Elsevier

Materials Science and Engineering: A

Volume 620, 3 January 2015, Pages 129-139
Materials Science and Engineering: A

Stability of machining induced residual stresses in Inconel 718 under quasi-static loading at room temperature

https://doi.org/10.1016/j.msea.2014.09.118Get rights and content

Abstract

Tensile residual stresses are very often generated on the surface when machining nickel alloys. In order to determine their influence on the final mechanical behaviour of the component residual stress stability should be considered. In the present work the evolution of surface residual stresses induced by machining in Inconel 718 under static loading at room temperature was studied experimentally and numerically. An Inconel 718 disc was face turned employing industrial working conditions and specimens for tensile tests were extracted from the disc. Surface residual stresses were measured by X-ray diffraction for initial state and after applying different loads over the material׳s yield stress. Then, a finite element model based on the surface–core approach was fitted to experimental results and the study was extended to analyse the influence of load level, degree of work-hardening and initial surface conditions. For the studied case, initial tensile surface residual stress (776 MPa) became even more tensile when applying loads higher than the material yield stress, but a shift was observed at the highest applied load (1350 MPa) and initial residual stress was relaxed about 170 MPa. This particular behaviour is associated to the modified stress–strain properties of the machined affected surface layer which was strongly work-hardened. Moreover, if the work-hardened properties are not considered in the finite element model results differ substantially from experiments. Surface residual stress stability also depends on the initial surface residual stress, but the degree of work-hardening induced by the machining process must be considered as well. If the difference between the yield stress of the surface and the yield stress of the core is lower than the initial surface residual stress, the surface begins yielding first and consequently the surface residual stress is decreased. In contrast, if the difference between the yield stress of the surface and the yield stress of the core is greater than the initial surface residual stress, the opposite behaviour is observed.

Introduction

Nickel based and titanium alloys are the most used in the aerospace industry as they offer higher mechanical strength, chemical resistance and thermal conductivity than steels [1]. However, due to these excellent properties, these superalloys are very difficult to machine and the final surface integrity of a machined component can be affected. In fact, over the last decade many research studies have been focused on workpiece surface integrity and functional performance [2] and extensive literature reviews overviewing different parameters of the surface integrity of machined nickel and titanium alloys have been carried out [1], [3].

Among the different parameters that describe the final surface integrity of a workpiece, residual stresses induced during machining have received considerable attention. Usually, when machining nickel and titanium alloys, tensile stresses are generated on the surface which can reduce the fatigue life of a part [4]. Indeed, this is an important aspect to be considered when manufacturing critical components. Nevertheless, residual stresses can vary over the lifetime and their stability should be analysed in order to quantify their real influence on the component behaviour [5]. Under the application of mechanical and/or thermal energy, elastic deformations related to residual stresses can be relaxed if they are transformed into plastic deformations [5]. Basically, residual stresses can be modified due to mechanical loads (static or cyclic), thermal exposure or crack extension [6].

Most of the studies are focused on the relaxation of residual stresses by thermal treatments and stability of compressive residual stresses induced by mechanical treatments. For instance, Vöhringer in an early research [7] published a work analysing thermal and mechanical relaxation mechanisms and models, completed with experimental examples. Later, in 2002, this study was updated by Löhe and Vöhringer in Totten׳s Handbook of Residual Stress and Deformation of steel [5]. Another interesting work was carried out by Schulze in 2006 [8], where stability of compressive residual stresses induced by various mechanical treatments was analysed deeply. More recently, in 2007, McClung [6] carried out a literature survey about the stability of residual stresses induced by shot peening and similar surface treatments, cold expansion of holes, welding and machining processes. He found very few studies about machining although, as mentioned previously, machining induced residual stresses are widely studied in the literature. For the last 10 years some studies dealing with relaxation of machining induced residual stress in steels have been carried out. In 2004, Capello et al. [9] observed that tensile surface residual stresses induced by turning in C45 and 39NiCrMo3 steels relaxed around 50% due to cyclic loading. Later, Denkena et al. [10] found that tensile and compressive surface residual stress locked in AISI 1060 steel specimens after turning at 3 different conditions were relaxed when applying pulsating and cyclic loads. More recently, Laamouri and co-workers [11] analysed relaxation of residual stresses in AISI 316L ground specimens due to cyclic loading and they observed greater relaxation when increasing the load amplitude.

The nickel based alloy Inconel 718 is the subject of this study, which is used 50% by weight in the case of modern engines [12]. Several authors have already studied the stability of residual stresses generated in Inconel 718, fundamentally those induced by mechanical treatments. Most of the residual stress relaxation studies in Inconel 718 are concerned with thermal relaxation, and there are only a couple of works about relaxation caused by mechanical loading.

One of the first investigations about thermal relaxation in shot peened Inconel 718 specimens was carried out by Khadhraoui et al. [13]. They employed the X-ray diffraction technique to measure the relaxation of residual stresses induced by two shot peening conditions after different exposure periods (10 and 100 h) at 600 °C and 650 °C respectively. They found a significant relaxation in the first period of exposure time (the maximum relaxation occurred at the surface), increasing when increasing temperature whilst the residual stress pattern was hardly affected by temperature. Later, Prevéy et al. [14] studied the thermal relaxation of compressive residual stresses induced in solution treated aged Inconel 718 by conventional airblast peening, gravity peening and laser shock peening. Specimens were exposed at 525, 575, 600, 625, 675 °C, typical working temperatures of a turbine engine, and held for 10, 20, 60, 200, 600, 2000 min time periods. They observed that the amount of relaxation is directly correlated to the degree of cold-work; the highest the initial cold-work the greatest the relaxation. Cai et al. [15] analysed the residual stress relaxation behaviour of solution treated and shot peened specimens of Inconel 718 during aging at two temperatures (700 °C and 740 °C). They found that residual stresses induced by the peening process were relaxed 71% at 700 °C and 83.7% at 740 °C after 12 h thermal exposure, and they were almost fully relaxed after 24 h. They also calculated parameters of the Zener–Wert–Avrami approach, determining different values for each temperature, which may indicate some interaction between precipitation of the γ′, γ″ and δ phases. More recently, Hoffmeister and co-workers [16] studied the thermal relaxation of compressive residual stresses induced by shot peening treatment in Inconel 718 specimens with the objective to obtain an analytical description of it. They used aged hardened specimens and each specimen was subjected to an isothermal heat treatment, testing a total of seven annealing times ranging from 0.1 h to 100 h and five temperatures below 750 °C. They found an anomalous behaviour for the intermediate temperature and they propounded a new modified Zener–Wert–Avrami equation which gave a good correlation with experimental measurements (mean deviation less than 6.5%). Lately, Zhou et al. [17] analysed the residual stress distributions in the superalloy IN718 due to laser shock peening by finite element simulation at 700 °C, 800 °C, 900 °C and 1000 °C. On the one hand, they did not find any relaxation at 700 °C, as previously observed by [14] when initial residual stresses were below 400 MPa or initial plastic deformations were low. In contrast, at higher temperatures they observed that residual stresses had a significant relaxation, mainly in the maximum peak stress, at the beginning of the exposure (1–3 min heating period) and the relaxation was stabilised at longer periods. Based on the calculated results, they also calculated the Zener–Wert–Avrami approach׳s parameters.

There are only a few studies which deal with residual stress stability under mechanical loading in Inconel 718. Zhuang and Halford [18] studied the residual stress relaxation due to cyclic loading in Inconel 718 after applying three different surface treatments: shot peening, laser shock peening and low plasticity burnishing. They put forward an analytical model which considers the main influencing parameters: initial residual stresses, degree of cold-work, load amplitude and ratio and material dependent constants. Results revealed that the major residual stress relaxation occurred in the early stage of the cyclic loading, and they observed that the higher the initial cold-work the greater the residual stress relaxation is, and the relaxation was also higher when increasing the load amplitude and load ratio. Although the analytical model proposed by them was capable of predicting the trends of residual stress relaxation, they did not verify experimentally the cycle-dependent residual stress relaxation. Hoffmeister et al. in a recent work [19], investigated the macro residual stress relaxation in shot peened Inconel 718 specimens under isothermal quasi-static and cyclic loading. Experimental results showed that initial compressive residual stresses (≈750 MPa) on the surface firstly relaxed due to thermal loading as studied in [16] around 300 MPa and additional relaxation occurred when applied strain reached the elastic limit. Moreover, tensile residual stresses were built-up when applying high strains (i.e. 750 MPa when applying 2% strain.). Regarding with the relaxation caused by cyclic-loading, within the first cycle similar stress relaxation as found in quasi-static loading was observed and residual stresses remained stable for the subsequent cycles. Therefore, in this particular case, cyclic loading could have been studied by quasi-static loading.

It can be concluded that although residual stresses induced by machining have been widely analysed in the literature, there are only a few works dealing with residual stress relaxation in steels. In particular, there are no studies about the stability of residual stresses generated by machining in Inconel 718, which is widely used in aeronautical critical parts. Moreover, residual stresses generated in this nickel based alloy after machining are usually tensile and the machined surface is also strongly work-hardened. Therefore, it is worth analysing the stability of residual stresses induced by machining in Inconel 718 in order to understand and quantify their influence on the final performance of the component.

The aim of the present paper is to analyse the stability of surface residual stresses generated during finish machining in Inconel 718 under mechanical static loading at room temperature. For this purpose, an Inconel 718 disc was face turned at cutting conditions employed in the aeroengine manufacturing industry. Then specimens were extracted from the disc and were loaded at different stress levels. Initial and final surface residual stresses were measured by X-ray diffraction. Finally, a finite element model was fitted to experimental results and the study was extended for more loading conditions and initial material states (degree of work-hardening and residual stress values).

Section snippets

Material

Inconel 718 rolled sheet with average grain size (ASTM7) and hardened by precipitation was selected for this study. The chemical composition of the material supplied by the manufacturer can be observed in Table 1. The sheet employed in the present study had a 44 HRC hardness value. The microstructure of the material can be seen in Fig. 1. It consists of an austenitic face centred cubic matrix with gamma double prime (γ″) strengthening precipitates.

Machining test and specimen preparation

The Inconel 718 rolled sheet was previously cut

The surface–core model and finite element model

The surface–core model is a simple model that enables the uniaxial residual stress stability to be analysed [7]. The model is based on a prismatic rod or sheet which is divided into a thinner surface layer and a core region, as it is depicted in Fig.4a. State-specific stress–strain properties and initial residual stresses are assigned to the surface layer and the stress–strain properties of the raw material to the core region. Fig. 4b shows an example of stress–strain curves (σ,ε)surf and (σ,ε)

Stress-strain properties of the core region and surface layer

The stress–strain curves assigned to the core region and surface layer are plotted in Fig. 8. The stress–strain response of the raw material was obtained through tensile tests. Nickel based alloys such as Inconel 718 are prone to work-hardening in response to the machining induced deformations and heat [1] and the stress–strain properties of the machined surface can vary considerably from raw material properties. Indeed, as reported by Pawade et al. [21], the local yield stress of the machined

Conclusions

In the present work the stability of surface residual stresses induced by turning in Inconel 718 under quasi-static loading were analysed, first experimentally and then by numerical simulation with finite element modelling. The main conclusions derived from this research are the following:

  • Experimental results revealed that initial surface residual stresses (average value of 776 MPa) did not vary when the applied load was below the yield stress of the material. In contrast to prior studies about

Aknowledgments

The authors wish to acknowledge the department of material science of the ETSI Caminos of UPM Madrid for their support in the measurements of residual stresses.

The authors thank the Basque and Spanish Governments for the financial support given to the projects PROFUTURE I and II (codes IE 10-271 and IE11-308), InProRet (code IE12-342), METINCOX (DPI2009-14286-C02-0 and PI-2010-11), DESAFIO, CRINCOPLUS (code UE2013-08), ISMEC (S-PE12MU007), ISMECII (S-PE13MU007) and BIA2011-26486.

The authors

References (26)

  • D. Ulutan et al.

    Int. J. Mach. Tools Manuf

    (2011)
  • I.S. Jawahir et al.

    CIRP Ann.-Manuf. Technol

    (2011)
  • V. García Navas et al.

    Int. J. Mach. Tools Manuf

    (2012)
  • O. Vöhringer
  • B. Denkena et al.

    Proc. Eng.

    (2011)
  • A. Laamouri et al.

    Int. J. Fatigue

    (2013)
  • Z. Zhou et al.

    Int. J. Impact Eng.

    (2011)
  • W.Z. Zhuang et al.

    Int. J. Fatigue

    (2001)
  • R.S. Pawade et al.

    Int. J. Mach. Tools Manuf.

    (2008)
  • A.R.C. Sharman et al.

    J. Mater. Process. Technol.

    (2006)
  • J. Liu et al.

    Mater. Sci. Eng. A

    (2010)
  • R. M’Saoubi et al.

    Int. J. Sustain. Manuf

    (2008)
  • D. Löhe et al.
  • Cited by (15)

    • Effect of surface integrity generated by machining on isothermal low cycle fatigue performance of Inconel 718

      2022, Engineering Failure Analysis
      Citation Excerpt :

      Therefore, it can be assumed that the stress-strain properties of the surface layer are identical in both discs. Due to the difficulty of obtaining accurate stress-strain properties of the surface layer, these can be estimated based on the study published in [18]. In that study, under similar cutting conditions to those employed in this work, the yield stress of the surface region was 1.8 times higher than the core region.

    • Statistical analysis and modeling of temperature distribution during various milling operations of thin walled aircraft parts

      2021, Physica A: Statistical Mechanics and its Applications
      Citation Excerpt :

      This thermo-mechanical action leads to numerous machining problems such as high temperature in the cutting zone, rapid tool wear, plastic deformation at the cutting zone and deflection and deformation of the thin panel due to its poor rigidity and the high cutting force applied [14]. Usually, the local overheating may leads to the reduction in the hardness of the cutting tool, the decrease of the tool life, the modification in the workpiece material properties, residual stress as well as a significant effect on the surface quality (surface damages) [15–18]. Furthermore, when machining thin-walled parts, regenerative effect occurs due to the non-constant chip thickness.

    • Measuring residual stress in Ti-6Al-4V with HR-EBSD, using reference patterns from annealed material

      2021, Materials Characterization
      Citation Excerpt :

      Residual stress fields within a component's metal structure can form during a variety of processing steps, affecting the capability of the material(s) to meet requirements. Welding, forging, machining, peening, and other mechanical processing operations typically impart residual stresses of some degree [1–6], and these stresses are known to alter important material characteristics such as corrosion resistance, recrystallization behavior, and fatigue properties among others [7–10]. Understanding the formation of these stresses and quantifying their magnitudes is, therefore, imperative for delivering high-performance components, such as those for aerospace, nuclear power, and oil-and-gas industries.

    View all citing articles on Scopus
    View full text