Sachet Packing Machine Selection Starts With the Seal Geometry, Not the Rated Speed

The Core Engineering Problem

The leaker rate on a sachet is set by the interaction of three variables: the seal geometry, the product physical state, and the timing of the dose relative to jaw closure. On a powder sachet the product sits as discrete particles and rarely bridges into a continuous film, so a 3-side seal — formed by two transverse seals and one longitudinal seal along a folded top edge — holds because there is nothing to track into the seal land. On a liquid or suspension sachet the product is continuous, and the instant the dosing nozzle retracts with a hanging droplet, that droplet or a capillary film reaches the seal land before the jaw closes. The longitudinal seal then cures around trapped liquid, and you get a micro-channel leak that passes a 0.5 bar pressure test only sometimes. The core problem is therefore not "bad sealing" — it is product in the wrong place at the wrong time, and the geometry has to prevent that, not recover from it.

I quantify this on every commissioning with a three-method integrity check: internal pressure at 35 kPa held 10 s, helium trace at 5 ppm sensitivity, and a dye-penetration visual on a 2,000-sachet batch. Anything above 0.5% on that batch is a validation fail in my book, and EU GMP Annex 1 leak-test expectations make that a hard stop, not a footnote. The numbers below are from those tests, not from supplier claim sheets.

How the Machine Types Actually Differ

The market labels everything a "sachet machine," but seal-forming mechanics and dosing platforms are two separate axes that have to be specified as one package. Here is the format comparison that actually changes your leaker rate.

*after suck-back and air-knife fit. The dosing axis matters as much as the seal: auger for powder, piston or peristaltic for liquid. A suspension needs a peristaltic or piston pump with an in-line agitator because the solid phase settles at 30–120 cP and a static nozzle dumps inconsistent phase. I have seen a suspension sachet swing from 8% to 14% active content because the pump drew the clear supernatant off the top of a settled tank.

Field Data From Real Installations

These are measured results from commissioning batches, not catalog figures. The dosing accuracy is reported as CV (coefficient of variation) over a 500-sachet sample.

Seal-zone temperature profile for the corrected suspension line (flat-web 4-side, PET12/AL9/PE60 laminate):

Commissioning case: Casablanca, Morocco

Product was an antacid oral suspension filled into a 10 ml aluminum-laminate sachet on a 4-side seal format at ~120 sachets/min. The line ran in a room at 21°C and 38% RH. The first validation batch showed 2.6% seal leakers concentrated on the longitudinal seal. Root cause was liquid wicking into the seal zone before jaw closure: the piston nozzle retracted leaving a 0.3 ml hang-drop that tracked to the seal land at 118 sachets/min, and the longitudinal jaw cured around it. The fix was three changes done together — a seal-zone air knife at 0.4 bar through a 12 mm nozzle with an 80 ms pulse, a nozzle suck-back of 0.25 ml, and a stepped jaw profile (3.5 mm land plus a 2 mm relief). Leakers dropped to 0.3% and dosing CV held at ±1.0%. No laminate change was needed; the geometry and timing were the whole problem.

Where Buyers Get It Wrong

1. Wrong seal geometry for a fluid product. Consequence: 6–9% leakers on a 3-side liquid sachet that no amount of temperature tuning fixes, because the open top fold is a permanent wick path. Avoid by specifying 4-side for any product below 500 cP that is not a dry powder.
2. No suck-back on the liquid nozzle. Consequence: the hang-drop bridges into the seal land on every cycle, giving intermittent leakers that pass PQ on Monday and fail on Thursday. Avoid by demanding a suck-back of 0.2–0.3 ml as a standard feature, not an option.
3. Laminate mismatch. Consequence: a PE sealant layer under 50 µm will not melt consistently against the jaw, and an AL layer under 7 µm tears on the transverse shear. Avoid by locking the laminate spec (e.g. PET12/AL9/PE60) into the URS before the machine is quoted.
4. Dosing pump wrong for suspension. Consequence: an auger in a suspension clumps and a static pump draws supernatant, swinging active content 8–14%. Avoid by pairing suspension with a peristaltic or piston pump and an in-line agitator at 20–40 rpm.
5. Ignoring the seal-zone temperature versus dwell balance. Consequence: over-temperature at 175°C degrades the AL layer and you pass seal but fail barrier; under-dwell at 200 ms leaves a cold seal. Avoid by qualifying dwell at 380–450 ms and surface temp at 159–164°C on the final jaw.
6. Specifying speed without a fill-time budget. Consequence: a 150 sachets/min claim collapses to 95 when the peristaltic dose of 10 ml needs 480 ms. Avoid by calculating dose volume divided by pump rate and confirming it fits the cycle time before signing.

Meeting Regional Compliance

European Union

For primary pharmaceutical packaging the sachet laminate and the machine surface contact must satisfy EU GMP (Annex 1 for sterile-adjacent risk) and ISO 15378 for primary packaging materials. The seal-integrity test data must be retained under EU GMP record rules, and any electronic batch record needs 21 CFR Part 11-style audit-trail discipline if the line reports to a global system. CE marking under the Machinery Directive 2006/42/EC is mandatory for the electrical and guarding package.

United States (FDA)

FDA 21 CFR Part 211 governs the dosage accuracy and the integrity of the filled sachet; Part 11 covers electronic records if you log dose weights. A liquid or suspension sachet for OTC use still needs demonstrable seal integrity, and the leaker-rate acceptance limit should be written into the validation protocol before FAT, not argued at PQ.

Southeast Asia

Indonesia (BPOM), Thailand FDA, and the Philippines FDA each accept ISO 15378 and GMP-aligned documentation, but they differ on sample sizes for stability and on laminate migration limits. I advise qualifying the laminate for heavy-metal and solvent migration per each market's food-contact or pharma-contact list rather than assuming one certificate covers all.

Middle East

Morocco's DMP and the SFDA export pathway for the GCC require the Certificate of Analysis to reference the actual leaker-rate batch test, and many Gulf tenders ask for ISO 15378 plus a halal or local registration dossier. The Casablanca case above was built against DMP sampling expectations from the start, which is why the 0.3% result was accepted on first submission rather than after a re-test loop.

Specification & RFQ Checklist

• Product state and viscosity range (cP) and whether a suspension settles.
• Seal geometry decision: 4-side for liquid/suspension, 3-side for powder — stated explicitly.
• Dosing system: piston/peristaltic with suck-back 0.2–0.3 ml and in-line agitator if suspended.
• Laminate structure locked in the URS (PET/AL/PE micron values).
• Seal land temperature and dwell set points, with measured-surface verification method.
• Integrity test method: 35 kPa/10 s pressure, helium trace, dye-penetration batch size.
• Cycle-time budget proving dose volume fits the target sachets/min.
• Compliance mapping: ISO 15378, EU GMP / FDA 21 CFR 211, regional authority.
• Changeover time between powder and liquid formats, and spare jaw sets.

Questions Buyers Ask Before Signing

Can one sachet machine run both powder and liquid? Technically yes with format change, but I do not recommend it for validated pharma lines — the powder auger and the liquid piston share a product contact path that needs full disassembly and re-clean between states, and cross-contamination risk is real. Run two dedicated machines.

When do I choose 3-side versus 4-side seal? 3-side only for free-flowing powder and granules where there is no continuous film to wick. 4-side for every liquid, suspension, gel, or paste. The 4-side flat-web longitudinal seal is the only geometry that lets you put an air knife and suck-back on the fill path.

How do you prove a sachet does not leak? A single bubble test is not enough. I use pressure hold at 35 kPa for 10 s on a sample, helium trace at 5 ppm for micro-channels, and dye penetration on a 2,000-sachet commissioning batch. The acceptance limit should be below 0.5%.

What is special about suspension handling? The solid phase settles, so you need an in-line agitator at 20–40 rpm and a pump that draws a representative phase. A static tank or a peristaltic tube that sits at the bottom will drift active content by double digits percent over a run.

Why did my liquid sachets leak only sometimes? Intermittent leakers are almost always the hang-drop reaching the seal land before jaw closure. The fix is suck-back plus an air-knife pulse timed to nozzle retract, not a higher seal temperature.

How fast can I realistically run a 10 ml suspension sachet? If the peristaltic dose needs 480 ms, your mechanical cycle is capped near 120–125 sachets/min regardless of the brochure's 200/min claim. Budget from the dose, not the rating.

What laminate do I specify for an aluminum sachet? For a liquid or suspension I lock PET12/AL9/PE60 — the 60 µm PE gives a stable sealant melt, the 9 µm AL keeps the barrier, and the 12 µm PET handles web tension. Thinner AL tears on transverse shear.

Written by Helen Xu | Chief Industrial Application Engineer

Helen Xu is a Chief Industrial Application Engineer with 9 years of specialized experience in packaging machinery and liquid filling machine design, equipment model selection, and full production line process optimization. He focuses on delivering customized packaging & filling solutions for pharmaceutical, food, and chemical manufacturing industries, with mature practical expertise in GMP compliance, ISO 9001 quality management standards, and turnkey large-scale filling & packaging production line integration.