Cup Side-Seam Overlap FAQ: Width, Bond and the Leak Path

Published: 2026-10-24

The seam overlap is the length of the leak path, and in most cup factories it is set by the sheet and the die and never measured again. These questions cover the width, the section, the hot fill, the squeeze and the machine settings that hold the seam together.

Why does a narrow seam leak when a wide one does not?

Because the overlap is the distance a liquid must travel to escape. A heat-sealed seam is never bonded perfectly across its full width: there are micro-channels at the seal edges and where the coating is thin. A wide overlap surrounds a channel with bonded material and the liquid stops. A narrow overlap lets the same channel reach the outside edge and the cup weeps. The defect in the seal is identical in both cases, and only the geometry decides whether the customer ever sees it.

What overlap should be specified?

Typically 5.0 to 6.5 millimetres with a tolerance of plus or minus 0.5, taken from the converter's proven design and confirmed at the lower limit with a hot fill test. A tolerance is not decoration: without one, a tool service that removes a shim can move the overlap by half a millimetre and the drawing is still satisfied. The value should be set for the worst product the cup will hold, because a width that is adequate for a cold drink can be marginal for a hot one.

Why is a hot drink harder on the seam?

Because heat softens the polyethylene coating, lets the bond creep under load and raises the vapour pressure inside the cup. The effect concentrates at the top of the seam, where the cup leaves the jaw, where the jaw is most likely to be slightly cooler or less loaded, where the liquid sits when the cup is full, and where the cup is squeezed when it is lifted. A leak that starts as a bead within 20 millimetres of the rim is showing all four at once.

ISO 9001, published by the International Organization for Standardization, requires inspection of characteristics that affect fitness for use, which for a cup includes the side seam overlap.
International Organization for Standardization, https://www.iso.org

How is the overlap measured?

Cut a cup through the seam with a sharp blade, lay the section flat and measure with a steel rule or a magnifier with a scale. Measure at three heights: 20 millimetres below the rim, the middle, and 20 millimetres above the bottom. The top reading matters most because that is where field leaks appear. One cup per shift and one per lot is reasonable for a stable process, recorded with the value and the height.

When should the frequency increase?

After a tool service, a sheet lot change or a coating lot change, to one cup per hour for the first eight hours of the new condition. A process that has just changed is the most likely to be outside a tolerance nobody has checked in a year, and the change is exactly when a single sample per shift can miss a shift-long excursion. Return to the normal frequency when the readings have been stable for the eight hours.

What does a section show besides the width?

The bond quality. A good bond shows the two layers merged with the coating bridging the fibers and no line of separation. A weak bond shows a visible boundary, a lighter band where the coating did not flow, or a section that separates when flexed. A bond line that stops short of the outer edge is the classic weep: the seal exists and does not reach the exit point. A crease in the overlap leaks regardless of width.

Why did nobody notice the change?

Because every other check was measuring something else. The rim diameter is set by the forming collar, the height by the blank, the bottom by the bottom die and the grammage by the board supplier, and none of those moves when the overlap changes. The finished cup looks the same. When a property is set by the tool rather than by the operator, the inspection has to look at the tool's output in a way that reveals it, and a section is the cheapest such look.

Why is a room temperature water fill a weak test?

Because it reproduces neither the temperature nor the load that the field applies. A static fill at twenty degrees can sit for ten minutes and show nothing on a seam that weeps at ninety-five degrees within twenty minutes of pouring. The test is not wrong, it is insensitive, and a factory that relies on it alone is testing a cup nobody handles the way the test does.

What should the hot fill test look like?

Water at 95 degrees Celsius, the cup on a dry surface, and inspection of the seam at 5, 15 and 30 minutes. Look for a bead on the outside, darkening of the board at the seam, and softening that lets the seam move when the cup is lifted. A cup that passes at 5 minutes and fails at 30 is showing creep, which points to coating and pressure rather than to width.

Why add a squeeze?

Because a cup is squeezed when it is lifted, and a seam that holds still can open under a squeeze. Grip the cup at the height where a hand naturally grips, apply a defined load for a few seconds and inspect the seam immediately. This is the gesture that produces most field leaks, and without it the test covers a handling pattern that no customer follows. Ten newtons at grip height is a repeatable starting value.

The U.S. Food and Drug Administration administers food-contact requirements for paper and coatings used in cups that contact beverages, which apply to the seam area as well as the wall.
U.S. Food and Drug Administration, https://www.fda.gov

What machine settings control the bond?

Temperature, pressure and dwell, in the range of 180 to 220 degrees Celsius, 0.3 to 0.5 MPa and 0.6 to 1.2 seconds, adjusted for the coating grade. They interact: more heat to fix a weak bond can burn the coating, more pressure to compensate for heat can thin the board at the seam, and more dwell adds cycle time and heat to the whole cup. Log all three hourly against the section width and plot them.

How do you tell a heater drift from a sheet drift?

By comparing the two curves. A falling bond quality with an unchanged overlap points to the jaw heater or the pressure, and it appears in the log as a temperature or load trend before it appears in the cup. A changing overlap with unchanged settings points to the sheet width or the forming die. The two causes look alike in a finished cup and are distinguishable only in the records.

What triggers a stop?

A change over half a millimetre in the measured overlap, a visible unbonded line on a section, any moisture on the outside of the seam in the hot fill, or a seam that opens under the squeeze. Write each with its sample frequency on the shift sheet, so the decision is a comparison with a number rather than a judgement about a cup. Set the values per product, because a cold drink cup can carry a narrower overlap than a hot drink cup.

Where should the overlap appear?

On the drawing with a tolerance, not only in the process sheet. A dimension on the drawing is checked at first article and at every tool service, while a process note is checked when someone remembers. The difference between those two is the difference between a specification and a hope, and it is the reason the drawing in the scenario above described a cup that could not be inspected for the property that failed.

What belongs in the specification?

The overlap with tolerance and method, the three measuring heights, the seal settings with ranges, the section frequency, the hot fill and squeeze procedures with pass criteria, the trigger values, a first-article section photograph, and the requirement that the section be kept with the lot record for a year. A cup specification that describes the rim and the print has described the top of the cup, and the seam runs the full height.