Case Studies in Engineering Failure Analysis 2 (2014) 25–32

Contents lists available at ScienceDirect

Case Studies in Engineering Failure Analysis

journa l homepage: www.e lsev ier .com/ locate /csefa

Case study

Failure analysis of a bridge crane shaft§

O.A. Zambrano *, J.J. Coronado, S.A. Rodrıguez

Research Group of Fatigue and Surfaces, Mechanical Engineering School, Universidad del Valle, AA 25360 Melendez, Cali, Colombia

A R T I C L E I N F O

Article history:

Received 12 June 2013

Received in revised form 16 October 2013

Accepted 5 December 2013

Available online 9 January 2014

Keywords:

Failure analysis

Shaft

Inclusions

Fatigue

Finite element analysis

A B S T R A C T

Failure analysis of a shaft used in a bridge crane has been carried out. The shaft fractured in

the keyway with evidence of fatigue. Chemical analysis, micro-structural characterization,

fractography, hardness measurements, and finite element simulation were used for the

analysis. The microstructure was predominantly tempered martensite; large amounts of

oxides, micropores, and manganese sulfide inclusions were found. The geometry of the

keyway also promoted the initiation crack because the width and height were erroneously

designed. It was concluded that all these factors produced fatigue failure. It is

recommended to first guarantee the chemical composition and microstructure of the

material. Secondly, use magnesium or calcium additions in the steel casting process to

obtain better shape control of inclusions and, finally, accomplish the geometric

parameters recommended by the standard to avoid high stress concentration factors.

� 2013 The Authors. Published by Elsevier Ltd. All rights reserved.

1. Introduction

Shafts are used to transmit power to other mechanical elements and are generally subjected to torsional and bendingloads. One of the most common failure mechanisms in shafts is fatigue. Fatigue failures start at vulnerable points wheremetallurgical and structural defects exist that favor high localized stresses [1]. Normally, the points of stress concentration inshafts are present in sharp changes of the cross-sectional area or at the keyways [2]. Additionally, when defects appear inthese sensitive sites, the fatigue life is severely compromised.

Other failure analyses have been performed on shafts, especially related to the corners of the keyway, where thepredominant causes of the onset of fatigue failure were due to low radius of keyway curvature [3], inclusions [4,5], incorrectrepair welding [6], brittle microstructures [7], and machining marks [8]. All of these failures were present across the entirecross section of the shafts and started at the corners of the keyway. In this failure analysis, only one side of the keyway wascompletely fractured by fatigue, not the entire transversal section. Besides, this type of failure has been recurrent in thismechanical element for years.

The shaft analyzed (replacement part) belongs to a bridge crane fractured after one year of operation. Fundamentally, thebridge crane system consists of an electric motor that transmits power to the shaft and this shaft transmits power to areducer gearbox; a representation of the system is shown in Fig. 1. The keyway connects the system’s brake. According to thematerial specification provided by the manufacturer, the material is an AISI 4340 steel normalized and tempered. Enginepower is 3.73 kW and operates from 600 to 1800 rpm.

§ This is an open-access article distributed under the terms of the Creative Commons Attribution-NonCommercial-No Derivative Works License, which

permits non-commercial use, distribution, and reproduction in any medium, provided the original author and source are credited.* Corresponding author. Tel.: +57 2 4849665.

E-mail addresses: oscar.zambrano@correounivalle.edu.co, oazzambrano2009@hotmail.com (O.A. Zambrano).

2213-2902/$ – see front matter � 2013 The Authors. Published by Elsevier Ltd. All rights reserved.

http://dx.doi.org/10.1016/j.csefa.2013.12.002

[(Fig._1)TD$FIG]

Fig. 1. Bridge crane and shaft analyzed, adapted from [9].

O.A. Zambrano et al. / Case Studies in Engineering Failure Analysis 2 (2014) 25–3226

The scope of this investigation was to determine the cause of failure occurring in this shaft, an element commonly used inalmost all industries, to prevent similar failures that could be the cause of damage of the complete equipment and, not lessimportant, to guarantee industrial security for workers.

2. Experimental procedure

Chemical analysis, visual inspection, fractography, metallographic analysis, hardness measurements, and finite elementsimulation were used for the analysis. The shaft’s chemical analysis was carried out via optical emission spectroscopy (ARL3460 Advantage spectrometer); fractography was performed by using a stereoscope (Nikon SMZ1000) and scanning electronmicroscopy (Jeol jsm-6490LV); for the metallography, the samples were polished and etched (2% Nital reagent during 40 s),micro indentation Vickers was used with 10 g during 15 s and 10 indentations for hardness determination.

3. Results and discussion

3.1. Visual and stereography examination

One edge of the keyway was fractured, a general view is shown in Fig. 2(a) and (b), some important features of the fractureare shown: machined marks on the keyway, crack origin, fatigue propagation zone, plastic deformation in the final fracturezone. Also, a longitudinal mark can be noted in the keyway, with characteristics of plastic deformation near the fracture zone.

Machining marks influenced in the nucleation of fatigue cracks. Besides, a longitudinal mark in the keyway, possiblycaused when the key was put in the keyway, produced additional damage in the corner of the keyway. Additionally, Fig. 3shows a beach pattern and ratchet marks, evidence of high local stress [8–11].

Fig. 4 reveals growth of a secondary crack on the other side of the keyway because of the reversibility of the torque theshaft must withstand. This torque alternation promoted change of the stress in each corner of the keyway and, finally,induced fatigue failure.

The keyway radius was measured at 0.6 mm and the shaft diameter is 25.4 mm; therefore, the r/d ratio is 0.024. Using thisratio, for a standardized keyway, it was found that the stress concentration factor for a torsion shaft (Kts) is 2.52 [12]. The USAstandard recommends maximum stress concentration factors up to 2.6 when the torque is transmitted without the key.However, the keyway width recommended for the same standard is 6.3 mm, but the shaft had a keyway width value of9.1 mm. Likewise, the keyway height must be 3.2 mm, but the shaft had a keyway height value of 3.3 mm.

[(Fig._2)TD$FIG]

Fig. 2. (a) General view of the fractured keyway and (b) view of the fractured surface of the failed shaft.

O.A. Zambrano et al. / Case Studies in Engineering Failure Analysis 2 (2014) 25–32 27

3.2. Fractography

The shaft fractography is shown in Fig. 5. The fractured surface of the broken keyway shows that the crack initiated fromthe keyway zone in Fig. 5(a) and striations parallel to the beach marks are observed in Fig. 5(b).

[(Fig._3)TD$FIG]

Fig. 3. Beach pattern and ratchet marks in the fracture surface.

[(Fig._4)TD$FIG]

Fig. 4. Transversal view of the keyway, showing growth of a secondary crack, fracture zone, and keyway geometry.[(Fig._5)TD$FIG]

Fig. 5. SEM fractography showing (a) crack initiation region and (b) striations.

O.A. Zambrano et al. / Case Studies in Engineering Failure Analysis 2 (2014) 25–3228

3.3. Metallographic analysis

A sample was cut from the shaft fracture zone. This sample was metallographically prepared and observed in an opticalmicroscope, in unetched and etched conditions. The microstructure, without etching, revealed high quantity of defects suchas micropores and non-metallic inclusions, as shown in Fig. 6(a). Also, elongated manganese sulfide inclusions in thelongitudinal direction of the shaft, near the fracture zone, were found (Fig. 6(b)). The MnS inclusion average length was about112� 2 mm; this kind of inclusion promoted fracture occurrence [13].

[(Fig._6)TD$FIG]

Fig. 6. Optical microstructure showing (a) oxide inclusions, MnS inclusions, and micropores (longitudinal section 100�) and (b) optical microstructure

showing MnS stringer near the fracture zone (longitudinal section 500�).

O.A. Zambrano et al. / Case Studies in Engineering Failure Analysis 2 (2014) 25–32 29

These inclusions microscopically affect stress distribution and favor crack nucleation; therefore, it is important tointroduce the critical inclusion size parameter [14], related to the threshold value for fatigue crack propagation. This criticalvalue is around 50–100 mm [15]. However, when inclusions are just below the surface, inclusions smaller than 45 mm causefatigue fracture [16]. Inclusions larger than these values were found near the fracture zone.

The microstructure shown in Fig. 7 revealed tempered martensite and some ferrite grains that could be related to high-tempered temperature or a long period of tempering. Although this microstructure is apparently adequate due to its highertendency of plasticity-induced crack closure and low crack growth than other tempered microstructures for AISI 4140 [17],the yield strength and hardness decrease drastically almost to one half of the initial values, affecting the fatigue life for whichit was initially designed. Besides, fatigue cracks nucleate and propagate in zones where the strain is most favored, thisphenomenon is more evident with increased tempering temperature.

3.4. Hardness measurements

Results of microhardness measurements in the central zone and near the fracture are shown in Table 1.As shown in Table 1, the hardness measurements were similar in both zones. This hardness is inconsistent with the AISI

4340 steel normalized and tempered specified by the manufacturer. High hardness was expected, similar to the hardness oftempered martensite at 425 8C [18]. Additionally, these hardness values could be related to the AISI 4140 steel hardnesswhen tempered at 595–650 8C for 2 h [18].

[(Fig._7)TD$FIG]

Fig. 7. Optical micrograph showing the microstructure that consists of tempered martensite (450�).

Table 1

Hardness in the central zone and near the fracture zone.

Zone of indentation Micro HV mean Standard deviation

Central Zone 288 6.3

Near the Fracture Zone 293 5.8

Table 2

Chemical composition of the shaft material, AISI 4340 and AISI 4140.

Material C Mn Si S P Ni Cr Mo V Cu

4140 (obtained) 0.38 0.79 0.29 0.026 0.012 0.08 0.83 0.17 0.004 0.16

4340 (expected) 0.38–0.43 0.6–0.8 0.15–0.30 0.040 Max 0.035 Max 1.65–2.0 0.7–0.9 0.20–0.30 – –

The underlined is used to emphasize the absence of nickel in the alloy.

O.A. Zambrano et al. / Case Studies in Engineering Failure Analysis 2 (2014) 25–3230

3.5. Chemical analysis of the shaft

Measured and specified compositions of the shaft are shown in Table 2. Differences in chemical composition were found,particularly the absence of nickel (this element extends the gamma loop, increases the hardenability of the alloy, and delaysthe pearlitic and bainitic transformation [19]), producing two conditions: (i) a yield strength lower than the AISI 4340 steeland (ii) a different response of the material (microstructure); when subjected to the same AISI 4340 heat treatment,particularly, the absence of nickel possibly favors the diffusion process and facilitates the formation of some ferrite grains atlower tempering temperatures and holding times than AISI 4340.

3.6. Finite element analysis (FEA)

A static and elastic model, using finite element analysis, was carried out in two geometries: (i) the shaft with the standardrecommended geometrical dimensions and (ii) the shaft with the geometrical dimensions found in the shaft out of thestandard (Section 3.1). The effect of the dissimilar geometry on the theoretical stress concentration factor for a torsion shaft(Kts) and fatigue notch factor (Kf) was determined. It must be remembered that the USA standard only shows how the Kts

factor varies with the keyway radius for standard shaft geometries; for this reason, a finite element analysis was required onthis irregular shaft geometry.

Quadratic tetrahedral elements in the upper zone of the keyway were used and a structural meshed was used in the restof the shaft. The radius, width, and height of the keyway were refined. Rotational movement and displacements wereconstrained in the shaft gear zone (Fig. 8). The torsion load was applied in homogenous pressure form on a selected face inthe keyway (Fig. 8). Boundary conditions and the refinement zone are shown in Figs. 8 and 9, respectively.

According to the literature, Kts is a function of the r/d ratio, where r is the keyway radius and d is the shaft diameter, forr = 0.6 mm and d = 25.4 mm the stress concentration factor of the standard geometry (Kt_st) is 2.52 [12]. For a modified

[(Fig._8)TD$FIG]

Fig. 8. Boundary conditions applied to the shafts in the FEA model.

[(Fig._9)TD$FIG]

Fig. 9. Refinement of the interest zone.

[(Fig._10)TD$FIG]

Fig. 10. Distribution of von Mises stress on the stress concentration zone of (a) standard geometry and (b) modified geometry.

O.A. Zambrano et al. / Case Studies in Engineering Failure Analysis 2 (2014) 25–32 31

geometry (out of the standard), the stress concentration factor (Kt_m) can be estimated as:Kt m ¼ Kt rKt stwhere Kt_r is therelation of the both stress concentration factors. Given that,

Kt ¼smax

snoninal(1)

O.A. Zambrano et al. / Case Studies in Engineering Failure Analysis 2 (2014) 25–3232

and the nominal stress (snominal) is equal in both analyses:

Kt r ¼smax m

smax st(2)

From the results of the finite element analysis, the von Mises maximum stress of the modified geometry (smax_m) and thevon Mises maximum stress of the standard geometry (smax_st) were obtained. The Kt_r calculated is equal to 2.20 and Kt_m isequal to 5.54. The effect of Kt_st and Kt_m on von Mises stress field is show in Fig. 10.

Thus, it is confirmed that the actual stress concentration factor (Kts = 5.54) generated by the irregular dimensions of thekeyway width and height, was higher than the maximum stress concentration factor recommended by the USA standard forthe keyway radius in this shaft for standard geometries (Kts = 2.6) [12].

The fatigue notch sensitivity of the material was estimated at 0.93 [12]; for this reason, the fatigue notch factor (Kf)increased 116.4% from the standard geometry to the modified geometry, decreasing the fatigue safety factor by 53.4%.

4. Conclusions

This study analyzed the failure of a bridge crane shaft, the analysis allowed concluding as follows.Stereography examination revealed the presence of beach and ratchet marks on the fracture surface and the fractography

examination shows striations; all of this evidence indicates the shaft was fractured by fatigue.The shaft material did not satisfy the chemical composition standard; on the other hand, the microstructure was not

adequate for this application because it has low mechanical properties and large amount of defects near to the fracture zone.Specifically, the study found the length of MnS inclusions above the value of critical inclusion size parameter, whichproduced a drastic decrease in fatigue life.

The stress concentration factor was higher than recommended and the effect of the keyway width and height variationcaused 116.4% increase in the stress concentration factor, decreasing the fatigue safety factor by 53.4%.

5. Recommendations

� U

se adequate materials (guarantee the chemical composition and microstructure). � U se magnesium or calcium additions in the steel casting process to obtain better shape control of the inclusions or use

vacuum melting.

� U se the geometry recommended by the standard keyway to minimize stress concentration and try to reduce machine

marks.

Acknowledgements

The authors thank Universidad del Valle for the support obtained through project no. CI 2652 and the Surface PhenomenaLaboratory at the University of Sao Paulo, Brazil.

References

[1] 8th ed. Metals handbook, vol. 11: failure analysis and prevention, 8th ed. Materials Park, OH: American Society for Metals; 1975. p. 1482.[2] Berndt F, Van Bennekom A. Pump shaft failures a compendium of case studies. Engineering Failure Analysis 2001;8:135–44.[3] Goksenli A, Eryurek IB. Failure analysis of an elevator drive shaft. Engineering Failure Analysis 2009;16(4):1011–9.[4] Bhattacharyya S, Banerjee A, Chakrabarti I, Bhaumik SK. Failure analysis of an input shaft of skip drive gearbox. Engineering Failure Analysis

2008;15(4):411–9.[5] Casanova F, Coronado JJ. Investigacion de la falla de un eje pinon conectado a una extrusora de doble tornillo. DYNA 2010;(77):88–97.[6] Atxaga G, Irisarri AM. Failure analysis of the end of a shaft of an engine. Engineering Failure Analysis 2010;17(4):714–21.[7] Parida N, Tarafder S, Das SK, Kumar P, Das G, Ranganath VR, et al. Failure analysis of coal pulverizer mill shaft. Engineering Failure Analysis

2003;10(6):733–44.[8] Bhaumik SK, Rangaraju R, Parameswara MA, Venkataswamy MA, Bhaskaran TA, Krishnan RV. Fatigue failure of a hollow power transmission shaft.

Engineering Failure Analysis 2002;9(4):457–67.[9] SHAW-BOX1 Wire Rope Electric Hoists Manual of Series 700.

[10] Ktari A, Haddar N, Ayedi HF. Fatigue fracture expertise of train engine crankshafts. Engineering Failure Analysis 2011;18(3):1085–93.[11] Asi O. Failure analysis of a crankshaft made from ductile cast iron. Engineering Failure Analysis 2006;13(8):1260–7.[12] Peterson PW. Stress concentration factors. 3rd ed. New York: John Wiley & Sons; 2008.[13] Barbosa CJ, do Nascimento L, Caminha IMV, Abud IC, de Carvalho SS. A microstructural and fractographic study on the failure of a drive shaft. Journal of

Failure Analysis and Prevention 2011;11(6):693–9.[14] Mihakovacic SS. Critical inclusion size in spring steel and genetic programming. RMZ – Materials and Geoenvironment 2010;57:17–23.[15] Kiessling R. Nonmetallic inclusions and their effects on the properties of ferrous alloys, encyclopedia of materials: science and technology. 2nd ed.

Oxford: Elsevier; 2001. p. 6278–83.[16] Juvonen P. Effects of non-metallic inclusions on fatigue properties of calcium treated steels. Espoo, Finland: Helsinki University of Technology; 2004.[17] London B, Nelson DV, Shyne JC. The effect of tempering temperature on near-threshold fatigue crack behavior in quenched and tempered 4140 steel.

Metallurgical and Materials Transactions A 1988;19(10):2497–502.[18] Heat treater’s guide: practices and procedures for irons and steels. 2nd ed. ASM International; 1995. p. 321–49, ISBN: 978-0-87170-520-4.[19] Verhoeven J. Steel metallurgy for the non-metallurgist. Ohio: ASM International; 2007. p. 86–7.