Globe valve bellows integrity testing verifies the leak-tightness, pressure resistance, and fatigue endurance of the metal bellows stem seal that prevents fugitive emissions in critical process services. This technical article explains the sample types, test items, helium leak detection and pressure testing methods, applicable standards such as ISO 15848, API 624, EN 12266, and GB/T 13927, and answers the most common questions about bellows quality assurance.
Technical Overview
A bellows-sealed globe valve replaces the conventional sliding stem packing with a welded metal bellows that expands and contracts with every stroke, creating a hermetic barrier between the process fluid and the atmosphere. Because the bellows is the only dynamic pressure-containing seal in the valve, any through-wall crack, weld defect, or fatigue failure results in an uncontrolled release of medium. Bellows integrity testing therefore focuses on demonstrating zero-leakage performance under pressure, temperature, and cyclic operating conditions before the component enters service.
Typical bellows failure modes include fatigue cracking at convolution roots and peaks, lack-of-fusion or porosity in the convolution-to-end-plate welds, intergranular corrosion thinning, stress corrosion cracking in chloride-bearing environments, and column squirm under excessive internal pressure. A complete integrity test program addresses all of these mechanisms through a combination of leak testing, proof pressure testing, cyclic endurance testing, non-destructive weld examination, and material verification, typically performed by accredited third-party laboratories or in-house quality departments.
- Leak-tightness verification using helium mass spectrometry leak detection
- Pressure strength verification through hydrostatic and pneumatic proof testing
- Fatigue endurance verification through full-stroke mechanical cycling
- Weld quality verification through liquid penetrant and radiographic examination
- Material conformity verification through chemical, mechanical, and metallographic analysis
The governing references for bellows integrity testing include ISO 15848-1 and ISO 15848-2 for fugitive emissions type and production testing, API 624 for rising-stem valve type testing, EN 12266-1/-2 and ISO 5208 for industrial valve pressure and leak testing, GB/T 13927 and GB/T 26479 for Chinese national valve and bellows globe valve requirements, EN 1779 for leak testing method selection, ASME BPVC Section V Article 10 for leak testing procedures, and EJMA guidelines for bellows design fundamentals.
Testing Samples
Individual Metallic Bellows
Standalone bellows elements, either hydroformed seamless-convolution types or welded-convolution (ripple-weld) types, supplied in single-ply or multi-ply configurations of two to three plies. Common materials include austenitic stainless steels such as 304, 316L, and 321, nickel alloys such as Inconel 625 and Inconel 718, Monel 400, Hastelloy C-276, and precipitation-hardening grades for demanding temperature or corrosion service. Component-level testing at this stage screens out defective bellows before value is added in downstream assembly.
Bellows Sub-Assemblies
Bellows units welded to their end collars, washers, inner support tubes, and anti-rotation guides. The convolution-to-end-plate circumferential welds are the most crack-sensitive locations in the entire assembly, so sub-assemblies are examined by penetrant testing and leak tested as a welded unit prior to installation into the valve bonnet.
Complete Bellows-Sealed Globe Valves
Fully assembled valves submitted for factory acceptance testing, production fugitive-emissions testing to ISO 15848-2, or full type testing to ISO 15848-1 or API 624. Valve-level testing validates the interaction between the bellows, stem, bonnet, backup packing, and body joint under realistic pressure and cycling conditions.
Material and Weld Coupons
Sample coupons from the same bellows material heat lot and production weld procedure qualification records, used for chemical composition analysis, tensile and hardness testing, intergranular corrosion evaluation per ASTM A262, and metallographic cross-sectioning of weld penetration depth.
Testing Items
A comprehensive bellows integrity test program typically covers the following items, selected according to the applicable product standard, purchase specification, and service severity.
- Helium leak rate and gas tightness of the bellows seal assembly
- Shell hydrostatic strength of the complete valve at 1.5 times rated pressure
- Seat tightness of the complete valve at 1.1 times rated pressure
- Bellows proof pressure at 1.5 times the maximum working differential pressure
- Burst pressure verification during design qualification
- Fatigue cycle life under full-stroke mechanical cycling at operating pressure and temperature
- Column stability and squirm resistance under internal pressure
- Liquid penetrant examination of all bellows circumferential and longitudinal welds
- Radiographic examination of end-plate attachment welds
- Dimensional verification of convolution geometry, free length, and stroke travel
- Spring rate and stem actuation force measurement
- Material verification by optical emission or XRF positive material identification
- Corrosion resistance evaluation for aggressive media services
Testing Methods
Helium Mass Spectrometry Leak Testing
The vacuum (inside-out) method is the primary technique for bellows qualification. The bellows or valve cavity is connected to a helium mass spectrometer leak detector and evacuated; helium is then sprayed onto the external bellows surface or filled inside the bellows, and any through-wall path allows helium to reach the detector. The vacuum method achieves sensitivities down to 1×10⁻⁹ Pa·m³/s and lower, which is orders of magnitude beyond the capability of water immersion testing. Typical production acceptance limits are 1×10⁻⁶ Pa·m³/s for general chemical service, tightened to 1×10⁻⁸ or 1×10⁻⁹ Pa·m³/s for high-purity gas, vacuum, and nuclear applications. The sniffer probe method is used on assembled valves to localize leaks at bellows welds and body joints where vacuum connections are impractical.
Hydrostatic Shell and Seat Testing
Complete valves undergo shell strength testing at approximately 1.5 times the rated pressure and seat tightness testing at approximately 1.1 times the rated pressure in accordance with EN 12266, ISO 5208, or GB/T 13927, with hold times specified by leak class and nominal size. The bellows itself is additionally pressurized at its maximum differential pressure to verify containment and check for visible convolution deformation or squirm.
Pneumatic and Immersion Bubble Testing
For sub-assemblies where a quick screening check is required, the bellows is pressurized with clean dry air or nitrogen and submerged in a water bath for a defined observation period. This method is economical but its sensitivity, roughly 1×10⁻³ Pa·m³/s, is sufficient only as an intermediate check and never replaces helium testing for final acceptance. Strict energy-isolation and shielding precautions apply because stored pneumatic energy is hazardous.
Pressure-Decay Testing
The sealed bellows volume is pressurized, isolated, and monitored with calibrated pressure transducers over time, with temperature compensation to eliminate false indications from ambient drift. Pressure-decay testing is suitable for high-volume production screening and for verifying multi-ply bellows by monitoring the inter-ply cavity, where a rising interlayer pressure indicates leakage from either the inner or outer ply.
Fatigue Cycle Endurance Testing
Bellows samples or complete valves are cycled through full rated travel, commonly 110 percent of nominal stroke, while pressurized at the rated or slightly elevated differential pressure, with optional elevated-temperature test environments up to the design temperature. Cycling continues to the specified number of operations or until leakage is detected. Purchase specifications commonly require between 2,000 and 20,000 full-stroke cycles depending on service, while valve-level fugitive emissions type testing to API 624 requires 610 mechanical cycles combined with 310 thermal cycles at elevated temperature with leakage monitored throughout.
Non-Destructive Weld Examination
Liquid penetrant testing to ISO 3452 or ASTM E165 is applied to 100 percent of bellows circumferential welds and longitudinal seam welds, examined from both sides where accessible, to reveal surface cracks, porosity, and undercut. Radiographic testing to ISO 17636 may be specified for thicker end-plate welds to confirm full penetration and absence of internal defects. Metallographic cross-sections prepared during procedure qualification verify weld penetration depth and fusion profile.
Material and Metallurgical Verification
Chemical composition is confirmed by optical emission spectrometry or XRF analysis against the certified material test report, and mechanical properties are verified by tensile and hardness testing. For corrosion-critical services, intergranular corrosion testing per ASTM A262 and microstructural examination of grain size and delta ferrite distribution are added to rule out sensitization that would compromise long-term bellows ductility.
Applications
Bellows integrity testing is essential wherever valve stem leakage cannot be tolerated, whether for environmental compliance, personnel safety, product purity, or process economics. Verified bellows-sealed globe valves are specified across the following industries.
- Petrochemical and refinery fugitive emission control programs, LDAR compliance, and procurement to ISO 15848 and European directives such as TA Luft and the IED
- Toxic and lethal media handling, including chlorine, phosgene, hydrogen fluoride, hydrogen sulfide, and ammonia services
- Nuclear power plants for radioactive borated water, sampling lines, and other contamination-critical circuits
- Semiconductor and high-purity gas delivery systems requiring leak rates at or below 1×10⁻⁹ Pa·m³/s
- High-vacuum processing equipment and cryogenic service where ingress or egress of molecules must be minimized
- Pharmaceutical and fine chemical plants handling toxic intermediates and high-value media
- Steam, heat transfer oil, and hydrogen energy systems with demanding temperature cycling
FAQ
What helium leak rate is acceptable for a bellows-sealed globe valve?
For general industrial service, an acceptance limit of 1×10⁻⁶ Pa·m³/s measured by the vacuum method is common. High-purity gas, vacuum, and nuclear applications typically specify 1×10⁻⁸ to 1×10⁻⁹ Pa·m³/s. Valve-level fugitive emissions acceptance is defined by the leak classes of ISO 15848-1 for type testing and ISO 15848-2 for production acceptance.
Why is the bellows leak tested separately before valve assembly?
Once the bellows is welded into the bonnet and the backup packing is loaded, small bellows leaks can be masked by the packing and become undetectable during routine valve tests. Component-level helium testing before assembly guarantees that every installed bellows is individually verified leak-tight, and it avoids the cost of scrapping a finished valve due to a defective bellows.
Can a water pressure test replace helium leak testing?
No. Hydrostatic testing verifies pressure strength but its leak detection sensitivity is limited to roughly 1×10⁻³ Pa·m³/s at best, which is a million times coarser than typical helium acceptance limits. The two methods are complementary: hydrostatic testing proves structural integrity, while helium mass spectrometry proves hermetic integrity.
Do standards require backup packing behind the bellows?
Yes, good engineering practice and most specifications require a secondary safety packing set above the bellows so that the valve remains sealed even if the bellows fails. This also allows monitoring of bellows condition in service through a leak-off port between the bellows and the packing in higher-criticality designs.
How many operating cycles must a bellows survive?
Requirements vary by specification, but typical purchase specifications demand between 2,000 and 20,000 full-stroke cycles verified by endurance testing at rated pressure, with type test environments up to the design temperature. Endurance classes defined in ISO 15848-1 are used to declare mechanical cycling capability during type testing.
How are multi-ply bellows verified?
Multi-ply bellows can be checked by pressurizing the inter-ply cavity and performing pressure-decay or helium testing on that cavity; pressure rise or helium detection between the plies confirms a breach in either the inner or outer wall before it becomes an external leak. Final acceptance is still based on whole-assembly helium leak testing of the complete bellows unit.
Should bellows integrity be re-verified in service?
Yes. For toxic, nuclear, and high-consequence services, periodic surveillance is recommended, including monitoring of leak-off connections between bellows and packing, review of cycle counts against qualified fatigue life, and replacement of the bellows cartridge when accumulated cycles approach the tested endurance limit. Any bellows exhibiting external leakage should be replaced rather than repaired.