The Defect That Wasn't in the Printing Press

The Defect That Wasn't in the Printing Press

How Thinking Beyond the Obvious Solved a Complex LEP Printing Failure

Ronen Dolev | 2026

Four Dark Dots That Stopped an Entire Production Line

At a factory manufacturing laminate tubes for personal-care and hygiene products, a quality problem had remained unresolved for almost two weeks. Predominantly white tubes, printed using Liquid Electrophotography (LEP), were being rejected after small dark marks appeared on their surfaces, mainly near the longitudinal seam.

The defect was inconsistent. Hundreds of tubes might pass inspection before several new marks appeared. Some were black, some brown, and a few showed traces of color resembling the inks used in the printed image. From the end customer's perspective, however, the technical distinction was almost irrelevant: a dark speck on a clean white hygiene package looked like contamination.

The local team was convinced that the origin lay in digital printing. The assumption seemed reasonable. The marks appeared on a printed surface and apparently contained colored material. The investigation therefore began at the LEP press.

Engineers checked image formation and ink transfer. They repeated calibrations, examined printing conditions, reviewed ink adhesion, and adjusted parts of the surface-preparation process. When those actions failed to provide a consistent result, they increased corona treatment and revisited primer application, assuming that an unstable interface might be releasing printed material. The UV varnishing and curing process was also brought into the investigation.

Some interventions helped for a few hours. Each apparent improvement revived the hope that the cause had been found. Then another finished tube carried a dark mark.

Rejected material accumulated. Production time was lost, maintenance and quality teams disagreed on the next step, and the customer demanded answers. Remote troubleshooting didn't resolve the problem. An application expert from the printing press manufacturer arrived at the factory.

The Question That Changed the Investigation

The expert didn't start by changing press parameters. He asked to see the finished product.

Several rejected tubes were laid out on the inspection table. He studied the marks, rotated the tubes under the light, and compared their locations with the package geometry and the longitudinal seam.

Then he asked a deceptively simple question: "At what stage do we know, for certain, that these marks already exist?"

The production manager answered that they were found during final inspection of the tubes.

"That tells us where you discover them," the expert replied. "I want to know where they are created."

Until that moment, the team had been asking why the digital press was producing black dots. Nobody had actually demonstrated that the press was producing them. A mark containing ink did not prove that the ink had been printed at the position where the mark was eventually found.

The expert proposed rebuilding the complete production sequence, from substrate preparation to the finished tube. He was not looking for a new setting. He was looking for the first point in the process at which the defect could be proven to exist.

Stop Looking for a Culprit. Start Looking for Evidence.

The product was manufactured from an approximately 275-micron polyethylene (PE) laminate web, already constructed as the material that would eventually form the tube body. No separate post-print lamination stage occurred. The web received controlled corona treatment before priming, followed by LEP printing. A second surface-treatment step prepared the printed surface for the protective UV varnish, which was then applied, cured, and rewound.

Later, the printed and coated web entered a tube-forming machine. The material was formed into a cylindrical body, and the longitudinal seam was sealed by a combination of heat, pressure, contact, and cooling.

On a process diagram, each step appeared straightforward: surface preparation, primer, printing, surface preparation for coating, varnish, curing, rewinding, and tube forming. In reality, every layer had to survive not only its own manufacturing step but also the demands of the next.

The expert requested samples from several defined points: directly after printing, after UV coating and curing, after rewinding, immediately before tube forming, and after sealing. The team compared those samples rather than relying on photographs of rejected finished products.

The pattern was revealing. The printed and varnished web did not show the same dark spots seen on the finished tubes. No convincing link to a position in the artwork, no repeat interval pointing to a rotating press component, and no clear pattern running along the printed web. The marks became visible after tube manufacturing.

This did not establish that every meter of upstream material was perfectly clean. It did, however, substantially weaken the assumption that the LEP press was placing the marks directly onto the substrate. For the first time, the investigation began to move away from printing.

Everything Passed — and the Product Still Failed

Before shifting focus completely, the expert asked the team to establish a controlled production baseline. The corona-treatment conditions before priming were reviewed against the qualified working window. Primer application, drying, and ink adhesion were checked. The surface preparation before coating, applied varnish amount, and UV curing conditions were also documented.

There was no justification for assuming that more corona energy, more primer, or higher UV exposure must improve the result. Each process had an intended operating window. Moving outside that window could create additional problems without addressing the actual failure mechanism.

Fresh material was then printed under controlled, approved conditions. The print looked good. Adhesion results were acceptable, the coating appeared uniform, and curing checks met the defined requirements. The web was rewound and inspected. The characteristic dark marks were not present.

The material proceeded into the tube-forming machine. After production continued, the marks returned.

For some team members, this was discouraging. The latest tests had not eliminated the defect. To the expert, however, the result represented progress. The marks could be reproduced after a downstream operation even when the printing and coating stages met their qualified criteria.

The experiment had not solved the problem. It had shown that the original direction of the investigation could not explain all the evidence.

That is often the decisive step in a complex industrial failure: recognizing that a plausible explanation is incomplete.

The Detail Nobody Thought Was Connected

During a conversation with an operator from the tube-forming line, the expert heard a small detail absent from the troubleshooting reports. The operator mentioned that dark deposits periodically collected on one of the moving belts in the sealing section. He cleaned the material away during maintenance and had always regarded it as routine machine housekeeping.

The expert asked to see the belt.

At a safe inspection stop, they examined the upper conveyor belt that carried the formed tube through the heated longitudinal-sealing region. The belt helped provide contact and movement while the PE seam was exposed to the energy needed to form a reliable seal.

One feature immediately drew attention. The belt was wider than the functional seam itself. Part of its width extended beyond the intended sealing area and could contact printed, primed, and varnished regions of the tube body.

Near one edge, the belt surface was visibly rougher than in the central contact region. Dark material had accumulated on that roughened area. It was not loose dust. In places, the deposit looked cohesive and polymer-like, as though small layers had transferred, adhered, and hardened over repeated cycles.

Suddenly, the investigation took a new direction: what if the dots weren't generated during printing at all? What if material was being collected on the sealing belt and transferred later onto subsequent tubes?

The Dots Contained Ink — but Were Never Printed There

Understanding this possibility required looking at the printed construction as a complete system rather than treating ink, primer, varnish, and sealing as unrelated subjects.

The longitudinal seam needed thermal energy to bond the PE structure. But heat does not stop at the precise geometric boundary of a seam simply because the engineering drawing defines one. The belt can also expose adjacent printed and varnished surfaces to heat, pressure, and mechanical shear.

A UV-cured varnish may show excellent adhesion, abrasion resistance, and handling durability at room temperature yet behave differently under sustained thermal-mechanical contact. Depending on resin chemistry, crosslink density, flexibility, glass-transition behavior, and the underlying layers, elevated temperature may reduce its effective mechanical resistance or increase its tendency to transfer under shear.

The varnish does not have to melt to become vulnerable. A slight loss of resistance at a rough, hot contact interface may be enough for microscopic amounts of coating to adhere to the belt. Those deposits can also capture ink or material from the surrounding printed structure.

One tube contributes a little. Another adds more. The accumulation thickens and becomes less stable until a fragment breaks off or detaches, leaving a dark, apparently random mark on a later tube.

That mechanism could explain observations that had remained disconnected for two weeks. The marks were intermittent rather than tied to the artwork. Some contained different colors. The printed-and-coated web looked clean before tube forming. Cleaning the sealing belt sometimes produced a temporary improvement, but the defect subsequently returned as material accumulated again.

The central insight was simple but profound: a speck may contain printed material without being printed where it appears.

How Do You Prove a Mechanism Like This?

A persuasive theory is not the same as a verified root cause. The expert asked the team to establish whether belt deposits were actually linked to the finished-tube defects.

First, the team inspected printed and coated samples before tube forming. The team then examined tubes produced with the same construction, recording belt condition at the start of the run and at set intervals. The process operated within its normal qualified parameters; nobody deliberately imposed unsafe or extreme sealing conditions to force a failure.

With a clean belt, the characteristic transfer marks were initially absent. As production continued, deposits began to build near the rough belt edge where it overlapped the printed-and-varnished region. Later, dark spots again appeared on finished tubes.

The positions were compared carefully. The accumulation zone on the upper belt corresponded to the region where the marks were found on the tube body. Comparable contact locations elsewhere in the machine did not show the same pattern. Inspection of a removed belt revealed strongly attached deposits containing a polymeric-appearing material, mixed with colored residue in some areas.

The evidence supported a transfer-and-accumulation mechanism: material was collected from the printed construction during tube sealing, built up on the upper belt, and was released later onto the product.

The investigation had finally identified a credible failure path. Now the harder question remained: how could the factory eliminate it without weakening the seam or creating a different packaging failure?

The Easy Fix — and Why It Was Not Enough

Several immediate actions were available. The factory could increase belt inspection and cleaning frequency, replace worn belts, and maintain the lowest qualified sealing temperature that still ensured reliable closure. Each action provided temporary containment.

But none answered the underlying materials question.

More frequent cleaning would remove deposits but would not necessarily prevent new ones. A smooth replacement belt could improve contact conditions but would not show that the printed construction resisted the stresses of sealing. And reducing sealing temperature without validation could compromise bond strength, tightness, or product integrity.

The expert reframed the problem again:

Had the protective varnish been qualified for the thermal and mechanical conditions it actually experienced during tube manufacture?

Until then, the varnish had been evaluated mainly against printing-related and packaging-protection requirements: adhesion to the image, proper UV curing, appearance, abrasion resistance, and handling performance. Those criteria were important, but did they include repeated contact with a heated, moving, locally rough belt under pressure and shear?

The varnish itself might not have been defective. It might be performing exactly as specified for the conditions under which it was originally qualified. The specification, rather than the material's intrinsic quality, might be incomplete.

When the Solution Requires a New Materials Specification

The application expert and technical team worked with coating specialists to identify a UV-protective varnish whose thermal-mechanical performance suited the actual tube-forming route.

The objective was not to select the hardest coating or the one with the highest nominal temperature rating. The material had to deliver a balanced set of properties: adhesion to the printed construction, sufficient flexibility during forming, resistance to handling and abrasion, suitable curing behavior, appropriate controlled-migration performance for the intended use, and sufficient resistance to transfer during hot-belt contact.

A label such as "low migration" could not, by itself, demonstrate suitability. Migration compliance depends on the formulation, applied amount, curing history, set-off risk, construction, storage, intended contents, and relevant regulatory requirements. Likewise, a cured coating that passes a room-temperature rub test has not necessarily passed a combined heat-pressure-shear test.

The permanent corrective action was therefore not another adjustment on the LEP press. It was to qualify the protective coating as part of the complete packaging structure under realistic downstream manufacturing conditions.

That represented a fundamental change in the investigation. For nearly two weeks, the factory had asked which aspect of the printing process was wrong. It was now asking whether the material specification had omitted a critical post-printing requirement.

A Solution Is Not Approved by One Good Sample

The replacement varnish was not immediately available, so the plant continued controlled containment while a comparative trial was prepared. The test plan preserved the qualified substrate, primer, and print baseline. The candidate coating was applied and cured per the supplier's approved conditions.

The central comparison was the protective-coating route and its ability to withstand the thermal-mechanical loading seen during tube sealing. The printed and varnished web was converted under representative production conditions, without artificially reducing process demands.

The team monitored the upper belt, seam strength, coating integrity, ink adhesion, and finished-tube appearance. The belt was inspected over a duration long enough that the previous material would already have shown visible build-up.

This time, the upper belt remained clean. The familiar dark points did not appear. Further production trials reproduced the result, while quality checks confirmed that the sealing function and other essential requirements had not been sacrificed to obtain a clean surface.

Only after the complete construction had passed the relevant manufacturing and application checks could the candidate varnish be considered qualified for this use.

The evidence did not establish that the original varnish was a poor product. It established that the revised construction performed more robustly in the specific application and process window under investigation.

The Real Failure Was in the Definition of the Requirements

In hindsight, it would be easy to summarize the case as a varnish-selection problem. That would miss the larger lesson.

The difficulty was not simply choosing another coating from a supplier's catalog. The difficulty was recognizing that an important performance requirement had never been adequately represented in the qualification plan.

The printing team had checked the parameters it knew how to measure: image quality, adhesion, treatment level, primer application, and UV curing. The coating performed its intended functions during printing, inspection, and rewinding. The tube machine produced a longitudinal seal under its approved operating conditions.

But the interface between those processes had not been fully qualified. A coated printed surface extended into a zone of repeated contact with a heated moving belt. That combination of temperature, pressure, surface roughness, and shear created a demand different from what was anticipated when the varnish was originally selected.

Each department could point to records showing that its own process was operating within defined limits. The finished product still failed.

This is what makes some industrial failures difficult: the problem is not always inside a single process. It can be created at the boundary between processes, where one department's acceptable output becomes another department's unexpected input.

Thinking Outside the Box Is Not Looking for Magic

The expression "thinking outside the box" is sometimes associated with unusual tricks or clever shortcuts. In manufacturing, a more useful definition is the willingness to question the assumptions that frame the investigation.

In this case, the expert did not solve the problem by finding a hidden press parameter. He changed the central question.

Instead of asking why the ink created black dots, he asked when the dots first existed. Instead of assuming that ink-containing contamination had to originate at the digital press, he tested the possibility that printed material moved from one location to another during a later operation. Instead of optimizing the press indefinitely, he followed the package into tube forming and sealing.

The breakthrough also depended on listening to the converting-machine operator. That operator had seen the belt deposits repeatedly but had never connected them to the dots found at final inspection. The observation existed inside the factory long before the expert arrived. What was missing was the link to the customer's quality complaint.

The most valuable contribution of an external expert is not always an answer that nobody else could know. Sometimes it is the ability to ask a question that the existing investigation has stopped asking.

What This Investigation Teaches

Complex production failures do not always result from a defective component or an incorrect parameter. They may emerge when two individually acceptable operations interact under conditions that were not included in the original application specification.

When an investigation stalls, the team should consider whether it is repeatedly trying to confirm one favorite explanation with increasingly elaborate tests. A better question may be what evidence would show that explanation to be incomplete.

It is equally important to understand exactly what a successful test proves. Good ink adhesion does not establish thermal resistance. A successful UV cure does not automatically demonstrate shear resistance. Good room-temperature abrasion performance does not guarantee that a coating will survive contact with a hot moving belt. Each test measures a defined property under defined conditions.

The finished package, however, must survive the entire process sequence.

This is why material qualification should begin with the final application and work backward through printing, coating, forming, sealing, storage, and use. Not every interaction can be predicted in advance. When an unexpected failure occurs, the investigation must remain open to mechanisms outside the part of the process where the defect first became visible.

Conclusion: The Ink Was There, but the Press Did Not Print the Dot

For almost two weeks, the factory searched for the source of black spots in its digital printing process. Operators and engineers checked the image, adjusted familiar parameters, revisited primer and surface treatment, and verified the UV varnishing process. Some actions seemed to help. None produced a reliable solution.

The key insight was that the spots could contain material from the printed structure without being printed there. Material was being transferred to an upper sealing belt, accumulated under combined thermal and mechanical stresses, and later detached onto finished tubes.

The permanent solution required a protective-coating construction qualified for that downstream exposure, not another round of press adjustments.

The most important lesson is not about any particular varnish, ink, or sealing machine. It is about the way we investigate a problem that refuses to fit our expectations.

When a failure does not respond to repeated corrections, a better answer in the same direction may not be what is needed. Real progress may begin when we ask how the symptom could be created in the first place—and follow the product beyond the boundaries of the machine we initially blamed.

Experience tells us where to begin looking. Thinking beyond the obvious helps us find what others have overlooked.

In industrial printing, where success depends on an entire chain of materials and processes, that shift in thinking can mean the difference between weeks of trial and error and a validated engineering solution.

Author's Note

This article adapts a fictionalized professional case from "The Dots That Were Not Printed," The Personal Journey of an Industrial Printing Expert. It is an educational scenario, not a verified account of an identifiable customer or production site. The mechanisms described reflect established principles of LEP printing, UV coatings, thermal-mechanical behavior, and industrial converting. Any corrective action must be validated against the actual construction, equipment, process conditions, supplier guidance, and intended use.