Key Engineering Materials Vols.
525-526 (2013) pp 417-420
© (2013) Trans Tech Publications, Switzerland
doi:10.4028/www.scientific.net/KEM.525-526.417
Effects of Surface Roughness on the Fatigue Life of Alloy Steel
W.L. Xiao1, a, H.B. Chen1, b, Y. Yin1, c
1
CAS Key Laboratory Mechanical Behavior and Design of Materials, Department of Modern
Mechanics, University of Science and Technology of China, Hefei, Anhui 230026, China
a
[email protected],
[email protected],
[email protected]Keywords: surface roughness, fatigue life, power function curve
Abstract: Surface roughness characterizes the micro-geometric appearance variation and
significantly affects the fatigue properties of machined specimen. Low cycle fatigue (LCF) tests of
alloy steel specimens with different surface roughness were carried out in this paper to investigate
the effect of roughness on fatigue life. The dumbbell plate specimens were tested in uniaxial
stress-controlled mode on the hydraulic servo machine at room temperature. Obviously discrepant
lifetime results corresponding to different surface roughness implied that the greater the roughness
was, the lower the fatigue life was. An approximate power function relationship between the
roughness and the fatigue life was established through the fitting of the experimental data.
Introduction
Surface roughness, which is an unavoidable phenomenon at machining, is usually strictly required
when the processed materials are applied in structural components subjected to cyclic loads. The
reason is that the fatigue life of structures is highly dependent on the surface quality. Several
research works with respect to the influence of machined surface roughness on the fatigue life have
been published in the past decade. Early studies [1] used empirical reduction factors which
modified the endurance limit of the material to evaluate the effects. Sigmund and Bjørn [2] and
Suraratchai, et al. [3] simulated the stress state of micro surface profile and established a numerical
method for fatigue life prediction of components with rough surfaces using the finite element
method. Smaga and Eifler [4] investigated the effects of the residual stress induced from surface
roughness. However, those researches do not work for low cycle fatigue and the present work is to
seek a quantitative relationship between the roughness and the low cycle fatigue life through
experiment test.
Material and experimental setup
The material investigated in this paper is alloy steel. Its yield strength is about 420MPa obtained
from monotonic tensile test. In the fatigue test, dumbbell plate specimens were used, with a middle
parallel length of 14mm and a thickness of 4mm, as show in Fig. 1. The aim using this
configuration was to avoid unstability in the fatigue process and conveniently measure the surface
roughness. Several group specimens with different surface roughness were created firstly by means
of different machining method. Then uniaxial stress-controlled fatigue tests were carried out at
room temperature with a stress amplitude of 500MPa and a load ratio of R=-1 using a frequency of
1Hz. The load and displacement change with time and life cycles were recorded by the sensors
embedded in the hydraulic servo material test machine, MTS809.
All rights reserved. No part of contents of this paper may be reproduced or transmitted in any form or by any means without the written permission of TTP,
www.ttp.net. (ID: 218.22.21.3-14/08/12,09:34:32)
418 Advances in Fracture and Damage Mechanics XI
Fig. 1 Configuration of the dumbbell plate specimen
Results
Surface roughness is usually characterized through average geometric parameter named Ra. Thus in
this investigation, the values of Ra of the specimens were measured firstly. Based on the
measurement results, 34 specimens were divided into 10 groups with each group including 3 or 4
specimens, and each group corresponded to a roughness grade, from minimum 0.1µm to maximum
1.7µm, as listed in the first row of Table 1.
The second row in Table 1 presents the experimental fatigue lives which are the average values of
the fatigue lives of 3 or 4 specimens in each group. Obviously, the fatigue lives are less than 10,000,
showing that the failure mode of these tests belongs to low cycle fatigue. On the whole, there is a
trend that the greater the roughness is, the lower the fatigue life is, although some individual data is
biased because of intrinsic discreteness for the fatigue test. In other words, the life has an inverse
proportionality with the surface roughness. This phenomenon can be theoretically explained by a
concept called the stress concentration factor [5]. It assumes the uneven geometric profile as a lot of
side-by-side micro notches lying on the surface of machined specimen. Stress concentration
initiates from the notches. Then the stress concentration factor can be defined as
Rz
Kt = 1 + 2 γ (1)
ρ
where Kt denotes the stress concentration factor, γ refers to the ratio between spacing and depth of
the asperities, which approximately equals to 1, Rz is the 10-point roughness, ρ is the effective
profile valley radius of the surface texture. Generally speaking, the machined surface becomes
rougher, which will lead to Rz larger and ρ smaller. So it is easy to conclude that Kt has an
increscent trend. Then the stress amplitude ∆σ in the notch is the nominal stress amplitude ∆e
multiplied by Kt, as expressed in Equation (2).
∆σ = K t i∆e (2)
Key Engineering Materials Vols. 525-526 419
Attention that Kt is always greater than 1. Based on the theory of local stress-strain method, fatigue
life depends on the local stress-strain field. And a greater local stress-strain leads to a lower fatigue
life. In a word, fatigue life is reducing as the surface roughness is increasing under the premise of
the same load condition.
Table 1 Experimental fatigue life results corresponding to each roughness grade
Roughness,
0.1 0.3 0.5 0.7 0.9 1.0 1.2 1.4 1.6 1.7
Ra [µm]
Fatigue
life, Nf 3375 2856 2751 2748 2525 2555 2617 2648 2415 2185
[cycle]
The theory of stress concentration factor presents qualitative explanation to the phenomenon that
the fatigue life is inversely proportional to the roughness. However, little attention has been paid to
the quantitative relationship between them. This research tried to find the character of this kind of
relationship through the fatigue experiment and the preliminary results were shown in Fig. 2. The
horizontal and vertical axes denote the roughness Ra and the fatigue life Nf, respectively. Ten group
data, displayed by the solid circle dots, scatter in the figure. A power function curve was used to fit
these data. As a result, the curve is expressed by Equation (3).
N f = 2559i Ra −0.1166 (3)
Fig. 2 Relationship between the roughness and the fatigue life
It can be seen that the curve well reflects the relationship between two variables. The result
provides some reference significance for engineering application, because the roughness of most
structures or components belongs to the range from 0.2µm to 1.6µm.
420 Advances in Fracture and Damage Mechanics XI
However, we should stress that this work is our preliminary research on the influence of surface
roughness on specimen’s low cycle fatigue life and some further works are needed to be done to
confirm or improve our preliminary judgments. The first one is the material. Equation (3) is derived
from the fatigue process of the material of alloy steel with yield strength 420MPa. Whether other
class steels or non-steels follow this rule? The second one is the roughness grade, just from 0.1µm
to 1.7µm, not covering wider range of roughness grades. The third one is the most important. The
material is subjected to a high level cyclic load, whose peak value exceeds the yield strength. The
material accumulates certain amount plastic deformation in the fatigue process and finally fractures
after thousands of load cycles. The power function curve implies a specific change trend of fatigue
life along with surface roughness in the low cycle fatigue region at room temperature. In our
experiment, only a stress amplitude was used. Whether the trend varies if the stress amplitude is
replaced by other level stresses but still keeping not beyond the low cycle fatigue region? We
imagine that the curve may be more abrupt for a higher level load amplitude and more gently for a
lower level load amplitude.
Conclusions
This research tried to investigate the effects of surface roughness on the low cycle fatigue life.
Experiments with different surface roughness specimens were carried out at room temperature.
Based on the concept of stress concentration factor, qualitative explanation has been presented
according to the experimental results. An approximate power function curve was fitted using the
experimental data. The results show that there is a certain quantitative relationship between the
surface roughness and the low cycle fatigue life. Another potential material that has being used in
high heat load components at the third generation synchrotron radiation facility will be studied in
our further research. And more roughness grades will be involved to further confirm our
conclusions.
Acknowledgment
This work was supported by the National Natural Science Foundation of China (10975130).
References
[1] A. Scott, S. Huseyin, In. J. Fatigue, Vol. 22 (2000), P. 619-630.
[2] K.A. Sigmund, S. Bjørn, In. J. Fatigue, Vol. 30 (2008), P. 2200-2209.
[3] M. Suraratchai, J. Limido, C. Mabru, In. J. Fatigue, Vol. 30 (2008), P. 2119-2126.
[4] M. Smaga and D. Eifler, J. Phys.: Conf. Ser. 240, (2010), IOP Publishing.
[5] D.C. Zhang, X.M. Pei, China Mech. Eng., Vol. 14 (2003), P. 1374-1377 (in Chinese).
