-
ORIGINAL ARTICLE
Numerical and experimental investigations of extensibledie
clinching
Xiaocong He & Fulong Liu & Baoying Xing &Huiyan Yang
& Yuqi Wang & Fengshou Gu & Andrew Ball
Received: 19 December 2013 /Accepted: 16 June 2014#
Springer-Verlag London 2014
Abstract With an increasing application of clinching in
differ-ent industrial fields, the demand for knowledge of static
anddynamic characteristics of clinching is increased. In the
presentwork, the extensible die clinching process was
numericallyinvestigated using finite element method. To validate
the com-putational simulation of the extensible die clinching
process,experimental tests on extensible die clinched specimens
havebeen carried out. Good agreement is achieved between
thepredictions and the experimental results. Monotonic tensile
testswere carried out to measure the ultimate tensile strengths of
theextensible die clinching joints and clinching-bonded
hybridjoints. Deformation and failure of the extensible die
clinchedjoints undermonotonic tensile loadingwere studied. The
normalhypothesis tests were performed to examine the rationality of
thetest data. This work was also aimed at evaluating
experimentallyand comparing the strength and energy absorption of
the exten-sible die clinched joints and clinching-bonded hybrid
joints.
Keywords Extensible die clinching . Process simulation .
Finite element method . Load-bearing capacity .
Energyabsorption
1 Introduction
Some relative new joining techniques have drawn more atten-tion
in recent years because they can join advanced sheet
materials that are dissimilar, coated and hard to weld
withconventional spot welding [15].
Many efforts have also been spent to develop hybrid join-ing
techniques and alternatives for application into light-weight
structures. Mucha et als paper [6] presented thepressed joint
technology using forming process with or with-out additional
fastener. The capabilities for increasing theload-carrying ability
of mechanical joints by applying specialrivets and dies were
presented. The joint forming was per-formed with the solid round
die and rectangular split die forriveted joint forming. The effect
of joint forming process onjointed material strain was compared by
measuring the micro-hardness of the joints. Mucha andWitkowski [7]
analyzed theshearing strength of double joints made of various
joiningtechniques. The capabilities of S350 GD sheet metal
joiningusing the ClinchRivet technique were presented. The
resultsachieved for joints arranged in parallel and perpendicular
tothe load for specified joint spacing were discussed. The
as-sessment of joint effectiveness was performed for both
ho-mogenous double joints and for various combinations of
thesejoints. Neugebauer et al.s paper [8] showed the advantages
ofthe two-piece dies especially in solid punch riveting of
differ-ent materials with distinct differences in strength. The use
ofthese dies effects convenient technological conditions and
anextended range of application for solid punch riveting.
The use of clinching is of interest to different industriessuch
as aerospace, automotive, packaging and domestic ap-pliance. This,
together with increasing use of light-weightmaterials, has produced
a significant increase in the useof clinching in light-weight
structures in recent years[912].
In industrial applications of the clinched structures,
knowl-edge of the mechanical characteristics of clinched joints
isvery important. The static and dynamic behaviour of
clinchedjoints has been the subject of a great amount of numerical
andexperimental studies. Previous publications mostly focussed
X. He (*) : F. Liu : B. Xing :H. Yang :Y. WangInnovative
Manufacturing Research Centre, Kunming University ofScience and
Technology, Kunming 650093,Peoples Republic of Chinae-mail:
[email protected]
F. Gu :A. BallCentre for Efficiency and Performance Engineering,
University ofHuddersfield, Queensgate, Huddersfield, HD1 3DH,
UK
Int J Adv Manuf TechnolDOI 10.1007/s00170-014-6078-y
-
on clinching with fixed grooved dies. An investigation
onclinching mechanism has been conducted by Gao and Budde[13]. Some
elementary terms were used to establish a basictheory for analyzing
the clinching mechanism. The influenceof the clinching process
parameters on the join-ability of high-strength steel was studied
by Mucha [14] using finite element(FE) method. The results showed
that some parameters, suchas die radius, die depth and die groove
shape were mainlyaffected on the join-ability. Markowski et al.
[15] presentedthe results of FE analysis for clinching joint
machines C-frame. Several versions of frame geometry were accounted
forwhen analyzing the straining of material, including the
massreduction. The purpose of this FE simulation was to
determinethe effect of mass reducing material recess on the
structurerigidity. The suitability and economics of clinching
processeswere studied by Varis [16, 17].
A dieless clinching process has been proposed byNeugebauer et
al. [18]. Using the dieless clinching, it is
possible to produce a one-sided flat connection, which is
notproducible with any other joining technology. Addition-ally, it
is possible to enlarge the application potential ofmechanical
joining technologies as for example semi-finished parts made of
magnesium can be partially heat-ed and directly joined without an
increase in processtime or a reduction in the process stability.
The tools costs,the necessary tolerances and the tool wear are
significantlyreduced.
Another clinching configuration has been developedinvolving an
extensible die for improving the mechani-cal behaviour of clinched
joints. The use of extensibledie clinching has increased in recent
years. But a liter-ature survey on the extensible die clinching has
showna very limited number of publications. In Zheng et al.spaper
[19], the extensible die clinching process has beensimulated by FE
method. The material flowing patternshave been compared between the
fixed grooved die clinchingand the extensible die clinching. The
influence of processparameters in extensible die clinching has been
systematicallyinvestigated by Lambiase and colleagues [20, 21]. The
exten-sible die clinched joints were produced under
differentforming loads for evaluating the evolution of the joints
profileexperimentally.
In the present study, the extensible die clinching processhas
been computationally studied using FE analysis software.A two
dimensional (2D) axisymmetric model was generatedbased on the
Cowper-Symonds material models. An implicittechnique with Lagrange
method and r-self-adaptivity wasused. To validate the computational
simulation of the exten-sible die clinching process, experimental
tests on specimens
Upper sheet
Lower sheet
20
20110
110
20
Fig. 1 A single lap clinched joint
(a) Fixed die clinching tools (b) Bottom view of fixed die
clinched joint
(c) Extensible die clinching tools (d) Bottom view of
extensible
die clinched joint
Fig. 2 Comparison of tools andbottom views between fixed
dieclinching and extensible dieclinching
Int J Adv Manuf Technol
-
made of aluminium alloy Al5754 were carried out. The struc-tural
analysis has also been performed for comparing the
strength and energy absorption ability of the extensibledie
clinched joints and clinching-bonded joints.
(a) Extensible die clinching Machine
Sliding sectorsDie anvil
Fixed die
Rubber spring
(b) Schemac of extensible die
(c) Geometrical dimensions (d) FE model
Punch Blank holder
Upper sheet
Lower sheet
Rubber spring
Sliding sectors
Fixed die
(e) Radial displacement of sliding sector in FE simulaon
s=0 mm s=0 mm s=0.2 mm s=0.8 mm
Fig. 3 FE simulation ofextensible die clinching process
Int J Adv Manuf Technol
-
2 Extensible die clinching process simulation
The single lap clinched joint comprises an upper sheet,
lowersheet as shown in Fig. 1. The sheet materials tested
wereAl5754 aluminium alloy plates of dimensions 110 mmlength20 mm
width2 mm thickness and were clinched inthe central part of lap
section. The mechanical properties ofthe aluminium alloy Al5754
were as follow: Youngs modu-lus, E=70 GPa; Poissons ratio,
v=0.33.
Comparison of tools and bottom views between fixed dieclinching
and extensible die clinching is shown in Fig. 2. Inthe fixed die
clinching process, the interlock is produced bydriving the material
towards the die groove. The extensible dieis composed of a series
of sliding sectors. In the extensible dieclinching, material is
spread radially rather than towards thedie groove, resulting in a
better interlock than in the fixed dieclinching process. The
extensible die clinched joints are char-acterized by different
geometrical and mechanical propertiesas compared with fixed die
clinched joints. In order to achievedesigned durability, the punch,
blank holder, sliding sectorsand fixed die were made of
high-strength steel materials. Therubber spring must be replaced at
regular intervals.Figure 3a, b show extensible die clinching
machine and sche-matic of extensible die. Figure 3c shows the basic
geometry ofthe extensible die clinching model.
A 2D axisymmetric extensible die clinching process modelwas
generated, as shown in Fig. 3d, using the commercial FEsoftware
LS-Dyna. The model was meshed using the planeelement 2D Solid162,
involving 5,427 elements with 5,905nodes in the model. The
extensible die clinching process
Fig. 4 Cross-section comparison between simulations and tests of
ex-tensible die clinching processes
(a) Monotonic tensile process of the clinched joints
(b) Monotonic tensile process of the clinch
(c) Failure mode of clinched and clinch-bonded hybrid joints
-bonded hybrid joints
Fig. 5 Monotonic tensile processand failure mode of the
clinchedand clinch-bonded hybrid joints
Int J Adv Manuf Technol
-
involves a large deformation with high local plastic strains
insheets, resulting in severe local mesh distortions. The
ALEadaptive technique in ANSYS/LS-DYNA was used. ASS2Dsingle
contact function was conducted to judge the contactsbetween the
surfaces.
The punch, blank holder and die were modelled as rigidbodies,
whilst the sheets were modelled as elasto-plastic ma-terials. The
piecewise-linear plasticity material model whichadopts the
Cowper-Symbols model to consider the influenceof strain rate was
used. The relationship between the Cowper-Symbols model and yield
stress is shown in the followingequation:
y 1 t
C
!1p
24
35 0 f Peff 1
where 0 is the yield stress in constant strain rate, t is
theeffective strain rate and C and P are the parameters of
strainrate; f(eff
P ) is the hardening coefficient which is based on theeffective
plastic strain. Mooney-Rivlin elastic rubber modelwas used for the
rubber spring.
Some criteria such as the von Mises yield criterion,
thepiecewise-linear isotropic strain-hardening rule and the
asso-ciated flow rule were adopted in simulations. The
frictionbetween different parts in the model has an effect on
the
profile of the extensible die clinched joints. In the lack
ofexperimental data, tentative values of the Coulomb
frictioncoefficient between different parts in the extensible
dieclinching process model were assumed as follows:
f=0.25punch-upper sheet, f=0.15 upper sheet-blank holder, f=0.15
upper sheet-lower sheet and f=0.25 lower sheet-die. These values
were kept constant for the simulationsin this study.
To save simulation time, start the analysis at the momentwhen
the punch was very close to the top surface of the uppersheet and
apply a specified initial velocity to simulate theextensible die
clinching process. The extensible die clinchingprocess is modelled
by applying a downward initial velocityto every node within the
punch. Figure 3e shows the radialdisplacement of sliding sector in
the extensible die clinchingprocess FE simulation.
3 Extensible die clinching process tests
A clinching equipment RIVCLINCH 1106 P50 system wasemployed as
clinching machine as shown in Fig. 3a. Allclinching joints were
made with constant pre-clamp (4 kN)and setting load (50 kN). As
shown in Fig. 3c, the diameter ofthe punch is 5 mm and all
clinching joints were formed for the
Fig. 6 Force-displacement curves and tensile strengths normal
probability density distributions of clinched joints and
clinch-bonded hybrid joints
Int J Adv Manuf Technol
-
same depth sensor. The average value of the bottom thicknessis
1.4 mm. The cross-section comparison between simulationsand tests
of extensible die clinching processes is shown inFig. 4. It is
clear that the result obtained from tests agree fairlywell with the
computational simulation. The results show thecapability of the FE
model for simulating the extensible dieclinching process for
different geometries and workconditions.
4 Deformation and failure of clinched joints
Clinching has found applications in heavy-duty situation, suchas
car bodies. Load-bearing capacity and energy absorption(EA) are the
two most important features in structural analysisof clinched
joints. During the clinching process, the uppersheet undergoes a
significant thinning near the punch cornerradius. The strength of
an extensible die clinched joint de-pends on the joint profile and
particularly on the neck thick-ness and the magnitude of the
produced undercut.
In order to improve the mechanical properties of theclinched
joints, it is also important for clinching to benefitfrom the
advantages of other fastening techniques, for exam-ple adhesive
bonding [22]. Adhesives are used to increase therigidity and
tightness of the structure [23, 24]. It is commonlyunderstood that
the addition of adhesive in clinched joints isbeneficial but it is
not clear if there are negative effects onmechanical properties of
clinched joints. Deformation andfailure of homogeneous clinched
joints under tensile loadingwere investigated for validating the
load-bearing capacity andEA of clinched joints and clinch-bonded
hybrid joints.
The clinch-bonded hybrid joints were produced followingexactly
the same procedure as the respective clinched joints.The adhesive
used in the present study was two componentsacryloid cement. The
mechanical properties of the adhesiveinvestigated were Youngs
modulus 2 GPa and Poissons ratio0.30 which had been proved as an
excellent adhesive property[25]. The adhesive was applied on
degreased surfaces and thetwo sheets were pressed together in order
to squeeze sufficientadhesive out to avoid undue quilting of the
finished clinch-
bonded hybrid joints. The flow of the adhesive was removed.The
clinching processes were then produced before adhesivecuring. The
thickness of the adhesive layer was controlled bythe clinching
process. The average values of the bottomthickness of the
clinch-bonded hybrid joints is 1.5 mm thusthe thickness of the
adhesive layer is estimated to be 0.1 mm.Thereafter, the adhesive
was cured at room temperature for atleast 24 h. After curing, the
adhesive layer can give strongadhesive forces between sheets.
Figure 5 shows the clinchedjoints and clinch-bonded hybrid
joints.
A servo-hydraulic testing machine was used for the mono-tonic
tensile tests of the clinched joints and clinch-bondedhybrid
joints. For each test, six samples were mechanicallytested. The
distance between two grips was about 100 mm.The tests were
performed with a constant displacement rate of1 mm/min and
terminated when the sheets were separated orthe force drops to 20 %
of the peak force value. Continuousrecords of the applied
force-displacement curves were obtain-ed during each test. Figure
5a, b show the monotonic tensileprocess and failed joints
separately. It is clear from Fig. 5a, bthat the failure modes of
the clinched joints and clinch-bondedhybrid joints were neck
fracture mode, as shown in Fig. 5c.
Fig. 7 Energy absorption normal probability density
distributions of clinched joints and clinch-bonded hybrid
joints
Maximum Force
[kN]
EA
[J]
Fig. 8 Intercepts of strength and EA for clinched joints and
clinch-bonded hybrid joints
Int J Adv Manuf Technol
-
Such failure of the clinched joints and clinch-bonded
hybridjoints could be attributed to a too small clearance of the
toolsdiameters or a too deep die.
Under the tensile-shear load, the neck of the upper sheetbear a
main shear load by geometrical interlocking. When theshear stress
reaches the yield criterion of aluminium alloyAl5754, a crack is
initiated from the interfacial surface of theupper sheet and grows
into the upper sheet thickness. Afterrowing into the upper sheet,
crack kinks towards the buttoncentre and then propagates along the
circumference of thebutton neck of the upper sheet. Finally, the
inner button issheared off at the neck. In Fig. 5a, sheets were
separated forfive samples and not completely separated for one
sample.
Figure 6 shows the force-displacement curves of theclinched
joints and clinch-bonded hybrid joints. In the caseof the clinched
joints, after the peak, the force decreasesgradually. In the cases
of the clinch-bonded hybrid joints,however, after the peak, the
force suddenly drops. It is clearfrom Fig. 6 that the load-bearing
capacity of clinch-bondedhybrid joint is higher than that of the
clinched joint. It is alsoclear from Figs. 5 and 6 that the
repeatability of the clinchedjoints and clinch-bonded hybrid joints
are big though therepeatability of the adhesive joints was not very
big [5].
To examine the rationality of the test data, the
normalhypothesis tests were performed using MATLAB 7.0. Theresults
indicated that the tensile strengths of all the clinchedjoints and
clinch-bonded hybrid joints follow normal distribu-tions. The mean
values () and standard deviations () havethe following numerical
values: for the clinched joints C=1,895.30 N, C=43.81; for the
clinch-bonded hybrid jointsCB=2,022.50 N, CB=49.41. All test data
fitting the regionwas estimated by the degree of confidence of 95
%. Thetensile strengths normal probability density distributions
ofthe clinched joints and clinch-bonded hybrid joints are alsoshown
in Fig. 6.
5 EA of clinched joints and clinch-bonded hybrid joints
The normal hypothesis tests were performed to examine
therationality of the EA values of the clinched joints and
clinch-bonded hybrid joints. The results show that the EA values
ofall the clinched joints and clinch-bonded hybrid joints
follownormal distributions. For the clinched joints EAC=1.28
J,EAC=0.04; for clinch-bonded hybrid joints EACB=1.37 J,EACB=0.16.
All test data fitting the region was estimated bythe degree of
confidence of 95 %. The EA values normalprobability density
distributions of the clinched joints andclinch-bonded hybrid joints
are shown in Fig. 7.
Fig. 8 shows the intercept for load-bearing capacity and EAof
the clinched joints and clinch-bonded hybrid joints. It isclear
that both the maximum load and EAvalues of the clinch-bonded hybrid
joints are higher than that of the clinched joint.
This means that the addition of adhesive resulted in an
in-crease in both the load-bearing and the energy
absorptioncapacities of the clinched joints.
6 Summary
The extensible die clinching process has been
computationallyinvestigated in this paper using the commercial FE
softwareLS-Dyna. Experimental tests on the extensible die
clinchedjoints made of aluminium alloy Al5754 have been carried
outto validate the numerical simulation of the extensible
dieclinching process. The result obtained from tests agreed
fairlywell with the computational simulation.
Deformation and failure of homogeneous clinched jointsunder
tensile loading were investigated for validating the load-bearing
capacity and EA of the clinched joints and clinch-bonded hybrid
joints.
As mentioned above, the clinched joints were producedbefore
adhesive curing. In the extensible die clinching pro-cess, adhesive
layer can be fully sandwiched between twosheets. After curing, the
adhesive layer can increase thestrength of the clinched joints due
to the adhesion mechanism.However, after the peak load, the failure
of adhesive layeroccurs in a brittle manner. In this case, though
the clinch stillkeeps the sheets connected, the joint can only bear
low load,resulting in some more elongation.
Acknowledgments Financial support of the National Natural
ScienceFoundation of China (Grant No. 50965009) is gratefully
acknowledged.
References
1. He X, Gu F, Ball A (2014) A review of numerical analysis of
frictionstir welding. Prog Mater Sci 65:166
2. Mucha J (2013) The effect of material properties and joining
processparameters on behavior of self-pierce riveting joints made
with thesolid rivet. Mater Des 52:932946
3. Mucha J (2014) The numerical analysis of the effect of the
joiningprocess parameters on self-piercing riveting using the solid
rivet.Archive Civil Mech Eng
4. He X, Gu F, Ball A (2012) Recent development in finite
elementanalysis of self-piercing riveted joints. Int J Adv Manuf
Technol 58:643649
5. He X, Xing B, Zeng K, Gu F, Ball A (2013) Numerical and
exper-imental investigations of self-piercing riveting. Int J Adv
ManufTechnol 69:715721
6. Mucha J, Kascak L, Spisak E (2013) The experimental analysis
offorming and strength of clinch riveting sheet metal joint made
ofdifferent materials. Adv Mech Eng, Article ID 848973
7. Mucha J, Witkowski W (2013) The experimental analysis of
thedouble joint type change effect on the joint destruction process
inuniaxial shearing test. Thin-Walled Struct 66:3949
8. Neugebauer R, Jesche F, Israel M (2010) Enlargement of the
appli-cation range of solid punch riveting by two-piece dies. Int J
MaterForm 3:9991002
Int J Adv Manuf Technol
-
9. He X (2010) Recent development in finite element analysis
ofclinched joints. Int J Adv Manuf Technol 48:607612
10. Mucha J, Witkowski W (2014) The clinching joints strength
analysisin the aspects of changes in the forming technology and
load condi-tions. Thin-Walled Struct 82:5566
11. Lee C, Kim J, Lee S, Ko D, Kim B (2010) Design of
mechanicalclinching tools for joining of aluminium alloy sheets.
Mater Des31(4):18541861
12. He X, Zhao L, Yang H, Xing B, Wang Y, Deng C, Gu F, Ball
A(2014) Investigations of strength and energy absorption of
clinchedjoints. Comput Mater Sci
13. Gao S, Budde L (1994) Mechanism of mechanical press joining.
Int JMach Tools Manuf 34(5):641657
14. Mucha J (2011) The analysis of lock forming mechanism in
theclinching joint. Mater Des 32(10):49434954
15. Markowski T, Mucha J, Witkowski W (2013) FEM analysis
ofclinching joint machines C-frame rigidity.
EksploatacjaNiezawodnosc-Maint Reliab 1:5157
16. Varis JP (2003) The suitability of clinching as a joining
method forhigh-strength structural steel. J Mater Process Technol
132(13):242249
17. Varis JP (2006) Economics of clinching joint compared to
rivetedjoint and example of applying calculations to a volume
product. JMater Process Technol 172(1):130138
18. Neugebauer R, Todtermuschke M, Mauermann R, Riedel F
(2008)Overview on the state of development and the application
potential ofdieless mechanical joining processes. Archive Civil
Mech Eng 8(4):5160
19. Zheng J, He X, Xu J, Zeng K, Ding Y, Hu Y (2012) Finite
elementanalysis of energy saving jointing method base on energy
materials:clinching. Adv Mater Res 577:912
20. Lambiase F, Di Ilio A (2013) Finite element analysis of
material flowin mechanical clinching with extensible dies. J Mater
Eng Perform22(6):16291636
21. Lambiase F (2012) Influence of process parameters in
mechanicalclinching with extensible dies. Int J Adv Manuf Technol
66:21232131
22. He X (2011) A review of finite element analysis of
adhesivelybonded joints. Int J Adhes Adhes 31:248264
23. Bartczak B, Mucha J, Trzepiecinski T (2013) Stress
distribution inadhesively-bonded joints and the loading capacity of
hybrid joints ofcar body steels for the automotive industry. Int J
Adhes Adhes 45:4252
24. He X (2012) Numerical and experimental investigations of the
dy-namic response of bonded beams with a single-lap joint. Int J
AdhesAdhes 37:7985
25. He X, Ichikawa M (1993) Effect of thickness control of
adhesivelayer on strength of adhesive joints. In: Proceedings of
the 70th JSMESpring Annual Meeting, Tokyo, Japan, 166-168, (in
Japanese)
Int J Adv Manuf Technol
Numerical and experimental investigations of extensible die
clinchingAbstractIntroductionExtensible die clinching process
simulationExtensible die clinching process testsDeformation and
failure of clinched jointsEA of clinched joints and clinch-bonded
hybrid jointsSummaryReferences