Quantcast
Advertisement

Stress Relaxation Cracking, a Forgotten Phenomenon

By Steven Bradley | Reviewed by Martin RodriguezCheckmark
Published: June 17, 2021
Key Takeaways

Stress relaxation cracking is often mis-identified. However, with proper understanding of the mechanisms behind it, it is possible to prevent SRC or conduct a more thorough failure analysis to find it and determine the root cause.

Stress relaxation cracking (SRC), also referred to as stress relief cracking or reheat cracking, is a high-temperature (500-700°C or 932-1292°F) failure mode. It is associated with the welding of thick-walled, highly restrained components of austenitic stainless steel, ferritic stainless steel, heat-resistant steel (Cr-Mo-V) and nickel-base alloys. Cracking often occurs in the heat-affected zone (HAZ) during welding, and sometimes in the heat-affected zone during post-weld heat treatment as well as in the heat-affected zone after extended service at elevated temperatures. Failure leads to repair work and plant shutdowns. SRC occurs at the operating temperature of many boilers and pressure vessels and there is no standard testing available for qualifying materials. Often, failure analysts will misidentify such failures occurring during service as stress corrosion cracking. In this article, we'll take a look at stress relaxation cracking mechanisms and the factors that can impact whether they occur in various materials.

Advertisement

Stress Relaxation Cracking Mechanisms

The two primary mechanisms that cause SRC are grain boundary sliding and creep cracking that initiates and grows within grains from the formation of a “precipitation-free zone.” For stabilized austenitic alloys such as Type 321 and 347 stainless steels, the formation of Nb(C,N) or Ti(C,N) precipitates within the HAZ grains but not at the grain boundaries, resulting in the interior of the grains becoming stronger than the grain boundaries. Thus, any deformation occurs via grain boundary sliding. This mechanism might be the reason that the stabilized grades of stainless steel are so much more susceptible to SRC.

The alternate failure mechanism for non-stabilized alloys involves sensitization and the formation of intergranular carbides with acute triangular, carbide-shaped particles being most associated with the observed nano-cavitation. When the residual welding strain relaxes, the deformation concentrates at the grain boundaries because the Cr-depleted zone adjacent to the carbide precipitates is much weaker than the grain interior. Then, when the strain exceeds the creep ductility of the grain boundary regions, SRC will tend to initiate at triaxial stress risers and intergranular cracking will result.

Advertisement

For heat-resistant alloys, such as 21/4Cr-1Mo and 9Cr-1Mo, carbides will begin to precipitate in the coarse-grained heat-affected zone (CGHAZ) when exposed to elevated temperatures during post-weld heat treatment and/or during service. The carbides that form on prior austenite grain boundaries tend to be incoherent with the matrix. When these carbides coarsen, the boundaries become devoid of alloying elements (a precipitation-free zone) and become weaker than the matrix. As a result, SRC will occur along prior austenite grain boundaries.

Stress relaxation cracking can initiate shortly after welding or startup and may be recognized sometime later when a leak is detected. This failure mode is most likely with rework in weld sections that are highly restrained and with thicknesses greater than about a ½ inch. Multiple restarts can also cause SRC for restrained joints under high stress conditions. Relaxation strains of as little as 0.1% can initiate SRC. Such failures are the most likely to be misidentified.

Susceptible Austenitic Alloys

Alloy composition significantly impacts SRC cracking sensitivity. Austenitic alloys strengthened by the dispersal of fine, intergranular precipitates are most susceptible. Alloy 800H, Type 321 stainless steel and Type 347 stainless steels are considered to be the most vulnerable. However, Types 304 and 316 welded components have been reported to fail by SRC but to a lesser extent and are considered less susceptible to SRC than the stabilized stainless steel grades. The high carbon version of these stainless steels (321H, 347H, 304H and 316H) are generally specified for high temperature applications due to their enhanced mechanical properties. Therefore, reports of SRC are more frequently associated with the high carbon materials. Alloys 601 and 617 are susceptible as a result of strain aging embrittlement.

Advertisement

The temperature range of susceptibility to SRC is alloy-dependent. Reported ranges are as follows: 550 and 600°C (1022-1112°F) for Types 304H and 321H stainless steels, 575 to 650°C (1067-1202°F) for Type 347 stainless steel, 550 to 650°C (1022-1202°F) for alloy 800H, and 550 and 700°C (1022-1292°F) for alloy 617.

Critical Factors that Increase Stress Relaxation Cracking

There are many factors that impact the susceptibility of any given alloy to SRC. Because the failure mechanism has a creep component, larger grain size metallurgy tends to be more susceptible. This is particularly the case for CGHAZ. High residual stresses from fabrication, such as cold working and welding, can be a major contributor, as are the stresses from the joint design. Thermal expansion of a restrained joint can add to the initial stresses. Thicker sections can control weld restraint and increase the state of stress. Notches at the weld or welding defects such as undercut, lack of fusion or lack of penetration can act as stress concentrators. Many SRC failures occur as a result of rework welding forcing high stress to the weld.

Advertisement

Conducting Failure Analysis for SRC

As noted, the failure mode for SRC is by intergranular (along grain boundaries) crack propagation. Cracking is almost always confined to the weld and heat affected zones. However, high temperature stress corrosion cracking failures such as by polythionic acid or chloride stress corrosion cracking of sensitized stainless steel also propagate by an intergranular mode and are often associated with the weld. Differentiating these mechanisms requires careful analysis, particularly if the failure occurs during service. Note that stress corrosion cracking occurrence requires a susceptible material and a particular aggressive agent (e.g. chloride or polythionic acid) in the environment, besides the applied/residual tensile stress. On the other hand, SRC involves little or no chemical interaction with the environment since it is the outcome of particular features in the alloy metallurgy and the stress system.

Identification of the extent of the crack and where it ends on both the ID and OD of the welded component is important for selecting samples for further analysis. Metallographic analysis of the crack tip is a critical aspect of an SRC failure analysis. For austenitic alloys, the intergranular cracks exhibit little to no evidence of deformation. Crack tips are usually preceded by small, isolated cavities or creep voids and very often a metallic filament is present near the crack tip. The composition of the metallic filament is a function of the alloy. For stainless steel, it is usually Fe-rich and it is encircled with a chromium layer indicating Cr depletion; for Alloy 800H, it is often enriched in Ni with a Cr-rich oxide layer; for Alloy 617, no clear metallic filament may be observed. Depending on the environmental exposure, the filament may have a surface oxide or sulfide layer. Observation of the metallic filament is best with an as-polished, unetched cross-section examined in a scanning electron microscope. Another potential fingerprint is that the hardness in the cracked region tends to be greater than 200HV. In order to avoid the mis-identification of SRC cracks with stress corrosion cracks, it is useful to bear in mind that the latter are generally branched cracks.

Read: Techniques for Analyzing Corrosion Products

Minimizing the Potential of SRC

Reducing the risk of SRC starts with alloy selection. For example, can a less susceptible alloy such as Type 304 be used instead of Type 347 stainless steel? Because that is not always practical, other approaches need to be employed. For austenitic alloys, CGHAZ is highly susceptible to SRC. As such, limiting maximum interpass temperature, maximum arc energy, and electrode diameter during welding and using the stringer bead technique rather than weaving should be employed. The temperature range of SRC susceptibility of the selected alloy should be outside of the operating windows of the process. From the point of view of the design, thick piping should be avoided in high temperature interconnecting lines whenever possible, as well as stress concentrating points and sharp corners (e.g. pipe bends or sudden change in pipe thickness).

Selection of weld consumable E16.8.2 has been found to reduce SRC when welding Type 321 and 347 stainless steel. For austenitic stainless steels, a postweld heat treatment (PWHT), particularly field welds, at 900°C (1652°F) has been found to be beneficial. PWHT of Alloys 800H and 800HT is required by ASME Code for pressure vessels if process temperature is above 540°C (1004°F). Rapid heating and cooling during PWHT can introduce additional strains and negate the impact of the PWHT. Designing the weld joint to minimize strain on the joint such as from static loads, thermal expansion and residual loads from assembly is considered an excellent practice to minimize the potential of SRC. Prior to startup, all welds should be inspected by ultrasonic testing and/or penetrant non-destructive testing to detect any initial cracking.

For the high-temperature ferritic alloys, a higher PWHT of around 700°C (1292°F) for 11/4Cr-1/2Mo steel can be used to minimize the potential of SRC. Such a PWHT, however, may lower the high temperature strength and impact toughness.

Conclusion

Stress relaxation cracking is often mis-identified. However, with proper understanding of the mechanisms behind it, it is possible to prevent SRC or conduct a more thorough failure analysis to find it and determine the root cause.

Share This Article

  • Facebook
  • LinkedIn
  • Twitter
Advertisement

Written by Steven Bradley | Principal Consultant, Bradley Consulting Services

Steven Bradley

Steven Bradley is currently the principal consultant at Bradley Consulting Services having retired after 44 years from UOP-Honeywell as Senior Research Fellow.

Steven is also membership chair of the ASM Failure Analysis Society and is a registered Professional Engineer in the state of Illinois. He has presented and authored/co-authored over 60 technical papers involving failure analyses of complex systems, materials characterization and advanced electron microscopy of materials and catalysts and holds 16 patents. Steven completed his BSSE and PhD in Materials Science and Engineering from Northwestern University.

In addition to being a member of ASM and ASTM, he is also a member of NACE and ACS.

Related Articles

Go back to top