C. ZHAO et al.: MECHANICAL PROPERTIES OF A NOVEL JOINT OF A SINGLE-LAYER ALUMINUM-ALLOY ...
811–819
MECHANICAL PROPERTIES OF A NOVEL JOINT OF A
SINGLE-LAYER ALUMINUM-ALLOY COMBINED
LATTICE-SHELL STRUCTURE
MEHANSKE LASTNOSTI NOVEGA SPOJA IZ ENE PLASTI
ALUMINIJEVE ZLITINE KOMBINIRANE MRE@NO-LUPINASTE
STRUKTURE
Caiqi Zhao
*
, Jun Ma, Shengchen Du, Ye Gu, Yunwen Zhou
Key Laboratory of Concrete and Prestressed Concrete Structures, Ministry of Education, School of Civil Engineering, Southeast University,
Nanjing 210096, China
Prejem rokopisa – received: 2019-02-01; sprejem za objavo – accepted for publication: 2019-06-17
doi:10.17222/mit.2019.031
To overcome the shortcomings of traditional aluminum-alloy gusset joints, i.e., the low bearing capacity caused by the lack of a
shear connector that makes these joints unsuitable for an aluminum-alloy combined lattice-shell structure of a larger span, this
paper proposes a novel joint with a high bearing capacity that has a cylindrical socket-and-spigot type shear connector at the
center of the joint. Experimental research was carried out on the novel joint. The results show that the overall stiffness and shear
resistance of the novel joint are greatly improved compared with those of a traditional gusset joint and that the out-of-plane
bending capacity is more than 50 % higher. The failure mode of the novel joint is mainly manifested by tension and shear block
failure of the lower cover plate when the thickness of the gusset plate is small and torsional buckling failure of the connected
aluminum-alloy beam. An elastoplastic finite-element parametric analysis based on multiple contacts shows that the bearing
capacity of the proposed joint is related to the thickness of the connecting plates and the diameter and number of bolts. Based on
the working mechanism of the novel joints, this paper offers a simplified calculation formula for the ultimate bearing capacity of
the joint. The results of this paper can provide a theoretical basis for making and revising the design codes for this novel joint.
Keywords: single-layer combined lattice shell, aluminum-alloy, novel joint, cylindrical socket-and-spigot type shear connector,
bearing capacity test
Avtorji predlagajo nov na~in spajanja, da bi se izognili slabostim tradicionalnega trikotnega spoja iz Al zlitin oz. njihove majhne
nosilnosti zaradi odsotnosti stri`nega konektorja, ki naredi te spoje neprimerne za mre`no-lupinaste strukture z ve~jim razpo-
nom. Nov, predlagan na~in spajanja oz. vezave z veliko nosilnostjo, ima cilindri~ni stri`ni tip SaS (angl.: socket-and-spigot)
konektorja v sredi{~u spoja. Avtorji so izvedli eksperimentalno raziskavo predlaganega spoja. Rezultati raziskave so pokazali,
da je celokupna togost novega spoja in njegova odpornost proti strigu, precej izbolj{ana v primerjavi s tradicionalnim trikotnim
spojem in, da je zunanja ravninska odpornost spoja pred upogibanjem bolj{a za ve~ kot 50 %. Na~in poru{itve novega spoja je v
glavnem posledica nateznih obremenitev. Strig spodnje pokrivne plo{~e je blokiran pri majhni debelini trikotne plo{~e in tudi
poru{itev zaradi torzijskega upogibanja povezovalne gredi iz Al zlitine je prepre~ena. Elastoplasti~na parametri~na analiza na
osnovi metode kon~nih elementov mnogoterih kontaktov, je pokazala, da je nosilnost novega predlaganega spoja odvisna od
debeline povezovalnih plo{~ ter premera in {tevila veznih vijakov. Na osnovi delovnega mehanizma novega spoja avtorji v tem
~lanku podajajo enostavno ena~bo za izra~un nosilnosti predlaganega spoja. Rezultati, podani v tem ~lanku, lahko predstavljajo
teoreti~no osnovo za revizijo in izdelavo ra~unalni{ke kode za dizajniranje predlagane vezi.
Klju~ne besede: enoslojna kombinirana mre`no lupinasta struktura, Al zlitina, nova vrsta spoja, cilindri~ni SaS-tip stri`nega
konektorja, test nosilnosti
1 INTRODUCTION
A wide range of joint forms are used in spatial grid
structures, including the German Mero system, the Ame-
rican Unistrut system, the Canadian Triodetic system,
and the Japanese NS system.
1–2
An aluminum-alloy
lattice-shell structure, due to its poor weldability, is
generally mechanically connected. Common joint types
include hub joints, Temcor joints (gusset joints), and cast
aluminum joints.
3–8
In the research on single-layer alu-
minum-alloy lattice-shell joints, G. G. Shi et al.
9–10
of
Tsinghua University conducted a series of studies on cast
aluminum joints and found that the out-of-plane bending
stiffness is higher than the in-plane bending stiffness;
they also identified the most unfavorable position of the
joint. Y. Q. Wang et al.
11
performed a nonlinear finite-
element analysis on aluminum-alloy gusset joints based
on experimental data. The results showed that the force
acting on the cover plate is transmitted to the aluminum-
alloy beam through the bolt group and that the specimen
is first formed with a plastic zone at the last row of bolt
holes of the flange and at the bottom of the web.
J. H. Qian et al.
12
showed that in the design of Tem-
cor joints, the diameter and thickness of the circular
cover plate and the specification and quantity of the con-
nection bolts are all key parameters. In the literature
13–16
authors have analyzed the factors affecting the stiffness
of the circular aluminum-alloy cover plate-gusset joint,
Materiali in tehnologije / Materials and technology 53 (2019) 6, 811–819 811
UDK 67.017:669.715:620.168 ISSN 1580-2949
Original scientific article/Izvirni znanstveni ~lanek MTAEC9, 53(6)811(2019)
*Corresponding author's e-mail:
101000815@seu.edu.cn (Caiqi Zhao)

but did not consider the influence of the bolts. X. N. Guo
et al.
17–19
studied the bending mechanism of gusset joints,
obtained the bending-moment rotation-angle curve via
experiments and used parameter analyses to obtain the
pattern of influence of factors such as the thickness of
the gusset plate, the height of the aluminum-alloy beam
and the number of bolts on the initial stiffness.
These research studies were generally carried out on
existing traditional aluminum-alloy gusset joints. How-
ever, the web of the aluminum-alloy beam in a gusset
joint is discontinuous in the joint region due to the
absence of a shear connector in the central region of the
joint; thus, the joint is supported only by the upper and
lower flanges of the connected beam. This feature
provides a limited shear bearing capacity, and it cannot
meet the bearing-capacity demands of a larger span
lattice-shell structure. This paper proposes a new type of
aluminum-alloy joint with a built-in core shear connec-
tor. Studies have shown that this joint has a higher
overall stiffness and load-bearing capacity than tradi-
tional gusset joints. It is applicable not only to general
single-layer aluminum-alloy lattice-shell structures but
also to aluminum-alloy composite lattice-shell structures
that incorporate aluminum-alloy honeycomb panels.
20–24
By means of a theoretical analysis and experimental
research, this paper further explores the force perform-
ance, force transmission mechanism and failure mecha-
nism of the novel joint and proposes a simplified formula
to provide a theoretical basis for the design procedure of
these joints in the future.
2 EXPERIMENTAL PART
2.1 Design and production of test specimen
To explore the stress characteristics, bearing capacity
and failure mechanism of the novel joint, six aluminum-
alloy joints are designed. Three of them are novel joints
(numbered JD1, JD3 and JD5), and the other three are
traditional joints (numbered JD2, JD4 and JD6). The
plane of each specimen is a regular hexagon with a dia-
meter of 2000 mm (Figure 1a). Each joint is connected
with 6 H-shaped aluminum-alloy beams (the cross-sec-
tional dimensions are 110 mm × 60 mm×5m m×
5 mm), and the other end of the beam is supported by
6 steel columns (Figures 1b and 1c). The novel joint
consists of two upper and lower circular cover plates and
a cylindrical socket-and-spigot type shear connector
located between them (Figure 1d). The circular cover
plates have a diameter of 280 mm and three thicknesses
of 6 mm (JD1 and JD2), 4 mm (JD3 and JD4) and 3 mm
(JD5 and JD6). Each of the cover plates is drilled with 8
holes at a 60° angle (Figures 1e and 1f) ,a n d8M 6
stainless steel bolts are used to connect the flanges of the
aluminum-alloy beam ends. The diameter of the
cylindrical socket-and-spigot type shear connector is 70
mm, and the height is 100 mm. Six "T-shaped" grooves
are machined using a numerical control machine tool
along a 60° angle (Figures 1g and 1h). After inserting
six "T-shaped" shear connection plates into the grooves,
the M6 bolts are used to connect the plates to the webs of
the beam ends (Figures 1i and 1j). The conventional
joint is not provided with a cylindrical shear connector;
C. ZHAO et al.: MECHANICAL PROPERTIES OF A NOVEL JOINT OF A SINGLE-LAYER ALUMINUM-ALLOY ...
812 Materiali in tehnologije / Materials and technology 53 (2019) 6, 811–819
Figure 1: Test model overview: a) top view, b) side view, c) the loading site, d) the test joint, e) and f) the cover plate, g) and h) the
socket-and-spigot type shear connector, i) an aluminum-alloy beam, j) a novel joint, k) a traditional joint, and l) support at the end of the beam

only the upper and lower circular cover plates are con-
nected to the flanges of the beam ends, and there are no
connections between the webs at the ends of the beams
(Figure 1k). The test joints are largely made of 6061-T6
aluminum alloy, except for the connection bolts, which
are made of 304 stainless steel.
2.2 Loading and testing methods
The displacement measuring points of the test joint
are arranged at the mid-span loading and the locations of
the six supports. The strain gauges and the strain rosette
are arranged in locations with a large stress, i.e., on the
upper and lower cover plates of the joint, on the flanges
and webs of the aluminum-alloy beams, and close to the
vicinity of the connection bolts (Figure 2).
The test uses the MTS autoload control system to
apply a centralized load at the center of the joint.
Loading is applied according to a displacement sensor at
a speed of 0.02 mm/s. To prevent the six steel columns
from being turned by the eccentric pressure during the
loading process, a support system consisting of lateral
inclined supports and horizontal coupling beams is
arranged between the steel columns. A rigid circular
block is placed at the loading point of the upper cover to
prevent the loading head from hitting the connection
bolts on the cover plate and affecting their normal force
transmission.
3 RESULTS AND DISCUSSION
3.1 Failure mode of test joints
3.1.1 Failure mode of novel joints
Due to space limitations, only the damage process of
JD1 is provided here; those of the other two cases (JD3
and JD5) are similar. The loading no longer increases
and begins to decrease when the load reaches 71 kN, and
it stops when the load decreases to 60 kN. The bending
deformation of the aluminum-alloy beams is obvious.
The deformation of several bolt holes on the flange of
the aluminum-alloy beams away from the center of the
joint is severe, and some individual nuts are embedded in
the holes. If the loading continues, the aluminum-alloy
beam will be broken. Due to the large thicknesses of the
upper and lower cover plates, only slight bending defor-
mations occurred. At the lower end of the cylindrical
socket-and-spigot-type shear connector, the tensile
deformation is obvious; however, there are no apparent
cracks, and the shear connection plates show no obvious
deformation. The bolts connecting the upper and lower
cover plates and the flanges of the aluminum-alloy
beams show compression and shear deformation and
tensile and shear deformation, and the bolts between the
shear connection plates and the aluminum-alloy beam
webs also underwent compression and shear deformation
(Figure 3).
3.1.2 Failure mode of traditional joints
Only the damage process of JD6 is provided here;
those of the other two cases (JD2 and JD4) are similar.
C. ZHAO et al.: MECHANICAL PROPERTIES OF A NOVEL JOINT OF A SINGLE-LAYER ALUMINUM-ALLOY ...
Materiali in tehnologije / Materials and technology 53 (2019) 6, 811–819 813
Figure 3: Novel joint (JD1) failure mode: a) deformation after loading, b) torsional deformation of aluminum-alloy beams, and c) load–strain
curves
Figure 2: Strain measurement point layout: a) inner surface strain measuring point of the upper cover, b) outer surface strain measuring point of
the lower cover, c) inner surface strain measuring point of the aluminum-alloy beam flange, and d) outer surface strain measuring point of the
aluminum-alloy beam flange

Since JD6 is a traditional joint connected only by the
upper and lower cover plates, its bearing capacity is
significantly lower than that of the novel joint, and its
damage form is also very different. A rustling sound was
heard when the loading reached 41 kN; a large metal
breaking sound was produced when the joint was loaded
to 52 kN, and the load value plummeted to 13 kN, at
which point the bolt holes of the lower cover were torn.
Significantly different from the failure mode of the novel
joint, the upper and lower cover plates of the traditional
joints were damaged by block shearing along the direc-
tion of the bolt-hole connection. The bolts connecting
the upper and lower cover plates and the flanges of the
aluminum-alloy beams produced compression and shear
deformation (Figure 4).
3.2 Test result analysis
1) A comparison of the novel joint and the traditional
gusset joint shows that the shearing capacity and the
out-of-plane bending capacity are significantly improved
due to the setting of the cylindrical socket-and-spigot-
type shear connector in the central area of the novel
joint, with an average increase of greater than 50 %.
2) Three different aluminum-alloy joint experimental
tests show that the failure mode of the traditional gusset
joint is either the tension and shear block failure of the
lower cover plate or the bending damage of the alumi-
num-alloy beam. The failure mode of the novel joint is
mainly represented by the torsional buckling failure of
the connected aluminum-alloy beam. The block shearing
damage of the lower cover plate occurs only when the
thickness of the gusset plate is small. In contrast to the
traditional joint, the thicknesses of the upper and lower
cover plates of the novel joint have little effect on its
bearing capacity. This is mainly because the shear
connectors in this type of joint share a large part of the
load-bearing capacity.
3.3 Elastoplastic finite-element analysis of novel joints
3.3.1 Analysis method
The constitutive relationship of the component
materials in the test joint is shown in Figure 5.T o
improve the accuracy and convergence speed of the
finite-element analysis results, when the finite-element
model is constructed, a fine meshing is used for the parts
with holes and the parts where stress concentration can
easily occur
25–29
(Figure 6). Additionally, to fully con-
sider the contact between the upper and lower covers and
the aluminum-alloy beams and the contact between the
stainless-steel bolts and the bolt holes, face-to-face
contact unit connections are used.
30–35
According to the
loading method and the bearing condition in the test, the
boundary constraint is set at the distal end of the
aluminum-alloy beams, and the vertical displacement
load is applied at the center position of the joint.
3.3.2 Analysis results
It can be seen from the stress contour diagrams in
Figures 7a to 7c that the stresses on the upper and lower
cover plates of JD1 and the bolt holes are the greatest
(260MPa), and the stress near the loading point in the
C. ZHAO et al.: MECHANICAL PROPERTIES OF A NOVEL JOINT OF A SINGLE-LAYER ALUMINUM-ALLOY ...
814 Materiali in tehnologije / Materials and technology 53 (2019) 6, 811–819
Figure 5: Constitutive relationship curves of the joint materials: a) aluminum-alloy beam, b) aluminum-alloy cover plate and connector and
c) stainless-steel bolt
35
Figure 4: Traditional joint (JD6) failure mode: a) overall deformation, b) and c) shear failure of the lower cover plate

central area of the upper cover plate is also large
(238 MPa). It can be seen from the different stress
distributions at the bolt holes of the upper and lower
cover plates that the upper cover-plate bolt holes are
more compressed at the near-center ends. The bolt holes
of the lower cover plate are also subjected to com-
pression at the telecentric end, as shown in a partially
enlarged view.
Figure 7d shows a von Mises stress-contour diagram
of an aluminum-alloy beam after loading. The upper-
flange stress developed in the aluminum-alloy beam is
asymmetrical, and thus, the aluminum-alloy beam is
twisted, which is consistent with the phenomenon
observed in the test. A plastic zone is formed near the
bolt holes of the lower flange of the beam away from the
cover plate; it is presumed that the damage location
should be located here, which is also consistent with the
phenomenon observed in the test.
Figure 7e shows that the stress is not large, except
where the shear connector is in contact with the inner
surface of the flange of the aluminum-alloy beam, where
it reaches the ultimate stress. This result shows that the
shear connector has sufficient stress reserve and
conforms to the design concept of a "strong joint-weak
member".
Figure 7f shows that the theoretical load–displace-
ment curves of the novel joint agree well with the
measured experimental curves. The average error
between them is 3.6 %.
3.4 Influential factors and simplified formula for novel
joint’s bearing capacity
3.4.1 Influential factors of novel joint’s bearing
capacity
Studies
17–19
have shown that the bearing capacity of a
traditional joint is related mainly to the thicknesses of
the upper and lower cover plates and the distribution of
the bolts. The novel joint is based on the traditional joint
with a shear connector added to the core area; thus, its
bearing capacity is related to the installed shear
connection plates and their connection bolts. To examine
the influence of the thickness of the shear connection
plate and the diameter and quantity of the connection
bolts on the bearing capacity of the novel joint under the
circumstance where the thicknesses of the upper and
lower cover plates are constant, this study selected 5
C. ZHAO et al.: MECHANICAL PROPERTIES OF A NOVEL JOINT OF A SINGLE-LAYER ALUMINUM-ALLOY ...
Materiali in tehnologije / Materials and technology 53 (2019) 6, 811–819 815
Figure 6: Finite-element model of the joint: a) the overall structure, b) an aluminum-alloy beam, c) a shear connection plate and d) upper and
lower cover plates
Figure 7: Stress and displacement conditions of JD1: a) the inner surface of the upper cover plate, b) the outer surface of the lower cover plate,
c) the vicinity of the bolt holes of the lower cover plate, d) stress-contour diagram of beam, e) stress-contour diagram of the shear connector and
f) load–displacement curves

shear connection plates of different thicknesses (t =
3 mm, 4 mm, 5 mm, 6 mm, 8 mm) and connection bolts
of 3 different diameters (d = 6 mm, 8 mm, 10 mm) and
quantities (n = 4, 6, 8); thus, a total of 45 joint models
were used to carry out the finite-element parametric
analysis.
Figure 8 shows part of the results for the analysis of
the joint’s bearing-capacity parameter when shear
connection plates of different thicknesses and bolts of
different diameters and quantities were used when the
joint cover plates were 4 mm thick. Figure 8a indicates
that in the case where the thicknesses of the upper and
lower cover plates and the bolt diameter are unchanged,
the bearing capacity of the joint can be improved by
increasing the number of bolts on the connection plates;
the average increase in the bearing capacity is approx-
imately 6 % after the bolt number is increased. However,
taking on-site construction factors into consideration, the
workload of the on-site installation will expand if too
many bolts are involved. Figure 8b reveals that in the
circumstance where the thickness of the connection
plates and the number of bolts is unchanged, the bearing
capacity of the joint increases as the bolt diameter
increases, yet its increment is not large.
Figure 8c illustrates that under the condition where
the diameter and number of connection bolts are fixed,
the bearing capacity of the joint can be improved by
increasing the thickness of the shear connection plates.
When the thickness of the shear connection plates
reaches the beam web thickness of 5 mm, the slope of
the load–displacement curve of the joint is greater than
that of the other curves. Therefore, in the future design
of such joints, it is recommended that the thickness of
the connection plates remains consistent with the beam
web thickness.
3.4.2 Simplified formula for novel joint’s bearing
capacity
This study indicates that the joint’s failure mode is
determined as follows: when the thickness of the upper
and lower cover plates is less than that of the alumi-
num-alloy beam flange, tension and shear block failure
of the cover plates will occur. When the thickness of the
cover plates is greater than that of the beam flange,
fracture failure of the aluminum-alloy beams or failure
of the connection bolts will occur.
3.4.3 Out-of-plane bearing capacity of the joint at the
failure of the cover plates
According to the experiment, when loading is applied
to the ultimate load, fracture-zone failure occurs on the
cover plates. The tensile cover plates of the joint can
present three types of shearing failure under the action of
the bolt group: single-fan-shaped area failure, double-
fan-shaped area failure, and tri-fan-shaped area failure,
as shown in Figure 9.
1) Single-fan-shaped area tensile and shear failure
Vk t flN
ii
i
u1 u
=≥
=
∑ 1
3
1
(1)
In Equation (1), V
u1
is the bearing capacity of the
single-fan-shaped area tensile and shear failure, is the
thickness of the cover plate, f
u
is the ultimate tensile
strength of the material, li is the net length of the i
th
C. ZHAO et al.: MECHANICAL PROPERTIES OF A NOVEL JOINT OF A SINGLE-LAYER ALUMINUM-ALLOY ...
816 Materiali in tehnologije / Materials and technology 53 (2019) 6, 811–819
Figure 9: Tensile and shear failures of the lower cover plate: a) single-fan-shaped area failure, b) double-fan-shaped area failure, and c) tri-fan-
shaped area failure
Figure 8: Joint bearing capacity parametric analysis result: a) t = 4 mm, d = 6 mm, b) t = 4 mm, n = 6, and c) n=6,d=6mm

failure side, l
1
 l
3
are the failure-side net lengths of the
single-fan-shaped area tensile and shear failure, and N
1
is
the tensile force of aluminum-alloy beam L
1
.  i
is the
equivalent failure intensity coefficient of the material at
the i
th
failure side, which is the normal stress and the
shearing stress on the failure side. Under the circum-
stance where von Mises yielding conditions are met, i
is
the ratio of the maximum value of the projection of the
resultant force in the failure-bearing direction and the
ultimate tensile strength of the material with respect to
the single-fan-shaped area tensile and shear failure,  1
=
 3
= 0.58 and 2
=1.k is the bearing-capacity reduction
factor caused by the partial bending of the cover plate.
2) Double-fan-shaped area tensile and shear failure
Vk t flNN
ii
i
u2 u
=≥ +
=
∑ 1
5
12
(2)
V
u2
is the bearing capacity of the double-fan-shaped area
tensile and shear failure, and N
1
and N
2
are the tensile
forces of aluminum-alloy beams L
1
and L
2
, respectively.
l
1
 l
5
are the failure-side net lengths of the double-
fan-shaped area tensile and shear failure. 2
depends on
the angle  1
between two adjacent aluminum-alloy
beams of the double-fan-shaped area.
Tri-fan-shaped area tensile and shear failure
Vk t flNNN
ii
i
u3 u
= ≥++
=
∑ 1
7
123
(3)
In Equation (3), V
u3
is the bearing capacity of the
tri-fan-shaped area tensile and shear failure, l
1
 l
7
are the
failure-side net lengths of the tri-fan-shaped area tensile
and shear failure, and N
1
, N
2
and N
3
are the tensile forces
of aluminum-alloy beams L
1
, L
2
and L
3
, respectively. The
value of  i
depends on the angle between the adjacent
aluminum-alloy beams corresponding to the tri-fan-
shaped area. i = 1, 2, 3 relates to 1
, and i = 4, 5, 6 relates
to 2
.
3.4.4 When the aluminum alloy beam fractures
When the thickness of the cover plate is greater than
that of the aluminum-alloy beam flange and the bearing
capacity of the bolts is sufficiently large, fracture failure
of the aluminum-alloy beams will occur. The fracture
zone is distributed along the outermost region of the
cover plate center where the bolt holes are located. Con-
sidering that hole weakening is present at the web con-
necting to the shear connector, the formula is expressed
as follows:
[]
 =>
M
w
1
(4)
Wbs dtht
thtn d
1
2
2
6
=−⋅⋅−+
⋅−−
()()
()
ff
w f
(5)
The bending moment generated by the friction bet-
ween the bolt and aluminum-alloy beam was not con-
sidered in Equations (4) and (5), where W
1
is the bending
section modulus of the aluminum-alloy beam cross-
section, n is the number of bolt rows on the web, and [ ]
is the allowable stress of the material.
3.4.5 When the bolts fail
Assuming that the requirements for the bearing
capacity of the cover plates and aluminum-alloy beams
are met, the bearing capacity of the joint is controlled
mainly by the bolt’s strength. Due to the introduction of
a shear connector in the novel joint, the out-of-plane
bending moment is carried collectively by the bolts bet-
ween the upper and lower cover plates and aluminum-
alloy beam flanges, together with the bolts between the
webs and the shear connector. Set  = V
y
·R/M
y
,  =
N
x
·h/M
y
, where R is the radius of the cover plate, and h is
the height of the aluminum-alloy beam, as shown in
Figure 10. The resulting equation is as follows:
MMV
R
N
h
yy x total
=++
22
(6)
MM
y total
=⋅±±
⎛ ⎝ ⎜ ⎞ ⎠ ⎟ 1
22
  (7)
In Equation (6), M
total
is the total out-of-plane
bending moment carried by the joint, M
u
is the
out-of-plane bending-resistance bearing capacity of the
joint, M
y
is the out-of-plane bending moment, V
y
is the
shear force in the direction of the y-axis, N
x
is the axial
force in the x direction, R is the radius of the cover plate,
h is the height of an aluminum-alloy beam, n
1
is the
number of bolts for each fan-shaped area on the joint
plate, N
T1
is the shear bearing capacity of the bolt on the
cover plate, n
2
is the number of bolts on the shear
connector, N
T1
is the shear bearing capacity of a bolt on
the shear connection plates, and h
2
is the spacing of the
bolt centers on the shear connector, as shown in Figure
10. The formulas are as follows:
MnNhnNh
u TT
=⋅⋅+⋅⋅
11222
(8)
or MnNh
u T
=⋅⋅⋅+
11
1()  (9)
where  =
⋅⋅
⋅⋅
nNh
nNh
222
111
T
T
(10)
By substituting relevant parameters of the test joints
into the above formulas, the mean error between the
theoretical calculated bearing capacity and the measured
C. ZHAO et al.: MECHANICAL PROPERTIES OF A NOVEL JOINT OF A SINGLE-LAYER ALUMINUM-ALLOY ...
Materiali in tehnologije / Materials and technology 53 (2019) 6, 811–819 817
Figure 10: Diagram of novel joint’s bearing-capacity calculation

bearing capacity is within 5 %, indicating that the sim-
plified formula has good computational accuracy.
4 CONCLUSIONS
To resolve the shortcomings of traditional gusset
joints, which are widely used in current single-layer
aluminum-alloy lattice-shell structures, a novel joint of a
cylindrical socket-and-spigot-type shear connector was
proposed in this paper for the first time. An experimental
study and an elastoplastic finite-element analysis were
conducted to provide an initial examination of the stress
characteristics and failure mechanism of this joint. The
following main conclusions were obtained:
1) By means of experiments performed on the novel
joint and a traditional gusset joint, it is revealed that
because a cylindrical socket-and-spigot-type shear con-
nector is installed in the central area of the novel joint,
its overall stiffness and shearing resistance are signifi-
cantly higher than those of the traditional joint, and the
average increment of its out-of-plane anti-bending
bearing capacity is above 50 %.
2) Three different sets of aluminum-alloy joint tests
indicated that the failure modes of traditional gusset
joints are tensile and shear block failure of the lower
cover plate or bending failure of the aluminum-alloy
beams. The main failure mode of the novel joint is
torsional buckling failure of the connected aluminum-
alloy beams. The tensile and shear block failure of the
lower cover plate occurs only when the thickness of the
joint plate is small. The bearing capacity is not affected
greatly by the thicknesses of the upper and lower cover
plates.
3) The multiple-element contact-based elastoplastic
finite-element parametric analysis indicates that the
theoretical load–displacement curves of the novel joint
agree well with the measured experimental curves.
Although the bearing capacity of the joint increases with
the thickness of the connection plates, the increase is
very limited after the thickness exceeds that of the web.
Therefore, it is recommended that the thickness of the
connection plates be the same as that of the web.
4) A simplified formula for the ultimate bearing
capacity is offered based on the working mechanism of
the novel joint, providing the future structural design of
this novel joint with a theoretical basis.
Acknowledgments
This study was supported by the Natural Science
Foundation of China under Grant No. 51578136.
5 REFERENCES
1
L. P. Yang, S. Wei, Q. L. Zhang, Research status and key issues of
aluminum alloy spatial lattice structure, Journal of Building
Structure, 34 (2013) 2, 1–19, doi:10.1108/s0735-
004x20140000027001
2
R. Bradshaw, D. Campbell, M. Gargary, Special structures: past,
present, and future, Journal of Structural Engineering, 128 (2002)6,
691–709, doi:10.1108/s0735-004x20140000027001
3
EN 1999-1-1:2000: Design of aluminium alloy structures, General
rules for buildings, Europe
4
AS/NZS 1664.1: 1999, Aluminum structures, Part 1: Limit state
design, Australia and New Zealand
5
AS/NZS 1664.1: 1999, Aluminum structures, Part 1: Limit state
design, Australia and New Zealand
6
K. Sugizaki, S. Kohmura, Y. Hangai, Experimental study on
structural behaviour of an aluminum single-layer lattice shell,
Transactions of AIJ, 61 (1996) 480, 113–122, doi:10.3130/
aijs.61.113_1
7
Y. Hiyama, H. Takashima, T. Iijima, Experiments and analyses of
unit single layered reticular domes using aluminum ball joints for the
connections, Transactions of AIJ, 64 (1999) 518, 3–40, doi:10.3130/
aijs.64.33_1
8
S. Xu, Experimental study on the temco node and analysis of the
performance of aluminum alloy single-layer reticulated shell
structure, Diss, Tianjin University, 2015
9
G. Shi, C. Luo, Y. Q. Wang, Experimental study on mechanical
behavior of new cast aluminum joints in aluminum alloy reticulated
shell structures, Journal of Building Structure, 33 (2012) 3, 70–79,
doi:10.14006/j.jzjgxb.2012.03.010
10
G. Shi, C. Luo, Y. Q. Wang, Research on design method of new cast
aluminum joint bearing capacity in aluminum alloy reticulated shell
structure, Architectural Science, 16 (2000) 5, 78–84, doi:10.13849/
j.issn.1006-6578.2012.01.013
11
Y. Q. Wang, X. C. Liu, Y. J. Shi, J. Yin, Finite element analysis of
mechanical behavior of disc joints in aluminum alloy reticulated
shells, Journal of Tianjin University, Natural Science and Engi-
neering Technology Edition, 48 (2015)7,2–5
12
J. H. Qian, P. F. Zhao, C. X. Hao, Research on the structure of long
span aluminum alloy dome reticulated shell, Architecture Science,
16, (2000) 5, 7–12, doi:10.13614/j.cnki.11-1962/tu.2000.05.002
13
J. Zhang, The stiffness of reticulated joints and its effect on overall
stability, China, Shanghai Jiaotong University, 2003
14
L. W. Wang, W. Yang, Y. Feng, Design of aluminum alloy single-
layer reticulated shell structure for swimming fencing hall of China
modern Pentathlon Center, Building Construction, 40 (2010)9 ,
73–76, doi:10.19701/j.jzjg.2010.09.018
15
J. C. Zhao, H. M. Xu, Stress analysis of single-layer reticulated
shells in Shanghai Science and Technology museum, Industrial
Construction., 10, 31 (2001) 7–9, doi:10.13204/j.gyjz2001.10.003
16
S. Lai, Stress analysis of single-layer aluminum Reticulated shell
joints of 60 m diameter Tank top, Petrochemical Equipment Tech-
nology, 28 (2007) 4, 11–15
17
X. N. Guo, Z. Xiong, Y. F. Luo, W. Xu, Experimental study on
bearing capacity of aluminum alloy plate joints, Journal of Tongji
University, Natural Science Edition, 42 (2014) 7, 1024–1030
18
X. N. Guo, Z. Xiong, Y. F. Luo, Initial stiffness of aluminum alloy
plate joints, Journal of Tongji University, Natural Science Edition, 42
(2014) 8, 1162–1165
19
X. N. Guo, L. Q. Qiu, Experimental study on flexural bearing capa-
city of aluminum alloy plate joints, Journal of Hunan University,
Natural Science, 41 (2014) 47–53
20
C. Q. Zhao, J. Ma, Assembled honeycombed sheet light empty
stomach building and roof structure system, Patent, Application No.
200810100745.X, Publication No. CN101270595A
21
C. Q. Zhao, S. Du, J. Chen, Mechanical relationships between the
fiber-lamination methods and the structural functions of oryctes
rhinoceros horns, Materials and technology, 53 (2019) 1, 17–23,
doi:10.17222/mit.2018.119
22
C. Q. Zhao, W. D. Zheng, J. Ma, Y. J. Zhao, Lateral compressive
buckling performance of aluminum honeycomb panels for long-span
hollow core roofs, Materials, 9 (2016) 444, doi:10.3390/ma9060444
C. ZHAO et al.: MECHANICAL PROPERTIES OF A NOVEL JOINT OF A SINGLE-LAYER ALUMINUM-ALLOY ...
818 Materiali in tehnologije / Materials and technology 53 (2019) 6, 811–819

23
C. Q. Zhao, W. D. Zheng, J. Ma, Y. J. Zhao, Shear strength of
different bolt connectors on large span aluminium alloy honeycomb
sandwich structure., Appl. Sci., 7 (2017) 450, doi:10.3390/
app7050450
24
C. Q. Zhao, J. Ma, Y. J. Zhao, The stability of new single-layer
combined lattice shell based on aluminum alloy honeycomb panels.,
Appl. Sci., 7 (2017) 1150, doi:10.3390/app7111150
25
T. S. Kim, H. Kuwamura, Investigation on ultimate strength of
thin-walled steel single shear bolted connections with two bolts using
finite element analysis, Thin-Walled Struct, 47 (2009) 1191–1202,
doi:10.1016/j.tws.2009.04.009
26
T. S. Kim, H. Kuwamura, Finite element modeling of bolted connec-
tions in thin-walled stainless steel plates under static shear,
Thin-Walled Struct., 45 (2007) 407–421, doi:10.1016/j.tws.2007.
03.006
27
K. Fallahnezhad, A. Steele, R. H. Oskouei, Failure mode analysis of
aluminium alloy 2024-T3 in double-lap bolted joints with single and
double fasteners, A numerical and experimental study, Materials, 8
(2015) 3195–3209, doi:10.3390/ma8063195
28
M. Rezaeifard, S. J. Salami, M. B. Dehkordi, M. Sadighi, A new
nonlinear model for studying a sandwich panel with thin composite
faces and elastic–plastic core, Thin-Walled Struct., 107 (2016)
119–137, doi:10.1016/j.tws.2016.06.012
29
U. K. Mallela, A. Upadhyay, Buckling load prediction of laminated
composite stiffened panels subjected to in-plane shear using artificial
neural networks, Thin-Walled Struct., 102 (2016) 158–164,
doi:10.1016/j.tws.2016.01.025
30
S. Hassanli, B. Samali, Buckling analysis of laminated composite
curved panels reinforced with linear and non-linear distribution of
Shape Memory Alloys, Thin-Walled Struct., 106 (2016) 9–17,
doi:10.1016/j.tws.2016.04.022
31
R. P. Khandelwal, A. Chakrabarti, P. Bhargava, Effect of interfacial
imperfection on bending behavior of composites and sandwich
laminates by an efficient C0 FE model, Thin-Walled Struct., 82
(2014) 170–182, doi:10.1016/j.tws.2014.04.011
32
Y. C. Cai, B. Young, Bearing factors of cold-formed stainless steel
double shear bolted connections at elevated temperatures. Thin-
Walled Struct., 98 (2016) 212–229, doi:10.1016/j.tws.2015.04.021
33
Y. H. Cho, T. S Kim, Estimation of ultimate strength in single shear
bolted connections with aluminum alloys (6061-T6), Thin-Walled
Struct., 101 (2016) 43–57, doi:10.1016/j.tws.2015.11.017
34
B. Pan, Y. Shi, Y. Wang, The effect of yield-to-ultimate ratio on the
bolted connection under static shear, J. Shenyang Jianzhu Univ., 28
(2012), 501–506, doi:10.13614/j.cnki.11-1962/tu.2012.09.024
35
GB/T3098.6-2014: Mechanical properties of fasteners–stainless steel
bolts, screws and studs, China
C. ZHAO et al.: MECHANICAL PROPERTIES OF A NOVEL JOINT OF A SINGLE-LAYER ALUMINUM-ALLOY ...
Materiali in tehnologije / Materials and technology 53 (2019) 6, 811–819 819