Inter-Ply Leak Detection Testing for Multi-Ply Bellows in Bellows-Sealed Globe Valves

Inter-ply leak detection is a specialized non-destructive test that verifies the individual layer integrity of multi-ply metal bellows used as the primary stem seal in bellows-sealed globe valves. This article explains the failure mechanisms behind inter-laminar leakage, the applicable standards such as ISO 5208, ISO 15848, and API 598, and the laboratory methods used to detect it, including helium mass spectrometry, interlayer pressure-decay testing, and bubble emission testing. It is written for valve manufacturers, plant reliability engineers, and quality inspectors responsible for zero-leakage sealing performance and fugitive emission control.

Technical Overview

Bellows-sealed globe valves use a welded metal bellows as the primary stem seal instead of, or in combination with, conventional gland packing. The bellows is typically a multi-ply construction consisting of two to five layers of thin metallic foil, each 0.05–0.15 mm thick, formed into convolutions and joined by longitudinal seam welds and circumferential end welds. The multi-ply design provides built-in redundancy: if one ply develops a defect, the remaining plies continue to contain the process medium while the valve remains in service until scheduled maintenance.

Inter-ply leakage refers to the migration of fluid or tracer gas through a defect in one ply into the space between adjacent layers. Defects can originate from incomplete fusion in longitudinal seam welds, pinholes in the foil, micro-cracks induced during convolution forming, or fatigue cracking after extended cycling. Inter-ply leakage is dangerous precisely because it is hidden: the valve may still pass conventional shell and seat pressure tests while the protective redundancy of the multi-ply construction has already been lost.

Inter-ply leak detection therefore isolates each layer boundary of the bellows and verifies that every ply is individually gas-tight. This is achieved by pressurizing the interlayer cavity through a dedicated monitoring port, or by introducing a tracer gas on one side of a ply and detecting its appearance on the other side. The test is performed on bellows components at the manufacturing stage and repeated at the assembled-valve level as part of production acceptance and type qualification programs.

Testing Samples

Multi-Ply Bellows Assemblies

Standalone bellows assemblies are submitted in the clean, dry, as-welded condition after seam welding, forming, and heat treatment. Common materials include:

• Austenitic stainless steels such as 304, 304L, 316, and 316L
• Nickel-based alloys such as Alloy 625, Alloy 718, and Alloy C-276 for corrosive or high-temperature service
• Two-ply, three-ply, and five-ply configurations with total wall thicknesses adapted to the valve pressure class

Complete Bellows-Sealed Globe Valves

Finished valves of DN 8 (NPS 1/4) through DN 400 (NPS 16), typically in pressure classes from PN 10 to PN 400 or ASME Class 150 to Class 2500, are tested with the bellows mounted and the interlayer monitoring port accessible. Both rising-stem and bellows-welded-to-bonnet designs can be evaluated.

Qualification and Prototype Test Specimens

For type testing programs, instrumented specimens equipped with permanent interlayer pressure ports and thermocouples are used to characterize leak behavior across thermal and mechanical cycling, in accordance with qualification test plans referencing ISO 15848-1.

Testing Items

Depending on the stage of production and the client specification, an inter-ply leak detection program may include the following test items:

• Helium leak rate measurement of each ply boundary, typically with acceptance limits of 1 × 10⁻⁶ mbar·L/s or stricter, down to the 1 × 10⁻⁹ mbar·L/s range for critical service
• Interlayer cavity pressure-decay testing at 1.1 times the bellows design pressure with a calibrated hold period
• Immersion bubble emission testing for gross leak screening of seam and end welds
• Bellows proof pressure and burst pressure verification
• Valve shell, seat, and backseat tightness testing per ISO 5208, API 598, or GB/T 13927
• Fugitive emission type testing and production acceptance testing per ISO 15848-1 and ISO 15848-2
• Post-cycling leak verification after mechanical and thermal endurance cycling to confirm that ply integrity survives the qualified stroke count

Acceptance criteria are agreed before testing. For zero-leakage applications such as toxic, lethal, or high-purity service, a common requirement is no detectable leakage above the background sensitivity of the instrument, no measurable pressure decay over a 5 to 15 minute hold, and zero continuous bubble streams during immersion testing.

Testing Methods

Helium Mass Spectrometry Leak Detection

Helium tracer gas methods offer the highest sensitivity available for inter-ply testing. In the pressurized sniffer configuration, the internal volume of the bellows is charged with helium or a helium-nitrogen mixture at the specified test pressure, and a mass spectrometer sniffer probe is applied to the interlayer monitoring port or weld seams. In the vacuum hood configuration, the interlayer cavity is evacuated and connected to the mass spectrometer while helium is applied to the opposing side of the ply. Test procedures follow recognized leak testing practices such as ASTM E493 and ISO 20485, with calibrated reference leaks used to verify instrument sensitivity before and after each test sequence.

Interlayer Pressure-Decay Testing

The space between two adjacent plies is sealed and pressurized with dry nitrogen or air through the monitoring port, typically to 1.1 times the design pressure differential. The port is then isolated and the cavity pressure is recorded with a high-resolution absolute pressure transducer over a defined stabilization and hold period. Temperature is monitored continuously so that thermal drift can be corrected, since a small temperature change can mimic a pressure loss. Any monotonic pressure decay beyond the calculated threshold indicates a through-wall ply defect.

Immersion Bubble Emission Testing

For gross leak screening, the bellows interlayer is pressurized and the entire assembly is submerged in a temperature-stabilized water or surfactant bath. A continuous, repeatable stream of bubbles tracing back to a specific convolution or seam location identifies the defective ply and often localizes the leak path. This method, consistent with ASTM E515, is frequently used as a pre-screen before helium testing and as a diagnostic follow-up to confirm the physical location of a detected leak.

Assembled-Valve Verification

After the bellows is welded into the valve, the complete assembly undergoes hydrostatic shell and seat testing per ISO 5208 or API 598, followed by monitoring of the bellows vent or monitoring port. Any medium or gas appearing at this port confirms that the inner ply boundary has been breached, and the valve is rejected regardless of whether the outer plies still hold. For fugitive emission compliance, the assembled valve is additionally challenged per ISO 15848-1 or ISO 15848-2, where bellows-sealed valves are exempted from packing leak rate classes only if bellows integrity is demonstrated throughout the cycling and thermal regimes.

Applications

Inter-ply leak detection is essential wherever bellows-sealed globe valves are relied upon for absolute containment. Key application sectors include:

• Petrochemical and chemical processing, including chlorine, vinyl chloride monomer, phosgene, and other acutely toxic services
• Lethal service piping systems where regulatory and corporate standards require sealed-stem valves with verified redundancy
• High-purity and specialty gas systems in semiconductor and pharmaceutical manufacturing
• Steam, heat transfer oil, and vacuum services where external leakage cannot be tolerated
• Hydrogen and fuel gas isolation, where small molecular size demands the highest seal integrity
• Nuclear power auxiliary systems governed by nuclear-grade valve qualification requirements

Accredited third-party laboratories routinely perform these tests for factory acceptance, third-party witness inspection, failure analysis of returned bellows, and periodic requalification of valves in long-term storage.

FAQ

Why can a valve pass standard shell and seat tests but still have inter-ply leakage?

Shell and seat tests verify the pressure boundary between the process and the outside of the valve, not the boundaries between individual bellows plies. A leak in the inner ply allows medium into the interlayer space, but the outer plies may still contain it completely, so external tests show no leakage. Only a test that pressurizes or monitors the interlayer cavity itself can reveal the defect.

What leak rate is considered acceptable for inter-ply tightness?

For general industrial service, manufacturers commonly specify no detectable leakage at instrument sensitivity, which is typically 1 × 10⁻⁶ mbar·L/s or better with helium methods. For toxic, nuclear, and high-purity applications, criteria of 1 × 10⁻⁸ to 1 × 10⁻⁹ mbar·L/s are frequently imposed, and pressure-decay tests must show zero measurable decay within the resolution of the transducer.

Which standards govern inter-ply leak testing of bellows globe valves?

No single standard is dedicated exclusively to inter-ply leakage, so test programs are typically built from a combination of ISO 5208 or API 598 for valve pressure testing, ISO 15848-1 and ISO 15848-2 for fugitive emission qualification, ASTM E493 and ISO 20485 for helium tracer methods, ASTM E515 for bubble emission testing, GB/T 13927 for valve pressure testing under the Chinese standards system, and EJMA guidance for bellows design and qualification.

Can inter-ply leakage be monitored on valves already installed in a plant?

Yes. Many bellows-sealed globe valves are supplied with a permanent interlayer monitoring port or telltale. During operation, a pressure gauge, portable helium sniffer, or periodic bubble check applied to this port provides an early warning that the inner ply has failed while the redundant outer plies continue to provide containment, allowing the valve to be scheduled for replacement before any external leakage occurs.