The heat–pressure–time triple
- Heat: the jaw must reach the laminate's seal-initiation temperature and hold it across the full width.
- Pressure: enough to bring the layers into contact without squeezing product into the seam.
- Time (dwell): the contact duration; at higher speed, time falls, so heat or pressure must rise to compensate.
The three trade off. A laminate with seal-initiation at 125 °C sealed at 135 °C for 0.8 s holds 5.2 N/15 mm. Raise speed so dwell drops to 0.5 s and the same 135 °C gives 3.1 N/15 mm — a weak seal. You must raise temperature or pressure to recover it, and that is a machine-and-laminate co-decision.
Laminate structure sets the window
| Structure | Seal layer | Jaw temp | Dwell | Seal strength | Note |
|---|---|---|---|---|---|
| PET/PE | LDPE | 130–150 °C | 0.4–0.8 s | 4–6 N/15 mm | Common, easy |
| PET/AL/PE | PE | 140–160 °C | 0.5–1.0 s | 5–7 N/15 mm | Barrier, flat |
| Paper/AL/PE | PE | 150–170 °C | 0.6–1.0 s | 4–6 N/15 mm | Humidity sensitive |
| PP-based | CPP | 160–185 °C | 0.6–1.0 s | 5–8 N/15 mm | Higher temp |
| PE/PE mono | PE | 125–145 °C | 0.6–1.0 s | 3–5 N/15 mm | Recyclable, tight window |
Jaw design and temperature uniformity
A ±8 °C spread across the jaw shows up as a weak corner that passes the eye and fails the leak test. We specify cartridge heaters with a second thermocouple loop and target a spread under ±3 °C. Cooling under pressure matters too: releasing the jaw while the seal is still molten lets the layers relax and weakens the bond. Serrated jaws grip powders; flat jaws give a cleaner peel for easy-open sachets.
Defect atlas: read the seal, find the cause
| Symptom | Probable cause | Corrective action |
|---|---|---|
| Wrinkled / channel leak | Product in seal, low pressure | De-dust, raise pressure |
| Burn-through | Over-temp, long dwell | Lower temp, shorten dwell |
| Incomplete seal at end | Jaw temp spread | Rebalance heaters |
| Delamination | Wrong laminate / over-temp | Check structure, lower temp |
| Poor tear / no notch | Knife / perforation worn | Replace, re-set |
| Registration drift | Web tension | Re-tension, check mark |
Verification: how to prove the seal
- Peel strength: target above 4 N/15 mm with an acceptance band, not a single point.
- Burst / pressure-decay: hold pressure and measure decay; deterministic for pharma per USP <1207>.
- Dye penetration: cut and inspect for channels; cheap, destructive, good for setup.
- Vacuum decay: non-destructive leak test on a sample plan.
- SPC: chart seal strength per hourly sample; a drift warns before failures appear.
Anonymised case: EU food plant converting to mono-material
A Dutch snack maker moved from PET/AL/PE to recyclable PE/PE. The old 150 °C setting burned the thinner mono film; seal strength fell from 5.8 to 2.9 N/15 mm and leaks rose to 4 %. Resetting to 138 °C with 0.9 s dwell and a second thermocouple loop brought strength back to 4.6 N/15 mm and leaks to 0.3 %. Throughput dropped 6 % — the price of the recyclable structure.
Compliance
EU 10/2011 and FDA 21 CFR 177 for food-contact seal layers, CE under 2006/42/EC (transition to Regulation (EU) 2023/1230), ISO 9001 for the maker. Pharma sachets add EU GMP and USP <1207> deterministic leak testing in the OQ.
Procurement traps
- Buying on jaw width alone, ignoring temperature spread and cooling.
- No spare heater or thermocouple kit in the scope.
- No seal-strength tester specified, so you cannot verify at接收.
- Buying a machine that cannot hold the laminate's temperature window.
FAQ
Why did raising temperature cause burns?
You exceeded the laminate window. Heat, pressure and time trade off — recover a short dwell with a smaller temperature rise, not a leap.
What seal strength is acceptable?
Above 4 N/15 mm for most laminates, with a band, not a point. Pharma uses deterministic leak testing per USP <1207>.
How even must jaw temperature be?
Within ±3 °C across the active width. A wider spread produces weak corners that pass the eye and fail the test.
Why does my mono-material seal worse?
Its window is tighter. Re-set temperature and dwell to the PE/PE spec; expect a small throughput loss.
Dye or vacuum decay?
Dye is cheap and destructive, good for setup; vacuum decay is non-destructive for batch verification. Use both.
Does cooling under pressure matter?
Yes — releasing a molten seal weakens the bond. Hold pressure through crystallisation.
What spare parts actually matter?
Heaters and thermocouples, plus the knife/perforation set. Budget them by run hours.
Can one machine seal all my laminates?
Within a temperature band, yes. Across PET/PE to PP/CPP you need a wider window and re-validation per structure.