KEYWORDS: Fatigue failure, High temperature fatigue, crack growth rates, crack closure, TIG welding, SS304L(N), SS316L(N), hold time.
Safety critical structures and components used in power plant boilers, pressure vessels, and heat transfer circuits are commonly subjected to numerous mechanical and thermal loads. Apart from static load periods during normal operating conditions, the components are commonly exposed to fluctuating loads generated from multiple sources such as start-stop cycles, flow-induced vibrations, and thermal transients. The power plant heat transfer circuits are normally operated over a temperature range of 673 K (400 °C) to 873K (600 °C) and are generally fabricated using austenitic stainless steels, such as SS304L(N) and SS316L(N) as they possesses good corrosion resistance and adequate high temperature mechanical strength. The presence of high temperature accelerates the fatigue damage, while creep damage occurs due to holding of stress at a steady state during operation. Thus, understanding high temperature fatigue response as well as creep-fatigue interaction is important. Manufacturing of the power plant components involve joining process and welding is one of the commonly used joining technique. Tungsten Inert Gas (TIG) welding is commonly used for the joining of stainless steel. However, the conventional TIG welding of stainless steel is limited by the maximum depth of penetration it can achieve in a single pass. Thus, multiple weld passes are employed for welding thicker plates. More recently, Activated flux TIG (A-TIG) welding has been shown to be a joining technique for achieving enhanced weld penetration, thus, enabling joining of thicker plates with reduced weld passes. However, in general, welding process is susceptible to defects, which may act as possible sites for damage initiation when subjected to fatigue cycling, thereby, affecting the overall component life. The damage tolerant design concept that assumes pre-existing defect in a component and estimates the fatigue life using fatigue crack growth rate (FCGR) characteristics is commonly used for the structural integrity evaluation. Amongst the various parameters that affect the fatigue life of a welded joint, weld quality and weld microstructure plays a prominent role, which in turn is governed by the welding process. The stainless-steel joints produced using A-TIG and
iii
conventional multiple-pass TIG (MP-TIG) process exhibit variation in mechanical properties and microstructure that can lead to a difference in fatigue crack growth (FCG) behavior between two joints. The problem is further complicated by the presence of different microstructure at the weld bead (WM), heat affected zone (HAZ) compared to the base metal (BM). The objective of this research is to evaluate the FCG rate characteristics of A-TIG and MP-TIG welded joints at the typical operating temperature of 673 K for SS304L(N) and at 823 K for SS316L(N) to understand the role of weld microstructure on exhibited FCG characteristics. The fatigue crack propagation rate is governed by crack tip stress intensity range (ΔK = Kmax-Kmin) in cyclic loading, apart from stress ratio, R (ratio of minimum to maximum load applied in a cycle of loading). However, in view of the plasticity ahead of the crack tip and its interaction during cyclic crack growth process, premature crack closure takes place even during tension-tension fatigue cycling and as a consequence, only a portion of the applied stress intensity range contributes to the crack propagation. This is referred to as crack closure and is due to plasticity ahead of the crack tip and in the wake of the advancing crack. Crack growth rate corrected for crack closure provides a true indication of crack driving force in fatigue. The work aims to evaluate the crack closure observed in welded joints to obtain the intrinsic fatigue crack growth rates in welds.
Fatigue crack growth experiments were conducted on 304L(N) and 316L(N) welded joint region (WM), the heat affected zone (HAZ) apart from the base metal (BM). Compact Tension specimens were extracted from the welded plates to align the starter notch at the WM, HAZ and base metal. The crack propagation rates observed for the welded joints were slower compared to the base metal rates. Among the welded joints, reduced crack propagation rates were observed for the MP-TIG weld joints. Better FCG resistance observed for the MP-TIG joint can be associated with the fine-grained microstructure observed in the MP-TIG joint. The FCGR resistance for A-TIG joints was not appreciably different.
To further investigate the above aspect, crack opening loads obtained from the load-displacement plots were used to evaluate the effective stress intensity range for FCGR following ASTM E647-15 test standards. The crack closure levels were found to be influenced by the fracture surface undulations and roughness. The higher crack
iv
closure levels were observed for the fusion zone of welded joints compared to the HAZ. After accounting for crack closure effects, the growth rates observed for the fusion zone of SS 304L(N) MP-TIG joint were higher, while for the case of 316L(N) welds, higher growth rates were observed for the fusion zone of A-TIG joint.
The present work further evaluates the interplay between crack tip creep damage and fatigue crack growth by introducing a tensile dwell period between fatigue cycles. A dwell period of 600 seconds was introduced between repeated blocks of constant amplitude fatigue cycles. A marginal slowdown in crack growth rates was observed during the hold time test. Experiments were conducted with varying hold times (900 seconds, 1200 seconds) as well as at different test temperatures (923 K). The increase in crack opening displacement and load line displacement observed during the hold period suggest crack extension due to creep which was however not reflected on the overall crack growth rates. This aspect was investigated in detail and it was found that there was an increase in crack closure levels immediately after hold time; this can be due to the increase in crack tip plastic zone during the hold period which can induce higher closure effects during the subsequent fatigue cycles. The SEM images observed from the failed specimen indicated evidence of both trans-granular and inter-granular mode of failure indicating the presence of both fatigue and creep damage.
In summary, this work evaluated the fatigue crack propagation rate of A-TIG and MP-TIG welded joints, their neighboring regions at operating temperatures in a power-plant system to provide valuable inputs to damage tolerant design. Anomalies observed while understanding the FCGR behavior from applied stress intensity factor range were examined through crack closure concepts. The effect of tensile hold-time on FCGR was investigated to understand the interplay between creep and fatigue crack closure. It is hoped that the findings of this study are useful to the design of power plant components that operate at sustained elevated temperatures.