Sie sind auf Seite 1von 8

A Prying Action Force and Contact Force Estimation Model for a T-Stub Connection with

High-Strength Bolts

Jae-Guen Yang*1, Jae-Ho Park2, Hyun-Kwang Kim2 and Min-Chang Back2

1
Professor, Department of Architectural Engineering, College of Engineering, Inha University, Korea
2
Graduate Student, Department of Architectural Engineering, College of Engineering, Inha University, Korea

Abstract
A T-stub connection with high-strength bolts under tensile force is affected by prying action force and
the contact force, among others, between members. If a design equation that does not consider such prying
action force and contact force between members is not proposed, the T-stub under tensile force is liable to
be fractured under a strength lower than the estimated design strength. To prevent this, many studies have
proposed contact force estimation equations between members as well as the prying action force of the
T-stub connection with high-strength bolts. However, no design equations based on such research have been
proposed in South Korea. Therefore, this study aims to propose an estimation model for more accurate prying
action force and contact force, and to improve on previously proposed estimation models by implementing
three-dimensional, nonlinear finite element analysis.

Keywords: prying action force; contact force; design strength; connection stiffness

1. Introduction force acquired from the action test. Assuming that the
Bolted connections under tensile force are liable to central line of the axial bolt force moves by the contact
be fractured under a strength lower than the estimated force between members, Swanson (1999, 2002)
design strength due to the prying action force or proposed an estimation model, which improved the
contact force between members. Thus, there have estimation model by Struik and Back. However, the
been many studies on estimating such prying action application of the estimation model has become more
force and contact force between members, which have complex. Jaspart and Maquoi (1995) observed the
proposed various estimation models. change in contact force between connection members
Douty and McGuire (1965) estimate prying according to bolt pretension, and showed that such
action force based on elastic analysis. In proposing a change affects the tensile stiffness of connections
an estimation model, they considered the material under tensile force. Faella et al. (1996, 1998) observed
properties of high-strength bolts and connecting the change in contact force between connection
members as well as the geometric shape of connections. members according to the change in the stiffness ratio
Similarly, Agerskov (1976) estimated prying action () between the flexural stiffness of the connecting
force based on elastic analysis. Agerskov's estimation member flanges and the axial tensile stiffness of high-
model considered the deformation of T-stub flanges strength bolts, and estimated the change in the axial
connected with high-strength bolts and shear tensile stiffness of the connections.
deformation, but did not consider non-linearity based This study aimed to improve the previous contact
on flange yielding. Struik and Back (1969) estimated force estimation models between members and the
the prying action force of connections that have prying action force. Toward this end, it performed the
generally geometric shapes based on the combination three-dimensional non-linear finite element analysis.
of high-strength bolts and T-stub flanges. Struik and The feasibility of the proposed estimation model was
Back's estimation model, which is most widely used, verified after a comparison and review of the analysis
however, offers a value larger than the prying action results of other researchers.

2. Three-Dimensional Non-linear Finite Element


*Contact Author: Jae-Guen Yang, Professor, Department of Analysis Modeling of Connections with High-
Architectural Engineering, College of Engineering, Strength Bolts
Inha University, #253 Young Hyun-Dong, Nam-Ku, The prying action force of the connection with high-
Incheon, 402-751, South Korea strength bolts and the contact force between members
Tel: +82-32-860-7588 Fax: +82-32-866-4624 were estimated by the three-dimensional non-linear
E-mail: jyang@inha.ac.kr finite element analysis. Generally, the contact effect of
( Received October 10, 2012 ; accepted July 23, 2013 ) the members and the prying action is largest when the

Journal of Asian Architecture and Building Engineering/September 2013/316 309


connecting flange is thinnest, and the gauge distance
between high-strength bolts is largest. Based on Kulak's
research, Thornton (1985) and Astaneh (1985) proposed
parameter ', as in Eq. (1), to determine the effect of
prying action. In proposing Eq. (1), , the parameter
that shows the effect of the bolt hole for high-strength
bolts on the elements under prying action force, is the
ratio between the net section area of the flange and the
gross sectional area, and is the ratio of the distance
Fig.1.Fig. 1. Geometric shape of the connection
Geometric Shape ofthethe Connection
between the central line of the axial load of the high- Fig. 1. Geometric shape of connection

strength bolt and the distance between the central line of


the axial load of the high-strength bolt to the T-stub stem
surface. The prying action effect is highly likely to occur
when ', which is affected by the geometric shape of the
connection, the pretension force of the high-strength bolts,
and the material properties of the connection are larger
than 1.0. Therefore, as shown in Fig.1. and Table 1., this
study selected the geometric shape of the connection with
two ' values such as ' > 1.0 and 0 < ' < 1.0, respectively.
The steel material used in the connection is SS400, and as
shown in Fig.2., the material properties are in a bilinear Fig.
Fig. 2.2.Stress-deformation
Stress-deformationcurve
curve of
of steel
steel material
material
Fig.2. Stress-Strain Curve of Steel Material
model, which is described by the regression analysis in
Table 2. F10T-M20 high-strength bolts were idealized as The connection with high-strength bolts was modeled
having been connected by 165kN pretension. by ABAQUS (ver. 6.9.2). The T-stub that comprises the
t 2

connection is a C3D8R (eight-node brick element with
1 c
' 1 (1) reduced integration) member element. In performing
1 t f
the three-dimensional non-linear finite element analysis,
where
the effect of the contact and bearing between members
as well as the pretension of the high-strength bolts were
8B0b ' considered. The frictional coefficient between members
tc (2)
pFy was set to 0.5. This frictional coefficient was applied under
the assumption that the faying surfaces between members
b' are unpainted, blast-cleaned steel surfaces. The Korean
(3)
a' Building Code (KBC2009) only provides the case of = 0.5.
For high-strength bolts used in the T-stub analytical model,
d' the horizontal movement was restrained by inserting the
1 (4)
p plate, which is 1.5 times thicker than the T-stub flange
between the lower T-stub and the upper T-stub. As shown in
db Fig.3., the overall connection analysis model consists of a
a ' a (5)
2 total of 40,716 elements and 55,376 nodes, and the analysis
of each model lasted about 240 minutes. In modeling the
db connection with high-strength bolts, the applied ABAQUS
b ' b (6)
2 options are as shown in Table 3.
Table 1. Geometric Variables of the T-Stub Analytical Model (Unit: mm)
Analytical model bf tw tf gt r a b bT p e lT '
Q
G260-T15-B350 ' 350 b' 15 15 260 22 45 122.5 100 100 50 400 7.58
(7)
T Struik & Back (1 ') 400
G310-T15-B400 a' 15 15 310 22 45 147.5 100 100 50 400 8.14
G260-T19-B350 350 12 19 260 20 45 124 100 100 50 400 4.59
G310-T19-B400 400 12 19 310 20 45 149 100 100 50 400 4.96
G260-T21-B350
EA 350 13 21 260 22 45 123.5 100 100 50 400 3.67
kb 1.6 b
G310-T21-B400 400 13 21 310 (8)22 45 148.5 100 100 50 400 3.98
Lb
G110-T28-B200 200 18 28 110 22 45 46 100 100 50 400 0.58
G140-T28-B230
where 230 18 28 140 22 45 61 100 100 50 400 0.98
G110-T35-B200 200 20 35 110 22 45 45 100 100 50 400 0.08
G140-T35-B230 t230
n th 20 35 140 0.37
Lb tT ,top t fl ,col ' n 2twh (9)22 45 60 100 100 50 400
2
Table 2. Material Properties of the T-Stub Specimen
Fy (N/mm2) Fu (N/mm2) y u E (N/mm2) Eh (N/mm2)
347.33 518.78 0.001612 0.094512 215,446.40 1,814.04

310 JAABE vol.12 no.2 September 2013 Jae-Guen Yang


b'
b' d ' (3)
1aa'' (3)
(4)
p
d'
1 dd' (4)
a ' 1ap b (4)
(5)
p2

Also, d it is assumed that the prying action force is


a ' a dbb db (5)
a 'applied
b 'a b 2 to the end of the flange. However, as pointed
(5)
(6)
2 2
out by Kulak et al. (2001), Eq. (7) results in excessive
prying daction force.
b ' b db (6)
b ' b 2b (6)
2
Q ' b '
(7)
(7)
T
Struik & Back (1 ') a '
Q ' b '
QKulak
et al. ' b'
proposed (7)
the axial tensile force on the
T
Struik & Back (1 ') a ' (7)
high-strength ') aof
T Struik & Back (1bolts ' the connection under tensile
EA
force,
kb 1.6 as bin the following Eq. (8). The coefficient 1.6 in (8)
Lb
Eq. (8) includes the effect of prying action force.
EA
kb 1.6 EAbb
where (8)
(a) G260-T15-B350 analysis model kb 1.6 Lb (8)
(8)
Lb
t t
L
where t t fl ,col ' n 2twh n h (9)
where
b
where
T ,top
2
t t
Lb tT ,top t fl ,col ' n 2twh tnn thh (9)
(9)
Lb tT ,top t fl ,col ' n 2twh 2 (9)
2

(b) G140-T28-B230 analysis model


Fig. 3. Finite element analysis model of the connection with high-strength bolts
Fig.4. Prying Action
Fig. 4. Prying Force
action forceEstimation Model
estimation model by Struik
by Struik and Backand Back
Fig.3. Finite Element Analysis Model of the Connection with
High-Strength Bolts
Faella et al. assumed that, as in Fig.5., the contact
Table 3. ABAQUS Options for Finite Element Analysis force between the high-strength bolt head and the
Contact surface Command Option connection flange at an angle of 45 would occur due
Washer contact to the pretension of the high-strength bolt. Also, they
Constraints Tie
surface
Allow Adjust only proposed K p /K b, the ratio of the high-strength bolt
Finite stiffness to the connection flange stiffness under such
T-stub T-stub Contact separation to remove
sliding
after contact over closure contact force, as in Eq. (10), which offers similar
Allow Adjust only values to those of the equations proposed by Agerskov
T-stub High- Finite
Contact separation to remove
strength bolt sliding
after contact over closure
and Bursi (1990). In the application of Eq. (10), Kp can
Adjust only be obtained through the application of deformability
Nut - High-
strength bolt
Contact Small sliding to remove characteristics by decompression. The value, A(z),
over closure used in Eq. (11) signifies the area of the T-stub flange
3. Estimation of the Prying Action Force on the compressed by the axial force of high-strength bolts.
Kp tp
Connection with High-Strength Bolts and the Fig. 5. 4.10
KContact 3.25
force tp
distribution (10)bolt
characteristics due to the pretension of the high-strength
K

p
4.10 3.25 db (10)
Contact Force between Members b
Kb db
(10)
In the case of axial tensile force on the connection where
with high-strength bolts, as shown in Fig.4., Struik and where
where
Back proposed the ratio Q/T, as in Eq. (7), where Q is 1
K p t p twh 1 (11)
the prying action force, and T is the axial tensile force. K p t p twh 1 (11)
(11)
2 1 dz
In using Eq. (7), if ' is larger than 1.0, that is, the 2 0 EA( z) dz
connection is fractured by the plastic deformation after 0
EA ( z)
the flexural yielding of the flange due to the prying 4. Three-Dimensional
Q ' b 'm Nonlinear Finite Element
action force, ' is set to 1.0. On the other hand, if ' is Q RResult
Analysis ' bon ' the Connection with High- (12)
m T R(1 ') a 'mm ,
Yang Bolts (12)
between 0 and 1, that is, the connection is fractured by Strength
m T Yang (1 ') a 'm ,
partial yielding of the flange followed by bolt failure, If a connection with ' > 1.0, linked with the high-
B ' b 'm Q
' is set to the value directly obtained from Eq. (1). strength
B bolts,
1 R is ' under 1the
b ' axial
Q tensile force, the
(13)
T Yang
m 1 R (1 ') a 'mm
1 m T Yang , (13)
T Yang
m (1 ') a 'm m T Yang ,

where,
where,R
2.05 if ' 1.0
JAABE vol.12 no.2 September 2013 R 2.05 ifif 0 ' 1.0
0.42 ' 1.0 Jae-Guen Yang 311
R 0.42 if 0 ' 1.0
Fig. 4. Prying action force estimation model by Struik and Back

< ' < 0.98. Such concentrated stress shows the plastic
hinge line in a beam pattern, as predicted by Faella et
al. as well as Ballio and Mazzolani (1994), and results
in a fracture.
Shown in Fig.7. is the applied load-bolt force
relationship curves of the T-stub connection, resulting
from the three-dimensional non-linear finite element
analysis result, which shows very similar behavioral
characteristics to those of the same curve from Struik
and Back's test. In other words, with the increase
in the applied load, the axial bolt force of the high-
strength bolt also gradually increases, and after the
reference load, such an increase of the axial bolt force
accelerates, resulting in the failure of the high-strength
Fig.Fig.5. Contact
5. Contact Forcecharacteristics
force distribution Distribution Characteristics
due to the pretension of theDue to thebolt
high-strength bolt. Such a rapid increase of the axial bolt force of the
Pretension of the High-Strength Bolt axial bolt force on the high-strength bolt is believed to
be due to the prying action effect.
connection is fractured by plastic deformation after the
According to the three-dimensional non-linear finite
flexural yielding of the flange. Whereas a connection is
element analysis result, the average increase rate of
fractured by partial yielding of the flange followed by
the axial bolt force on the high-strength bolt due to the
bolt failure if the connection with 0 < ' < 1.0. Shown in
prying action effect is 228.2% when 3.67 < ' < 8.14,
Fig.6. is the stress distribution on the connection when
as shown in Table 4., resulting in 3.282 of the axial
the applied load reaches the reference load, which, as
bolt force ratio of the high-strength bolt to the applied
defined by Richard et al. (1988), is the value at which
load. This ratio is much larger than that referred to by
the slope of the region that shows the strain hardening
Eq. (8). That is, the prying action effect is excessive
symptom of the connection meeting the y-axis and the
since each connection with 3.67 < ' < 8.14 has much
load axis. As shown in Fig.6.(a), the flexural yielding
thinner flange thickness and larger bolt gauge distance
occurs by the concentrated stress in the area, which is
than the connection with ' = 1.0. On the other hand,
at a certain distance from the T-stub fillet and the area
the average increase rate of the axial bolt force on
at which the end of the interior of the high-strength
the high-strength bolt due to the prying action effect
bolt head and the T-stub flange meet when 3.67 < '
is 11.4% when 0.08 < ' < 0.98, as shown in Table 5.
< 8.14. On the other hand, as shown in Fig.6.(b), the Therefore, the following Eq. (12) was proposed by
flexural yielding only occurs by the concentrated stress
applying a revised parameter to Eq. (7), which offered
in the area at which the end of the interior of the high-
excessive Q/T, based on the three-dimensional finite
strength bolt head and the T-stub flange meet when 0.08

(a) T0=383.3kN, G260-T15-B350 analysis model (b) T0=1,958.1kN, G140-T28-B230 analysis model

(c) Tfinal=770.3kN, G260-T15-B350 analysis model (d) Tfinal=1,970.5kN, G140-T28-B230 analysis model
Fig. 6. Connection stress distribution under reference load
Fig.6. Connection Stress Distribution Under Reference Load

312 JAABE vol.12 no.2 September 2013 Jae-Guen Yang


T Yang
m
m
Yang
(1 ') a 'mm ,
Q ' b 'm
R (12)
m T (1 ')''a 'mbb''m,
1 R
BBYang QQ
T Yang 1 R (1 ') a 'm
m
11 T ,
(13)
(13)
m T (1 ') a 'm
m Yang m T Yang ,
m Yang

(a) T0=383.3kN, G260-T15-B350 analysis model


B T0=1,958.1kN,
(b) b 'm
' G140-T28-B230 Q analysis model

where, 1 R 1 (13)
where,
m T Yang (1 ') a 'm m T Yang ,

R 2.05 if
R 2.05 if ' 1.0 ' 1.0
element analysis result. In addition, if the equilibrium
where
R 0.42 if 0
R 0.42 if 0 ' 1.0 ' 1.0
where,
equation (B = T + Q) among T, the applied load, B, the R 2.05 if ' 1.0
axial force of high-strength bolts, and Q, the prying R 0.42 if 0 ' 1.0
action force, is applied, the correlation between the
(a) T0=383.3kN, kkG260-T15-B350 22 model EEbbAAbb
analysis
2
(b) T20=1,958.1kN,EEbbAAbbG140-T28-B230 analysis model
b,Yang
applied load and the axial force of high-strength bolts b,Yang
1 R ' b 'mm Lbb 1 Q LLbb
' b ' L Q
(14)
(14)
(14)
is summarized in Eq. (13). As shown in Eq. (13), the 1 R (12 ') a 'm Eb Ab 1 m2T Yang Eb Ab
kb,Yang (1 ') a 'm m T Yang
above-mentioned value of 3.282 is the sum of 2.282, ' b 'm Lb Q Lb (14)
1 R 1
the average increase in the ratio of the prying action where where (1 ') a 'm
where
m T Yang
force, as summarized in Table 4., and 1.0. As shown
(c) Tfinal=770.3kN, G260-T15-B350 analysis model (d) Ta''m =1,970.5kN,
aa0.3
0.3ddb G140-T28-B230 analysis model (15)
(15)
in the following Eq. (14), the tensile stiffness of high- where afinal m b (15)
Fig. 6. Connection stress distribution underb ' b reference
0.3 d load (16)
(16)
strength bolts is summarized from the correlation b 'mm b 0.3dbb (16)
a 'm a 0.3db (15)
b 'm b 0.3analysis
(c) Tfinal=770.3kN, G260-T15-B350 db model (16) analysis model
(d) Tfinal=1,970.5kN, G140-T28-B230
Kp Fig.
6. Connection stress distribution
t under reference load
K 0.642 t pp
3.893 0.642

K
p 3.893
(17)
(17)
m b, pretension Yang
Kb, pretension Yang ddbb
m
Kp tp
3.893 0.642 (17)
K d
m b, pretension Yang b

(a) G310-T21-B400 analysis model (b) G110-T28-B200 analysis model


(a) G310-T21-B400 analysis model (b) G110-T28-B200 analysis model
Fig. 7. Applied load vs. axial bolt force relationship
Fig. 7. curve on
Applied the
load vs. high-strength
axial bolt of the
bolt force relationship T-stub
on theconnection
Fig.7. Applied Load vs. Axial Bolt Force Relationship Curve on the High-Strength Bolt of curve
the T-Stub high-strength
Connectionbolt of the T-stub connection

between the constitutive equation on the high-strength


bolts and the axial force of high-strength bolts. That is,
the tensile stiffness of a couple of high-strength bolts
is affected by the prying action force, as shown in Eq.
(14), and it is reflected in the value, 1.6, the coefficient
mentioned in Eq. (8). In implementing Eq. (15) and
Eq. (16), a'm and b'm are defined as the distance from
the central line of the axial force of the high-strength
bolt to the end of the flange and the distance from the
central line of the axial force of the high-strength bolt
to the stem, respectively. As the load acting on T-stub Fig.8. Stress
Fig. 8.Distribution of the
Stress distribution Axial
of the axial Load
load onon
thethe High-Strength
high-strength bolt Bolt
increases, the stress that occurs in the shank of high- Table 4. Prying Action Effect Coefficient from the T-Stub
strength bolts moves from the centroid point of the
Fig. 8. Stress distribution of the axial Connection Analysis Result
load on the high-strength bolt when 3.67 < ' < 8.14
shrank of high-strength bolts to the right, as shown in Average increase ratio of the axial load
Fig.8., showing non-uniform distribution. In this case, Analytical model
on high strength bolt (%)
K
the
Kpp axial force ttp of high-strength bolts is assumed to G260-T15-B350 233.2
4.10
4.10 3.25 d 3.25 p (10)
(10)
K
be
Kbb acting on the
db
b centroid point of non-uniform stress G310-T15-B400 314.6
distribution. The centroid point of non-uniform stress G260-T19-B350 197.6
where
where
distribution obtained from the stress distribution that G310-T19-B400 247.5
occurred 11in the shank of high-strength bolts of each G260-T21-B350 160.6
KKpp tt tt (11) G310-T21-B400 215.4
analytical
p
p
wh
wh model
11 was the point that moved as much (11)
0 EA
as an22 average
EA ((zz))
dz
dzof 0.3d b from the central axis of high- Average increase ratio 228.2
strength0 bolts. Table 5. Prying Action Effect Coefficient from the T-Stub
Connection Analysis Result when 0.08 < ' < 0.98
Q
Q '' bb''m
RR(1 ') a ' m , (12)
(12)
(12) Analytical model
Average increase ratio of the axial load
m T Yang (1 ') a 'mm ,
m T Yang on high strength bolt (%)
G110-T28-B200 11.8
B
B '' bb''m 1 Q
Q G140-T28-B230 27.5
11 R
R (1 ') a ' m
1 T ,
(13)
(13)
(13) G110-T35-B200 1.4
T
m T Yang (1 ') a 'mm m T
Yang
m Yang m Yang ,
G140-T35-B230 4.9
where,
where, Average increase ratio 11.4

RR 2.05 if
2.05 if '' 1.0
1.0

RR 0.42 if 00
0.42 if '' 1.0
1.0

JAABE vol.12 no.2 September 2013 Jae-Guen Yang 313


22 E
EbbA
Abb 22 E
EbbA
Abb
kkb,Yang
b,Yang ' b' L
Q LLb (14)
1 R ' b 'mm Lb 1 Q b (14)
b
Kp tp
4.10 3.25 (10)
Kb db

where

1
Kp t p twh
(11)
1
2
0
EA( z)
dz

Q ' b 'm
R (12)
m T Yang (1 ') a 'm ,

B (a) G260-T15-B350
' b 'm analysis
Q model (b) G110-T28-B200 analysis model

9.Contact R
1 force 1 (13)
Fig.
m T Yang (1 distribution
') a ' characteristics
T , by the pretension of the high-strength bolt, resulting from the finite element analysis
Fig.9.mContactm
ForceYang
Distribution Characteristics by the Pretension of the High-Strength Bolt,
result
where, Resulting from the Finite Element Analysis Result
R 2.05 if ' 1.0
TheR ratio of 0the
0.42 if 'contact
1.0 force between members to to estimate the prying action effect, the normalizing
the stiffness of the connection flange and the high- factor * should be obtained by performing a finite
strength bolt is affected by such stress distribution. element analysis for each connection. The test
According to Fig.9., 2 theEcontact
b Ab
force
2 between Eb Ab the head performed the three-dimensional non-linear finite
kb,Yang
of the high-strength
' b 'm bolt
L b
and
the connection
Q Lb flange element analysis again by applying the identical
(14)
R
occurs1at 9. 1 geometric shape, material properties, boundary
(1 ') a 'm m T Yang
As described in Table 6. and Table 7., the contact conditions, and loading conditions to the T-stub finite
force is exerted at a stress distribution angle, which
where element analysis model by Hantouche, shown in
results from the finite element analysis result, and is Fig.10.(a) and 10.(b). The stress distribution of the
smallera 'm than
a 0.3that
db proposed by Faella et al. Based (15) on T-stub model, resulting from the re-performed finite
such ba'm stress
b 0.3ddistribution
b angle, the stiffness ratio (16) ofelement analysis, is shown in Fig.11. As shown in
the connection flange to the high-strength bolt can be Fig.12., the prying action effect of the high-strength
expressed as Eq. (17). bolt, resulting from the re-analysis of the T-stub by
Kp tp
Yang, offers a very similar prying action effect to that
3.893 0.642 (17)
(17) of the high-strength bolt by Hantouche. Although
K db
m b, pretension Yang Yang's analysis model showed a slightly larger Q/T
Table 6. Stress Distribution Angle Resulting from the T-Stub value than Hantouche's analysis model for the W24x76
Connection Analysis Result when 3.67 < ' < 8.14 analysis model, the Q/T from Yang's analysis model
Analytical model Angle () offered a value close to that of Hantouche's analysis
G260-T15-B350 10 model. Therefore, it is determined that using Yang's
G260-T15-B350 10 analysis model is feasible. In addition, it is much easier
G260-T19-B350 7 to estimate the prying action effect of a connection
G310-T19-B400 10
a) Hantouches T-stub since Yang's analysis
(b) Hantouches T-stub model does not need to use the
G260-T21-B350 8
specimen with continuity normalizing
specimen without
(a) G260-T15-B350
factor
continuity
analysis model
*. The normalizing factor *
(b) G110-T28-B200 analysis model
G310-T21-B400 9
plates can
Fig. 9. Contact only
force distributionbe obtained
characteristics
plates by performing
by the pretension a complex
of the high-strength bolt, resulting from thefinite
finite element analysis
result
Average angle 9 10. Hantouches T-stub
Fig. element specimenanalysis.
Table 7. Stress Distribution Angle Resulting from the T-Stub
Connection Analysis Result when 0.08 < ' < 0.98
Analytical model Angle ()
G110-T28-B200 1
G140-T28-B230 1
G110-T35-B200 1
G140-T35-B230 2
Average angle 1

The feasibility of the application of Eq. (12), the


proposed estimation of the prying action effect, was
assessed by comparing and reviewing the result from
that of the T-stub analysis, in Fig.10., which was
conducted by Hantouche (2011). In the development
of Hantouche's model, a normalizing factor * was a) Hantouches T-stub
specimen with continuity
(b) Hantouches T-stub
specimen without continuity
used to relate the primary prying strength model to the plates plates
Fig. 10. Hantouches T-stub specimen
primary prying finite element model. That is, in order Fig.10. Hantouche's T-Stub Specimen

314 JAABE vol.12 no.2 September 2013 Jae-Guen Yang


(a) With continuity plates (b) Without continuity plates
(a) With continuity
Fig. 11.plates (b) Without
Stress distribution from Yangs T-stub re-analysis continuity plates
result
Fig. 11.
Fig.11. Stress
Stress distribution
Distribution from
from Yangs
Yang's T-stubRe-Analysis
T-Stub re-analysis Result
result

(a) With continuity plates (b) Rigid Column


(a) With continuity plates (b) Rigid Column
Fig. 12. Comparison of Hantouches analysis result for the W30x108 analytical model with tcf/tTf=0.76
Fig. 12. Comparison of Hantouches analysis result for the W30x108 analytical model with tcf/tTf=0.76
Fig.12. Comparison of Hantouche's Analysis Result for the W30x108 Analytical Model with tcf/tTf=0.76

Table 8. Comparison of the Analysis Result by Hantouche and the Analysis Model (With Continuity Plates)

Analytical model tcf/tTf Differences (%)

W24x76 0.67 36.65 11.87 24.78


W30x108 0.6 30.28 31.32 1.04
W36x150 0.72 23.69 22.09 1.60

Table 9. Comparison of the Analysis Result by Hantouche and the Analysis Model (Without Continuity Plates)

Analytical model tcf/tTf Differences (%)

W24x76 0.67 36.65 19.85 16.80


W30x108 0.6 30.28 40.70 10.42
W36x150 0.72 23.69 39.63 15.94

5. Conclusion high-strength bolt to the applied load by the prying


This study aimed to offer an improved analysis action effect is 3.282 when 3.67 < ' < 8.14. Hence,
model that could estimate the prying action force of it is recommended that the coefficient referred to by
the T-stub connection with high-strength bolts under Eq. (8) is set to 0.609 for a connection with 3.67 <
tensile force, as well as the contact force between ' < 8.14.
members. The study resulted in the following 2) The three-dimensional non-linear finite element
conclusions: analysis result showed that the contact force is
1) Based on the three-dimensional non-linear finite delivered at a maximum angle of 9, the angle of
element analysis, the ratio of the axial force on the the stress distribution between the head of the high-

JAABE vol.12 no.2 September 2013 Jae-Guen Yang 315


strength bolt and the connection flange, which is 11) Richard et al. (1988) Derived moment-rotation curves for double-
framing angles. Comput & Struct, Elsevier Science Ltd., 3,
smaller than the angle proposed by Faella et al.
pp.485-494.
3) As shown in Table 8. and Table 9., in estimating 12) Struik, J. H. A., and de Back, J. (1969) Tests on Bolted T-stubs
Q/T of the T-stub, Yang's analysis model offered with Respect to Bolted Beam-to-Column Connections. Report
a closer value to the actual test than Hantouche's 6-69-13, Stevin Laboratory, Delft University of Technology, Delft,
analysis model did. Therefore, it is determined The Netherlands.
13) Swanson, J. A. (1999) Characterization of the Strength, Stiffness,
that estimating Q/T of the T-stub by using Yang's
and Ductility Behavior of T-stub Connections. Ph. D. Dissertation,
analysis model is valid. Georgia Institute of Technology, Atlanta, USA.
14) Swanson, J.A. (2002) Ultimate Strength Prying Models for Bolted
Acknowledgment T-Stub Connections. Engineering Journal, AISC, 39 (3), 3rd
The authors deeply appreciate the generous support Quarter, pp.136-147.
15) Thornton, W. A. (1985) Prying action: A general treatment.
of the National Research Foundation of Korea (under
Engineering Journal, AISC, 22, pp.67-75.
Research No. 2012047786).

List of Nomenclatures
db : Diameter of the bolt
d': Diameter of the bolt hole
kb: Axial tensile stiffness of the bolt
r: Radius of the fillet of the T-stub
B0 : Bolt pretension force used in the T-stub
B: Bolt force developed by the applied load
E: Young's modulus of steel
Eh: Secant modulus of steel
Fy: Yield strength of steel
Fu: Ultimate strength of steel
Q: Prying action force of the T-stub
T: Load applied to the T-stub
T0: Reference load of the T-stub
: Value that either maximizes the bolt's available tensile strength
for a given thickness or minimizes the thickness required for a
given bolt's available tensile strength
y: Yield strain of steel
u: Ultimate strain of steel

References
1) Agerskov, H. (1976) High strength bolted connections subject to
prying. Journal of Structural Engineering. Div., ASCE, 102(1),
pp.161-175.
2) Astaneh, A. (1985) Procedure For a Design and Analysis of
Hanger-Type Connections. Engineering Journal, AISC, 22(2),
pp.63-66.
3) Ballio, G., Mazzolani, F.M. (1994) Strutture in A cciaio. Hoepli,
Milan.
4) Bursi, O. (1990) Behaviour of High strength Bolts in Bolted
Beam-to-Column connections, Proc. Of Applied Stress Analysis
Conference, Nottingham.
5) Douty, R. T. and McGuire, W. (1965) High Strength Bolted
Moment Connections. J. Struct. Div., ASCE, 91(2), pp.101-128.
6) Faella, C., Piluso, V., Rizzano, G. (1996) Some Proposals to
Improve EC3-Annex J Approach for Predicting the Moment-
Rotation Curve of Extended End Plate Connections, Costruzioni
Metalliche, No. 4.
7) Faella, C., Piluso, V., Rizzano, G. (1998) Experimental Analysis
of Bolted Connections: Snug versus Preloaded Bolts. Journal of
Structural Engineering, ASCE, 124(7), pp.765-774.
8) Hantouche, EG. (2011) Behavioral characterization of built-up
T-stub connections for use in moment resisting frames. Ph.D.
dissertation, University of Cincinnati, Cincinnati, USA.
9) Jaspart, J. P. and Maquoi, R. (1995) Effect of Bolt Preloading
on Joint Behaviour. Steel Structures, Eurosteel '95, edited by
Kounadis, Balkema.
10) Kulak, G. L., Fisher, J. W. and Struik, J. H. A. (2001) Guide To
Design Criteria For Bolted and Riveted Joints 2nd Ed. New York:
American Institute of Steel Construction.

316 JAABE vol.12 no.2 September 2013 Jae-Guen Yang