On the role of crack tip creep deformation in hot compressive dwell fatigue crack growth acceleration in aluminum and nickel engine alloys

https://doi.org/10.1016/j.ijfatigue.2020.106082Get rights and content

Highlights

  • Crack growth acceleration due to hot compressive dwell (HCD) effect was observed.

  • HCD effects were studied on fatigue crack growth in 319 cast Al and Inconel 718.

  • In 319 Al, a creep-based waveform effect was observed at RT with a non-dwell cycle.

  • A physics-based model was refined to predict HCD effects on crack growth behavior.

  • Good agreement was found between model predictions and experimental results.

Abstract

Ambient and Hot Compressive Dwell (HCD) fatigue experiments were carried out to understand the effects of HCD on the fatigue crack growth behavior of cast 319 Al and Inconel 718 engine alloys. HCD was found to reduce endurance and increase rates of crack growth at engine service temperatures, while far less crack growth acceleration was found at room temperature. A fracture mechanics-based model was developed, considering tensile residual stress contributions due to compressive creep in order to predict crack growth with HCD. Good agreement was found between predictions and experimental results.

Introduction

The function of many engineered systems can be reduced to three critical steps: start-up, running, and shut-down. While running, continuously applied stress (dwell) can create creep damage at elevated temperature. Additionally, fatigue damage accumulates due to repeated start-up and shut-down cycles. In non-isothermal applications involving hot regions surrounded by cooler material, a Hot Compressive Dwell (HCD) condition is potentially critical, especially if notched-features/stress-concentrators fall within the hot region. Such is the case in automotive combustion chambers and aerospace engine applications, among others.

Conditions involving combined creep and fatigue damage have been widely studied for traditional high temperature materials such as steels, nickel-based superalloys, and titanium alloys, but mostly for tensile creep conditions. Wareing’s studies [1] on 316 stainless steel show that the introduction of a tensile hold period into the fatigue cycle can reduce life to failure. In this study it was also shown that the ASME N47 [2] design line is extremely conservative compared to the results from the studies on 316 stainless steel. Pédron and Pineau [3] carried out tensile creep-fatigue studies on Inconel 718 and found that fatigue crack growth rates are greatly increased (by two orders of magnitude) with a five-minute tensile hold time within the fatigue cycle. A tensile creep-fatigue failure model for nickel-based superalloys was developed by Winstone [4], concluding that the creep and fatigue damage can be summed linearly to predict the crack growth per cycle when both creep and fatigue contribute to the crack propagation process.

Studies investigating compressive fatigue phenomena are less common in the literature. Lord and Coffin’s studies [5] on cast René 80 include experiments with both tensile and compressive hold times. They showed an increase in fatigue life with the tensile strain hold experiments, while compressive strain hold experiments resulted in shorter lifetimes. An extensive review of compressive fatigue experimental studies is given in [6], highlighting conflicting results and discussing possible mechanisms and experimental issues.

It should be pointed out that the most of the creep-fatigue models discussed so far, and others in the literature [7], [8], [9], are focused on crack initiation rather than propagation, which is the focus of the present work. Here, tensile dwell again receives greater attention. Goswami proposed correcting baseline crack growth rates with a Normalized Crack Growth Ratio [10]. More mechanistic approaches for tensile dwell include the superposition of cycle-dependent and time-dependent crack growth [11] and numerical simulation methods [12].

Rhymer [13] is one of the few who addresses the dwell-induced compressive creep contribution to crack growth in a direct mechanistic approach. Rhymer identified a creep-induced residual stress mechanism that accelerates crack growth in nickel superalloys under HCD conditions. His approach was to simulate structural response in an elastic/visco-plastic finite element model with contact nonlinearity. In [14], the authors developed a simple fracture mechanics-based creep-fatigue crack growth model that preserved the basic mechanism of Rhymer’s work, but in a reduced-order implementation for computational efficiency. In principle, the method involves the following elements:

  • 1.

    Requires constitutive models for creep and plasticity, where applicable. For thermal mechanical fatigue (TMF) crack growth, the variation of the modulus of elasticity with temperature can also be accounted for.

  • 2.

    Requires a thermo-elastic analysis of a duty cycle (typically a finite element analysis).

  • 3.

    The constitutive model and the stress field are analyzed to predict the residual stress field that results from plasticity and creep over the duty cycle.

  • 4.

    The stress-intensity factor (K) is calculated as a superposition of the elastic stress and residual stress components associated with plasticity and creep.

  • 5.

    Using the maximum and minimum stress intensity factors of the duty cycle, the crack growth life is calculated based on a crack growth model.

Whereas current practice applies stress superposition methods [15] to account for local tensile yielding (which decelerates crack growth), the current research merely applies an analogous practice to account for residual stresses built up due to local compressive creep (which accelerates crack growth). The aim of the present work is to further develop this approach with additional experimental evidence, and validate it on different materials and at different temperatures. In doing so, model complexity with regard to material crack growth and creep characterization is purposely kept to a minimum to highlight the salient features of the method with minimal distraction. It is left to individual practitioners to implement higher fidelity material characterization models as their applications may require.

This research is a continuation of the work introduced in [14], where the reader may find additional information regarding this method, including a detailed discussion on the application of linear elastic fracture mechanics to the HCD problem.

Section snippets

Testing materials and specimens

An automotive engine material, cast 319 aluminum alloy [16] (with secondary dendrite arm spacing (SDAS) of ~45 μm, grain size of ~500 μm, and modified eutectic silicon morphology) and a jet engine material, Inconel 718 (with grain size ~10 μm and ~2 μm δ phase), were used in this study, as shown in Fig. 1. Both applications (automotive and jet engines) experience creep-fatigue during their service, and the material selection for each application was guided by industry sponsors. Elevated

Methodology and testing conditions

Crack growth tests under HCD conditions were carried out for both materials at elevated temperature and room temperatures. Local applied (elastic) stresses at the blunt notch of the BCT specimens for all crack growth tests are listed in Table 2 in terms of the yield stress values given in Table 1. These test parameters are based on a combination of industrial practice and consideration of conducting the test in a reasonable amount of time. Crack growth tests without dwell (control experiments)

Experimental results

Results from crack growth tests with and without compressive dwell for cast 319 Al and Inconel 718 are plotted in Figs. 8(a,c) and 9(a) with crack growth rate da/dN as a function of crack length a. At elevated temperature, both alloys showed significant crack growth acceleration due to compressive dwell, 319 Al showing approximately six times faster growth with 120 s dwell than the baseline (0.5 Hz). At room temperature, the effects of dwell were limited, as had been expected, and further

Conclusions

A crack growth model for compressive creep-fatigue interactions using the stress intensity factor was further developed and validated in this work. It was found that crack growth rates significantly accelerate due to compressive dwell at elevated temperature. Hot compressive dwell causes the accumulation of creep deformation/tensile residual stress at elevated temperature, resulting in the observed crack growth acceleration.

Successful and slightly conservative predictions were made using the

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

Many thanks to the consortium members of the Integrative Materials Design Center (iMdc) at Worcester Polytechnic Institute for their support and guidance of this work.

References (21)

There are more references available in the full text version of this article.

Cited by (3)

  • Evaluation of fatigue and creep-fatigue damage levels on the basis of engineering damage mechanics approach

    2023, International Journal of Fatigue
    Citation Excerpt :

    There is a growing tendency towards complicated service conditions when improving the efficiency of hot-section components in industrial fields, such as aero-engines, steam turbines and heat exchangers. LCF and CF interaction caused by the cyclic loading phase in combination with a steady running period at elevated temperature constitute the major damage modes [1–3]. In the case of long-term operation and loading variation at elevated temperatures, the material mechanical properties degrade progressively with the prolongation of service time [4–6].

  • A modified damage-coupled viscoplastic constitutive model for capturing the asymmetric behavior of a nickel-based superalloy under wide creep-fatigue loadings

    2022, International Journal of Fatigue
    Citation Excerpt :

    In aero-engines and nuclear power plants, fatigue induced by frequent start-up and shut-down and creep induced by long-term operations would cause creep-fatigue interaction (CFI) that decreases the strength of structural materials, as seen the example of turbine disk in Fig. 1a. However, the situation seems to be more complicated aiming at a specific engineering application, such as the aviation turbine disk with complex damage mode, as schematically shown in Fig. 1b. The bore of the rotating disk is generally subjected to low cycle fatigue (LCF) at relatively low temperature, the simplified loading waveform of which is shown in Fig. 1c. However, the rim, hub as well as mortise are subjected to CFI at high temperature (≥0.4–0.5 Tm) [3], as seen in Fig. 1d-f. Contributed to inhomogeneous distribution of thermal stress, the dwell period may occur in tension or compression [4–6]. Accordingly, actual loading conditions for depicting CFI are generally simplified to tensile-dwell-only (CP), compressive-dwell-only (PC) and balanced strain dwell (CC), as shown in Fig. 1d-f. Under this circumstance, the complex series of processes takes the risk in causing multi-source failure mode under various loading waveforms [7–9].

View full text