Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
ISO 28921-1:2022 specifies requirements for design, material selection, manufacturing, and production testing of isolating valves for low-temperature applications. This second edition extends the scope to larger sizes (DN 950-1800, NPS 38-72) and higher pressure classes (PN 400, Class 2500) compared to the 2013 edition. Developed by ISO/TC 153 in collaboration with CEN/TC 69, this standard covers gate valves, globe valves, check valves, butterfly valves, and ball valves designed for service temperatures as low as -196°C (liquid nitrogen temperature).
The standard mandates specific material requirements for low-temperature service. Metallic materials must meet Charpy V-notch impact test requirements at the minimum design temperature or lower. Austenitic stainless steels (304/304L, 316/316L) are the standard choice for -196°C service, while 9% nickel steel and aluminum alloys are specified for specific applications. Internal non-metallic materials (seats, seals, gaskets) must be proven suitable for the intended temperature range through thermal cycling tests. The standard also specifies extended bonnet (stem extension) lengths to ensure that the stem seal operates above freezing even when the valve body is at cryogenic temperature.
| Component | Material Requirement | Temperature Limit | Testing Required |
|---|---|---|---|
| Body/bonnet | SS 304/304L, SS 316/316L, 9% Ni steel, or Al alloy | -196°C to +150°C | Charpy impact at Tmin – 5°C |
| Stem | SS 316/316L, SS 630 (17-4 PH), or Inconel 718 | -196°C to +250°C | Tensile + impact + intergranular corrosion |
| Seat rings | PTFE, PEEK, reinforced PTFE, or stellite | Depends on material | Thermal cycling (10 cycles min) |
| Gaskets | Expanded PTFE, flexible graphite, or spiral wound | -196°C to +400°C | Compression + leakage at temperature |
| Bolting | SS 304/304L or low-temperature alloy steel | -196°C to +150°C | Charpy impact + tensile |
The extended bonnet (also called the cold column or neck tube) is the most distinctive feature of cryogenic valves. Its length must ensure that the stem seal temperature stays above 0°C when the valve body is at -196°C, preventing ice formation on sealing surfaces. The required extension length depends on material thermal conductivity — for stainless steel, typical lengths range from 150 mm (DN 50) to 500 mm (DN 600). The standard provides detailed equations for calculating minimum extension length based on heat transfer analysis.
The stem design must accommodate differential thermal contraction between the stem (typically stainless steel) and the body. At -196°C, the stem contracts approximately 3 mm/m relative to room temperature. For a 1 m valve height, this means 3 mm of stem retraction that must not compromise the sealing force or operational torque.
The standard defines production testing (100% of valves undergo hydrostatic shell test, seat leakage test at room temperature, and low-temperature seat leakage test on a sampling basis) and type testing (per ISO 28921-2 for design validation). The sampling plan requires testing one valve from each lot of 50 or fewer valves. If any valve fails the low-temperature test, the entire lot is rejected, and the cause of failure must be investigated.
A manufacturer of cryogenic ball valves for LNG plant applications implemented ISO 28921-1:2022 requirements in their production line. The valves, designed for -196°C service in LNG transfer lines, featured extended bonnets of 300 mm length for DN 150 size, austenitic stainless steel bodies (CF8M), and PTFE seats with glass-fiber reinforcement. Production testing following Annex A demonstrated that 98.5% of valves passed the low-temperature seat leakage test on first attempt (allowable leakage of 0.3 cc/min/inch of seat diameter for soft-seated valves).
A critical engineering insight from the production experience was the importance of the thermal cycle rate during low-temperature testing. The standard specifies cooling at a rate not exceeding 1°C per minute to avoid thermal shock, followed by a stabilization period of at least 1 hour after reaching the minimum test temperature. The total test cycle time of 8-10 hours (including warm-up) constrained production testing capacity. To optimize throughput, the manufacturer implemented a sampling plan based on the standard’s lot acceptance provisions: testing one valve from each production lot of 50, with 200% lot testing for initial validation and 100% lot testing reduced to 25% after 12 consecutive lots passed without failure.