You are here

Light Water Reactor Sustainability (LWRS) Program – R&D Roadmap for Non-Destructive Evaluation (NDE) of Fatigue Damage in Piping

Light water reactor sustainability (LWRS) nondestructive evaluation (NDE) Workshops were held at Oak Ridge National Laboratory (ORNL) during July 30th to August 2nd, 2012. This activity was conducted to help develop the content of the NDE R&D roadmap for the materials aging and degradation (MAaD) pathway of the LWRS program. The workshops focused on identifying NDE R&D needs in four areas: cables, concrete, reactor pressure vessel, and piping. A selected group of subject matter experts (SMEs) from DOE national laboratories, academia, vendors, EPRI, and NRC were invited to each workshop. The LWRS NDE workshop on piping fatigue was held on August 2nd, 2012 with twenty one SMEs attending that particular workshop.

Fatigue (caused by mechanical, thermal, or environmental factors) is the number one cause of failure in metallic components. Examples of past experience with this form of degradation in reactor coolant system (RCS) include cracking at: BWR feedwater nozzle; BWR steam dryer support bracket; BWR recirculation pipe welds; PWR surge line to hot leg weld; PWR pressurizer relief valve nozzle welds; PWR cold leg drainline; PWR surge, relief, and safety nozzle-to-safe-end dissimilar metal butt welds; PWR decay heat removal drop line weld; PWR weld joins at decay heat removal system drop line to a reactor coolant system hot leg. The effects of environment on the fatigue resistance of materials used in operating PWR and BWR plants are uncertain. There is a need to assess the current state of knowledge in environmentally assisted fatigue of materials in LWRs under extended service conditions. It is also important to develop a mechanistic understanding of the role of water chemistry on the microstructural changes in the materials and on their fatigue properties. In parallel, implementation of effective inspection and monitoring programs (i.e., NDE techniques, sensors and systems for surveillance and monitoring, improved condition monitoring and operational assessment strategies, etc.) for timely detection and mitigation of fatigue damage to safety critical components is of vital importance to achieving safe and economical long term operation (LTO) of the existing fleet of LWRs.

The first presentation at the piping fatigue NDE workshop was made to help identify material degradation issues associated with environmentally assisted fatigue. Dr. S. Majumdar, a senior mechanical engineer at Argonne National Laboratory (ANL), with extensive expertise in this area gave an overview of that subject. In summary, the presentation identified locations in LWRs where problems are more likely to occur. Those locations include: weld heat affected zones; vulnerable spots associated with dead flow zones or places where the local chemistry is different from bulk chemistry; thermal stratification and thermal striping zones; locations affected by off-design transients; initiation sites caused by manufacturing flaws (e.g., scratches and dings). The presentation also identified certain gaps in the ability to detect and monitor environmentally assisted fatigue damage that needs special attention. Those include measurement of oxygen/hydrogen content in the water during transients (start up, shut down, etc.), wide-area sensors and instruments for detection of cracks that initiate from places not identified during design as high stress regions (i.e., surprise failure incidents), and novel NDE techniques for early detection of cracks (microcracks) and for loss of protective oxides in view of the fact that the actual degradation locations are often hard to predict.

A series of presentations were made subsequently on promising NDE and monitoring techniques for detection, diagnostics and prognostics of fatigue damage. The talks – S. Bakhtiari (ANL), J. Wall (EPRI), B. Regez/S. Krishnaswamy (Northwestern Univ.), and A. Chattopadhyay (Arizona State Univ.) – covered a wide range of conventional NDE and monitoring methods as well as emerging sensors and techniques, many of which are currently being evaluated for non-nuclear applications. A brainstorming session was conducted next during which the SMEs were asked to provide their input to help define R&D actions needed to address the gaps discussed earlier. A major objective of the brainstorming session was to identify those sensors and techniques that have the most promising commercial viability and fill a critical inspection or monitoring need.

Some common themes regarding R&D needs identified by the working groups included techniques for early (pre-cursor) detection, fatigue crack initiation and growth monitoring (below current conventional NDE limits and for welds, base metals, bends/elbows, and long pipe sections), sensors for in situ materials characterization and for feature sizes usually examined by laboratory techniques (e.g., oxide coating assessment), global screening —as opposed to local examination methods—for early detection of damage, robust sensors for harsh environments (elevated temperatures – >200oC), and development of improved signal processing, data analysis, and sensor fusion algorithms for better sensitivity to detection of incipient degradation. It was also noted that a critical challenge in damage detection is the fact that damage at the microscale may not be detected by conventional sensors. Although considerable research has been conducted on developing different sensing techniques, existing sensors pose considerable limitation on the size of detectable damage. A viable approach proposed to overcome this problem is to develop a hybrid framework that integrates physics based model with data from physical sensors, resulting in a robust framework for damage detection and remaining useful life prediction.

Following the brainstorming session, the proposed NDE R&D needs were prioritized and ranked based on an open voting process. The top three proposed R&D needs are as follows:

1. NDE capability to detect and characterize damage/degradation (fatigue and stress corrosion cracking) at an early stage. The efforts should address the physics of measurement–sensitivity to early degradation and extraction of “real” signals from noise (unwanted signals) associated with structural features. New sensor technologies may be needed to measure material property changes of interest (i.e., size and dimension of grain boundaries, commonly measured in tens of microns).

2. Measurement capability for fatigue crack initiation and growth monitoring in welds, base metals, bends and elbows, and along long sections of piping. Sensors and systems are needed to measure below the current conventional NDE detection limits for macrocracks. It is also imperative to ensure that the defined detection limits are adequate for LTO.

3. Measurement capability for in situ material characterization for features of the size usually studied in the laboratory. Of particular interest are sensors that provide material state awareness for selected early degradation modes (e.g., oxide coating assessment).

A common theme among all the proposed R&D needs was the development of a sample library for the evaluation of all NDE and monitoring techniques (i.e., design of experiments to define numbers and classes). A number of NDE and monitoring techniques employing acoustic/ultrasonic, thermal, electromagnetic, micromagnetic, and optical (visual, laser) sensors were also identified as viable candidates for further evaluation. An overview of the proposed NDE and monitoring techniques is presented in this report. The NDE data will ultimately serve as input to mechanistic models to help more accurately predict the remaining useful life of components and in turn increase confidence in LTO of the existing fleet LWRs.