A field guide for QA teams running bubble emission testing on flexible packaging
Your bubble leak test showed a pass. Then your customer found a leaking package in the field.
Before you blame the sealer, check your test setup. The most common reason a bubble test misses a real leak isn’t a machine problem. It’s vacuum level, water temperature, or a package with no internal air to push gas through a defect. A test that isn’t configured correctly doesn’t tell you whether the package is sealed. It just tells you the test ran.
This guide covers what ASTM D3078 actually requires, how to run it so results mean something, where it fits against other leak detection methods, and the setup mistakes that quietly undermine QA programs.
What Is a Bubble Leak Test?
A bubble leak test is a method for detecting gross leaks in flexible packaging by submerging a sealed package in fluid inside a vacuum chamber. As vacuum is drawn, internal headspace gas expands and escapes through any seal defect, producing a steady, continuous stream of bubbles at the leak point. The test follows ASTM D3078 and delivers a fast, visual pass/fail result in around 30 seconds.
Two things the test is not. It does not measure leak rate or seal strength, and it does not specify a guaranteed micron sensitivity. ASTM D3078 detects gross leaks: visible seal failures, open channels, and breaches large enough to allow headspace gas to escape under vacuum. Sensitivity depends on vacuum level, headspace volume, product type, and packaging material. Small leaks may not be detected.
For most flexible food packaging QA programs, that’s exactly what’s needed. Gross leaks are the failures that compromise product safety, accelerate spoilage, and put manufacturers at risk of recall. Catching them before a shipment leaves the facility is what D3078 is designed for.
The bubble emission test is sometimes called bubble leak testing, bubble emission leak testing, or “the soap bubble leak test procedure” when wetting agents are added to lower surface tension. They all refer to the same underlying method. ASTM D3078 is the formal standard published by ASTM International that codifies how the test should be run.
When Is ASTM D3078 the Right Method?
D3078 is the right starting point when you’re testing flexible packages (pouches, bags, flow wraps, retort pouches, and similar formats) that contain internal headspace gas. The test relies on that internal air to create the pressure differential that pushes gas through a defect and produces a visible bubble stream. No headspace, no test signal.
D3078 applies when:
- Your package has internal headspace (air or modified atmosphere gas)
- You need fast, visual confirmation of gross seal integrity
- You’re running batch-level QA for food, pharmaceutical, or medical device packaging
- Your customer or auditor has specified ASTM D3078 by name
- You need to know where a leak is, not just whether one exists
D3078 has limits when:
- Your package is vacuum-sealed with little or no internal air
- You need higher sensitivity than gross-level detection supports
- You’re testing rigid glass containers that might crack under vacuum
- Your application requires nondestructive, instrument-based results for regulatory documentation
Vacuum-sealed packages are the case teams most often get wrong. A package with no headspace has no meaningful pressure differential under vacuum. Product can also plug small openings, which means a genuine leak can produce no bubbles at all. The fix is the FlexPak Vacuum Pack Attachment, which introduces a controlled shot of air into the package before testing. That restores the headspace D3078 needs without modifying the seal.
When porous packaging like Tyvek is in play, ASTM F2096 internal pressurization is often the better fit. When your application calls for nondestructive, higher-sensitivity detection, vacuum decay testing may be what your regulatory framework expects.
Bubble Leak Test vs. Other Leak Detection Methods
Different industries land on different leak detection methods, and the reasons usually have less to do with raw sensitivity and more to do with what kind of evidence your QA program needs to produce. Bubble emission wins on speed, cost, and one thing the instrument-based methods can’t do: showing you exactly where the leak is.
| Test Method | Package Type | Leak Type Detected | Destructive? |
|---|---|---|---|
| Bubble Emission (ASTM D3078) | Flexible packages with headspace | Gross leaks, visible bubble streams | No |
| Internal Pressurization (ASTM F2096) | Trays and pouches | Gross leaks (~250 μm with ~81% probability per published round-robin data) | Yes |
| Vacuum Decay (ASTM F2338) | Rigid, semi-rigid, flexible nonporous | Higher-sensitivity leaks (validated per package type) | No |
| Altitude Simulation (ASTM D6653) | Any | Conditioning only; pair with detection method | No |
| Closure Integrity (ASTM D5094) | Rigid containers with threaded closures | Gross liquid leaks at closure points | No |
What this comparison highlights: bubble emission is the only fast, low-cost method that also tells you where the leak is. Vacuum decay testing confirms whether a leak exists with higher sensitivity for specific package types, but it cannot show you where on the package. For QA teams diagnosing sealing machine problems, that location data is what makes corrective action possible. A steady bubble stream from the side seal of a pouch tells your line operators exactly where to look on the sealing machine. A vacuum decay fail tells them only that something, somewhere, is wrong.
[SME INSERT: short anecdote from Gordon on a case where bubble emission location data helped resolve a recurring seal failure that another method had only flagged as intermittent.]
Bubble Leak Test Equipment You Need
You don’t need a full lab to run ASTM D3078. But every piece of equipment has to do its job reliably. A leaky chamber seal or a foggy window costs you confidence in every result you record.
Transparent vacuum chamber. Must withstand full atmospheric pressure differential and stay fully clear for bubble observation. Clarity matters. Faint bubble streams are easy to miss through a clouded or scratched chamber wall. The FlexPak Package Leak Detector chamber is built specifically for this application.
Vacuum-tight lid with gauge. The lid must seal completely and include fittings for the vacuum pump, pressure gauge, and atmosphere release. A lid that doesn’t seal produces false passes consistently.
Vacuum pump. Creates the pressure differential that draws gas from the package. The pump needs to reach the vacuum level your test protocol specifies. Verify this during qualification before running production samples.
Pressure gauge and regulator. Fine-tune vacuum levels and confirm the chamber is holding. An uncalibrated or slow-responding gauge means you don’t actually know what vacuum level the package experienced during the test.
Test fluid. Room-temperature water works well. Cold water has higher surface tension, which can prevent small bubble streams from forming visibly. This is one of the most common sources of false passes. Lower-surface-tension fluids (water with a wetting agent, or a soap bubble leak test procedure variant) are generally more sensitive than plain water. Denatured alcohol and mineral oil are also acceptable under the standard.
Hold-down fixture or grid. Keeps the package fully submerged. A package that floats partially out of the water means part of the seal area isn’t being observed. The hold-down plate is not optional.
Observation lighting. Adequate lighting around the chamber ensures even faint bubble streams are visible. Dim conditions are a setup problem, not an operator problem.
For teams choosing equipment, the leak tester selection guide walks through the trade-offs across chamber sizes, automation levels, and accessory options like the semi-automatic control package.
How to Run the Bubble Leak Test Step by Step
Start with a clean, sealed, unpunctured package that contains headspace gas. If you’re testing vacuum-sealed packages, introduce air with the Vacuum Pack Attachment before beginning. Bring both the package and test fluid to room temperature. Around 20 to 25°C is a commonly used range, though your validated protocol may specify differently based on your package type.
Step 1: Fill and prepare the chamber.
Fill the vacuum chamber with enough fluid to fully submerge the package with at least one inch of coverage above its highest point. Check that the chamber seals, gaskets, and fittings are clean and intact before starting.
What success looks like: clear fluid, clean fittings, no fog on the chamber wall.
Step 2: Submerge the package.
Place the package in the center of the chamber. Lower the hold-down plate or grid to keep it fully submerged. Close and seal the lid. Confirm the package isn’t folded or positioned so that a seal area sits outside the fluid line.
What success looks like: package fully underwater, all seal surfaces visible from at least one angle.
Step 3: Apply vacuum slowly.
Apply vacuum at a steady rate (around 1 in. Hg per second is typical). Ramping too quickly can rupture the package before the test is complete. Your target vacuum level should be established during process validation for your specific package, not picked arbitrarily. A commonly used benchmark for North American flexible food packaging is around 12 in. Hg, but the right number depends on package strength, headspace volume, and product behavior. Hold vacuum for around 30 seconds once the target level is reached.
What success looks like: steady gauge reading at the target level, no package deformation beyond what your validation expects.
Step 4: Observe.
Watch the entire package surface (seal areas, folds, and seams) for a steady, continuous stream of bubbles from a consistent location. That’s a leak. Isolated bubbles that appear once and stop are typically entrapped air releasing from surface folds and don’t count as a leak indication. The key is continuity from a fixed point.
What success looks like: either no continuous bubble stream (pass) or a clear, locatable stream (fail with leak location identified).
Step 5: Release and remove.
Release the vacuum slowly, remove the package, and inspect it for fluid ingress. Fluid inside the package after a vacuum test is a secondary failure indicator. Record results immediately.
What success looks like: package removed intact, with or without fluid ingress noted.
Step 6: Document everything.
Log vacuum level, hold time, fluid temperature, package ID, result, leak location if observed, operator initials, and date. This log is your audit defense. A result without documentation is a result that doesn’t exist when an auditor asks for it.
[SME INSERT: operational detail from Gordon on how FlexPak’s Custom Bubble Tests program adapts these steps for non-standard package geometries.]
How to Read Your Results Accurately
Results come down to what you observe, but the interpretation requires more than “bubbles or no bubbles.”
Pass: No continuous bubble stream observed from any location during the full hold time. No fluid ingress after vacuum release.
Fail: A steady, continuous stream of bubbles from a fixed location (seal area, seam, or film surface). The location matters. Record exactly where it is so sealing machinery adjustments can be targeted.
Inconclusive: Isolated bubbles that appear briefly and stop, or bubbles that can’t be traced to a fixed location. This is most often entrapped air releasing from folds or seams during the initial vacuum pull. Retest before calling a result. If the same pattern repeats in the same location, treat it as a fail.
Here’s what a documentation log should look like:
| Package ID | Vacuum | Hold Time | Fluid Temp | Result | Leak Location | Operator | Date |
|---|---|---|---|---|---|---|---|
| 34201 | 12 in. Hg | 30 sec | 22°C | Pass | N/A | A. Patel | Mar 4, 2026 |
| 34202 | 12 in. Hg | 30 sec | 22°C | Fail | Side seal | A. Patel | Mar 4, 2026 |
| 34203 | 12 in. Hg | 30 sec | 22°C | Inconclusive | Bottom fold | A. Patel | Mar 4, 2026 |
The value of bubble emission testing isn’t just the pass/fail. It’s the location data. A steady bubble stream tells you exactly where on the package the seal failed, which tells your line operators exactly where to look on the sealing machine. For a broader view of how D3078 integrates with the rest of a package integrity program, see the food package leak testing guide.
4 Setup Mistakes That Produce False Results
These aren’t edge cases. They show up regularly, and they’re all fixable.
1. Vacuum level not validated for your package type.
This is the one that quietly undermines entire QA programs. Vacuum level directly determines what you can and can’t detect. Too low, and genuine leaks produce no visible bubble stream. Too high, and you’re rupturing packages before the test is complete. The right vacuum level isn’t a default. It’s validated for your specific package using control samples with known defects. If your team picked a number and hasn’t revisited it since, that’s the first thing to audit. The bubble leak testing mistakes article covers this in more depth.
2. Testing vacuum-sealed packages without air introduction.
A vacuum-sealed package has no meaningful internal air. Under vacuum, there’s no pressure differential to push gas through a defect, and product can physically plug small openings. The result is a false pass on a package that may have a genuine leak. The Vacuum Pack Attachment resolves this by introducing a controlled shot of air before the test, restoring the headspace D3078 needs. If your line includes vacuum-sealed SKUs and you’re not using air introduction, your bubble test results for those packages aren’t reliable.
3. Cold or high-surface-tension fluid.
Cold water resists bubble formation. A genuine leak in cold water may produce bubbles too small or too slow-moving to observe clearly. What’s happening underneath: the surface tension at the leak point is high enough that escaping gas deforms the fluid surface without breaking through it visibly. Room-temperature water, or water with a small amount of wetting agent, lowers surface tension and makes bubble streams visible at lower flow rates. It’s a simple fix that meaningfully improves detection.
4. Inadequate observation conditions.
Bubble emission testing is inherently visual. A distracted operator, poor lighting, or a clouded chamber wall are all setup failures, not human failures. If the observation conditions aren’t controlled (consistent lighting, clear chamber, operator’s full attention for the full hold time), the test outcome depends on luck. Control the conditions, and the test controls itself.
Field-Tested Tips for Different Package Types
After 25+ years running bubble emission testing across food, pharmaceutical, medical device, and consumer goods packaging, certain patterns hold by package format. Treat these as starting points to incorporate into your own validated protocols.
Pouches. Focus observation on corners and transition areas where seal width changes. These are the most common leak locations on stand-up and flat pouches. Angle changes create stress concentration points where the seal can thin or skip.
Tray seals. Watch the entire perimeter, especially where the film meets curved edges. Heat seal coverage tends to vary at curves, and that variation produces inconsistent seal strength.
Gusseted bags. Pay extra attention to gusset folds where multiple film layers meet. Three- and four-layer overlap points are common failure sites, and trapped air at those layers can mask continuous streams if you’re not watching specifically.
Retort pouches. Retort pouch leak testing has its own considerations around post-process testing temperature and seal stress patterns. The bubble test still works, but validate at the temperature you’ll be testing at, not just ambient.
MAP packaging. Modified atmosphere packages have gas headspace by definition, which makes them well-suited to D3078. MAP packaging leak detection covers the protocol variations specific to controlled-atmosphere applications.
A few visibility improvements that apply across package types: filtered fluid reduces particulate that can obscure bubble streams, positioning packages at slight angles prevents bubbles from hiding underneath, and photo or video documentation of unusual patterns helps when you need to investigate recurring failures.
[SME INSERT: specific example from Gordon where pattern recognition across a documented bubble emission archive helped identify a systematic sealing machine issue.]
Frequently Asked Questions
What size leaks does a bubble leak test detect?
ASTM D3078 detects gross leaks: visible seal failures, open channels, and breaches large enough to allow headspace gas to escape under vacuum. It does not specify a guaranteed micron sensitivity, and detection capability depends on vacuum level, headspace volume, product type, and packaging material. Small leaks may not produce a visible bubble stream under standard test conditions.
Can you bubble test vacuum-sealed packages?
Yes, with air introduction.
Vacuum-sealed packages have no internal headspace, which means no pressure differential to push gas through a defect under vacuum. The FlexPak Vacuum Pack Attachment introduces a controlled shot of air into the package before testing, making vacuum-sealed packs reliably testable under ASTM D3078. Without air introduction, a bubble test on a vacuum-sealed package will produce a pass regardless of whether the seal is intact. This is one of the most common false-pass patterns in QA programs that haven’t accounted for vacuum-sealed SKUs in their protocol.
How long should you hold vacuum during a bubble leak test?
ASTM D3078 specifies holding vacuum for around 30 seconds once the target level is reached, though your validated protocol may specify a longer hold time depending on your package type, headspace volume, and product. Hold time should be established during validation, not selected based on convenience. A shorter hold reduces the observation window and increases the likelihood of missing a slow-developing leak.
What vacuum level does ASTM D3078 specify?
D3078 doesn’t mandate a single vacuum level. The standard expects users to validate the appropriate vacuum for their specific package. A commonly used benchmark for North American flexible food packaging is around 12 in. Hg, though the right number for your package depends on package strength, headspace volume, and product behavior. Run control samples with known defects to determine the level that catches your worst-case failures without rupturing intact packages. Higher vacuum levels improve sensitivity to smaller leaks but raise the risk of package rupture.
How does ASTM D3078 differ from ASTM F2096?
Both are bubble-based leak tests, but they pressurize the package from opposite directions, and one is destructive.
D3078 pulls vacuum externally, forcing internal headspace gas out through defects. It’s nondestructive and works well for flexible food packaging with headspace. F2096 punctures the package, inflates it underwater to a predetermined pressure, and watches for bubbles at failure points. F2096 is destructive (the package must be punctured) and is more commonly used for trays and pouches in medical and pharmaceutical applications, including those with porous materials like Tyvek. Published round-robin data for F2096 shows approximately 81% probability of detecting a 250 μm defect, which is the only ASTM bubble test where a specific micron figure is directly supported by interlaboratory validation. See the full ASTM F2096 practical uses guide for a detailed comparison.
Is a bubble leak test required by regulation?
ASTM D3078 is voluntary as a standard, but specific industries adopt it as part of regulated quality systems. FDA guidance on container closure integrity testing for pharmaceutical packaging references bubble emission as one acceptable method among several. Food safety programs operating under HACCP commonly use D3078 as a critical control point verification. Medical device packaging compliance under ISO 11607 accepts bubble emission as part of a broader package integrity testing program. The applicable framework depends on your product category and target markets.
About This Article
Written with input from Gordon Bruce, who handles sales and customer education at FlexPak Inc. FlexPak manufactures ASTM-compliant bubble emission leak detection equipment for package integrity testing across food, pharmaceutical, and medical device industries, with 25+ years of experience helping QA teams match testing methods to package formats. Reach Gordon at gordon@flexpakinc.com.