Low-Leakage Testing of Manual Bellows Globe Valves: Standards, Methods, and Fugitive Emission Compli

Manual bellows globe valves are the preferred zero-leakage shut-off solution for chemical, petrochemical, hydrogen, and high-purity applications where fugitive emissions must be kept to an absolute minimum. Low-leakage testing, performed in accordance with ISO 15848, ISO 5208, EN 12266, and TA Luft requirements, quantitatively verifies that the bellows seal, stem area, and seating surfaces achieve leak rates in the 10⁻⁴ to 10⁻⁶ mbar·L/s range, both when new and after simulated service cycling. This article explains the sample requirements, test items, test methods, acceptance criteria, and typical applications of low-leakage testing for manual bellows globe valves.

Technical Overview

A manual bellows globe valve is a linear-motion shut-off valve in which a handwheel-driven stem moves a disc onto a seat, and the stem penetration through the bonnet is sealed by a welded metal bellows. The bellows—typically a multi-ply (two- or three-ply) convolution assembly made of austenitic stainless steel or nickel alloy—has its lower end welded to the stem and its upper end welded to the bonnet, creating a frictionless, all-metal primary barrier between the process fluid and the atmosphere. A graphite packing set above the bellows serves as a secondary safety seal in the unlikely event of bellows failure, and many designs incorporate a leak-detection port between the two seals.

Because the bellows replaces the sliding stem-to-packing interface that is the dominant leakage path in conventional globe valves, this design is specified wherever emission to atmosphere must approach zero: toxic, flammable, or explosive media; high-purity and vacuum systems; heat transfer circuits; and process units subject to fugitive emission control and leak detection and repair (LDAR) programs. Low-leakage testing is the formal, quantitative verification that a valve design—or an individual production valve—keeps external leakage (bellows/stem area) and internal leakage (seat) within defined limits, both in the as-manufactured condition and after simulated service consisting of mechanical and thermal cycling.

The testing framework for manual bellows globe valves rests on the following key documents:

• ISO 15848-1 — Fugitive emission type testing and classification of valves, covering measurement procedures, tightness classes, and qualification through mechanical and thermal cycling.
• ISO 15848-2 — Production acceptance testing of valves for fugitive emissions, applying simplified leak measurement to individual manufactured valves.
• ISO 15848-3 — Alternative qualification procedures for low-emission valves.
• ISO 5208 — Pressure testing of metallic valves, defining shell strength and seat tightness tests and leak rate designations.
• EN 12266 — Industrial valves, testing of metallic valves, specifying pressure tests and leakage rates.
• TA Luft — The German Technical Instructions on Air Quality Control, which sets widely referenced maximum permissible leak rates for valve stem seals.
• Relevant GB/T and JB/T product standards for steel globe valves and bellows globe valves, which define design, materials, and factory test requirements.

Testing Samples

Type-Test Samples (Design Qualification)

Type testing per ISO 15848-1 is performed on representative production valves that embody the exact design to be qualified: identical body-bonnet construction, bellows geometry and ply configuration, welding procedure, packing type, and materials of construction. The selected specimens must cover the pressure class and size range claimed in the qualification scope. Where required, an additional bellows assembly may be separately subjected to pressure and cycle-life verification to establish the fatigue basis of the seal design.

Production Acceptance Samples

Production acceptance testing per ISO 15848-2 applies to individual valves or defined lots leaving the factory. Shell and seat pressure tests are normally carried out on 100% of valves, while the external low-leakage check—using helium or methane as the test gas—is performed at ambient temperature with a limited number of operations, typically measuring the stem/bellows leak rate before and after a specified number of open-close cycles.

Component-Level Samples

Bellows subassemblies and their attachment welds may be examined as discrete samples. Typical component testing includes proof-pressure testing of the bellows at shell test pressure, non-destructive examination of bellows convolution and end-plate welds by liquid penetrant methods in accordance with ASTM E165, and verification of bellows cycle life against the manufacturer's declared fatigue rating.

All samples must be clean, dry, and fully assembled as delivered, complete with handwheel, packing, and gaskets, and must be operated at the torques recorded during factory assembly so the test reflects real service behavior.

Testing Items

Depending on whether the objective is design qualification or production verification, a low-leakage test program for manual bellows globe valves covers the following items:

• Shell (body-bonnet) strength and tightness under hydrostatic test pressure.
• Seat (closure) tightness with liquid and with gas, graded against the leakage rates of ISO 5208 or EN 12266.
• External leak rate of the bellows/stem seal area, measured with helium or methane and expressed in std cm³/s, mbar·L/s, or ppmv.
• Mechanical endurance: repeated full open-close cycling with intermediate leak measurements.
• Thermal endurance: cycling between ambient and the specified temperature extremes with leak measurement under pressure at each extreme.
• Bellows pressure integrity and weld soundness, including penetrant examination of bellows welds.
• Operating torque verification throughout the test sequence.
• Optional material verification of bellows, body, and seat sealing surfaces.

Testing Methods

Shell and Seat Pressure Testing

Shell strength is verified hydrostatically, typically at 1.5 times the rated pressure at test temperature, with no visible leakage permitted through the pressure-retaining envelope. Seat tightness is then verified with gas (commonly at 1.1 times the rated pressure) or liquid, and results are graded per ISO 5208—rate A, meaning no visible leakage during the specified test duration, being the normal requirement for bellows globe valves intended for severe service.

Helium Vacuum Method (Mass Spectrometer Leak Detection)

In the vacuum method of ISO 15848-1, the valve stem and bonnet area are enclosed in a sealed chamber that is evacuated and connected to a helium mass spectrometer leak detector. The valve is charged with helium at the specified test pressure, and any helium escaping through the bellows or secondary packing is captured in the chamber and quantified. Because the technique measures total leakage with a sensitivity reaching the 10⁻⁹ std cm³/s order, it is the most rigorous method and is used to qualify the tightest helium classes (e.g., Class AH, on the order of 10⁻⁶ std cm³/s).

Sniffing Detection Method (Methane)

In the sniffing method, the valve is pressurized with methane (or helium, where specified) and a calibrated portable detection instrument is used to sample the atmosphere around the stem/bellows interface with a defined probe distance and traverse speed, following the procedure defined in ISO 15848-1. The measured concentration, corrected against a calibrated background, is reported in ppmv and compared with the methane tightness classes (AM, BM, CM). This method best simulates real hydrocarbon service and is also the basis of portable on-site monitoring used in LDAR programs.

Mechanical and Thermal Cycling Sequence

A full type-test sequence begins with a baseline leak measurement in the as-delivered condition. The valve is then subjected to the required number of mechanical open-close cycles—executed at the documented operating torque, with leak checks repeated at prescribed intervals. Thermal endurance follows: the pressurized valve is cycled between ambient and its qualified temperature extremes, with leak measurement at each plateau. Final leak verification after all cycling confirms that the bellows seal, packing backup, and seat have retained their integrity; ISO 15848-1 endurance (CO) and temperature (TH) class designations record the severity the valve has withstood.

Acceptance Criteria and Classification

• Helium tightness classes AH, BH, and CH, expressed in std cm³/s, with Class AH representing the tightest criterion—commonly on the order of 10⁻⁶ std cm³/s.
• Methane tightness classes AM, BM, and CM, expressed in ppmv, assessed by the sniffing method.
• TA Luft benchmark: a maximum stem-seal leak rate of 1 × 10⁻⁴ mbar·L/s, with well-made bellows-sealed designs routinely achieving rates in the 10⁻⁶ range.
• Seat leakage rate A (no visible leakage) or rate B per ISO 5208, according to the purchase specification.

Equipment, Calibration, and Traceability

Leak detectors must be calibrated using certified reference leaks, pressure instrumentation must carry valid traceable calibration, and the test bench must record pressure, temperature, torque, and cycle count continuously. Testing is conducted by accredited third-party laboratories or by manufacturer laboratories with accredited quality systems, and all measurement uncertainty is documented in the test report so results can be defended to regulators and end users.

Applications

Low-leakage testing of manual bellows globe valves is demanded wherever regulated emissions, personnel exposure limits, product purity, or process safety impose quantitative leakage limits:

• Petrochemical and refining units, for control of VOC fugitive emissions and compliance with emission inventories and LDAR regulations.
• Chemical processing, for toxic, corrosive, or carcinogenic media where any stem leakage is unacceptable.
• Oil and gas production and gas transmission, for methane slip reduction and corporate emission-reduction commitments.
• Hydrogen energy systems, where the small hydrogen molecule demands the most stringent seal verification.
• Semiconductor fabrication and specialty gas supply, where parts-per-million purity requires near-hermetic isolation.
• Pharmaceutical and fine chemical plants, for solvent vapor containment under occupational exposure limits.
• Nuclear and conventional power generation, for radioactive, steam, and heat transfer circuits.
• Vacuum systems and cryogenic service, where ingress or egress of even trace gas affects process integrity.

FAQ

What leak rate counts as "low leakage" for a bellows globe valve?

It depends on the specification invoked. Under ISO 15848-1, the tightest helium class (AH) corresponds to a leak rate on the order of 10⁻⁶ std cm³/s, while TA Luft sets the widely used benchmark of 1 × 10⁻⁴ mbar·L/s for stem seals. A properly manufactured and tested bellows globe valve typically meets or beats both figures.

Should the test gas be helium or methane?

Helium with vacuum-method mass spectrometry offers the highest sensitivity and, because helium is a smaller, inert molecule, provides a conservative proxy for most media. Methane sniffing better represents actual hydrocarbon service and is required when the purchase specification invokes methane-based classes. Many qualification programs combine both.

How many operating cycles must the valve withstand?

For production acceptance testing per ISO 15848-2, only a limited number of operations is required. For type testing per ISO 15848-1, the valve must pass leak checks throughout the mechanical endurance class claimed by the manufacturer—ranging from hundreds to tens of thousands of cycles—and most bellows designs are certified for a fatigue life of 10,000 full-stroke cycles or more.

What is the difference between type testing and production testing?

Type testing (ISO 15848-1) qualifies a valve design on representative samples through the full sequence of mechanical and thermal cycling, assigning tightness, endurance, and temperature classes. Production testing (ISO 15848-2) verifies individual manufactured valves or lots with a simplified ambient-temperature leak test plus the standard ISO 5208 shell and seat pressure tests.

Can leakage be checked after the valve is installed?

Yes. Portable sniffing instruments using the same detection principle as laboratory sniffing tests are routinely used for periodic LDAR monitoring of installed valves. However, formal qualification or certification data must always come from controlled laboratory testing.

What documents are delivered after testing?

A complete low-leakage test report includes the valve identification and design data, test gas, pressures and temperatures, leak measurement method and instrument calibration details, measured leak rates at every measurement point, the number of mechanical and thermal cycles completed, and the resulting ISO 15848 tightness and endurance classifications, together with a low-emission qualification certificate where the specification requires one.