For the past several years baffle-former bolt (BFB) cracking in pressurized water reactors has become a significant concern for of PWR plants. In 2016, three similar Westinghouse designed plants (Indian Point 2, Salem 1, and D. C. Cook Unit 2) experienced significant numbers of cracked BFBs, attributed to irradiation-assisted stress corrosion cracking (IASCC). These plants had common characteristics that included the 4-loop plant design, downflow configuration, and Type 347 stainless steel bolting material. BFB cracking is not an entirely new phenomenon as it was initially detected in the French PWR fleet in the 1990s. However, the extent of cracking found in some of the US plants has greatly exceeded prior cracking. Extensive industry programs have identified and categorized by tier group the most susceptible plants, and the EPRI Materials Research Program (MRP) has published guidance regarding baffle-former bolt UT inspections for PWR plants for detection of degraded and cracked bolts in the baffle-former assembly (MRP-2017-009).
A typical arrangement of the baffle-former assembly showing the BFBs and other bolt types is shown in Figure 1. There are between 700 – 1100 baffle-former bolts (depending on design) that hold the baffle plates in place for structural support. While the baffle-former assembly is a highly redundant structure, a significant number of the bolts must remain intact to assure functionality of the baffle-former assembly during a design basis event such as a loss-of-coolant accident (LOCA) or seismic activity.
The baffle-former assembly provides core support and directs coolant flow through and around the core. The susceptibility to BFB cracking depends on several external factors that influence crack initiation (i.e., fluence and temperature) and design or operating conditions that contribute to tensile stresses in the bolts. The factors that load the bolts are bolt preload, the differential pressure across the baffle-former plates, thermal expansion of the assembly, stress relaxation, and void swelling. Differential pressure loading across the baffle plates increases the stress in the bolts and is more significant for plants operating in the downflow configuration than upflow configuration. The difference between downflow and upflow configuration for the coolant flow is shown in Figure 2.
Managing BFB cracking has become a significant concern for these susceptible plants because of the uncertainty and cost of mitigation measures. Many options are being considered. However, it is difficult to understand the value of each option. As other options become available, how can they be compared to the traditional approaches for BFB management? SI uses a decision management tool for analysis of run/repair/replace decisions. This tool is well-suited to the evaluation of decisions requiring cost and uncertainty and incorporates the probability of failure model results with the costs associated with each alternative strategy.
Several actions may be considered as options for managing BFB cracking once a plant is identified as being susceptible:
To evaluate the optimal strategies and implement these options(s), it is crucial to first determine the likelihood (or probability) of future bolt failures, consider changes that could occur by implementing one or more mitigation measures, and compare the projected costs and outcomes associated with the various strategies. Then, by performing variations of plant-specific cost/benefit analysis and making these comparisons, informed decisions can be made when considering the different combinations of technical objectives and economic trade-offs.
The cost/benefit approach for BFBs is as follows:
First, the projection of BFB cracking is required to understand the level of failures expected for each option. The industry previously projected the number of BFB failures for various plant designs to help determine the inspection interval. SI participated in helping develop those guidelines using an in-house bolt failure prediction model. For this prediction model, a bolting failure is considered based on IASCC susceptibility and the time or effective fluence to cause crack initiation. As shown in Figure 3, once cracking initiates, the spreading of failed bolts can occur more rapidly, as demonstrated by the upward trend in % failed bolts. The results are presented as a probability of failure (i.e., number or pct. of failed bolts) as a function of fluence accumulation, which is plotted in Figure 3 as a function of effective full power years (EFPY). The probabilistic prediction model is benchmarked with the measured failure trends for the 4-loop downflow type plants. These trend lines of BFB failure predictions are used as input when considering the various options to mitigate BFB failures.
As discussed previously, there are many options that utilities can consider to address BFB failures. These options include: do nothing and replace what is cracked, preemptively replace bolts, combine an upflow conversion with preemptive bolt replacement. To compare these options, the cost and impact for the utility are essential parameters.
The alternative analysis cost/benefit study includes a financial model that considers the discounted cash flows and expected outcomes for each of the alternatives evaluated, compared to the base case. The costs and timing of implementing the various mitigation options take into account the future cash flows and estimated cost savings (per year) expected.
Each option for managing BFB cracking includes several variables:
Each of these options is evaluated to determine the “value” of implementing an alternative strategy compared to several other options. Each of these options or strategy is examined in terms of the probability of expected outcome related to BFB failures and their relative expected lifetime costs in terms of a planned (net present) value, or NPV. These results can be used to help make decisions on the best option to manage baffle bolt degradation using the best estimate costs and expected value results; the resulting strategies can be ranked by the relative NPV cost, as illustrated in Figure 4.
Figure 5 shows a more detailed description of the various options over the remaining time horizon of the operating plant.
The multiple options, as shown here, are considered for a cost comparison. In these sample scenarios, the proactive bolt replacement with no upflow conversion is significantly more expensive than the other options. After about 2058, option 2 (upflow conversion and replace bolts as needed) is the least costly option. At different times, the relative rankings of the different options are shown to change. For example, the Upflow Conversion options are more expensive early in life when the conversion costs are incurred, but those options recoup that investment over time as failure rates and numbers of bolts to be replaced are reduced over time.
Once the costs and expected values of the various options are determined, SI works with the utility to identify the alternative(s) that best fit the performance objectives and cost (or cash flow) objectives for planning purposes. The results are presented in terms of time horizons for implementation of the planning options and management decisions, recognizing that the analytical models can be refined later using actual inspection results and real-time information and related operating experience for each of the various strategies.
BFB cracking has been shown to cause significant outage impacts to utilities of these susceptible plants. Understanding the options and managing the issue can help utilities save money in the long run and eliminate surprises. For more information on how SI could help in your planning, please contact Tim Griesbach or Chris Lohse.