See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/315635676

Behaviors of one-side bolted T-stub through thread holes under tension

strengthened with backing plate

Article  in  Journal of Constructional Steel Research · July 2017

DOI: 10.1016/j.jcsr.2017.03.010

CITATION READS

1 1,190

5 authors, including:

Peijun Wang
Shandong University
43 PUBLICATIONS   178 CITATIONS   

SEE PROFILE

All content following this page was uploaded by Peijun Wang on 24 November 2017.

The user has requested enhancement of the downloaded file.

 Journal of Constructional Steel Research 134 (2017) 53–65

Contents lists available at ScienceDirect

Journal of Constructional Steel Research

Behaviors of one-side bolted T-stub through thread holes under tension

strengthened with backing plate

Xulin Zhu a, Peijun Wang a,b,⁎, Mei Liu a, Wulan Tuoya a,b, Shuqing Hu b
a
Civil Engineering College of Shandong University, Jinan, Shandong Province 250061, China
b
Zhongtong Steel Construction Co. Ltd., Liaocheng, Shandong Province 252000, China

a r t i c l e i n f o a b s t r a c t

Article history: Tests were carried out to investigate the one-side bolted T-stub through thread holes under tension strengthen
Received 24 January 2017 with backing plate. For the thread length is limited by the flange thickness of T-stub, which might lead to the pre-
Received in revised form 14 March 2017 mature failure of the connection due to the failure of thread. The thread length could be increased by adding a
Accepted 16 March 2017
backing plate. Two new failure models of the bolted T-stub were proposed to consider the potential failure of
Available online 24 March 2017
hole thread. And the design methods for tension strength of corresponding failure modes were presented. Stud-
Keywords:
ied parameters included the bolt pre-tension force and the T-stub flange thickness. The bolt pre-tension force al-
Tension behavior most did not affect the tension strength of a bolted T-stub, while the initial stiffness could be improved. Effects of
One-side bolt the backing plate on the tension strength and load-displacement curve of T-stub under tension were presented
Bolted T-stub through comparing those without backing plate. Test results showed that the backing plates could be efficiently
Thread holes improve the tension strength of the T-stubs. However, for T-stubs with same screw depth, it was more efficient to
Screwed backing plates increase the flange thickness directly. The comparison between test results and calculated results from design
Design method methods indicated that the proposed design methods could be used to predict the tension strength and failure
modes of the one-side bolted T-stub through thread holes and strengthened with screwed backing plates.
Under the design load, the bolted T-stub was still in elastic state, which could meet the design requirement.
© 2017 Elsevier Ltd. All rights reserved.

1. Introduction
Bolted endplate connections are widely used in steel-framed struc-
tures for it avoids welding on site, which in general requires highly
skilled labor and slows down the construction speed [1]. Hollow sec-
tions are generally more efficient as structural members than conven-
tional open sections due to their superior torsional rigidity and hence
resistance to flexural-torsional and torsional buckling modes. However,
the application of the hollow section component has traditionally been
hampered by access constraints for a fully bolted connection. The blind
bolt provided an alternate way to connect beam to a tubular column
using bolted endplate, which could be inserted and tightened from out-
side of the tube.
Application of the blind-bolt, such as the Hollo-bolt, the Enhanced
Hollo-bolt (EHB), the Blind Bolt, the Flow drill [2] and so on, in bolted
beam-column connections has been widely studied recently. Wang and
Wang [3] carried out tests to investigate the yield and ultimate strengths
of beam-tubular column connections using Hollo-bolts under tension.
⁎ Corresponding author at: Civil Engineering College of Shandong University, Jinan,
Shandong Province 250061, China.
E-mail address: Pjwang@sdu.edu.cn (P. Wang).
Design recommendations for the beam endplate using blind bolted
were proposed. Tizani et al. [4] studied the performance and reliability
of EHB connection to concrete filled columns under cyclic loading. Test re-
sults showed that the connection could provide energy dissipation capac-
ity and ductility appropriate for its potential use in seismic design.
The mechanism of blind bolt discussed above is through tightening
the bolt head with appropriate torque. The threaded cone flared the
sleeves and clamped against the inside of the hole, which was analogous
to the clamping produced by the nut of standard bolt.
Except for using flared sleeves, the bolt can also be clamped by
thread holes. In space grid structures with bolted ball joints, the grid
members are connected using bolts that are screwed in to a hole with
thread, as shown in Fig. 1. The design codes require that the thread
length should be longer than 1.1 times of bolt diameter [5–8]. For a
steel plate being thinner than 12.5 mm, the thread holes could be
manufactured by flow drilling technique [9,10]. However, the bolted
beam-tubular column endplate connection using thread holes has not
been fully investigated yet.
The thread length is limited by the wall thickness. For a tubular col-
umn with thin walls, the failure of thread might lead to the premature
failure of the connection. The thread length could be increased by
adding a backing plate, as shown in Fig. 2.

http://dx.doi.org/10.1016/j.jcsr.2017.03.010
0143-974X/© 2017 Elsevier Ltd. All rights reserved.
54 X. Zhu et al. / Journal of Constructional Steel Research 134 (2017) 53–65

Fig. 1. Space grid net structures with bolt ball connection.

Fig. 2. Bolted beam-tubular column connection with backing plates.

Fig. 3. Dimensions of studied T-stubs.

 X. Zhu et al. / Journal of Constructional Steel Research 134 (2017) 53–65 55

Fig. 4. Details of studied bolts.

components, i.e. failure of the thread in the hole and effects of plastic fail-
Table 1
Parameters of studied T-stubs. ure of the T-stub on the thread. Two new failure modes might occur and
should be investigated, which were (1) the hole thread failure and (2) the
T-stubs with screwed backing plates T-stubs without screwed backing plates
T-stub flange yielding accompanied with hole thread failure.
No. t/mm Pre-tension Screw No. t/mm Pre-tension Screw In this paper, tension behaviors of the one-side bolted T-stub
force (P) depth force (P) depth
through thread holes strengthened by backing plate were investigated
kN tc/mm kN tc/mm
T06-0 6 0 16
experimentally and theoretically. The two new failure models and the
T06-70 70 corresponding design methods were proposed. Influences of parame-
T06-140 140 T06-140C 6 140 6 ters including the bolt pre-tension force and the T-stub flange thickness
T08-0 8 0 18 on failure modes and load-displacement curves of the T-stub connec-
T08-70 70
tions were presented. Comparisons of tension strength between T-
T08-140 140 T08-140C 8 140 8
T10-0 10 0 20 stubs with and without backing plates were conducted. Tension
T10-70 70 strength of bolted T-stubs obtained from tests and design equations
T10-140 140 T10-140C 10 140 10 were compared to show the applicability of one-side bolts using thread
T12-0 12 0 22
holes.
T12-70 70
T12-140 140 T12-140C 12 140 12
T14-0 14 0 24 2. Test set-up
T14-70 70
T14-140 140 T16-140C 16 140 16
T16-0 16 0 26 2.1. Test specimens
T16-70 70
T16-140 140 T18-140C 18 140 18 Specimens in the tests were designed as T-stubs, as shown in Fig. 3.
T18-0 18 0 28
Three parts were included in the tested T-stubs:
T18-70 70
T18-140 140 T20-140C 20 140 20 (1) the M20 high-strength bolts of Class 8.8;
T20-0 20 0 30 (2) the base T-stub with flange thickness of 30 mm and web thick-
T20-70 70 ness of 20 mm to ensure that it was kept in elastic state. The hole diam-
T20-140 140 T22-140C 22 140 22 eter on the flange was 22 mm;
Total 24 Total 8
(3) the target T-stub with the flange thickness varying from 6 mm to
20 mm to create different failure modes. The web thickness was 20 mm to
In bolted endplate connections, the endplate could be simplified as an ensure it would be in elastic state in the test. The backing plate thickness
equivalent T-stub. The equivalent T-stub in tension might be used to was all 10 mm. The backing plates were welded to the T-stub flange by
model the design resistance of the column flange in bending, the end- fillet weld first, and then drilling and tapping with the flange together.
plate in bending, the flange cleat in bending and the base plate in bending The purpose of adding a backing plate to the T-stub flange was to in-
under tension, etc. [11]. T-stubs connected through standard bolts with crease the screw length and improve loading capacity of the thread. The
nuts had been wildly studied [12–15]. In general, bolted T-stubs will en- total depth of the threads tc = t + 10, which varied from 16 mm to
counter three failure modes, which are (1) bolt failure, (2) T-stub flange 30 mm.
yielding failure, and (3) T-stub flange yielding with bolt failure. However, Types of the thread in the hole were corresponding to the bolt used
this knowledge could not be applied directly to connections to tubular in the tests, as shown in Fig. 4. D = 20 mm and D1 = 17.3 mm were the
members using one-side bolts with thread holes. This was due to the external diameter and the internal diameter of bolt, respectively. P =
lack of full understanding of the behavior of two most relevant 2.5 mm was the pitch of screw thread. H = 0.866P was the thread

Table 2
Properties of steel used in tests.

Components Yield strength (fy/MPa) Ultimate tensile strength (fu/MPa) Elastic modulus (E/GPa) Poisson's ratio (υ)

Base T-stub 285.3 499.1 206 0.3

Target T-stub with thickness of 6 mm 302.5 471.3
Target T-stub with thickness of 12 mm 282.8 438.5
Target T-stub with thickness of 18 mm 270.8 437.6
High-strength bolts 640 800
Washer 640 800
56 X. Zhu et al. / Journal of Constructional Steel Research 134 (2017) 53–65

Fig. 5. Test set up.

Fig. 8. Failure Mode 1: bolt failure.

triangle height [16]. Tension behaviors of T-stubs with and without

screwed backing plates were compared through experiment for the T-
stubs with flange thickness t ≤ 12 mm, as shown in Fig. 3(b).
Dimensions of the T-stubs were shown in Fig. 3. Parameters of the tar-
get T-stub with screwed backing plates were listed in Table 1. Properties
of the steel used in tests were listed in Table 2 from the coupon tests.
2.2. Test set-up
A universal servo-controlled hydraulic testing machine with loading
capacity of 1000 kN was used to test the T-stubs, as shown in Fig. 5(a).
Before the tension tests, the high-strength bolts were tightened to the
designed pre-tension force through a presetting torque wrench, as
shown in Fig. 5(b). Three different levels of pre-tension forces were ap-
plied, which were 0 kN, 70 kN and 140 kN. P = 140 kN was the designed
pre-tension force according to the China design code [17]. The bolt used
in the test was the M20 high-strength bolt of Class 8.8 with the yield
strength fyb = 640 MPa. The pre-stresses of the three levels of pre-ten-
sion forces were 0 MPa, 285.7 MPa and 571.4 MPa respectively. The cor-
responding pre-stress levels were 0, 0.446 and 0.893, respectively. The
loading speed was 0.05 kN/s.
3. Preliminary design of tested T-stubs
3.1. Design methods in Eurocodes
The EC3-1-8 [11] provides design methods for bolted endplate
beam-column connections. The tension resistance of a bolted T-stub
with backing plate can be obtained through the design methods, as

Fig. 6. Column flange with backing plates.

Fig. 7. Prying force in T-stubs under tension.

 X. Zhu et al. / Journal of Constructional Steel Research 134 (2017) 53–65 57

Fig. 11. Failure Mode 3: flange yielding with bolt failure.

Fig. 9. Failure Mode 2: flange yielding.

shown in Fig. 6, where hbp and bbp are the height and width of the back-
ing plates, respectively.
For the T-stub in tension, EC3-1-8 [11] gives three different fail-
ure modes: (1) Mode 1, bolt failure; (2) Mode 2, yielding of the
flange; and (3) Mode 3, flange yielding with bolt failure, in which
the prying effects are taken into consideration in the design equa-
tions, as show in Fig. 7. Ft is the tension force applied to the web of
the T-stub. Q is the prying force. Pb is the actual tension force in the
bolt [18,19]. And FT,Rd is the design tension strength of the T-stub.
Design equations for the three failure modes were given below. The
tension resistance of the T-stub takes the minimum value of FT,1,Rd,
FT,2,Rd and FT,3,Rd.

3.1.1. Tension strength of T-stub failed by Mode 1: bolt failure

As indicated in Fig. 8, the tension strength of a T-stub failed by Mode
1 is calculated by
Fig. 12. Failure Mode 4: hole thread failure.

F T;1;Rd ¼ ∑F t;Rd ð1Þ

where ∑Ft,Rd is the total design tension resistance of the bolts in the T-
k2 f ub As stub. fub is the ultimate tensile strength for bolts. As is the effective area
F t;Rd ¼ ð2Þ
γ M2 of bolt. The coefficients k2 = 0.9 and γM2 = 1.25.

Fig. 10. Yielding patterns of T-stub flange.

58 X. Zhu et al. / Journal of Constructional Steel Research 134 (2017) 53–65

3.1.2. Tension strength of T-stub failed by Mode 2: complete yielding of the

flange
As indicated in Fig. 9, the tension strength of a T-stub failed by Mode
2 is calculated by

4Mpl;2;Rd þ 2Mbp;Rd
F T;2;Rd ¼ ð3Þ
m

Mpl;2;Rd ¼ 0:25∑leff ;2 t 2f f y =γM0 ð4Þ

Mbp;Rd ¼ 0:25∑leff ;2 t 2bp f y;bp =γ M0 ð5Þ

where m is geometrical parameter of the T-stub joints indicated in Fig. 9.

γM0 = 1.0 is the partial factor. ∑leff is the total effective length of the T-
stub flange in the joint. tf and fy are the thickness and the yield strength
of the T-stub flange. tbp and fy,bp are the thickness and the yield strength
Fig. 13. Failure Mode 5: flange yielding with hole thread failure. of the backing plates, respectively.

Fig. 14. Threads under bending and shear.

Fig. 15. Failure modes of T-stubs with screwed backing plates.

 X. Zhu et al. / Journal of Constructional Steel Research 134 (2017) 53–65 59

Fig. 16. Failure modes of T-stubs without backing plates.

Table 3
Failure modes and tension resistance for T-stubs with different flange thickness t.

No. Screw depth tc/mm Tension strength (FT,Rd) / kN

Failure modes Calculated results Tested results

Tested results Calculated results Mode 1 Mode 2 Mode 3 Mode 4 Mode 5 Ultimate strength Yield strength

T06-0 16 Mode 2 Mode 2 282.2 33.3 137.4 152.8 76.1 183.1 103
T06-70 173.4 106
T06-140 178.9 100
T06-140C 6 Mode 2 Mode 2 282.2 13.9 137.4 62.6 33.3 67.6 38
T16-140C 16 Mode 2 Mode 2 282.2 99.1 159.8 167 105.2 166.3 135
T08-0 18 Mode 2 Mode 2 282.2 44.1 140.2 171.9 87.9 195.9 118
T08-70 201.3 115
T08-140 186.8 113
T08-140C 8 Mode 2 Mode 2 282.2 24.8 140.2 83.5 46.1 103.1 63
T18-140C 18 Mode 5 Mode 5 282.2 125.4 166.7 187.9 122 223.3 165
T10-0 20 Mode 2 Mode 2 282.2 58.1 143.9 191 100.7 232.8 155
T10-70 232.4 160
T10-140 223.1 158
T10-140C 10 Mode 2 Mode 2 282.2 38.7 143.9 104.4 59.6 114.6 75
T20-140C 20 Mode 5 Mode 5 282.2 154.8 174.4 208.8 139.6 285.2 240
T12-0 22 Mode 2 Mode 2 282.2 75.1 148.4 210.1 114.2 257.6 180
T12-70 248.1 183
T12-140 241.6 182
T12-140C 12 Mode 2 Mode 2 282.2 55.7 148.4 125.3 74.0 131.3 95
T22-140C 22 Mode 5 Mode 5 282.2 187.3 183 229.7 158.1 330.5 260
T14-0 24 Mode 2 Mode 2 282.2 95.2 153.7 229.2 128.5 261.1 210
T14-70 257.4 200
T14-140 267.4 215
T16-0 26 Mode 2 Mode 2 282.2 118.4 159.8 248.3 143.7 312.4 250
T16-70 309.1 260
T16-140 314.4 253
T18-0 28 Mode 2 Mode 2 282.2 144.7 166.7 267.4 159.7 342.8 283
T18-70 340.9 285
T18-140 340.9 284
T20-0 30 Mode 5 Mode 2 282.2 174.2 174.4 286.5 176.5 353.4 326
T20-70 351.6 313
T20-140 354.1 312

The bold text in the table was the calculated design tension strength corresponding to the failure mode observed in the test.
60 X. Zhu et al. / Journal of Constructional Steel Research 134 (2017) 53–65

Fig. 17. Deformation of stiffened T-stub with backing plate at different load level (t = 6 mm).

Fig. 18. Deformation of T-stub without backing plate at different load level (t = 6 mm).

SCI P398 [20] provides equations to calculate the efficient yield length
leff,cp and leff,nc according of the yield patterns of the T-stub, as shown in
Fig. 10. For the tested T-stubs here, e = 45 mm, ex = 45 mm and mx =
50 mm, the calculate yield length is leff = 90 mm and the corresponding
yield pattern is the double curvature pattern, as shown in Fig. 10(d).
3.1.3. Tension strength of T-stub failed by Mode 3: flange yielding with bolt
failure
As shown in Fig. 11, the tension strength of a T-stub failed by Mode 3
is calculated by
2Mpl;3;Rd þ e∑F t;Rd
F T;3;Rd ¼
mþe
where m and e are geometrical parameters of the T-stubs indicated in Fig.
11, ∑Ft,Rd is the total design tension resistance for the bolts in the T-stub.
3.2. New failure modes of one-side bolted T-stub with thread holes
For the one-side bolted T-stub with thread holes, two new failure
modes may occur: (1) Mode 4, hole thread failure, as shown in Fig.
12; and (2) Mode 5, flange yielding with hole thread failure, as shown
in Fig. 13, where Fs,Rd is the design resistance of the threads.
3.2.1. Tension strength of T-stub failed by Mode 4: hole thread failure
For failure Mode 4, as shown in Fig. 12, the tension resistance of the
ð6Þ T-stub can be calculated by:

Mpl;3;Rd ¼ 0:25∑leff ;3 t 2f f y =γM0 ð7Þ F T;4;Rd ¼ ∑F s;Rd ð8Þ

Fig. 19. Deformation of T-stub with additional nuts (t = 6 mm).

 X. Zhu et al. / Journal of Constructional Steel Research 134 (2017) 53–65 61

And Fs,2,Rd is the bending strength of a thread,

Wf y;p
F s;2;Rd ¼ ð11Þ
b

Av ¼ πDh ð12Þ

2
πDh
W¼ ð13Þ
4

Shear failure of a thread is assumed to be occurred at the section of

thread root. For calculating the bending strength of a thread, the plastic
bending modulus of the section W is adopted. fyv,p and fy,p are the design
shear stress and yield stress of the T-stub flange respectively. Av is the
efficient shear area of one circle of the internal threads. D is the external
diameter of the thread. b and h are geometric parameters, as shown in
Fig. 14(b). For threads studied in this paper, h = 7P/8 and b = (D −
Fig. 20. Load-displacement curves of stiffened T-stub with flange thickness of 10 mm.
D1) / 2 = 5H/8. The bending strength ∑ Fs,2,Rd was smaller than the
shear strength ∑Fs,1,Rd, which showed that the internal threads in T-
stubs was failed in bending failure.


F s;Rd ¼ min ∑F s;1;Rd ; ∑F s;2;Rd ð9Þ 3.2.2. Tension strength of T-stub failed by Mode 5: flange yielding with hole
thread failure
For failure Mode 5 shown in Fig. 13, the tension strength of T-stub
Fs,1,Rd is the shear strength of a thread, can be calculated by:

2Mpl;5;Rd þ e∑F s;Rd

F s;1;Rd ¼ Av f yv;p ð10Þ F T;5;Rd ¼ ð14Þ
mþe

Fig. 21. Load-displacement curves of stiffened T-stubs with different bolt pre-tension force.
62 X. Zhu et al. / Journal of Constructional Steel Research 134 (2017) 53–65
Fig. 21 (continued).
Mpl;5;Rd ¼ 0:25∑leff ;5 t 2f f y =γM0
4. Tests results and discussion
4.1. Failure modes of one-side bolted T-stubs with thread holes
As shown in Fig. 15, two failure modes were observed form tests
with changing of T-stub flange thickness, which were (1) failure Mode
2, complete yielding of the T-stub flange and (2) failure Mode 5, flange
yielding with hole thread failure.
For T-stubs with flange thickness t = 20 mm, the two bolts were also
stretched and necked, as shown in Fig. 15(b), which indicated that the
bolts also reached their tension ultimate state.
Tested results on T-stubs without backing plates were also present-
ed, as shown in Fig. 16. Compared with the T-stub with the same flange
thickness, it is apparently that the backing plate increased the yield
strength of the flange, as shown in Fig. 16(a). However, effects of the
adding a backing plate is far behind increase the flange thickness direct-
ly, as shown in Fig. 16(b). The T-stub with flange thickness of 16 mm de-
veloped a much smaller deformation, compared with the T-stub with
flange thickness of 6 mm and backing plate thickness of 10 mm.
Detail failure modes and tension strengths of every T-stub with
backing plates obtained from design methods and tests were listed in
Table 3. The bold text in the table was the design tension strength
ð15Þ corresponding to the failure mode observed in the test. It could be
seen that the calculated failure modes agreed well with that observed
in the tests. The calculated tension strengths of T-stub with screwed
backing plates were all lower than the tested tension resistance. For
the T-stub with flange thickness of 20 mm, the calculated values for fail-
ure Mode 2, Mode 3 and Mode 5 were almost the same, which could ex-
plain why the tendency of flange yielding (Mode 2) accompanied by the
necking and stretching of the bolt shank (Mode 3) were also observed,
as shown in Fig. 15(b).
The tested ultimate strength presented in this paper was the maxi-
mum tension force that observed from the load-displacement curve.
That is, the failure criterion adopted in determination of the ultimate
tension strengths in model test was that when the tension force reached
the apex of the load-displacement curves.
From the load-displacement curves of the T-stub connection, the
yielding occurred in the T-stub flange when the slope of the curve
started to reduce. The corresponding yielding load was about 80% of
the ultimate strength; and it was listed in Table 3.
Deformation of T-stub with backing plates at different load level was
shown in Fig. 17, where the T-stub flange thickness was 6 mm. Test re-
sults of T-stubs without backing plate and with an additional nut were
also presented in Figs. 18 and 19 for comparison.
Under the load calculated from design equations [20], there was
no out-of-plane deformation of the T-stub, as shown in Fig. 17(b), in-
dicating that the T-stub was in elastic state and it had sufficient load-
ing bearing capacity to carry the design load. With the increasing of
 X. Zhu et al. / Journal of Constructional Steel Research 134 (2017) 53–65
Fig. 22. Load-displacement curves of stiffened T-stubs with different flange thickness.
tension force, large deformation in T-stub flange and screwed back-
ing plates developed. In the end, the T-stub failed in complete yield-
ing of the flange.
Comparing with the T-stub without backing plate, the existence of
backing plate greatly improves the load bearing capacity of the T-stub,
as shown in Fig. 18. And the effects of a backing plate could be as good
as a nut, as shown in Fig. 19.
63
4.2. Load-displacement curves
Load-displacement curves of T-stubs with screwed backing plates
were shown in Fig. 20, where the T-stub flange thickness was 10 mm.
Displacement s was the axial displacement along the tension force.
From the load-displacement curves, it could be found that the bolted
T-stub experienced three stage, which were (1) the elastic stage; (2) the
plastic stage and (3) the failure stage. In the elastic stage, the axial ten-
sion load increased linearly with the increase in the axial displacement,
and the design load should not exceed this stage. When the load
reached about 80% of the ultimate strength, the tension displacement
increased quickly, which indicated that the T-stubs entered to the plas-
tic stage. The one-side bolted T-stub could develop a large plastic defor-
mation, as shown in Fig. 20, which showed that it had sufficient
ductility.
Corresponding loads to the five failure modes were also illustrated in
Fig. 20. In the tests, failure Mode 2 was observed for T-stub with flange
thickness of 10 mm and backing plate thickness of 10 mm, as listed in
Table 3, which was coincide with the calculated results. The T-stub
was still in elastic state under the design load.
5. Parameter studies on one-side bolted T-stub with backing plate
Influences of the bolt pre-tension force and the thickness of T-stub
flange on the load-displacement curves of a bolted T-stub through
thread holes with backing plate were investigated through tests. Param-
eters of every specimen were listed in Table 1.
5.1. Effects of bolt pre-tension force
Effects of the bolt pre-tension force on load-displacement curves
of the T-stubs with backing plates were displayed in Fig. 21. Three
levels of bolt pre-tension forces were studied, which were 0 kN,
70 kN and 140 kN. The 140 kN was the design pre-tension force for
a M20 Class 8.8 high strength bolt. Load-displacement curves of the
T-stub with flange thickness of 6 mm, 8 mm, 10 mm, 12 mm,
14 mm, 16 mm, 18 mm and 20 mm were shown in Fig. 21(a)–(h),
respectively.
The pre-tension force almost did not affect the ultimate tension
strength of bolted T-stubs through thread holes with backing plate,
while the initial stiffness of the connections would be improved slightly
by applying a higher pre-tension force.
All the bolted T-stubs could develop a large deformation after the
tension forces exceeded the yield forces under the three levels of
pre-tension force, which indicated that the bolted T-stubs through
thread holes could meet the ductile requirement of the beam-col-
umn joints.
5.2. Effects of flange thickness
Load-displacement curves of the T-stubs with different flange thick-
ness were shown in Fig. 22. The flange thickness studied varied from
6 mm to 20 mm; and totally 8 plate thicknesses were studied. Both
the tension resistance and the stiffness of the T-stub joints could be im-
proved by increasing the flange thickness.
Design loads of the bolted T-stubs calculated by method in P398 [20]
were also marked on the load-displacement curves, which showed that
the bolted T-stubs were still in elastic state under the design load. The
ultimate load of T-stubs through thread holes was much greater than
the design load. However, with the increase in T-stub flange thickness,
the ultimate tension displacement reduced.
5.3. Effects of backing plates
Except for increasing the thread length, the backing plate could also
increase the yield strength and stiffness of the T-stub flange.
64 X. Zhu et al. / Journal of Constructional Steel Research 134 (2017) 53–65
Fig. 23. Load-displacement curves of T-stubs with and without backing plate (P = 140 kN).
Comparison of load-displacement curves between T-stubs with and
without screwed backing plates were shown in Fig. 23. Design loads of
the bolted T-stubs calculated by method in P398 [20] were also marked
on the load-displacement curves.
For T-stubs with same flange thickness, the backing plates improved
the tension strengths significantly. For the T-stub with flange thickness
of 6 mm and without backing plate, the tested ultimate tension strength
was only 67.6 kN. However, through adding a backing plate with thick-
ness of 10 mm, the tested ultimate tension strength increased to
178.9 kN, as shown in Fig. 23(a). For the T-stub with flange thickness of
8 mm, 10 mm and 12 mm, the tested ultimate strengths were also greatly
improved through adding backing plate. However, the T-stub flange
yielded in double curvature pattern, as shown in Fig. 10(d). There were
two yield lines on the T-stub flange: one across the bolt hole and the
other along the edge of web. The yield strength and stiffness of the flange
along the edge of web could not be increased by the backing plate. The ul-
timate strengths could be more efficient increased with the increase in T-
stub flange thickness directly. For the T-stub with flange thickness of
8 mm and backing plate with thickness of 10 mm, the total thread length
was 18 mm. The tested ultimate strength was 186.8 kN. However, for the
T-stub with flange thickness of 18 mm, the tested ultimate strength was
much greater as 223.3 kN, as shown in Fig. 23(b).
6. Conclusions
Tests were carried out to investigate tension behavior of stiffened
one-side bolted T-stub with thread holes and stiffened using screwed
backing plates. Two new failure modes and the corresponding design
equations to calculate tension strength of the one-side bolted T-stub
were proposed. Effects of the bolt pre-tension force and the T-stub
flange thickness on failure modes and load-displacement curves were
discussed. Tension strengths of T-stubs with screwed backing plates
were compared with that without backing plates. Main conclusions of
this paper were summarized as follows:
(1) Two new failure modes for stiffened one-side bolted T-stub
with internal holes were proposed, which were (a) the hole
thread failure and (b) the flange yielding with hole thread failure.
Corresponding design equations for tension strength were
presented.
(2) Two failure modes were observed in the tests, which were (a)
Mode 2, the flange yielding for T-stubs with flange thickness t
b20 mm, and (b) Mode 5, the flange yielding with thread failure for T-
stubs with flange thickness t = 20 mm.
(3) The bolt pre-tension force almost did not affect the tension
strength of the T-stub, while it could improve the initial stiffness of
the connection. T-stubs with thicker flange had higher tension strength
and stiffness.
(4) The backing plates could efficiently improve the tension strength
of the bolted T-stub connection. However, it is more efficient to increase
the flange thickness directly.
(5) The proposed design equations could be used to predict tension
strength and failure modes of the bolted T-stubs through thread holes
and strengthen using strewed backing plate.
Acknowledgement
The authors wish to acknowledge the support from the National
Natural Science Foundation of China (51578322, 51608305), the Natu-
ral Science Foundation of Shandong Province (ZR2015EM041,
2016GGX103014) for the work reported in this paper.
 X. Zhu et al. / Journal of Constructional Steel Research 134 (2017) 53–65
References
[1] Y. Hu, L. Shen, S. Nie, B. Yang, W. Sha, FE simulation and experimental tests of high-
strength structural bolts under tension, J. Constr. Steel Res. 126 (2016) 174–186.
[2] W. Tizani, A. Al-Mughairi, J.S. Owen, T. Pitrakkos, Rotational stiffness of a blind-
bolted connection to concrete-filled tubes using modified Hollo-bolt, J. Constr.
Steel Res. 80 (1) (2013) 317–331.
[3] Z.Y. Wang, Q.Y. Wang, Yield and ultimate strengths determination of a blind bolted
endplate connection to square hollow section column, Eng. Struct. 111 (2016)
345–369.
[4] W. Tizani, Z.Y. Wan, I. Hajirasouliha, Hysteretic performance of a new blind bolted
connection to concrete filled columns under cyclic loading: an experimental inves-
tigation, Eng. Struct. 46 (1) (2013) 535–546.
[5] S.A. Mostafavian, M.R. Davoodi, J.V. Amiri, Ball joint behavior in a double layer grid
by dynamic model updating, J. Constr. Steel Res. 76 (76) (2012) 28–38.
[6] G. Yan, C. Fang, R. Feng, X. Hua, Y. Zhao, Detection of member overall buckling in
civil space grid structures based on deviation in normal strain along the member,
Eng. Struct. 131 (2017) 599–613.
[7] Construction industry standard of People's Republic of China, Bolted Spherical Node
of Space Grid Structures. JGT 10, 2009.
[8] Industry standard of the People's Republic of China, Technical Specification for
Space Frame Structures. JGJ 7-2010, China building industry press, Beijing, 2010.
[9] J.E. France, J.B. Davison, Patrick A. Kirby, Strength and rotational stiffness of simple
connections to tubular columns using flowdrill connectors, J. Constr. Steel Res. 50
(1) (1999) 15–34.
View publication stats
65
[10] Steel Construction Institute, PUB 358 Joints in Steel Construction: Simple Joints to
Eurocode 3, 2014.
[11] European Committer for Standardization, Eurocode3: Design of Steel Structures,
Part 1–8: Design of Joints. BS EN 1993-1-8, 2005.
[12] V. Piluso, C. Faella, G. Rizzano, Ultimate behavior of bolted T-stubs – II. Model valida-
tion, J. Struct. Eng. 127 (6) (2001) 694–704.
[13] A.M.G. Coelho, F.S.K. Bijlaard, N. Gresnigt, L.S.D. Silva, Experimental assessment of
the behaviour of bolted T-stub connections made up of welded plates, J. Constr.
Steel Res. 60 (2) (2004) 269–311.
[14] A.B. Francavilla, M. Latour, V. Piluso, G. Rizzano, Simplified finite element analysis of
bolted T-stub connection components, Eng. Struct. 100 (2015) 656–664.
[15] M.S. Zhao, C.K. Lee, S.P. Chiew, Tensile behavior of high performance structural steel
T-stub joints, J. Constr. Steel Res. 122 (2016) 316–325.
[16] National Standard of People's Republic of China, General Propose Metric Screw
Threads - Basic Profile. GB 192-2003, China building industry press, Beijing, 2003.
[17] National Standard of People's Republic of China, Code for Acceptance of Construc-
tion Quality of Steel Structures. GB 50205-2001, China building industry press, Bei-
jing, 2001.
[18] S.F. Chen, Principles of Steel Structure Design, The Science Publishing Company, Bei-
jing, 2005.
[19] H. Agerskov, High-strength bolted connections subject to prying, J. Struct. Div. ASCE
102 (10) (1976) 2066–2068.
[20] Steel Construction Institute, PUB 398 Joints in Steel Construction: Moment-resisting
Joints to Eurocode 3, 2013.