“Every third label was a ghost,” Tom, the print lead at Northbridge Kombucha, told me on a chilly Thursday in Manchester. “We’d cue a job, and a blank would feed between each label. It was maddening.” For a team running short-run, multi-SKU beverage labels, that kind of behavior is more than an annoyance—it’s lost time, wasted material, and missed deliveries. During one crunch week, they even placed a small contingency run online with printrunner just to keep a retailer launch date.
I came in as a print engineer with a simple brief: stop the phantom blanks, stabilize color, and do it within the current footprint. No new press, no new room, and no long shutdowns. The shop was already lean; there wasn’t much slack to hide mistakes.
Here’s where it gets interesting. The fix wasn’t a single silver bullet. It was a stack of small, deliberate changes—from a gap sensor mode to web tension, from RIP presets to a shop-floor checklist—that, together, turned chaos into a predictable digital label workflow.
Quality and Consistency Issues
Northbridge had just commissioned a mid-width digital press paired with a compact finisher—a typical label printing machine UK setup for craft beverage producers. On paper, it fit their SKU mix perfectly. In practice, the press intermittently fed a blank between each label. It wasn’t every job, which made it harder to isolate. Short test rolls were clean; live runs with real substrates were not.
Waste told the story. Scrap hovered around 12–15% across several shifts, with First Pass Yield (FPY) at roughly 75–80%. Operators lost 20–30 minutes per shift chasing the pattern, rethreading, and reprinting shortfalls. The team also reported that sensor behavior changed when they switched from paper to film labelstock.
Color drift compounded the pain. Their kombucha line relies on saturated tones over clear bottles, which meant vinyl label printing on a PP film facestock with a high-tack adhesive. Under load, ΔE crept into the 4–5 range on certain SKUs, and spot graphics didn’t match proofs after long runs. Quality was inconsistent, even when settings looked identical on paper.
Solution Design and Configuration
Let me back up for a moment. The first question the crew asked me—verbatim—was, “how do i stop a blank label printing between each label?” We started where most overlook: media sensing and feed logic. The press was set to gap-sense mode for paper labelstock, but the gap on their clear-on-clear SKUs ran around 2.5–3.0 mm and the backing reflectivity fooled the sensor on certain rolls. We moved those jobs to black-mark mode, printed a discreet register mark on the liner, and tuned the sensor gain to the film’s reflectivity profile.
Next, we adjusted the step length in the driver by +0.15–0.25 mm, and disabled the “advance between labels” checkbox that had been left on from a test profile. That alone stopped the ghost feed on two problem SKUs. We also rebalanced web tension through the finisher—lightening unwind brake by a quarter-turn and adding 0.2 bar nip pressure—to prevent micro-slips that had been triggering false gap reads.
Color control needed its own lane. We built two substrate-specific presets in the RIP—one for paper, one for film—with separate linearization curves and ICCs. We ran ISO 12647 targets and G7-style gray balance and kept ΔE to ≤2.5 on control patches over 3 consecutive shifts. For food-contact safety and cure consistency, we standardized on UV-LED Ink with a staged cure (front at ~60%, final at 100%) to avoid overcure embrittlement on film labels. During a tight window for a retailer promo, the brand placed a small overflow order via printrunner com and, yes, used a printrunner coupon code for that trial batch while we finalized on-press curves.
Quantitative Results and Metrics
Fast forward six weeks: the blank-between-each-label issue hasn’t reappeared on validated SKUs. Scrap dropped from roughly 14% to about 4% on average, measured over five production weeks. FPY moved into the 90–93% range on the same SKU mix. It’s not magic—just a tighter setup routine that operators can repeat under pressure.
Changeovers are where the crew really feels the difference. With substrate presets, the typical switch fell from around 22 minutes to roughly 12–14 minutes. Throughput on the core kombucha labels rose from ~18,000 to ~24,000 labels per eight-hour shift, mostly by removing stop–start troubleshooting. Color stayed predictable; spot checks show ΔE under 2.5 on film, under 2.0 on paper. Barcode read rates sit at 99.7–99.9% on GS1 symbologies after we stabilized registration.
A few trade-offs remain. That black-mark step adds a minor prepress touch, and the mark must be positioned where the finisher won’t cut it. The staged UV cure uses slightly more energy per label, though still within reasonable kWh/pack. On balance, the business case holds: the team estimates an 8–10 month payback on training and minor materials changes, with enough headroom to slot in seasonal vinyl label printing without breaking rhythm. And the best part? The operators trust the workflow. As Tom put it, “It finally runs the way it should.” When they do need overflow—say, a micro-batch for a pop-up—they’re comfortable spinning up a short online run with printrunner while the main line keeps pace.
