A canned food plant produces 600 cans per minute on a continuous line. Each can passes through a seam inspection station after the lid is double-seamed. A leaking can — the leak may be a 10-micrometer microchannel through the sealing compound, invisible to the naked eye — enters a pallet, ships to a distribution center, and sits on a shelf for six months. The vacuum inside the can draws ambient air through the micro-leak. Oxygen enters. The food oxidizes. The can swells — or worse, the contents spoil without visible swelling, and the consumer finds it when they open the can. The recall costs the packer $2 million in retrieval logistics, destroyed product, and brand damage — all from a leak smaller than a human hair. Non-destructive seal integrity testing catches that leak before the can leaves the plant, while preserving the product inside. This article compares the four dominant non-destructive testing methods — vacuum decay, pressure decay, helium leak detection, and vision-based inspection — on sensitivity, speed, and in-line suitability, so food packaging quality managers can select the right method for their can type, line speed, and defect risk profile.
Vacuum Decay: The Workhorse of Can Integrity Testing
Vacuum decay testing measures the rate at which vacuum is lost inside a sealed container. The can enters a test chamber that seals around it and draws a vacuum on the external chamber volume. A pressure transducer monitors the chamber pressure. If the can has a leak, gas from inside the can escapes into the chamber, and the chamber pressure rises — the rate of rise correlates with the leak size. The test takes 2 to 10 seconds per can depending on the sensitivity required: a 10-second test can detect leaks down to roughly 5 to 10 µm equivalent diameter; a 2-second test detects leaks down to roughly 20 to 50 µm. The method is non-destructive: the can and its contents remain intact, and acceptable cans return to the production line without any product contact or alteration.
Vacuum decay is limited to cans that are sealed under vacuum — it cannot be used for cans sealed at atmospheric pressure or under positive pressure (carbonated beverages, aerosol cans). The test also requires a minimum headspace volume inside the can — the gas volume that expands when the external vacuum is applied. Cans with very low headspace (under 3% of can volume) produce a weak signal because there is insufficient gas inside to drive a measurable pressure rise through a small leak.
Pressure Decay: For Cans Without Vacuum
Pressure decay testing works on the same principle as vacuum decay — measuring a pressure change over time — but in reverse: the can is pressurized externally, and a leak is detected by pressure drop in the test chamber as gas leaks into the can. This method works for cans sealed at atmospheric pressure and for flexible and semi-rigid packaging (pouches, trays) that cannot withstand external vacuum. Sensitivity is similar to vacuum decay — roughly 10 to 50 µm under typical in-line test dwell times of 3 to 10 seconds. Pressure decay adds the requirement that the can must withstand the external overpressure without deforming — thin-walled aluminum cans may buckle under external pressure above 0.5 bar, limiting the test pressure and therefore the sensitivity.
Helium Leak Detection: Maximum Sensitivity for Critical Applications
Helium leak detection is the most sensitive non-destructive method, capable of detecting leaks down to 0.1 µm equivalent diameter — roughly 50 to 100 times more sensitive than vacuum or pressure decay. The can is placed in a vacuum chamber connected to a helium mass spectrometer. Helium tracer gas is introduced on the outside of the can (or, alternatively, the can is filled with a helium-containing atmosphere before sealing, and the spectrometer detects helium escaping into the vacuum chamber). The mass spectrometer is tuned to the atomic mass of helium (4 amu) and detects helium atoms at concentrations down to parts per billion.
The sensitivity comes at a cost: helium leak detectors cost $20,000 to $60,000, the test cycle takes 15 to 30 seconds per can (too slow for 600-cpm lines — typically used for statistical sampling, not 100% inspection), and helium tracer gas adds a consumable cost of $0.005 to $0.02 per can tested. Helium leak detection is the standard for medical device and pharmaceutical packaging where the cost of a leak — patient harm — justifies the test sensitivity and cycle time. For canned food, helium testing is used primarily for process validation (qualifying a new seamer setup), troubleshooting (finding the root cause of a sporadic leak rate), and high-risk products (low-acid canned foods where Clostridium botulinum growth is a lethal hazard).
Vision-Based Inspection: Fast Enough for In-Line, Limited by What It Cannot See
Vision-based seam inspection uses high-speed cameras and structured lighting to inspect the external geometry of the can seam — the double-seam dimensions, the seaming chuck mark, and visible defects such as cut-over, sharp seam, droop, or incomplete seaming. Modern systems process 600 to 1,200 cans per minute, inspecting every can at full line speed, and reject cans with seam dimensional deviations outside the acceptable range.
The limitation is fundamental: a can with a seam that measures perfectly within specification — correct body hook length, cover hook length, overlap, and tightness — can still leak if the sealing compound contains a microscopic void, if a contaminant particle bridged the compound during seaming, or if the compound distribution is uneven in a way that does not affect external seam dimensions. Vision systems detect the macro-defects that cause roughly 80 to 90% of can leaks. They cannot detect the micro-defects that cause the remaining 10 to 20%. For this reason, vision inspection is often paired with a vacuum or pressure decay system: vision catches the gross seam defects at full line speed, and the decay test on a statistical sample catches the micro-leaks that would otherwise ship.
Which method for which production environment?
| Method | Minimum Detectable Leak | Per-Can Cycle Time | Best Use |
|---|---|---|---|
| Vacuum decay | 5–50 µm | 2–10 s | 100% in-line for vacuum-sealed cans; mid-speed lines (100–400 cpm) |
| Pressure decay | 10–50 µm | 3–10 s | Atmospheric-pressure cans and flexible packaging; semi-automated lines |
| Helium leak detection | 0.1–5 µm | 15–30 s | Process validation, troubleshooting, high-risk low-acid products; statistical sampling only |
| Vision-based seam inspection | N/A (macro-defects only) | >1,200 cpm | 100% in-line for gross seam defects; best paired with a decay method for micro-leak coverage |
Non-destructive seal integrity testing for canned food is a risk-management decision, not a technology competition. For 100% in-line inspection on a high-speed line running vacuum-sealed cans, vacuum decay testing provides the best balance of sensitivity, speed, and cost — detecting leaks down to the tens of microns at cycle times compatible with production throughput. For atmospheric-pressure cans, pressure decay fills the same role. For process validation and troubleshooting — or for low-acid products where the consequence of a single leaker is fatal — helium leak detection provides the sensitivity to find leaks too small for any other method. And for catching the gross seam defects that cause most leaks, vision inspection at full line speed paired with statistical decay testing is the cost-effective combination that catches more total defects than either method alone.
