Journal Pre-proofs

Study on shear behavior of riveted lap joints of aircraft fuselage with different
hole diameters and squeeze forces

Ming Li, Wei Tian, Junshan Hu, Changrui Wang, Wenhe Liao

PII: S1350-6307(21)00359-9
DOI: https://doi.org/10.1016/j.engfailanal.2021.105499
Reference: EFA 105499

To appear in: Engineering Failure Analysis

Received Date: 13 February 2021

Revised Date: 2 May 2021
Accepted Date: 17 May 2021

Please cite this article as: Li, M., Tian, W., Hu, J., Wang, C., Liao, W., Study on shear behavior of riveted lap
joints of aircraft fuselage with different hole diameters and squeeze forces, Engineering Failure Analysis (2021),
doi: https://doi.org/10.1016/j.engfailanal.2021.105499

This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover
page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version
will undergo additional copyediting, typesetting and review before it is published in its final form, but we are
providing this version to give early visibility of the article. Please note that, during the production process, errors
may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

© 2021 Elsevier Ltd. All rights reserved.

 Study on shear behavior of riveted lap joints of aircraft fuselage with

different hole diameters and squeeze forces

Ming Li , Wei Tian *, Junshan Hu , Changrui Wang , Wenhe Liao

College of Mechanical and Electronical Engineering, Nanjing University of Aeronautics and Astronautics, Nanjing,
210016, China
*Corresponding author: Wei Tian (Email: [email protected]).

Abstract

Riveting is the most important method to connect metal sheets and is widely used in the joining of
aircraft components. In the present research, the effect of different hole diameters and squeeze forces on
the shear behavior of riveted lap joints is investigated. The variation of shear load with hole diameter
under different squeeze forces was obtained by shear tests. Besides, the fracture mode and microstructure
of the rivet shank were characterized by SEM, and the formation process of brittleness and plastic fracture
was discussed. A 2D axisymmetric finite element model was established with the help of ABAQUS
commercial finite element software and the validity of the finite element models was verified by
experiments. To investigate the influence of interference distribution on the mechanical behavior of
riveted lap joints, the interference distribution and material flow characteristics during the riveting
process were analyzed in detail by the finite element methods. The results showed that all the specimens
presented a mixed fracture mode of brittleness and plasticity at rivet shank. The area of the brittle fracture
region was much smaller than the plastic fracture region. Also, the shear load was enhanced when the
hole diameter increased from 4.82 mm to 5.10 mm. The maximum shear load increased by 17.77% and
17.24% under 15 kN and 23 kN squeeze forces respectively. In addition, compared with the squeeze
force, increasing the hole diameter could effectively raise the shear load of the riveted lap joints.
Keywords: material flow characteristics, hole diameter, mixed fracture modes, shear load

1. Introduction

Aircraft assembly is an important part of modern aircraft manufacturing. The assembly between
components inevitably employs various connecting techniques, including riveting [1-6], bolting [7-12],
welding [13,14], bonding [15,16], etc. Among them, riveting is still the main connection method,
especially for metal components, because riveting has the characteristics of simple process, good sealing,
and reliable strength [17]. However, hole diameter tolerance and the squeeze force during the riveting
process can lead to the change of interference-fit dimensions, which in turn have an impact on the
mechanical behavior of the riveted lap joints [4,18]. In order to ensure the safe service of aircraft, a
comprehensive understanding of the mechanical response of riveted lap joints is essential.
Many researchers established the relationship between squeeze forces and fatigue life [19-20]. The
squeeze force, which is closely related to the driven rivet head dimension, is a fundamental control factor
in the riveting process [19-22]. Based on the plastic deformation exponential hardening law and the
principle of constant volume assumption, an analytical model of the relationship between the squeeze
force and the driven rivet head dimensions is established [17-19], and the riveting quality can be further
controlled by the squeeze force, which also provides a method for investigating the mechanical behavior
of riveted aircraft structures [21,23]. Exhaustive fatigue tests show that the squeeze force affects the
fatigue life of riveted lap joints, which is attributed to the residual compressive stress around the hole

1
leading to crack initiation position away from the edge of the hole [24-26]. Changes in production
variables may cause variations in the relationship between squeeze force and fatigue life of riveted lap
joints [20,27-33]. For thin sheets, increasing the squeeze force can significantly prolong the fatigue life
of riveted specimens. However, with the increase of the sheet thickness, the benefit of large squeeze force
to fatigue life gradually decreases [33].
Another advantage of large squeeze force is the ability to increase the interference-fit size. By changing
the shape of the rivet head to change the interference fit state of the rivet/hole, the fatigue life of the
riveted lap joints can be improved [28-30,34]. Fatigue tests on universal head rivets and countersunk
head rivets showed that the fatigue life of universal head rivets was longer under the same rivet diameter.
This is because the interference-fit size is greater and interference distribution is more uniform in the
sheet's thickness direction compared with countersunk head rivets [33]. Fatigue life is not only related to
the amount of interference-fit, but also related to the uniformity of the interference distribution. To
improve the uniformity of interference distribution, electromagnetic riveting was widely used in the
riveting of riveted joints [35-37]. A comparative study of conventional pressure riveting and
electromagnetic riveting found that the riveting method had little effect on the shear strength of riveted
lap joints. Since the high interference-fit size delayed the formation of the gap between rivet and hole
wall, the fatigue life of the specimen could be increased by 1 ~ 3 times by electromagnetic riveting [37].
However, some references [24,38] showed that the fatigue life gain weakened with the increase of the
interference-fit size, and even reduced the fatigue life, which was ascribed to the fretting wear on the
sheet contact surface [39]. Fretting is not the only cause of crack initiation during cyclic loading. The
stress concentration at rivet/hole interface of fuselage lap joints may be the main factor causing fatigue
damage [40]. The mismatch of rivet holes causes the asymmetric deformation of rivet shank during the
riveting process, and the presence of drill shavings between the stacked sheets changes the interference
distribution, which ultimately leads to the uneven distribution of residual stress [41,42].
From the perspective of hole expansion, both the squeeze force and the hole diameter tolerance will
affect the final amount of interference-fit and the stress around the hole [1,31,39]. To this end, many
scholars have used analytical models [43], numerical simulations [44,45], and experiments [39] to study
the influence of hole size tolerances on residual stress distribution and fatigue life. The results showed
that the hole diameter had an important influence on the fatigue strength of the riveted lap joints [39],
and the selection of the initial hole diameter is related to the squeeze force, rivet diameter, and rivet
length [21]. In order to drive the maximum benefit of riveted lap joints, it is particularly important to
select the appropriate hole diameter.
Previous research mainly focused on the fatigue behavior of riveted lap joints. However, there is still
a scarcity of studies on the effect of hole diameters and squeeze forces on the shear behavior of riveted
lap joints. Besides, the discussion on the shear failure mechanisms hasn’t got enough attention. Especially
in the design of aircraft structure, ensuring a good static strength threshold is an important guarantee for
improving aircraft performance and provides basic parameters for fatigue design.
Therefore, the primary objective of the current study is to investigate the effect of different hole
diameters and squeeze forces on the shear behavior of riveted lap joints. The effect of hole diameters on
the interference distribution of the riveted lap joints was investigated through finite element simulations
and the material flow characteristics during the riveting process were further analyzed. The maximum
shear load of the riveted lap joints was obtained by shear tests, and the shear behavior of the riveted lap
joints under quasi-static loading was discussed. Meanwhile, the fracture mode and microstructure of the
rivet were characterized by SEM. Finally, the shear failure mechanism of riveted lap joints was analyzed
to ensure their safety in service and to provide guidance for engineering applications.
2
2. Materials and methods

2.1. Specimen preparation

The specimen joints were made of 2024-T3 aluminum alloy sheets and MS20470 AD6-7 aluminum
alloy rivet. The thickness of aluminum alloy sheets was 2 mm. The nominal diameter of the rivet was
4.76 mm and the length of the rivet shank was 11.1 mm. The mechanical properties of 2117-T4 and
2024-T3 aluminum alloys are shown in Table 1 [46,47]. To eliminate the influence of fixture on the stress
condition in the overlap region, the geometry of the sheet was 174 mm × 24 mm, in which the overlap
region size was 24 mm × 24 mm, and the clamped region size was 50 mm × 24 mm. The sheets were cut
by a wire cutting machine. Metal gaskets with the same thickness as the sheets were installed during the
shear tests to avoid the deviation of the specimen's axis. The shear specimens were designed as a single
rivet lap joints structure, as shown in Fig. 1, where the end distance (m) was 12 mm and the edge margin
(n) was 12 mm. The length of the free end was 50 times the sheet's thickness (t) to ensure that the stresses
during the riveting process were independent of the dimension of the model.

Fig. 1. Specimen dimensions and geometry (unit: mm).

Table 1. Mechanical properties of 2117-T4 and 2024-T3 Al alloy.

E σs C Kh Hardening
Material  m εtrue
(GPa) (MPa) (MPa) (MPa) type
544 0.23 0.02 ≤ εtrue ≤ 0.10
2117-T4 71.7 172 0.33
551 0.15 0.10 ≤ εtrue ≤ 1.0
- exponential
676 0.14 εs ≤ εtrue ≤ 0.02
2024-T3 72.4 310 0.33 745 0.164 0.02 ≤ εtrue ≤ 0.10
- - 1034 0.10 ≤ εtrue ≤ 1.0 linear
During the drilling process, a special fixture was used to clamp the lap sheets to ensure the accuracy
of hole position. The drilling experiment platform is shown in Fig. 2. In order to ensure the quality of
hole making, cemented carbide tri-point drill was used. The drilling experiment was completed in Dexi
vertical machining center, in which the spindle speed was 3000 rpm and the feed rate was 120 mm/min.

3
 Fig. 2. Drilling experiment platform.

2.2. Riveting experiment

The riveting experiment was carried out on a pressure riveting experiment platform, which has two
control modes, namely displacement control and force control. In the current experiment, force control
was used since the displacement control was easily affected by the mechanical structure [20,34].
Therefore, the squeeze force Fsq was selected as the input parameter to obtain the required dimensions of
the driven rivet head. In the present research, the squeeze force was set at two levels of 15 kN and 23 kN,
and the hole diameter was set at five levels of, 4.82 mm, 4.90 mm, 4.94 mm, 5.00 mm, and 5.10 mm.
Each test configuration was repeated with five specimens. The detailed experimental setup is shown in
Table 2. In order to conveniently describe the experiment configuration of riveted lap joints, the
following terms were used: D4.82-F15-NO.X denotes the diameter of the hole is 4.82 mm and squeeze
force is 15 kN. NO.X denotes the specimen label, the value of X was from 1 to 5.
Table 2. The test configuration of riveted lap joints and their corresponding specimen groups.
Squeeze force (Fsq)
Test configurations
15 kN 23 kN
4.82mm D4.82-F15-NO.X D4.82-F23-NO.X
4.90 mm D4.90-F15-NO.X D4.90-F23-NO.X
Diameter of hole
4.94 mm D4.94-F15-NO.X D4.94-F23-NO.X
(D)
5.00 mm D5.00-F15-NO.X D5.00-F23-NO.X
5.10 mm D5.10-F15-NO.X D5.10-F23-NO.X

2.3. Shear behavior tests

Quasi-static shear tests were carried out on a UTM5504Y electronic universal testing machine
produced by Sansi Zongheng Technology Co., Ltd, which is capable of generating a maximum tensile
load of 50 kN. The whole tests were carried out at room temperature, the tensile rate of the uniaxial
tensile tests was set to 2 mm/min, and the loading was terminated after the specimen was broken. The
fracture morphology after the shear tests was characterized by FEI Quanta 200 scanning electron
microscope (SEM). All specimens were cleaned with an ultrasonic cleaner before observation.

3. Finite element analysis

3.1. Finite element models

Considering the symmetry and the calculation time of the model, a 2D axisymmetric finite element
model was established, as shown in Fig. 3. The riveted lap joints were composed of five parts. The rigid
set and rigid bucking bar were set as rigid bodies. The two sheets and one rivet were deformable bodies.

4
The detailed material parameters are shown in Table 1. The enhanced hourglass control and an Arbitrary
Lagrange Eulerian (ALE) adaptive mesh technology were used in the simulation. The type of surface-to-
surface explicit contact was chosen for analysis, with the friction coefficient of 0.2 for each contact
interface [41,48,49]. Due to the nonlinear characteristics of the riveting process, the Nlgeom option was
selected [17,50]. The current model includes five contact pairs, as shown by the purple line in Fig. 3.
During the riveting simulation, the axis of the rivet was constrained to displacement and rotation along
the x the z axes. The rigid set was constrained to all degrees of freedom. The rigid bucking bar was
constrained to all degrees of freedom, except for the displacement of y-direction. Meanwhile, the rigid
bucking bar was subjected to a y-direction squeeze force. During the pressure maintenance stage, the
displacement of the rigid bucking bar was restricted in the y-direction, and squeeze force was set to
"deactivated". In order to complete the riveting unloading, the y-direction constraint of the rigid set was
canceled. Also, the rigid set and the rigid bucking bar were applied with an opposite displacement of y-
direction.

Fig. 3. 2D axisymmetric finite element mesh model and path description.

3.2. Verification of the finite element models

In the present research, the validity of the finite element simulation is confirmed by measuring the
dimensions of the driven rivet head, because the dimensions of the driven rivet head can be measured
intuitively [51-53]. The final deformed of driven rivet head diameter and driven rivet head height
measured by the experiment were compared with the finite element simulation results for the specified
range of squeeze forces, as shown in Fig. 4. There is a nonlinear relationship between driven rivet head
dimensions and squeezes forces. As the squeeze force increases from 13 kN to 25 kN, the diameter of
the driven rivet head increases by 29.2%, and the height of the driven rivet head decrease by 42.8%. The
increased amplitude of driven rivet diameter decreases when the squeeze force increases to a certain
value, as the riveting process is a large plastic deformation process accompanied by strain hardening, and
the material's ability to resist deformation is enhanced. The error comparisons between the experiment
measurement values and the finite element simulation values under different squeeze forces are shown
in Table 3, where Hdh and Ddh represent the height and diameter of the driven rivet head respectively.
The maximum errors of driven rivet head diameter and driven rivet head height are 2.13% and 2.29%,
respectively. In conclusion, the experimental measurement results are in good agreement with the finite
element simulation results.

5
 Fig. 4. Comparison of experimental results and finite element simulation results:
(a) Height of driven rivet head, (b) Diameter of driven rivet head.
Table 3. Comparison of finite element method and experimental method for dimension errors of driven rivet head.
Squeeze force (Fsq)
Experimental variables
13 kN 14 kN 15 kN 16 kN 17 kN 18 kN 19 kN
Ddh 1.25 0.78 0.89 1.40 2.13 1.85 1.66
Hdh 0.22 0.29 -1.02 -1.11 -1.58 -1.53 -1.99
Driven rivet head
20 kN 21 kN 22 kN 23 kN 24 kN 25 kN
dimension
Ddh 1.49 1.94 0.72 1.20 1.50 0.94
Hdh -2.29 -0.76 -0.10 -0.89 -0.31 -1.64

4. Results and discussions

4.1. Influence of hole diameter on interference distribution and material flow characteristics

The amount of interference is an important parameter to evaluate the quality of riveting, and it has an
important influence on the mechanical behavior of riveted lap joints [2,4]. The interference amount
during the riveting process can be expressed as absolute interference amount and relative interference
amount. The absolute interference amount is twice the radial displacement of the hole wall. The relative
interference amount is the difference between the diameter of the rivet shank after riveting and the initial
hole diameter divided by the initial hole diameter. Since the diameter of the hole after riveting is equal
to the diameter of the rivet shank after riveting, the relative interference amount is shown in Eqs. (1) and
(2), respectively.
2 H
I  100% (1)
D
d D
I  100% (2)
D
Where H is the radial displacement of the hole wall after riveting, D is the initial hole diameter and d is
the diameter of the rivet shank after riveting.
The degree of hole expansion of different hole diameters under 15 kN and 23 kN squeeze forces are
presented in Fig. 5. The left and right sides of the vertical line indicate the direction of the thickness of
the outer sheet and inner sheet respectively. From Fig. 5, the expansion degree of the inner sheet is greater
than that of the outer sheet under each hole diameter. With the increase of squeeze force, the difference
in expansion degree of the hole wall of the inner and outer sheets increases. There is little difference in
the expansion degree of the outer sheet along the thickness direction under five-hole diameters. However,

6
the difference of hole expansion degree of inner sheet increases, especially near the driven rivet head
side. For the squeeze force of 15 kN or 23 kN, the expansion degree of the hole wall on the side of the
driven rivet head is greater than that on the side of the manufactured rivet head. The above phenomena
are related to the material flow characteristics during riveting.

Fig. 5. Radial expansion of rivet holes with different hole diameters at various squeeze forces.
Generally, the rivet and hole belong to clearance fit in the initial state, and the diameter of the hole is
larger than the diameter of the rivet shank. In the early stage of the riveting process, the whole rivet shank
is in the free upsetting stage. At this time, the rivet shank does not in contact with the hole wall, and the
material flows mainly in the axial direction. When the rivet shank is in contact with the hole wall, the
radial expansion of the rivet shank is restricted, and the axial material flow of the rivet shank tends to be
saturated. When the squeeze force is exerted on the rivet shank, only a small amount of material flows
into the hole, and the remaining material forms the driven rivet head. The axial paths of rivet shank at
different radial positions are shown in Fig. 3, where the oy and ox represent rivet axial direction and rivet
radial direction respectively. The radial positions of path 1, path 2, path 3 and path 4 were set at x1 =
2.380 mm, x2 = 1.785 mm, x3 = 1.190 mm and x4 = 0.595 mm respectively. The radial displacement
changes of different paths are shown in Fig. 6. The left side of the vertical line represents inside the hole
and the right side represents outside the hole. As Fig. 6 shows, the radial displacement of the rivet shank
inside the hole does not change much, and the increase of radial displacement mainly occurs outside the
hole. The radial displacement of the rivet material outside the hole firstly increases and then decreases,
which is caused by the uneven distribution of friction during the forming process of the driven rivet head.
For the same path, the radial displacement produced by 23 kN squeeze force is greater than that of 15 kN
squeeze force. The radial displacement values of different paths are significantly different, the radial
displacement value of path 1 is the largest, and the radial displacement value of path 4 is the smallest.
The mutation points of radial displacement in paths 2, 3, and 4 are earlier compared with path 1. Therefore,
it is more appropriate to describe the radial flow with the increase of the radial displacement of the
material inside the rivet shank, which is consistent with the conclusions of previous studies [22]. It is
revealed that the radial deformation of rivet shank increases with the increase of hole diameter, especially
when the hole diameter is 5.00 mm and 5.10 mm (Fig. 6(g)~ Fig. 6(j)). The main reason is that when the
gap between the rivet and the hole increases, the reaction force of the hole wall on the rivet decreases,
and the rivet shank is easier to be upset.

7
8
 Fig. 6. The radial displacement changes with the axial position:
(a)D4.82-F15, (b) D4.82-F23, (c) D4.90-F23, (d) D4.90-F23, (e) D4.94-F23, (f) D4.94-F23, (g) D5.00-F23, (h)
D5.00-F23, (i) D5.10-F23, (j) D5.10-F23.
In the present research, the radial displacement value of path 2 is selected for the study. The radial
displacement changes with the axial position for different hole diameters, as shown in Fig. 7. With the
increase of squeeze force, the position of mutation of radial displacement will be later. When the squeeze
force is 15 kN, the mutation point is located at the axial position of 5 mm, as shown in Fig. 7(a). However,
when the squeeze force is 23 kN, the mutation point is located at the axial position of 6 mm, as shown in
Fig. 7(b). In other words, when the squeeze force increases, more material will flow into the hole and the
expansion of the hole wall will be greater. This is also consistent with the results of the interference-fit
analysis. As the increase of the hole diameter, the amount of material flowing into the hole increases
slightly.

Fig. 7. The radial displacement changes with the axial position under different diameters of path 2:
(a)15 kN, (b)23 kN.

4.2. Effect of hole diameter on shear behavior of riveted lap joints

The results of the shear tests are shown in Fig. 8. From the corresponding shear load-displacement
curve, the peak loads of specimens with different hole diameters can be obtained. As shown in Fig. 8,
the changing trend of the shear load-displacement curve under five-hole diameters is basically the same.
In the stage of loading, there is a 0.1mm parallel straight section, which is caused by the clamping slip
of the riveted lap joints, and it does not affect the final shear load. Then, it enters the elastic deformation
stage. In this stage, the rivet begins to deform with the increase of the tensile force until it finally breaks.
This is mainly because the interference-fit size between the rivet and the hole becomes smaller when the
9
squeeze force decreases, the rivet tends to tilt in the hole under the action of tensile force. When the rivet
shank breaks, it will produce a larger displacement. On the contrary, with the increase of the squeeze
force, the diameter of the rivet shank increases, and the hole wall produces a reaction force on the rivet
shank, and the rivet is not easy to tilt in the hole. In this case, the contact area between the rivet shank
and the hole wall will increase, so the bearing capacity will be stronger.

10
 Fig. 8. Relationship between shear load and displacement:
(a)D4.82-F15, (b) D4.82-F23, (c) D4.90-F15, (d) D4.90-F23, (e) D4.94-F15, (f) D4.94-F23, (g) D5.00-F15, (h)
D5.00-F23, (i) D5.10-F15, (j) D5.10-F23.
In order to further illustrate the influence of hole diameter on the bearing capacity of riveted lap joints,
a histogram of the influence of hole diameter on the shear load of riveted lap joints under 15 kN and
23kN squeeze forces was obtained using the average value of the shear load under the same parameters,
as shown in Fig. 9. The results show that the influence of the hole diameter on the shear load under 15
kN and 23 kN squeeze forces is the same, that is, the shear load of riveted lap joints increases with the
increase of hole diameter. It is worth noting that the deformation degree of the hole wall decreases as the
hole diameter increases from 4.82mm to 5.10mm (see Fig. 5 and Fig. 7). Also, the rivet shank on the
outside of the sheet has more material flowing into the hole, and a larger rivet shank diameter can be
obtained. The bearing strength of the riveted lap joints is usually determined by the diameter of the rivet
shank. The larger the diameter of the rivet shank, the greater the tensile force required for the failure of
the specimen, indicating that the shear load of the riveted lap joints is enhanced with the increase of the
hole diameter. On the other hand, as Fig. 9 reveals that the shear load increases with the increase of
squeeze force. Under five-hole diameters, the maximum increase of shear load is 2.1%. Obviously,
compared with raising the squeeze force, increasing the hole diameter can effectively enhance the shear
load of the riveted lap joints, which is contrary to Zeng's conclusion [39]. The main reason is that the
fatigue process is significantly different from the static loading process, and the factors affecting the
fatigue process are more complicated. Only from the perspective of improving the static strength of the
specimens, increasing the hole diameter is beneficial to improve the shear resistance of the specimens.
11
 Fig. 9. Mean value of shear load under different hole diameters.
Fig. 10 shows the effect of five-hole diameters on the load-bearing capacity of the specimens. As
shown in Fig. 10, under the condition of 15 kN squeeze force, the shear load of the specimens with
diameters of 4.90 mm, 4.94 mm, 5.00 mm, and 5.10 mm increase by 3.51%, 6.52%, 9.99%, and 17.77%
respectively, compared with the specimen with a diameter of 4.82 mm. When the squeeze force is
increased to 23 kN, the shear load of the specimens with diameters of 4.90 mm, 4.94 mm, 5.00 mm, and
5.10 mm increase by 4.47%, 7.04%, 10.59%, and 17.24% respectively, compared with the specimens
with a diameter of 4.82 mm. Except for the specimen with a diameter of 5.10 mm, the increased amplitude
of shear load under the 23 kN squeeze force is greater than that under the 15 kN squeeze force.

Fig. 10. The percentage increase of shear load under different hole diameters.

4.3. Shear fracture analysis

Fig. 11 shows the shear failure specimens of partially riveted lap joints under different combinations
of hole diameters and squeeze forces. For all the riveted specimens, the fracture occurs at the rivet shank
of the inner and outer sheet interface. All the specimens show no obvious necking phenomenon,
indicating that the rivet is brittle fracture under the tensile force. There is an obvious gap between the
cross-section of the rivet shank and the hole wall, which is the result of the plastic deformation of the
rivet shank being squeezed by the hole wall. After breaking the rivet shank, the plastic deformation region
can not be recovered, and the round rivet shank becomes an ellipse. Except for the deformation of the
rivet shank, the sheet has no obvious deformation. As can be seen from the radial expansion diagram of
the hole wall (Fig. 5), there is a large interference-fit step at the junction of the inner and outer sheets.
Meanwhile, the single-lap riveted joints are prone to secondary bending in the tensile process, and the
specimen breaks from the middle of the rivet shank. What's more interesting is that under the squeeze
force of 15 kN, the rivet shank connected to the manufactured rivet head stays in the outer sheet. Except
for some specimens with a hole diameter of 5.10 mm, the manufactured rivet head falls off from the outer
12
sheet, as shown in Fig. 11(a). Under the 23 kN squeeze force, the manufactured rivet head of all the
specimens does not fall off from the outer sheet, which is significantly different from the specimens under
the 15 kN squeeze force. However, under the five-hole diameters, the rivet shank at the driven rivet head
end for all shear specimens is popped out from the hole under the squeeze force of 15 kN. The above
phenomena are not observed under 23 kN squeeze force conditions. The reason is that under the 23 kN
squeeze force, the interference-fit size near the driven rivet head is greater than that under the squeeze
force of 15 kN. When the riveted lap joints are subjected to tensile force, the hole wall squeezes the rivet
shank, causing plastic deformation of the rivet shank and forming a gap between the rivet shank and the
hole. The specimens with small interference-fit sizes are not enough to offset the effect of the gap at the
moment of fracture, causing the rivet head to fall off from the hole. The manufactured rivet head falls off
from the outer sheet for the specimens with a hole diameter of 5.10 mm. This is mainly because when
the hole diameter is increased to 5.10 mm, the extrusion of the rivet shank on the outer sheet decreases.
It is estimated that if the hole diameter continues to increase, all the manufactured rivet heads will fall
off from the outer sheet.

Fig. 11. Shear failed for the investigated riveted lap joints:
(a)15 kN, (b)23 kN.

13
 Fig. 12 shows the macroscopic fracture morphology of the shear specimens under different
combinations of hole diameters and squeeze forces. As Fig. 12 shows, the shear failure position and
fracture morphology under different hole diameters and squeeze forces are similar, which is the mixed
mode of brittle fracture and plastic fracture. The shear fracture of all specimens can be divided into BF-
region and PF-region. At the beginning of shearing, the inner sheet and the outer sheet will be displaced
under the tensile force. In this case, the local stress in the rivet shank at the interface between the inner
and outer sheets will increase. When the shear stress is much higher than the fracture strength of the
material, cracks will initiate and rapidly propagate in the BF-region of the fracture, forming a shear band
similar to brittle fracture. After that, the local stress began to weaken, and the rivet began to enter a state
of overall stress. With the decrease of the initial stress and the extension of the crack propagation time,
the fracture mode changes from shear bands to dimples, which appears as a plastic fracture on the
macroscopic view. When the hole diameter is constant, the area of BF-region decreases as the squeeze
force increases. This means that when the rivet breaks, the shear displacement decreases, which can also
be seen in the shear load-displacement curve.

Fig. 12. Macroscopic fracture morphology of shear specimens:

(a) D4.90-F15, (b) D4.90-F23, (c) D5.10-F15, (d) D5.10-F23.
Detailed micro fracture characteristics are shown in Fig. 13. BF-region is a shear slip region, also
called a brittle fracture region which is smooth, with obvious shear texture, and the direction is consistent
with the loading direction. PF-region is relatively rough, like goose down, darker in color, and is an
obvious plastic fracture region. The aluminum alloy used in the present research belongs to the face-
14
centered cubic structure. When the shear stress exceeds the critical shear stress, the crystal will slip along
the direction of shear force. During the shearing process, friction marks are formed due to the friction
between the fracture surfaces, and tearing ridges can be seen as a result of the rapid propagation of cracks
in the rivet shank, as shown in Fig. 13(c). As the shear load increases, it will enter the transition region
from brittle fracture to ductile fracture. There is a distinct boundary in the transition region. The boundary
is marked with a yellow dashed line, as shown in Fig. 13(b). There are many elongated dimples along
the direction of shear force below the boundary of the slip region, which is the result of plastic fracture
of the rivet shank. When the shear stress exceeds the strength limit of the rivet shank, the bearing capacity
of the rivet shank will be greatly reduced, and the rivet shank will instantaneously break. The area of the
BF-region is much smaller than that of the PF-region, which is different from the results of Li [37]. In
the current research, the rivet material is aluminum alloy 2117-T4, while the rivet material is Q235 carbon
steel in the experiment of Li. The tensile stress of Q235 carbon steel is larger than that of aluminum alloy
2117-T4. There is no doubt that Q235 has a stronger resistance to tensile deformation, so the area of
region 1 is larger.

Fig. 13. Microfracture characteristics of typical specimens:

(a) Macro fracture, (b) Magnified area A, (c) Magnified area B, (d) Magnified area C.

4.4. Shear failure mechanism

Fig. 14 shows the shear load-displacement curve and the explanation of shear failure mechanism for
typical shear specimens. According to the changing trend of the curve, the shear load-displacement curve
15
can be divided into four stages, as shown in Fig. 14(a). The first stage is called the adjustment stage. In
this stage, the gap between the fixture and the specimen is adjusted under the tensile force. Meanwhile,
there is no deformation of the rivet and sheets due to the small displacement and shear load. The second
stage is called the stationary stage. After the adjustment, the riveted lap joints enter the elastic stage. Due
to the interference-fit between the rivet and the sheets, the rivet is sufficiently resistant to the effects of
tensile force. The shear load and displacement increase linearly with the enhance of tensile force,
indicating that the riveted lap joints have good stiffness characteristics at this stage. The third stage is
called the degradation stage. With the further increase of the tensile force, the riveted lap joints enter the
plastic deformation stage, and the joint stiffness continues to decrease. In this stage, due to the low yield
strength of the rivet, the rivet first undergoes plastic deformation under the extrusion of the hole wall,
creating a gap at the interface between the rivet shank and the hole wall. Under the tensile force, the
riveted lap joints produce secondary bending, creating a gap between the contact surfaces, overcoming
the friction between the two sheets. The inner sheet and the outer sheet exert the same shear force with
the opposite direction on the rivet shank, resulting in a tilting of the rivet shank, as shown in Fig. 14(b).
As the shear load increases, the stress of the rivet shank at the interface between the two sheets increases
gradually, which will become a potential damage site. When the shear stress reaches the limit of shear
strength of the rivet, the rivet shank will be significantly damaged, resulting in shear slip. With the
occurrence of shear slip, the bearing capacity of the rivet shank is greatly reduced and the shear load
drops sharply, leading to the shear failure of the rivet shank finally. In this stage, the shear load reaches
its maximum value.

Fig. 14. Description of typical shear load-displacement curve and shear failure mechanism:
(a) Shear load-displacement curve, (b) Shear failure mechanism.

5.Conclusions

The influence of different hole diameters on the shear behavior of riveted lap joints was studied by
finite element methods and experiments to ensure the safety of service, the following conclusions can be
drawn:
The analysis of the material flow characteristics of the riveting process indicated that as the squeeze
force increased from 15 kN to 23 kN, the amount of material flowing into the hole increased, and the
position of the radial displacement mutation point moved from 5 mm to 6 mm.
For all the riveted specimens, the fracture occurred at the rivet shank of the inner and outer sheet
interface. All the specimens belong to the brittle-plastic mixed fracture modes of rivet shank. The area of
the brittle fracture region was much smaller than the plastic fracture region.

16
 Shear tests showed that the shear load was enhanced when the hole diameter increased from 4.82 mm
to 5.10 mm. Compared with the increase of the squeeze force, the shear load of the riveted lap joints
could be raised effectively by increasing the hole diameter. The quantitative studies of the shear tests
showed that the maximum shear load increased by 17.77% and 17.24% under 15 kN and 23 kN squeeze
forces respectively.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships
that could have appeared to influence the work reported in this paper.

Acknowledgments

The authors are grateful for the support National Science Foundation of China (Grant No. 52005259),
and the National Major Science and technology project (Grant No. 2018ZX04014001) for sponsoring
this project. Moreover, the authors would like to acknowledge the editors and the anonymous reviewers
for their insightful comments.

References

[1] C. Zeng, W. Tian, W.H. Liao. The effect of residual stress due to interference fit on the fatigue
behavior of a fastener hole with edge cracks, Eng. Fail. Anal. 66(2016) 72-87.
[2] X. Zhang, H. Jiang, T. Luo. Theoretical and experimental investigation on interference fit in
electromagnetic riveting, Int. J. Mech Sci 156 (2019) 261-271.
[3] K.F. Zhang, H. Cheng, Y. Li. Riveting Process Modeling and Simulating for Deformation Analysis
of Aircraft's Thin-walled Sheet-metal Parts, Chin. J. Aeronaut. 24 (2011) 369-377.
[4] M. Skorupa, T. Machniewicz, A. Skorupa. Fatigue life predictions for riveted lap joints, Int. J.
Fatigue 94 (2017) 41-57.
[5] G. Liu, H.L. Huan, Y.L. Ke. Study on analysis and prediction of riveting assembly variation of
aircraft fuselage panel, Int. J. Adv. Manuf. Tech. 75 (2014) 991-1003.
[6] J.M. Kafie, P.B. Keating, V.P Chakra. Stress distributions and crack growth in riveted lap joints
fastening thick steel plates, Eng. Fail. Anal. 91(2018) 370-381.
[7] Y.J. Zuo, Z.Q. Cao, G. Zheng. Damage behavior investigation of CFRP/Ti bolted joint during
interference fit bolt dynamic installation progress, Eng. Fail. Anal. 111(2020) 104454.
[8] T.N. Chakherlou, M. Shakouri, A. Akbari. Effect of cold expansion and bolt clamping on fretting
fatigue behavior of Al 2024-T3 in double shear lap joints, Eng. Fail. Anal. 25 (2012) 29-41.
[9] T.N. Chakherlou, M.J. Razavi, B. Abazadeh. Finite element investigations of bolt clamping force
and friction coefficient effect on the fatigue behavior of aluminum alloy 2024-T3 in double shear
lap joint, Eng. Fail. Anal. 29 (2013) 62-74.
[10] J. Li, F. K. Zhang, Y. Li. Influence of interference-fit size on bearing fatigue response of single-lap
carbon fiber reinforced polymer/Ti alloy bolted joints, Tribol. Int. 93 (2015) 151-162.
[11] M. Chishti, C.H. Wang, R.S. Thomson. Numerical analysis of damage progression and strength of
countersunk composite joints, Compos. Struct. 94 (2012) 643-653.
[12] J.S. Hu, K.F. Zhang, Y.W. Xu. Modeling on bearing behavior and damage evolution of single-lap
bolted composite interference-fit joints, Compos. Struct. 212 (2019) 452-464.
[13] W.S. Shin, K.H. Chang, S. Muzaffer. Fatigue analysis of cruciform welded joint with weld
penetration defects, Eng. Fail. Anal. 120 (2021) 105111.
[14] J. Zhao, C.S Wu, H. Su. Ultrasonic effect on thickness variations of intermetallic compound layers

17
 in friction stir welding of aluminium/magnesium alloys, J. Manuf. Process 62 (2021) 388-402.
[15] D.F. Wang, W.C. Xu. Fatigue failure analysis and multi-objective optimisation for the hybrid
(bolted/bonded) connection of magnesium-aluminium alloy assembled wheel, Eng. Fail. Anal.
112(2020) 104530.
[16] J.J. Cui, S. Gao, H. Jiang. Adhesive bond-electromagnetic rivet hybrid joining technique for
CFRP/Al structure: Process, design and property, Compos. Struct. 244 (2020) 112316.
[17] F. Aman, S.H. Cheraghi, K.K. Krishnan. Study of the impact of riveting sequence, rivet pitch, and
gap between sheets on the quality of riveted lap joints using finite element method, Int. J. Adv
Manuf. Tech. 67 (2013) 545-562.
[18] M.P. Szolwinski, T.N. Farris. Linking Riveting Process Parameters to the Fatigue Performance of
Riveted Aircraft Structures, J. Aircraft 37 (2012) 130-137.
[19] D. Rijck, J.J. Homan, J. Schijve. The driven rivet head dimensions as an indication of the fatigue
performance of aircraft lap joints, Int. J. Fatigue 29 (2007) 2208-2218.
[20] M. Skorupa, T. Machniewicz, A. Skorupa. Fatigue life prediction model for riveted lap joints, Eng.
Fail. Anal. 53 (2015) 111-123.
[21] S.H. Cheraghi. Effect of variations in the riveting process on the quality of riveted joints, Int. J.
Adv. Manuf. Tech. 39 (2008) 1144-1155.
[22] C. Zeng, W. Tian, W.H. Liao. Improved Model Concerning Driven Rivet Head Dimensions Based
on Material Flow Characteristics, J. Aircraft. 53 (2016) 1180-1185.
[23] B. Zheng, H.D. Yu, X.M. Lai. Assembly deformation prediction of riveted panels by using
equivalent mechanical model of riveting process, Int. J. Adv. Manuf. Tech. 92 (2017) 1955-1966.
[24] R.P.G. Muller. An experimental and analytical investigation on the fatigue behaviour of fuselage
riveted lap joints, Delft University of Technology 1995.
[25] G. Li, G.Q. Shi, N.C. Bellinger. Stress in Triple-Row Riveted Lap Joints Under the Influence of
Specific Factors, J. Aircraft 48 (2011) 527-539.
[26] H.D. Yu, B. Zheng, X. Xu. Residual stress and fatigue behavior of riveted lap joints with various
riveting sequences, rivet patterns, and pitches, P. I Mech. Eng. B-J Eng. 233(2019) 2306-2319.
[27] M. Skorupa, A. Skorupa, T. Machniewicz. Effect of production variables on the fatigue behaviour
of riveted lap joints, Int. J. Fatigue 32 (2010) 996-1003.
[28] S. Elzbieta, S. Grzegorz, J. Jerzy. Comparison of the riveting process of a rivet with and without a
compensator, Journal of Kones 16 (2009) 455-462.
[29] R.F. Simenz, M.A. Steinberg. Alloy Needs and Design: The Airframe. Fundamental Aspects of
Structural Alloy Design, 1977.
[30] W. Wronicz, J. Kaniowski, Korzeniewski B. Experimental and Numerical Study of Stress and Strain
Field around the Rivet, 2011.
[31] C. Rans, P.V. Straznicky, Alderliesten R. Riveting process induced residual stresses around solid
rivets in mechanical joints, J. Aircraft 44 (2007) 323-329.
[32] M. Skorupa, A. Skorupa, T. Machniewicz. Improving the fatigue performance of riveted joints in
airframes, Final report on EUREKA project IMPERJA, No.E!3496. University of Science and
Technology AGH 2011.
[33] M.R. Urban. Analysis of the fatigue life of riveted sheet metal helicopter airframe joints, Int. J.
Fatigue 25 (2003) 1013-1026.
[34] M. Skorupa, T. Machniewicz, A. Skorupa. Fatigue strength reduction factors at rivet holes for
aircraft fuselage lap joints, Int. J. Fatigue 80 (2015) 417-425.
[35] Z.Q. Cao, M. Cardew-Hall. Interference-fit riveting technique in fiber composite laminates, Aerosp.
18
 Sci. Technol. 10 (2006) 327-330.
[36] Z.Q. Cao, Y.J. Zuo. Electromagnetic Riveting Technique and Its Applications, Chin. J. Aeronaut.
33 (2020) 5-15.
[37] Y.G. Li, J. Hao, Z. Xu. Mechanical properties and fatigue behavior of electromagnetic riveted lap
joints influenced by shear loading, J. Manuf. Processes 26 (2017) 226-239.
[38] T.N. Chakherlou, M. Mirzajanzadeh, J. Vogwell. Experimental and numerical investigations into
the effect of an interference fit on the fatigue life of double shear lap joints, Eng. Fail. Anal. 16
(2009) 2066-2080.
[39] C. Zeng, J.T. Xue, X.Y. Liu. Design variables influencing the fatigue of Al 2024-T3 in riveted
aircraft lap joints: Squeeze force and fit tolerance, Int. J. Fatigue 140 (2020) 105751.
[40] A. P. Atre. A Finite Element and Experimental Investigation on the Fatigue of Riveted Lap Joints
in Aircraft Applications, Georgia Institute of Technology 2006.
[41] A.P. Atre, W.S. Johnson. Analysis of the Effects of Interference and Sealant on Riveted Lap Joints,
J. Aircraft 44 (2012) 353-364.
[42] G. Sławiński, T. Niezgoda, E. Szymczyk, J. Jachimowicz. Numerical study of the influence of shape
imperfections on residual stress fields in a rivet hole, Journal of KONES 17 (2010) 427-434.
[43] C. Zeng, W.H. Liao, W. Tian. Influence of fit tolerance and squeeze force on the residual stress in
a riveted lap joint, Int. J. Adv. Manuf. Tech. 81 (2015) 1643-1656.
[44] B. Zheng, H.D. Yu, X.M. Lai. Analysis of Residual Stresses Induced by Riveting Process and
Fatigue Life Prediction, J. Aircraft 53 (2016) 1-8.
[45] A. Manes, M. Giglio, F. Viganò. Effect of riveting process parameters on the local stress field of a
T-joint, Int. J. Mech Sci 53(2011) 1039-1049.
[46] G. Li, G. Shi, N.C. Bellinger. Study of the Residual Strain in Lap Joints, J. Aircraft 43 (2006) 1145-
1151.
[47] C. Zeng, W. Tian, X.Y. Liu. Experimental and numerical studies of stress/strain characteristics in
riveted aircraft lap joints, J. Mech. Sci. Technol. 33 (2019) 3245-3255.
[48] J.T. Liu, A.A. Zhao, Z.Z. Ke. Influence of Rivet Diameter and Pitch on the Fatigue Performance of
Riveted Lap Joints Based on Stress Distribution Analysis, Materials 13 (2020) 3625.
[49] V. Blanchot, D. Alain. Riveted assembly modelling: Study and numerical characterisation of a
riveting process, J. Mater. Process Tech. 180 (2006) 201-209.
[50] I.G.A.A Jaya. Safety risk in aircraft structural joints associated with the use of corrosion treatment,
RMIT University 2013.
[51] C.Y. Lei, Y.B. Bi, J.X. Li. Effect of riveting parameters on the quality of riveted aircraft structures
with slug rivet, Adv. Mech. Eng. 9 (2017).
[52] A. Skorupa, M. Skorupa, T. Machniewicz. Fatigue crack location and fatigue life for riveted lap
joints in aircraft fuselage, Int. J. Fatigue 58 (2014) 209-217.
[53] C.D. Rans. The role of rivet installation on the fatigue performance of riveted lap joints, Carleton
University 2007.

19
[54]

20
Highlights

 Material flow characteristics and interference distribution in the riveting process.

 The pop-up phenomenon of rivet manufactured head and driven rivet head in the
shearing process.
 Brittle fracture and plastic fracture transition mechanism of rivet shank in the shearing
process.
 Shear fracture stage division and shear failure mechanism.
[55]

21