Engineer-

ing
Procedia Engineering 00 (2017) 1–8
Available online at www.sciencedirect.com
Procedia
Procedia
www.elsevier.com/locate/procedia
Engineer-
Engineer-
ScienceDirect
7th International Conference on Fatigue Design, Fatigue Design 2017, 29-30 November ing 2017,
Senlis,
Procedia Engineering France
00 (2017) 1–8 ing
Procedia Engineering 213 (2018) 262–269
Procedia Engineering 00 (2017) 1–8
www.elsevier.com/locate/procedia
www.elsevier.com/locate/procedia

Fatigue
7th test results
7th International
of surface
International Conference
Conference on
on Fatigue
hardened
Fatigue Design,
Design, Fatigue
components
Fatigue Design
Design 2017,
to evaluate
2017, 29-30
29-30 November
a2017,
two
November 2017,
layer approachSenlis,for strength
Senlis, France
France assessment
Fatigue test
test results
Fatigue Dénes
results
Dobberke of
ofa,∗surface
surface hardened
hardened
, Jens Wiebesiek a components
components
, Jürgen Fröschla , Martin to
to evaluate
Leitnerb a
evaluate a two
two
layer
layer
BMW Group, a approach
approach
Research,
for
for
New Technologies,
strength
strength
Innovations,
assessment
assessment
Parkring 17-19, Garching 85748, Germany
b Montanuniversität Leoben, Chair of Mechanical Engineering, Franz-Josef-Straße 18, 8700 Leoben, Austria

a,∗ a a b
Dénes
Dénes Dobberke
Dobberkea,∗,, Jens
Jens Wiebesiek
Wiebesieka ,, Jürgen
Jürgen Fröschl
Fröschla ,, Martin
Martin Leitner
Leitnerb
a BMW Group, Research, New Technologies, Innovations, Parkring 17-19, Garching 85748, Germany
a BMW Group, Research, New Technologies, Innovations, Parkring 17-19, Garching 85748, Germany
b Montanuniversität
b Montanuniversität Leoben, Chair of Mechanical Engineering, Franz-Josef-Straße 18, 8700 Leoben, Austria
Abstract Leoben, Chair of Mechanical Engineering, Franz-Josef-Straße 18, 8700 Leoben, Austria

In the automotive industry, forged steels are commonly used for powertrain components such as gears, axles or crankshafts. In
order to minimize wear and increase fatigue strength, these mechanical engineering parts are usually surface-hardened. The surface
heat treatment leads to a significant change of the local material properties in the heat effected zone of the surface area.
Abstract
This paper addresses an effective two-layer model for evaluating fatigue strength of surface hardened components based on local
Abstract
stresses. Hence, one layer represents the induction-hardened surface and the second characterizes the base material. The aim of
In
thisthe automotive
elaborated industry,
method forged
is a more steels computational
reliable are commonly used for powertrain components suchassessments
as gears, axles
basedor crankshafts. In
In the automotive industry, forged steels are commonlyestimation of fatigue life
used for powertrain among other
components such as gears, axles oron technological
crankshafts. In
order
benefit tofactors
minimize for wear and increase
components fatigue
designed for strength,
the high these
cycle mechanical
fatigue (HCF)engineering
regime byparts are into
taking usually surface-hardened.
account local material The surface
properties,
order to minimize wear and increase fatigue strength, these mechanical engineering parts are usually surface-hardened. The surface
heat treatment
defects leads stresses.
and residual to a significant change of the local material properties in the heat effected zone of the surface area.
heat treatment leads to a significant change of the local material properties in the heat effected zone of the surface area.
This paper
In order addresses
verify the an
to addresses effectivemethod
presented two-layer
andmodel for evaluating
to determine the localfatigue strength ofprocess-dependent
manufacturing surface hardened components based on local
This paper an effective two-layer model for evaluating fatigue strength of surface hardened fatigue strength,
components specimens
based on local
stresses.
are Hence,
extracted from one layerstressed
highly represents the induction-hardened
component areas considering surface
forged and thestructure.
grain second characterizes
The sample the base
notch material.
shape The typical
represents aim of
stresses. Hence, one layer represents the induction-hardened surface and the second characterizes the base material. The aim of
this
notch elaborated method is aengineering
types in mechanical more reliable computational
parts regarding form estimation of fatigue
factor, stress life among
gradient otherstressed
and highly assessments
volume.based
Theon technological
this elaborated method is a more reliable computational estimation of fatigue life among other assessments based onidea behind the
technological
benefit
fatigue factors
tests is toforstudy
components
materialdesigned for the higha comparably
samples exhibiting cycle fatigue (HCF) regimestress
by taking into account local material properties,
benefit factors for components designed for the high cycle fatigueminor (HCF) residual condition
regime by taking in both hardened
into account and unhardened
local material properties,
defects
condition and residual stresses.
defects andtoresidual
separatestresses.
the cause variables residual stresses, strength of base material and martensitic phase on fatigue strength.
In order to verify
Compared the presented method andthe to determineteststhe local manufacturing
stress ratios process-dependent fatigue strength, specimens
In order to to the unhardened
verify the presented base material,
method and to fatigue
determine at different
the local manufacturing revealed higher fatigue
process-dependent strength
fatigue within
strength, low and
specimens
are extracted
highextracted from
cycle fatigue highly stressed component
material. areas considering forged grain structure. The sample notch shape represents typical
areas considering forged grain structure. The sample notch shape represents
are from of the martensitic
highly stressed component typical
notch types inthese
Considering mechanical
input dataengineering parts regarding
in combination form factor,
with the fatigue stress
strength of gradient and highly
each material stressed
section, volume.
the layer The idea
approach behind
enables the
a more
notch types in mechanical engineering parts regarding form factor, stress gradient and highly stressed volume. The idea behind the
fatigue
reliable tests
local isfatigue
to study material samples
assessment among exhibiting
other fatiguea comparably
evaluation minor residual
methods based stress
on condition in
technological both factors.
benefit hardened and unhardened
fatigue tests is to study material samples exhibiting a comparably minor residual stress condition in both hardened and unhardened
condition to separate the cause variables residual stresses, strength of base material and martensitic phase on fatigue strength.
condition to separate the cause variables residual stresses, strength of base material and martensitic phase on fatigue strength.
Compared
c 2017 The toAuthors.
the unhardened basebymaterial,
Published Elsevierthe Ltd.fatigue tests at different stress ratios revealed higher fatigue strength within low and
Compared to the unhardened base material, the fatigue tests at different stress ratios revealed higher fatigue strength within low and
high cycle fatigue
Peer-review of the martensitic material.
high cycle fatigue of the martensitic material. committee of the 7th International Conference on Fatigue Design.
under responsibility of the scientific
Considering these input data in combination with the fatigue strength of each material section, the layer approach enables a more
Considering these input data in combination with the fatigue strength of each material section, the layer approach enables a more
reliable
Keywords: localfatigue
fatiguestrength,
assessment among other fatigue
surface-hardened evaluation
components, methods
residual based on technological benefit factors.
stresses
reliable local fatigue assessment among other fatigue evaluation methods based on technological benefit factors.
c 2018
© 2017 Published byby Elsevier Ltd.
c 2017 The
The Authors.
Authors. Published
Published by Elsevier

 Elsevier Ltd.
Peer-review under
Peer-review under responsibility
responsibility of
of the
the scientific
scientific committee
committee of
of the
the7th
7thInternational
InternationalConference
Conferenceon
onFatigue
FatigueDesign.
Design.
Peer-review under responsibility of the scientific committee of the 7th International Conference on Fatigue Design.
Keywords: fatigue strength, surface-hardened components, residual stresses
Keywords: fatigue strength, surface-hardened components, residual stresses

∗ Correspondingauthor. Tel.: +49-89-382-37414.

Email address: denes.dd.dobberke@bmw.de (Dénes Dobberke)

1877-7058 © 2018 The Authors.

∗ Corresponding
∗ Corresponding author.
Published by Elsevier Ltd.
Tel.: +49-89-382-37414.
Peer-review underauthor.
Email address:
Tel.: +49-89-382-37414.
responsibility of the scientific(Dénes
denes.dd.dobberke@bmw.de committee of the 7th International Conference on Fatigue Design.
Dobberke)
Email address: denes.dd.dobberke@bmw.de (Dénes Dobberke)
10.1016/j.proeng.2018.02.027
2 Dobberke et al. / Procedia Engineering 00 (2017) 1–8
Dénes Dobberke et al. / Procedia Engineering 213 (2018) 262–269 263
1. Introduction

Present automotive products within continuous development process are subjected to the claim of consistent vehicle
light weight design[1]. Hence the declaration of component safety has a high priority. In the automotive industry
highly-stressed structural parts such as gears, axles or shafts are usually surface-hardened with the aim of minimizing
wear and pitting[2] and increasing fatigue strength[3]. In comparison to weight-critical or package limited design
changes surface-hardening achieves efficiently an ascending fatigue strength[4]. Thermochemical surface-hardening
techniques[5, 6] causing further strength enhancement due to heat treatment are e.g. nitriding, case and induction
hardening[7]. Within the heat affected zone processes of microstructure transformation and development of compres-
sive residual stresses[8] occur. As depicted in Fig. 1 crack initiation beneath the surface layer according to different
mean stress sensitivies of martensitic structure and base material at stress ratio R = −1 is possible. Mentioned effects
regarding the different hardening processes are taken into account within fatigue design guidelines[9]. For ensuring
an appropriate computational fatigue design this paper evinces a two layer material approach considering different
material properties of the surface-hardened layer and the subjacent core material. A forged crankshaft with induction
hardened main and conrod bearing areas is considered. Latter areas possess a martensitic structure as a result of the
induction hardening process. Small-scale round specimens are extracted from crankshafts and uniaxial fatigue tests
for both layers incorporating varied specimen hardness are pursued. As base material a AFP steel is used, which is
common for forged and induction hardened components.

σ[MPa]

hardness zone base material

m martensite > m base material

load R = −1

local durability
considering σres
R = −1

possible stress based crack

initiation beneath surface

local durability
considering σres load R = 0
R=0

0 mm =ˆ surface 1 x[mm]
χ

Fig. 1. Mechanical load vs. mechanical durability of surface hardened components

Nomenclature

σres residual stresses [MPa]

α angle of notch [◦ ]
D,d diameter [mm]
r1 radius unnotched specimen [mm]
r2 radius notched specimen [mm]
L specimen length [mm]
R stress ratio [−]
k slope [−]
σa nominal stress [MPa]
σm mean stress [MPa]
N load cycle [−]
 Dobberke et al. / Procedia Engineering 00 (2017) 1–8 3

264 Dénes Dobberke et al. / Procedia Engineering 213 (2018) 262–269

2. Specimen preparation

During specimen preparation blanks are extracted from crankshaft considering forged grain structure, see Fig. 2. The
partially detached cuboidal blocks represent base material of forged crankshaft steel. Half of the blanks are subse-
quently transformed into martensitic microstructure via oven hardening. The obtained crankshaft blanks represent on
one hand the induction hardened surface layer and on the other hand the forged base material. Latterly the notched
and unnotched specimen geometry is manufactured out of present cuboidal blanks. The final shape of the notched
specimen is approximated to crankshaft fillet regarding form, stress gradient an highly-stressed volume, see Fig. 3.
The idea behind the presented layer approach is to study specimens in untreated and hardened condition, which exhibit
a comparably minor residual stress condition, in order to separate cause variables like residual stresses, strength of
base material and martensitic phase on fatigue strength.

sampling of forged raw part specimen

quenching and tempering

specimen

Fig. 2. Specimen preparation

L L
r1

r2
∅D

∅D
∅d

∅d

Fig. 3. Unnotched (left) and notched (right) specimen geometry for fatigue tests

Fig. 4 outlines via sectioning the hardened main and conrod bearing areas. The foundation for hardness profile in the
depth direction from hardened surface to base material illustrates Fig. 5. The results are normalized to the maximum
degree of hardness of the surface layer and hardness path length. Hardness measurements of the hardened specimens
reveal an equal level in comparison to the crankshaft bearing area, see Fig. 6. Metallographic investigations attest
the resulting martensitic microstructure of the fully hardened specimens. For the sake of completeness the hardening
profile of the unhardened specimen is mentioned, see Fig. 7. Fig. 8 illustrates the to yield strength normalized axial
residual stress depth profile of a crankshaft bearing. Final residual stress measurements of prepared specimens reveal
a low residual stress state for unhardened specimens. The compressional residual stresses for hardened specimens are
inferior in comparison to the maximum crankshaft condition.
4 Dénes Dobberke
Dobberke et al. / Procedia
et al. / Procedia Engineering
Engineering 2131–8
00 (2017) (2018) 262–269 265

1.2
1.0

hardness [−]
0.8
martensitic structure
base material
0.6
0.4
hardness hardness
zone zone 0.2
0.0
0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
path [−]

Fig. 4. Crankshaft section Fig. 5. Normalized hardness profile of crankshaft bearing in depth direction

1.2 1.2
1.0 1.0
hardness [−]

hardness [−]
0.8 0.8
0.6 0.6
0.4 0.4
0.2 0.2
0.0 0.0
0.2 0.4 0.6 0.8 1.0 0.2 0.4 0.6 0.8 1.0
path [−] path [−]

Fig. 6. Normalized hardness profile of hardened specimen Fig. 7. Normalized hardness profile of unhardened specimen

0.4
0.2
axial residual stress [−]

0.0
−0.2
−0.4
−0.6
−0.8
−1.0
−1.2
−1.4
0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
path [−]

Fig. 8. Normalized residual stress depth profile of crankshaft bearing

266 Dénes Dobberke et al. / Procedia Engineering 213 (2018) 262–269
Dobberke et al. / Procedia Engineering 00 (2017) 1–8 5

3. Fatigue Tests

Table 1. Summary of fatigue test scheme

tension-compression bending torsion axial + torsion
stress ratio R=0 R = −∞ R = −1 R = −1 R = −1 R=0 R = −1 R = −1 R = −1
phase shift ϕ = 0◦ ϕ = 90◦
temperatur
20 ◦ C 20◦ C 20◦ C 150◦ C 20◦ C 20◦ C 20◦ C 20◦ C 20◦ C
specimen

unnotchted, untreated, Rz1    ♦ ♦ ♦ ♦  

unnotched, hardened, Rz1 ♦    ♦ ♦ ♦  
unnotched, hardened, Rz16 ♦ ♦ ♦
notched, untreated, Rz1    ♦ ♦ ♦  
notched, hardened, Rz1 ♦  ♦  ♦ ♦ ♦  

For the characterization of the aforementioned material conditions uniaxial and multiaxial fatigue tests are performed.
Tab. 1 lists the completed (), current (♦) and future () experiments. Cyclic tension-compression tests at stress
ratio of R = −1 are executed for comparative purpose. Fig. 9 and 10 depicts the normalized results to the minimal
run-out load level (nominal stress) of the notched unhardened specimen. The results reveal the fatigue strength of the
hardened specimens is expactedly greater than the run-out load level of untreated base material. The comparison of
run-out load levels leads to a normalized value of 1.86 for the unhardened specimen without notch geometry. The
hardened counterpart reaches 3.67 for the run-out load level, see Tab. 2. Resulting cyclic tests at R = 0, see Fig. 11
lead to the Haigh-diagram for the unhardened base material according to Fig. 12. The fatigue tests demonstrate that
the heat treated surface area possesses a generally increased fatigue strength compared to the untreated base material.
The knowledge of fatigue strength of the two material layers enables the consideration of different material properties
within computational fatigue design for surface-hardened components.

10 10

k = 15.1
1.16

10 %
50 %
90 %
1.25

k = 13.8
1.17

10 %
σa [−]

σa [−]

50 %
1.23

10 %
50 % 90 %
90 %
k = 9.8

10 %

1 k = 4.5
50 %
90 %
1

R = −1 R = −1 R = −1 R = −1
failure failure failure failure
retested runout retested runout retested runout retested runout
runout runout runout runout

104 105 106 107 104 105 106 107

N [−] N [−]

Fig. 9. Normalized fatigue test data of unhardened specimen Fig. 10. Normalized fatigue test data of hardened specimen
6 Dénes Dobberke
Dobberke et al. / Procedia
et al. / Procedia Engineering
Engineering 2131–8
00 (2017) (2018) 262–269 267

Table 2. Normalized resulting run-out load levels of hardened specimens and base material
specimen normalized unhardened run-out load level normalized hardened run-out load level
[−] [−]
1.86 3.67
1.00 2.06

10 5.0

ult
im
ate
4.0

ten
sile
stre
ngt
k = 13.8
R = −∞ R=0

h
3.0
1.17
σa [−]

σa [−]
10 %
1.08

50 %
90 %
10 %
50 %
90 %

k = 10.8
2.0 m=
1 0.2
9

R = −1 R=0 1.0
failure failure
retested runout retested runout
runout runout

104 105 106 107 −5.0 −2.5 0.0 2.5 5.0

N [−] σm [−]

Fig. 11. Normalized fatigue test data of unhardened specimens Fig. 12. Normalized haigh-diagram of unnotched, unhardened specimen

4. Computational Fatigue Design

The computational fatigue design is based on stress gradient approach according to [10, 11]. Input data are static
and cyclic strength, which are investigated based on small-scale round specimens according to section 3. By usage
of the specimen/component diameter and the local stress gradient of the highly-stressed component notch the local
component Woehler-curve can be assessed, which acts as basis for the fatigue design. To include the hardenend
surface layer the experimentally evaluated specimen Woehler-curve for the hardened material is included. The result
is a number of local Woehler-curves of the component which will serve damage accumulation and component safety
estimation, see Fig. 13. The advantages of the presented two-layer model can be summarized as follows:

• Separation of surface-hardened and subjacent core material, which improves the accuracy of the fatigue assess-
ment

• Consideration and mapping of component residual stress state based on simulation

• Individual local stress-based fatigue design of both layers considering microstructural properties, load mean
stress and residual stress condition as well as notch sensitivity
• Presented method acts as engineering-feasible fatigue design concept, which can be easily applied also for other
material types and surface-hardening processes
Prospective work deals with the validation of the presented two-layer model with component fatigue data and the set-
up of a feasible simulation technique to estimate the local residual stress condition of surface-hardened components,
which acts as input data for the two-layer fatigue design approach.
268 Dénes Dobberke
Dobberke et al. / Procedia
et al. / Procedia Engineering
Engineering 00 (2017)213
1–8(2018) 262–269 7

fatigue tests
specimen base & martensitic material state

input data:
 fatigue strength (tension-compression,
 material parameter
bending, torsion)
(yield strength, poisson’s ratio)  mean stress sensitivity
 geometry  notch sensitivity
 microstructure characteristics  multiaxial fatigue behaviour
 surface roughness

specimen SN curves
base & martensitic material state

simulation: measurement:
 hardening profile  hardening profile
 residual stress profile  residual stress profile

local fatigue strength

base & martensitic material state

local Haigh-diagram

simulation: mechanical loads:

 residual stresses  stress collective
 stress ratio for constant
amplitude loading

600

local SN curves
500
slope

400 local SN curve

σa [MPa]

knee point

300
run-out load level
specimen SN curve

damage calculation 200

104 105 106 107 108

damage accumulation N [−]

stress hypothesis (multiaxial)

component safety & damage

Fig. 13. scheme for computational fatigue design according to [10, 11]

5. Conclusion
This paper introduces an approach for the consideration of local different fatigue properties induced by surface harden-
ing processes. To characterize the fatigue behaviour of the hardened surface layer reliable small-scale round specimens
are extracted from automotive steel crankshafts and tempered to an equal hardening level as the bearing areas. Cyclic
fatigue tests at a load stress ratio R = −1 revealed an increased fatigue strength of the martensitic material com-
pared to the untreated base material. Based on the experimental results, the methodology of the two-layer approach is
presented, which contributes to an improved fatigue design of surface-hardened components.
Further investigations will focus on cyclic fatigue tests at different stress ratios as well as additional loading types and
elevated temperature effects. The aim is the characterization of mean stress sensitivity of the hardenend and untreated
carboned steel. Cyclic testing under bending, torsion and combined loads will complete the scope of experiments and
enable the application of the presented two-layer approach. A comprehensive validation of the concept acts as major
task for the subsequent research work.
 Dénes Dobberke et al. / Procedia Engineering 213 (2018) 262–269 269
8 Dobberke et al. / Procedia Engineering 00 (2017) 1–8

References
[1] S. B. Muttana, S. Mubashir, Design strategy for light weight vehicles, S. Auto Tech Review Vo. 1 (2012) 20-27.
[2] L. Xie, D. Palmer, F. Otto, Z. Wang, J. Wang, Effect of surface hardening technique and case depth on rolling contact fatigue behavior of alloy
steels, Tribology Transactions Vol 58 (2015) 215-224.
[3] V. F. Pavlov, V. A. Kirpichev, V. S. Vakulyuk, V. P. Sazanov, A. S. Bukatyi, Assessment of the influence of surface hardening on fatigue limit
by residual stresses, Strength of Material Col. 45 (2014) 649-653.
[4] K. H. Kloos, P. K. Braisch, Über die Wirkung einer Randschichtverfestigung auf die Schwingfestigkeit von Proben und Bauteilen, Härterei
Technische Mitteilungen 37 (1982), 83-92.
[5] E. Haibach, Betriebsfestigkeit - Verfahren und Daten zur Bauteilberechnung, 2nd edition, Springer Verlag (2006).
[6] D. Radaj, M. Vormwald, Ermüdungsfestigkeit - Grundlagen für Ingenieure, 3rd edition, Springer Verlag (2007).
[7] A. Vieweg, G. Ressel, P. Prevedel, P. Raninger, M. Panzenböck, S. Marsoner, R. Ebner, Induction hardening: Differences to a conventional
heat treatment process and optimization of its parameters, International Conference on Materials, Processing and Product Engineering (2015).
[8] G. V. B. Lemos, T. K. Hirsch, A. d. S. Rocha, R. M. Nunes, Residual stress analysis of drive shafts after induction hardening, Materials
Research Vol. 17 (2014) 70-74.
[9] R. Renner, E. Kullig, M. Vormwald, A. Esderts, D. Siegele, FKM-Richtlinie: Rechnerischer Festigkeitsnachweis für Maschinenbauteile aus
Stahl, Eisenguß- und Aluminiumwerkstoffen, 6th edition, VDMA Verlag (2012).
[10] W. Eichlseder, Rechnerische Lebensdaueranalyse von Nutzfahrzeugkomponenten mit der FE-Methode, TU Graz (1989).
[11] W. Eichlseder, Fatigue analysis by local stress concept based on finite element results, J. Computers & Structures Vol. 80 (2002) 2109-2113.