Reliable and Relevant Stress Corrosion Cracking Testing

Peter L. Andresen*

Shanghai Jiao Tong University, Shanghai 200030, China

ABSTRACT: This history of SCC equipment failures in hot water extends back over 180 years, and the earliest design codes didn’t appear until 75 years later.  The early view was that SCC occurred only in specific materials and above threshold conditions of stress and environment.  Based on this weak understanding, and for convenience (to obtain quick results), laboratory tests were simplistic and not well controlled.  Therefore, they produced misleading guidance, and SCC problems persist.  The effects of environment are always time dependent, and designing for a 40 year lifetime based on a test lasting a few weeks or months is inherently risky, if not foolish. The last 30 – 40 years has produced dramatic improvements in testing sophistication and relevance, to the point where crack growth testing can use fully relevant materials, environments and stresses with sufficient on-line sensitivity to detect growth rates consistent with a 40 – 80 year lifetime.  The response of cracks at static load (SCC) can be much more difficult to study than at cyclic load (corrosion fatigue), and extensive experience has led to detailed testing guidance. SCC initiation testing is more challenging because the only way to perform a fully-relevant test is to run it for the full (40 – 80 year) lifetime, which is clearly not feasible.  Thus, all initiation tests employ a large acceleration factor, often >100X, and it is almost never known how this acceleration affects the results (What is the actual acceleration?  How does it alter the effect of stress, or environment, or temperature?).  Small details of the surface condition can have large effects, and often are not known for plant components.  The exposed area evaluated in SCC initiation tests is important, and test specimens generally involve a tiny fraction of the area of plant components.  SCC initiation is stochastic, but testing rarely uses a sufficient number of specimens to obtain statistically meaningful information. A key element of the modern understanding of SCC is that immunity is rare, although the concept if immunity is embedded in the design codes.  Designs should be based on achieving the desired lifetime, which must account for the possible initiation and growth of cracks. Guidance is provided to make SCC testing more reliable, reproducible, efficient and relevant.