- Email: [email protected]
Failure analysis in aluminium turbocharger wheels
EFA-02749; No of Pages 11 Engineering Failure Analysis xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Engineering Failure Analysis journal homepage: www.elsevier.com/locate/engfailanal
Failure analysis in aluminium turbocharger wheels M.F. Moreira Corrosion and Protection Laboratory, Instituto de Pesquisas Tecnológicas do Estado de São Paulo - IPT, Brazil
a r t i c l e
i n f o
Article history: Received 30 September 2014 Received in revised form 29 October 2015 Accepted 7 November 2015 Available online xxxx Keywords: AA 2618 aluminium alloy Fatigue Intergranular corrosion Turbocharger compressor wheel
a b s t r a c t This paper presents a failure analysis conducted in aluminium compressor wheels used in diesel turbocharged engines. These wheels were made with machined AA 2618T652 alloy and installed in light truck engines, which were used in rural and industrial atmospheres. The premature failures of the wheels happened after life between 40,000 km and 300,000 km, while the expected life was about 1,000,000 km. The present investigation showed that a fatigue process was triggered by intergranular corrosion on the upper camber surface of the wheel. A set of immersion corrosion tests was carried out to evaluate intergranular corrosion susceptibility of the alloy. The root-cause for the formation of the intergranular corrosion cracking in the compressor wheels could not be identified in the present investigation. The overall results indicated that an “in service” contamination of the compressor surface started the intergranular corrosion cracking, promoting the fatigue failure of the wheels. © 2015 Published by Elsevier Ltd.
1. Introduction The turbocharger is basically an air pump, which makes the air/fuel mixture more combustible by introducing more air into the engine's chamber, creating more power and torque. Hot exhaust gases that leave the engine are routed directly to promote the rotation of the turbine wheel [1]. A typical diesel turbocharger rotates at speeds in the range of 100,000 rpm to 250,000 rpm. The rotation of the compressor wheel pulls in ambient air and compresses it before pumping it into the engine's chambers. The compressed air leaving the compressor wheel housing is hot as a result of compression and friction. Fig. 1 shows a cross section of a turbocharger and the location of the aluminium compressor wheel. Light turbocharged diesel trucks presented failures of 51 compressor wheels in a total of 17,146 wheels. These wheels were made of AA2618 aluminium alloy, forged, heat treated (T652) and fully machined. This particular wheel presents 14 blades: 7 full blades and 7 small blades. The premature failures happened in light trucks running in rural and industrial atmospheres after life between 40,000 km and 300,000 km, while the expected life is about 1,000,000 km. The goals of this failure analysis were: to check whether the material of the failure wheels was in accordance with standards; and to identify the fracture mechanisms involved in the premature failure of the wheels. Nine compressor wheels samples were submitted to chemical analyses, microstructural characterization, X-ray diffraction, hardness testing and fractographic examination Table 1 presents the wheels identification, available vehicle mileage and failure location atmosphere. The major stresses of rotating components are created by centrifugal forces and the highest tensile values are located at the bore, in the plane of maximum mass concentration at the largest diameter. A second relatively highly loaded region of wheels is the blade root which is connected to the backwall, near the outer diameter [2]. In service, the component is heated by the air compression and subjected to centrifugal forces imposed by angular speed. In a light turbocharged truck the compressor can be subjected to temperatures
http://dx.doi.org/10.1016/j.engfailanal.2015.11.024 1350-6307/© 2015 Published by Elsevier Ltd.
Please cite this article as: M.F. Moreira, Failure analysis in aluminium turbocharger wheels, Engineering Failure Analysis (2015), http://dx.doi.org/10.1016/j.engfailanal.2015.11.024
2
M.F. Moreira / Engineering Failure Analysis xxx (2015) xxx–xxx
Fig. 1. Cross section of a turbocharger indicating the location of a compressor wheel.
up to 140 °C and to angular speed peak of 100,000 rpm. An estimation of tensile stress in the blade root imposed by the angular speed can be calculated by Eq. (1) [3]. Z σ¼
r tip r root
2
ρ ω r dr ¼
ρ ω2 2 2 r tip −r root 2
ð1Þ
where: σ is the tensile stress in the blade root [Pa]; ρ is the material density [kg/m3]; ω is the angular speed [rad/s] and r is the radius (tip or root) [m]. Considering the geometry of the failed aluminium compressor wheel with a density of ρ = 2700 kg/m3, a root radius of 0.0125 m, a tip radius of 0030 m and the maximum angular speed (ω) of 100,000 rpm (10,472 rad/s), the maximum tensile stress in the blade root is 110 MPa. Since the cross section area is, generally, no greater than 500 mm2, the load is equivalent to 5.6 tf hanging on each blade. To resist this load the selected aluminium alloy must be precipitation hardened to achieve the maximum yield and tensile strength. In addition, turbocharger manufacturers use a speed sensor to limit the overspeed and extend the fatigue lifetime. For even more critical commercial truck applications, Ti–6Al–4V alloy has been employed [2,4]. The aluminium alloy AA 2618 presents high creep resistance and has been used primarily as forgings, impellers and skin for the aircraft industry and compressor wheels and pistons to automotive applications [5]. The T652 temper applied in the alloy AA 2618 is equivalent to peak aged T61 temper and results in a tensile strength peak of 380 MPa, yield strength of 290 MPa [6]. The published mechanical property data of AA 2618 [6,7] shows that yield strength of T61 samples (exposed at 140 °C by 1000 h) is between 200 MPa and 220 MPa, the tensile strength is between 290 MPa and 310 MPa and the stress amplitude of fatigue endurance limit (room temperature, unnotched test pieces, LT direction and R = 0,05) is about 100 MPa [8]. The aluminium 2xxx alloys contain copper as the main alloying element. These alloys are precipitation hardened by a solution and artificial ageing heat treatments to increased strength via formation of coherent and semi-coherent Al2CuMg (S phase) precipitates in the T6 and T8 tempers. These precipitates are initially anodic to the surrounding matrix. However, as a result of the copper-depletedzones along the grain boundaries, they can become cooper rich and cathodic. In this case, a galvanic couple might be established, creating conditions for the formation of corrosion pits along the precipitate/matrix interfaces. The intergranular and stress corrosion Table 1 Wheels identification and vehicle data for the failed compressor wheels. Wheel #
Mileage [km]
Failure location (city–state)
Atmosphere
#2 #4 #6 #21 #22 #31 #32 #33 #51
39,080 52.417 156,102 152.193 68,226 204.949 273.012 62.294 266.292
Brasília — DF São Bernardo do Campo — SP Formiga — MG Cláudio — MG Franca — SP Belo Horizonte — MG Brasília — DF São José do Rio Preto — SP São Bernardo do Campo — MG
Urban Industrial Rural Rural Rural Urban Urban Rural Industrial
Please cite this article as: M.F. Moreira, Failure analysis in aluminium turbocharger wheels, Engineering Failure Analysis (2015), http://dx.doi.org/10.1016/j.engfailanal.2015.11.024
M.F. Moreira / Engineering Failure Analysis xxx (2015) xxx–xxx
3
Table 2 Alloy chemical composition results. Element
Results [wt.%]
Specification for AA 2618 [wt.%]
Cu Fe Mg Ni Si Ti Others Al
2.34 ± 0.02 1.0 ± 0.03 1.46 ± 0.02 1.11 ± 0.05 0.21 ± 0.01 0.04 ± 0.1
1.9–2.7 0.9–1.3 1.3–1.8 0.9–1.2 0.25 max. 0.04–0.1 0.15 max. Balance
Balance
Table 3 Brinell hardness testing results. Tested compressor wheels
Brinell hardness [HB]
Product specification AA 2618 T652
#2, #4, #21, #22, #31, #33 and #51
142–166
125 HB minimum
cracking resistance of 2xxx alloys varies significantly, depending on the service atmosphere and ageing conditions [9]. Previous investigation showed, for instance, that the alloy AA 2618 peak alloy aged (T61) presented “poor” pitting corrosion resistance [10] and was susceptible to stress corrosion cracking in saline solution at 75% of its yield strength [8]. Intense intergranular corrosion attack of a similar AA 2618 alloy was reported by Ber [11], however the IGC test conditions were not detailed.
Fig. 2. Typical microstructure of AA 2618 compressor wheels, wheel #4 (see Table 1). Dispersion of Al9FeNi precipitates in equiaxed Al-alpha matrix. (A) Optical microscope image, Backer etching. (B) SEI image, Backer etching. (C) and (D) STEM images featuring the S phase dispersion and a Al9FeNi precipitate.
Please cite this article as: M.F. Moreira, Failure analysis in aluminium turbocharger wheels, Engineering Failure Analysis (2015), http://dx.doi.org/10.1016/j.engfailanal.2015.11.024
4
M.F. Moreira / Engineering Failure Analysis xxx (2015) xxx–xxx
Fig. 3. X-ray diffraction results for AA 2618 alloy, featuring the presence of Al7Cu4Ni, Al7Cu2Fe, Al9FeNi, Al2CuMg phases in α matrix.
2. Materials and methods Nine compressor wheels were submitted to chemical analyses, microstructural characterization, X-ray diffraction, hardness testing and fractographic examination. Before the fractographic examination, the specimens were cleaned with polyethylene brush in water and detergent; and submitted to ultrasound cleaning in acetone during 10 min. The hardness testing was performed in the Brinell scale with 2.5 mm ball diameter and load of 612.93 N. The microstructural characterization was performed using the Olympus BX51M optical with electrolytic Backer etching and FEI Quanta 3D SEM with 20 kV with secondary electron image. Transmission microscopy of two samples extracted from wheel #4 was carried using the scanning transmission electron microscopy detector (STEM) in bright field. STEM thin foil preparation employed the Quanta 3D 30 kV Ga ion beam. STEM images were made using 5 kV and 5 mm
Fig. 4. General view of four failed compressor wheels in the as received condition. (A) Wheel #2, life of 39,080 km. (B) Wheel #6, life of 156,102 km. (C) Wheel #22, life of 68,226 km. (D) Wheel #31, life of 204.949 km.
Please cite this article as: M.F. Moreira, Failure analysis in aluminium turbocharger wheels, Engineering Failure Analysis (2015), http://dx.doi.org/10.1016/j.engfailanal.2015.11.024
M.F. Moreira / Engineering Failure Analysis xxx (2015) xxx–xxx
5
work distance. Finally, 4 mm thick sample extracted from #6 wheel was analysed using a Shimadzu XRD 6000 diffractometer with Co Kα radiation in the range of 20° to 60° (2θ) and scan speed of 0.2°/min. In addition, three unused compressor wheels (AA2618 T652 alloy featuring the same hardness and microstructure) were used for the corrosion immersion tests. The ASTM G110 standard [12] immersion test and a variant of this test using the machining oil and the degreaser used in the wheel manufacturing process were conducted to check the susceptibility of the microstructure to intergranular corrosion. This practice specifies a specimen preparation followed by the immersion test for a period of at least 6 h in a solution of NaCl and hydrogen peroxide. Another two sets of immersion tests were performed for a period of 30 days substituting the ASTM G 110 test solution by the machining oil or the degreaser liquid (used in the manufacturing process). After the corrosion immersion tests, the samples were submitted to microstructural evaluation. 3. Results 3.1. Chemical analyses The chemical composition was determined using instrumental analytic methods (ICP) in a sample extracted from wheel #6. The results and the chemical composition requirements of AA 2618 are given in Table 2 and the chemical composition results comply with the design specification.
Fig. 5. Typical “flat fracture” observed in one of the full blades of the compressor wheel samples. Note the presence of “half-moon” regions (light grey, pointed by arrows) characteristic of stable crack growth. (A) Wheel #2, life of 39,080 km. (B) Wheel #33, life of 62.294 km. (C) Wheel #4, life of 52.417 km. (D) Wheel #6, life of 156,102 km.
Please cite this article as: M.F. Moreira, Failure analysis in aluminium turbocharger wheels, Engineering Failure Analysis (2015), http://dx.doi.org/10.1016/j.engfailanal.2015.11.024
6
M.F. Moreira / Engineering Failure Analysis xxx (2015) xxx–xxx
3.2. Hardness testing The hardness testing was performed in the transversal section located at 15 mm from the top, near the crack observed in the failed wheels. The results are shown in Table 3 and the results are in accordance with the minimum hardness requirement for AA 2618T652 alloy product specification. 3.3. Microstructural characterization Fig. 2A and B shows the typical microstructure of compressor wheels in the longitudinal section. The optical and SEM microstructures show a dispersion of coarse Al9FeNi precipitates in an equiaxed Al-alpha matrix, featuring a grain size of approximately 40 μm. Fig. 2B shows evidence of differential attack at grain boundaries. STEM observations under high magnification (Fig. 2C and D) showed the dispersion of coherent phase Al2CuMg (S phase). Unfortunately, the STEM sampling did not allow the examination of the grain boundaries to observe the presence of intergranular S phase. Al7Cu4Ni and Al7Cu2Fe phases were identified by X-ray diffraction (Fig. 3).
Fig. 6. Fracture surface of compressor wheel #33 (service life of 62.294 km), showing details of region #1. Note the presence of cleavage-like cracking mode and facets with cleavage steps, which presented brittle fatigue striations. SEM-SEI.
Please cite this article as: M.F. Moreira, Failure analysis in aluminium turbocharger wheels, Engineering Failure Analysis (2015), http://dx.doi.org/10.1016/j.engfailanal.2015.11.024
M.F. Moreira / Engineering Failure Analysis xxx (2015) xxx–xxx
7
3.4. Visual and fractographic examinations The damaged compressor wheels were visually examined. Fig. 4 shows four of the failed compressor wheels in the as received condition. It can be observed that most parts of blade tips were plastic deformed or fractured. In addition, some staining can be seen in the upper camber surface of the failed compressor wheels. In all samples, one of the full blades showed a “flat fracture” near the nose. Fig. 5 shows the details of these “flat fractures” in four compressor wheels. Note that the “flat fracture” features half-moon region (light grey) and a dark grey region. Detailed fractographic examination of these “half-moon” regions (region #1) were carried (see Fig. 6). These regions presented a cleavage-like transgranular fracture. A more detailed examination of these facets revealed the presence of cleavage steps and striation marks. The macrofractographic and microfractographic results indicated that the fracture in the half-moon region (see Fig. 5) took place by fatigue. The morphology of the striation marks is typical of brittle fatigue fracture found in aluminium alloys [13,14]. The origin of the fatigue fracture is located on the upper camber surface. Fig. 7 shows the fractographic examination of the dark grey region
Fig. 7. Fracture surface of compressor wheel #33 showing details of the region #2, indicating fracture by overloading. Note the presence of coarse and fine dimples and evidence of intergranular ductile fracture. SEM-SEI.
Please cite this article as: M.F. Moreira, Failure analysis in aluminium turbocharger wheels, Engineering Failure Analysis (2015), http://dx.doi.org/10.1016/j.engfailanal.2015.11.024
8
M.F. Moreira / Engineering Failure Analysis xxx (2015) xxx–xxx
Fig. 8. — (A) Fatigue origin site on the wheel #2 (service life of 39,080 km), showing intergranular cracking. (B) Fatigue origin site on the wheel #33 (service life of 62.294 km), showing intergranular cracking. (C) Fatigue origin site on the wheel #22 (service life of 68,226 km), showing intergranular cracking. (D) Fatigue origin site on the wheel #32 (service life of 273.012 km), showing intergranular cracking. (E) and (F) Grain detachment on the upper camber surface of the wheel #51 (service life of 266.292 km). SEM-SEI.
(region #2). The microfractographic examination indicated the presence of coarse and fine dimples and intergranular ductile fracture [15], corresponding to the region of unstable fracture of the blade by overload. Additionally, SEM examination performed on the fatigue origin sites identified the presence of intergranular cracking associated with the presence of intergranular corrosion on the upper surface of the camber (see Fig. 8A to D). Additional examination of all failed compressor wheel samples showed the presence of intergranular cracking in the fatigue origin sites. In addition, grain detachment was observed on the upper camber surface (see Fig. 8E and F). Microstructural characterization on fatigue origin site of the wheel #6 shows in more detail the intergranular corrosion cracking (see Fig. 9). Please cite this article as: M.F. Moreira, Failure analysis in aluminium turbocharger wheels, Engineering Failure Analysis (2015), http://dx.doi.org/10.1016/j.engfailanal.2015.11.024
M.F. Moreira / Engineering Failure Analysis xxx (2015) xxx–xxx
9
Fig. 9. Metallographic examination on the fatigue origin site of wheel #6 (service life of 156,102 km), as polished. Aluminium matrix and Al9FeNi phase precipitation, featuring intergranular cracking near the fatigue origin site. The grain boundary seems to present a network of small precipitates. OM and SEM images.
3.5. Immersion corrosion tests Corrosion immersion tests were performed to evaluate the susceptibility to the intergranular corrosion of AA 2618T652 alloy on the saline solution and the processing fluids in order to clarify the origin of the intergranular corrosion attack on the upper camber surface. Three immersion corrosion tests were performed in unused compressor wheels (AA2618 T652 alloy featuring the same hardness and microstructure). One set of these tests followed ASTM G110 standard and the results of extension and depth of the intergranular corrosion were used to evaluate the susceptibility of the microstructure to intergranular corrosion. The general view and microstructural evaluation of the blade surface after ASTM G110 test are shown in Fig. 10. Note the intense intergranular attack in
Fig. 10. General view and microstructural examination performed in the blade after ASTM G110 standard immersion corrosion test (6 h immersion). Note the intense surface intergranular corrosion attack. Backer etching.
Please cite this article as: M.F. Moreira, Failure analysis in aluminium turbocharger wheels, Engineering Failure Analysis (2015), http://dx.doi.org/10.1016/j.engfailanal.2015.11.024
10
M.F. Moreira / Engineering Failure Analysis xxx (2015) xxx–xxx
Fig. 11. General view and microstructural examination performed in the blade after the machining oil corrosion test (30 days immersion). No evidence of intergranular attack. Backer etching.
the compressor wheel surface: the mean depth of intergranular corrosion after 6 h immersion test in a solution of NaCl and hydrogen peroxide was 220 μm and the maximum depth was 250 μm. The general view and microstructural evaluation of the blade surface after the immersion corrosion test performed in the machining oil are shown in Fig. 11, without any evidence of intergranular corrosion. Fig. 12 shows the results of immersion corrosion tests conducted with the degreaser fluid (used after machining), without any evidence of intergranular corrosion. 4. Discussion The chemical composition analyses and hardness testing results indicated that the failed wheels comply with the requirements specified to AA 2618 aluminium alloy, T652 temper (see Tables 2 and 3). The wheels failed by fatigue and the fatigue origin sites were located in the upper camber surface (see Fig. 8). The cleavage-like fatigue crack propagates (see region #1 in Fig. 6) until it reaches the critical size of the crack, followed by unstable crack propagation by overload fracture (see region #2 in the Fig. 7), releasing a piece of the blade that damaged the remaining compressor blades (see Fig. 4). The metallographic and fractographic examinations indicated intergranular corrosion, which acted as a stress raiser on the upper camber surface, and promoted the fatigue crack initiation (see Figs. 8 and 9). Environmentally assisted fatigue cracking is a complex phenomenon, being influenced by the loading cycle, the microstructure and the environment [16,17]. Hénaff et al. [18] compared the fatigue crack propagation path of two 2xxx aluminium alloys in different environments (vacuum, air, distilled water and permanent immersion in a 3.5% NaCl saline solution) and they reported a fractographic change from transgranular cleavage-like fatigue cracking to an intergranular cracking as soon as the corrosionassisted crack growth mechanism is activated by the saline solution. Lynch [19] also presented many examples of environmentally
Fig. 12. General view and microstructural examination performed in the blade after the degreaser corrosion test (30 days immersion). No evidence of intergranular attack. Backer etching.
Please cite this article as: M.F. Moreira, Failure analysis in aluminium turbocharger wheels, Engineering Failure Analysis (2015), http://dx.doi.org/10.1016/j.engfailanal.2015.11.024
M.F. Moreira / Engineering Failure Analysis xxx (2015) xxx–xxx
11
assisted fatigue cracking and he observed that corrosion-fatigue in aluminium alloys involves the creation of corrosion pits at crack origin and the presence of “brittle” fatigue striations and progression marks on fracture surfaces. The present investigation does not show evidence of corrosion products or corrosion progression marks at the fracture surface that indicate corrosion-assisted fatigue cracking growing, suggesting that the stable crack propagation mechanism is essentially by fatigue. The fatigue crack was triggered by an intergranular corrosion, but the environmental features that promoted the onset of the intergranular corrosion mechanism could not be identified. The corrosion tests showed intense intergranular corrosion after 6 h immersion in a solution of NaCl and hydrogen peroxide and the resulting intergranular cracking (Fig. 10) was similar with the ones observed in the fatigue origin sites of the wheels (see Fig. 9). This result indicated that the T652 aged microstructure is susceptible to intergranular corrosion in the ASTM G110 saline solution. STEM examinations do not reveal the grain boundaries, but the differential attack observed in the SEM images (Fig. 2B), the intergranular ductile fracture mode at region #2 and the intergranular corrosion susceptibility indicates a grain boundary precipitation. By contrast, corrosion immersion tests performed with machining oil and degreaser did not show any evidence of intergranular corrosion after a period of 30 days, indicating that fluids do not act as an electrolyte. The intergranular corrosion attack was promoted by an electrolyte contamination during the life of the wheels. The source of contamination during service and the protection of the wheels to avoid corrosion must be further investigated.
5. Conclusions ¬ The present investigation showed that a fatigue process was triggered by an intergranular corrosion on the upper camber surface of the AA 2618T652 aluminium alloy compressor wheel. ¬ A set of corrosion immersion tests was carried out to evaluate intergranular corrosion susceptibility of the alloy, but only immersion in a solution of NaCl promoted intergranular corrosion cracking. ¬ The root-cause for the formation of the intergranular corrosion cracking in the compressor wheels could not be identified in the present investigation.
Acknowledgements I would like to thank my colleagues for their help during this work, particularly MSc. S. Pagotto Jr. (IPT) concerning the corrosion tests, Dr. T. Y. Fukuhara (IPT) for her patience with the STEM images and Prof C. R. F. Azevedo (EPUSP) for his support.
References [1] H. Yamagata, The Science and Technology of Materials in Automotive Engines, Woodhead Publishing Limited and CRC Press LLC, 2005. [2] B. Engels, Lifetime Prediction for Turbocharger Compressor Wheels — Why Use Titanium? 2002 (accessed at http://www.turbos.bwauto.com/pt/press/ knowledgeLibrary.aspx). [3] R.C. Reed, The Superalloys Fundamentals and Applications, Cambridge University Press, 2006. [4] R. Christmann, F. Längler, M. Habermehl, P. Fonts, L. Fontvieille, B. Warner, et al., Low-cycle Fatigue of Turbo Charger Compressor Wheels – Online Prediction and Lifetime Extension, 9th Int. Conf. Turbochargers Turbocharging, Institution of Mechanical Engineers, London, UK 2010, pp. 251–262. [5] J.S. Robinson, R.L. Cudd, J.T. Evans, Creep resistant aluminium alloys and their applications, Mater. Sci. Technol. 19 (2003) 143–155, http://dx.doi.org/10.1179/ 026708303225009373. [6] MMPDS, Metallic Materials Properties Development and Standardization, Federal Aviation Administration, 2008. [7] WFB Jr., H. Mindlin, C.Y. Ho (Eds.), Aerospace Structural Metals Handbook, West Lafayette: CINDAS/USAF CRDA Handbooks Operation Purdue University, 1993. [8] J.A. Lumm, Mechanical Properties of 2618 Aluminum Alloy — Technical Report AFML-TR-66-238, 1966. [9] J. Moran, Effects of metallurgical variables on the corrosion of aluminum alloys, ASM Handb. v.13 A Corros. Fundam. Test. Prot. 10th ed.ASM International 2003, p. 2597. [10] I. Weisshaus, L. Gal-Or, A. Kaufman, The effect of microstructure on the pitting tendency of a heat-resistant aluminium alloy (AAA 2618), Corros. Sci. 20 (1980) 1119–1127, http://dx.doi.org/10.1016/0010-938X(80)90142-0. [11] L. Ber, Accelerated artificial ageing regimes of commercial aluminum alloys. I. Al–Cu–Mg alloys, Mater. Sci. Eng. A 280 (2000) 83–90, http://dx.doi.org/10.1016/ S0921-5093(99)00660-7. [12] ASTM, ASTM G110 — Standard Practice for Evaluating Intergranular Corrosion Resistance of Heat Treatable Aluminum Alloys by Immersion in Sodium Chloride, Hydrogen Peroxide Solution, 12003 92–94 (+, 92). [13] K.R.L. Thompson, J.V. Craig, Fatigue crack growth along cleavage planes in an aluminum alloy, Metall. Trans 1 (1970) 1047–1049. [14] P.J. Forsyth, Fatigue damage and crack growth in aluminium alloys, Acta Metall. 11 (1963) 703–715, http://dx.doi.org/10.1016/0001-6160(63)90008-7. [15] A.S. Zamarripa, C. Pinna, M.W. Brown, M.P.G. Mata, M.C. Morales, T.P. Beber-Solano, Identification of modes of fracture in a 2618-T6 aluminum alloy using stereophotogrammetry, Mater. Charact. 62 (2011) 1141–1150, http://dx.doi.org/10.1016/j.matchar.2011.09.005. [16] F. Menan, G. Hénaff, Influence of frequency and exposure to a saline solution on the corrosion fatigue crack growth behavior of the aluminum alloy 2024, Int. J. Fatigue 31 (2009) 1684–1695, http://dx.doi.org/10.1016/j.ijfatigue.2009.02.033. [17] F. Menan, G. Hénaff, Influence of frequency and waveform on corrosion fatigue crack propagation in the 2024-T351 aluminium alloy in the S–L orientation, Mater. Sci. Eng. A 519 (2009) 70–76, http://dx.doi.org/10.1016/j.msea.2009.04.058. [18] G. Hénaff, F. Menan, G. Odemer, Influence of corrosion and creep on intergranular fatigue crack path in 2xxx aluminium alloys, Eng. Fract. Mech. 77 (2010) 1975–1988, http://dx.doi.org/10.1016/j.engfracmech.2010.03.039. [19] S.P. Lynch, Failures of structures and components by environmentally assisted cracking, Eng. Fail. Anal. 1 (1994) 77–90.
Please cite this article as: M.F. Moreira, Failure analysis in aluminium turbocharger wheels, Engineering Failure Analysis (2015), http://dx.doi.org/10.1016/j.engfailanal.2015.11.024
Contents lists available at ScienceDirect
Engineering Failure Analysis journal homepage: www.elsevier.com/locate/engfailanal
Failure analysis in aluminium turbocharger wheels M.F. Moreira Corrosion and Protection Laboratory, Instituto de Pesquisas Tecnológicas do Estado de São Paulo - IPT, Brazil
a r t i c l e
i n f o
Article history: Received 30 September 2014 Received in revised form 29 October 2015 Accepted 7 November 2015 Available online xxxx Keywords: AA 2618 aluminium alloy Fatigue Intergranular corrosion Turbocharger compressor wheel
a b s t r a c t This paper presents a failure analysis conducted in aluminium compressor wheels used in diesel turbocharged engines. These wheels were made with machined AA 2618T652 alloy and installed in light truck engines, which were used in rural and industrial atmospheres. The premature failures of the wheels happened after life between 40,000 km and 300,000 km, while the expected life was about 1,000,000 km. The present investigation showed that a fatigue process was triggered by intergranular corrosion on the upper camber surface of the wheel. A set of immersion corrosion tests was carried out to evaluate intergranular corrosion susceptibility of the alloy. The root-cause for the formation of the intergranular corrosion cracking in the compressor wheels could not be identified in the present investigation. The overall results indicated that an “in service” contamination of the compressor surface started the intergranular corrosion cracking, promoting the fatigue failure of the wheels. © 2015 Published by Elsevier Ltd.
1. Introduction The turbocharger is basically an air pump, which makes the air/fuel mixture more combustible by introducing more air into the engine's chamber, creating more power and torque. Hot exhaust gases that leave the engine are routed directly to promote the rotation of the turbine wheel [1]. A typical diesel turbocharger rotates at speeds in the range of 100,000 rpm to 250,000 rpm. The rotation of the compressor wheel pulls in ambient air and compresses it before pumping it into the engine's chambers. The compressed air leaving the compressor wheel housing is hot as a result of compression and friction. Fig. 1 shows a cross section of a turbocharger and the location of the aluminium compressor wheel. Light turbocharged diesel trucks presented failures of 51 compressor wheels in a total of 17,146 wheels. These wheels were made of AA2618 aluminium alloy, forged, heat treated (T652) and fully machined. This particular wheel presents 14 blades: 7 full blades and 7 small blades. The premature failures happened in light trucks running in rural and industrial atmospheres after life between 40,000 km and 300,000 km, while the expected life is about 1,000,000 km. The goals of this failure analysis were: to check whether the material of the failure wheels was in accordance with standards; and to identify the fracture mechanisms involved in the premature failure of the wheels. Nine compressor wheels samples were submitted to chemical analyses, microstructural characterization, X-ray diffraction, hardness testing and fractographic examination Table 1 presents the wheels identification, available vehicle mileage and failure location atmosphere. The major stresses of rotating components are created by centrifugal forces and the highest tensile values are located at the bore, in the plane of maximum mass concentration at the largest diameter. A second relatively highly loaded region of wheels is the blade root which is connected to the backwall, near the outer diameter [2]. In service, the component is heated by the air compression and subjected to centrifugal forces imposed by angular speed. In a light turbocharged truck the compressor can be subjected to temperatures
http://dx.doi.org/10.1016/j.engfailanal.2015.11.024 1350-6307/© 2015 Published by Elsevier Ltd.
Please cite this article as: M.F. Moreira, Failure analysis in aluminium turbocharger wheels, Engineering Failure Analysis (2015), http://dx.doi.org/10.1016/j.engfailanal.2015.11.024
2
M.F. Moreira / Engineering Failure Analysis xxx (2015) xxx–xxx
Fig. 1. Cross section of a turbocharger indicating the location of a compressor wheel.
up to 140 °C and to angular speed peak of 100,000 rpm. An estimation of tensile stress in the blade root imposed by the angular speed can be calculated by Eq. (1) [3]. Z σ¼
r tip r root
2
ρ ω r dr ¼
ρ ω2 2 2 r tip −r root 2
ð1Þ
where: σ is the tensile stress in the blade root [Pa]; ρ is the material density [kg/m3]; ω is the angular speed [rad/s] and r is the radius (tip or root) [m]. Considering the geometry of the failed aluminium compressor wheel with a density of ρ = 2700 kg/m3, a root radius of 0.0125 m, a tip radius of 0030 m and the maximum angular speed (ω) of 100,000 rpm (10,472 rad/s), the maximum tensile stress in the blade root is 110 MPa. Since the cross section area is, generally, no greater than 500 mm2, the load is equivalent to 5.6 tf hanging on each blade. To resist this load the selected aluminium alloy must be precipitation hardened to achieve the maximum yield and tensile strength. In addition, turbocharger manufacturers use a speed sensor to limit the overspeed and extend the fatigue lifetime. For even more critical commercial truck applications, Ti–6Al–4V alloy has been employed [2,4]. The aluminium alloy AA 2618 presents high creep resistance and has been used primarily as forgings, impellers and skin for the aircraft industry and compressor wheels and pistons to automotive applications [5]. The T652 temper applied in the alloy AA 2618 is equivalent to peak aged T61 temper and results in a tensile strength peak of 380 MPa, yield strength of 290 MPa [6]. The published mechanical property data of AA 2618 [6,7] shows that yield strength of T61 samples (exposed at 140 °C by 1000 h) is between 200 MPa and 220 MPa, the tensile strength is between 290 MPa and 310 MPa and the stress amplitude of fatigue endurance limit (room temperature, unnotched test pieces, LT direction and R = 0,05) is about 100 MPa [8]. The aluminium 2xxx alloys contain copper as the main alloying element. These alloys are precipitation hardened by a solution and artificial ageing heat treatments to increased strength via formation of coherent and semi-coherent Al2CuMg (S phase) precipitates in the T6 and T8 tempers. These precipitates are initially anodic to the surrounding matrix. However, as a result of the copper-depletedzones along the grain boundaries, they can become cooper rich and cathodic. In this case, a galvanic couple might be established, creating conditions for the formation of corrosion pits along the precipitate/matrix interfaces. The intergranular and stress corrosion Table 1 Wheels identification and vehicle data for the failed compressor wheels. Wheel #
Mileage [km]
Failure location (city–state)
Atmosphere
#2 #4 #6 #21 #22 #31 #32 #33 #51
39,080 52.417 156,102 152.193 68,226 204.949 273.012 62.294 266.292
Brasília — DF São Bernardo do Campo — SP Formiga — MG Cláudio — MG Franca — SP Belo Horizonte — MG Brasília — DF São José do Rio Preto — SP São Bernardo do Campo — MG
Urban Industrial Rural Rural Rural Urban Urban Rural Industrial
Please cite this article as: M.F. Moreira, Failure analysis in aluminium turbocharger wheels, Engineering Failure Analysis (2015), http://dx.doi.org/10.1016/j.engfailanal.2015.11.024
M.F. Moreira / Engineering Failure Analysis xxx (2015) xxx–xxx
3
Table 2 Alloy chemical composition results. Element
Results [wt.%]
Specification for AA 2618 [wt.%]
Cu Fe Mg Ni Si Ti Others Al
2.34 ± 0.02 1.0 ± 0.03 1.46 ± 0.02 1.11 ± 0.05 0.21 ± 0.01 0.04 ± 0.1
1.9–2.7 0.9–1.3 1.3–1.8 0.9–1.2 0.25 max. 0.04–0.1 0.15 max. Balance
Balance
Table 3 Brinell hardness testing results. Tested compressor wheels
Brinell hardness [HB]
Product specification AA 2618 T652
#2, #4, #21, #22, #31, #33 and #51
142–166
125 HB minimum
cracking resistance of 2xxx alloys varies significantly, depending on the service atmosphere and ageing conditions [9]. Previous investigation showed, for instance, that the alloy AA 2618 peak alloy aged (T61) presented “poor” pitting corrosion resistance [10] and was susceptible to stress corrosion cracking in saline solution at 75% of its yield strength [8]. Intense intergranular corrosion attack of a similar AA 2618 alloy was reported by Ber [11], however the IGC test conditions were not detailed.
Fig. 2. Typical microstructure of AA 2618 compressor wheels, wheel #4 (see Table 1). Dispersion of Al9FeNi precipitates in equiaxed Al-alpha matrix. (A) Optical microscope image, Backer etching. (B) SEI image, Backer etching. (C) and (D) STEM images featuring the S phase dispersion and a Al9FeNi precipitate.
Please cite this article as: M.F. Moreira, Failure analysis in aluminium turbocharger wheels, Engineering Failure Analysis (2015), http://dx.doi.org/10.1016/j.engfailanal.2015.11.024
4
M.F. Moreira / Engineering Failure Analysis xxx (2015) xxx–xxx
Fig. 3. X-ray diffraction results for AA 2618 alloy, featuring the presence of Al7Cu4Ni, Al7Cu2Fe, Al9FeNi, Al2CuMg phases in α matrix.
2. Materials and methods Nine compressor wheels were submitted to chemical analyses, microstructural characterization, X-ray diffraction, hardness testing and fractographic examination. Before the fractographic examination, the specimens were cleaned with polyethylene brush in water and detergent; and submitted to ultrasound cleaning in acetone during 10 min. The hardness testing was performed in the Brinell scale with 2.5 mm ball diameter and load of 612.93 N. The microstructural characterization was performed using the Olympus BX51M optical with electrolytic Backer etching and FEI Quanta 3D SEM with 20 kV with secondary electron image. Transmission microscopy of two samples extracted from wheel #4 was carried using the scanning transmission electron microscopy detector (STEM) in bright field. STEM thin foil preparation employed the Quanta 3D 30 kV Ga ion beam. STEM images were made using 5 kV and 5 mm
Fig. 4. General view of four failed compressor wheels in the as received condition. (A) Wheel #2, life of 39,080 km. (B) Wheel #6, life of 156,102 km. (C) Wheel #22, life of 68,226 km. (D) Wheel #31, life of 204.949 km.
Please cite this article as: M.F. Moreira, Failure analysis in aluminium turbocharger wheels, Engineering Failure Analysis (2015), http://dx.doi.org/10.1016/j.engfailanal.2015.11.024
M.F. Moreira / Engineering Failure Analysis xxx (2015) xxx–xxx
5
work distance. Finally, 4 mm thick sample extracted from #6 wheel was analysed using a Shimadzu XRD 6000 diffractometer with Co Kα radiation in the range of 20° to 60° (2θ) and scan speed of 0.2°/min. In addition, three unused compressor wheels (AA2618 T652 alloy featuring the same hardness and microstructure) were used for the corrosion immersion tests. The ASTM G110 standard [12] immersion test and a variant of this test using the machining oil and the degreaser used in the wheel manufacturing process were conducted to check the susceptibility of the microstructure to intergranular corrosion. This practice specifies a specimen preparation followed by the immersion test for a period of at least 6 h in a solution of NaCl and hydrogen peroxide. Another two sets of immersion tests were performed for a period of 30 days substituting the ASTM G 110 test solution by the machining oil or the degreaser liquid (used in the manufacturing process). After the corrosion immersion tests, the samples were submitted to microstructural evaluation. 3. Results 3.1. Chemical analyses The chemical composition was determined using instrumental analytic methods (ICP) in a sample extracted from wheel #6. The results and the chemical composition requirements of AA 2618 are given in Table 2 and the chemical composition results comply with the design specification.
Fig. 5. Typical “flat fracture” observed in one of the full blades of the compressor wheel samples. Note the presence of “half-moon” regions (light grey, pointed by arrows) characteristic of stable crack growth. (A) Wheel #2, life of 39,080 km. (B) Wheel #33, life of 62.294 km. (C) Wheel #4, life of 52.417 km. (D) Wheel #6, life of 156,102 km.
Please cite this article as: M.F. Moreira, Failure analysis in aluminium turbocharger wheels, Engineering Failure Analysis (2015), http://dx.doi.org/10.1016/j.engfailanal.2015.11.024
6
M.F. Moreira / Engineering Failure Analysis xxx (2015) xxx–xxx
3.2. Hardness testing The hardness testing was performed in the transversal section located at 15 mm from the top, near the crack observed in the failed wheels. The results are shown in Table 3 and the results are in accordance with the minimum hardness requirement for AA 2618T652 alloy product specification. 3.3. Microstructural characterization Fig. 2A and B shows the typical microstructure of compressor wheels in the longitudinal section. The optical and SEM microstructures show a dispersion of coarse Al9FeNi precipitates in an equiaxed Al-alpha matrix, featuring a grain size of approximately 40 μm. Fig. 2B shows evidence of differential attack at grain boundaries. STEM observations under high magnification (Fig. 2C and D) showed the dispersion of coherent phase Al2CuMg (S phase). Unfortunately, the STEM sampling did not allow the examination of the grain boundaries to observe the presence of intergranular S phase. Al7Cu4Ni and Al7Cu2Fe phases were identified by X-ray diffraction (Fig. 3).
Fig. 6. Fracture surface of compressor wheel #33 (service life of 62.294 km), showing details of region #1. Note the presence of cleavage-like cracking mode and facets with cleavage steps, which presented brittle fatigue striations. SEM-SEI.
Please cite this article as: M.F. Moreira, Failure analysis in aluminium turbocharger wheels, Engineering Failure Analysis (2015), http://dx.doi.org/10.1016/j.engfailanal.2015.11.024
M.F. Moreira / Engineering Failure Analysis xxx (2015) xxx–xxx
7
3.4. Visual and fractographic examinations The damaged compressor wheels were visually examined. Fig. 4 shows four of the failed compressor wheels in the as received condition. It can be observed that most parts of blade tips were plastic deformed or fractured. In addition, some staining can be seen in the upper camber surface of the failed compressor wheels. In all samples, one of the full blades showed a “flat fracture” near the nose. Fig. 5 shows the details of these “flat fractures” in four compressor wheels. Note that the “flat fracture” features half-moon region (light grey) and a dark grey region. Detailed fractographic examination of these “half-moon” regions (region #1) were carried (see Fig. 6). These regions presented a cleavage-like transgranular fracture. A more detailed examination of these facets revealed the presence of cleavage steps and striation marks. The macrofractographic and microfractographic results indicated that the fracture in the half-moon region (see Fig. 5) took place by fatigue. The morphology of the striation marks is typical of brittle fatigue fracture found in aluminium alloys [13,14]. The origin of the fatigue fracture is located on the upper camber surface. Fig. 7 shows the fractographic examination of the dark grey region
Fig. 7. Fracture surface of compressor wheel #33 showing details of the region #2, indicating fracture by overloading. Note the presence of coarse and fine dimples and evidence of intergranular ductile fracture. SEM-SEI.
Please cite this article as: M.F. Moreira, Failure analysis in aluminium turbocharger wheels, Engineering Failure Analysis (2015), http://dx.doi.org/10.1016/j.engfailanal.2015.11.024
8
M.F. Moreira / Engineering Failure Analysis xxx (2015) xxx–xxx
Fig. 8. — (A) Fatigue origin site on the wheel #2 (service life of 39,080 km), showing intergranular cracking. (B) Fatigue origin site on the wheel #33 (service life of 62.294 km), showing intergranular cracking. (C) Fatigue origin site on the wheel #22 (service life of 68,226 km), showing intergranular cracking. (D) Fatigue origin site on the wheel #32 (service life of 273.012 km), showing intergranular cracking. (E) and (F) Grain detachment on the upper camber surface of the wheel #51 (service life of 266.292 km). SEM-SEI.
(region #2). The microfractographic examination indicated the presence of coarse and fine dimples and intergranular ductile fracture [15], corresponding to the region of unstable fracture of the blade by overload. Additionally, SEM examination performed on the fatigue origin sites identified the presence of intergranular cracking associated with the presence of intergranular corrosion on the upper surface of the camber (see Fig. 8A to D). Additional examination of all failed compressor wheel samples showed the presence of intergranular cracking in the fatigue origin sites. In addition, grain detachment was observed on the upper camber surface (see Fig. 8E and F). Microstructural characterization on fatigue origin site of the wheel #6 shows in more detail the intergranular corrosion cracking (see Fig. 9). Please cite this article as: M.F. Moreira, Failure analysis in aluminium turbocharger wheels, Engineering Failure Analysis (2015), http://dx.doi.org/10.1016/j.engfailanal.2015.11.024
M.F. Moreira / Engineering Failure Analysis xxx (2015) xxx–xxx
9
Fig. 9. Metallographic examination on the fatigue origin site of wheel #6 (service life of 156,102 km), as polished. Aluminium matrix and Al9FeNi phase precipitation, featuring intergranular cracking near the fatigue origin site. The grain boundary seems to present a network of small precipitates. OM and SEM images.
3.5. Immersion corrosion tests Corrosion immersion tests were performed to evaluate the susceptibility to the intergranular corrosion of AA 2618T652 alloy on the saline solution and the processing fluids in order to clarify the origin of the intergranular corrosion attack on the upper camber surface. Three immersion corrosion tests were performed in unused compressor wheels (AA2618 T652 alloy featuring the same hardness and microstructure). One set of these tests followed ASTM G110 standard and the results of extension and depth of the intergranular corrosion were used to evaluate the susceptibility of the microstructure to intergranular corrosion. The general view and microstructural evaluation of the blade surface after ASTM G110 test are shown in Fig. 10. Note the intense intergranular attack in
Fig. 10. General view and microstructural examination performed in the blade after ASTM G110 standard immersion corrosion test (6 h immersion). Note the intense surface intergranular corrosion attack. Backer etching.
Please cite this article as: M.F. Moreira, Failure analysis in aluminium turbocharger wheels, Engineering Failure Analysis (2015), http://dx.doi.org/10.1016/j.engfailanal.2015.11.024
10
M.F. Moreira / Engineering Failure Analysis xxx (2015) xxx–xxx
Fig. 11. General view and microstructural examination performed in the blade after the machining oil corrosion test (30 days immersion). No evidence of intergranular attack. Backer etching.
the compressor wheel surface: the mean depth of intergranular corrosion after 6 h immersion test in a solution of NaCl and hydrogen peroxide was 220 μm and the maximum depth was 250 μm. The general view and microstructural evaluation of the blade surface after the immersion corrosion test performed in the machining oil are shown in Fig. 11, without any evidence of intergranular corrosion. Fig. 12 shows the results of immersion corrosion tests conducted with the degreaser fluid (used after machining), without any evidence of intergranular corrosion. 4. Discussion The chemical composition analyses and hardness testing results indicated that the failed wheels comply with the requirements specified to AA 2618 aluminium alloy, T652 temper (see Tables 2 and 3). The wheels failed by fatigue and the fatigue origin sites were located in the upper camber surface (see Fig. 8). The cleavage-like fatigue crack propagates (see region #1 in Fig. 6) until it reaches the critical size of the crack, followed by unstable crack propagation by overload fracture (see region #2 in the Fig. 7), releasing a piece of the blade that damaged the remaining compressor blades (see Fig. 4). The metallographic and fractographic examinations indicated intergranular corrosion, which acted as a stress raiser on the upper camber surface, and promoted the fatigue crack initiation (see Figs. 8 and 9). Environmentally assisted fatigue cracking is a complex phenomenon, being influenced by the loading cycle, the microstructure and the environment [16,17]. Hénaff et al. [18] compared the fatigue crack propagation path of two 2xxx aluminium alloys in different environments (vacuum, air, distilled water and permanent immersion in a 3.5% NaCl saline solution) and they reported a fractographic change from transgranular cleavage-like fatigue cracking to an intergranular cracking as soon as the corrosionassisted crack growth mechanism is activated by the saline solution. Lynch [19] also presented many examples of environmentally
Fig. 12. General view and microstructural examination performed in the blade after the degreaser corrosion test (30 days immersion). No evidence of intergranular attack. Backer etching.
Please cite this article as: M.F. Moreira, Failure analysis in aluminium turbocharger wheels, Engineering Failure Analysis (2015), http://dx.doi.org/10.1016/j.engfailanal.2015.11.024
M.F. Moreira / Engineering Failure Analysis xxx (2015) xxx–xxx
11
assisted fatigue cracking and he observed that corrosion-fatigue in aluminium alloys involves the creation of corrosion pits at crack origin and the presence of “brittle” fatigue striations and progression marks on fracture surfaces. The present investigation does not show evidence of corrosion products or corrosion progression marks at the fracture surface that indicate corrosion-assisted fatigue cracking growing, suggesting that the stable crack propagation mechanism is essentially by fatigue. The fatigue crack was triggered by an intergranular corrosion, but the environmental features that promoted the onset of the intergranular corrosion mechanism could not be identified. The corrosion tests showed intense intergranular corrosion after 6 h immersion in a solution of NaCl and hydrogen peroxide and the resulting intergranular cracking (Fig. 10) was similar with the ones observed in the fatigue origin sites of the wheels (see Fig. 9). This result indicated that the T652 aged microstructure is susceptible to intergranular corrosion in the ASTM G110 saline solution. STEM examinations do not reveal the grain boundaries, but the differential attack observed in the SEM images (Fig. 2B), the intergranular ductile fracture mode at region #2 and the intergranular corrosion susceptibility indicates a grain boundary precipitation. By contrast, corrosion immersion tests performed with machining oil and degreaser did not show any evidence of intergranular corrosion after a period of 30 days, indicating that fluids do not act as an electrolyte. The intergranular corrosion attack was promoted by an electrolyte contamination during the life of the wheels. The source of contamination during service and the protection of the wheels to avoid corrosion must be further investigated.
5. Conclusions ¬ The present investigation showed that a fatigue process was triggered by an intergranular corrosion on the upper camber surface of the AA 2618T652 aluminium alloy compressor wheel. ¬ A set of corrosion immersion tests was carried out to evaluate intergranular corrosion susceptibility of the alloy, but only immersion in a solution of NaCl promoted intergranular corrosion cracking. ¬ The root-cause for the formation of the intergranular corrosion cracking in the compressor wheels could not be identified in the present investigation.
Acknowledgements I would like to thank my colleagues for their help during this work, particularly MSc. S. Pagotto Jr. (IPT) concerning the corrosion tests, Dr. T. Y. Fukuhara (IPT) for her patience with the STEM images and Prof C. R. F. Azevedo (EPUSP) for his support.
References [1] H. Yamagata, The Science and Technology of Materials in Automotive Engines, Woodhead Publishing Limited and CRC Press LLC, 2005. [2] B. Engels, Lifetime Prediction for Turbocharger Compressor Wheels — Why Use Titanium? 2002 (accessed at http://www.turbos.bwauto.com/pt/press/ knowledgeLibrary.aspx). [3] R.C. Reed, The Superalloys Fundamentals and Applications, Cambridge University Press, 2006. [4] R. Christmann, F. Längler, M. Habermehl, P. Fonts, L. Fontvieille, B. Warner, et al., Low-cycle Fatigue of Turbo Charger Compressor Wheels – Online Prediction and Lifetime Extension, 9th Int. Conf. Turbochargers Turbocharging, Institution of Mechanical Engineers, London, UK 2010, pp. 251–262. [5] J.S. Robinson, R.L. Cudd, J.T. Evans, Creep resistant aluminium alloys and their applications, Mater. Sci. Technol. 19 (2003) 143–155, http://dx.doi.org/10.1179/ 026708303225009373. [6] MMPDS, Metallic Materials Properties Development and Standardization, Federal Aviation Administration, 2008. [7] WFB Jr., H. Mindlin, C.Y. Ho (Eds.), Aerospace Structural Metals Handbook, West Lafayette: CINDAS/USAF CRDA Handbooks Operation Purdue University, 1993. [8] J.A. Lumm, Mechanical Properties of 2618 Aluminum Alloy — Technical Report AFML-TR-66-238, 1966. [9] J. Moran, Effects of metallurgical variables on the corrosion of aluminum alloys, ASM Handb. v.13 A Corros. Fundam. Test. Prot. 10th ed.ASM International 2003, p. 2597. [10] I. Weisshaus, L. Gal-Or, A. Kaufman, The effect of microstructure on the pitting tendency of a heat-resistant aluminium alloy (AAA 2618), Corros. Sci. 20 (1980) 1119–1127, http://dx.doi.org/10.1016/0010-938X(80)90142-0. [11] L. Ber, Accelerated artificial ageing regimes of commercial aluminum alloys. I. Al–Cu–Mg alloys, Mater. Sci. Eng. A 280 (2000) 83–90, http://dx.doi.org/10.1016/ S0921-5093(99)00660-7. [12] ASTM, ASTM G110 — Standard Practice for Evaluating Intergranular Corrosion Resistance of Heat Treatable Aluminum Alloys by Immersion in Sodium Chloride, Hydrogen Peroxide Solution, 12003 92–94 (+, 92). [13] K.R.L. Thompson, J.V. Craig, Fatigue crack growth along cleavage planes in an aluminum alloy, Metall. Trans 1 (1970) 1047–1049. [14] P.J. Forsyth, Fatigue damage and crack growth in aluminium alloys, Acta Metall. 11 (1963) 703–715, http://dx.doi.org/10.1016/0001-6160(63)90008-7. [15] A.S. Zamarripa, C. Pinna, M.W. Brown, M.P.G. Mata, M.C. Morales, T.P. Beber-Solano, Identification of modes of fracture in a 2618-T6 aluminum alloy using stereophotogrammetry, Mater. Charact. 62 (2011) 1141–1150, http://dx.doi.org/10.1016/j.matchar.2011.09.005. [16] F. Menan, G. Hénaff, Influence of frequency and exposure to a saline solution on the corrosion fatigue crack growth behavior of the aluminum alloy 2024, Int. J. Fatigue 31 (2009) 1684–1695, http://dx.doi.org/10.1016/j.ijfatigue.2009.02.033. [17] F. Menan, G. Hénaff, Influence of frequency and waveform on corrosion fatigue crack propagation in the 2024-T351 aluminium alloy in the S–L orientation, Mater. Sci. Eng. A 519 (2009) 70–76, http://dx.doi.org/10.1016/j.msea.2009.04.058. [18] G. Hénaff, F. Menan, G. Odemer, Influence of corrosion and creep on intergranular fatigue crack path in 2xxx aluminium alloys, Eng. Fract. Mech. 77 (2010) 1975–1988, http://dx.doi.org/10.1016/j.engfracmech.2010.03.039. [19] S.P. Lynch, Failures of structures and components by environmentally assisted cracking, Eng. Fail. Anal. 1 (1994) 77–90.
Please cite this article as: M.F. Moreira, Failure analysis in aluminium turbocharger wheels, Engineering Failure Analysis (2015), http://dx.doi.org/10.1016/j.engfailanal.2015.11.024