The Chocolate Factory Glue Problem

A confectionery packaging line in Shanghai was producing rigid boxes for chocolate gift sets. The boxes looked perfect coming off the forming machine, but within 48 hours of storage at 18°C, the glue seam on roughly 7% of cartons began to separate — a hairline gap of 0.3 to 0.5 mm along the side seam. The buyer rejected two pallets. The production manager’s first instinct was to increase glue application volume. That made the problem worse: excess glue squeezed out beyond the seam edge, creating visible residue, and the seam still separated because excess glue does not equal better adhesion.

The actual problem was a 12°C glue temperature deviation at the application nozzle combined with a pump pressure setting 0.3 MPa above the specification for the 300 g/m² coated board they were using.

The Four Parameters That Matter

Hot-melt glue application on a carton forming machine is controlled by four interdependent parameters. Changing one without checking the others is the source of most glue-related defects.

Temperature (Setpoint vs Actual). The HMI displays tank setpoint, not nozzle temperature. With a 2-meter heated hose on Nordson ProBlue systems, the nozzle runs 8-12°C below the tank. For EVA hot melt, optimal nozzle temperature is 150-160°C. Below that, viscosity rises and penetration suffers. Above that, adhesive degrades and chars inside the nozzle. Verify with a contact thermocouple — don’t trust the tank display alone.

Pump Pressure. For uncoated board, 0.8-1.2 MPa is typical. For coated board (clay, UV, laminated), reduce pressure by 0.2-0.3 MPa — the coating limits absorption, and excess pressure forces glue sideways. Shanghai was running 1.2 MPa on 300 g/m² coated board; the correct setting was 0.9 MPa.

Nozzle-to-Board Distance. Keep the nozzle 1.5-2.0 mm from the board surface. Closer risks contact and smearing; further lets the bead skin over before bonding. Measure with a feeler gauge during a dry cycle, not during glue flow.

Open Time. EVA hot melt at 150°C has a 2-4 second open window to bond the second surface. On the ZF-420 at 30 ppm, glue-to-compression time is ~1.8 seconds — safe. At 15 ppm, it stretches to 3.6 seconds, approaching the limit. Below 15°C ambient, open time shrinks further. Use a faster-setting grade for low-speed or cold-environment production.

Diagnosing Glue Overflow

Glue overflow (squeeze-out beyond the seam edge) has three possible causes, listed in order of frequency:

1. Excess pump pressure. Reduce pressure by 0.1 MPa increments until the bead width at the seam is 1.0 to 1.5 mm. Measure bead width on 10 consecutive cartons and take the average.

2. Nozzle orifice wear. A worn nozzle produces a wider, less controlled bead. After approximately 500 operating hours, the nozzle orifice can erode by 0.05 to 0.10 mm, which increases bead width by 20 to 30% at the same pressure setting. Inspect the nozzle under 10x magnification. Replace if the orifice is visibly oval or if the bead width has increased without any pressure change.

3. Incorrect nozzle angle. The nozzle should discharge perpendicular to the board surface. An angled nozzle (more than 5 degrees off perpendicular) produces an asymmetric bead that is wider on one side. Check with a small machinist’s square held against the nozzle body.

Diagnosing Uneven Output

Uneven glue output (variation in bead width or intermittent gaps) points to different causes:

• Check the tank filter. A partially clogged filter (replace every 200 operating hours or when pressure drop across the filter exceeds 0.3 MPa) causes pressure fluctuation at the nozzle.
• Check for char in the system. If the glue tank has been held at temperature for more than 8 hours without production, adhesive can begin to degrade and form carbon particles. These particles intermittently block the nozzle. Purge the system: run the pump at 50% pressure with the nozzle removed until clean adhesive flows, then reinstall the nozzle.
• Verify hose heating. A failing heater element in the heated hose causes cold spots. The hose temperature should be uniform within ±5°C along its entire length. Use an infrared thermometer to scan the hose surface every 20 cm.

Daily Glue System Checks

• Tank temperature stable at setpoint ±3°C before starting production
• Pump pressure gauge steady (needle should not oscillate more than ±0.05 MPa)
• Visual inspection of glue bead on first 5 cartons: continuous bead, no gaps, no squeeze-out beyond 1.5 mm
• Nozzle tip wiped clean with a brass brush (never steel — it scores the orifice)
• Purge cycle run for 30 seconds at shift end to clear the nozzle