Gross-leak testing addresses relatively large leakage paths; fine-leak testing addresses smaller leakage rates using a suitably sensitive method. For a sealed package, these functions are complementary. A low fine-leak reading cannot always distinguish a genuinely tight enclosure from a package that has already lost its tracer gas through a large opening.
For laser hermetic sealing, the practical task is to establish coverage across the required failure range, with test conditions that the package can tolerate. This guide compares the two functions and explains how to review conflicting results. The separate helium leak-testing guide covers detector setup and measurement records in more detail.
What the two tests are intended to detect
The terms gross and fine describe parts of a test strategy, not universal numerical categories. The boundary depends on the specified method, package and acceptance requirement. Neither word alone defines a test gas, pressure or pass limit.
| Question | Gross-leak function | Fine-leak function |
|---|---|---|
| What failure range is addressed? | Relatively large paths or an effectively open package | Smaller flows that need a more sensitive measurement |
| What approaches may be used? | Qualified liquid-based or dry methods appropriate to the package | Frequently helium-based measurements; other qualified methods may apply |
| What can complicate the result? | Media compatibility, cavity size, pressure response and observation conditions | Tracer concentration, exposure, transfer time, background and calibration |
| What does a passing result mean? | No rejectable gross leak detected by that procedure | No rejectable fine leak detected under those test conditions |
| What does it not establish alone? | Acceptance throughout the fine-leak range | Reliable detection of every gross leak or every other package defect |
A single validated instrument may perform more than one function. The requirement is complete, demonstrated coverage; it does not necessarily mean two separate machines.
Why a large leak can produce a small helium signal
In a helium bombing process, a sealed package is exposed to pressurized helium. Tracer enters through available leak paths. The package is then transferred to a detector that measures escaping helium.
Consider the timeline rather than only the final reading. With a sufficiently large path, helium can leave rapidly after pressurization ends. If little tracer remains when measurement begins, the detector may show a small signal even though the package is not acceptably sealed. Internal free volume and the delay before measurement influence this behavior.
This is a reason to include a suitable gross-leak function and to control timing. It is not evidence that all low readings are false or that every helium test arrangement has the same limitation. Packages filled with tracer before sealing, continuously supplied test arrangements and combined methods require their own validated interpretation.
Specify coverage before selecting the test sequence
Begin with the governing drawing, contract or product test specification. Identify the method, required revision, package free volume, permitted pressure differential, test media and acceptance rules. Electronic and semiconductor packages may reference a seal-test standard, but that standard is not automatically applicable to every laser-welded enclosure.
Do not impose a universal instruction to run gross testing first or fine testing first. Conditioning, exposure to liquids, pressure history and tracer retention can affect subsequent measurements. Use the sequence required by the qualified procedure and document any allowed retest route.
The supporting equipment overview should be used to plan the detector, fixture and handling process together. Buying an instrument with a low sensitivity specification does not establish the package-level test coverage.
Evaluate whether the package can tolerate the method
An enclosure may be sensitive to immersion, elevated temperature, pressure changes or residues. Thin lids can deform, and windows or feedthroughs can impose limits different from those of the metal seam. Discuss these restrictions before selecting conditioning and test equipment.
For electronic packaging, review the whole assembly. Internal free volume is the space available to gas, which can differ substantially from a volume estimated using external dimensions. Internal components and cavities may also affect gas exchange.
When a product cannot tolerate a proposed method, identify an alternative that is both permitted and demonstrated for its failure modes. A convenient shop-floor check should not silently replace the required acceptance test.
Demonstrate the measurement process
A useful development plan separates instrument performance from package-method performance. A calibrated reference leak checks detector response. It does not by itself demonstrate that the package receives sufficient tracer, retains it during transfer or is correctly presented to the fixture.
NIST describes calibrated leak artifacts as tools for checking helium leak detectors and notes the temperature dependence of permeation artifacts. Follow the certificate and stabilization requirements of the actual reference used. Then evaluate the complete test cycle with suitable reference parts or known conditions defined by the test specialist.
Record exposure and transfer timing, fixture identification, background, calibration checks and package condition. Investigate variation near the acceptance limit rather than repeatedly testing until a passing number appears.
Interpret the two results together
| Test outcome | Interpretation to investigate | Appropriate next step |
|---|---|---|
| Gross fail, fine reading low | A large path may have depleted tracer, or either test may be invalid | Hold the part and review both procedures; do not override the gross failure |
| Gross pass, fine fail | A smaller path may be present; background or fixture leakage may also contribute | Confirm measurement validity and localize the leakage |
| Both fail | A package problem or a shared test problem requires investigation | Check setup controls and inspect all package boundaries |
| Both pass | The part meets those test criteria if both measurements were valid | Apply the remaining drawing and product acceptance checks |
| Results change on repeated tests | Timing, tracer depletion, conditioning or unstable setup may be involved | Use the documented retest procedure and preserve the original results |
This table is a troubleshooting framework. The actual disposition of a production part must follow its approved acceptance procedure.
Separate leak detection from leak localization
A complete-package failure may originate at the lid weld, a feedthrough, a window, a fill port or another seal. Review the package construction and use an appropriate localization method before changing the laser recipe.
If evidence points to the weld, compare the location with the start, stop, corner and fit-up records. Use representative sections or other qualified inspection methods to investigate the mechanism. A leak reading alone cannot distinguish a crack from incomplete fusion, and increasing energy is not a reliable response to every leak failure.
Questions for a test-equipment discussion
Ask the supplier or test laboratory to explain:
- Which gross- and fine-leak failure ranges the proposed method covers for this package.
- Which package volumes, materials and pressure limits were assumed.
- How exposure, transfer and measurement timing will be controlled in production.
- How fixtures, reference leaks and background checks establish measurement validity.
- How failures, invalid tests and retests are recorded and handled.
- Which other inspections remain necessary after leakage acceptance.
JMLASER can incorporate these requirements into discussions of the welding system and workflow. A clear test plan gives process development a useful target: a reproducible package result supported by evidence, rather than an isolated detector display.