Abstract ICCS
THE EFFECT OF PROTRUSION DENSITY ON COMPOSITE-METAL JOINTS WITH
SURFI-SCULPT REINFORCEMENT
Wei Xiong 1*, 2*, Xichang Wang3, John P. Dear1*, Bamber R.K.
Blackman 1*
1 Department of Mechanical Engineering, Imperial College London,
London SW7 2AZ, United Kingdom
2 First Aircraft Design Institute, Aviation Industry Corporation
of China, Xi'an 710089, China
3 Beijing Aeronautical Manufacturing Technology Research
Institute, Beijing 100024, China
Abstract
The effect of protrusion density on the static mechanical
properties of composite-metal joints strengthened by surfi-sculpt
protrusions has been experimentally studied with single lap joints.
The CFRP composite adherends were constant thickness with a
quasi-isotropic layup. The metallic adherends were Ti-6Al-4V alloy
with a variable number of protrusions per unit area, manufactured
by electron beam surfi-sculpt. Digital Image Correlation was used
to measure the debonding on the overlap during the tests. Although
the surfi-sculpt protrusions did not significantly affect the onset
of debonding, they did resist the initial unstable failure
mechanism and converted it into stable growth. The analysis
indicated that the efficiency of the surface protrusions was
different at the metal and composite ends of the overlap. This
finding opens the possibility to vary the protrusion density across
the overlap to meet specific damage tolerance criteria and optimise
joint efficiency. Increasing the protrusion density significantly
increased the ultimate failure load, joint extension and hence
absorbed energy.
Keywords: Composite-metal joint, Surfi-sculpt, electron beam
surfi-sculpt (EBS), density
1. Introduction
Joining dissimilar material such as carbon fibre reinforced
plastic (CFRP) to metals effectively has been a significant
challenge. The most common joining technologies are adhesive
bonding and mechanically fastened joints. For bonded joints,
adhesive bonding is usually sensitive to surface preparation and
many such joints are limited in use to secondary loading
applications due to their relatively low strengths. Moreover,
inspection of the bond-line in service is also difficult. The
subsequent failure of bonded joints is usually abrupt and
catastrophic. For mechanically fastened joints, there is a need for
holes to be formed in the composite and the intrinsic brittleness
of CFRP makes them sensitive to these stress concentrations [1].
The use of mechanical fasteners therefore reduces structural
integrity and also increases the weight, offsetting the potential
weight saving gained from using the composite.
An innovative joining technology known as Surfi-Sculpt adopts
the idea of z-pinning technology in a composite-composite joint by
creating arrays of macro-scale surfi-sculpts on the surface of
metal part. This allows a mechanical joint to form between the
metal and the composite without the need for holes [2]. This
advanced hybrid joint thus combines the advantages of a bonded and
mechanically fastened joint. In this work, the electron beam
surfi-sculpt (EBS) technique has been employed to create
protrusions on the metal adherends using a process that uses
multiple relative movements between an electron beam and the metal
surface [3, 4]. Although some studies on the mechanical properties
and damage tolerance of joints formed using this technology have
been reported [5-14], little has been reported on the effects of
protrusion density in hybrid joints.
The purpose of this study is to evaluate experimentally the
effect of surfi-sculpt protrusion density on the mechanical
properties of composite-metal joints strengthened by surfi-sculpt.
Four joint types with different surfi-sculpt protrusion densities
together with a reference joint have been tested and the initial
damage, ultimate failure, debonding propagation and absorbed energy
have been compared.
2. Single lap joint Manufacture
The single lap joints, as shown in Figure 1, has been adopted in
the present work. Although single lap joints introduce normal
stresses at the ends of the overlap which affects the apparent
shear strength of the joint, they are easy to manufacture and hence
are widely used [15]. Moreover, this manufacturing ease leads to
joints with higher quality and greater reliability than using
double lap joints to achieve a desired joint shape [10]. Single lap
joints provide more freedom for composite layup design (stacking
sequences) and are convenient for damage growth detection [12].
Figure 1 Single lap joint
The objective of this study is to investigate the effect of the
surfi-sculpt protrusion density on the static mechanical properties
of the advanced hybrid joints. The test matrix studied is shown in
Table 1. Four types of joint with different surfi-sculpt protrusion
densities were tested. Unstrengthened joints were also manufactured
and tested as reference. For each type of joint, at least two
specimens were tested.
Table 1 Test matrix
Joint type
CFRP nominal thickness (mm)
Composite layup
Surfi-sculpt array
Number of Protrusions
Joint No.
Reference
2.6
[45/0]3/45/[0/45]3
no surfi-sculpt
0
R
Protrusion
density effect
2.6
[45/0]3/45/[0/45]3
5, 6, 5, 6, 5, 6
33
DEN1
[45/0]3/45/[0/45]3
6, 7, 6, 7, 6, 7, 6, 7
52
DEN 2
[45/0]3/45/[0/45]3
7, 8, 7, 8, 7, 8, 7, 8
60
DEN 3
[45/0]3/45/[0/45]3
9,10,9,10,9,10,9,10,9,10
95
DEN 4
2.1. Materials
Ti-6Al-4V [4, 16] was used to manufacture the metal adherend.
This alloy is frequently used in the manufacture of hybrid
(metal-CFRP) structures in aerospace applications. The composite
used to manufacture CFRP adherend was a hot-melt, epoxy prepreg SE
84LV [17]. The fibre of the prepreg is RC200T which is a woven
fibre.
Electron beam surfi-sculpt (EBS) was adopted to manufacture
surfi-sculpt protrusions on the bonding surface of the metal
adherends [3, 4]. The electron beam impacts the metal surface
through a special track. The external metal on the track is melted
and this molten metal is displaced due to surface tension and
vapour tension forming a projection [18]. After a portion of the
molten metal solidifies, the EB scan is repeated one or more times.
Desired protrusions can be formed with special tracks and using the
optimized processing parameters of the EB. Figure 2 shows the
surfi-sculpt protrusion shape obtained for the present work.
Figure 2 Surfi-sculpt protrusions
2.2. Joint manufacture
The integration of the metal and composite adherends was
completed using vacuum bag processing [17]. Before the vacuum bag
processing, the full thickness of the CFRP adherend was built up
using the required number and sequence of prepregs (Table 1) and
this was placed onto the metal adherends. During the vacuum bag
processing, the protrusions were pressed into the uncured laminate
with the pressure provided by the vacuum bag. The assembly was
finally co-bonded on a hot plate, after which the composite was
trimmed to size. Bespoke tooling was used to ensure thorough
consolidation of the CFRP matrix and to minimise any misalignment
between the laminate and the metal adherend. The lap joints
manufactured had a width of 25.4 mm and a nominal overlal length of
30 mm.
Figure 3 Initial damage to CFRP
The insertion of the surfi-sculpt protrusions into the composite
introduces resin rich zones and can cause some fibre damage, as
shown in the scanning electron micrograph in Figure 3. To prevent
the resin rich zones joining together the distance between two
adjacent protrusions in both the horizontal and vertical directions
was carefully controlled. Every other row of protrusions was offset
by a distance equal to half the inter-protrusion distance along the
row direction. Taking joint ‘DEN1’ for example, the final
surfi-sculpt protrusion array design of joint DEN1 is shown in
Figure 4, and contains three rows of 5 protrusions and three rows
of 6 protrusions, giving 33 protrusions in total over the overlap
area. As the bonding area and the protrusion distribution pattern
of all strengthened joints were nominally the same, the
surfi-sculpt protrusion number (i.e. number of protrusions on the
overlap) is used to represent the corresponding surfi-sculpt
protrusion density in this paper.
Figure 4 Surfi-sculpt protrusion array for joint design DEN1
3. Experimental
Single lap shear tests were conducted under displacement control
on an Instron testing machine (model 3369) at ambient temperature.
The cross head rate was 0.3 mm/min. This is a much lower rate than
specified in ASTM D5868 [19] to allow for smaller changes in the
compliance of the joint to be recorded. Prior to loading, an
extensometer with a gauge length of 50 mm was attached to the
specimen as shown in Figure 5. During the tests, the extension of
the gauge length was recorded. DIC (Digital Image Correlation) [20]
was also employed on the lateral side of the joint to identify the
occurrence and the evolution of the debonding of the joint.
Figure 5 Single lap shear test setup
4. Results and discussion4.1. Load-displacement behaviour
Figure 6 Typical quasi-static load vs displacement curves for
the joints
Figure 6 compares the typical load-displacement curve measured
for the reference joint (control- no protrusions) with the curves
for the strengthened joints manufactured with four different
protrusion densities (DEN1 to DEN4).
The curve of the reference joint rises linearly up to 4 kN, at
which point the stiffness of the joint experiences an abrupt change
due to the initiation of debonding. The load at this point is
referred to as the damage initiation load. The joint is then
further damaged by unstable debonding after a further small
increase in load (about 200 N). The resulting debonding propagates
rapidly and the joint fails catastrophically because the joint
geometry under the tensile load presents an increasing energy
release rate with crack propagation and the interface is relatively
brittle [21].
The curves for the strengthened joints also increase linearly up
to the initial damage load point. It is noted that all joints, with
same composite layup design, show the same slope (within the range
of experimental error) indicating that the initial joint stiffness
is not affected significantly by the different surfi-sculpts
densities.
Joints DEN1-3 exhibited damage initiation at about 4.4 kN
whereas joint DEN4 exhibited damage initiation at about 5 kN.
Following damage initiation, all strengthened joints exhibit
further deformation with a lower stiffness. This reduced stiffness
value was broadly constant for all pin densities studied. Joints
DEN1-3 failed after significant additional load increase (at
approximately constant stiffness) at load values in the range of
9-11 kN. However, joint DEN4 exhibited an additional
load-displacement behavior at about 13 kN. Thus, for joints DEN4,
three different loading stages were observed. Firstly the initial
linear stage up to damage initiation. Then the load increases
further, almost linearly but with a lower slope, to about 13 kN.
Then finally, with further displacement, the trace becomes nearly
horizontal. Experimental observation shows that the surfi-sculpt
protrusions near both the metal and composite ends of the overlap
begin to be pulled out after this point. The load point of 13 kN is
referred to here as the transition point. Because the protrusions
pull out of the composite in an unstable manner and because there
is very little further increase in strength following the
transition point (only about 6%), it is suggested that the
transition point is a more reliable point to define to characterize
the joint failure. The data analysis of debonding propagation,
joint extension and absorbed energy also support this suggestion,
as is discussed further below.
4.2. Damage initiation load
Figure 7 Damage initiation load for the joints
Figure 7 shows the damage initiation load plotted against the
number of protrusions present on the overlap area for the reference
and strengthened joints. As the number of protrusions increases
(i.e. as the density increases) the damage initiation load
increases somewhat but appears to plateau for DEN4. According to
the best fitted line through the data, when the protrusion number
is less than 25, the damage initiation load remains approximately
constant at 4 kN. When the protrusion number increases from 25 to
80, the damage initiation load increases almost linearly. Then the
damage initiation load reaches a plateau value at about 5 kN at a
protrusion number of 80, irrespective of further increases in
protrusion number. The maximum increase in the damage initiation
load was 23% for joint DEN4, relative to the reference joint.
4.3. Ultimate failure load
Figure 8 Ultimate failure load for the joints
Figure 8 compares the ultimate failure load for the reference
and strengthened joints. All the strengthened joints were much
stronger than the reference joints, and the ultimate failure load
increased with increasing number of surfi-sculpt protrusions. DEN4
was the strongest joint (the ultimate failure load was 1. 34 kN)
and achieved the maximum increase of 212% relative to the reference
joint, R, (for which the ultimate failure load was 4.29 kN). The
ultimate failure loads for joints DEN1, DEN2 and DEN3 were
increased by 115%, 150% and 171% respectively, relative to the
reference joint. However, the rate of increase in the ultimate
failure load per additional protrusion decreased as the number of
protrusions increased.
Figure 9 Ultimate failure load increase rate for the joints
To compare the efficiency of additional protrusions for the
various joints, a parameter called the ultimate failure load
increase rate (UFLIR) has been used. The UFLIR is defined as:
UFL : Ultimate failure load
: The number of surfi-sculpt protrusions of the joint
DEN(i);
DEN(0 ) refers to the Reference joint
Figure 9 shows the comparison of UFLIR values for DEN1 to DEN4.
It is clear that the values of UFLIR and hence the efficiency of
additional protrusions decreases with increasing protrusion number.
DEN1, with the minimum number of protrusions achieved the largest
UFLIR. On average, per additional protrusion, the ultimate failure
load was increased by 150 N compared with reference joint. DEN4,
with the maximum number of protrusions, achieved the smallest value
of UFLIR. On average, per additional protrusion, the ultimate
failure load increased by 96 N compared with the reference joint.
Moreover, it should be noticed that higher density will cause more
serious damage into composite adherend and short distance between
protrusion make the linkage of resin rich zone more possible, thus
decreasing the joint’s mechanical properties significantly.
Besides, during manufacture, it is found that the protrusion
density higher than DEN4 makes it extremely difficult for
protrusions to insert composite adherend and make protrusions lose
their function completely. For this material system and manufacture
method, the optimum protrusion density is near that of DEN4.
4.4. Damage mechanisms and damage evolution4.4.1. Damage
mechanisms
The initially overlapped surfaces of the failed joints were
inspected after each test. Visual inspection of the bonding areas
on the metal and composite adherends showed evidence of the damage
mechanisms that had occurred in the strengthened joints. In each
case all the protrusions had failed and the plane of the fracture
surfaces through the residual protrusions was not parallel to the
bonding surface, as is shown in Figure 10 (a). For a protrusion,
its initial height is defined as h. Following fracture, the
residual protrusion shows a different height at its left and right
sides, as is shown in Figure 10 (b). Here the height on the right
side is termed the damage initiation height, . It is determined by
the external force acting on a protrusion. The height at the left
side is termed the rear fracture height . It results mainly from
the internal microstructure of the re-solidified protrusion.
Because was found not to be repeatable, the height of the residual
protrusion (residual height) has been defined as the height at the
damage initiation point on protrusions,, as shown in Figure 10
(b).
hrear
(a) Force diagram of a protrusion (b) The residual
protrusion
Figure 10 Protrusion force diagram and residual height
definition
For joints DEN1-4, the distribution pattern of the residual
heights were similar (shown in Figure 11): the residual protrusions
near the metal adherend end are the highest, then the residual
height decreases from the metal adherend end to composite adherend
end. The lowest residual protrusions heights were near composite
adherend end, at about 2/3 length of the bond line from the metal
adherend end. Then the residual height increased from the lowest
position to the composite adherend end, but the height of the
residual protrusions near the composite adherend end is lower than
those near the metal adherend end. Figure 11 also shows that the
residual height on the joint DEN4 (with the maximum surfi-sculpt
protrusion number) is higher than that on the joint DEN1 (with the
minimum surfi-sculpt protrusion number).
(a) Joint DEN1
(b) Joint DEN4
Figure 11 The joint damage and protrusion residual height
distribution
The variation in the residual protrusion heights indicate that
they experience different load conditions across the joint. This
results in different failures being attained. It is believed all
the differences are the result of the joint deformation in
thickness direction.
Figure 12 shows a typical elastic deformation of a single lap
joint under a tensile load. The joint experiences bending
deflection normal to the external load in addition to a tensile
deformation parallel to the external load, caused by the
unsymmetrical geometry of the single lap joint. To compare the
joint deformation in the normal (Z direction) and the protrusion
residual height, the coordination is defined as shown in Figure 13:
The origin is located at the bonding interface at the metal end of
the overlap. The joint bending deflection are in the Z
direction.
Figure 12 Typical single lap joint deformation under tensile
load
Figure 13 Joint coordination
The residual height value is always positive but the joint
deflection is negative near metal end, so the absolute value of the
normalized joint bending deflection, shown as the dashed line in
Figure 14, is compared with the residual heights along the bond
line for the various joints. These deflection values are normalized
by the deflection value at X=0, such that the normalized deflection
=1 when X=0. The joint deflection shows a bilinear trend. The point
(X=21) shows no deflection and is defined as the neutral point.
From the neutral point to the metal end (X=0), the deflection
increases linearly and reaches the largest value at the metal end.
The same trend is evident from the neutral point to the composite
end (X=29), but the deflection at the composite end is smaller than
at the metal end. It is evident that the residual height of the
protrusions in the joints is proportional to the protrusion
density. The residual height of protrusions in the joints (given by
the data points in Figure 14) show bilinear trends along the bond
line as was noted for the joint deflections. The minimum residual
height occurs in the vicinity of the neutral point, according to
the bilinear lines fitted through the data for joints DEN1-4. The
maximum residual height occurs near the metal end and the residual
height near the composite end is smaller than that near the metal
end. It is believed that the crack opening displacement at a large
deflection point is larger than that at a smaller deflection point,
so the force (applied to the metal protrusion by the composite
adherend) is applied further from the metal protrusion root, thus
the critical damage initiation location is higher, which leads to
larger residual height .
Figure 14 The comparison of joint deflection with residual
height
Table 2 Damage mechanisms in hybrid joints
Joint
Damage Mechanisms
R
Unstable bond line debonding
DEN1
Composite compression failure; protrusion breakage (maximum 0.15
mm)
DEN2
Composite compression failure; protrusion breakage (maximum 0.17
mm)
DEN3
Composite compression failure; protrusion breakage (maximum 0.19
mm)
DEN4
Composite compression failure; protrusion breakage (maximum 0.34
mm)
The damage mechanisms observed are summarized in Table 2. The
reference joints failed by rapid, unstable debonding and two
adherends separate catastrophically, immediately following damage
initiation. Joint DEN1 only sees the breakage of protrusions.
Joints DEN2-4 exhibit similar damage mechanisms with the residual
height increasing with increasing protrusion density.
4.4.2. Damage evolution
The DIC measurements taken from a camera facing the X-Z plane
allowed the crack growth from both ends of the overlap to be
determined during the tests. From the recorded images, the strains
were analyzed. Figure 15 shows a typical distribution for a test.
The length of the crack formed by the debonding of the interface
from the composite end of the overlap is defined as and the length
of the similarly formed crack from the metal end of the overlap is
defined as . The total crack length is given by the sum of .
Figure 15 DIC data from overlap region showing debonding
Joints DEN1-4 exhibited debonding from the composite end first,
prior to attaining the damage initiation load. Then the joints
started to debond from the metal end at a load equivalent to the
damage initiation load and the debonding continued to propagate
from both ends of the overlap until ultimate failure. The maximum
total crack length for joints DEN1-3 reached about half of the
overlap length prior to final failure. However, for joint DEN4, the
total crack length grew across the entire overlap prior to
failure.
For the reference joint, there is only debonding from one end of
the overlap (the composite end) prior to the ultimate failure of
the joint. The initial debonding is about 2 mm and then the
debonding crack propagates quickly to 3 mm with a 200N load
increase and then the joint fails with total debonding.
Debonding at composite end
DIL
Debonding at composite end
DIL
(a) (b)
Figure 16 Nonlinear behaviour caused by debonding from composite
end for joints
The reason that the debonding started from the composite end in
all joints is that this end experienced the largest normal (mode I
opening) stress (Although the deflection on the metal end is
higher, an elastic finite element analysis model of the joints
shows that the opening stress is higher at composite end). For
joints DEN1-4 the debonding did not lead to early catastrophic
failure but resulted in the smooth and continuum nonlinear
load-displacement behaviour prior the damage initiation point, as
shown in Figure 16. The debonding propagation was delayed by the
nearest row of protrusions. The joint deflection at the composite
end was so small that the joint stiffness was only slightly
affected by the debonding at the composite end.
Debonding from the metal end of the overlap occurred after the
damage initiation load and was also resisted by the nearest row of
protrusions which the crack encountered. However, the joint
deflection at the metal end is larger than that on the composite
end and the joint stiffness is therefore affected more
significantly by the debonding. The debonding from the metal end
thus leads to significant changes in the load-displacement curve
(due to reduction in stiffness) and the load at this point
corresponds to the damage initiation load, as shown in Figure
16.
Figure 17 shows development of the total debonding length for
joints DEN1-4 versus the applied load. Joints DEN1-3 all show
bilinear behaviour, which comprises of two stages. The first stage
is a short and rapid debonding propagation stage, during which the
debonding length rises quickly to about 4 mm. The second stage is
characterized by the slow propagation of debonding with a
significantly reduced slope with increasing load up to the ultimate
failure load. However, joint DEN4 shows an additional third stage:
stage two leads finally to a transition point rather than ultimate
failure. Following this transition point, the propagation of the
debonding became rapid and unstable. This instability following the
transition suggests that the transition point should be considered
as the ultimate failure point for this joint.
Bondline length
Figure 17 Debonding length of strengthened joints vs load
Figure 17 also indicates that there is no clear correlation
between the initial debonding (i.e. intersection of each line on
the load axis) and the number of protrusions present. It is
believed that these initial values are influenced significantly by
manufacturing details such as the existence of a resin fillet at
the ends of the overlap. The knee between stages one and two is
observed to always occur after about 4 mm of debond growth. Beyond
the knee (i.e. in stage two) the surfi-sculpt protrusions are
effectively resisting the rapid debond growth, resulting in a much
more gradual crack growth than was observed in stage one. In stage
two, the slope of the line fitted to the total crack length versus
load data, defined as the debonding length growth rate, decreases
with increasing protrusion density.
Figure 18 Normalized debonding length growth rate
Figure 18 shows the normalized (with respect to DEN1) crack
growth rate for the joints DEN1-4. Compared with DEN1, joints DEN2
and DEN3 achieve only a 5% and 12% decrease in crack growth rate
respectively. However, DEN4 achieves a 47% decrease in crack growth
rate. Thus, it is clear that although the protrusions do not affect
the load at which debonding initiates in the joint, their existence
significantly decreases the subsequent crack growth rate.
(a) Debonding length from composite end
(b) Debonding length from metal end
Figure 19 Debonding growth from each end of the overlap vs
load
It is of interest to partition the total crack growth into the
two components: acomp and am. Figure 19 shows these partitioned
values. Both acomp and am have similar propagation behaviour and
are similar to the total crack growth behaviour shown in Figure 17.
The debonding from the composite end and the metal end in joints
DEN1-3 show bilinear behavior. The debonding in joint DEN4 again
shows three stages: the bilinear stages (the first two stages) and
the last unstable debonding growth stage after the transition
point. It is evident that during stage one, there is about twice as
much crack growth from the composite end (4 mm) than from the metal
end (2 mm). Both debonding lengths are 1mm less than the distance
from the corresponding adherend end to the nearest protrusion
respectively, which indicates that the initial unstable debonding
on both ends is resisted by the nearest protrusions during the
first stage.
Figure 20 Debonding length from each end of bond line
Figure 20 shows the average debonding length from each end of
the overlap at the end of the second stage (i.e. at the ultimate
failure point) versus the protrusion number for the joints. When
the protrusion number increased from DEN1 to DEN2, both debonding
lengths increased significantly and reached a similar length of
about 11 mm. This corresponded to an increase in the ultimate
failure load for the DEN2 joint of 34% compared to DEN1. When the
protrusion number was increased further to DEN3 and then DEN4, the
crack lengths from both ends of the overlap decrease but remain
similar.
Although increased protrusion density can increase the
resistance of the joint to debonding, at the same time it increases
the ultimate load and this increased load leads to a longer
debonding length. From DEN1 to DEN2, the increased resistance is
not able to resist the debonding resulting from this increased
ultimate load, leading to longer debonding lengths from both ends
of the overlap. However, when the protrusion density is higher than
DEN2, the protrusion density is high enough to resist the debonding
caused by the increased ultimate failure load.
Figure 21 Normalized debonding length growth rate vs protrusion
density
Figure 21 shows the debonding propagation rate during the second
stage (stable debonding propagation stage) for joints DEN1-4. The
slope (mm/N) obtained from the fit to the data in the second stage
of Figure 19 (a) and (b) is normalized by the debonding rate of
DEN1 at the metal end. The results indicate that the debonding from
the metal end of the overlap grows faster than the debonding from
the composite end for all joints. The debonding growth rate (mm/N)
from the composite end was about 58% of that from metal end. It is
also clear that the reduction in the rate of debonding growth is
greater at the metal end than that at the composite end. In joint
design, Figure 21 can allow for the tailoring of the protrusion
density to achieve an allowable debonding growth rate by using
different protrusion densities at the two ends of the overlap
(higher at the metal end than the composite end). This finding may
have significant potential to inform future joint design with
sculpted metallic adherends.
4.5. Joint extension and energy absorption
Figure 22 Joint extension at failure for all joints
Figure 22 shows the comparison of extension versus protrusion
density for all joints tested. All the strengthened joints can
withstand higher extension than reference joints. It should be
noticed that although joint DEN4 shows a higher extension value at
the maximum load point, it has a much longer error bar, indicating
that the extension value is not repeatable. In contrast, the
extension value at the transition point is more repeatable. The
joint extension at failure values increase with the protrusion
number. Joint DEN4, with highest protrusion density, shows that the
average extension was increased by 423% compared with reference
joint. Joint DEN1, with the lowest protrusion density, achieves the
least extension increase of 223% compared with reference joint. The
extension of joints DEN2-3 were increased by 287%. The figure also
shows that the rate of increase of extension decreases with
increasing protrusion number.
Figure 23 Energy absorption
The energy absorbed by the joint is defined as the area under
the load-displacement curve. The increase in both the ultimate
failure load and extension of the joints results in a remarkable
increase in energy absorption. Figure 23 shows the values of the
energy absorbed by the various joints. It also shows that although
joint DEN4 shows a much more absorbed energy at the maximum load
point, it has a much larger uncertainty, indicating that the
absorbed energy is not repeatable at the maximum point. However,
the absorbed energy is more repeatable when measured at the
transition point. This confirms the greater reliability of defining
final failure as the transition point for joints exhibiting a three
stage failure process. The absorbed energy increases linearly with
the surfi-sculpt protrusion number, indicating that most energy
absorption is due to the failure of surfi-sculpt protrusions. Joint
DEN4, with the most surfi-sculpt protrusions, shows that the
average absorbed energy was increased by 1616% compared with
reference joint. Joint DEN1, with the minimum surfi-sculpt
protrusions, achieved the least energy absorption increase of 672%
compared with reference joint. The absorbed energy of joints DEN2
and DEN3 were increased by 957% and 1012% respectively.
5. Conclusions
The effect of the surfi-sculpt protrusion density on the static
mechanical properties of advanced hybrid joints has been
experimentally studied. Four joint types with different protrusion
densities were tested in tension and the corresponding results were
compared with that of reference joints without protrusions present.
The maximum load capacity and best damage tolerance was achieved
with the highest surfi-sculpt density, i.e. the joints of type
DEN4.
Digital image correlation analysis was used to measure the
growth of the debonding along the joint overlap during the tests.
It was shown that although the protrusions did not affect the joint
stiffness and the onset of debonding significantly, the protrusions
did resist the initial unstable failure mechanism and converted it
into stable growth. With increasing protrusion density, the
debonding propagation rate decreased and the failure mode was shown
to change from debonding to intra-laminar failure of the composite
and fracture of the metallic protrusions in a mixed shear and
bending mode. The analysis also showed the efficiency of
protrusions was different at the two ends of the overlap, i.e. at
the metal and composite ends respectively. This opens the
possibility that the protrusion density could be designed to vary
across the overlap to meet a damage tolerance criterion and also to
optimise protrusion efficiency. Finally, increasing the protrusion
density was shown to significantly increase the ultimate failure
load, joint extension and hence absorbed energy. Simulations are
being conducted to further understand these observations.
Acknowledgements
The strong support from the Aviation Industry Corporation of
China (AVIC) the First Aircraft Institute (FAI) and Beijing
Aeronautical Manufacturing Technology Research Institute (BAMTRI)
for this funded research is much appreciated. The research was
performed at the AVIC Centre for Structural Design and Manufacture
at Imperial College London.
References
[1] Hou L, Liu D. Size Effects and Thickness Constraints in
Composite Joints. Journal of Composite Materials.
2003;37:1921-38.
[2] Dance BGI, Kellar EJC. Workpiece Structure Modification. In:
Organization WIP, editor.2004. p. 58.
[3] Dance BGI, Kellar EJC. WORKPIECE STRUCTURE MODIFICATION. In:
States U, editor.2003. p. 28.
[4] Wang X, Guo E, Gong S, Li B. Realization and Experimental
Analysis of Electron Beam Surfi-Sculpt on Ti-6Al-4V Alloy. Rare
Metal Materials and Engineering. 2014;43:819-22.
[5] Parkes PN, Butler R, Meyer J, de Oliveira A. Static Strength
of Metal-Composite Joints with Penetrative Reinforcement. Composite
Structures. 2014.
[6] Parkes PN, Butler R, Almond D. Fatigue of Metal-Composite
Joints with Penetrative Reinforcement. Aiaa/asme/asce/ahs/asc
Structures, Structural Dynamics, and Materials Conference2013.
[7] Graham DP, Rezai A, Baker D, Smith PA, Watts JF. The
development and scalability of a high strength, damage tolerant,
hybrid joining scheme for composite–metal structures. Composites
Part A: Applied Science and Manufacturing. 2014;64:14.
[8] Parkes PN, Butler R, Almond DP. Fatigue of metal-composite
joints with penetrative reinforcement. Proceedings of 54th
AIAA/ASME/ASCE/AHS/ASC Structures, Structural Dynamics and
Materials Conference. Boston, USA2013.
[9] Graham DP, Rezai A, Baker D, Smith PA, Watts JF. A hybrid
joining scheme for high strength multi-material joints. Proceedings
of 18th International Conference on Composite Materials. Jeju,
South Korea2011.
[10] Ucsnik S, Scheerer M, Zaremba S, Pahr DH. Experimental
investigation of a novel hybrid metal–composite joining technology.
Composites Part A: Applied Science and Manufacturing.
2010;41:369-74.
[11] Zhang H, Wen W, Cui H. Study on the strength prediction
model of Comeld composites joints. Composites Part B: Engineering.
2012;43:3310-7.
[12] Parkes PN, Butler R, Almond DP. GROWTH OF DAMAGE IN
ADDITIVELY MANUFACTURED METAL-COMPOSITE JOINTS. ECCM15 - 15TH
EUROPEAN CONFERENCE ON COMPOSITE MATERIALS. Venice, Italy2012.
[13] Ewen JCK, Faye S. Comeld - A new approach to damage control
for composite to metal joints. Design & Performance of
Composite Materials Conference. The University of
Sheffield2005.
[14] Wang X, Ahn J, Kaboglu C, Yu L, Blackman BRK.
Characterisation of composite-titanium alloy hybrid joints using
digital image correlation. Composite Structures.
2016;140:702-11.
[15] ASTM. D4896 − 01 Standard Guide for Use of Adhesive-Bonded
Single Lap-Joint Specimen Test Results. 2008. p. 6.
[16] Inc AASM. Titanium Ti 6AL 4V - AMS 4911. 2014.
[17] SE 84LV LOW TEMPERATURE CURE EPOXY PREPREG. GURIT;
2015.
[18] Wang XC, Guo EM, Gong SL, Bruce D. Study of Electron Beam
Surfi-sculpt during Composite Materials Joining. Rare Metal
Materials and Engineering. 2011;40.
[19] ASTM. D5868 − 01 Standard Test Method for Lap Shear
Adhesion for Fiber Reinforced Plastic (FRP) Bonding. 2014. p.
2.
[20] McCormick N, Lord J. Digital Image Correlation. Materials
Today. 2010;13:52-4.
[21] Chang P, Mouritz AP, Cox BN. Properties and failure
mechanisms of z-pinned laminates in monotonic and cyclic tension.
Composites Part A: Applied Science and Manufacturing.
2006;37:1501-13.
Figure
Figure 1 Single lap joint
Figure 2 Surfi-sculpt protrusions
Figure 3 Initial damage to CFRP
Figure 4 Surfi-sculpt protrusion array for joint design DEN1
Figure 5 Single lap shear test setup
Figure 6 Typical quasi-static load vs displacement curves for
the joints
Figure 7 Damage initiation load for the joints
Figure 8 Ultimate failure load for the joints
Figure 9 Ultimate failure load increase rate for the joints
(a) Force diagram of a protrusion (b) The residual
protrusion
Figure 10 Protrusion force diagram and residual height
definition
(a) Joint DEN1
(b) Joint DEN4
Figure 11 The joint damage and protrusion residual height
distribution
Figure 12 Typical single lap joint deformation under tensile
load
Figure 13 Joint coordination
Figure 14 The comparison of joint deflection with residual
height
Figure 15 DIC data from overlap region showing debonding
Debonding at composite end
DIL
Debonding at composite end
DIL
(a) (b)
Figure 16 Nonlinear behaviour caused by debonding from composite
end for joints
Bondline length
Figure 17 Debonding length of strengthened joints vs load
Figure 18 Normalized debonding length growth rate
(a) Debonding length from composite end
(b) Debonding length from metal end
Figure 19 Debonding growth from each end of the overlap vs
load
Figure 20 Debonding length from each end of bond line
Figure 21 Normalized debonding length growth rate vs protrusion
density
Figure 22 Joint extension at failure for all joints
Figure 23 Energy absorption
Tables
Table 1 Test matrix
Joint type
CFRP nominal thickness (mm)
Composite layup
Surfi-sculpt array
Number of Protrusions
Joint No.
Reference
2.6
[45/0]3/45/[0/45]3
no surfi-sculpt
0
R
Protrusion
density effect
2.6
[45/0]3/45/[0/45]3
5, 6, 5, 6, 5, 6
33
DEN1
[45/0]3/45/[0/45]3
6, 7, 6, 7, 6, 7, 6, 7
52
DEN 2
[45/0]3/45/[0/45]3
7, 8, 7, 8, 7, 8, 7, 8
60
DEN 3
[45/0]3/45/[0/45]3
9,10,9,10,9,10,9,10,9,10
95
DEN 4
Table 2 Damage mechanisms in hybrid joints
Joint
Damage Mechanisms
R
Unstable bond line debonding
DEN1
Composite compression failure; protrusion breakage (maximum 0.15
mm)
DEN2
Composite compression failure; protrusion breakage (maximum 0.17
mm)
DEN3
Composite compression failure; protrusion breakage (maximum 0.19
mm)
DEN4
Composite compression failure; protrusion breakage (maximum 0.34
mm)
6x6-1(1)3469.11063499.58154000000013529.91699000000023559.50293000000013917.32227000000013958.9013700.355731225296442691.18577075098814232.01581027667984182.49011857707509864.7430830039525693DEN13958.901374507.27880999999985021.72753999999995548.64794999999965736.444345779.12597999999986043.18505999999986573.37694999999997102.91503999999997641.17235999999968187.53075999999968721.39160000000089205.94530999999924.74308300395256935.33596837944664065.45454545454545415.6916996047430835.86896634327464377.11462450592885447.82608695652173928.18181818181818179.249011857707509810.19762845849802311.02766798418972313.16205533596837914.5849802371541497x8-1(1)2509.17922565.17383000000022959.439940000000200.588235294117647083.1764705882352939DEN22959.43994000000023502.838623743.65601000000023794.406254237.35498000000014808.31444999999996294.60595999999996858.13965000000017416.343758265.81151999999939013.97655999999929536.198239999999710095.8564499999993.176470588235293944.05882352941176454.58823529411764675.88235294117647017.05882352941176458.35294117647058710.11764705882352910.58823529411764912.58823529411764913.76470588235294215.52941176470588216.1176470588235298x8-1(1)3295.48754999999983314.77417000000013586.832033608.814940000000200.472440944881889760.826771653543307062.8346456692913384DEN33608.81494000000024689.87207000000035331.156255347.69141000000045510.90332000000036388.93896000000047810.57031000000018869.74022999999949993.785159999999510851.3232411707.274412.83464566929133843.77952755905511814.01499999999999974.72440944881889726.25984251968503898.385826771653544410.39370078740157411.33858267716535414.0551181102362214.76377952755905614.881889763779528DEN45993.96093999999996743.03075999999967989.20604999999989227.757809999999211449.6562513230.4404300000015.08366533864541835.768.2800000000000011911.0412.2410x10-1(3)13230.44043000000113688.7968814066.19140999999914140.5927699999991414212.2414.87999999999999917.5219.7999999999999973010x10-1(1)4145.36376999999994206.63231999999974766.28369000000025038.52001999999995683.49608999999965734.23095999999995993.960939999999900.61.562.0434.43999999999999955.08366533864541836x6-1-303469.11063499.58154000000013529.91699000000023559.50293000000013917.32227000000013958.901374507.27880999999985021.72753999999995548.64794999999965736.444345779.12597999999986043.18505999999986573.37694999999997102.91503999999997641.17235999999968187.53075999999968721.39160000000089205.945309999999200.355731225296442691.18577075098814232.01581027667984182.49011857707509864.74308300395256935.33596837944664065.45454545454545415.6916996047430835.86896634327464377.11462450592885447.82608695652173928.18181818181818179.249011857707509810.19762845849802311.02766798418972313.16205533596837914.5849802371541497x8-1-302509.17922565.17383000000022959.43994000000023502.838623743.65601000000023794.406254237.35498000000014808.31444999999996294.60595999999996858.13965000000017416.343758265.81151999999939013.97655999999929536.198239999999710095.85644999999900.588235294117647083.176470588235293944.05882352941176454.58823529411764675.88235294117647017.05882352941176458.35294117647058710.11764705882352910.58823529411764912.58823529411764913.76470588235294215.52941176470588216.1176470588235298x8-1-303295.48754999999983314.77417000000013586.832033608.81494000000024689.87207000000035331.156255347.69141000000045510.90332000000036388.93896000000047810.57031000000018869.74022999999949993.785159999999510851.3232411707.2744100.472440944881889760.826771653543307062.83464566929133843.77952755905511814.01499999999999974.72440944881889726.25984251968503898.385826771653544410.39370078740157411.33858267716535414.0551181102362214.76377952755905614.88188976377952810x10-1-304145.36376999999994206.63231999999974766.28369000000025038.52001999999995683.49608999999965734.23095999999995993.96093999999996743.03075999999967989.20604999999989227.757809999999211449.6562513230.44043000000113688.7968814066.19140999999914140.5927699999991414200.61.562.0434.43999999999999955.08366533864541835.768.2800000000000011911.0412.2414.87999999999999917.5219.79999999999999730
Load (N)
Total debonding length (mm)
3352609510.955050237969328420.882601797990481110.52564780539397149
Surfi-sculpt protrusion number
Normalized debonding rate
DEN13958.901374507.27880999999985021.72753999999995548.64794999999965736.444345779.12597999999986043.18505999999986573.37694999999997102.91503999999997641.17235999999968187.53075999999968721.39160000000089205.94530999999924.74308300395256935.33596837944664065.45454545454545415.6916996047430835.86896634327464375.92885375494071195.92885375494071196.04743083003952546.40316205533596877.11462450592885357.2332015810276688.18181818181818178.89328063241106656x6-1(1)3469.11063499.58154000000013529.91699000000023559.50293000000013917.32227000000013958.9013700.355731225296442691.18577075098814232.01581027667984182.49011857707509864.74308300395256937x8-1(1)2509.17922565.17383000000022959.439940000000200.588235294117647083.1764705882352939DEN23502.838623743.65601000000023794.406254237.35498000000014808.31444999999996294.60595999999996858.13965000000017416.343758265.81151999999939013.97655999999929536.198239999999710095.85644999999944.05882352941176454.1176470588235294.1176470588235294.58823529411764674.82352941176470564.94117647058823554.94117647058823556.94117647058823557.29411764705882347.76470588235294118.2352941176470588x8-1(1)3295.48754999999983314.77417000000013586.832033608.814940000000200.472440944881889760.826771653543307062.8346456692913384DEN33608.81494000000024689.87207000000035331.156255347.69141000000045510.90332000000036388.93896000000047810.57031000000018869.74022999999949993.785159999999510851.3232411707.274412.83464566929133843.77952755905511814.01499999999999974.0157480314960634.13385826771653524.37007874015748054.84251968503937045.66929133858267696.96850393700787417.08661417322834637.086614173228346310x10-1(1)4145.36376999999994206.63231999999974766.28369000000025038.52001999999995683.49608999999965734.23095999999995993.960939999999900.61.562.04333.3505976095617527DEN45993.96093999999996743.03075999999967989.20604999999989227.757809999999211449.6562513230.4404300000013.35059760956175273.64.24.44000000000000045.4610x10-1(3)13230.44043000000113688.7968814066.19140999999914140.5927699999991414266.848.279999999999999410.199999999999999156x6-1-103469.11063499.58154000000013529.91699000000023559.50293000000013917.32227000000013958.901374507.27880999999985021.72753999999995548.64794999999965736.444345779.12597999999986043.18505999999986573.37694999999997102.91503999999997641.17235999999968187.53075999999968721.39160000000089205.945309999999200.355731225296442691.18577075098814232.01581027667984182.49011857707509864.74308300395256935.33596837944664065.45454545454545415.6916996047430835.86896634327464375.92885375494071195.92885375494071196.04743083003952546.40316205533596877.11462450592885357.2332015810276688.18181818181818178.89328063241106657x8-12509.17922565.17383000000022959.43994000000023502.838623743.65601000000023794.406254237.35498000000014808.31444999999996294.60595999999996858.13965000000017416.343758265.81151999999939013.97655999999929536.198239999999710095.85644999999900.588235294117647083.176470588235293944.05882352941176454.1176470588235294.1176470588235294.58823529411764674.82352941176470564.94117647058823554.94117647058823556.94117647058823557.29411764705882347.76470588235294118.2352941176470588x8-13295.48754999999983314.77417000000013586.832033608.81494000000024689.87207000000035331.156255347.69141000000045510.90332000000036388.93896000000047810.57031000000018869.74022999999949993.785159999999510851.3232411707.2744100.472440944881889760.826771653543307062.83464566929133843.77952755905511814.01499999999999974.0157480314960634.13385826771653524.37007874015748054.84251968503937045.66929133858267696.96850393700787417.08661417322834637.086614173228346310x10-14145.36376999999994206.63231999999974766.28369000000025038.52001999999995683.49608999999965734.23095999999995993.96093999999996743.03075999999967989.20604999999989227.757809999999211449.6562513230.44043000000113688.7968814066.19140999999914140.5927699999991414200.61.562.04333.35059760956175273.64.24.44000000000000045.466.848.279999999999999410.19999999999999915
Load (N)
Debonding length acomp (mm)
6x6-1(1)5736.444345779.125979999999801.1857707509881423DEN16043.18505999999986573.37694999999997102.91503999999997641.17235999999968187.53075999999968721.39160000000089205.94530999999921.89723320158102762.13438735177865622.84584980237154153.08300395256916993.79446640316205524.98023715415019735.6916996047430837x8-1(1)3743.65601000000023794.406254237.354980000000100.470588235294117641.7647058823529411DEN24237.35498000000014808.31444999999996294.60595999999996858.13965000000017416.343758265.81151999999939013.97655999999929536.198239999999710095.8564499999991.76470588235294112.47058823529411783.52941176470588225.17647058823529445.64705882352941215.64705882352941216.47058823529411787.76470588235294117.8823529411764718x8-1(1)5331.156255347.69141000000045510.903320000000300.708661417322834612.1259842519685042DEN35510.90332000000036388.93896000000047810.57031000000018869.74022999999949993.785159999999510851.3232411707.274412.12598425196850424.0157480314960635.55118110236220465.66929133858267697.08661417322834637.67716535433070837.795275590551181510x10-1(1)5683.49608999999965734.230959999999901.44DEN45993.96093999999996743.03075999999967989.20604999999989227.757809999999211449.6562513230.4404300000011.73306772908366542.164.084.55999999999999965.646.2410x10-1(3)13230.44043000000113688.7968814066.19140999999914140.592769999999141426.248.03999999999999919.249.6156x6-1-203469.11063499.58154000000013529.91699000000023559.50293000000013917.32227000000013958.901374507.27880999999985021.72753999999995548.64794999999965736.444345779.12597999999986043.18505999999986573.37694999999997102.91503999999997641.17235999999968187.53075999999968721.39160000000089205.945309999999201.18577075098814231.89723320158102762.13438735177865622.84584980237154153.08300395256916993.79446640316205524.98023715415019735.6916996047430837x8-1-202509.17922565.17383000000022959.43994000000023502.838623743.65601000000023794.406254237.35498000000014808.31444999999996294.60595999999996858.13965000000017416.343758265.81151999999939013.97655999999929536.198239999999710095.85644999999900.470588235294117641.76470588235294112.47058823529411783.52941176470588225.17647058823529445.64705882352941215.64705882352941216.47058823529411787.76470588235294117.8823529411764718x8-1-203295.48754999999983314.77417000000013586.832033608.81494000000024689.87207000000035331.156255347.69141000000045510.90332000000036388.93896000000047810.57031000000018869.74022999999949993.785159999999510851.3232411707.2744100.708661417322834612.12598425196850424.0157480314960635.55118110236220465.66929133858267697.08661417322834637.67716535433070837.795275590551181510x10-1-204145.36376999999994206.63231999999974766.28369000000025038.52001999999995683.49608999999965734.23095999999995993.96093999999996743.03075999999967989.20604999999989227.757809999999211449.6562513230.44043000000113688.7968814066.19140999999914140.5927699999991414201.441.73306772908366542.164.084.55999999999999965.646.248.03999999999999919.249.615
Load (N)
Debonding length am (mm)
From composite
end335260958.928571428571428810.7661290322580649.44881889763779456From
metal
end335260956.190476190476190711.3709677419354848.85826771653543336.24
Surfi-sculpt protrusion number
Debonding length (mm)
From composite
end335260950.581337737407101480.513625103220478940.4351775392237820.30305532617671344From
metal
end3352609510.86374896779521060.758051197357555660.51775392237819973
Surfi-sculpt protrusion number
Normalized debonding rate
6x6-1(1)3469.11063499.58154000000013529.91699000000023559.50293000000013917.32227000000013958.9013700.355731225296442691.18577075098814232.01581027667984182.49011857707509864.7430830039525693DEN13958.901374507.27880999999985021.72753999999995548.64794999999965736.444345779.12597999999986043.18505999999986573.37694999999997102.91503999999997641.17235999999968187.53075999999968721.39160000000089205.94530999999924.74308300395256935.33596837944664065.45454545454545415.6916996047430835.86896634327464377.11462450592885447.82608695652173928.18181818181818179.249011857707509810.19762845849802311.02766798418972313.16205533596837914.5849802371541497x8-1(1)2509.17922565.17383000000022959.439940000000200.588235294117647083.1764705882352939DEN22959.43994000000023502.838623743.65601000000023794.406254237.35498000000014808.31444999999996294.60595999999996858.13965000000017416.343758265.81151999999939013.97655999999929536.198239999999710095.8564499999993.176470588235293944.05882352941176454.58823529411764675.88235294117647017.05882352941176458.35294117647058710.11764705882352910.58823529411764912.58823529411764913.76470588235294215.52941176470588216.1176470588235298x8-1(1)3295.48754999999983314.77417000000013586.832033608.814940000000200.472440944881889760.826771653543307062.8346456692913384DEN33608.81494000000024689.87207000000035331.156255347.69141000000045510.90332000000036388.93896000000047810.57031000000018869.74022999999949993.785159999999510851.3232411707.274412.83464566929133843.77952755905511814.01499999999999974.72440944881889726.25984251968503898.385826771653544410.39370078740157411.33858267716535414.0551181102362214.76377952755905614.881889763779528DEN45993.96093999999996743.03075999999967989.20604999999989227.757809999999211449.6562513230.4404300000015.08366533864541835.768.2800000000000011911.0412.2410x10-1(3)13230.44043000000113688.7968814066.19140999999914140.5927699999991414212.2414.87999999999999917.5219.7999999999999973010x10-1(1)4145.36376999999994206.63231999999974766.28369000000025038.52001999999995683.49608999999965734.23095999999995993.960939999999900.61.562.0434.43999999999999955.08366533864541836x6-1-303469.11063499.58154000000013529.91699000000023559.50293000000013917.32227000000013958.901374507.27880999999985021.72753999999995548.64794999999965736.444345779.12597999999986043.18505999999986573.37694999999997102.91503999999997641.17235999999968187.53075999999968721.39160000000089205.945309999999200.355731225296442691.18577075098814232.01581027667984182.49011857707509864.74308300395256935.33596837944664065.45454545454545415.6916996047430835.86896634327464377.11462450592885447.82608695652173928.18181818181818179.249011857707509810.19762845849802311.02766798418972313.16205533596837914.5849802371541497x8-1-302509.17922565.17383000000022959.43994000000023502.838623743.65601000000023794.406254237.35498000000014808.31444999999996294.60595999999996858.13965000000017416.343758265.81151999999939013.97655999999929536.198239999999710095.85644999999900.588235294117647083.176470588235293944.05882352941176454.58823529411764675.88235294117647017.05882352941176458.35294117647058710.11764705882352910.58823529411764912.58823529411764913.76470588235294215.52941176470588216.1176470588235298x8-1-303295.48754999999983314.77417000000013586.832033608.81494000000024689.87207000000035331.156255347.69141000000045510.90332000000036388.93896000000047810.57031000000018869.74022999999949993.785159999999510851.3232411707.2744100.472440944881889760.826771653543307062.83464566929133843.77952755905511814.01499999999999974.72440944881889726.25984251968503898.385826771653544410.39370078740157411.33858267716535414.0551181102362214.76377952755905614.88188976377952810x10-1-304145.36376999999994206.63231999999974766.28369000000025038.52001999999995683.49608999999965734.23095999999995993.96093999999996743.03075999999967989.20604999999989227.757809999999211449.6562513230.44043000000113688.7968814066.19140999999914140.5927699999991414200.61.562.0434.43999999999999955.08366533864541835.768.2800000000000011911.0412.2414.87999999999999917.5219.79999999999999730
Load (N)
Total debonding length (mm)
3352609510.955050237969328420.882601797990481110.52564780539397149
Surfi-sculpt protrusion number
Normalized debonding rate
DEN13958.901374507.27880999999985021.72753999999995548.64794999999965736.444345779.12597999999986043.18505999999986573.37694999999997102.91503999999997641.17235999999968187.53075999999968721.39160000000089205.94530999999924.74308300395256935.33596837944664065.45454545454545415.6916996047430835.86896634327464375.92885375494071195.92885375494071196.04743083003952546.40316205533596877.11462450592885357.2332015810276688.18181818181818178.89328063241106656x6-1(1)3469.11063499.58154000000013529.91699000000023559.50293000000013917.32227000000013958.9013700.355731225296442691.18577075098814232.01581027667984182.49011857707509864.74308300395256937x8-1(1)2509.17922565.17383000000022959.439940000000200.588235294117647083.1764705882352939DEN23502.838623743.65601000000023794.406254237.35498000000014808.31444999999996294.60595999999996858.13965000000017416.343758265.81151999999939013.97655999999929536.198239999999710095.85644999999944.05882352941176454.1176470588235294.1176470588235294.58823529411764674.82352941176470564.94117647058823554.94117647058823556.94117647058823557.29411764705882347.76470588235294118.2352941176470588x8-1(1)3295.48754999999983314.77417000000013586.832033608.814940000000200.472440944881889760.826771653543307062.8346456692913384DEN33608.81494000000024689.87207000000035331.156255347.69141000000045510.90332000000036388.93896000000047810.57031000000018869.74022999999949993.785159999999510851.3232411707.274412.83464566929133843.77952755905511814.01499999999999974.0157480314960634.13385826771653524.37007874015748054.84251968503937045.66929133858267696.96850393700787417.08661417322834637.086614173228346310x10-1(1)4145.36376999999994206.63231999999974766.28369000000025038.52001999999995683.49608999999965734.23095999999995993.960939999999900.61.562.04333.3505976095617527DEN45993.96093999999996743.03075999999967989.20604999999989227.757809999999211449.6562513230.4404300000013.35059760956175273.64.24.44000000000000045.4610x10-1(3)13230.44043000000113688.7968814066.19140999999914140.5927699999991414266.848.279999999999999410.199999999999999156x6-1-103469.11063499.58154000000013529.91699000000023559.50293000000013917.32227000000013958.901374507.27880999999985021.72753999999995548.64794999999965736.444345779.12597999999986043.18505999999986573.37694999999997102.91503999999997641.17235999999968187.53075999999968721.39160000000089205.945309999999200.355731225296442691.18577075098814232.01581027667984182.49011857707509864.74308300395256935.33596837944664065.45454545454545415.6916996047430835.86896634327464375.92885375494071195.92885375494071196.04743083003952546.40316205533596877.11462450592885357.2332015810276688.18181818181818178.89328063241106657x8-12509.17922565.17383000000022959.43994000000023502.838623743.65601000000023794.406254237.35498000000014808.31444999999996294.60595999999996858.13965000000017416.343758265.81151999999939013.97655999999929536.198239999999710095.85644999999900.588235294117647083.176470588235293944.05882352941176454.1176470588235294.1176470588235294.58823529411764674.82352941176470564.94117647058823554.94117647058823556.94117647058823557.29411764705882347.76470588235294118.2352941176470588x8-13295.48754999999983314.77417000000013586.832033608.81494000000024689.87207000000035331.156255347.69141000000045510.90332000000036388.93896000000047810.57031000000018869.74022999999949993.785159999999510851.3232411707.2744100.472440944881889760.826771653543307062.83464566929133843.77952755905511814.01499999999999974.0157480314960634.13385826771653524.37007874015748054.84251968503937045.66929133858267696.96850393700787417.08661417322834637.086614173228346310x10-14145.36376999999994206.63231999999974766.28369000000025038.52001999999995683.49608999999965734.23095999999995993.96093999999996743.03075999999967989.20604999999989227.757809999999211449.6562513230.44043000000113688.7968814066.19140999999914140.5927699999991414200.61.562.04333.35059760956175273.64.24.44000000000000045.466.848.279999999999999410.19999999999999915
Load (N)
Debonding length acomp (mm)
6x6-1(1)5736.444345779.125979999999801.1857707509881423DEN16043.18505999999986573.37694999999997102.91503999999997641.17235999999968187.53075999999968721.39160000000089205.94530999999921.89723320158102762.13438735177865622.84584980237154153.08300395256916993.79446640316205524.98023715415019735.6916996047430837x8-1(1)3743.65601000000023794.406254237.354980000000100.470588235294117641.7647058823529411DEN24237.35498000000014808.31444999999996294.60595999999996858.13965000000017416.343758265.81151999999939013.97655999999929536.198239999999710095.8564499999991.76470588235294112.47058823529411783.52941176470588225.17647058823529445.64705882352941215.64705882352941216.47058823529411787.76470588235294117.8823529411764718x8-1(1)5331.156255347.69141000000045510.903320000000300.708661417322834612.1259842519685042DEN35510.90332000000036388.93896000000047810.57031000000018869.74022999999949993.785159999999510851.3232411707.274412.12598425196850424.0157480314960635.55118110236220465.66929133858267697.08661417322834637.67716535433070837.795275590551181510x10-1(1)5683.49608999999965734.230959999999901.44DEN45993.96093999999996743.03075999999967989.20604999999989227.757809999999211449.6562513230.4404300000011.73306772908366542.164.084.55999999999999965.646.2410x10-1(3)13230.44043000000113688.7968814066.19140999999914140.592769999999141426.248.03999999999999919.249.6156x6-1-203469.11063499.58154000000013529.91699000000023559.50293000000013917.32227000000013958.901374507.27880999999985021.72753999999995548.64794999999965736.444345779.12597999999986043.18505999999986573.37694999999997102.91503999999997641.17235999999968187.53075999999968721.39160000000089205.945309999999201.18577075098814231.89723320158102762.13438735177865622.84584980237154153.08300395256916993.79446640316205524.98023715415019735.6916996047430837x8-1-202509.17922565.17383000000022959.43994000000023502.838623743.65601000000023794.406254237.35498000000014808.31444999999996294.60595999999996858.13965000000017416.343758265.81151999999939013.97655999999929536.198239999999710095.85644999999900.470588235294117641.76470588235294112.47058823529411783.52941176470588225.17647058823529445.64705882352941215.64705882352941216.47058823529411787.76470588235294117.8823529411764718x8-1-203295.48754999999983314.77417000000013586.832033608.81494000000024689.87207000000035331.156255347.69141000000045510.90332000000036388.93896000000047810.57031000000018869.74022999999949993.785159999999510851.3232411707.2744100.708661417322834612.12598425196850424.0157480314960635.55118110236220465.66929133858267697.08661417322834637.67716535433070837.795275590551181510x10-1-204145.36376999999994206.63231999999974766.28369000000025038.52001999999995683.49608999999965734.23095999999995993.96093999999996743.03075999999967989.20604999999989227.757809999999211449.6562513230.44043000000113688.7968814066.19140999999914140.5927699999991414201.441.73306772908366542.164.084.55999999999999965.646.248.03999999999999919.249.615
Load (N)
Debonding length am (mm)
From composite
end335260958.928571428571428810.7661290322580649.44881889763779456From
metal
end335260956.190476190476190711.3709677419354848.85826771653543336.24
Surfi-sculpt protrusion number
Debonding length (mm)
From composite
end335260950.581337737407101480.513625103220478940.4351775392237820.30305532617671344From
metal
end3352609510.86374896779521060.758051197357555660.51775392237819973
Surfi-sculpt protrusion number
Normalized debonding rate
1
30
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0
2000
4000
6000
8000
10000
12000
14000
Damage Initiation
Transition Point
Load (N)
Displacement (mm)
R
DEN1
DEN2
DEN3
DEN4
0
20
40
60
80
100
0
1000
2000
3000
4000
5000
DEN4 (22.9%)
DEN3 (16%)
DEN2 (10%)
DEN1(4%)
R
Initiation Damage load (N)
Surfi-sculpt protrusion number
0
20
40
60
80
100
0
2000
4000
6000
8000
10000
12000
14000
Ultimate failure load (N)
Surfi-sculpt protrusion number
R
DEN1(114%)
DEN2 (150%)
DEN3 (171%)
DEN4 (212.03%)
0
10
20
30
40
50
60
70
80
90
100
0
20
40
60
80
100
120
140
160
Ultimate failure load increase rate(N)
Surfi-sculpt protrusion number
DEN2 (123N)
DEN1(150N)
DEN3 (122N)
DEN4 (96 N)
0
5
10
15
20
25
0.0
0.2
0.4
0.6
0.8
1.0
Bonding line deflection
DEN1
DEN2
DEN3
DEN4
X (mm)
Normalized joint Deflection (absolute value)
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
Residual' Height (mm)
0.0
0.1
0.2
0.3
0
5000
10000
Nonlinear behavior
Load (N)
Displacement (mm)
R
DEN1
DEN2
DEN3
DEN4
DEN1
DEN2
DEN3
DEN4
0
20
40
60
80
100
0.0
0.1
0.2
0.3
0.4
DEN4
(Maximum load point)
(714.25%)
Extension (mm)
Surfi-sulpt density
R
DEN1
(223%)
DEN2
(287%)
DEN3
(288%)
DEN4
(Transition point)
(423%)
0
20
40
60
80
100
0
1000
2000
3000
4000
DEN4
(Maximum load point)
(3339%)
Absorbed engery (N.mm)
Surfi-sulpt density
R
DEN1
(673%)
DEN2
(957%)
DEN3
(1012%)
DEN4
(Transition point)
(1616%)
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0
2000
4000
6000
8000
10000
12000
14000
Damage Initiation
Transition Point
Load (N)
Displacement (mm)
R
DEN1
DEN2
DEN3
DEN4
0
20
40
60
80
100
0
1000
2000
3000
4000
5000
DEN4 (22.9%)
DEN3 (16%)
DEN2 (10%)
DEN1(4%)
R
Initiation Damage load (N)
Surfi-sculpt protrusion number
0
20
40
60
80
100
0
2000
4000
6000
8000
10000
12000
14000
Ultimate failure load (N)
Surfi-sculpt protrusion number
R
DEN1(114%)
DEN2 (150%)
DEN3 (171%)
DEN4 (212.03%)
0
10
20
30
40
50
60
70
80
90
100
0
20
40
60
80
100
120
140
160
Ultimate failure load increase rate(N)
Surfi-sculpt protrusion number
DEN2 (123N)
DEN1(150N)
DEN3 (122N)
DEN4 (96 N)
0
5
10
15
20
25
0.0
0.2
0.4
0.6
0.8
1.0
Bonding line deflection
DEN1
DEN2
DEN3
DEN4
X (mm)
Normalized joint Deflection (absolute value)
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
Residual' Height (mm)
0
20
40
60
80
100
0.0
0.1
0.2
0.3
0.4
DEN4
(Maximum load point)
(714.25%)
Extension (mm)
Surfi-sulpt density
R
DEN1
(223%)
DEN2
(287%)
DEN3
(288%)
DEN4
(Transition point)
(423%)
0
20
40
60
80
100
0
1000
2000
3000
4000
DEN4
(Maximum load point)
(3339%)
Absorbed engery (N.mm)
Surfi-sulpt density
R
DEN1
(673%)
DEN2
(957%)
DEN3
(1012%)
DEN4
(Transition point)
(1616%)