-
Construction and Building Materials 50 (2014) 530–539
Contents lists available at ScienceDirect
Construction and Building Materials
journal homepage: www.elsevier .com/locate /conbui ldmat
Feasibility study of grouted splice connector under tensile
load
0950-0618/$ - see front matter � 2013 Elsevier Ltd. All rights
reserved.http://dx.doi.org/10.1016/j.conbuildmat.2013.10.010
⇑ Corresponding author. Tel.: +60 128072616.E-mail address:
[email protected] (J.H. Ling).
Ling Jen Hua a,⇑, Ahmad Baharuddin Abd. Rahman b, Izni Syahrizal
Ibrahim ba School of Engineering and Technology, University College
of Technology Sarawak, 96000 Sibu, Sarawak, Malaysiab Faculty of
Civil Engineering, Universiti Teknologi Malaysia, 81310 Skudai,
Johor Darul Ta’zim, Malaysia
h i g h l i g h t s
� Use of standard size steel section to confine splicing of
bars.� Feasibility assessment of grout splice connection under
tensile load.� Derive equations to evaluate the response of splice
connection.
a r t i c l e i n f o
Article history:Received 1 March 2010Received in revised form 1
October 2013Accepted 4 October 2013Available online 26 October
2013
Keywords:Grouted spliceSleeveConnectorBondPrecast connection
a b s t r a c t
The conventional bar lapping approach to connect steel bars
requires long development length andalways leads to bar congestion
problems. For this reason, grouted splice connectors are used to
confinethe grout surrounding the bars to improve the bond between
the grout and the bars. Four series of spec-imens with a total of
35 specimens were tested under incremental tensile load. These
specimens vary interms of configurations and were assessed for
feasibility in the aspects of bond strength, ductilityresponse and
failure modes. Equations are derived to evaluate the structural
performance of the speci-mens. The typical modes of failure are bar
tensile failure, grout-bar bond failure, grout-sleeve bond
fail-ure, and sleeve tensile failure. These failures reveal the
factors to be considered during the design of asplice connector.
Under confinement, the required anchorage length of the bars can be
shortened tonearly nine times the diameter of the spliced bar.
� 2013 Elsevier Ltd. All rights reserved.
1. Introduction
Ever since the first use of iron reinforced concrete structure
byThaddeus Hyatt in 1853, many researchers started to study thebond
between steel and concrete, experimentally and analytically.The
scopes of study include the mechanisms of bond [1], the
inter-locking effects of the geometry and patterns of the ribs on
steelbars [2,3], the relative rib area to generate bond strength
[4,5],the distributions of bond stress along a bar [6,7], the crack
propa-gation surrounding the steel bars [7–11], etc.
The research developed further to connect steel bars in
con-crete, by lapping adjacent bars [12–14]. This method requires
along development length of bar for stress to entirely transfer
froma bar to another in concrete. It often leads to detailing and
bar con-gestion problems, especially when large diameter steel bars
areused in heavily reinforced structures. To solve this problem,
lateralforces and confinement were induced to increase the
bondstrength and to reduce the development length [15–18].
Initially, transverse reinforcements were used to control
thedevelopment of splitting cracks surrounding the anchorage
region
[19,20]. This approach could only give passive confinement to
alarger region of concrete, but is unable to directly confine a
smallgrouted region along a bar. Nevertheless, it provides
essential fun-damental for a splice connector, where the bond
strength can beincreased by controlling the circumference splitting
cracks aroundthe bar.
At present, the confinement can be produced in a small
regionalong a bar. It is by surrounding the splice with spirals
[21,22],cylindrical pipes [23–25] and fiber reinforced polymer
[26]. Theseapproaches need grout as the bonding and load
transferring mate-rials, for its high strength and fine particles.
It is also utterly essen-tial to ensure the full capacity of bar
stress is properly distributedfrom a bar to another without being
compromised by bond capac-ity. In the ideal condition, the bond
strength should outperform itsspliced bars.
The splice connectors available in the marketplace [27–34]
areproprietary products owned by the inventors and several
estab-lished companies. The designs of the shapes are rather
complexand they generally need advance steel molding techniques for
fab-rications. Furthermore, there is limited information on the
loadresisting mechanisms, the distributions of internal stress and
thedesign calculations of a splice connector published
academically,except for some feasibility test reports on these
established
http://crossmark.crossref.org/dialog/?doi=10.1016/j.conbuildmat.2013.10.010&domain=pdfhttp://dx.doi.org/10.1016/j.conbuildmat.2013.10.010mailto:[email protected]://dx.doi.org/10.1016/j.conbuildmat.2013.10.010http://www.sciencedirect.com/science/journal/09500618http://www.elsevier.com/locate/conbuildmat
-
J.H. Ling et al. / Construction and Building Materials 50 (2014)
530–539 531
products [35,36] on the basis of several prescribed acceptance
cri-teria proposed by relevant standards [37,38].
Based on the reviews from previous research, several
principlesare extracted:
(a) The alignments of bar in a splice connection, in-line or
adja-cent, leads to different responses. Adjacent alignment of
barslead to eccentricity. Under tensile load, the load tends
toself-align and causes undesired bending deformation of thespliced
bars while rotating the couple [39,40]. This affectsthe performance
of the spliced bar, generates excessivedeformation, and causes
regional failure of the surroundingconcrete of the structural
elements [39].
(b) The performance of grouted splice is heavily influenced
bythe quality of the grout in the sleeve. The strength of thegrout
and the completeness or segregation of grout in thesleeve would
affect the capacity of the splice [25,41].
(c) The bar slippage resistance can be generated by the
bondbetween the grout and the bar, or by the frictional
grippingbetween the coupler and the bar, or by using a threaded
sys-tem to connect steel bars and the couple [23,34,36,39].Under
tensile load, undesired sudden slippage of bar usuallyoccurs when
the frictional gripping approach is used [35].
(d) The performance of the bond in a sleeve can be increased
byreducing in the diameter of the sleeve and increasing
theanchorage length of the splice bars [25].
Based on these fundamentals, grouted splice was studied andthe
bars were spliced in-line. This study uses standard-sizedsteel
sections (pipes, square hollow sections and aluminumsleeves) to
bridge the discontinuity and to confine the groutedregion around of
steel bars, as initiated by Einea et al. in 1995[23]. The objective
was to determine whether these sectionsare suitable as the
connectors. For this reason, incremental ten-sile load tests were
carried out on a total of 35 specimens withdifferent
configurations. The results were analyzed and the feasi-bility of
the specimens was determined based on several evalu-ation
criteria.
2. Experimental program
A test program was carried out on four series of 35 grouted
splice connectorswith various sizes and configurations to
understand the effects of the interactingvariable; material
properties, sleeve configurations, confinement mechanism, etc.
2.1. Specimens
Fig. 1 illustrates the details of the grouted splice connectors.
These connectionswere used to splice 16 mm diameter high strength
bars of 460 N/mm2 specifiedyield strength. The configurations of
the connectors are briefly described as follows:
� AS-Series – bars were spliced, in-line or adjacent, with the
surrounding groutconfined by a corrugated aluminum sleeve with 43
mm diameter and 1 mmthick.� BS-Series – 65 mm diameter and 4.5 mm
thick mild steel pipes with the speci-
fied yield strength of 250 N/mm2 were used to splice in-line
bars. 10 mm Diam-eter high strength steel bars were welded to
sleeves BS-01 to BS-06 from bothends to interlock with the grout.
The amounts and the provided lengths of the10 mm diameter bars
varied among these specimens. For specimens BS-08 toBS-11, 30 mm
holes were provided at 50 mm and 100 mm from the ends ofthe pipes.
The holes provide space to be occupied by the grout to engage a
largeshear area for interlocking with the pipes to prevent slippage
of the grout. Spec-imens BS-12 and BS-13 used rings of welded ribs
of 2 mm height. These weldedribs were located at 25 mm and 50 mm
from the ends of pipes to interlock withgrout. Also, taper nuts of
37� inclined angle were welded on the bars to givemore bearing area
for interlocking with grout.� CS-Series – specimens CS-01 to CS-09
comprised sets of two semi-cylindrical
mild steel pipes. Two steel plates, 5 mm thick and with 22 mm
opening, werewelded to the pipes to lock the movement of nuts on
the spliced bars. Thethreaded length of the bars was 70 mm so that
the nuts can be flexibly adjustedalong it to fit in the compartment
between the two steel plates.
� DS-Series – specimens DS-01 to DS-11 were modified from mild
steel squarehollow section (SHS). Plates of 3 mm thick were
inserted through the four cor-ners at 20 mm, 50 mm and 100 mm from
the ends of SHS. These plates interlockwith grout to resist
slippage of grout. For specimens DS-08 to DS-11, severalbolts were
used to laterally compress on the spliced bar. This generates an
addi-tional resistance to prevent slippage of the spliced bars.
2.2. Test plan and setup
The specimens were made by inserting bars from both ends of the
sleeve beforehigh strength Sika Grout-215 (proportion of 25 kg
grout: 4 l water) was poured in,to fill the void in the sleeve. The
specimens were ready for testing after the grouthad achieved the
intended strength of 40 N/mm2. The incremental tensile loadwas
applied at a rate of 0.5 kN/s by a hydraulic actuator with a
capacity of250 kN (Fig. 2). The relationships of the applied loads
versus the longitudinal dis-placement of the bars were plotted and
recorded for analysis.
3. Test results
The displacements of the bars corresponding to the applied
loadwere measured throughout the test and the largest axial
forcemeasured was considered as the ultimate capacity of the
speci-mens, as shown in Table 1.
Based on the results, due to limited specimens available, it
isdifficult to distinguish the effects of different configurations
ofBS-Series. All the specimens failed in the same mode – bar
tensilefailure. For this reason, the enhancements like (a) the
tapered nutat the end of the bars, (b) the interlocking bars welded
throughoutthe length of BS-01, (c) the extra shear area provided by
the groutoccupying the holes on sleeve BS-08 to BS-11, and (d) the
inter-locking ribs welded to the inner surface of sleeve BS-12 and
BS-13, might not be necessary.
Fundamentally, as long as (a) the sleeve is strong enough
towithstand the tensile stress induced by the load, (b) the bond
be-tween the bar and the grout is sufficient to prevent slippage
ofbar, and (c) the bond between the grout and the sleeve is
sufficientto resist the grout from slipping out of the sleeve, the
splice con-nection would adequate. Fabrication wise, it is more
practical to in-duce a minimum efforts of modification to the pipe
section. Thus,the interlocking bars welded to the inner surface of
the sleeveshould be of a minimum length, but just adequate to
prevent thegrout from slipping out of the sleeve. No tapered nut is
requiredfor this series of specimens.
As observed from DS-Series, tapered nut indeed improves thebond
between the bar and the grout, especially when the develop-ment
length is inadequate. The specimens with the tapered nuts atthe
ends of the bars (DS-01 and DS-03) always outperform thespecimens
without the tapered nuts (DS-02 and DS-04).
The ribs at the corner of the SHS of DS-Series seems to
providebond-slip resistance for the grout only. As observed from
DS-01,DS-03, DS-05, DS-06 and DS-07, the capacities of the
connectionwere about the same regardless the amount and the
positions ofthe ribs provided. Similarly, when the ribs generate
sufficientbond-slip resistance to the grout, an addition of shear
areas pro-vided by the grout that occupied the holes of the sleeve
wouldnot be necessary.
The test results show that BS-Series is more efficient in
provid-ing confinement to the splicing of bars in the sleeve as
comparedwith DS-Series. The bar embedded lengths of BS-03, DS-02
andDS-04 were same, and tapered nut was not provided at the endsof
the bars. BS-03 apparently outperformed DS-02 and DS-04 witha
higher ultimate capacity without symptoms of bond failure.
Thesquare section (DS-Series) appeared to be less superior to the
cir-cular section (BS-Series), especially when it is used to
generateconfinement to the splicing of bars. The circular section
is moreefficient in generating tangential tensile resistance to
confine thegrout surrounding the spliced bar.
-
Fig. 1. Details of specimens (dimensions in mm).
532 J.H. Ling et al. / Construction and Building Materials 50
(2014) 530–539
3.1. Failure modes
Fig. 3 shows the typical failure modes of the specimens;
bartensile failure, grout-bar bond failure, sleeve tensile failure
andgrout-sleeve bond failure, each of which offered different
rangesof ultimate capacity. Specimens with bar tensile failure
failed at
high tensile capacities, ranging from 118 kN to 143 kN. The
corre-sponding displacements ranged from 22.4 mm to 43.7 mm.
Thephenomenon of bar tensile failure signifies an excellent
bondstrength developed in the sleeve that outperformed the
tensilestrength of the splice bars. The peripheral tensile strength
of thesleeves provided resistance to control radial splitting of
the grout,
-
Hydraulic actuator
Pressured grips
Grouted splice specimen
Pulling force
Reaction force
Fig. 2. Setup of tensile load test.
Table 1Test results of specimens under incremental tensile
load.
Specimens Grout strength, fcu (N/mm2) Ultimate load, Pu (kN)
Displacement at failure, d (mm) Failure mode
16 mm Bar 133.2 55.1 Bar tensile failureAS-01 58.10 11.9 2.7
Sleeve tensile failureAS-02 58.10 135.6 35.5 Bar tensile
failureBS-01 62.97 125.6 36.0 Bar tensile failureBS-02 62.97 119.2
22.4 Bar tensile failureBS-03 62.97 141.2 32.0 Bar tensile
failureBS-04 62.97 118.0 23.5 Bar tensile failureBS-05 62.97 143.6
42.2 Bar tensile failureBS-06 62.97 124.9 36.9 Bar tensile
failureBS-07 62.97 124.4 35.8 Bar tensile failureBS-08 62.97 121.2
36.3 Bar tensile failureBS-09 62.97 124.4 37.8 Bar tensile
failureBS-10 62.97 136.3 31.2 Bar tensile failureBS-11 62.97 120.0
37.3 Bar tensile failureBS-12 62.97 125.1 38.1 Bar tensile
failureBS-13 62.97 135.4 29.5 Bar tensile failureCS-01 43.32 87.0
8.72 Bar-grout bond failureCS-02 43.32 98.4 26.3 Bar-grout bond
failureCS-03 43.32 80.0 4.54 Bar-grout bond failureCS-04 43.32 85.3
17.1 Bar-grout bond failureCS-05 43.32 84.3 6.8 Bar-grout bond
failureCS-06 43.32 40.9 1.0 Grout-sleeve bond failureCS-07 43.32
72.5 7.0 Bar-grout bond failureCS-08 43.32 70.5 6.7 Bar-grout bond
failureCS-09 43.32 101.8 5.6 Bar-grout bond failureDS-01 49.28
129.1 41.6 Bar tensile failureDS-02 49.28 112.0 16.6 Bar-grout bond
failureDS-03 49.28 123.4 41.9 Bar tensile failureDS-04 49.28 110.6
4.8 Bar-grout bond failureDS-05 49.28 130.1 43.7 Bar tensile
failureDS-06 49.28 123.7 29.4 Bar-grout bond failureDS-07 49.28
127.5 33.3 Bar-grout bond failureDS-08 49.28 122.3 31.6 Bar-grout
bond failureDS-09 49.28 120.7 37.1 Bar-grout bond failureDS-10
49.28 118.2 36.2 Bar tensile failureDS-11 49.28 122.4 38.8 Bar
tensile failure
J.H. Ling et al. / Construction and Building Materials 50 (2014)
530–539 533
and subsequently, improve the efficiency of the mechanical
inter-lock between grout keys and bar ribs [23]. It also enhanced
the fric-tion resistance between the grout and the sleeve [42].
Moreover,the 10 mm bars welded on specimens BS-01 to BS-07
generate
an addition confinement stress that derived from the
resultantsacting perpendicularly to the inclined surface of the
ribs [25].
The bar-grout bond failed at a moderate high strength,
rangingfrom 70.6 kN to 135.6 kN. The mode of failure can be ductile
or
-
(b) Bar-grout bond failure
(c) Grout-sleeve bond failure
(d) Sleeve tensile failure
(a) Bar tensile failure
Bar necking and tensile fracture
Damage of grout keys as the bond failed
Tensile fracture of grout as it slipped out of sleeve
Tensile fracture of sleeve and grout
Fig. 3. Typical modes of failure.
534 J.H. Ling et al. / Construction and Building Materials 50
(2014) 530–539
brittle. Specimens DS-02, DS-04, DS-06, DS-07 DS-08 and
DS-09,gave ductile responses because the bond failed shortly after
sur-passing the yield strength of the spliced bars. On the other
hand,CS-Series failed in a brittle mode, as the bond failed before
thespliced bars yielded.
CS-Series failed earlier than DS-Series due to inappropriate
barsurface condition. The threaded surface of the bars changed
thegeometry of grout keys that occupied the spacing between
thethreads to be sharp and narrow. This led to a larger
bearing/sheararea of grout keys, and hence, led to a smaller
regional shear-slipresistance to withstand the high stress
concentrated at the endsof the sleeve near to the source of the
pulling force [43]. This in-creased the tendency of the load to
trigger a progressive failureof the grout keys starting from either
end of the sleeve. Eventually,the bar-grout bond failed completely
as the spliced bar slipped outof the sleeve. This phenomenon is
proven when CS-09, where thesurface of the bars were not threaded,
offered a higher bondstrength of 17% than CS-01.
The comparison of bond strength among specimens DS-01 toDS-04
revealed the effects of the taper-headed bars. The
interaction between taper-headed bars and SHS ribs could
improvethe bond between the grout and the bars. The tapered nut
providedbearing area for interlocking with the grout. It also
converted thelongitudinal pulling force into stress acting
perpendicularly to itsinclined surface. The stress triggered the
counteraction of thesleeve, specifically from the welded ribs on
the SHS at 100 mmfrom the ends of the sleeve. This generated a
regional compressivefield in the sleeve along the spliced bars to
control the peripheralsplitting cracks around the spliced bars, and
hence, decreased thedeterioration rate of the bond between the bar
and the grout.
Sleeve AS-01 fractured at the discontinuity of bars. The
corru-gated aluminum sleeve generated excellent bond with the
grout,while the grout bonded with the spliced bars. The stress
generatedby the tensile load transferred from the spliced bars to
the groutand to the sleeve. Due to low tensile resistance
properties andinadequate effective cross-sectional area of the
aluminum sleeve(about 1 mm thick), it fractured together with the
grout as its ten-sile capacity was exceeded.
The grout-sleeve bond failure demonstrated by CS-06 revealsthat
the interlocking mechanism between sleeve and grout is
-
J.H. Ling et al. / Construction and Building Materials 50 (2014)
530–539 535
utterly important. Without it, the splice would have to depend
onthe chemical adhesion and the surface friction between the
groutand the sleeve to prevent the grout from slipping out of the
sleeve[1], both of which were apparently less efficient than
mechanicalinterlocking mechanism.
The load–displacement curves in Fig. 4 show the behaviour ofthe
specimens. These curves represent the information such as(a) the
ultimate tensile capacity, which is rendered from the high-est
peaks of the curves, (b) the stiffness, which is represented bythe
tangent slopes of the curves, (c) the yield points, which is
iden-tified upon the first drastic decrease in the slope after the
elasticresponse, and (d) the ductility characteristic, which is
determinedfrom the degree of the post-yielding. The specimens with
the load–displacement responses similar a 16 mm high strength steel
bar
0
20
40
60
80
100
120
140
160
Displacement (mm)
Load
(kN
)
Y16-1BS-11BS-12BS-13
0
20
40
60
80
100
120
140
160
Displacement (mm)
Load
(kN
)
Y16-1BS-05BS-06BS-07
0
20
40
60
80
100
120
140
160
0 5 10 15 20 25 30 35
0 5 10 15 20 25 30 35
0 5 10 15 20 25 30 35Displacement (mm)
Load
(kN
) Y16-1AS-01AS-02
Fig. 4. Load–displacement graph of test sp
(control specimen) were considered adequate. This is based onthe
fundamental that splice connection should not compromisethe ability
(structural performance) of the spliced bars.
3.2. Analysis for feasibility
Table 2 shows the feasibility evaluations of the specimens,which
are derived from the measured variables presented in Ta-ble 1. The
yielding strength, Py, and displacement, dy, of the speci-mens are
obtained from the load–displacement curves presentedin Fig. 4, when
the stiffness of the specimens degraded suddenly.The bar stress at
the ultimate and yielding states, fbu and fby respec-tively, is
calculated by dividing the applied force with the nominalcross
sectional area of the bar, as given in the following equations:
0
20
40
60
80
100
120
140
Displacement (mm)
Load
(kN
)
Y16-1CS-01CS-02CS-03
0
20
40
60
80
100
120
140
160
Displacement (mm)
Load
(kN
)Y16-1BS-01BS-02BS-03BS-04
0
20
40
60
80
100
120
140
160
Displacement (mm)
Load
(kN
)
Y16-1BS-08BS-09BS-10
0 5 10 15 20 25 30 35
0 5 10 15 20 25 30 35
0 5 10 15 20 25 30 35
ecimens up to 35 mm displacement.
-
0
20
40
60
80
100
120
140
Displacement (mm)
Load
(kN
) Y16-1CS-04CS-05CS-06
0
20
40
60
80
100
120
140
Displacement (mm)
Load
(kN
) Y16-1CS-07CS-08CS-09
0
20
40
60
80
100
120
140
Displacement (mm)
Load
(kN
)
Y16-1DS-09DS-10DS-11
0
20
40
60
80
100
120
140
Displacement (mm)
Load
(kN
) Y16-1DS-01DS-02DS-03DS-04
0
20
40
60
80
100
120
140
0 5 10 15 20 25 30 35 0 5 10 15 20 25 30 35
0 5 10 15 20 25 30 35
0 5 10 15 20 25 30 35 0 5 10 15 20 25 30 35
Displacement (mm)
Load
(kN
)
Y16-1DS-05DS-06DS-07DS-08
Fig. 4 (continued)
536 J.H. Ling et al. / Construction and Building Materials 50
(2014) 530–539
fbu ¼Pu � 103
p db2� �2 ¼ 4Pu � 10
3
pd2bð1Þ
fby ¼Psy � 103
p db2� �2 ¼ 4Psy � 10
3
pd2bð2Þ
Steel bars are connected in-line in the grouted splice
connector.Due to the discontinuity, the materials that bridge the
bars wouldhave to endure the tensile load. This involved the grout
and thesleeve.
The tensile resistance of the grout is estimated by
multiplyingits effective cross sectional area, Agr, with its
tensile capacity, which
is conservatively approximated as 1/10 of its compressive
strength[44], as expressed in Eq. (3). The effective cross
sectional areas ofthe connectors vary depending on the size and the
shape of thesleeve as shown in Table 3.
Pgr ¼ Agr � fgr ¼ Agr �1
10fcu ð3Þ
The tensile load resisted by the sleeve is expressed in the
fol-lowing equation:
Psl ¼ Pu � Pgr ð4Þ
It is assumed that the tensile stress is equally shared by
theeffective cross sectional area of the sleeve, although some of
whichwere made from different materials. For example, BS-01 was
made
-
Table 2Feasibility evaluation of test specimens.
Specimen ld Asl (mm2) Agr (mm2) Psl (kN) Py (kN) fy(N/mm2)
dy (mm) fbu(N/mm2)
fsu(N/mm2)
ub(N/mm2)
esy(mm)
Reff Rst D Remark–b
AS-01 150 67 1184 5 11.5 57 1.9 59 75 1.6 0.8 0.3 0.14 1.4
NAAS-02 300 67 983 129.9 109.1 543 4.3 674 –a 9.0 31.2 –a 1.64 8.3
ABS-01 150 1297 2803 107.9 106.1 528 3.1 625 83 16.7 32.9 0.33 1.52
11.6 ABS-02 150 983 3117 99.6 122.1 607 3.5 593 101 15.8 18.9 0.40
1.45 6.4 ABS-03 150 983 3117 121.6 98.0 487 2.5 702 124 18.7 29.5
0.50 1.71 12.8 ABS-04 150 983 3117 98.4 98.1 488 3 587 100 15.7
20.5 0.40 1.43 7.8 ABS-05 150 983 3117 124.0 112.2 558 3.5 714 126
19.0 38.7 0.50 1.74 12.1 ABS-06 150 983 3117 105.3 106.2 528 3 621
107 16.6 33.9 0.43 1.51 12.3 ABS-07 150 983 3117 104.8 108.1 538
2.8 619 107 16.5 33 0.43 1.51 12.8 ABS-08 150 983 3117 101.6 102.2
508 2.6 603 103 16.1 33.7 0.41 1.47 14.0 ABS-09 150 983 3117 104.8
106 527 3.6 619 107 16.5 34.2 0.43 1.51 10.5 ABS-10 150 983 3117
116.7 118.1 587 3.3 678 119 18.1 27.9 0.48 1.65 9.5 ABS-11 150 983
3117 100.4 102.1 508 2.7 597 102 15.9 34.6 0.41 1.46 13.8 ABS-12
150 983 3117 105.5 108.1 538 3.2 622 107 16.6 34.9 0.43 1.52 11.9
ABS-13 150 983 3117 115.8 118.1 587 3.4 673 118 18.0 26.1 0.47 1.64
8.7 ACS-01 100 1070 1473 80.6 80.1 398 4.6 433 75 17.3 4.12 0.30
1.06 1.9 NACS-02 50 1070 1473 92.0 60.0 298 4.2 489 86 –c 22.1 0.34
1.19 6.3 NACS-03 150 1070 1473 73.6 76.1 378 3.1 398 69 10.6 1.44
0.28 0.97 1.5 NACS-04 100 1052 1473 78.9 74.1 369 3.3 424 75 17.0
13.8 0.30 1.03 5.2 NACS-05 100 1283 1473 77.9 74.1 369 2.7 419 61
16.8 4.1 0.24 1.02 2.5 NACS-06 100 1070 1473 34.5 40.9 203 1 203 32
8.1 0 0.13 0.50 1.0 NACS-07 100 1070 1473 66.1 72.5 361 3.1 361 62
14.4 3.9 0.25 0.88 2.3 NACS-08 100 1070 1473 64.1 70.5 351 6.7 351
60 14.0 0.0 0.24 0.86 1.0 NACS-09 100 1070 1473 95.4 100.1 498 4.4
506 89 20.3 1.2 0.36 1.23 1.3 NADS-01 150 981 2299 117.8 112.0 557
4.4 642 120 17.1 37.2 0.48 1.57 9.5 ADS-02 150 981 2299 100.7 112.0
557 3.2 557 103 14.9 13.4 0.41 1.36 5.2 ADS-03 150 981 2299 112.1
106.1 528 3.3 614 114 16.4 38.6 0.46 1.50 12.7 ADS-04 150 981 2299
99.3 110.6 550 4.8 550 101 14.7 0.0 0.40 1.34 1.0 NADS-05 150 981
2299 118.8 112.0 557 4.2 647 121 17.3 39.5 0.48 1.58 10.4 ADS-06
150 981 2299 112.4 108.1 538 3.9 615 115 16.4 25.5 0.46 1.5 7.5
ADS-07 150 981 2299 116.2 112.1 558 5.8 634 118 16.9 27.5 0.47 1.55
5.7 ADS-08 150 981 2299 111.0 106.0 527 3.8 608 113 16.2 27.8 0.45
1.48 8.3 ADS-09 150 981 2299 109.4 104.1 518 4.5 600 112 16.0 32.6
0.45 1.46 8.2 ADS-10 150 981 2299 106.9 100.1 498 3.7 588 109 15.7
32.5 0.44 1.43 9.8 ADS-11 150 981 2299 111.1 104.1 518 4.4 609 113
16.2 34.4 0.45 1.49 8.8 A
a The stress in the sleeve longitudinal to the direction of the
force is not measurable due to the adjacent arrangement of the
bars. This arrangement does not require thesleeve to resist the
pulling force.
b The rating for the feasibility of the specimens: ‘‘A’’
indicates that the specimen meets the evaluation criteria (fbu/fs P
1.25, and du/dy P 4.0), while ‘‘NA’’ represents thatthere is at
least one of the evaluation criteria is not met.
c The maximum bond stress was hard to determine because of early
failure of the bond and the tensile resistance of the specimen
relied upon the interlocking between thenuts on the bars and the
welded plates on the sleeve.
Table 3Cross sectional area of grout in sleeves.
Specimens Cross sectional area of grout in sleeves, Agr
DS-SeriesD2si � p
db2
� �2BS-01
p Dsi2� �2
� p db2� �2� �
� 4p dwb2� �2
All other specimens p Dsi2� �2
� p db2� �2
J.H. Ling et al. / Construction and Building Materials 50 (2014)
530–539 537
up of high strength steel bars and mild steel pipes. Therefore,
thestress developed in sleeves, fsu, is estimated by dividing the
loadwith the effective cross sectional area, Asl, as derived in
Table 4.
fsu ¼Psl � 103
Aslð5Þ
Substitute, Asl, from Table 4 into Eq. (5), acquires equations
inTable 5, which is generalized into the following equation:
fsu ¼ K � ðPu � PgrÞ ð6Þ
where K ¼ 4�103pðD2so�D2siÞþC
, C ¼ 4pd2wb, for specimen BS-01; 16bwphwp, forCS-Series; 0, for
others specimens.
For DS-Series, Eq. (6) is multiplied with p/4 in order to
cancelout the factor of 4/p in the equation in order to obtained
the equa-tion as presented in Table 5.
The average bond stress, ub, is calculated from the existing
for-mula as a function of the total surface area of embedded in
thegrout (Eq. (7)).
ub ¼Pu � 103
pdblbð7Þ
The post yielding elongation of the bar, epy, is expressed in
Eq.(8), with the assumptions that the bond-slip of the bar and
theelongation of the pipe are negligible.
epy ¼ du � dsy ð8Þ
The efficiency ratio of the sleeve indicates the level of
capacityutilization. It is calculated by dividing the stress
developed in thesleeve at the ultimate state, fsl, with the
specified yield strengthof the material used for the sleeve, as
expressed in the followingequation:
Reff ¼fslfsly
ð9Þ
ACI-318 [37] and AC-133 [38] recommend the acceptance crite-ria
for a mechanical splice. The capacity a splice should be at
least125% of the specified yield strength of the bar. BS-8110
[45],
-
Table 4Effective cross sectional area of grouted splices.
Specimen Cross sectional area of sleeve, Asl
AS-01, AS-03p Dso
2
� �2� p Dsi
2
� �2
BS-01Aps þ 4Awb ¼ p
Dso2
� �2� p Dsi
2
� �2" #þ 4 p dwb
2
� �2" #
BS-02 to BS-13p Dso
2
� �2� p Dsi
2
� �2
CS-SeriesAps þ 4Awp ¼ p
Dso2
� �2� p Dsi
2
� �2" #þ 4bwphwp
DS-Series D2so � D2si
Table 5Stress developed in sleeves, fsu, for different
specimens.
Specimen Stress in sleeves, fsu
AS-01 4ðPu�Pgr Þ�103pðD2so�D
2siÞ
BS-01 4ðPu�Pgr Þ�103pðD2so�D
2siþ4d
2wbÞ
BS-02 to BS-13 4ðPu�Pgr Þ�103pðD2so�D
2siÞ
CS-Series 4ðPu�Pgr Þ�103pðD2so�D
2siÞþ16bwp hwp
DS-Series ðPu�Pgr Þ�103D2so�D
2si
538 J.H. Ling et al. / Construction and Building Materials 50
(2014) 530–539
however, in Clause 2.4.2.2 recommended that the design
strengthof a material and the limit state for analysis should be
derived fromthe characteristic strength divided by the partial
safety factor, cm,of 1.4. Therefore, the ultimate strength of the
specimens shouldbe at least 1.4 times the specified yield strength
of the spliced barsfor the design analysis. For these reasons, the
calculated relativestrength ratio of the specimens, Rst, which is
expressed in Eq.(10), should be at least 1.25 and 1.4
accordingly.
Rst ¼fufy
ð10Þ
It is generally accepted that the ductility requirement of a
struc-ture in low-moderate seismic regions should be at least 4.0
[46].Hence, the ductility ratio of the specimens, D, (Eq. (11))
shouldbe greater than 4.0.
D ¼ dudsy
ð11Þ
4. Discussion
In this study, the feasibility of the test specimens is
determinedbased on two major criteria, where the tensile strength
ratio, Rst,should be at least 1.25 and the ductility ratio, D,
should be greaterthan 4.0. Twenty-three out of thirty-four
specimens meet therequirements, where most of which were BS-Series
and DS-Series.
The results show that not all of the specimens which
sufferedbar-grout bond failure were not feasible. Specimens like
AS-01,DS-06, DS-07, DS-08 and DS-09 could still give Rst and D
largerthan 1.25 and 4.0, respectively. It only happened when the
bondfailed after yielding of the spliced bar.
Table 2 shows that the efficiency ratios of the sleeves, Reff,
ran-ged from 0.13 to 0.50. Less than 50% of the capacity of the
sleevewas utilized, which was rather inefficient. The sleeve
provides ten-sile resistance in two directions; longitudinally and
perpendicularto the alignment of the bars [23,25]. The tensile
resistance in thelongitudinal direction is more critical as it
needs to sustain all
the stress induced by the applied tensile load, while the
perpendic-ular direction is solely used to resist the lateral
expansion of thegrout due to propagation of splitting cracks. Thus,
as long as suffi-cient resistance in the longitudinal direction is
provided, the thick-ness of the sleeve can be reduced. An excessive
provision of thethickness of the sleeve would offer no further
enhancement tothe capacity of a splice connection.
5. Conclusion
In this study, thirty-five splice connectors were made
fromstandard size steel sections. These specimens were tested
underincremental tensile load to assess for feasibility according
to ACI-318 and BS-8110. Twenty-three out of them were feasible.
It is concluded that:
(i) From the failure modes, the bar-grout bond,
grout-sleevebond, sleeve’s tensile capacity and the spliced bar’s
tensilecapacity are recognized as the factors that govern the
failureof splice connector. Hence, these factors must be
consideredduring the design of splice connectors.
(ii) Different configurations of sleeve leads to different forms
ofstress distribution and different degree of confinement.
Thissubsequently causes different efficiency of bond betweenbar and
grout.
(iii) By taking the advantages of confinement effects
generatedfrom sleeve and tapered nuts, the required developmentcan
be as short as nine times the bar diameter or less, whichis
approximately 26% of the anchorage length recommendedby BS8110 (35
times diameter of bar).
(iv) The equations derived in Section 3.2 provide the basis for
thedesign and evaluation of a typical grouted splice connectorwith
the bars spliced in-line.
(v) The experiments and analysis presented are based on
thestatic incremental load without considering the
potentialslippage of the splice bars. Moreover, the degree of
confine-ment might vary depending upon the tensile force
applied[25]. Hence, unless the behaviour of these splice
connectionsis proven adequate under cyclic and fatigue loads, the
spliceshould not be used in structures subjected to cyclic or
fati-gue loads.
Acknowledgments
The authors gratefully thank the Construction Industry Re-search
Institute of Malaysia (CREAM) and Construction IndustryDevelopment
Board (CIDB) for financial support under researchGrant Vot
73713.
References
[1] Lutz LA. The mechanics of bond and slip of deformed
reinforcing bars inconcrete. ACI J Proc 1966.
[2] Hamad BS. Comparative bond strength of coated and uncoated
bars withdifferent rib geometries. ACI Mater J
1995;92(6):579–90.
[3] Darwin D, Graham EK. Effect of deformation height and
spacing on bondstrength of reinforcing bars. ACI J
1993;90(6):646–57.
[4] Abrams DA. Tests of bond between concrete and steel. Urbana:
University ofIllinois; 1913. p. 71.
[5] Clark AP. Bond of concrete reinforcing bars. Detroit,
Michigan; 1949.[6] Ferguson PM, Breen JE, Jirsa JO. Reinforced
concrete fundamentals. John Wiley
& Sons, Inc.; 1988.[7] Mains RM. Measurement of the
distribution of tensile and bond stresses along
reinforcing bars. ACI J Proc 1951;48(11):225–52.[8] Goto Y.
Cracks formed in concrete around deformed tension bars. ACI J
Proc
1971;68(4):244–51.[9] Tepfers R. A theory of bond applied to
overlapped tensile reinforcement splices
for deformed bars. Chalmers University of Technology; 1973.
http://refhub.elsevier.com/S0950-0618(13)00920-3/h0005http://refhub.elsevier.com/S0950-0618(13)00920-3/h0005http://refhub.elsevier.com/S0950-0618(13)00920-3/h0010http://refhub.elsevier.com/S0950-0618(13)00920-3/h0010http://refhub.elsevier.com/S0950-0618(13)00920-3/h0015http://refhub.elsevier.com/S0950-0618(13)00920-3/h0015http://refhub.elsevier.com/S0950-0618(13)00920-3/h0020http://refhub.elsevier.com/S0950-0618(13)00920-3/h0020http://refhub.elsevier.com/S0950-0618(13)00920-3/h0025http://refhub.elsevier.com/S0950-0618(13)00920-3/h0025http://refhub.elsevier.com/S0950-0618(13)00920-3/h0030http://refhub.elsevier.com/S0950-0618(13)00920-3/h0030http://refhub.elsevier.com/S0950-0618(13)00920-3/h0035http://refhub.elsevier.com/S0950-0618(13)00920-3/h0035http://refhub.elsevier.com/S0950-0618(13)00920-3/h0040http://refhub.elsevier.com/S0950-0618(13)00920-3/h0040
-
J.H. Ling et al. / Construction and Building Materials 50 (2014)
530–539 539
[10] Gambarova PG, Rosati GP, Schumm CE. Bond and splitting: a
vexing question.Int Concr Res Inform Portal 1998;SP-180:23–44.
[11] Rosati GP, Schumm CE. Modeling of local bar-to-concrete
bond in RC beams.In: International conference ‘‘bond in concrete:
from research to practice’’,Riga, Latvia; 1992.
[12] Chamberlin SJ. Spacing of spliced bars in tension pull-out
specimens. Proc J AmConcr Inst 1952;49(4):261–74.
[13] Chinn J, Ferguson PM, Thompson JN. Lapped splices in
reinforced concretebeams. Proc J Am Concr Inst
1955;52(2):201–12.
[14] Orangun CO, Jirsa JO, Breen JE. A reevaluation of test data
on developmentlength and splices. ACI J 1977;74(3):114–22.
[15] Lormanometee S. Bond strength of deformed reinforcing bar
under lateralpressure. University of Texas at Austin; 1974.
[16] Untrauer RE, Henry RL. Influence of normal pressure on bond
strength. ACI J1965;65(5):577–85.
[17] Robins PJ, Standish IG. The influence of lateral pressure
upon anchorage bond.Mag Concr Res 1984;36(129):195–202.
[18] Moosavi M, Jafari A, Khosravi A. Bond of cement grouted
reinforcing barsunder constant radial pressure. Cem Concr Compos
2005;27:103–9.
[19] Eligehausen R, Popov EP, Bertero VV. Local bond stress-slip
relationships ofdeformed bars under generalized excitations. Univ.
of California; 1983.
[20] Soroushian P, Choi KB, Park GH, Aslani F. Bond of deformed
bars to concrete:effects of confinement and strength of concrete.
ACI Mater J 1991:227–32.
[21] Lim CT. The effect of pitch distance of steel spiral
reinforcement to theperformance of grouted sleeve connector under
direct tensile load. UniversitiTeknologi Malaysia; 2010.
[22] Lee GS. Parametric studies of sleeve connector using steel
pipe with spiral steelfor precast concrete connection. Universiti
Teknologi Malaysia; 2009.
[23] Einea A, Yamane T, Tadros MK. Grout-filled pipe splices for
precast concreteconstruction. Precast/Prestr Concr Inst J
1995;40(1):82–93.
[24] Loh HY. Development of grouted splice sleeve and its
performance under axialtension. Universiti Teknologi Malaysia;
2008.
[25] Ling JH. Behaviour of grouted splice connections in precast
concrete wallsubjected to tensile, shear and flexural loads.
Universiti Teknologi Malaysia;2011.
[26] Tibbetts AJ, Oliva MG, Bank LC. Durable fiber reinforced
polymer bar spliceconnections for precast concrete structures. In:
Compos Polycon. FL,USA: Tampa; 2009.
[27] Yee AA. Splice sleeve for reinforcing bars. US Pat.
3,540,763; 1968.[28] Yee AA. A splice sleeve for reinforcing bars
with cylindrical shell. US Pat.
4627212; 1986.
[29] Colarusso LJ, Gregel JJ. Reinforcing bar splice and method.
EP1231332 B1;2007.
[30] Ali Eldin SS, Al-Tuhami AA. Mechanical reinforcing bar
coupler based on bardeformations. US 20100104357 A1; 2010.
[31] Yee AA. Splice sleeve for connecting reinforcing bars to
another entity. EP0570848 A2; 1993.
[32] Albrigo J, Ricker ED, Colarusso LJ. Method of forming
concrete structures with agrout splice sleeve which has a threaded
connection to a reinforcing bar. USPat. 5,366,672; 1994.
[33] Albrigo J, Ricker ED, Colarusso LJ. Reinforcing bar splice
and system for formingprecast concrete members and structures. US
Pat. 5,468,524; 1995.
[34] Lancelot HB. Mechanical splices of reinforcing bars. The
Aberdeen, Group;1985.
[35] Coogler KL. Investigation of the behaviour of offset
mechanicalsplices. University of Pittsburgh; 2006.
[36] Jansson PO. Evaluation of grout-filled mechanical splices
for precast concreteconstruction. Michigan Department of
Transportation MDOT; 2008.
[37] ACI-318. Building code requirements for structural concrete
andcommentary. American Concrete Institute; 2002.
[38] AC-133. Acceptance criteria for mechanical connector
systems for steelreinforcing bars. ICC Evaluation Service, Inc.;
2008.
[39] Coogler KL, Harries KA, Gallick M. Experimental study of
offset mechanical lapsplice behavior. ACI Struct J
2008;105(4):478–87.
[40] Ling JH, Abd. Rahman AB, Abd. Hamid Z. Failure modes of
aluminium sleeveunder direct tensile load. In: 3rd International
conference on postgraduateeducation (ICPE-3). Malaysia: Penang;
2008.
[41] Kim YM. A study of pipe splice sleeves for use in precast
beam–columnconnections. The University of Texas at Austin;
2000.
[42] Abd. Rahman AB, Ling JH, Ibrahim IS, Abd. Hamid Z.
Performance of groutedsleeve connectors subjected to incremental
tensile loads. Malaysian Constr ResJ (MCRJ) 2010;6(1):39–55.
[43] Ling JH, Abd. Rahman AB, Mirasa AK, Abd. Hamid Z.
Performance of CS-sleeveunder direct tensile load: Part I – failure
modes. Malaysian J Civ Eng2008;20(1):89–106.
[44] Technical Data Sheet. Sikagrout-215. Sika Kimia Sdn Bhd;
2009.[45] BS8110-1:1997. Structural use of concrete – Part 1: code
of practice for design
and construction. British Standards BSI; 1997.[46] Soudki KA,
Rizkalla SH, LeBlanc B. Horizontal connections for precast
concrete
shear wall subjected to cyclic deformations Part 1: mild steel
connections. PCIJ 1995;41(1):78–96.
http://refhub.elsevier.com/S0950-0618(13)00920-3/h0045http://refhub.elsevier.com/S0950-0618(13)00920-3/h0045http://refhub.elsevier.com/S0950-0618(13)00920-3/h0050http://refhub.elsevier.com/S0950-0618(13)00920-3/h0050http://refhub.elsevier.com/S0950-0618(13)00920-3/h0055http://refhub.elsevier.com/S0950-0618(13)00920-3/h0055http://refhub.elsevier.com/S0950-0618(13)00920-3/h0060http://refhub.elsevier.com/S0950-0618(13)00920-3/h0060http://refhub.elsevier.com/S0950-0618(13)00920-3/h0065http://refhub.elsevier.com/S0950-0618(13)00920-3/h0065http://refhub.elsevier.com/S0950-0618(13)00920-3/h0070http://refhub.elsevier.com/S0950-0618(13)00920-3/h0070http://refhub.elsevier.com/S0950-0618(13)00920-3/h0075http://refhub.elsevier.com/S0950-0618(13)00920-3/h0075http://refhub.elsevier.com/S0950-0618(13)00920-3/h0080http://refhub.elsevier.com/S0950-0618(13)00920-3/h0080http://refhub.elsevier.com/S0950-0618(13)00920-3/h0085http://refhub.elsevier.com/S0950-0618(13)00920-3/h0085http://refhub.elsevier.com/S0950-0618(13)00920-3/h0090http://refhub.elsevier.com/S0950-0618(13)00920-3/h0090http://refhub.elsevier.com/S0950-0618(13)00920-3/h0095http://refhub.elsevier.com/S0950-0618(13)00920-3/h0095http://refhub.elsevier.com/S0950-0618(13)00920-3/h0095http://refhub.elsevier.com/S0950-0618(13)00920-3/h0100http://refhub.elsevier.com/S0950-0618(13)00920-3/h0100http://refhub.elsevier.com/S0950-0618(13)00920-3/h0105http://refhub.elsevier.com/S0950-0618(13)00920-3/h0105http://refhub.elsevier.com/S0950-0618(13)00920-3/h0110http://refhub.elsevier.com/S0950-0618(13)00920-3/h0110http://refhub.elsevier.com/S0950-0618(13)00920-3/h0115http://refhub.elsevier.com/S0950-0618(13)00920-3/h0115http://refhub.elsevier.com/S0950-0618(13)00920-3/h0115http://refhub.elsevier.com/S0950-0618(13)00920-3/h0120http://refhub.elsevier.com/S0950-0618(13)00920-3/h0120http://refhub.elsevier.com/S0950-0618(13)00920-3/h0120http://refhub.elsevier.com/S0950-0618(13)00920-3/h0125http://refhub.elsevier.com/S0950-0618(13)00920-3/h0125http://refhub.elsevier.com/S0950-0618(13)00920-3/h0130http://refhub.elsevier.com/S0950-0618(13)00920-3/h0130http://refhub.elsevier.com/S0950-0618(13)00920-3/h0135http://refhub.elsevier.com/S0950-0618(13)00920-3/h0135http://refhub.elsevier.com/S0950-0618(13)00920-3/h0140http://refhub.elsevier.com/S0950-0618(13)00920-3/h0140http://refhub.elsevier.com/S0950-0618(13)00920-3/h0145http://refhub.elsevier.com/S0950-0618(13)00920-3/h0145http://refhub.elsevier.com/S0950-0618(13)00920-3/h0150http://refhub.elsevier.com/S0950-0618(13)00920-3/h0150http://refhub.elsevier.com/S0950-0618(13)00920-3/h0155http://refhub.elsevier.com/S0950-0618(13)00920-3/h0155http://refhub.elsevier.com/S0950-0618(13)00920-3/h0155http://refhub.elsevier.com/S0950-0618(13)00920-3/h0160http://refhub.elsevier.com/S0950-0618(13)00920-3/h0160http://refhub.elsevier.com/S0950-0618(13)00920-3/h0165http://refhub.elsevier.com/S0950-0618(13)00920-3/h0165http://refhub.elsevier.com/S0950-0618(13)00920-3/h0165http://refhub.elsevier.com/S0950-0618(13)00920-3/h0170http://refhub.elsevier.com/S0950-0618(13)00920-3/h0170http://refhub.elsevier.com/S0950-0618(13)00920-3/h0170http://refhub.elsevier.com/S0950-0618(13)00920-3/h0175http://refhub.elsevier.com/S0950-0618(13)00920-3/h0180http://refhub.elsevier.com/S0950-0618(13)00920-3/h0180http://refhub.elsevier.com/S0950-0618(13)00920-3/h0185http://refhub.elsevier.com/S0950-0618(13)00920-3/h0185http://refhub.elsevier.com/S0950-0618(13)00920-3/h0185
Feasibility study of grouted splice connector under tensile
load1 Introduction2 Experimental program2.1 Specimens2.2 Test plan
and setup
3 Test results3.1 Failure modes3.2 Analysis for feasibility
4 Discussion5 ConclusionAcknowledgmentsReferences