You are on page 1of 10

materials

Article
Fatigue Property of Open-Hole Steel Plates Influenced
by Bolted Clamp-up and Hole Fabrication Methods
Zhi-Yu Wang 1,2 , Lihui Li 1,2 , Yong-Jie Liu 1,2 and Qing-Yuan Wang 2,3, *
1 Department of Civil Engineering & Mechanics, Sichuan University, Chengdu 610065, China;
zywang@scu.edu.cn (Z.-Y.W.); 18200340772@163.com (L.L.); liuyongjie@scu.edu.cn (Y.-J.L.)
2 Sichuan Provincial Key Laboratory of Failure Mechanics and Engineering Disaster Prevention & Mitigation,
Sichuan University, Chengdu 610065, China
3 School of Architecture and Civil Engineering, Chengdu University, Chengdu 610106, China
* Correspondence: wangqy@scu.edu.cn; Tel.: +86-28-8540-6919

Academic Editor: Yoshiharu Mutoh


Received: 1 July 2016; Accepted: 8 August 2016; Published: 16 August 2016

Abstract: Steel plates with open holes are commonly used in structural assemblies. The fatigue
properties of such details are influenced by bolted clamp-up and hole fabrication methods.
The fracture surface, stiffness degradation and fatigue life of test specimens are investigated in
detail and compared with the contemporary test data. The analysis results show that the presence
of draglines greatly influences the fatigue crack initiation at the open-hole cut by laser. The bolted
clamp-up condition greatly enhances the stiffness and the fatigue life of the open-hole details.
A discussion is also made from a comparison with the referred fatigue life of hole fabrication details,
such as the influence of plate thickness and plasma cutting, drilling and oxy-fuel gas cutting, with the
details studied herein. This work could enhance the understanding of the fatigue property and design
of such details.

Keywords: open-hole; bolted clamp-up; fatigue property; life assessment

1. Introduction
Steel plates with open holes are commonly used in structural components to enable an assembly
with bolts. The gusset plates and cross-frames are components which can be regarded as examples of
such types. As evidenced from the basic configurations of these components, the weaknesses of the
components are determined by fabrication methods as well as clamping conditions [1,2], which become
an important issue for fatigue loading cases.
For the sake of evaluating the mechanical characteristics of the components with hole fabrication,
tensile tests have been widely used in obtaining the basic mechanical properties of the connections.
Punched holes as a fast and cost-effective method of forming holes are widely used for secondary
tension members. Also, holes may be formed by full-size drilling and thermal cutting, which a concern
of most early studies. From the tests, Chesson and Munse [3] concluded that punching reduced
the ductility of the net section which causes the early arrival of ultimate stress near the holes and
thus limits the effectiveness in strength development. Frank [4] showed that the strength ratio of
drilled-hole specimens was much lower than that of punched-hole specimens and pure punching
was recommended in secondary connection members. Later, Rassati et al. [5] confirmed this point
and observed a decrease in the tensile ductility for punched-hole specimens. In agreement with
all these findings in the literature, the AASHTO LRFD (American Association of State Highway
and Transportation Officials Load-and-resistance Factor Design) [6] restricts pure punched holes in
the primary loading carrying members and gives limits of material thickness in some other uses.
In addition, a considerable amount of recent research [7–11] was developed based on the open-hole

Materials 2016, 9, 698; doi:10.3390/ma9080698 www.mdpi.com/journal/materials


Materials 2016, 9, 698 2 of 10

tension and bolt-filled hole tension tests in accordance with ASTM (American Society of Testing
Materials) standards [12] to provide mechanical characteristics of composite laminates.
Although the weakness of punched holes may be improved using reamed full-size to remove
the damaged zone surrounding the hole, laser cutting as a substitute has shown good applications in
engineering [13,14]. The merits of laser cutting are obvious, including the exemption from physical
contact and mechanical force, low-cost fast processing and precise operation without changing tools.
Yilbas et al. [15] showed that the small holes made by laser cutting in aluminium foam are parallel sided
and free from defects such as sideways burning and large burr formation. Alegre et al. [16] concluded
from the fatigue tests that the punched-hole specimens have a much lower life span than replicated
drilled-hole ones and they have fatigue cracks initiated at the transition point. Sánchez et al. [17]
observed that the drilled specimens have twice the fatigue resistance of the punched specimens.
Brown et al. [18] confirmed the conclusion from [16,17] and showed that specimens with laser-cut
holes have better surface conditions but the same average strength when compared with punched-hole
specimens. Garcia et al. [19] and Cicero et al. [20] reported research comparing the fatigue performance
of drilled and punched holes with typical thermal cutting methods. Due to limited studies, however,
a full understanding of the fatigue property of open-hole steel plates when the modern fabrication
process and clamping conditions are concerned has not yet been developed.
In this paper, the fatigue properties of open-hole steel plates using laser cutting as a modern and
common fabrication method will be studied in terms of failure mode and life results. Also, the fatigue
life of the open-hole steel plates with the bolted washer clamp-up condition is also examined. Finally,
the effects of clamping and fabrication methods on the fatigue life of such structural details will be
discussed further. The results obtained in this study can be taken as a basic reference for further study
involved in the fatigue life extension of related structural details.

2. Material and Fatigue Test Procedures


The material used for the test specimens was Grade 345B low alloy steel which conforms to the
Chinese national standard GB/T1591. This material has good toughness and ductility. The chemical
composition and mechanical properties are listed in Table 1. Related information on steel materials
of S355N, S460Q and S460M in [17,19] is also given herein as a reference and for discussion in the
subsequent section.

Table 1. Chemical composition and mechanical properties of steel materials.

Chemical Composition (%) Mechanical Properties


Steel Grade
C Si Mn P S σy (MPa) Es (MPa) σu (MPa)
Q345B 0.17 0.25 1.15 0.015 0.014 388 2.1 × 105 553
S355N [17] 0.17 0.30 1.35 0.02 0.011 423 2.0 × 105 602
S460Q [17] 0.08 0.30 1.31 0.014 0.003 619 2.0 × 105 685
S460M [19] 0.12 0.45 1.49 0.012 0.001 484 2.05 × 105 594

The test specimens were designed to consider two configuration details, i.e., the open-hole details
and the bolted washer clamp-up details. All test specimens had similar dimensions, as shown in
Figure 1. The steel plate in tension was 6-mm thick. The plates were machined to their design
dimensions of 300 mm long and 30 mm wide prior to the holes being made. The hole at the centre
of the plate was cut using a standard laser cutter. During the cutting process, the metal workpiece
was completely penetrated by the laser. A stream of gas blew off the melted materials to form a hole
as the laser beam moved around the melting material. The metal laser-cutting machine of 2000 W
was employed in this cutting process. In this machine, the nozzle diameter of 1.6 mm was chosen to
ensure that the gas diffusion area and size can be properly controlled. Meanwhile, the laser cutting
was handled with the speed of 1.5 m/min to enable stable cutting. For the test specimens with bolted
Materials
Materials2016, 9, 9,
2016, 698698 3 of 1010
3 of

specimens with bolted washer clamp-up details, the square washer placed beneath the bolt head
washer
and clamp-up
nut in details, thethe
double-clamping square
steelwasher placed
plate was beneath
kept the thick
at 6 mm bolt head
and and nut in
30-mm double-clamping
wide. All washer
the steel plate was kept at 6 mm thick and 30-mm wide. All washer surfaces were flattened
surfaces were flattened without rust and with a bevelled edge to ensure a good contact condition. without
rustM12
The and high
with astrength
bevelledbolt
edgewith
to ensure a good contact
the tightened torquecondition.
of 79 N·m Thewas
M12chosen
high strength
for thebolt with
bolted
the tightened torque of 79 N · m was chosen for the bolted connection following the
connection following the Chinese code GB/T 16823.2-1997 [21] which gives the general rules of Chinese code
GB/T 16823.2-1997
tightening [21]fasteners.
for threaded which gives the general rules of tightening for threaded fasteners.

M12 bolt

12 30 30

300 300

6 6
Unit: mm

(a) (b)
Figure 1. 1.
Figure Geometry of typical
Geometry testtest
of typical specimens. (a) Open-hole
specimens. details;
(a) Open-hole (b) Bolted
details; washer
(b) Bolted clamp-up
washer details.
clamp-up details.

The test specimens were tested using the Shimadzu EHF-UV050k2-020-0A fatigue testing
The test specimens were tested using the Shimadzu EHF-UV050k2-020-0A fatigue testing machine
machine of 50 kN capacity. Output and instant information from the testing machine were
of 50 kN capacity. Output and instant information from the testing machine were monitored and
monitored and recorded by an automatic data acquisition system controlled by Windows Software
recorded by an automatic data acquisition system controlled by Windows Software for 4830 3.40b.
for 4830 3.40b. Tensile fatigue tests were conducted following the Chinese code GB/T 3075-2008 [22].
Tensile fatigue tests were conducted following the Chinese code GB/T 3075-2008 [22]. Constant
Constant amplitude sinusoidal stress cycles with the frequency of 8 Hz were conducted during the
amplitude sinusoidal stress cycles with the frequency of 8 Hz were conducted during the fatigue
fatigue test. Five target stress levels between 200 and 340 MPa were mainly applied. The stress ratio
test. Five target stress levels between 200 and 340 MPa were mainly applied. The stress ratio was set
was set at 0.1 for all tests. Fifteen specimens with open-hole details followed by six specimens with
at 0.1 for all tests. Fifteen specimens with open-hole details followed by six specimens with bolted
bolted washer clamp-ups were tested. The fatigue life was determined as the specimen was tested
washer clamp-ups were tested. The fatigue life was determined as the specimen was tested to rupture.
to rupture. The actual measured cross-sectional dimensions were used for the calculation of the
The actual measured cross-sectional dimensions were used for the calculation of the nominal stress of
nominal stress of the testing specimens.
the testing specimens.
3. Experimental Test Results
3. Experimental Test Results
3.1.
3.1.Fracture Surface
Fracture SurfaceObservation
Observation
The
Thefractures
fracturesofofthe thesteel
steelplate
plateexclusively
exclusivelyoriginated
originatedatatthe theedge
edgeofofthethehole.
hole.The
Thetypical
typicalcrack
crack
initiation
initiationthrough
throughthe thethickness
thickness ofofthe
theplate
platearound
aroundthe thehole
holeappeared
appearedtotobebeunsymmetrical
unsymmetricalononboth both
sides of the hole. The presented fracture surface was taken from the test
sides of the hole. The presented fracture surface was taken from the test specimens and was able specimens and was able to
demonstrate a typical observation similar to the others. For the test specimen
to demonstrate a typical observation similar to the others. For the test specimen without the bolted without the bolted
clamp-up
clamp-uploaded loaded under
under the
thestress
stressrange
rangeofof220220MPa,
MPa,forforexample,
example,the thetypical
typicalfracture
fracturesurface
surfaceand anditsits
magnification of approximately 30× under the scanning electron microscope
magnification of approximately 30× under the scanning electron microscope (SEM TM3000, Hitachi, (SEM TM3000, Hitachi,
Tokyo,
Tokyo,Japan)
Japan)are areshown
shownin inFigure
Figure 2a.
2a. On
On closer
closer examination,
examination, the the hole
hole experienced
experiencedcrackcrackinitiation
initiationin
inthe
the vicinity
vicinity of theof middle
the middle thickness
thickness of the
of the steel steel
plate. Theplate. The crackseemed
crack initiation initiation
to beseemed
moderately to be
tied
moderately tied to the amount of disturbance at the hole surface. On the right
to the amount of disturbance at the hole surface. On the right side of the fracture surface, a straight side of the fracture
surface,
primaryacrack straight primary
initiated crack initiated
at the middle at the
of the surface of themiddle of the
steel plate surface ofto the
perpendicular steel surface
the upper plate
perpendicular to the upper surface of the plate, while the on the left side,
of the plate, while the on the left side, some scattered cracks originated from the middle thickness some scattered cracks
originated
point and frompropagatedthe middle
deeplythickness pointInand
into the plate. propagated
addition, deeply
several tiny intocan
cracks the plate.
also In addition,
be identified along
several tiny cracks can also be identified along with the primary cracks in the
with the primary cracks in the vicinity of the cavities, as shown in Figure 2a. In contrast, the specimenvicinity of the cavities,
aswith
shownthe in Figure
bolted 2a. In contrast,
clamp-up seems tothe specimen
exhibit typicalwith
cracktheinitiation
bolted clamp-up
towards the seems
upperto edge
exhibitof typical
the hole,
crack initiation towards the upper edge of the hole, as shown in Figure
as shown in Figure 2b. This difference can be attributed to the preload of the bolt applied onto 2b. This difference can be the
attributed to the preload of the bolt applied onto the upper and lower faces
upper and lower faces of the bolt clearance hole. In this sense, the fatigue crack propagation under of the bolt clearance
hole. In this
tension stresssense, the likely
is more fatigue tocrack propagation
take place under
close to the tensionhole.
clearance stress is more likely to take place
close to the clearance hole.
Materials 2016, 9, 698 4 of 10
Materials 2016, 9, 698 4 of 10

Materials 2016, 9, 698 4 of 10

(a) (b)
Figure 2. Typical fracture surface of test specimens. (a) Without bolted clamp-up (Δσ = 220 MPa);
Figure 2. Typical fracture surface of test specimens. (a) Without bolted clamp-up (∆σ = 220 MPa);
(b) With bolted clamp-up
(a)(∆σ(Δσ = 300 MPa). (b)
(b) With bolted clamp-up = 300 MPa).
Figure 2.Degradation
3.2. Stiffness Typical fracture surface
Behaviour of Damage
and test specimens.
Progress(a) Without bolted clamp-up (Δσ = 220 MPa);
3.2. Stiffness Degradation
(b) With Behaviour
bolted clamp-up (Δσ =and
300 Damage
MPa). Progress
Stiffness is a well-defined engineering property, easily measured, and not involved in the
Stiffness
destruction
3.2. isof
Stiffness a the
well-defined
test specimen
Degradation engineering
Behaviour [23]. Duringproperty,
and Damage the fatigue
Progress easily measured,
loading, damage and not involved
accumulates with thein the
crack propagation,
destruction of the test which
specimenin turn progressively
[23]. During thereduces
fatiguethe stiffnessdamage
loading, of the test specimens. with
accumulates The the
Stiffness is a well-defined engineering property, easily measured, and
of not
the involved in the
crackstiffness degradation
propagation, which inbehaviour is influenced
turn progressively by thethe
reduces connection details
stiffness of the test specimens.
specimens. TheThe
stiffness
destruction of the
stiffness degradation test specimen [23].
characteristics During
resulting the fatigue loading,
from the fatigue damage
damage accumulates
can be supposed with the
tostiffness
be
degradation
crack behaviour
propagation, is
which influenced
in turn by the
progressivelyconnection
reduces details
the of
stiffness the
of specimens.
the test The
specimens. The
related with the damage propagation of the number of cycles. The stiffness at the ith cycle can be
degradation
stiffness characteristics resultingisfrom the fatigue damage can be supposed thetospecimens.
be related The
with the
obtained degradation behaviour
as the ratio between influenced
the measured by range
force the connection details
and the deformation of range, which can be
damage propagation
stiffness of the
degradation number of cycles.
characteristics resultingThefrom
stiffness
the at the ith
fatigue cycle can
damage can be
be obtained
supposed as
to the
be ratio
given by:
between thewith
related measured forcepropagation
the damage range and the deformation
of the number of range, which
cycles. The can beatgiven
stiffness the ithby:
cycle can be
, ,
obtained as the ratio between the measured = force range and , the deformation range, which can(1) be
, ,
given by: (σi,max − σi,min ) Am
where σi,max and σi,min are the maximum Ri =and minimum stress components , at the ith cycle; i,max and (1)
δi,max
,
−, δi,min
i,min are the deformation components corresponding = ,
to σi,max and σi,min, respectively, which were (1)
, ,
where measured
σi ,max andfrom the testing
σi,min are themachine
maximum andand while eliminating
minimum stressthe components
initial error of at deformation
the ith cycle;arising
δi ,max and
wherethe σi,max and σi,min areAthe maximum and minimumarea stress components at the ith cycle; i,max and
δi,minfrom
are the grip assembly;
deformation m is the plate net
components cross-section
corresponding to σexcepting the hole.
i ,max and σi,min , respectively, which were
i,min The
are comparison
the deformation components corresponding to σ i,max and σi,min, respectively, which were
measured
measuredfrom the thetestingofmachine
the stiffness
and degradation
while behaviour
eliminating of specimens
the initial error with open-holearising
errorofofdeformation
details
arising from
and boltedfrom washer testing
clamp-up machine
detailsand while
is shown eliminating
in Figure 3. the initial
It can be observed deformation
that the open-hole
the grip
from assembly; Am is the Amplate net cross-section area excepting the hole.
plate the grip assembly;
exhibits is thedegradation
notable stiffness plate net cross-section
when thearea excepting
nominal the hole.
loading cycle is greater than 0.85.
The comparison
The comparisonof the stiffness degradation behaviour ofofspecimens with
with open-hole details and
When approaching theofultimate
the stiffness
state,degradation
the stiffness behaviour
is decreased specimens
from approximately open-hole
65 kN/mmdetailsto
boltedandwasher
bolted clamp-up
washer details
clamp-up is shown
details in
is Figure
shown in3. It can
Figure be3. observed
It can
59 kN/mm. In contrast, the stiffness of the bolted clamp-up plate under similar loading force is be that the
observed open-hole
that the plate
open-hole exhibits
plate
notable
nearly exhibits
stiffness
76 kN/mm,notable
degradation
whichstiffness
17%degradation
iswhen thanwhen
the nominal
greater loading
that thethe
of nominal
cycle loading
is greater
open-hole cycle
plate.than is greater
0.85.
Meanwhile, thethan
When 0.85.
approaching
nominal
When
the ultimate approaching
state,for
loading cycle thethethe ultimate
stiffness state,
is decreased
initiation the
of the notable stiffness
from is decreased
approximately
stiffness degradation from
65iskN/mm approximately 65
to 59 kN/mm.
also enhanced kN/mm
to 0.9 and Inonly toa
contrast,
59 kN/mm.
the stiffness
2 kN/mm In contrast,
ofstiffness
the bolted the
clamp-up
reduction stiffness
until plateof the bolted
under loading
the ultimate clamp-up
similar case. loading plate
Thus, force under
it is is similar
nearly that
apparent loading
76 kN/mm, force is
the detailswhich
of is
nearly
the 76
bolted kN/mm,
clamp-up which is
condition17% greater than
significantly that
increase of the
the open-hole
initial plate.
stiffness Meanwhile,
and improve the
the nominal
stiffness
17% greater than that of the open-hole plate. Meanwhile, the nominal loading cycle for the initiation
loading cyclebehaviour.
for the initiation of the notable stiffness degradation is also enhanced to 0.9 and only a
of thedegradation
notable stiffness degradation is also enhanced to 0.9 and only a 2 kN/mm stiffness reduction
2 kN/mm stiffness reduction until the ultimate loading case. Thus, it is apparent that the details of
until the
thebolted
ultimate loading
clamp-up case.90Thus,
condition it is apparent
significantly increase that
Boltedthe the details
initial
washer clamp-up
of the
stiffness
details
bolted
and clamp-up
improve condition
the stiffness
significantly
degradation increase the initial80stiffness and improve the stiffness degradation behaviour.
behaviour. Open-hole details

70
90
Ri (kN/mm) Ri (kN/mm)

Bolted washer clamp-up details


60 80 Open-hole details
80
50 70
70
60
40 800.85
60 0.9 0.95 1
30 70
50
0.6 0.7 0.8 0.9 1
60 Ni/N
40
0.85 0.9 0.95 1
30
Figure 3. Comparison of stiffness degradation behaviour of specimens with open-hole details and
0.6 0.7
bolted washer clamp-up details (Δσ = 300 MPa). N0.8
/N
0.9 1
i

Figure 3. Comparison of stiffness degradation behaviour of specimens with open-hole details and
Figure 3. Comparison of stiffness degradation behaviour of specimens with open-hole details and
bolted washer clamp-up details (Δσ = 300 MPa).
bolted washer clamp-up details (∆σ = 300 MPa).
Materials 2016, 9, 698 5 of 10

3.3. Fatigue Life Results and Analysis


Materials 2016, 9, 698 5 of 10
For each class of structural components, the general equation can be expressed in relating the
3.3. Fatigue Life Results and Analysis
number of cycles to failure, N, and the applied stress range, ∆S, as:
For each class of structural components, the general equation can be expressed in relating the
N = stress
number of cycles to failure, N, and the applied )−m , ΔS, as:
C (∆Srange, (2)
= (∆ ) , (2)
where the exponent m is the slope of the S-N relation; C is the material constant-related parameter [23].
where
Taking the exponent
the logarithm m issides
on both the of
slope of the(2),
Equation S-Ntherelation; C is
following the material
equation can beconstant-related
written as:
parameter[23]. Taking the logarithm on both sides of Equation (2), the following equation can be
written as: log ( N ) = log (C ) − mlog (∆S), (3)
log( ) = log( ) − log(∆ ), (3)
Following the general fatigue design rule, the test data can be processed based on a statistical
Following the general fatigue design rule, the test data can be processed based on a statistical
analysis to provide the best fit mean S-N curve by the method of least squares. The fatigue detail
analysis to provide the best fit mean S-N curve by the method of least squares. The fatigue detail
categories of JSSC
categories (Japanese
of JSSC Society
(Japanese SocietyofofSteel
Steel Construction) [24]were
Construction) [24] were also
also added
added for for the purpose
the purpose of of
comparison. A graphical presentation in the form of the stress range versus the number of
comparison. A graphical presentation in the form of the stress range versus the number of cycles is cycles is
shown in Figure
shown 4 with
in Figure codified
4 with detail
codified detailclasses.
classes.

1000
10 3
: No failure
Stress range, Δσ (MPa)

1002
10

Test open-hole details Test clamp-up details


JSSC-A JSSC-B
JSSC-D JSSC-E
10
10
10000 100000 1000000 100000
104 10 5 10 6 10 7
Number of cycles, N
Figure 4. S-N relations of test specimens with open-hole details and bolted washer clamp-up details.
Figure 4. S-N relations of test specimens with open-hole details and bolted washer clamp-up details.

Using regression analysis, the corresponding S-N relations with the curve slope equal to 3.0 for
Using regression
test specimens analysis,
can be corresponding S-N relations with the curve slope equal to 3.0 for
theas:
expressed
test specimens can be expressed as:
(i) For the specimens with open-hole details:

(i) For the specimens with open-hole ( ) = 12.45 − 3 log(∆ ),


logdetails: (4)
(ii) For the specimens with bolted washer clamp-up details:
log ( N ) = 12.45 − 3log (∆S), (4)
log( ) = 13.36 − 3 log(∆ ) (5)
(ii) For the specimens with bolted washer clamp-up details:
The comparison of the test data with assigned fatigue detail categories by calculating the given
constant Log(C) of JSSC [24], AASHTO [6,25] and Eurocode 3 [26] is also listed in Table 2. It can be
log ( N ) = 13.36 − 3log (∆S), (5)
seen that the JSSC detail category E with a constant amplitude fatigue limit (CAFL) of 80, the
Eurocode 3 detail category 100 with a constant amplitude fatigue limit (CAFL) of 100, and the
The
AASHTO comparison of theB’test
detail category withdata with assigned
a constant amplitude fatigue detail
fatigue limit categories
(CAFL) of 82.7 by
can calculating
be referred the
giventoconstant
as a lowerLog(C)
bound ofin the evaluation
JSSC of the test
[24], AASHTO specimens
[6,25] with open-hole
and Eurocode 3 [26]details.
is alsoIn listed
contrast,
in the
Table 2.
consideration of the bolted washer clamp-up is effective in the improvement of
It can be seen that the JSSC detail category E with a constant amplitude fatigue limit (CAFL) of 80,the fatigue life
which can3be
the Eurocode referred
detail to by JSSC
category detaila category
100 with constantA,amplitude
AASHTO fatigue
detail category A and Eurocode
limit (CAFL) of 100, and 3 the
detail 160 with constant amplitude fatigue limits (CAFL) of 190, 165 and 160, respectively. For
AASHTO detail category B’ with a constant amplitude fatigue limit (CAFL) of 82.7 can be referred
illustration, the evidence of the significant enhancement of the fatigue life of test specimens using
to as a lower bound in the evaluation of the test specimens with open-hole details. In contrast,
the bolted clamp-up details can be observed in Figure 5.
the consideration of the bolted washer clamp-up is effective in the improvement of the fatigue life
Materials 2016, 9, 698 6 of 10

which can be referred to by JSSC detail category A, AASHTO detail category A and Eurocode 3 detail
160 with constant amplitude fatigue limits (CAFL) of 190, 165 and 160, respectively. For illustration,
the evidence of the significant enhancement of the fatigue life of test specimens using the bolted
Materials 2016, 9, 698 6 of 10
clamp-up details can be observed in Figure 5.

Table 2. Comparison of S-N relations with referred fatigue categories.


Table 2. Comparison of S-N relations with referred fatigue categories.
Categories/Relations Standard Deviation Log(C) CAFL (MPa)
Categories/Relations
JSSC-A Standard
- Deviation Log(C)
13.13 CAFL (MPa)
190
JSSC-B JSSC-A - - 13.13
12.87 190 155
JSSC-D JSSC-B - - 12.87
12.30 155 100
JSSC-D - 12.30 100
JSSC-E JSSC-E - - 12.01
12.01 80 80
AASHTO-A - 12.91 165
AASHTO-A - 12.91 165
AASHTO-BAASHTO-B - - 12.59
12.59 110 110
AASHTO-B’
AASHTO-B’ - - 12.30
12.30 82.7 82.7
EurocodeEurocode
3-160 3-160 - - 12.91
12.91 160 160
EurocodeEurocode
3-112 3-112 - - 12.45
12.45 112 112
EurocodeEurocode
3-100 3-100 - - 12.30
12.30 100 100
EquationEquation
(4) (4) 0.11 0.11 12.45
12.45 112.11112.11
EquationEquation
(5) (5) 0.06 0.06 13.36
13.36 225.22225.22

8
Open-hole details
Bolted washer clamp-up details
6
Log (N)

0
300MPa 315MPa 350MPa
Stress range

Figure 5.
Figure 5. Comparison
Comparison ofof fatigue
fatigue life
life results
results of
of specimens
specimens with
with open-hole
open-hole and bolted washer
and bolted washer
clamp-up details.
clamp-up details.

4. Discussion
4. Discussion
The fatigue
The fatiguebehaviour
behaviourofof thethe open-hole
open-hole steel
steel plates
plates differs
differs whenwhen the clamp-up
the clamp-up condition
condition and
and varied
varied
hole hole fabrication
fabrication methods methods are involved.
are involved. PunchingPunching uses standard
uses standard turret tooling
turret tooling to make toparts
makewhich
parts
which inevitably produce louvers, extrusions, tabs and some other shapes in
inevitably produce louvers, extrusions, tabs and some other shapes in addition to holes. As a result, addition to holes. As a
result, the quality problems of tool marks and scratches are more prone
the quality problems of tool marks and scratches are more prone to be present. Previous research to be present. Previous
research
work work
[17,18] [17,18]that
showed showed that punched
the failed the failedspecimen
punchedhad specimen had a non-symmetrical
a non-symmetrical fatigue crackfatigue
which
crack which initiated near the transition zone between the cut and
initiated near the transition zone between the cut and the tearing zones due to the punchingthe tearing zones dueprocess.
to the
punching
In contrast,process. In contrast,
laser cutting laser cutting
as a non-contact as aisnon-contact
process ideal for parts process is ideal
that will for parts
be nested since that will be
the features
nested since the features requiring a forming operation such as louvers or extrusions
requiring a forming operation such as louvers or extrusions are removed. Garcia et al. [19] indicated are removed.
Garcia
that theetfracture
al. [19]surface
indicated that
of the the fracture
laser-cut hole issurface
featuredofbythethelaser-cut hole isstarting
fatigue cracks featured by the
in the fatigue
cut surface,
cracks starting
favoured by thein the cut surface,
roughened surface.favoured by the aroughened
In this research, surface. In
similar observation canthis
be research,
obtained; ahowever,
similar
observation can be obtained; however, the pattern of the fatigue crack is varied, as
the pattern of the fatigue crack is varied, as shown in Figure 2, within the range of the plate thickness shown in Figure 2,
within the range of the plate thickness and around the edge of the hole. This can
and around the edge of the hole. This can be explained as the standard laser-cut hole being featured by be explained as the
standard laser-cut hole being featured by the presence of predominant draglines within the
thickness, which may trigger the fatigue cracks not only near the surface but also at the middle
thickness point.
For the sake of further discussing the influence of the clamp-up and hole fabrication methods
on the fatigue life of the open-hole plates, the fatigue data of test specimens and the referred
specimens of similar materials are summarized in Figure 6. Following the discussion in the
Materials 2016, 9, 698 7 of 10

the presence of predominant draglines within the thickness, which may trigger the fatigue cracks not
only near the surface but also at the middle thickness point.
For the sake of further discussing the influence of the clamp-up and hole fabrication methods on
the fatigue life of the open-hole plates, the fatigue data of test specimens and the referred specimens of
similar
Materials materials
2016, 9, 698 are summarized in Figure 6. Following the discussion in the foregoing paragraph, 7 of 10
Materials 2016, 9, 698 7 of 10
the fatigue life in the logarithm of the test results and the referred punched plates are compared in
Figure
3%~10% 7a.greater
It can be observed
fatigue life inthat
thethe specimens
logarithm thanwith
thoselaser-cut holes exhibit
with punched holes3%~10%
under the greater
stressfatigue
ranges
3%~10% greater fatigue life in the logarithm than those with punched holes under the stress ranges
life in the
of 185 MPa,logarithm
225 MPa thanandthose
250with
MPa.punched holes
Likewise, thisunder the stress
observation ranges
can be dueof 185 MPa,
to the 225 MPa
removal ofand
the
of 185 MPa, 225 MPa and 250 MPa. Likewise, this observation can be due to the removal of the
250 MPa.
contact Likewise,
problem this observation
induced by the punchedcan be due
hole to the
which inremoval of thethe
turn extends contact
fatigue problem induced
life to some by
extent.
contact problem induced by the punched hole which in turn extends the fatigue life to some extent.
the punched hole which in turn extends the fatigue life to some extent. Therefore,
Therefore, the better fatigue behaviour of laser-cut holes as compared to punched holes can be the better fatigue
Therefore, the better fatigue behaviour of laser-cut holes as compared to punched holes can be
behaviour
confirmed of andlaser-cut holes as
can be further comparedfor
considered toindustrial
punched application
holes can besince confirmed
the laserand can able
is also be further
to run
confirmed and can be further considered for industrial application since the laser is also able to run
considered
large irregular for industrial
cut-outs with application sincespeed.
much faster the laser
On isthealso able
other to run
hand, thelarge irregular
influence cut-outs
of plate with
thickness
large irregular cut-outs with much faster speed. On the other hand, the influence of plate thickness
much faster speed. On the other hand, the influence of plate thickness on the
on the fatigue life of the test specimens can be identified from a comparison with referred test data fatigue life of the test
on the fatigue life of the test specimens can be identified from a comparison with referred test data
specimens
[19]. As shown can be in identified
Figure 7b,fromit cana be
comparison with the
seen that with referred testof
increase data
the[19].
plateAs shown in
thickness Figure
from 6 to7b,
15
[19]. As shown in Figure 7b, it can be seen that with the increase of the plate thickness from 6 to 15
it can be seen that with the increase of the plate thickness from 6 to 15
mm, the test fatigue life in the logarithm of the test specimens is moderately enhanced by mm, the test fatigue life in the
mm, the test fatigue life in the logarithm of the test specimens is moderately enhanced by
logarithm
approximately of the6%.test specimens is moderately enhanced by approximately 6%.
approximately 6%.
1000
10 3
1000
10 3
: No failure
: No failure
Stress range, Δσ (MPa)
Stress range, Δσ (MPa)

100
10 2
100
10 2
Test open-hole details Test clamp-up details
Test open-hole
Oxy-fuel details (S460M)
gas cutting Test clamp-up
Plasma cuttingdetails
(S460M) [19]
Oxy-fuel gas cutting
Drilling (S355N) [17](S460M) Plasma cutting (S460M)
Drilling (S460Q) [17] [19]
Drilling (S355N)
Punching [17][17]
(S355N) Drilling (S460Q)
Punching (S460Q)[17][17]
Punching
JSSC-A (S355N) [17] Punching
JSSC-B (S460Q) [17]
JSSC-A
JSSC-D JSSC-B
JSSC-E
10 JSSC-D JSSC-E
1010
10 10000
10 4 100000
10 5 1000000
10 6 10000000
10 7
10000
10 4 100000
10 5 1000000
10 6 10000000
10 7
Number of cycles, N
Number of cycles, N
Figure 6.
Figure 6. Comparison
Comparisonofof fatigue
offatigue life
fatiguelife results
resultsofof
liferesults test specimens
oftest
test specimens with
with reference
reference using
using different
different hole
hole
Figure 6. Comparison specimens with reference using different hole
fabrication methods.
fabrication methods.
fabrication methods.
8 8
8 Test open-hole details 8
Test open-hole details Current test (6mm thick)
Punching (S355N) [17] Current test (6mm thick)
Punching (S355N) [17] Referred test (15mm thick) [19]
6 Punching (S460Q) [17] Referred test (15mm thick) [19]
6
6 Punching (S460Q) [17] 6
Log (N)

Log (N)
Log (N)

Log (N)

4 4
4 4

2 2
2 2

0 0
0 0
185MPa 225MPa 250MPa 185MPa 270MPa 348MPa
185MPa 225MPa 250MPa 185MPa 270MPa 348MPa
Stress range Stress range
Stress range Stress range
(a) (b)
(a) (b)
Figure 7. Comparison of fatigue life data influenced by details. (a) Influence of laser-cut and
Figure 7. Comparison of fatigue life data influenced by details. (a) Influence of laser-cut and
Figure
punched 7. Comparison of fatigue
holes; (b) Influence life data
of plate influenced by details. (a) Influence of laser-cut and punched
thickness.
punched holes; (b) Influence of plate thickness.
holes; (b) Influence of plate thickness.
To compare the test fatigue life results with the referred specimens using other fabrication
To compare the test fatigue life results with the referred specimens using other fabrication
methods, the detail fatigue rating method–based analysis was adopted. The fatigue detail
methods, the detail fatigue rating method–based analysis was adopted. The fatigue detail
coefficient can be taken as the measure of the fatigue quality of a structural component [27]. Related
coefficient can be taken as the measure of the fatigue quality of a structural component [27]. Related
coefficients resulting from this method indicate the inherent characteristics of the fatigue capacity of
coefficients resulting from this method indicate the inherent characteristics of the fatigue capacity of
a structure independent of the applied fatigue load. The characteristic value of the Weibull
a structure independent of the applied fatigue load. The characteristic value of the Weibull
distribution can be given as the relation between the number of cycles to failure, N, the number of
distribution can be given as the relation between the number of cycles to failure, N, the number of
test specimens, n, and the Weibull scale parameter, m, which is taken as 4.0:
Materials 2016, 9, 698 8 of 10

To compare the test fatigue life results with the referred specimens using other fabrication methods,
the detail fatigue rating method–based analysis was adopted. The fatigue detail coefficient can be
taken as the measure of the fatigue quality of a structural component [27]. Related coefficients resulting
from this method indicate the inherent characteristics of the fatigue capacity of a structure independent
of the applied fatigue load. The characteristic value of the Weibull distribution can be given as the
relation between the number of cycles to failure, N, the number of test specimens, n, and the Weibull
scale parameter, m, which is taken as 4.0:
 ( 1 )
1 n m

Materials 2016, 9, 698


β=
n ∑ i =1
Nim , (6)
8 of 10

Assuming
Assuming the
therequired reliability,RsR, sis
requiredreliability, , isdefined
definedasas
95%, thethe
95%, reliability level
reliability factor,
level SR ,SisR,defined
factor, as:
is defined
as:  1
1 (m)

SR = ln( ) ( ) , (7)
= ln(RS ) , (7)

The
The fatigue
fatigue life
life with
with aa 95%95% level
level ofof reliability
reliability and and aa 95%
95% confidence
confidence levellevel for
for structural
structural fatigue
fatigue
details
details is calculatedas: N
is calculatedas: N95/95
95/95 = β/(S
= β/(S RS
RSc). Scc ).isSthe
c isconfidence
the confidence coefficient
coefficient under under the confidence
the 95% 95% confidence level
level Given the basic stress range level N
[27]. Given the basic stress range level N0 at0 50,000, the referred fatigue detail coefficient can be
[27]. at 50,000, the referred fatigue detail coefficient can be
defined (1/m) . Thus, σ
defined as asσσ max[N
max [N95/95 /N
95/95/N 0 ] . Thus,
0](1/m) σmax max = 290
= 290 MPa MPa waswasconsidered
consideredfor for comparison.
comparison. Using
Using the the
above-mentioned
above-mentionedmethod, method,the thefatigue
fatiguelife life
results of theoftest
results thespecimens are compared
test specimens with the reference
are compared with the
regarding other fabrication methods as shown in Figure 8.
reference regarding other fabrication methods as shown in Figure 8. It is demonstratedIt is demonstrated that the fatigue detail
that the
coefficients
fatigue detailofcoefficients
the test open-hole
of the testspecimens
open-holeare 7%, 12%are
specimens and 7%,42%
12%lower
and 42%thanlower
that than
of thethat
referred
of the
data withdata
referred plasma
with cutting, drilling drilling
plasma cutting, and oxy-fuel gas cutting,
and oxy-fuel respectively,
gas cutting, which can
respectively, be can
which duebe to due
the
inherent shortcomings
to the inherent of the heat
shortcomings of affecting
the heat zone induced
affecting zonebyinduced
the laser bycutting. In fact,
the laser the fatigue
cutting. In fact,cracks
the
are morecracks
fatigue prone areto occur
morewhen pronethese heat affecting
to occur when these zonesheatare under a certain
affecting zonesmagnitude
are under ofa tension
certain
load. On theofother
magnitude tensionhand, the On
load. introduction
the other of bolt washer
hand, clamp-upof
the introduction details significantly
bolt washer covers
clamp-up such
details
shortages
significantlyby covers
enhancingsuchthe fatigueby
shortages detail
enhancingcoefficient by 43%,detail
the fatigue 39% coefficient
and 23% for bythe
43%,referred
39% and details
23%
with plasma cutting, drilling and oxy-fuel gas cutting, respectively. This
for the referred details with plasma cutting, drilling and oxy-fuel gas cutting, respectively. This cancan be attributed to the local
compression
be attributed applied
to the localby the bolt preload
compression appliedwhichbylessens
the boltthe surfacewhich
preload tension crackthe
lessens initiation
surfaceto some
tension
extent, and typical
crack initiation cracks
to some are more
extent, and likely
typicaltocracksappear arevery
more close to the
likely edge ofvery
to appear the bolt
closeclearance
to the edge hole.
of
In addition, the fatigue life of the structural components with laser-cut holes
the bolt clearance hole. In addition, the fatigue life of the structural components with laser-cut holes could be improved with
the rational
could configuration
be improved with the of bolt clamp-up
rational details. of bolt clamp-up details.
configuration

500
Referred fatigue detail coefficient

400

300

200

100

0
Test open-hole Test bolted Oxy-fuel gas Plasma cutting Drilling Drilling
washer clamp- cutting (S460M) [19] (S355N) [17] (S460Q) [17]
up (S460M) [19]

Figure 8. Fatigue
Figure 8. Fatigue detail
detail coefficient of details
coefficient of details with
with bolted
bolted clamp-up
clamp-up and
and varied
varied cutting
cutting methods.
methods.

5. Conclusions
The fatigue property of open-hole steel plates was investigated in this research with a focus on
the influences of the bolted clamp-up condition and hole fabrication methods. Through analyzing
the fracture surface, stiffness degradation and fatigue life results, the following conclusions can be
drawn:
Materials 2016, 9, 698 9 of 10

5. Conclusions
The fatigue property of open-hole steel plates was investigated in this research with a focus on
the influences of the bolted clamp-up condition and hole fabrication methods. Through analyzing the
fracture surface, stiffness degradation and fatigue life results, the following conclusions can be drawn:

• The fatigue crack of test specimens mostly initiated not only at the edge of the hole but also in
the vicinity of the middle thickness of the steel plate with the presence of predominant draglines
within the thickness for the laser-cut hole.
• The open-hole plate using laser cutting exhibits notable stiffness degradation in its appearing
cycle and magnitude. This trend seems to be reduced to some extent when the bolted clamp-up
details are introduced.
• The bolted clamp-up details greatly enhance the fatigue life of the open-hole plate from category
E to approaching category A as codified in JSSC, from category B’ to approaching category A as
codified in AASHTO, and from category 100 to approaching category 160 as codified in Eurocode.
• The fatigue life of open-hole details with laser cutting is greater than that with punched holes
but lower than those with plasma cutting, drilling and oxy-fuel gas cutting. The introduction
of bolted clamp-up details can be used to compensate for the laser cutting holes when a lower
fatigue detail coefficient is concerned.

Based on the basic understandings obtained from this study, further research is still needed
incorporating a wider range of parameters, such as plate thickness, etc., in a follow-up study.

Acknowledgments: The research presented was sponsored by the National Natural Science Foundation of
PR China (No. 51308363 and No. 11327801), the Scientific Research Foundation for the Returned Overseas
Chinese Scholars (No. 2013-1792-9-4), the Program for Changjiang Scholars and Innovative Research Team at the
University (No. IRT14R37) and the Science and Technology Support Programs of Sichuan Province (2015GZ0245
and 2015JPT0001).
Author Contributions: Zhi-Yu Wang conceived this fatigue experiment work and prepared this manuscript.
Lihui Li assisted in the fatigue experiment and prepared test samples. Yong-Jie Liu coordinated the laboratory
experimental work and took part in technical discussions. Qing-Yuan Wang supervised the projects and the
students with contributions to guide the research program.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Wang, Z.Y.; Zhang, N.; Wang, Q.Y. Tensile behaviour of open-hole and bolted steel plates reinforced by CFRP
strips. Compos. Part B 2016, 100, 101–113. [CrossRef]
2. Wang, Z.Y.; Wang, Q.Y. Fatigue strength strengthened welded joints with corrugated steel plates.
Compos. Part B 2015, 72, 30–39. [CrossRef]
3. Chesson, J.E.; Munse, W.H. Riveted and bolted joints: Truss-type tensile connections. J. Struct. Eng. 1963,
89, 67–107.
4. Frank, K.H. Influence of Hole Making Process upon the Tensile Strength of Steel Plates; TxDOT Research
Publications: Washington, USA, 2002; Volume 5, pp. 1–9.
5. Rassati, G.A.; Swanson, J.A.; Yuan, Q. Investigation of Hole Making Practices in the Fabrication of Structural Steel;
AISC: Cincinnati, OH, USA, 2004.
6. American Association of State Highway and Transportation Officials (AASHTO). AASHTO LRFD Bridge
Design Specifications, 3rd ed.; AASHTO: Washington, DC, USA, 2005.
7. O’Higgins, R.M.; McCarthy, M.A.; McCarthy, C.T. Comparison of open-hole tension characteristics of high
strength glass and carbon bre-reinforced composite materials. Compos. Sci. Technol. 2008, 68, 2770–2778.
[CrossRef]
8. Dano, M.L.; Kamal, E.; Gendron, G. Analysis of bolted joints in composite laminates: Strains and bearing
stiffness predictions. Compos. Struct. 2007, 79, 562–570. [CrossRef]
9. Hallett, S.R.; Green, B.G.; Jiang, W.G.; Wisnom, M.R. An experimental and numerical investigation into the
damage mechanisms in notched composites. Compos. Part A 2009, 40, 613–624. [CrossRef]
Materials 2016, 9, 698 10 of 10

10. Yun, J.H.; Choi, J.H.; Kweon, J.H. A study on the strength improvement of the multi-bolted joint.
Compos. Struct. 2014, 108, 409–416. [CrossRef]
11. Gamdani, F.; Boukhili, R.; Vadean, A. Tensile strength of open-hole, pin-loaded and multi-bolted single-lap
joints in woven composite plates. Mater. Des. 2015, 88, 702–712. [CrossRef]
12. ASTM-D5766/D5766M-11. Standard Test Method for Open-Hole Tensile Strength of Polymer Matrix Composite
Laminates; ASTM International: West Conshohocken, PA, USA, 2011.
13. Dubey, A.K.; Yadava, V. Laser beam machining-a review. Int. J. Mach. Tools Manuf. 2008, 48, 609–628.
[CrossRef]
14. Rangari, V.K.; Jeelani, M.I.; Zhou, Y.; Jeelani, S. Fabrication and characterization of MWCNT/thermoplastic
microsphere nanocom-posite foams. Int. J. Nanos 2008, 7, 161–169. [CrossRef]
15. Yilbas, B.S.; Akhtar, S.S.; Keles, O. Laser cutting of small diameter hole in aluminum foam. Int. J. Adv.
Manuf. Technol. 2015, 79, 101–111. [CrossRef]
16. Alegre, J.M.; Aragon, A.; Gutierrez-Solana, F.A. Finite element simulation methodology of the fatigue
behavior of punched and drilled plate components. Eng. Fail. Anal. 2004, 11, 737–750. [CrossRef]
17. Sánchez, L.; Gutierrez-Solana, F.; Pesquera, D. Fatigue behaviour of punched structural plates. Eng. Fail. Anal.
2004, 11, 751–764. [CrossRef]
18. Brown, J.D.; Lubitz, D.J.; Cekov, Y.C.; Frank, K.H.; Keating, P.B. Evaluation of Influence of Hole Making upon the
Performance of Structural Steel Plates and Connections; Report No. 0-4624-1; Center for Transportation Research,
The University of Texas at Austin: Austin, TX, USA, 2007.
19. Garcia, T.; Cicero, S.; Alvarez, J.A.; Martín-Meizoso, A.; Bannister, A.; Klimpel, A.; Aldazabal, A. Fatigue
performance of thermally cut bolt holes in structural steel S460M. Proc. Eng. 2015, 133, 590–602. [CrossRef]
20. Cicero, S.; Garcia, T.; Alvarez, J.A.; Martín-Meizoso, A.; Aldazabal, A.; Bannister, A.; Klimpel, A. Definition
and validation of Eurocode 3 FAT classes for structural steels containing oxy-fuel, plasma and laser cut holes.
Int. J. Fatigue 2016, 87, 50–58. [CrossRef]
21. Wang, Z.Y.; Wang, Q.Y. Yield and ultimate strengths determination of a blind bolted endplate connection to
square hollow section column. Eng. Struct. 2016, 111, 345–369. [CrossRef]
22. GB/T 3075-2008. Metallic Materials-Fatigue Testing-Axial Force Controlled Method; Standards Press of China:
Beijing, China, 2009.
23. Wang, Z.Y.; Wang, Q.Y.; Liu, Y.J. Evaluation of fatigue strength improvement by CFRP laminates and
shotpeening onto the tension flanges joining corrugated steel webs. Materials 2015, 8, 5348–5362. [CrossRef]
24. Jharkhand Staff Selection Commission (JSSC). JSSC Fatigue Design Recommendations for Steel Structures;
Gihodo Shuppan: Tokyo, Japan, 1993.
25. Wang, Z.Y.; Wang, Q.Y. Fatigue assessment of welds joining corrugated steel webs to flange plates. Eng. Struct.
2014, 73, 1–12. [CrossRef]
26. prEN 1993-1-8: 2005. Eurocode 3: Design of Steel Structures, Part 1-9; CEN: Brussels, Belgium, 2005.
27. Wirsching, P.H. Probabilistic fatigue analysis. In Probabilistic Structural Mechanics Handbook; Sundararajan, C.,
Ed.; Chapman and Hall: New York, NY, USA, 1995.

© 2016 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC-BY) license (http://creativecommons.org/licenses/by/4.0/).

You might also like