Inline offset ink viscosity control during the press run — solvent corrected before density drifts off.

The problem

Viscosity climbs on its own during the run — and the hourly Ford cup arrives late.

In package and label printing — offset with liquid inks and coatings, flexo, gravure — the ink runs in an open circulation loop and solvent evaporates continuously. Viscosity rises, the ink transfers more pigment, and density moves throughout the run. At the press:

1. The drift is constant and silent — faster at higher speed, higher room temperature and larger exposed ink surface. It does not announce itself: it accumulates.
2. The Ford cup is manual and spaced out — it depends on the operator stopping to measure. Between cups, at high speed, a lot of material gets printed at a viscosity nobody knows.
3. The data goes unrecorded — cup seconds land on the press log at best; tied to the specific roll or sheet, never.
4. The correction is by feel — how much solvent to add is each operator's craft. Two shifts, two hands, two results.

The press room manager knows it; the quality manager knows it. The classic system works 99% of the time. That 1% is the run where the first thousand meters do not match the last — and the brand customer sees it on the shelf before anyone else.

How it fits the IRIS system

iLEAN does not replace the Ford cup — it captures every check and watches the drift between checks.

Viscosity control does not fail for lack of a method; it fails because every measurement is an isolated snapshot and nobody watches the movie. iLEAN acts as the putty that fills the gap between the cup, the inline viscometer if there is one, the press log and the ERP, without forcing you to change anything.

Connect captures the viscosity — from the inline viscometer or the Ford cup. The agent sees the drift per unit and warns before it leaves the range. The person corrects and signs — never the other way round.

The iLEAN pieces applied to in-run viscosity control:

• Edge — a terminal at the press. It reads the QR on the production order and identifies the run, the ink and the print unit. With a voice earpiece, the operator calls “unit 3, cup 22 seconds” and the data is tied in with no paperwork. It works without a network. If the plant loses its connection, it keeps capturing.
• Connect — if the press has an inline viscometer (rotational, vibrational or falling-piston), it reads the signal continuously; it also captures ink temperature, which moves viscosity as much as solvent does. Where there is only a Ford cup, it captures every called or timed check.
• Viscosity agent — builds the curve per print unit and watches the trend, not just the latest value. If the slope points out of range, it warns with the correction suggested by that ink's history on that press. The person executes and signs the correction; the effect is recorded for next time.

See the full IRIS architecture →

Before and after

Ford cup + press log vs. viscosity watched inline with iLEAN

Aspect	Ford cup + press log	With iLEAN (viscosity curve per run)
Measurement frequency	Hourly, workload permitting	Continuous (inline) or every cup, captured on the spot
Drift between checks	Invisible until the next cup	Trend per print unit in real time
Solvent correction	By feel, per operator's hand	Suggested from history, executed and signed by the person
Record per run	Press log note, if noted at all	Full curve tied to the order, with time and operator
Density drifting off	Seen on the printed roll, late	Warning before it leaves the agreed range
Veteran's judgment	In his wrist	Corrections recorded as a pattern per ink and press

Impact estimate

Impact estimate for your plant — to be validated with your numbers.

The block below is an estimate to be validated with the specific data of your plant. We put it forward so the committee has an order of magnitude; we refine it during the diagnostic.

• Package or label printing plant (offset with liquid inks and coatings, flexo, gravure), with viscosity control by Ford cup or an inline viscometer already installed.
• Pilot on one press (Edge terminal + voice cup capture or viscometer integration). First value expected within a few weeks.
• Indicative payback between 4 and 9 months, depending on meters/month, the cost of off-shade waste and the readjustment stops avoided.
• Hard levers: ≥ 20% reduction in waste tied to color and density; a viscosity curve per order deliverable to the brand customer; less dependence on individual craft for corrections.

And the quality manager's reasonable doubt

“What if the system misreads a cup check?” — false positives exist, and that is why the system never decides alone. Hallucination is a problem of free generation, not of anchored tasks. In tasks where the AI merely receives a value — the cup seconds or the viscometer signal — and compares it against an agreed range, the best models brought error below 1.5% ^[1]. And even so, what is critical is never decided alone: iLEAN warns and the operator or the quality manager sign. The three safety rings exist precisely for this.

^[1] OpenAI paper “Why Language Models Hallucinate”, 2025 — on the reliability of AI in anchored tasks.