Paper Cup Bottom Seal Leak Test: Pressure, Dwell and the Three Second Rule

Published: 2026-10-31 | Author: Yoco Group Editorial

A paper cup leaks at the bottom seal long before it leaks anywhere else. The bottom is a formed disc bonded to a curled wall skirt under heat and pressure, and the bond has to survive 95 degrees Celsius liquid for 20 minutes, a 1.2 metre drop and a customer squeezing the cup. Most leak complaints trace to one of three parameters: seal temperature, dwell time or seal pressure.

What the Bottom Seal Has to Do

The wall is coated on the inside with 18 to 25 grams per square metre of PE, and the bottom disc carries the same coating. The seal is formed by softening both coatings and pressing them together while the polymer chains interdiffuse. A finished seal has a 6 to 8 mm wide bond ring and should show 100 percent fibre tear when pulled apart, not a clean peel.

If the seal peels cleanly, the interface never reached the melting range. If the cup leaks at 0.3 bar of air pressure, the bond is continuous but porous.

Test Methods That Find Different Faults

The water leak test is the simplest: fill with water at 90 degrees Celsius, stand for 24 hours on absorbent paper, inspect for a wet ring. It finds gross channel leaks but misses a pinhole that opens under pressure.

The air decay test is faster and more sensitive. Pressurise the cup to 0.03 megapascals, isolate, and measure the pressure drop over 30 seconds. A drop above 10 percent indicates a leak. Run it on 3 cups per 1 000, and on the first 20 cups after every seal temperature change.

Dwell Time and Temperature

PE seal initiation begins around 105 degrees Celsius and full bond strength develops at 130 to 150 degrees Celsius at the interface. The sealing jaw is usually set at 180 to 220 degrees Celsius, because heat has to travel through the paper wall, and the dwell is 0.4 to 0.8 seconds.

Cutting dwell to raise line speed is the most common cause of a leak. At 0.25 seconds the interface reaches 110 to 120 degrees Celsius and the seal looks closed but holds only 40 to 60 percent of its strength. It will pass a visual check and fail in the customer's hand.

Pressure and Seal Coverage

Seal pressure runs 0.3 to 0.5 megapascals on the jaw. Below 0.25 the coating does not flow into the fibre valleys and the bond ring has gaps; above 0.6 the paper crushes and the wall thickness at the seal drops by 15 to 20 percent, which creates a weak line rather than a strong one.

Check the jaw for flatness every month. A 0.05 mm crown in the jaw produces a 0.05 mm gap at the edges and a channel leak that appears only on one side of the cup.

The Three Failure Modes

A channel leak runs from the inside edge to the outside along one line, usually from a wrinkle in the bottom disc. A cold seal shows a dull, easily peeled bond over part of the ring. A crushed seal shows a thin, translucent band where the fibre has been compressed and the coating pushed aside.

Each has a different fix. Channel: improve disc flatness and reduce forming speed. Cold seal: raise jaw temperature 10 degrees Celsius or lengthen dwell by 0.1 second. Crushed seal: reduce pressure and check jaw parallelism.

Inline Sampling and Control

Run the air decay test on 3 cups per 1 000 and record the pressure drop. Run the 90 degrees Celsius water test on 5 cups per shift, held for 24 hours. Pull a seal strength sample every 2 hours and record fibre tear percentage.

When a lot fails, quarantine 2 000 cups upstream and downstream of the sample. Leaks cluster: a jaw temperature drift of 8 degrees Celsius typically affects 15 to 30 minutes of production, which is 1 500 to 3 000 cups at 100 cups per minute.

Specification for the Purchase Order

Write: bottom seal 100 percent fibre tear, air decay under 10 percent at 0.03 megapascals over 30 seconds, no leakage after 24 hours at 90 degrees Celsius, seal ring width 6 to 8 mm, and no visible crushing.

Then require the supplier to report jaw temperature, dwell and pressure for the lot. Without those three numbers, a pass certificate is a statement of hope rather than a record of a process.

Related reading