Available online at www.sciencedirect.

com

Procedia Engineering 10 (2011) 1297–1302

ICM11

Fatigue Properties of High Strength Bolts

MARCELO, A.L.a, UEHARA, A.Y.a, UTIYAMA, R.M.b, FERREIRA, I.a*
a
Department fo Materials,College of Mechanical Engineering, University of Campinas (UNICAMP), Campinas, SP, CEP13083-
970, Brazil
b
NHK Fastener, Ribeirão Pires, SP, CEP09441-000, Brazil

Abstract

The aim of this paper is to analyze fatigue properties of high strength bolts in different metallurgical conditions. M8
bolts with coarse threads were separated in seven groups, three manufactured with AISI 4135 steel and heat treated
after the process of thread rolling with three tempering temperatures (TRBHT). Four groups manufactured with SCM
435H steel and same conditions. In one group, the threads were made after quenching and tempering (TRAHT).
Metallographic analyses, tension, hardness, and fatigue tests were performed. All TRBHT groups presented same
fatigue resistance. TRAHT group presented fatigue limit approximately 9% superior than other groups.
© 2011 Published by Elsevier Ltd. Open access under CC BY-NC-ND license.
Selection and peer-review under responsibility of ICM11
Keywords: Fatigue; Bolt; Preload; Tempering Temperature; Stress Ratio.

1. Introduction

The literature on fatigue in bolts with threads rolled before and after heat treatment is contradictory. In
many cases, axial fatigue tests were performed with stress ratio (R) equal to 0.1, which is not a desirable
condition of preload, since the bolts of high strength are used with higher preload and, consequently,
higher stress ratio. Under low stress ratio, a significant increase in the fatigue limit generally occurs.
These beneficial results, usually, have been extrapolated to high stress ratio without verification.
However, some researchers suggest that the benefits due to rolling the threads after the heat treatment will
be significantly reduced for high stress ratios. This reduction can be associated with a relief of the
residual stress due to high preload according to Stephens, et al [1]. With this uncertainty, the industry
prefers to roll the threads before the heat treatment, in order to maximize the rolling die life and minimize

* Corresponding author. Tel.: +55 19 35213312; fax: +55 19 32893722.

E-mail address: itamar@fem.unicamp.br.

1877-7058 © 2011 Published by Elsevier Ltd. Open access under CC BY-NC-ND license.
Selection and peer-review under responsibility of ICM11 doi:10.1016/j.proeng.2011.04.216
1298 Marcelo, A.L. et al. / Procedia Engineering 10 (2011) 1297–1302

the costs. Moreover, rolling the threads after the heat treatment (quenching and tempering) results in a
more frequent replacement of the rolling die, decreasing the production rates and increasing the costs
according to Kepheart [2] and Marcelo [3].
Given the constant need for optimization of the mechanical properties, specially the fatigue limit of the
high strength bolts, this study aims to characterize and analyze the influence of tempering temperature,
the sequence of the manufacturing process (threads rolled before and after heat treatment) at the axial
fatigue limit of M8 bolts, with 40mm of length and high strength (ISO class 10.9), manufactured with two
different steels: AISI 4135 and SCM 435H.

Nomenclature

I490, 520, 550 Bolts made of SCM 435H steel and tempered at 490, 520, and 550°C, respectively
I550 RAHT Bolts made of SCM 435H steel, tempered at 550°C, and Threads Rolled After Heat
Treatment
I550 RAHT Bolts made of SCM 435H steel, tempered at 550°C, and Threads Rolled After Heat
Treatment

2. Materials and Methods

M8 bolts, with 40mm of length and high strength (ISO class 10.9) were manufactured from two wires,
one made from the AISI 4135, and the other one made from SCM 435H. These wires have diameter from
7.80 to 7.85mm, and a favorable metallurgical condition for forging, which means low strength and high
levels of ductility [3]. For both wires, chemical analysis were performed in a LECO analyzer to determine
the carbon and sulfur contents. Moreover, for other elements, it was used a spectrometer of X-ray
Fluorescence Shimadzu EDX 900Hs. Also, the microstructural characterization of the samples was
performed using an optical microscope Zeiss, Neophot 32 model. Metallographic Samples of the
longitudinal section from the wires and bolts were prepared, and then etched by immersion in Nital 2%.
The tensile tests were performed in a servo-hydraulic MTS, Test Star II model, with 10 tons capacity
and 0.02mm/s piston speed. The tensile tests in the wires were carried out according to ISO 6892 [4]. The
hardness tests were carried out on a Buehler, model 2100, with load of 1kgf. The tests were performed at
the central region of the longitudinal, bolts and wires surface, of the metallographic samples. Three
measurements were performed for each studied condition. Axial fatigue tests were performed in intact
bolts according to ISO 3800 [5], in order to obtain the fatigue limit at 106 cycles. The tests were
conducted at the same machine where the tensile test was performed, sine wave, 25Hz frequency, and
tension preload (minimum stress) of 700MPa.

3. Results and Discussion

Table 1 present the results of chemical analysis corresponding to the wires used in the manufacturing
process of the bolts. It was possible to observe small differences in the concentration of some alloying
 Marcelo, A.L. et al. / Procedia Engineering 10 (2011) 1297–1302 1299

elements due to small variations in composition between different regions of the wire, which are
irrelevant when it comes to mechanical properties.

Figure 1 shows the micrographs taken from longitudinal sections of the wires observed by optical
microscopy after etching by immersion in Nital 2%. In both cases, the microstructure has a ferritic matrix
with spheroidized carbides. It was also possible to observe that the ferrite matrix was composed
predominantly by equiaxed grains with similar grain size. In addition to this, the wire microstructure
manufactured with AISI 4135 apparently revealed a higher number of spheroidized carbides when
compared to the one manufactured with SCM 435H, which results in different mechanical properties. The
micrographic analysis of the bolts did not reveal significant differences showing tempered martensite,
where the microstructural difference between the wires was removed due to heat treatment.

Also, it was possible to observe that the bolts made of SCM 435H steel, tempered at 550°C, and
threads Rolled After Heat Treatment (I550 RAHT) group showed characteristics rolling texture at the
nearby of the threads fillets due to the fact that the rolling was done after quenching and tempering, what
could be associated to the delay observed at the fatigue crack growth, according to Ifergane [6]. When it
comes to bolts rolled after heat treatment, the ISO 898-1 [7] does not recommend the machining of
specimens from the bolts for the tensile tests, due to the residual stress of compression at the threads roots
and also because an alignment of the grains at these regions. Therefore, tests were conducted in the intact
bolts. From the results provided by tensile tests performed in entire bolts, it is possible to observe that the
higher the tempering temperature, the lower is the strength of the bolts.

However, the group of bolts with threads rolled after heat treatment (I550 RAHT) showed higher
levels of strength than those heat treated with the same tempering temperature, but with threads rolled
before the heat treatment (N550 and I550), and also higher when compared to bolts with tempered
temperatures a little lower (N520 and I520), as observed in Table 2.

Besides, results shown in Table 2 are on average and with standard deviation for three tests, with the
conversion between Vickers microhardness (HV1) to Rockwell C (HRC). The AISI 4135 steel has
hardness levels lower than those found for wires made of SCM 435H steel in the same conditions (as
received), which is consistent with microstructure conditions shown in the metallographic analysis.

Also, the mechanical properties observed from the tension tests for this steel were better. From Vickers
microhardness, values of Table 2 for bolts with thread rolled before heat treatment, it is possible to
observe, as expected, that the higher the tempering temperature, the lower are the hardness levels.

Table 1. Chemical composition of the AISI 4135 and SCM 435H steels.

C Si Mn P S Cr Mo
AISI 4135 0.36 0.26 0.81 0.022 0.010 0.97 0.17
Standard AISI 4135 0.34 0.23 0.79 0.018 0.008 0.94 0.16
SCM 435H 0.35 0.19 0.76 0.032 0.011 0.98 0.15
Standard SCM 435H 0.35 0.20 0.74 0.012 0.006 0.99 0.17
1300 Marcelo, A.L. et al. / Procedia Engineering 10 (2011) 1297–1302

Table 2. Tensile strength of the bolts, Vickers microhardness results performed in the core region of the longitudinal section of
bolts and wires, and estimation of the average fatigue limit for 10 6 cycles for bolts of all studied conditions using preload of 700
MPa (R§0.8) and staircase method.

*
* * Rockwell C Fatigue limit (Sf)R=0.8
Tensile Strength Vickers Hardness *
Condition-Sample Hardness
(HV1) Maximun load Maximum stress
(MPa)
(HRC) [kN] [MPa]
AISI 4135 wire -- 173.5 r 2.1 -- -- ---
SCM 435 wire -- 180.0 r 2.0 -- --- ---
N490 1,211 ± 1 392.0 r 2.6 40.0 r 0.3 31.0 r 0.27 845.9 r 7.2
N520 1,119 ± 5 369.7 r 0.8 37.7 r 0.1 30.5 r 0.27 832.2 r 7.2
N550 1,025 ± 7 339.0 r 2.4 34.3 r 0.3 31.2 r 0.27 853.7 r 7.2
I490 1,233 ± 7 393.0 r 3.6 40.1 r 0.4 31.1 r 0.35 849.8 r 9.7
I520 1,070 ± 4 351.6 r 2.3 35.7 r 0.3 31.0 r 0.27 845.9 r 7.2
I550 1,051 ± 13 343.6 r 2.3 34.8 r 0.2 31.1 r 0.35 849.8 r 9.7
I550 RAHT 1,154 ± 9 383.5 r 1.2 39.1 r 0.1 33.7 r 0.65 920.0 r 17.8

Note: * Average r Standard deviation.

However, the group of bolts made of I550 RAHT presented hardness levels greater than the group of
bolts tempered at the same temperature and even for those bolts with tempering temperatures a little
lower with threads rolled before heat treatment (I550, N550, I520, and N520). The hardness is associated
with hardening of the material during the thread rolling performed after heat treatment. In other words,
thread rolling after heat treatment affects the core region of the bolt.

Table 2 also presents an estimative for the average of axial fatigue limits (with 50% of failure
probability) and the standard deviations for 106 cycles from all studied conditions. These results were
obtained through axial fatigue tests using the statistical staircase method.

The results in Table 2 shows that for bolts with threads rolled before heat treatment, the variations in
tempering temperature had a little or no influence in the fatigue limit with preload of 700 MPa and load
ratio of R=0.8.

Also, the results related to the axial fatigue tests in terms of fatigue limit curves for 106 cycles as a
function of the hardness levels using preload (700 MPa) and creating high load ratios around R=0.8 are
represented in Figure 2. From the observation of this figure, it is possible to note that the fatigue limits
based in the maximum load presented a little variation with the change in the hardness levels for these
materials, which means that little variations presented in the strength of the material did not modify the
fatigue limit significantly.
 Marcelo, A.L. et al. / Procedia Engineering 10 (2011) 1297–1302 1301

(a) (b)

(c) (d)

Fig. 1. (a) Micrograph from the longitudinal section of the wire AISI 4135; (b) Micrograph from the longitudinal section of the wire
SCM 435H; (c) Microstructure of the SCM 435H bolt tempered at 5500C with threads rolled before heat treatment; (d)
Microstructure of the SCM 435H bolt tempered at 550 0C with threads rolled after heat treatment.

Fig. 2. Variation of the fatigue limit with preload of 700 MPa (load ratio of approximately 0.8), based on the maximum load and 10 6
cycles, as a function of Rockwell C hardness, for the bolts M8 made of AISI 4135 and SCM 435H steels, quenched and tempered at
490, 520, and 550°C.
1302 Marcelo, A.L. et al. / Procedia Engineering 10 (2011) 1297–1302

3. Conclusions

The AISI 4135 and SCM 435H steels presented chemical compositions closed to each other and in
accordance with its respective standards. Metallographic analysis revealed that both wires have ferritic
matrix with approximate grain size. All groups, with heat treatment (quenching and tempering), presented
tempered martensite microstructure differing only by the well defined rolling texture in the boundary of
the root fillets of the group where threads were rolled after quenching and tempering at 550°C.
Besides, AISI 4135 steel presented lower strength and higher ductility than SCM 435H steel. Based on
the tensile properties of the bolts, as expected, the higher the tempering temperature, the lower the
strength levels. Furthermore, bolts with threads rolled after heat treatment presented higher strength than
those heat treated at the same tempering temperature and thread rolled before heat treatment. These levels
were also higher than those from bolts tempered at a lower temperature. Moreover, wires made of AISI
4135 steel had hardness levels lower than those from the SCM 435H steel, while bolts tempered with
higher temperatures presented lower hardness. However, bolts with thread rolled after heat treatment
presented higher hardness than those tempered at the same temperature (or with a little lower
temperatures) with thread rolled before heat treatment.
The results of axial fatigue tests in a neutral environment for bolts with thread rolled before heat
treatment revealed that variations in tempering temperatures (490, 520, and 550°C) of the steels had little
influence on the fatigue limit with constant preload (minimum stress) of 700 MPa and load ratio of about
0.8. On the other hand, fatigue tests (with load ratio of about 0.8) for bolts with thread rolled after heat
treatment presented an increase of about 9% in the fatigue limit compared with those with thread rolled
before heat treatment. This increase can be explained due to the introduction of compressive residual
stresses on the surface of the bolt, and by the alignment of the grains in the vicinity of the root of the
thread in the rolling direction, which difficults the nucleation and propagation of the fatigue cracks.

Acknowledgements

The authors would like to acknowledge NHK Fastener, Capes, and CNPq for financial supports.

References

[1] Stephens, R.I., Bradley, N.J., Horn, N.J., Arken, J.M., Gradman, J.J. Influence of cold rolling threads before or after heat
treatment on fatigue resistance of high strength coarse thread volts for multiple preload conditions. J ASTM Inter 2006; v.3, (3).
[2] Kephart, A.R. Optimum thread rolling process that improves SCC resistance. J ASTM Inter 2006, v.3, (7).
[3] Marcelo, A.L. Fatigue properties of high strength bolts. College of Mechanical Engineering, University of Campinas, 2008.
Dissertation (In Portuguese).
[4] ISO 6892 (1998). Metallic materials-tensile testing at ambient temperature. International Organization for Standardization,
1998.
[5] ISO 3800 (1993). Threaded fastners-axial load fatigue testing-test methods and evaluation of results. International
Organization for Standardization, 1993.
[6] Ifergane, S., Eliaz, N., Stern, N., Kogan, E., Shemesh, G., Sheinkopf, H., Eleizer, D. The effect of manufacturing processes
on fatigue lifetime of aeronautical bolts. Eng Fail Anal 2001, v.8:227-235.
[7] ISO 898-1 (1999). Mechanical properties of fastners made of carbon steel and alloy steel. Part 1: Bolts, screws and studs.
International Organization for Standardization, 1999.