Plates, Cylinders & Tooling — Troubleshooting & Failure Guide

What You’re Seeing on Press — and What It’s Actually Telling You #

Three failure symptoms come up repeatedly in our production reviews, and each one points in a different diagnostic direction.

Dot gain spiking mid-run. You set up the job at the correct press curve, the first pulls look clean, but by sheet 3,000 or metre 800 of a flexo run, shadow tones are plugging and highlight dots are starting to bridge. The substrate hasn’t changed. The ink viscosity is being held. Something mechanical is drifting.

Ghost banding at regular intervals. You see a faint repeat pattern — sometimes a tone shift, sometimes a partial image echo — appearing at fixed intervals across the web or sheet. The repeat distance is consistent enough to measure.

Register creep between colours. The front-to-back or colour-to-colour register that was within tolerance at makeready has walked 0.2–0.4mm by mid-run. On a four-colour folding carton job printed to G7 Master Colorspace standards, that’s already outside acceptable delta-E tolerance.

Each of these maps to a different root cause cluster:

The ghost banding interval is your fastest diagnostic tool. Measure the distance between bands in millimetres and divide by π (3.14159). If the result matches a cylinder or roller diameter anywhere in the print unit — within ±2mm — you have a rotating component fault, not an ink or substrate fault.

The Root Cause Most Tooling Reviews Miss — Plate Swell Under Solvent Load #

Of the three symptom clusters above, dot gain spike mid-run is the one that gets misdiagnosed most often. Our QC-11 failure classification log consistently shows it attributed to ink viscosity or anilox volume when the actual driver is photopolymer swell.

Here is the mechanism. Solvent-based flexo inks contain active solvents — typically ethyl acetate, n-propyl acetate, or toluene blends — that penetrate photopolymer plate material during printing. The swell rate depends on the solvent polarity, the plate formulation, and the contact time per revolution. On a narrow web line running at 150 metres per minute, a plate element contacts ink approximately 2.5 times per second. Over 30 minutes of continuous running, cumulative solvent absorption causes the plate relief depth to increase measurably. We have recorded swell of 0.04–0.09mm on standard analogue photopolymer plates running toluene-bearing inks after 45 minutes of contact time, using a digital calliper measured at the plate dot shoulder.

The consequence is not uniform. Fine highlight dots (1–3% at 150 lpi) swell proportionally more at their tips than large solids do, because the surface-area-to-volume ratio is higher. This means shadow tones hold relatively stable while highlights gain dot area at a rate of roughly 2–4 percentage points per hour of run time. The pressman sees plugging in the 70–80% tonal range and assumes the ink is thickening — so they add solvent to the ink, which accelerates the next swell cycle.

Thermal contribution compounds this. A plate running at press ambient of 32–35°C (common in our pressroom during summer production) will swell approximately 15% faster than one running at 22°C, per our internal baseline data from a controlled comparison run in Q3 2023 across 12 production jobs.

The measurement threshold for quarantine: if a calibrated dot area reading (using a FOGRA-compliant densitometer at the plate surface, not the substrate) shows the mounted plate printing above 2.5% dot gain relative to a fresh plate of the same type, remove it from press and inspect for swell. Plates that have absorbed beyond reversibility — indicated by a Shore A hardness drop of more than 4 points from the manufacturer’s specification — should not be remounted.

For UV-cured flexo systems, swell is not a significant variable. That calculus changes when you move to water-based inks on absorbent substrates, where different swelling dynamics apply and the failure mode presents differently.

Corrective Actions Ranked by Impact and Feasibility #

1. Requalify your plate material against your ink chemistry (high impact, low cost). Run a 60-minute immersion soak test using your actual press solvent blend and measure Shore A before and after per ASTM D2240. A drop of more than 3 Shore A points indicates incompatibility. This takes one day in a QC lab and costs nothing beyond technician time.

2. Switch to solvent-resistant digital photopolymer plate grades (high impact, moderate cost). Plates engineered for solvent resistance — typically with modified binder chemistry — show swell of under 0.02mm under the same 45-minute test conditions where standard plates show 0.04–0.09mm. The unit cost premium is roughly 20–35% per plate, but re-makeready time and substrate waste reduction over a production month typically recover the delta within 6–8 weeks on high-volume SKUs.

3. Implement a 45-minute plate rotation protocol (moderate impact, zero equipment cost). Keep a duplicate set of plates per station. Rotate off-press plates into a wash-and-dry cycle every 45 minutes of press running time. This fixes roughly 70–75% of swell-related dot gain cases in our flexo carton lines without any material change, though it requires an organised plate handling workflow.

4. Audit cylinder TIR at every planned maintenance interval (preventive, moderate cost). Using a dial gauge mounted in the print unit, measure TIR at three points along the cylinder face: both ends and centre. Our pass/fail threshold is 0.015mm. Cylinders exceeding this go to our tooling room for re-grinding before the next job. We run this check every 250 production hours per cylinder, logged under our PM-04 tooling maintenance schedule.

5. Inspect and torque plate clamps to specification before every makeready (low cost, often overlooked). Under-torqued clamps allow the plate leading edge to slip under the cyclic tension of impression, causing progressive register creep. The manufacturer’s torque specification for most plate-clamping systems is 4–6 Nm — we verify with a calibrated torque wrench at makeready, not by feel.

Prevention — What to Specify Upfront to Avoid These Failure Modes #

When you brief a job to us or any converter, the plate specification should be locked before artwork is released to prepress. Specify the ink system type (solvent, water-based, UV) explicitly — it determines plate material selection. Specify the target screen ruling and the minimum highlight dot percentage: jobs running below 3% highlights at 150 lpi require tighter plate exposure control and a harder plate grade to survive the run without dot bridging.

For gravure cylinder jobs, request a certificate of chrome plating thickness — we specify 6–8 microns of hard chrome on all our production cylinders, verified per ISO 1463 cross-section measurement. Below 5 microns, cell wall wear is measurable after 1.5 million impressions.

The document to request from your converter: a tooling qualification record confirming plate or cylinder TIR, Shore A (for flexo), chrome thickness (for gravure), and the date of last measurement.

Specification Notes for Brand Partners #

When you brief us on a new print job, the most common gap we encounter is an undefined ink system. Artwork sometimes arrives with a Pantone colour call-out and a substrate spec but no ink chemistry specified. That gap directly affects plate selection — a plate grade correct for UV flexo is the wrong choice for solvent-based flexo, and using the wrong grade is the upstream cause of most of the failure modes described here.

For a new flexo carton job, we need: substrate type and surface energy (dyne/cm), ink chemistry, target screen ruling, highlight dot minimum percentage, and run length. Run length matters because a 50,000-unit run warrants a different plate investment than a 500,000-unit run.

Our standard sampling timeline for a new flexo tooling set is 10–14 working days from confirmed artwork and approved prepress proof. Gravure cylinder tooling runs 18–25 working days. If a job requires specialty plate material (e.g., solvent-resistant grade for a toluene-ink application), add 5 working days for sourcing.

What’s the minimum run length where it makes sense to invest in hard-chrome gravure cylinders instead of flexo plates?

Gravure cylinder amortisation starts making economic sense at around 300,000–500,000 impressions for standard packaging formats, where cylinder durability (1.5–3 million impressions) offsets the higher tooling cost. Below that, the plate cost advantage of flexo is clear. Above 600,000 impressions on a repeating SKU, we typically recommend a gravure tooling review with the client.

If ghost banding appears after a cylinder re-grind, does that mean the re-grind was done incorrectly?

Not necessarily — re-grinding eliminates wear-pattern runout but can introduce new geometric error if the grinding spindle itself has bearing play. The correct check after any re-grind is a TIR measurement at three axial positions before the cylinder returns to press. If TIR reads below 0.015mm and banding persists, the source is upstream in the ink train, not the cylinder.

We’re seeing register creep only on the third and fourth colour units, not the first two. Does that point to a specific component?

Register creep isolated to downstream units almost always indicates thermal expansion in the sleeve or mandrel, not clamp slip. The first two units run cooler in the startup phase; by the time the third unit reaches thermal equilibrium (typically 20–30 minutes into a run), the sleeve has grown axially by 0.1–0.3mm depending on material and temperature delta. Specifying low-thermal-expansion carbon fibre sleeves instead of standard aluminium sleeves reduces this drift to under 0.05mm across the same temperature range.

Can you use the same photopolymer plate for both solvent-ink and UV-ink jobs on different runs?

This depends on the specific plate formulation, but in our standard practice the answer is no. Solvent-ink runs partially swell and plasticise the plate binder; once that has occurred, the plate’s dot geometry is altered and UV-ink performance on that same plate will show measurable dot gain increase versus a fresh plate. Our internal protocol (QC-11) classifies plates as solvent-service or UV-service after first use and does not allow crossover. The cost of a replacement plate is consistently lower than the cost of a re-run.

---

Planning a packaging project? Contact our team to request a complimentary specification review and sample quote.