What fails first on a cryogenic valve is exactly what a ambient test cannot see: different contraction rates at -196 °C, changed sealing stress across the temperature gradient, trapped liquid expanding in the body cavity, ice forming at the stem packing. None of this shows up on a 20 °C hydrostatic bench — it shows up on site at the first cool-down. Cryogenic (low-temperature) testing is therefore not “the ambient pressure test moved into a cold box”; it is a separate type-test sequence in which cold seat tightness, external leakage, cold cycling and torque each have their own criteria.
This article separates the subject: how cryogenic testing differs from ambient pressure testing, how ISO 28921-2 runs, what BS 6364 and MSS SP-134 govern, what GB/T 24925 requires in China, what the bench must provide, and what to check at acceptance. All figures follow the official texts referenced by the project; the tables are for quick orientation in sourcing and quotations.
1. Four differences from ambient pressure testing
- Temperature is itself a test condition: an ambient test only needs stable media temperature; a cryogenic test cools the body to the design minimum temperature (commonly -50 °C or -196 °C) and holds until stable — the temperature record is part of the report;
- Different acceptance criteria: soft seats are judged to ISO 5208 Rate A at ambient, but when cold the elastomer hardens and shrinks, and leakage is judged against size-scaled limits (table below) — ambient classes cannot simply be carried over;
- Operability must be measured: cold cycling count and external leakage are hard requirements, and cold breakaway torque is often well above the ambient figure — sizing an actuator on ambient torque leads to field stalls;
- Different safety envelope: liquid nitrogen, cold gas, helium detection, oxygen deficiency and cold burns make the test cell’s shielding and ventilation part of the facility design.
2. ISO 28921-2: the main type-test sequence
ISO 28921 has two parts: Part 1 covers design, manufacturing and production testing (including the DN/NPS and PN/Class coverage), Part 2 covers type testing over -50 °C to -196 °C. It does not apply to valves for cryogenic vessels designed to ISO 21011.
| Step | ISO 28921-2 requirement |
|---|---|
| Test temperature | -196 °C ±5 °C or -50 °C ±5 °C, or an intermediate temperature; never above the valve’s design minimum temperature |
| Ambient pre-tests | Shell: 1.1× CWP with gas, 1.5× CWP with water or alcohol; seat: (6 ± 1) bar; soft seats to ISO 5208 Rate A, metal seats per the manufacturer’s standard |
| System proving | External leak check of valve and connected piping at the lower of maximum CWP or maximum seat test pressure, using soap solution or helium |
| Cavity purge | Continuous (2 ± 0.5) bar test gas throughout cool-down to keep moisture out and prevent icing |
| Cooling | Liquid immersion with the level at least covering the body-to-bonnet joint; spraying must cover the whole valve; cold-gas method must expose the body and joint |
| Cold seat test | Low pressure (2 ± 0.5) bar first, then four pressure stages up to CWP, each recorded |
| Cold cycling | 200 operating cycles for on-off valves; 3 cycles for check valves |
| External leakage | Helium detection ≤50 ppmv (test gas purity ≥97%) |
| Seat leakage limits | Scaled by class and size: PN16/Class 150 ≤12 mm³/(s·DN); PN100/Class 600 ≤100; PN250/Class 1500 ≤250 |
The bench layout is also specified: the cold box at least 100 mm from the valve, the liquid level about 25 mm above the body-to-bonnet joint, and at least three thermocouples monitoring body, bonnet and medium. The cold box, blind-flange supports and cryogenic piping need insulation, the test area needs protective barriers, and operators need cryogenic-test training.
3. BS 6364 and MSS SP-134: the extended bonnet question
BS 6364 is the British cryogenic valve specification. It was withdrawn by BSI in 2021 but is still widely cited in LNG project specifications. Its type test is homologous to ISO 28921-2: a prototype immersed in liquid nitrogen, cycled cold, helium-measured for seat and stem leakage, and checked for cold operability. When checking the certificate, confirm the design family coverage (size, class, construction, seat material) as usual.
MSS SP-134 answers the engineering question of how long the bonnet must be: the extended bonnet creates a stable vapour column between the cold medium and the packing, keeping the packing above 0 °C. Two installation constraints follow — the stem must point upward (deviation from vertical per the manufacturer, commonly within ±45°), and the extension must not be wrapped in insulation, which would draw cold up and short-circuit the design. Bonnet length, installation angle and insulation practice belong in the technical agreement; “cryogenic type” alone is not a specification.
4. The Chinese reference: GB/T 24925-2019
GB/T 24925-2019 “Low temperature valve — Technical specifications” covers PN16–PN400, DN15–DN1200, Class 150–2500, NPS 1/2–48, for gate, globe, check, ball and butterfly valves from -196 °C to -29 °C. The clauses most relevant to testing:
- Double-seated valves must have a cavity relief hole of at least 3 mm, vented preferably to the upstream high-pressure side and marked; valves with relief features must undergo a relief test;
- Non-metallic-seated valves in flammable service need an anti-static design, with discharge path resistance ≤10 Ω;
- Where fire resistance is required, GB/T 26479 (fire test for resilient-seated quarter-turn valves) and GB/T 26482 apply, or as agreed;
- Stem operation: non-cold-box valves must be operable at no less than 45° from the ground plane; cold-box valves within 15°–90°.
In other words, a cryogenic valve is often accepted against three certificates stacked together: cryogenic type testing (ISO 28921-2 or BS 6364), fire resistance (GB/T 26479 / ISO 10497 / API 607), and fugitive emissions (ISO 15848-1 where specified). Their conditions and criteria do not substitute for one another.
5. What the cryogenic bench must provide
- Cold source and cold box: liquid nitrogen storage, vaporizer, insulated cold box and level control; liquid level, immersion depth and box spacing must reproduce the standard’s conditions;
- Temperature measurement: at least three thermocouples (body, bonnet, medium) with recording and curve output, plus a controlled cooling rate to avoid thermal shock;
- Pressure circuits: ambient pre-tests (shell 1.1/1.5× CWP, seat 6 bar) and the cold stage (2 bar low pressure up to CWP in four stages) must be switchable, with the piping itself pre-cooled and system-proven;
- Purge and drying: dry-gas purge at 2 ± 0.5 bar throughout cool-down, plus post-test drying and dew-point checks — residual water freezes on the first cool-down and gets misread as valve leakage;
- Measurement chain: seat leakage metering (volumetric or flow), helium detector with ≤50 ppmv resolution, and cold torque measurement with recording;
- Safety: cold-burn protection, oxygen monitoring and ventilation, helium venting and test-area barriers — the main reason a cryogenic bench costs far more than a hydrostatic one.
6. Six things to check at acceptance
- The type certificate covers the delivered valve’s size, class, construction and seat material (a 2-inch Class 150 certificate does not cover a 12-inch Class 300 valve);
- The test temperature is at or below the design minimum temperature, with complete temperature curves and soak times;
- Seat leakage is recorded per pressure stage, with units and limit scaling consistent with the contract (mm³/(s·DN) or bubble counts);
- Cold cycling (200 cycles / 3 for check valves) and cold torque records are complete;
- External helium leakage is ≤50 ppmv, with test gas purity ≥97%;
- Bonnet length, installation angle, relief-hole direction and anti-static resistance (≤10 Ω) match the agreement.
FAQ
Can a passed ambient pressure test replace a cryogenic test?
No. The ambient test verifies strength and tightness at room temperature; cold contraction, changed sealing stress, trapped cavity liquid and packing icing are all outside its scope. Cryogenic testing is a separate type test, and its report cannot be derived from ambient data.
Must cryogenic testing go to -196 °C?
Not necessarily. ISO 28921-2 allows -196 °C, -50 °C or an intermediate temperature, but the test temperature may not be above the design minimum temperature. LNG service (-162 °C) is normally covered by liquid nitrogen at -196 °C; refrigeration and low-temperature cooling duties can use the -50 °C stage.
Is the 50 ppmv helium limit for seat leakage or external leakage?
External leakage. The 50 ppmv figure in ISO 28921-2 is the post-test external leak detection limit; seat leakage is judged separately against the size-scaled mm³/(s·DN) values. Keep the two criteria on separate record sheets.
Specific parameters, criteria and scopes of application follow the official ISO 28921-1/-2, BS 6364, MSS SP-134 and GB/T 24925-2019 texts referenced by the project; the tables above are for quick orientation in sourcing discussions. Cryogenic valves are often required to hold two further type tests at the same time: fire resistance is covered in ISO 10497 vs API 607, ambient tightness classes in ISO 5208 leakage classes; the pressure-rating check logic is in the PN vs CWP checklist, and the scopes of all referenced standards are summarized in the Valve Test Standards Index.
Related test equipment
Each test above maps to a bench. Pick by valve type and by the item you need to prove:
- Pipe and connection leak test bench — pressure-decay and helium methods, judged against a leak-rate limit
- Valve comprehensive test bench — pressure, tightness, flow, torque and life cycling on one clamping system
- Valve torque test machine — break-to-open and seating torque curves for actuator sizing
- Valve testing equipment range — browse all models by valve type and test item