Cosmetic Jar Manufacturer: Why Jars Leak in Air Freight
We are a cosmetic jar manufacturer, and this failure repeats every season. A jar leaking in air freight is rarely a torque problem. The liner stops sealing when cabin pressure drops. It is not a loose cap. The gap closes again on landing. That is why the leak never shows up in your warehouse.
Four things cause it. Five tests find it before you ship.
1. What fails at altitude
A cargo hold sits well below sea level pressure. On the ground, inside and outside are the same. In the air, they are not.
The jar becomes a small pressure vessel. It pushes outward on every joint. The weakest joint is the liner edge.
Land the plane, open the hold, and the pressure equalises. The liner settles back. The seam looks intact. The product is already in the carton.
This is the whole mechanism. Everything below is a variation on it.
2. Four causes, ranked by how often we see them
Compression set in the liner. Foam takes a shape under load and keeps it. The first torque flattens it to the right thickness. By the third closure it no longer returns. A factory test passes on the first closing. Not the fourth.
Product in the thread. Filling leaves a film on the thread. It dries hard. Now the cap stops 0.3 mm to 0.5 mm short. The liner is under-compressed by that exact amount. You cannot see it. A fingernail can feel it.
Fill line too high. Product sits against the mouth. Tilt the jar and it runs into the thread. Saving 2 g of fill per unit buys a leak path on every jar in the carton.
Pressure drop itself. This is the trigger, not the cause. A sound liner absorbs it. A tired liner does not.
Note what is missing from that list. Torque. Almost every complaint blames it. Almost every failed jar was torqued correctly. Torque is easy to measure, so it gets measured. Then it gets blamed.
3. Why your warehouse test proves nothing
A jar leak test at sea level tests a jar at sea level. The pressure difference that causes the failure is not there.
Warm the jar to 40 °C and it may weep. Viscosity drops. That is a different failure. It says nothing about 30,000 feet.
So the test has to create the pressure difference. Two ways exist. Put the jar in a vacuum chamber. Or fly it and open the carton at the far end. Only one of those repeats.
4. Five tests, and what each one catches
A flight is the harshest leg a jar will ever travel. These five cover it.
As a cosmetic jar manufacturer we run all five, and we keep the records per SKU.
- Vacuum decay. The closed jar goes into a chamber. Air is pumped out. Then we watch the pressure climb. A leaking jar lets air back in, and the curve rises. It is the only bench method that reproduces a flight.
- Open-close cycling. Thirty cycles, then vacuum decay again. Catches the liner that only seals when new.
- Inverted hold. 40 °C, upside down, 72 hours. Catches edge seepage and a fill line set too high.
- Drop test. 1.2 m, six faces. Catches a cap that pops and a wall that cracks.
- Compatibility. Filled with your formula. Hold at 40 °C for four weeks, then eight. Catches deformation, colour shift and stress cracking.
| Test | Reproduces altitude? | Catches |
|---|---|---|
| Vacuum decay | Yes | Thread and liner leaks as a whole |
| Open-close cycling | No, but ages the liner | Liner that seals only when new |
| Inverted hold | No | Edge seepage, overfill |
| Drop test | No | Cap pop, wall crack |
| Compatibility | No | Deformation, stress cracking |
The bench conditions above are conventional. We do not claim them as a standard. 1.2 m, 40 °C and 72 h are industry habits.
5. Finding the cause from the symptom
Almost every cosmetic jar leaking in air freight reaches us as a photo of a lid. Leakage in transit gets blamed on the lid for the same reason. The photo rarely helps. The pattern does.
| What you see | Most likely cause | Check first |
|---|---|---|
| Only air shipments leak | Liner compression set | Liner thickness and compression ratio |
| Leaks cluster on one side | Product in the thread, or off-centre filling | Thread cleanliness, nozzle alignment |
| Product already inside the cap on opening | Fill line too high | Headspace above the fill |
| Sea freight leaks too | Not pressure | Liner and formula compatibility |
| Leaks in a hot warehouse, not a cold one | Viscosity drop | Inverted hold at 40 °C |
Two rows cover most real cases. Failure only in the air means the liner. Failure everywhere means the formula met the wrong liner.
6. What to change in the spec
Three changes fix most of it, in order of cost.
Reduce the fill. Leave the headspace the pressure needs. Safe fill runs 85% to 90% of overflow capacity. That is a convention, not a rule.
Change the liner. PE foam is the default, and the first to take a set. Aluminium composite holds better when the oil phase is high.
Change the closure. A double-lead thread opens in half a turn. Pleasant in the hand. Harder to seal, because less thread is engaged. Single lead with a deeper skirt seals better.
What not to change first: the cap torque. Raising it compresses the liner harder. A compressed liner takes a set faster. You trade a leak today for a leak in three months.
7. FAQ
Is leaking in air freight a manufacturing defect?
Usually not. A cosmetic jar leaking in air freight was often moulded and closed correctly. The liner simply cannot hold the pressure drop. That is a spec gap, not a production fault. It is also worth settling before the tool is cut.
Can I test this without a vacuum chamber?
Not accurately. You can fly a carton and inspect it. Expensive. Not repeatable. A vacuum decay test on one sample jar is the practical option.
Does a thicker liner always help?
No. Too thick means over-compressed on the first closing. Over-compression speeds up compression set. The right liner returns to thickness after thirty cycles.
Why did last year's order pass and this one fail?
Liner batches differ. So do torque settings and fill temperatures. Ask any cosmetic jar manufacturer for the vacuum decay record on the batch you are buying. Not last year's.
Does this apply to wide-mouth jars only?
No. Wide mouth jar sealing is less forgiving, though. The liner is larger, so the same error covers more area.
Where this sits
Liner and thread decisions belong to the tool, not to the packaging brief. That is covered in who owns the mould. If the formula decides the pack, start with the moisturizer jar guide.
To send a sample jar for a vacuum decay test, use the contact page.
