Engineering Failure Analysis 129 (2021) 105679

Contents lists available at ScienceDirect

Engineering Failure Analysis

journal homepage: www.elsevier.com/locate/engfailanal

Analysis of an automotive coil spring fracture

Felipe Bergh a, *, Gilmar Cordeiro Silva b, Caio Silva b, Pedro Paiva b
a
Graduate Program in Mechanical Engineering – Universidade Federal de Minas Gerais, Av. Antônio Carlos, 6627 Belo Horizonte, Brazil
b
Pontifícia Universidade Católica de Minas Gerais, Brazil

A R T I C L E I N F O A B S T R A C T

Keywords: Failure analysis of a coil spring which failed prematurely during service was carried out using
Coil Spring optical and scanning electron microscopy, hardness testing, metallography and optical emission
Spring Steel spectroscopy. Indentation marks, which described a helical pattern along the spring wire, were
Failure analysis
identified in the spring wire below the protective paint. By analyzing the manufacturing process,
Fracture
Radial forging
which included a radial forging stage, it was established that these indentation marks were due to
a crack in the forging hammer. One of these marks propagated fatigue cracks, which is confirmed
by the presence of ratchet marks on the fracture surface. Observation of a notch at the crack
origin, followed by fatigue mechanism and intergranular cracking implies that some sort of stress
concentration might have facilitated crack growth and led to failure. It is suggested that radial
forging should be carried out by recommended procedures and the status of the forging hammers
should be checked before each major run. Careful inspection of the surface after forging must also
be carried out to assure product quality and avoid premature failure.

1. Introduction

Coil springs are an important item in many types of suspension systems, used in front or rear sides of the chassis, whose function is
to absorb the impact energy received by the wheels when in contact with irregular terrain, softening the oscillations transmitted to the
car body [1]. Springs typically operate in the elastic regime, which means they return to their original shape after enduring service
loads. To comply with this specification, springs should be manufactured in a resilient material, and typical material choices involve
grades of martensitic steel [2].
Material selection of springs should comply with the needs of the automotive industry, whose focus on weight reduction has been
increasing in recent years [3–5]. This practice aims to minimize fuel consumption and gas emissions, a critical concern during the
development of cars driven by fossil fuel engines [6]. In order to achieve the aforementioned weight reduction, some mechanical parts,
like the coil spring, are downsized and subjected to higher stresses. These more complex performance demands can lead to spring
failures, whose root causes should be carefully investigated by using failure analysis methods.
Typical coil spring failure causes, such as raw material defects, surface imperfections, improper heat treatment, corrosion and
decarburization are presented by Prawoto et al. [7], alongside a general overview of the stress distribution in the whole spring. In a
case study investigated by Kosec et al. [8], a coil spring fracture was caused by a damage in the protective paint, which allowed the
formation of a corrosion pit that acted as a stress raiser and led to a fatigue fracture. A similar a failure mode was found in the study by
Vukelic & Brcic [9], where a spring fracture occurred after the formation of a corrosion pit due to a damaged coating.

* Corresponding author.
E-mail address: [email protected] (F. Bergh).

https://doi.org/10.1016/j.engfailanal.2021.105679
Received 7 April 2021; Received in revised form 28 July 2021; Accepted 9 August 2021
Available online 12 August 2021
1350-6307/© 2021 Elsevier Ltd. All rights reserved.
F. Bergh et al. Engineering Failure Analysis 129 (2021) 105679

Using an alternative approach, Pastorcic et al. [10] combined virtual analysis and laboratory testing in the failure analysis of a coil
spring. The laboratorial examination revealed a corrosion fatigue failure initiated in point defects. In addition to that, a finite element
analysis of the part was performed in order to quantify its fatigue life reduction due to the presence of that point defect.
A different failure mode was observed by Maciejewski and Akyuz [11], where coil spring fractures occurred due to the localized
formation of untempered martensite at some surface regions. This phase transformation was caused by in-service stray current arcing,
which allowed the initiation of intergranular brittle cracks and reduced the fatigue life of the spring. Failure due to the formation of
untempered martensite was also reported by Schroeder [12] when analyzing a sleet scraper spring. In that study, localized melting and
resolidification of the metal were caused by stray electrical currents through the spring. This event led to the formation of the brittle
untempered martensite, which developed cracks as a result of the uncontrolled heating–cooling cycles.
Another root cause of a compression spring fracture was pointed out by Rocha et al. [13]. In this study, the fatigue fracture observed
in the spring was caused by a surface decarburization that occurred during heat treatment. The decarburized region presented lower
hardness compared to the spring wire’s core and the strain hardening due to shot peening did not reach the appropriate effectiveness,
allowing the initiation of fatigue. An additional type of heat treatment abnormality was reported by Das [14], to explain the failure of
spring washer samples during twist testing. The failure mode occurred due to an anomaly in the heat treatment process, which allowed
the formation of non-uniform spheroidized annealed structure of high hardness and brittleness. The microstructural features were
responsible for the variation of the spring washer performance during the tests.
Bo-Chao [15] identified the electroplating process as another source of spring steel embrittlement. This study showed a difference
in the fatigue life between coated and uncoated springs and was able to point that the electroplating process executed to apply the
coating was causing the embrittlement of the material in the outer side of the coil. As a consequence of that anomaly, there were losses
of compressive residual stresses on the surface from insufficient shot peening, which culminated in the early failure of the spring.
Failures associated to poor surface finishing were also reported in literature [16,17]. Those studies showed how excessive
roughness in highly stressed areas of a spring can act as stress raisers and lead to fatigue propagation. In such situations, microcracks
are induced from surface indentations and propagate by fatigue until the final overload fracture occurs.
Therefore, even though the methodology of failure analysis of coil springs has been developing, there is still room for future im­
provements and discussion of different root causes of those failures. This paper takes place in this context, aiming to determine the root
cause of a premature coil spring fracture that occurred during the regular operation of an automotive rear suspension.

2. Materials and methods

The evaluation of a failed coil spring, retrieved from an automotive rear suspension, started by gathering vehicle data, such as:
vehicle model, engine type, vehicle identification number (VIN) and mileage traveled before failure. After the failed coil spring was
disassembled from the vehicle, it was taken to the laboratory environment. Then, the visual inspection was conducted by capturing
images of the failed part with a Sony Cybershot Dsc-Hx400v 20.4MP camera, showing the fracture location and the macroscopic
features of the fracture surface [18].
The macrofractography was performed in an Olympus SZ61stereoscope and the images were digitally captured with the equip­
ment’s built in imaging system [19]. To evaluate the surface finish of the spring wire, the protective paint was removed. The removal
process consisted in two steps: immersion in dichloromethane reagent of analytical purity and scrubbing with a soft brush. The wire
surface images were captured using the same equipment and method used for the macrofractography.
The microfractography was done in a FEI Quanta 250 scanning electron microscope (SEM) operating at 20.0 kV [20,21]. In order to
isolate the fracture from the rest of the spring and allow the observation of the fracture surface in the equipment, the spring was cut
transversely at a distance of approximately 30 mm from the fracture, using a Teclago CM60 manual tabletop cutter with an abrasive
cut-off wheel. After cutting, the fracture surface was scrubbed with a soft brush and washed with acetone for the removal of impurities
and oxides that could compromise the observation of fracture micromechanisms.
The chemical analysis of the spring material was performed by optical emission spectroscopy (OES) in an AMETEK Spectromaxx
spectrometer, according to ASTM E 415 [22]. The OES analysis was conducted at 23 ◦ C and 50% air humidity. The samples used for
chemical analysis were cut from the failed spring using the same method used to separate the fractured region from the remainder of
the spring.
The metallographic analysis was done via optical microscopy and the images were captured in an Olympus BX60 microscope with a
digital image acquisition system [23]. The preparation of the samples for metallography consisted of cutting, mounting, grinding
(180–2500 grinding papers), diamond paste polishing (3–1 μm) and chemical etching with a solution of Nital 4% (4 mL nitric acid and
96 mL ethyl alcohol) for around 10 s [24]. The cuts were made on a Setocom-50 Struers cutter, the mounting in a Struers CitoPress-30
mounting press, the grinding and polishing in a Struers Tegramin-30 polisher.
The manufacturer’s hardness specification was defined in two different scales and both of them were evaluated in the present study:
Rockwell C (HRC) and Vickers (HV). The HRC test intended to verify the heat treatment result and was conducted according to ASTM
E18 [25], using a Wilson® Rockwell® Series 2000 hardness tester. For the HRC measurements, material of the outer curved surface
was removed up to a depth of 1 mm and the hardness was measured six times The HV test was used for cross-section measurements.
The samples for this test were cut and prepared according to ASTM E3 [26], and the test was performed with a microhardness tester
Shimadzu HSV-20, according to ASTM E384 [27], using an indentation load of 30 kgf, applied for 5 s, in six points in the coil spring
core. After all the measurements were executed, the average values of each scale were considered as representative of the hardness of
the spring.

2
F. Bergh et al. Engineering Failure Analysis 129 (2021) 105679

3. Results and discussion

3.1. Visual examination and macrofractographic study

In the visual assessment, the part was classified as a pigtail model spring [28], whose wire and coil diameters are gradually reduced
at one end (in this case, the upper end). The fracture was located in the first active coil counted from the upper end, that is, in the region
of smaller diameter of the spring wire. No damage was noted in the coating (anti-corrosion paint) that could indicate some kind of
external impact or contact between coils during the compression of the spring.
Both fracture surfaces generated from the spring failure were analyzed, being labeled as upper end fracture surface, attached to the
smaller fragment, which had progressive diameter reduction until the end of the spring wire (Fig. 1A and C), and lower end fracture,
attached to the rest of the spring (Fig. 1B and D). In the fracture origin, complementary geometries were observed: the upper end
surface presented a protrusion and the lower end surface, a groove. Ratchet marks were observed in the fracture origin, indicating the
initiation of a fatigue process from the defect found [29]. After the fatigue propagation zone, there were radial marks, which indicate a
fast crack propagation. The edges presented a sharp and shiny appearance (shear lips), characteristic of the final fracture zone.
The fatigue origin occurred in the upper region of the cross section of the spring wire (Fig. 2), a fact that does not correspond to the
region of higher active stresses, considering that the combined loads of torsion and bending should generate a peak combined stress in
the inner part of the wire cross section [7]. The fracture was completed in a plane located at approximately 45◦ with respect to the
length of the spring wire, a typical characteristic of fractures caused primarily by torsion loads in materials with high hardness and low
tenacity (brittle torsional fracture).
In the surface inspection after the removal of the protective paint, indentation marks describing a helical pattern along the wire
were observed near the fracture (Fig. 3A) and in other regions of the coil spring (Fig. 3B).

3.2. Microfractographic study

The Scanning Electron Microscope (SEM) inspection showed a smooth surface with no fracture mechanisms in the fracture origin,
indicating a lack of cohesion of the material before the fracture, that is, a manufacturing defect (Fig. 4A and B). Starting from the
surface defect found, there was the development of ratchet marks, characteristic of fatigue (Fig. 4B). The micromechanism found in the
region was striation, also representative of fatigue [17]. The defect was 1.44 mm wide and 0.16 mm deep, as shown in Fig. 4C.
In the fatigue zone, fatigue striation was observed, as expected (Fig. 5A). Following the fatigue zone, a mixed mechanism of
intergranular brittle fracture and dimples (microvoids) was detected (Fig. 5B). These mechanisms are an evidence of accelerated crack

Fig. 1. Fracture surfaces generated after the complete rupture of the coil spring (A) top and (B) bottom fragments. Crack origin details of the (C) top
and (D) bottom fragments.

3
F. Bergh et al. Engineering Failure Analysis 129 (2021) 105679

Fig. 2. Location of the fracture origin.

Fig. 3. Indentation marks observed (A) near the crack origin and (B) in another region of the spring wire.

propagation in tempered steels with high hardness, such as the SAE 9254 steel alloy used in the coil spring [30]. The final fracture
region (shear lips), present at the edges of the fracture surface, presented a mechanism formed essentially by elongated dimples
(Fig. 5C). This morphology was expected due to the change in orientation of the principal stresses plane, which makes the final
separation of the surfaces occur by shearing [30].

3.3. Chemical analysis

In Table 1 there is a comparison between the chemical composition specified in the SAE J404 material standard [31] and that
obtained in the chemical characterization of the failed spring. From the weight percentage of the main elements present in the steel
alloy SAE 9254, it could be seen that the material of the failed spring agrees with the chemical composition specifications prescribed
for this spring.

3.4. Microstructural characterization

In the metallographic evaluation, a martensite microstructure was identified in the spring core, as prescribed in the coil spring
design (Fig. 6A). The cross section metallography of the fracture origin identified the presence of partial decarburization up to a depth
of 0.09 mm (Fig. 6B). This evidence indicates that there was an unacceptable material defect at the fracture origin. Cross-sections were
also made in the surface indents, which formed a helical pattern along the spring wire, and, similarly, partial decarburization was
found inside these defects (Fig. 6C). Those features, both at the beginning of the fracture and inside the surface defects, suggest that the

4
F. Bergh et al. Engineering Failure Analysis 129 (2021) 105679

Fig. 4. Fracture origin: (A) general view, (B) surface defect found at the fracture origin, (C) surface defect dimensions.

Fig. 5. Fracture mechanisms observed in the coil spring fracture surface: (A) fatigue, (B) mixed intergranular brittle and dimples, (C) elongated
dimples due to shear.

fracture started at one of these defects. In addition, the presence of decarburization implies the existence of ferrite grains, which are not
as resistant as martensite, inside the surface defects and also indicates that those defects were originated prior to the heat treatment
[23].

5
F. Bergh et al. Engineering Failure Analysis 129 (2021) 105679

Table 1
Chemical compositions of SAE 9254 steel and of a sample retrieved from the failed spring.
Source C Si Mn P S Cr

SAE 9254 0.51–0.59 1.3–1.6 0.6–0.8 0.03 max 0.03 max 0.6–0.8
Failed Spring 0.52 1.41 0.69 0.010 0.004 0.67

Fig. 6. Coil spring cross section metallography: (A) core, (B) fracture origin, (C) surface defect.

3.5. Hardness test results

Hardness tests were carried out in order to determine if the failed part met the technical specifications prescribed in the spring
design. The failed spring met the hardness specifications on both Rockwell C and Vickers scales. It is worth mentioning that the
hardness is an indirect indication of the mechanical strength of the material, therefore, it is considered that the mechanical properties
of the spring concurred with the technical specifications of the design. Table 2 presents the average values of Rockwell C hardness and
Vickers hardness of the failed spring, as compared to the technical specifications prescribed for this spring design.

4. Conclusions

With the aid of the failure analysis techniques applied in the present study, it was possible to determine that the helical spring
fracture occurred due to fatigue under predominantly torsional loads. Fatigue life of the part was compromised by the presence of a
surface defect (groove), which generated a stress raiser, increasing the magnitude of the alternated stresses acting on the spring during
the regular use of the vehicle [16,17,32]. Thus, the actual stresses acting on the coil spring exceeded those foreseen in its design.
The material in which the spring was manufactured was in accordance with the design specifications with respect to micro­
structure, hardness and chemical composition. Therefore, it was confirmed that the raw material used in the failed helical spring was a
SAE 9254 steel alloy, quenched and tempered.
No evidence of irregularity was found in the setting of the spring in the lower or upper plates, nor of excessive movement or

Table 2
Hardness values of the failed spring in Rockwell C and Vickers scales.
Scale Failed Part Design Specifications

Rockwell C 52.8 HRC 54 ± 2 HRC

Vickers 647.3 HV 30 645 ± 6 HV 30

6
F. Bergh et al. Engineering Failure Analysis 129 (2021) 105679

displacement of the spring in its seat. There were also no impact marks or damage to the protective coating (anti-corrosion paint) that
indicated poor use of the car. Therefore, the failure mode studied here could not be related to anomalies in the assembly process and/or
negligence of the driver in vehicle care.
After gathering the aforementioned evidence, it was possible to backtrack the origin of the defect to the radial forging process used
to reduce the diameter of one end of the spring wire. There was a fracture in one of the forging hammers, which left the helical pattern
marks, found only in the region of diameter reduction. The determination of the root cause associated with the radial forging process
can be used to guide improvements in the manufacturing process, in order to eliminate this failure mode.

Declaration of Competing Interest

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

Acknowledgments

The authors are grateful to FCA Group and its technical staff. This research did not receive any specific grant from funding agencies
in the public, commercial, or not-for-profit sectors.

References

[1] R. Bartolozzi, F. Frendo, Stiffness and strength aspects in the design of automotive coil springs for McPherson front suspensions: a case study, Proc. Instit. Mech.
Engineers, Part D: J. Automobile Eng. 225 (2011) 1377–1391.
[2] Y. Yamada, Materials for Springs, Springer, Berlin Heidelberg, 2007.
[3] A. Goto, H. Nihei, P. Zhao, T. Anami, Aluminum Sheet Material Technology Supporting the Weight Reduction of a Car Body, Int. J. Automotive Eng. 2 (2011)
81–86.
[4] M. Copeland, Weight reduction and improved performance in automotive applications, Plast. Eng. 69 (10) (2013) 44–46.
[5] D. Wang, S. Zhang, C. Wang, C. Zhang, Structure-material-performance integration lightweight optimisation design for frontal bumper system, Int. J.
Crashworthiness 23 (3) (2018) 311–327.
[6] P. Malnati, Automotive composites: mass reduction for mass production: reinforced plastics lower weight, improve handling and performance, and boost safety
on passenger cars, Plast. Eng. 72 (8) (2016) 22–26.
[7] Y. Prawoto, M. Ikeda, S.K. Manville, A. Nishikawa, Design and failure modes of automotive suspension springs, Eng. Fail. Anal. 15 (2008) 1155–1174.
[8] L. Kosec, A. Nagode, G. Kosec, D. Kovačević, B. Karpe, B. Zorc, B. Kosec, Failure analysis of a motor-car coil spring, Case Stud. Eng. Fail. Anal. 4 (2015) 100–103.
[9] G. Vukelic, M. Brcic, Failure analysis of a motor vehicle coil spring, Procedia Struct. Integrity 2 (2016) 2944–2950.
[10] D. Pastorcic, G. Vukelic, Z. Bozic, Coil spring failure and fatigue analysis, Eng. Fail. Anal. 99 (2019) 310–318.
[11] J. Maciejewski, B. Akyuz, Spring Fatigue Fractures Due to Microstructural Changes in Service, J. Fail. Anal. Prev. 14 (2) (2014) 148–151.
[12] C.J. Schroeder, Metallurgical Failure Analysis of a Fractured Sleet Scraper Spring, J. Fail. Anal. Preven. 13 (5) (2013) 555–560.
[13] J.M.A.M. Rocha, A.R. Pimenta, S.R. Correa, M. Cindra Fonseca, M.G. Diniz, Fracture failure analysis in compression spring of a wagon torpedo, Eng. Fail. Anal.
122 (2021), 105245.
[14] S. Das, S. Talukder, V. Solanki, M. Adhikary, S. Bhattacharya, Breakage of Automotive Spring Washer During Twist Test: A Metallurgical Analysis, J. Fail. Anal.
Preven. 18 (2018) 241–245.
[15] L. Bo-Chao, Y. Zhen-Guo, Failure Analysis of Shock Absorption Spring in Motorcycle, J. Fail. Anal. Preven. 16 (3) (2016) 337–345.
[16] M.P. Valles González, M. García-Martínez, M.A. Pastor, Study of a torsion spring fracture, Eng. Fail. Anal. 98 (2019) 150–155.
[17] S. Yan, Q. Wang, X. Chen, C. Zhang, G. Cui, Failure Analysis of an Automobile Coil Spring in High-Stress State, J. Fail. Anal. Preven. 19 (2) (2019) 361–368.
[18] R. McSwain, Photography in Failure Analysis, in: ASM (Ed.), ASM Handbook, vol. 11 Failure Analysis and Prevention, 9th ed., ASM International, Ohio, 2003.
[19] G. Vander Voort, Visual Examination and Light Microscopy, in: ASM (Ed.), ASM Handbook, vol. 12 Fractography, 9th ed., ASM International, Ohio, 2003.
[20] B. Gabriel, Scanning Electron Microscopy, in: ASM (Ed.), ASM Handbook, vol. 12 Fractography, 9th ed., ASM International, Ohio, 2003.
[21] J.I. Goldstein, D.E. Newbury, J.R. Michael, N.W.M. Ritchie, J.H.J. Scott, D.C. Joy, Scanning Electron Microscopy and X-Ray Microanalysis, 4th ed., Springer New
York, New York, NY, 2018.
[22] ASTM E415-14, Standard Test Method for Analysis of Carbon and Low-Alloy Steel by Spark Atomic Emission Spectrometry. American Society for Testing and
Materials, 2015.
[23] ASM, ASM Handbook, Vol. 9 Metallography and Microstructures, 9th ed., ASM International, Ohio, 2003.
[24] B.L. Bramfitt, A.O. Benscoter, Chapter 7: Metallographic Specimen Preparation, in Metallographer’s Guide: Practices and Procedures for Irons and Steels, ASM
International, 2002.
[25] ASTM E18-15, Standard Test Methods for Rockwell Hardness of Metallic Materials. American Society for Testing and Materials, 2015.
[26] ASTM E3-95, Standard Practice for Preparation of Metallographic Specimens. American Society for Testing and Materials, 1995.
[27] ASTM E384-16, Standard Test Method for Microindentation Hardness of Materials. American Society for Testing and Materials, 2016.
[28] M.A. Wittek, D. Gąska, B. Łazarz, T. Matyja, Coil springs in passenger cars – general theoretical principles and structural requirements, Archives Automotive
Eng. – Archiwum Motoryzacji. 72 (2) (2016) 141–158, https://doi.org/10.14669/AM.VOL72.ART9.
[29] R. Lund, Fatigue Fracture Appearances, in: ASM (Ed.), ASM Handbook, vol. 11 Failure Analysis and Prevention, 9th ed., ASM International, Ohio, 2003.
[30] T.-H. Nam, M.-S. Kwon, J.-G. Kim, Mechanism of corrosion fatigue cracking of automotive coil spring steel, Met. Mater. Int. 21 (6) (2015) 1023–1030, https://
doi.org/10.1007/s12540-015-5326-5.
[31] SAE: SAE J404-2000 chemical compositions of SAE alloy steels. Soc. Automot. Eng. (2000).
[32] Y. Nishimura, M. Endo, K. Yanase, Y. Ikeda, S. Miyakawa, N. Miyamoto, in: Advances in Structural Integrity, Springer Singapore, Singapore, 2018, pp. 541–548,
https://doi.org/10.1007/978-981-10-7197-3_45.