Helium Leak Testing: When You Need It, and When You Don't

Helium leak testing is often specified when a cheaper method would have done, and occasionally omitted where nothing else would work. The difference comes down to one number: the leak rate you need to detect.

Sensitivity, in figures

This is the comparison that decides the method. Typical detectable leak rates, in mbar·L/s:

Method	Typical detection limit
Visual inspection	Only gross, visible leakage
Soap bubble / foam	~10⁻⁴
Pressure decay	~10⁻³ to 10⁻⁴
Vacuum box	~10⁻⁴ to 10⁻⁵
Helium — sniffer (detector probe)	~10⁻⁶ to 10⁻⁷
Helium — hood / vacuum (mass spectrometer)	down to ~10⁻¹²

Figures vary with equipment, operator technique and test conditions, and should be confirmed against the acceptance criteria in your specification. But the scale of the difference is the point: helium in vacuum mode detects leaks several orders of magnitude smaller than bubble testing.

That is why helium gets specified — and equally why it is overkill when your acceptance criterion is 10⁻³.

Why helium

Helium works as a tracer because of a specific combination of properties, not because it is exotic:

• Small molecular size — it passes through leak paths other gases won’t.
• Inert and non-flammable — safe to introduce into most systems without reaction or hazard.
• Rare in atmosphere (roughly 5 parts per million) — background concentration is low, so detection has a clean baseline. This is the property most often overlooked, and it is what makes quantitative measurement possible.
• Non-toxic, so testing does not require the controls other tracer gases would.

The three methods

Vacuum (hard vacuum) method. The component is evacuated and connected to a mass spectrometer leak detector; helium is applied externally. This is the most sensitive configuration and gives a quantitative total leak rate. It requires the component to withstand vacuum.

Sniffer (detector probe) method. The system is pressurised with helium and a probe is passed over joints, welds and seals. Less sensitive by several orders of magnitude, but it locates the leak rather than just measuring total leakage. Usually the practical choice on installed plant that cannot be evacuated.

Hood method. The pressurised component is enclosed and the enclosure sampled. Gives a total leak rate for the whole assembly without locating individual leaks.

In practice these are frequently combined: hood or vacuum to establish whether total leakage is within specification, then sniffer to find the source if it isn’t.

Governing standards

Helium leak testing is not an informal procedure. The methods, acceptance criteria and personnel qualification are defined in published standards, and your specification should name which applies:

• ASME Boiler and Pressure Vessel Code, Section V, Article 10 — leak testing, including helium mass spectrometer techniques. The usual reference for pressure equipment.
• ASTM E493, E498 and E499 — mass spectrometer techniques: inside-out, detector probe, and tracer probe respectively.
• EN 1779 — criteria for selecting leak testing method and technique.
• ISO 20485 — tracer gas method.

Ask any provider which standard they will test to and what the acceptance criterion is, before work starts. If the answer is vague, so will the report be.

When helium is the right choice

• Acceptance criteria tighter than roughly 10⁻⁵ mbar·L/s
• Vacuum systems, where any inward leakage compromises function
• Sealed assemblies where leakage cannot be tolerated at all
• High-value or hazardous contents where the cost of a leak far exceeds the cost of testing
• Where the specification or client requires it — which, on ADNOC and similar projects, it often does

When something cheaper will do

• Acceptance criteria in the 10⁻³ to 10⁻⁴ range, where pressure decay or bubble testing is sufficient and considerably faster
• Structural integrity verification, which is what hydrostatic testing is for — that is a strength test, not a leak test, and the two are not interchangeable
• Storage tank floor and wall seams, where vacuum box testing is the established method
• Situations where the component cannot hold pressure or vacuum

A provider who specifies helium for every application is not assessing your requirement.

Typical applications

Pressure vessels, heat exchangers, condensers, storage systems, process piping, valves, compressors, vacuum chambers, refrigeration systems and precision-engineered assemblies. Most commonly at pre-commissioning, during shutdown, after repair or modification, and before restart.

The economics are usually straightforward: finding a leak before startup is dramatically cheaper than finding it during operation.

Accreditation, and what it does and doesn’t cover

itmad is an EIAC-accredited inspection body under ISO/IEC 17020 (listing here), accredited since 2014. Our accredited NDT scope covers ultrasonic testing including phased array, magnetic particle, dye penetrant, eddy current, visual examination, hardness testing, infrared thermography and leak testing by the vacuum box technique to ASME Section V.

Accreditation under ISO/IEC 17020 is granted for defined activities rather than for a company as a whole, so any inspection body’s scope will cover some methods and not others. Helium leak testing sits outside our accredited scope — as it does for most providers in this market, since accredited helium testing is rarely specified.

Where it matters is if your specification or client explicitly requires an accredited inspection report. Tell us what the specification says and we will confirm before quoting whether we can meet it as written.

Common questions

How small a leak can helium testing actually find? In vacuum mode with a mass spectrometer detector, down to around 10⁻¹² mbar·L/s under good conditions. Sniffer mode is far less sensitive — roughly 10⁻⁶ to 10⁻⁷ — but locates the leak rather than only quantifying it.

Is it the same as a hydrostatic test? No. Hydrostatic testing verifies structural strength under pressure. Helium leak testing detects and measures leakage. A vessel can pass one and fail the other.

Does the helium contaminate the system? No. Helium is inert and does not react with system materials or residues. It is evacuated or vented after testing.

Why is it more expensive? Mass spectrometer detectors are specialised instruments requiring calibration and qualified operators, and helium itself is a consumable with a real cost. Test setup — evacuation, enclosure, or systematic probing — also takes longer than spraying soap solution on a weld.

Which method should we use? It depends on whether you need to locate the leak or quantify total leakage, whether the component can be evacuated, and what your acceptance criterion is. Those three answers determine the method.

---

Request a technical proposal — tell us your acceptance criteria and equipment type, or see our leak testing capabilities.