24小时热门版块排行榜

返回列表

当前只显示满足指定条件的回帖，点击这里查看本话题的所有回帖

幸福小猪瑞

金虫 (正式写手)

应助: 8 (幼儿园)
金币: 2097.3
散金: 548
红花: 5
帖子: 813
在线: 153.3小时
虫号: 1322767
注册: 2011-06-14
性别: MM
专业: 自然地理学

[求助] 求钢结构方面的翻译，BB不是问题啊。。。。求快速哈

题目：Retrofitting tubular steel T-joints subjected to axial compression in chord and brace members using bonded FRP plates or through-wall steel bolts
a b s t r a c t：T-joints of Hollow Steel Sections (HSSs) are vulnerable to local instabilities of the web under the orthogonal compression in both members. Unlike W-sections stiffeners cannot be installed inside the closed sections. Therefore, alternative strengthening methods are needed. This experimental study explored
the effectiveness of two retrofitting methods by controlling the web buckling of the longitudinally compressed 203 76 3.09 mm chord, which is also subjected to transverse axial compression loading through the brace member. The web height-to-wall thickness (h/t) ratio of the chord is 65. In the first method, 8 mm diameter through-wall steel bolts were used to brace the webs of the chord at the vicinity
of the brace. In the second method, 76 185 9.5 mm glass fiber reinforced polymer (GFRP) plates were adhesively bonded to the two webs of the chord at the brace location. Two levels of sustained axial compression load were induced in the chord, representing 45% and 80% of its full axial capacity, in addition to control specimens without axial loads. The transverse brace load was then gradually increased to failure.The through-wall steel bolts increased the joint capacity by 13–25%, depending on the chord’s axial load level, while the bonded GFRP plate increased the capacity by 38–46%.
1. Introduction
Tubular steel structures are commonly used in the form of
trusses, vierendeel girders, and frames. The connection between
the chord and vertical member (brace) in vierendeel girders and
at mid-span of N-trusses, or between beam and column in frames,
take the form of a T-joint. The capacity of the structure may be
governed by the strength of its joints. Therefore, upgrading the
joint capacity may be essential in certain structures, particularly
if the members have been upgraded. This is crucial in thin-walled
members, where the bearing of the brace member on the chord
member produces web crippling of the thin walls of the chord. Unlike
steel W- and S-sections, it is not possible to install steel web
stiffeners inside tubular sections. In offshore structures with circular
tubes, the commonly used concepts include the ‘can’ whereby
the chord members are partially thickened at the joints, or the
‘doubler plate’, sometimes slightly modified to a system referred
to as the ‘collar’, where the brace is welded directly to the doubler
plate through a penetration weld, whereas the doubler plate is fillet-welded to the chord [1]. It should be noted that all these
strengthening techniques are primarily used in circular hollow
sections, commonly used in offshore structures.
For Rectangular Hollow Sections (RHSs), different strengthening
techniques may be employed. Failure of the chord side walls by
yielding or crippling is the most common for T-, Y- and X-joints,
especially when their b ratios (breadth of brace/breadth of chord)
are close or equal to unity. Welded rectangular HSS T-joints have
been studied experimentally by a number of researchers in the
past (e.g. [2,3]). When b is within 0.8 and 1.0, web buckling normally
governs. When b is less than 0.8, the failure modes depend
on the chord flange width-to-chord thickness ratio. A deformation
limit, in the form of the level of local indentation of flange chord
face, is adopted in the design of welded tubular joints as described
in the latest IIW static design procedures for welded tubular joints
[4–6]. Filling hollow sections with concrete to improve the web
crippling behavior was found to be efficient [7]. Another technique
was adopted [8,9], where a single bolt was used to brace the chord
sidewalls against the outward buckling, and resulted in an increase
of 18% in capacity. Another study [10] recommended that a wooden
brace be inserted into the RHS in addition to the through-wall
bolt to avoid the inward buckling of the chord sidewalls. The use of
externally bonded fiber reinforced polymers (FRPs) laminates has
also been successfully used to strengthen HSS columns against global buckling [11]. Limited work addressed local instabilities of
HSS members. For example, a study [12] demonstrated the benefits
of using carbon-FRP wraps to control web crippling of HSS sections
under end bearing. Another study [13] investigated axially loaded
carbon-FRP-wrapped short HSS columns to control local buckling.
Two previous studies by the authors on the subject [14,15]
investigated the strengthening of T-joints comprising RHS chord
and HSS brace members, where the brace was gradually loaded
axially to failure. The chord members did not include any axial
compression loads; hence simulating a beam-column joint in a
frame system. This paper also investigates the strengthening of
T-joints but with sustained axial compression loads in the chord
members, while the brace member is gradually loaded to failure.
This simulates joints in trusses or vierandeel girders, where the
combined longitudinal and transverse compression loads is more
critical to web stability at the T-joint. Both through-wall steel bolts
and adhesively bonded GFRP plates techniques are investigated.
The two techniques were selected for two reasons, namely, the
ease and simplicity of installation, and suitability to the geometric
nature of the section. Both techniques have minimal impact on the
esthetic appearance of RHS sections. Unlike W-sections, welded
steel stiffeners to the web at the joint are not convenient for RHS
sections. The study also addresses the effect of the level of sustained
axial compression in the chord on the strengthening
effectiveness.
2. Experimental program
The following sections describe test specimens and parameters,
material properties, fabrication of specimens, test setup and
instrumentation.
2.1. Test specimens and parameters
Table 1 provides a summary of the test matrix. T-joints were
fabricated and tested under combined brace and chord axial compression
loads. The T-joint consisted of a horizontal, 1220 mm long
chord member welded to a 400 mm long brace member (Fig. 1a).
Two key parameters were explored, namely, the type of retrofitting
reinforcement (through-wall steel bolts and adhesively bonded
GFRP plates (Fig. 1b), and the level of axial load in the chord member.
The chord member was a 203 76 3.09 mm RHS, with an
(h/t) ratio of 65. This section is classified as class 4 in accordance
with CAN/CSA-S16-01 [16] based on the web (h/t) ratio. The brace
member was a 76 76 8.9 mm HSS with a thick wall to avoid
failure of the brace. The study included three control specimens,
one with no axial load in the chord (T1), one with 200 kN (T2),
and one with 350 kN (T3). The axial compression loads in specimens
T2 and T3 represent 45% and 80%, respectively, of the pure axial strength of the chord, calculated according to CAN/CSA-S16-
01 [16]. The chord loads were kept constant during gradual
application of the load on the brace member to failure.
Specimens T4–T6 are counterparts of specimens T1–T3 and
were retrofitted with the through-wall steel bolts system (configuration
A in Fig. 1b), where the two bolts were located directly below
the brace member, 40 mm below the compression flange of
the chord, where the maximum local buckling was expected in
the chord. Specimens T7–T9 are also counterparts of T1–T3 and
were retrofitted by two 76 185 9.5 mm adhesively bonded
GFRP plates, one on each web, directly below the brace member
(configuration B in Fig. 1b). A previous study [15] has compared
GFRP and CFRP plates. It was shown that thick GFRP plates, which
are more economical than CFRP, are more effective in mitigating
local buckling in these T-joints.
2.2. Material properties
2.2.1. Cold-formed tubular sections
Two types of HSS sections were used, a rectangular one
(RHS), 203 76 3.09 mm, for the chord, and a square (SHS)
76 76 8.9 mm section for the brace. Both the RHS and SHS
were manufactured in accordance with ASTM A500 C [17]. Uniaxial
tension tests were performed according to ASTM E8/E8M-09 [18]
on dog-bone coupons cut from the flange and the web of the
RHS chord. A 50 mm extensometer was used to measure and
record strains. The stress–strain plots for the steel coupons are
shown in Fig. 2. The yield strength (offset secant at 0.2%) and
modulus of the chord were 426 MPa and 209 GPa. The reported
characteristic yield strength by manufacturer was 410 MPa.2.2.2. Through-wall bolts
The bolts used in the strengthened specimens are standard
8 mm diameter (Grade 8) high-strength bolts with a reported yield
and ultimate strengths of 896 and 1034 MPa, respectively.
2.2.3. GFRP plates
A 9.5 mm thick commercially available GFRP plate was used. It
consists of alternating layers of unidirectional E-glass roving and
random mats impregnated with polyester resin. The manufacturer
reported longitudinal and transverse tensile strengths of 138 and
69 MPa, and moduli of 12.4 and 6.9 GPa, respectively. The reported
longitudinal and transverse compressive strengths and modului by
the manufacturer are 165 and 110 MPa and 12.4 and 6.9 GPa,
respectively. To confirm these results, three coupons were cut from
the plates and tested in tension in the longitudinal direction
according to ASTM D3039/D3039M [19]. The stress–strain curves
are shown in Fig. 2. The average longitudinal tensile strength and
elastic modulus were 268 MPa and 20.6 GPa, respectively.Table 1
Test matrix.
Specimen
ID
Chord wall thickness
(mm)
Retrifitting system and configuration Axial compression Load of
Chord (kN)
Maximum transverse
Load (kN)
% Age gain in
strength
T1a 0 131.1 –
T2 N/A (control) 200 103.8 –
T3 350 58.3 –
T4a 3.09 0 164.3 25.3
T5 2 Bolts (Config. A in Fig. 1) 200 116.9 12.6
T6 350 66.0 13.2
T7b 76 185 9.3 mm GFRP Plate (Config. B
in Fig. 1)
0 180.9 38.0
T8 200 151.6 46.1
T9 350 80.6 38.3
a Aguilera et al. [14].
b Aguilera and Fam [15].
J. Aguilera, A. Fam / Engineering Structures 48 (2013) 602–610 603
2.2.4. Adhesive
The adhesive used for bonding the GFRP plates to the steel specimens
was Weld-On SS620. It is comprised of two components,
namely an adhesive (SS620-A) and an activator (SS620-B) mixed
at a 10:1 ratio. The reported tensile strength is 18–21 MPa and
the lap shear strength is 19–22 MPa.
2.3. Fabrication of T-joint specimens
The chord and brace members were cut to lengths of 1220 and
400 mm, respectively. The SHS brace member was directly welded
to the flange of the RHS chord member at mid-length. Cutting and
welding of specimens were performed by a professional, at a machine
shop. The holes necessary for the bolts to pass through the
chord in retrofitting configuration A were then hand drilled in both
webs. The 8 mm Grade 8 high-strength headed bolts were then inserted
into the holes and anchored from one side using a special
washer and nut for a snug fit. Special care was taken in order not
to over tighten the nut and cause inward buckling of the two webs.This was accomplished by manual tightening of the nut until slack
with removed between the washer and the web of the section, but
no pretension was applied to the bolt. For retrofitting configuration
B, specimens were first sandblasted to prepare the surface and
were cleaned with acetone to remove any dirt or debris. The GFRP
plates were then sanded with fine grit sandpaper to remove the
smooth polymeric coating and were cleaned with isopropyl alcohol.
The adhesive was then applied to the surfaces and the FRP
plates were positioned and pressed to maintain a consistent thickness
of adhesive. After curing, and before testing, vertical stiffener
steel plates, 110 191 12 mm, were inserted inside the chord
member at both ends (Fig. 3) to prevent premature failure due to
crippling at supports.
2.4. Test setup and Instrumentation
The axial compression load in the chord was first applied
through a 25 mm diameter high strength threaded rod positioned
concentrically inside the RHS chord member. The rod was anchored
against the RHS section on one end using a large load cell
with a central hole and was tensioned from the other end against
an end steel plate using a hydraulic ram (Fig. 3a). At both ends,
the rod was anchored using a special nut and washer system of
spherical surfaces (Fig. 3a), which was also lubricated. This was
to prevent any bending in the threaded rod, which could contribute
to the flexural capacity of the RHS chord member during transverse
loading on the brace member.
After loading the chord to the desired axial force, the brace was
concentrically loaded to failure using a 1000 kN Riehle testing machine
(Fig. 3b), at a 1 mm/min rate. The brace was clamped at both
ends using a special assembly of heavy SHS sections and threaded
rods. First, the specimen was rested on two 150 150 12 mm
HSS supports. The two supports were set apart to provide a clear
span (L) of 1000 mm, which is almost five times the chord depth
(h) of 203 mm. Another two SHS sections were set on top of the
specimens ends. The two upper SHSs were held down and anchored
to the Riehle testing machine using two vertical 25.4 mm
diameter threaded rods at each support. The threaded rods were evenly hand tightened using wrenches. This span-to-depth (L/h)
ratio of the chord was carefully selected as per recommendations
in the literature. The UK Department of Energy Offshore Technology
Report [20] and other researchers [21] suggested that (L/h)
ratio should not be less than four, in order to avoid any effect of
supports on the joint strength. On the other hand, others [22] indicated that the (L/h) ratio should not be excessive, otherwise chord
failure may occur prior to joint failure as the plastic moment of the
chord cross-section at the crown location is reached. Some
researchers [23] have indicated a limit of 5.75 for (L/h) to ensure
joint failure. In this study, the (L/h) ratio of 5 was used. Some
end fixities were provided by clamping, to increase the load at which yield and plastic moments occur, in order to focus on the
stability aspect of joint strength.
Two 100 mm linear potentiometers (LPs) were mounted at the
upper and lower flanges of the chord member at mid-span, to measure
the vertical deflection of the top and bottom flanges independently
(Fig. 3b). These two deflections may differ as the chord
sidewalls buckle. Four additional LPs were mounted in a horizontal fashion to measure the slip of the ends of the chord specimens. The
strains on the chord, in two directions, and on some through-wall
bolts, were measured using 5 mm electric resistance strain gauges.
Fig. 4 shows the locations of strain gauges on the chord.
3. Results of the experimental program
Table 1 provides a summary of the experimental results, in
terms of the maximum load achieved and the percentage of gain
in strength of retrofitted specimens relative to their control counterparts.
Fig. 5 provides the load–deflection responses of all specimens.
Fig. 6 demonstrates the stability of axial force in chord
members during the transverse loading of the specimen. Fig. 7
summarizes the study by showing the interaction diagram of axial
and lateral loads in the chord at peak values for control and retrofitted
specimens. Fig. 8 shows the load–strain responses. Fig. 9 depicts
the end slip of the top and bottom flanges of the chord
member at the supports. Figs. 10–12 show the various failure
modes of the control and retrofitted specimens. The following sections
describe in details the test results and effect of various
parameters on performance.
3.1. Load–deflection behavior
Fig. 5 shows the load–deflection responses of all test specimens
based on both the top and bottom flanges in each specimen (except
in specimen T3 in which only top flange deflection was recorded).
Generally, the load ascends almost linearly initially, followed by a
non-linear behavior until a peak load is reached and then a
descending response can be observed with various degrees of nonlinearity.
This general trend is similar in both control and retrofitted
specimens, regardless of chord axial load level or method of
retrofitting, except that the peak load was increased in retrofitted
specimens. The peak load consistently corresponds to instability
failure of the webs of the chord, or buckling of the top flange. It
should be noted that while the retrofitting system enhances
strength, it has virtually no effect on the initial stiffness or ductility
of the member. For each specimen, the deflection of the top and
bottom flanges of the chord differed slightly at any given load, as
a result of the out-of-plane displacements of the vertical webs. This
effect becomes more pronounced near the peak load and even
more in the descending part of the response. Although no specific
link exists in this study between the GFRP plate and the through
steel bolts designs, it can be noticed that the GFRP plate (specimen
T7) provided an enhanced post-peak deformation capacity relative
to the steel bolts (specimen T4) at zero axial loads in the chord.
Fig. 6 shows the variations of the axial force in the chord with
deflection. Also shown are the transverse load–deflection curves.
Specimens T4 and T5 with through-wall bolts had both sidewalls
forced to displace laterally in the same direction, by having
one of the sidewalls buckling outward, and the other side buckled
inward (Fig. 11a and b). Specimen T6 had the web fail by inward
buckling just below the brace and outward buckling in the adjacent
region (Fig. 11c).
In specimen T7 with bonded GFRP plate, the outward buckling
of the web was shifted just beyond the GFRP plate on one side
and occurred in both webs (Fig. 12a). Also, the top flange buckled
inwards. As axial load was introduced in specimens T8 and T9,
web buckling shifted inwards, while top flange buckling shifted
outwards (Fig. 12b and c). It is well established that geometric
compatibility dictates that the local buckling of the web and flange
occur in opposite directions.
It should be noted that once the peak loads were reached in all
axially loaded specimens, and due to local buckling occurrences, an
increased overall axial shortening occurred in the chords. To avoid
excessive loss of axial force due to this shortening, a faster rate of
jacking was needed for the hydraulic ram to maintain the axial
force. As shown in Fig. 6, this approach led to a steady level of axial
compression in all axially loaded specimens. This was particularly
important at the range where peak loads occurred.
4. Conclusions
Based on this experimental study on T-joints in which the chord
member was a Rectangular Hollow Section (RHS) with (h/t) ratio of
65 and subjected to various levels of axial compression loads, and
the brace member was an axially loaded Square Hollow Section
(SHS), the following conclusions are drawn:
1. Both methods investigated, namely through-wall steel bolts
and bonded GFRP plates, are practical in retrofitting RHS chord
members in T-joints under brace and chord axial compression
loads. Generally, the FRP plating technique was more effective
than the through-wall bolts technique, especially at higher axial
loads in chord.
2. Using through-wall steel bolts retrofitting system, the joint
strength was increased by 25%, 13% and 13% when the axial
compression loads in the chord were zero, 45% and 80% of its
pure axial strength, respectively.
3. Using bonded GFRP plates as a retrofitting system, the joint
strength was increased by 38%, 46% and 38% when the axial
compression loads in the chord were zero, 45% and 80% of its
pure axial strength, respectively.
4. In general, the transverse load capacity of the chord reduces as
the axial compression load increases, for both control and retrofitted
specimens.
5. The peak loads reached in all control and retrofitted specimens
were associated with stability failure in the form of local buckling
in the webs and compression flange of the chord. In retrofitted
specimens, the local buckling was shifted laterally, away
from the vicinity of the brace, leading to the higher strength.
6. Generally, local buckling occurred before or just at yielding of
the chord member at the joint. Therefore, for the (h/t) ratio used
in this study, it can be concluded that the retrofitting technique
enhanced strength significantly but did not allow the chord to
exceed yielding.
The authors are currently extending this research through finite
element modeling to capture the observed behavior accurately and
then carryout a parametric study to consider the effects of the following
parameters on the strengthening effectiveness: (a) heightto-
width ratio of the chord, (b) height of the chord-to-breadth of
the brace ratio, (c) span-to-depth ratio of the chord and (d) boundary
conditions, including the two extreme cases of fully hinged and
fully fixed. The results of the parametric study should facilitate the
development of design guides guidelines.
References
[1] Choo YS, Li BH, Liew JYR, van der Vegte GJ, Zettlemoyer N. Static strength of Tjoints
reinforced with doubler or collar plates. In: Eighth international
symposium on tubular structures. Singapore; 1998.
[2] Kato B, Nishiyama I. The static strength of R.R. joints with large beta ratio.
CIDECT program 5Y report. Japan: Department of Architecture, Faculty of
Engineering, University of Tokyo; 1979.
[3] Zhao XL, Hancock GJ. T-joints in rectangular hollow sections subject to
combined actions. J Struct Eng ASCE 1991;117(8):2258–77.
[4] Zhao XL, Wardenier J, Packer JA, van der Vegte GJ. Current static design
guidelines for hollow section joints. Struct Build 2010;163(6):363–73
[Institution of Civil Engineers, UK].
[5] Lu LH, de Winkel GD, Yu Y, Wardenier J. Deformation limit for the ultimate
strength of hollow section joints. In: Proceedings of the 6th international
symposium on tubular structures, Melbourne. Balkema, Rotterdam; 1994. p.
341–8.
[6] Zhao XL. Deformation limit and ultimate strength of welded T-joints in coldformed
RHS sections. J Construct Steel Res 2000;53(2):149–65.
[7] Packer JA. Concrete-filled HSS connections. J Struct Eng ASCE
1995;121(3):458–67.
[8] Bains S. Investigating into the effects of through bolting on rectangular hollow
steel beams in flexure. PhD Thesis, UK: University of Sussex; 1983.
[9] Bradfield CD, Morrell, PJB, Ibrahim A. Improvements in the flexural capacity of
rectangular hollow sections by through bolting. Tubular Struct 1994:109–14
[Rotterdam, Balkema].
[10] Zhao XL. Partially stiffened RHS sections under transverse bearing force. Thin-
Walled Struct 1999;35:193–204.
[11] Shaat A, Fam A. Slender steel columns strengthened using high modulus
CFRP plates for buckling control. ASCE J Compos Construct 2009;13(1):
2–12.
[12] Zhao XL, Fernando Dilum, Al-Mahaidi Riadh. CFRP strengthened RHS subjected
to transverse end bearing force. Eng Struct 2006;28:1555–65.
[13] Shaat A, Fam A. Axial loading tests on short and long hollow structural steel
columns retrofitted using carbon fibre reinforced polymers. Can J Civil Eng
Spec Issue Rec Adv Steel Struct Res 2006;33(4):458–70.
[14] Aguilera J, Shaat A, Fam A. Strengthening T-joints of rectangular hollow steel
sections against web buckling under brace axial compression using throughwall
bolts. Thin-Walled Struct 2012;56:71–8.
[15] Aguilera J, Fam A. Bonded FRP plates for strengthening rectangular hollow
steel section T-joints against web buckling induced by transverse
compression. ASCE J Compos Construct; in press [May 17].
[16] Canadian Standards Association, CAN/CSA-S16-01. Limit states design of steel
structures, Mississauga, Ontario.
[17] ASTM A500. Standard specification for cold-formed welded and seamless
carbon steel structural tubing in rounds and shapes.
[18] ASTM E8/E8M–09. Standard test methods for tension testing of metallic
materials.
[19] ASTM D3039/D3039M. Standard test method for tensile properties of polymer
matrix composite materials.
[20] UK Department of Energy Offshore Technology Report, Background to new
static strength guidance for tubular joints in steel offshore structures. OTH 89
308, DoN, UK; 1990.
[21] Moffat DG, Kruzelecki J, Blachut J. The effects of chord length and boundary
conditions on the static strength of tubular T-joints under brace compression
loading. Marine Structures 1996;9:935–47.
[22] Lalani M. Developments in tubular joints technology for offshore structures.
In: Proc. 2nd int. offshore and polar engineering conference, San Francisco;
14–19 June 1992.
[23] Madros MSZ, Zettlemoyer N, Healy BE. Effect of chord can length on strength of
T joints. In: Proceeding of the 27th annual offshore technology conference
(OTC); 1995 Huston, TX, USA [May 1–4

回复此楼