where unqualified suppliers tend to assign blame.
Diagnostic table for initial supplier screening:
| Symptom | Likely Root Cause | Qualification Test to Request |
|---|---|---|
| Color drift across run (>0.10 ΔE) | Anilox wear, ink viscosity instability | Anilox cell volume audit + ink viscosity log |
| Ink adhesion failure on film | Low corona treatment / wrong ink system | Surface energy certificate + ASTM D3359 tape test results |
| Register error >0.25 mm on 6-color | Sleeve/mandrel runout, worn bearings | Press mechanical inspection report, TIR ≤0.02 mm |
| Pinholing on solids | Anilox screen too coarse for ink viscosity | Cell volume / line screen compatibility matrix |
| Ink rub-off after converting | Undercure (UV) or insufficient drying (water-based) | OD cure log or dryer temperature chart per job |
The Root Cause Most Qualification Audits Miss: Anilox Roll Population Management #
Suppliers will show you their newest anilox rolls during a factory visit. The rolls running on your actual jobs, three months after qualification, are a different story.
Anilox roll wear is cumulative and nonlinear. A chrome-engraved ceramic roll running at 800 lines per centimeter (lpi) will lose roughly 10–15% of its original cell volume after 500 million impression cycles under normal cleaning conditions. That sounds manageable. The problem is that the loss is not uniform across the roll surface — wear concentrates at the doctor blade contact band, creating a lateral gradient where cell volume at roll center is measurably lower than at the edges. On a 1,200 mm repeat width, this gradient can produce a visible density differential across the web.
We track our anilox population under what we call the AX-Registry protocol: every roll is assigned a barcode, measured by profilometer (per ISO 12647-6 ink density guidance for flexo) at three lateral positions and two circumferential positions on commissioning, then re-measured at 250-million-impression intervals. The measurement threshold is simple: if any single measurement point falls more than ±0.3 BCM below the roll’s original specification, that roll is flagged Category C in our registry and removed from fine-detail or process-color work. It stays available for coarse-screen solid coverage jobs where the tolerance is wider — but it never runs brand color critical applications without a documented deviation approval.
When auditing a potential flexo supplier, ask for their anilox roll registry. If they cannot produce a document showing each roll’s original cell volume, current measured volume, and last measurement date, they are managing anilox condition by visual inspection — which catches nothing until the customer complains. The measurement method matters too: profilometer is the only method that gives you actual volumetric data. Optical microscopy gives cell geometry but not volume. Suppliers who quote BCM values without a profilometer in their metrology lab are using estimated or manufacturer-quoted values, not measured ones.
The confirmation threshold: if a supplier’s anilox population shows more than 20% of rolls with undocumented measurement history, treat that as a disqualifying gap in their process control infrastructure.
Corrective Actions Ranked by Impact and Feasibility #
-
Request a COA with specific numeric pass/fail limits before sampling. A COA that says “viscosity: pass” is not a COA — it’s a checkbox. The COA must state the method (e.g., Zahn Cup #2, 23°C), the specified range (e.g., 18–22 seconds), and the measured result. This costs the supplier nothing to provide and immediately separates process-controlled operations from those running on operator judgment. Fixes roughly 40% of qualification ambiguity.
-
Run an incoming substrate surface energy check. Film substrates arriving without a surface energy certificate should be checked in-house with a dyne pen set before press release. The minimum acceptable surface energy for water-based flexo ink adhesion on BOPP is 40 dynes/cm; for PE, 42 dynes/cm. This is a 5-minute test that catches corona treatment decay from substrate aging. We hold any substrate lot below 38 dynes/cm regardless of the supplier’s dispatch certificate.
-
Specify AQL inspection level on ink adhesion. Reference ASTM D3359 Method B (cross-cut tape test) with a minimum pass rating of 4B for flexible film substrates. AQL 2.5 at inspection level II per ANSI/ASQ Z1.4 is our standard for incoming print quality checks on customer-critical jobs. This requires telling the supplier the acceptance criteria upfront, not after a batch fails.
-
Audit press maintenance records for doctor blade and impression roller replacement intervals. Doctor blades should be replaced on a documented schedule — typically every 8–12 hours of press running time for steel blades on ceramic anilox. Suppliers who cannot provide maintenance intervals are running reactive maintenance, which means press condition degrades unpredictably between jobs. This is a medium-cost audit item: it requires a half-day factory visit and access to maintenance logs.
-
Qualify ink formulation traceability back to batch number. Under REACH Regulation (EC) No 1907/2006, ink components used on food-contact packaging require documented substance traceability. Ask for the ink supplier’s safety data sheet and batch certificate for every colorant. If a flexo supplier cannot provide batch-level ink traceability for food-contact jobs, they cannot supply compliant product for EU or UK markets regardless of what their quality manual states.
Prevention — What to Specify Upfront #
The qualification document that prevents most downstream problems is a Flexo Press and Process Specification Sheet shared at the RFQ stage — not after sampling. Specify: substrate type and corona treatment window (minimum 40 dynes/cm for BOPP), ink system (water-based, UV, or solvent), color standards with ΔE tolerance (typically ΔE ≤1.5 for brand colors per ISO 12647-6), anilox specification per color station (cell volume in BCM, lpi), register tolerance (±0.2 mm for process color, ±0.3 mm for spot), and adhesion test method with pass rating.
Request the supplier’s anilox registry, press maintenance log, and a sample COA with numeric results before committing to a sampling run. These three documents reveal more than a factory tour.
Specification Notes for Brand Partners #
When you brief us on a flexographic printing job, the first thing we need is the substrate specification: film type, thickness in microns, and whether it’s pre-treated or requires corona at our end. These directly determine ink system selection and anilox specification — without them, we can’t give you a meaningful sample timeline or an accurate quote.
The brief gap that causes the most sample iterations is incomplete color specification. Sending a JPEG or a PDF is not a color standard. We need either a confirmed Pantone reference, a calibrated press proof, or a spectral target with an agreed ΔE tolerance. When we receive visual-only color references, we flag this at our intake review (what we call the JB-01 brief completeness check) before scheduling press time — a missing color standard adds one to two sample rounds on average.
Our standard sampling timeline for a new flexo substrate-ink combination is 15–18 working days from approved brief. If you’re introducing a new film substrate that we haven’t run before, add 5 working days for incoming surface energy validation and ink adhesion qualification.
What should a COA for flexographic ink include as a minimum?
At minimum: ink batch number, viscosity (method and measured value in cP or cup-seconds at a stated temperature), pH for water-based inks (typically 8.2–9.0), adhesion test result per ASTM D3359, and color strength vs. reference standard. A COA without measured viscosity and pH is not usable for press setup or incoming release.
Our current supplier sends color samples that look correct, but the production run drifts. Where do we start?
Start with the anilox roll. Ask your supplier for the cell volume measurement on the roll that ran your job and compare it to the original specification. A 10–15% volume loss from wear changes ink laydown enough to shift density by 0.08–0.12 ΔE without any change to ink formulation. If the supplier cannot produce a measured BCM value — not an estimated one — that’s the root issue.
Is ±0.25 mm register tolerance achievable on long runs of 100,000+ impressions?
It depends on sleeve and mandrel condition, not just press calibration. A press running with mandrel TIR above 0.03 mm will drift in and out of ±0.25 mm tolerance across a long run even if it holds tight at start. For sustained register control over 100,000+ impressions, ask specifically about sleeve concentricity verification and bearing replacement intervals — not just the press’s rated specification.
Do we need REACH compliance documentation even if our packaging is not food-contact?
Yes, for EU market entry. REACH applies to chemical substances in articles and mixtures regardless of food-contact status. Ink formulations containing SVHC (Substances of Very High Concern) above 0.1% w/w in the finished article require notification. Non-food packaging sold into the EU is not exempt. Request the ink supplier’s REACH compliance declaration specifically referencing the current SVHC candidate list, not a generic SDS.
Planning a packaging project? Contact our team to request a complimentary specification review and sample quote.
