Containment leak tightness testing is a critical safety verification activity that confirms the integrity of the nuclear power plant containment structure, the final engineered barrier against the release of radioactive materials to the environment. By combining integrated leak rate tests (Type A) with local leak rate tests (Type B and Type C), the testing program demonstrates that leakage remains within the maximum allowable leakage rate at design basis accident pressure. This article explains the technical principles, applicable standards, testing samples, methods, acceptance criteria, and typical applications of containment leak tightness testing for nuclear power plants.
Technical Overview
The containment building of a nuclear power plant—typically a prestressed or reinforced concrete shell with an internal steel liner, or a freestanding steel vessel—encloses the reactor coolant system and serves as the last physical barrier preventing the escape of radioactive fission products during normal operation and design basis accidents. Because even small defects in the shell, penetrations, seals, or isolation valves could allow radioactive leakage under accident pressure, regulatory frameworks worldwide mandate a structured containment leak tightness testing program throughout the plant lifetime.
The testing program is generally organized into three categories. Type A tests measure the overall integrated leakage rate of the complete containment structure at design basis accident pressure. Type B tests verify the local leak tightness of penetration seals, airlocks, and sealed electrical and mechanical penetrations. Type C tests verify the leak tightness of containment isolation valves, including valve seats, packing, and closure seals. The program is governed by recognized codes and standards, including:
- 10 CFR Part 50, Appendix J, and Regulatory Guide 1.163 (United States regulatory framework)
- ANSI/ANS-56.8, Containment System Leakage Testing Requirements
- Risk-informed, performance-based industry guidelines such as NEI 94-01
- IAEA safety standards for commissioning, operation, and periodic testing of nuclear power plants
- Chinese nuclear safety regulations and industry standards, including the HAF series and NB/T and EJ/T standards for containment testing
- ISO/IEC 17025 for the competence of laboratories performing the measurements
Testing Samples
Containment Structure (Overall Envelope)
The primary test object is the complete containment envelope, including the concrete shell, steel liner, and all welded and sealed joints. Type A integrated testing evaluates the envelope as a whole, capturing leakage through the liner, construction joints, and any unidentified leakage paths that local tests cannot reach.
Airlocks, Hatches, and Access Openings
Personnel airlocks with double interlocked doors, emergency airlocks, equipment hatches, and temporary openings are high-risk leakage points because they rely on elastomeric or metallic gaskets and movable door mechanisms. Their seals and seats are individually leak tested during outages.
Mechanical and Electrical Penetrations
All piping, cable, and instrumentation penetrations that pass through the containment wall are tested, including the fuel transfer tube penetration, electrical penetration assemblies, and process pipe penetrations with their bellows, sleeve welds, and pressure-retaining seals.
Containment Isolation Valves
Every valve credited for containment isolation in safety analysis—gate valves, globe valves, check valves, butterfly valves, and dampers located inside and outside the containment boundary—is subject to seat, packing, and closure leakage testing.
Testing Items
The scope of containment leak tightness testing covers the following key items:
- Integrated (overall) containment leak rate at design basis accident pressure, compared against the maximum allowable leakage rate (La)
- Local leak rates of airlock door seals, gasketed covers, and penetration assemblies
- Containment isolation valve seat leakage, packing leakage, and closed-position leak tightness
- Weld integrity of penetration sleeves, containment penetrant bellows, and liner attachment welds where applicable
- Identification and localization of unexpected leakage paths, followed by verification testing after repair
- Supporting measurements such as containment pressure, internal temperature distribution, humidity, and ambient barometric pressure used to compute corrected leak rates
Testing Methods
Integrated Leak Rate Test (Type A)
The containment is pressurized with dry compressed air to the design basis accident pressure Pa, typically in the range of 300 to 500 kPa gauge for large dry pressurized water reactor containments. After pressurization, a stabilization period of 12 to 24 hours or longer is observed to allow thermal equilibration. Measurement then proceeds continuously for a minimum of 24 hours while the containment air mass is computed from the ideal gas law, corrected for water vapor content.
The dominant technique is the absolute method: dozens of high-accuracy pressure transducers, resistance thermometer or thermocouple trees distributed throughout the containment volume, and dew point or humidity sensors feed a centralized data acquisition system. The containment air mass is calculated at successive time points (mass points), and a least-squares regression of mass versus time yields the measured leak rate Lt. A reference vessel method, in which an internal calibrated tank of known stability is compared against the containment, can also be used to cancel common-mode environmental effects. Sensitivity is verified by introducing a simulated known leak. The test is typically performed during a scheduled refueling outage, and at the pre-operational stage it is often combined with the structural integrity pressure test.
Local Leak Rate Tests (Type B and Type C)
Local testing applies pressure across the individual leak path and quantifies leakage by one or more of the following methods:
- Pressure decay method: monitoring the pressure drop across an enclosed sealed volume over time
- Mass balance or makeup flow method: measuring the constant-pressure gas makeup flow required to hold a set test pressure
- Vacuum hood method: evacuating a sealed hood mounted over a weld or seal and measuring gas inflow
- Tracer gas method: using helium and a mass spectrometer leak detector, with sensitivity down to approximately 10⁻⁶ standard cubic centimeters per second, to pinpoint leakage paths
- Bubble or soap solution testing as a simple qualitative screening method for accessible joints
Type B and Type C tests are commonly performed at Pa or at a reduced pressure with conservative extrapolation, since gas leakage is assumed proportional to the pressure difference for the relevant flow regimes.
Data Evaluation and Acceptance Criteria
The Type A result, including its measurement uncertainty, must demonstrate that the measured integrated leak rate Lt does not exceed the maximum allowable leakage rate La, which is commonly specified as 0.1 percent of the containment air mass per day at Pa for large dry containments, with values up to about 0.5 percent per day for some designs. Earlier prescriptive programs required the upper 95 percent confidence bound of Lt to remain below 0.75 La. The sum of all Type B and Type C local leak rates must typically remain below 0.6 La, providing margin against unidentified distributed leakage. Statistical uncertainty analysis, instrument calibration traceability, and independent review of the test report are integral parts of the acceptance process.
Applications
Containment leak tightness testing is applied at every major stage of the nuclear facility life cycle:
- Pre-operational commissioning and construction acceptance testing of new units, including the initial pressurization campaign
- Periodic in-service testing, historically at intervals of approximately once every ten years, with extensions permitted under risk-informed, performance-based regulatory programs for plants with strong test history
- Post-maintenance and post-modification verification after replacement of penetration seals, airlock components, isolation valves, or liner repairs
- Aging management programs and long-term operation or license renewal assessments
- Condition assessment following beyond-design-basis events, external hazards, or any condition that could challenge containment integrity
- Adaptation of test methodologies to small modular reactors, advanced reactors, and research reactors with containment or confinement structures
Utilities frequently engage accredited third-party laboratories and specialized engineering teams to perform high-accuracy measurements, independent data analysis, and regulatory documentation, ensuring traceability and defensibility of results before the nuclear safety authority.
FAQ
What is the difference between Type A, Type B, and Type C containment tests?
Type A is the integrated leak rate test of the whole containment envelope at accident pressure. Type B covers local leak testing of penetration seals, airlocks, and electrical or mechanical penetration assemblies. Type C covers leak testing of containment isolation valves. Together, the three test types verify the total leak tightness of the containment boundary.
How often must the integrated leak rate test be performed?
Traditional prescriptive programs require three Type A tests per ten-year service period. Under risk-informed, performance-based frameworks adopted by many regulators, plants with acceptable test history may extend the interval, commonly up to fifteen years, provided local leak rate testing and monitoring continue at regular intervals.
What pressure is used for the tests?
The integrated test is performed at the design basis accident pressure Pa, the peak containment pressure calculated for events such as a loss of coolant accident. Local tests are usually performed at Pa or at a reduced pressure with conservative extrapolation to Pa.
How long does a complete test campaign take?
Pressurization typically requires one to two days, stabilization twelve to twenty-four hours or more, and data acquisition at least twenty-four hours. Including instrumentation setup, local leak rate testing, depressurization, and reporting, a full campaign generally spans one to two weeks within a refueling outage.
What happens if the measured leak rate exceeds the acceptance limit?
The plant must investigate, localize, and repair the leakage path—using methods such as helium tracer detection—and repeat the relevant tests until compliance is demonstrated. The results and corrective actions are documented and reported to the nuclear safety authority before the containment is declared fit for continued operation.