Structural Engineering and Mechanics, Vol. 21, No.

2 (2005) 221-235

221

Experimental investigation on the shear capacity of RC

dapped end beams and design recommendations
Quanfeng Wang and Zixiong Guo
College of Civil Engineering, Huaqiao University, 362011 Quanzhou, Fujian, China

Pierre C.J. Hoogenboom

Department of Civil Engineering, Delft University of Technology, 2600 GA Delft, Netherlands
(Received January 19, 2004, Accepted June 24, 2005)

Abstract. In this paper, the shear resistance behaviour of reinforced concrete (RC) dapped end beams is
investigated by 24 tests until failure load. The main parameters considered are the dapped end height, the
type and effective range to provided the stirrups and the bent form of the longitudinal reinforcement. The
failure behaviour of dapped end beams is presented and some conclusions are given. Inclined stirrups and
longitudinal bent reinforcement have more influence on the shear capacity than vertical stirrups.
Additionally, the shear mechanism of dapped end beams is analysed. Relatively simple semi-empirical
equations for shear strength have been derived based on the results of 22 dapped end beams. The
predicted results are in close agreement with the experimental ones. Finally, some design suggestions for
the ultimate shear strength of dapped end beams are presented.
Key words: reinforced concrete (RC); dapped end beams; shear capacity; shear failure; ultimate shear
strength; shear resistance mechanisms.

1. Introduction
The extensive study of the behaviour of reinforced concrete flexural members has clarified the
flexural behaviour mechanism to the extent that well-understood conclusions are now incorporated
in the design codes of many countries. Progress in the understanding and quantitative assessment of
the behaviour of members subjected to flexure and shear have been somewhat less spectacular. The
majority of structural members in RC must resist shear forces. These forces seldom act on their own
but rather in combination with flexural, axial load, and perhaps torsion. The problem of how shear
failures occur in RC still remains, despite numerous extensive studies over the last 50 years (Zararis
et al. 2001, Leonhardt et al. 1984, Schlaich et al. 1987, Thomas 1998). Because of these
complexities, codified methods continue to include expressions for shear resistance based on fitting
curves to experimental data. Strong evidence of this is the fact that international codes, such as the
code of the American Concrete Institute (ACI) (1999) or the Eurocode 2 (1992) are based on rather
Professor, Corresponding author, E-mail: [email protected]
Associate Professor

222

Quanfeng Wang, Zixiong Guo and Pierre C.J. Hoogenboom

(semi-) empirical considerations (Zararis et al. 2001). Unfortunately, empirical results cannot be
correctly extrapolated to beams of more complicated shape such as RC dapped end beams. In the
international design community for structural concrete a distinction between B-regions and Dregions is used. B-regions are the parts of a structure in which the classic beam theory applies. The
remaining part of the structure consists of D-regions in which the fore-mentioned classic state is
disturbed. Examples are joints between beams and columns, zones around holes in the web of a
beam, deep beams, corbels, brackets, shear walls having an irregular shape (Blaauwendraad et al.
2000). A short beam having dapped ends would be considered as one D-region. RC dapped end
beams differ from general flexural members in that transfer of shear force may be of more critical
importance. In a dapped end beam, where inclined cracking under combined stress states occurs, a
significant redistribution of internal forces can be expected after cracking, and a large part of the
shear force is transferred by what is known as the truss mechanism (Leonhardt et al. 1984, Schlaich
et al. 1987, Thomas 1998). In addition to identifying the effect of shear forces acting alone, it is
necessary to examine the possible interaction with other structural actions. In flexural members in
particular, the shear resisting mechanisms interact intimately with the bond between concrete and
the embedded reinforcement of the latter. Though extensive experimental work, particularly in
recent years, has greatly extended the identification of various shear resistance mechanisms (Zararis
et al. 2001), very few tests on RC dapped end beams have been performed. Up to present, only a
total of 38 dapped end beam tests were report in literal by Mattock et al. (Mottaock et al. 1979,
Mottaock 1986), which included both RC and prestressed concrete dapped end beam. The present
paper reports the test results of 24 RC dapped end beams. For these beams, the effects of the
dapped end height, the type and effective range to provide the stirrups, and the bent form of the
longitudinal reinforcement on the shear capacity were studied in detail.

2. Research significance
This study adds test data of 24 RC dapped end beams to a field in which few test studies have
been performed and thus has theoretical and practical significance. The failure behaviour of dapped
end beams is presented, and some design suggestions, as well as a series of semi-empirical
equations for shear strength are given. These results are valuable references for engineers who are
confronted with the design of RC dapped end beams and other shear dominated components.

3. Experimental program
Shear transfer in reinforced concrete dapped end beams, as well as in general reinforced concrete
beams, relies heavily on the tensile and compression strength of the concrete. Thus it is not
surprising that a shear failure in general is non-ductile. In practice, flexural failure could occur
before shear failure. However, for these experiments the shear failure must occur prior to any other
type of failure such as bond, flexure, splitting etc. This implies that the maximum flexural strength
of the beam must be somewhat in excess of the shear strength it could possibly develop. In the test
program, 22 of the 24 specimens collapsed after arriving at the shear capacity. The collapse of 2
specimens (B1.21 and B2.32) resulted from the compression crushed of the cracked reinforced
concrete and splitting failure of concrete, respectively.

223

Experimental investigation on the shear capacity

3.1 Test specimens

Three series, B1, B2 and B3 of RC dapped end beams were tested. The overall dimensions of
each series are shown in Fig. 1. All tested beams had the same length of 1500 mm. The location of
centre lines of loads and supports were the same for all test specimens. The details of reinforcement
and the sizes for each beam are shown in Fig. 1 and Table 1.
The specimens range from I to III. In Table 2, the specimen number Bk.ij, subscript k is the
number of the beam series 1, 2, 3; subscript i is the number of the beam and subscript j = 1 or 2,
which denote the first or second end of the beam, respectively.
In Table 1 fcu is the compressive strength of standard cube (150 150 150); fc is the cylinder
compressive strength; b and h are the width and depth of the dapped end beam as shown in Fig. 1,
respectively; h and h are the height and the effective height of the dapped end, respectively; is
the shear span/depth ratio of the dapped end beam; a is the distance from the load applied point to
the support as shown in Fig. 2; 1 6 means a 6 mm diameter vertical stirrup with 2 legs; and
1 14 (45 ) means one 14 mm diameter normal mild bar bent along 45 ; a is the distance of the
first stirrup to the end face of dapped end beam; and a is the effective range of stirrups.
1

10

Table 1 Specimen details and failure loads VTu

Range of
Concrete
Web
Dimensions (mm)

Stirrups
mm
Strength
MPa
Reinforcement
Series No.
= a/h
Provided
fc
b
h
h
h
a
a
fcu
50
50
B1.11 16.90 11.32 214 370 170 145 3.38
18+114 (90 )
B1.12
16.90
11.32
214
370
164
139
3.23
1

6
60
60
I B1.21 16.01 10.73 220 370 190 165 2.83
18+118 (45 )
105 105
B1.22 16.01 10.73 220 370 160 135 3.13
28@42
45
87
B2.11 19.53 13.09 150 300 150 125 2.0
16
30
30
B2.12 19.53 13.09 150 300 150 125 2.0
16
100 100
B2.21 19.53 13.09 150 300 150 125 2.0
26@70
30
100
II B2.22 19.53 13.09 150 300 150 125 1.9
26@64
40
104
B2.31 19.53 13.09 150 300 150 125 1.87
16+114 (45 )
40
40
B2.32 19.53 13.09 150 300 150 125 2.0
16+114 (45 )
75
75
B3.11 25.89 18.12 150 315 160 130 2.31
26@30
30
60
B3.12 25.89 18.12 150 315 160 130 2.31
26@30
22
52
B3.21 25.89 18.12 150 300 150 120 2.50
26@35
25
60
B3.22 25.89 18.12 150 300 150 120 2.50
26@30
45
75
B3.31 22.82 15.97 150 305 150 120 2.50
16 with 4 legs
50
50
B3.32 22.82 15.97 150 305 150 120 2.50 26@45 with 4 legs
20
65
B3.41 22.82 15.97 150 310 100 75 4.07
16@35
30
30
III B3.42 22.82 15.97 150 310 100 75 4.0
16@35
40
40
B3.51 22.82 15.97 150 305 155 125 2.4
16+114 (45 )
30
30
B3.52 22.82 15.97 150 305 150 120 2.5
16+114 (45 )
30
30
B3.61 22.67 15.19 150 300 150 128 2.34
26+114 (90 )
38
68
B3.62 22.67 15.19 150 300 150 128 2.42 26+18 with slop legs 23
58
B3.71 22.67 15.19 150 300 150 128 2.34
26+18 (45 )
23
58
B3.72 22.67 15.19 150 300 150 128 2.38
26+28 (45 )
13
56
10

10

o
o

o
o
o

o
o

Failure
Load
VTu (kN)
58.77
42.24
65.00
72.98
34.75
31.75
55.22
65.83
75.73
65.0
60.0
70.0
66.0
50.0
52.2
63.0
26.8
26.4
79.1
75.0
63.2
92.8
90.0
116.2

224

Quanfeng Wang, Zixiong Guo and Pierre C.J. Hoogenboom

Fig. 1 Specimen geometry and reinforcement detailing

The form and the amount of web reinforcement included three levels: a low amount, a large
amount, and with the longitudinal bar bent and inclined at an angle of 45 to the horizontal. For
Series II, called B2.ij, these were varied to obtain six different dapped end beams as shown in
Fig. 1(a). For Series I and III, called B1.ij or B3.ij, in addition, the height of the dapped end was
varied to obtain different shear spanto-depth ratios as shown in Figs. 1(a), (b). The difference
between Series I and III is the form of web reinforcement including the inclined stirrup, the stirrup
with 4 legs and fc. The longitudinal reinforcement was kept constant for each series except for the
beams in Series III (B3.7), where the amount of the reinforcement was increased. All main
longitudinal reinforcements extended over the full length of the beams bottom and were welded on
steel plates embedded at the ends of the beams to provide sufficient anchorage (Fig. 1).
o

3.2 Material
The constituents of the concrete were ordinary Portland cement, irregular gravel of 10 mm
maximum size, and sand. The mix proportions by weight are given in Table 2 for the beams in
Series I and II, and in Table 3 for the beams in Series III.
All specimens were cast in wooden formwork and specially differed in the dapped end height and
the overall depth. The following control specimens were made during casting of each beam: three
150 150 150 mm cubes for beams in every series. The cubes were tested on the same day of
3

Experimental investigation on the shear capacity

225

Table 2 Concrete constituents for beams in Series I and II

Ingredients
Weight (kN/m ) Weight/Cement content
Water
1.95
0.62
Cement
3.15
1.00
Sand
6.57
2.09
Gravel
11.19
3.55
3

Table 3 Concrete constituents for beams in Series III

Ingredients
Weight (kN/m ) Weight/Cement content
Water
1.80
0.46
Cement
3.90
1.00
Sand
6.29
1.61
Gravel
12.00
3.08
3

the beam test. The cube compressive strength fcu shown in Table 1 was obtained from the average
of the compressive strengths of the tested cubes. The applied relation between the cube compressive
stength fcu and cylinder compressive strength fc is
fc = 0.67fcu

All reinforcing bars were rolled out of normal mild steel. The yield stress fy was obtained by
tension tests.

3.3 Instrumentation
The specimens were heavily instrumented to obtain as much information as possible. Most of the
results was recorded automatically using the data logger and the rest was measured manually. The
load cell reading was recorded automatically using the data logger. Steel strains were measured
using 5 3 mm electrical resistance strain gauges (ERS). All ERS gauges used in one beam were
concentrated on one end of the dapped end beam. The number of ERS gauges used was varied from
one end to another end depending on the amount of the web reinforcement. The concrete strains
were measured across 100 mm gage length demic points. The midspan deflection of each beam
was measured using a demountable mechanical strain gauge with 10 mm capacity.
2

3.4 Test setup

The test specimens were subjected to four-point-bending in a compression machine with a total
capacity of 500 kN. A top steel spreader beam was used to divide the total applied load from the
machine head into two equal point loads as shown in Fig. 2. The load was applied in increments of
10 kN until failure occurred. After each increment, the load was kept constant to allow marking of
the new cracks and running of the data logger. The test was under load control until the specimen
reached its peak strength.

226

Quanfeng Wang, Zixiong Guo and Pierre C.J. Hoogenboom

Fig. 2 Test set-up

4. Results of tests and analyses

4.1 Crack formation and shear failure modes
The typical crack pattern and failure modes of these dapped end beams reported in this paper are
shown in Fig. 3. In the tests, cracks always initiated at the reentrant corner of the dapped end beam
at approximate 40 -60 to the axis of the beam, resulting in a redistribution of stresses causing
increasing steel stresses, bond stresses and some bond slip. Under an additional load these cracks
spread, increased in number and reduced the compression zone of the beam considerably. At one or
a few load increments before failure, more inclined cracks (about 3), occurred at the reentrant
corner of the majority of dapped end beams, and steeper (50 -70 ) than the first inclined crack up to
o

Fig. 3 Failure modes

227

Experimental investigation on the shear capacity

the vicinity of the load point. A critical inclined crack occurred suddenly, which activated resistance
to stresses in the web reinforcement, aggregate interlock along the diagonal crack, if any exists, as
well as resistance in the uncracked concrete above the head of the crack. Under the increasing load,
the diagonal crack propagated towards the loading point causing an increase in the dowel shear, and
final failure occurred when the head of the diagonal crack had decreased the uncracked compression
block of concrete to a critical point at which time a so-called shear-compression failure occurred
under a combined state of stress. As the reinforcement was yielding, cracks became wide and the
deflection significantly increased. When the cracks had developed to some extent, the tension in the
web reinforcement at the vicinity of the reentrant corner increased as the inclined cracks gradually
developed.
A majority of the beams developed some cracks at the bottom edge before reaching their ultimate
failure due to redistribution of internal forces (See Figs. 3c and 3d). Exceptions were B2.22 which
had more web reinforcements, and B3.42 which had the smallest nib height (See Figs. 3a and 3b).
The tests show that the first inclined crack initiated at 13-45% of the failure load. Such inclined
cracks, known as diagonal tension cracks, were generally controlled adequately by the shear
reinforcement. The results indicate that the height of the cross section of the specimens played an
important part in the crack control. The effect of web reinforcement on the control of diagonal
tension cracks was not significant compared to the dimension of cross section.
The specimen B3.72, which had detailed with two vertical and two inclined stirrups, showed the
largest failure load, whereas the specimen B1.21 detailed with a similar amount of web
reinforcement to specimen B3.72 had relatively lower shear strength than that of B3.72. This is
because that the above two specimens have a rather different failure mechanism. The specimen
B3.72 reached its ultimate shear strength after its web reinforcement yielded, whereas the concrete
of specimen B1.21 was crushed before the web reinforcement yielded for its relatively low concrete
strength ( fc is only 10.73 MPa).
The crack pattern and failure mode of specimen B3.41 and B3.42 were clearly different from
other specimens (See Fig. 3b). The inclined cracks are concentrated above the support platen of the
nib and no crack can be seen under the reentrant corner. Noted that the nib depth of B3.41 and
B3.42 are only 0.33 h and their shear span to depth ratio is up to 4.0. This result in its resistant
mechanism is rather different compared to that of other specimens (Fig. 7). The internal arching
effect of B3.41 and B3.42 is weakened for their large shear span to depth ratio, and thus showed
relative low shear strength.

4.2 Strains of reinforcement and concrete

In some cases, the electrical resistance strain (ERS) gauges were near but not at major crack
positions, and so yielding has occurred at the crack even though this was not shown by the ERS
gauges. In the following the variation of strains in several reinforcing bars against the total applied
load is presented for B3.32. B3.32 had vertical stirrups with 4 legs as shown in Fig. 1(c). Fig. 4(a)
shows the positions of different ERS gauges attached to the stirrups at approximately 45 to the axis
of the beam. Fig. 5 presents the variation of steel strains of different bars and positions with the
total applied load. The longitudinal reinforcement 103 yielded first. The vertical stirrup with 4 legs
near the dapped end was very close to yielding. Just before failure, major redistribution of the
vertical stirrup strain took place and stirrup 102 yielded.
o

228

Quanfeng Wang, Zixiong Guo and Pierre C.J. Hoogenboom

Fig. 4 Positions of the strain gauges

Fig. 5 Shear-strain curve for B3.32 reinforcement

Fig. 6 Shear-strain curves for B3.51 concrete

Fig. 4(b) shows the positions of some of the ERS gauges attached to the concrete surface at
approximately 45 to the axis of the beam. Fig. 6 presents the strains recorded on the concrete
surface at the beginning of loading for beam B3.51. The strain of the concrete increased as the
loading increased, and the largest strain occurred at the vicinity of the reentrant corner. Along the
inclined crack the values of the strains decreased.
o

229

Experimental investigation on the shear capacity

4.3 Effect of stirrup variation on shear capacity

The reinforcement details for each specimen are shown in Fig. 1 and Table 1. In the test, 14
specimens were provided with vertical stirrups or closed stirrups. Only two of these specimens had
stirrup with 4 legs; three specimens were provided with vertical stirrups and inclined stirrups, and
10 specimens had vertical stirrups and bent longitudinal reinforcement. By comparing the shear
capacity of specimens with vertical stirrups (B3.11-B3.22) to specimens with inclined stirrups
(B3.62, B3.71 and B3.72), it can be concluded that the inclined stirrups had more influence on the
capacity than the vertical stirrups had. The reason is that when inclined reinforcement crosses a
crack subjected to shear displacements, a significant part of the shear force may be transferred by
dowel action in these bars. From the tests, it also can be seen that closed stirrups will enhance the
shear resistance capacity, but less than the inclined stirrups and the bent reinforcement.

4.4 Effect of the bent form of longitudinal reinforcement on shear capacity

The shear capacities of specimens B1.11 and B3.61 are given in Table 1. B1.11 and B3.61 with
combined vertical stirrups and web bar bent along 90 had a smaller shear capacity than B1.21 and
B3.52 with combined vertical stirrups and web bars bent along 45 . Therefore, it is recommended
not to use the bent form along 90 of the longitudinal reinforcement.
o

Table 4 Comparisons of predicted shear strength and experimental shear strength

fcbh
V Tu
V uc
V uc
Asv fyv Asb fysin
kN
kN
kN
kN
kN
kN
B1.12
336.72
24.42
42.24
39.34
95.32
B1.21* 389.50
33.60
61.94
65.00
112.05
81.35
B1.22
318.68
67.05
72.98
74.95
99.84
B2.11
245.44
24.42
34.75
34.36
105.65
B2.12
245.44
24.42
31.75
34.36
63.39
B2.21
245.44
48.84
55.22
55.32
105.65
B2.22
245.44
48.84
65.83
55.32
96.46
B2.31
245.44
24.42
37.47
75.73
71.83
96.46
B2.32** 245.44
24.42
37.47
65.00
71.83
73.95
B3.11
353.34
54.0
60.00
65.64
109.88
B3.12
353.34
67.0
70.00
76.80
118.94
B3.21
326.16
54.0
66.00
64.16
106.49
B3.22
326.16
54.0
50.00
64.16
88.75
B3.51
299.44
27.0
37.45
79.10
76.97
105.65
B3.52
299.44
27.0
37.45
75.00
76.97
101.42
B3.62
291.65
53.9
23.80
92.80
85.99
115.91
B3.71
291.65
53.9
23.80
90.00
85.99
115.91
B3.72
291.65
53.9
47.60
116.20
109.79
129.08
Note:
V uc is the calculated shear strength from Eq. (9); V uc is the calculated shear
T
V u is the experimental shear strength.
* -- the compression crushed of the cracked reinforced concrete;
** -- the splitting failure of the concrete.

No.

10

V uc/ V uT

V uc/ V uc

0.931
2.423
1.724
0.726
1.027
1.332
0.989
3.075
1.082
1.845
1.002
1.910
0.840
1.744
0.949
1.343
1.105
1.030
1.094
1.674
1.097
1.549
0.972
1.660
1.283
1.383
0.973
1.373
1.026
1.318
0.927
1.348
0.955
1.348
0.945
1.176
strength from Eq. (11);

230

Quanfeng Wang, Zixiong Guo and Pierre C.J. Hoogenboom

4.5 Effect of nib depth on shear capacity

In addition to the various stirrups provided, variables of importance to the shear capacity included
the depth of the dapped end and the effective range of the stirrups. The shear capacities of
specimens B1.12, B2.11 and B2.12 are given in Table 1. B1.12 had a higher dapped end than B2.11
and B2.12. We can see that B1.12 had the same reinforcement as B2.11 and B2.12 in Table 4, the
test data showed that B1.12 had larger shear capacity than B2.11 and B2.12, although the later had
larger compressive strength fc'. The same can be seen from the comparison between B3.4 and B1.12
or B2.11. B3.41 and B3.42 had a dapped end height of 0.33 h and showed a very small shear
strength. The effect of the dapped end height on shear capacity is significant. From the tests it is
recommended that the dapped end height should not be less than 0.45 h (Mottaock et al. 1979).

4.6 Effective range of beam where stirrups are provided

The distance, a of the first stirrup from the end face of dapped end beam and the length, a
where stirrups are provided (Fig. 7), affected the shear capacity of the dapped end beams. The
conditions of B2.11 and B2.12 were the same except that specimen B2.11 had a smaller value of a ,
whereas B2.11 showed a larger shear capacity than B2.12. The same can be seen in the second
group from the comparison between B3.21 and B3.22. Therefore the effect of the distance, a of the
first stirrup from the end face of the dapped end beam cannot be ignored. To obtaining this value,
the concrete covers of all first stirrups were removed after testing. The measured distance, a was
between 15 mm (prescribed concrete cover of stirrup) and 40 mm. The strains in the stirrups
decreased as the distance from the end face of dapped end beam increased. Note that not all the
stirrups provided in a could contribute themselves to the shear capacity of the dapped end beams.
In this paper, the length a in which the stirrups could effectively contribute to the shear capacity
are referred to as effective range. By analysing the variation of strains of the stirrups and directly
measuring the spacing of the stirrups, the effective range to provided the stirrups is recommended to
be ao = 0.5 h (Wang et al. 1996).
1

Fig. 7 The position of a and ao

1

231

Experimental investigation on the shear capacity

5. Mechanism of shear resistance and design for shear

5.1 Truss mechanism
5.1.1 Vertical stirrups provided only

The analogous truss shown in Fig. 8(a) depicts the case of orthogonal reinforcement. In Fig. 8, T
denoted tension (solid line) and D denoted compression (dashed line). It will serve to illustrate the
relation between external shear forces Vs, to be resisted by the truss, and the various internal forces.
The forces in the truss can be determined from considerations of equilibrium only:
Vs = Ts
(1)
T1 = Vs ----e(2)
Z1
in which Ts is the resultant of all stirrup forces within the effective range ao; T is the resultant in
the flexural reinforcement at the end of the dapped end beam; e is the distance from the centre of
the support to T and Z is the distance from T to the centre of D.
1

5.1.2 Inclined web reinforcement only

The analogous truss shown in Fig. 8(b) depicts the general case of web reinforcement inclined at
an angle to the horizontal. It will serve to illustrate the relation between external shear forces Vs,
to be resisted by the truss, and the various internal forces. The forces in the truss can be determined
from considerations of equilibrium only:
(3)
Vb = D 1
D1 = T sb sin

and

(4)

Tsb = Asb f y
Vb = A sb f y sin

Fig. 8 Internal forces in analogous trusses of the dapped end beam

(4a)

232

Quanfeng Wang, Zixiong Guo and Pierre C.J. Hoogenboom

In which Vb is the bearing capacity of dapped end beam when the inclined web reinforcement was
provided; Tsb is the resultant of all inclined web reinforcements provided; Asb is the area of the
inclined web reinforcement, and fy is the stirrup stress.

5.2 Design for shear of the dapped end beam

After cracking, substantial redistribution of internal forces occurred. For the dapped end beam, the
total shear resistance VN in a typical region is comprised of the sum of that of the still uncracked
portion of the concrete section and interface shear transfer Vc across the crack by aggregate
interlock and the force Vs transferred by direct tension in the stirrup
VN = V c + Vs

(5)

If the vertical stirrups and the inclined web reinforcement are combined, the following equation
results
VN = V c + Vs + V b
(6)

5.2.1 Vertical stirrups only

The shear resisting mechanism of a beam without web reinforcement, relying only on aggregate
interlock action, will function as long as the width of the cracks does not become excessive. The
interface shear transfer can be expressed as
Vc = 1 f c b h10

In which is a coefficient to be determined by regression analysis of the test results. None of the
stirrups yielded within their effective range according to the test results in Table 2. Hence in the
presence of stirrups, we use
Vs = 2 A sv f yv
1

In which is also a coefficient to be determined by regression analysis of the test results; Asv is the
total cross section area of the vertical stirrups within the effective range. Substituting the above
equations into Eq. (5) gives
2

VN = 1 f c b h10 + 2 A sv f yv

(7)

5.2.2 Combined vertical stirrups and inclined web reinforcement

When vertical stirrups and inclined web reinforcement are combined, besides the shear resisted
partly by the vertical stirrup Vs and partly by the previously described concrete struts Vc, the third
term of Eq. (6) signifies that the shear was sustained by the inclined web reinforcement as
expressed by Eq. (4a)
V b = 3 Asb f y sin

Measurements on the dapped end beam indicated that the shear strength with inclined web

233

Experimental investigation on the shear capacity

reinforcement was higher than the shear strength with vertical stirrups, but was the same as the
shear strength with the longitudinal reinforcing bent at an angle of 45 . By taking = 1.0, Eq. (6)
can be expressed as
o

VN = 1 f c b h10 + 2 A sv f yv + A sb f y sin

(8)

Thus far, it had not been possible to rationally allow for all factors affecting each of the
components of the shear resisting mechanisms and their interaction. By using statistical regression
analysis of the test results of 16 specimens (without B1.21 and B2.32 as shown in Table 4), the two
most important parameters governing shear strength were derived: = 0.0546 and = 0.8583. So,
the following equation to account for the shear strength of dapped ends beam is proposed
1

VN = 0.0546 f c b h10 + 0.8583 A sv f yv + A sb f y sin

(9)

5.3 Demand of longitudinal reinforcement at the nib of the dapped end beam
If only vertical stirrups are provided, by approximately taking Z = 0.85 h , Eq. (2) becomes
e- V
T 1 = 1.18 -----(10)
h10 u
The minimum amount of longitudinal reinforcement in the nib of a dapped end beam can be
determined from the above equation.
If the shear strength of a dapped end beam is controlled by the flexural strength of its nib, the
shear strength can also be determined by Eq. (10) and can be expressed as follows
As 1 f y h10
(11)
VN = -------------------------1.18 e
In which As is the area of longitudinal reinforcement detailed at the nib of a dapped end beam; fy is
the yield stress of longitudinal reinforcement; h is the effective high of nib and e is the shear span
defined in Fig. 8.
1

10

10

6. Comparison of results
The predicted shear strength of the dapped end beams can be determined by smallest value of
shear strength calculated by Eq. (9) and Eq. (11). Table 4 shows the comparison of calculation
results and test results of 18 specimens. The other 6 specimens exist of exception, and thus are
excluded from comparisons. Exceptions are B3.31 and B3.32, having the stirrups with 4 legs, B3.41
and B3.42 with the smallest dapped end height (less than 0.45 h), B1.11 and B3.61 with vertical
stirrups combined with web bar bent along 90 .
It is noted that the values calculated by Eq. (9) are larger than that by Eq. (11) for all specimens
but specimen B1.21 due to adequate amount of longitudinal reinforcement in the nib.
Table 4 shows a good agreement between test and calculation results for most specimens except
that the result of specimen B1.21 showed large scatter between calculation and test result. It should
be noted that the specimen B1.21 was provided with a largest web reinforcement and smallest
o

234

Quanfeng Wang, Zixiong Guo and Pierre C.J. Hoogenboom

concrete strength than that of other specimens, and thus resulted in a compression failure before the
web reinforcement arriving its yield stress.

7. Conclusions
From the tests the following behavior was observed and some design recommendations are given:
(a) 22 test specimens collapsed after arriving at the ultimate load due to loss of their shear
capacity. Exceptions are specimen B1.21 and specimen B2.32 in which collapse resulted from
the compression crushed of the cracked reinforced concrete and splitting failure of concrete,
respectively. The first crack always initiated at the reentrant corner of the dapped end beam at
approximately 40 -60 to the axis of the beam. Obviously, the cracks developed as the load
increased.
(b) The inclined cracks, up to the vicinity of the load point, occurred at the reentrant corner of the
dent. Under increasing load, the diagonal crack propagated towards the loading point and final
failure occurred when the tip of the diagonal crack decreased the uncracked compression
block of concrete to a critical point when a so-called shear-compression failure occurred.
(c) The cross-sectional dimensions of the specimen played an important role in control of the
diagonal cracks, and the effect of the web reinforcement on crack control was not significant,
compared with the size of the cross section;
(d) From the tests, it also can be seen that closed stirrups enhance the shear resisting capacity, but
less than inclined stirrups and bent reinforcement.
(e) The effect of the height of the nib end on shear capacity is significant. It is suggested that the
height of the nib end should be larger than 0.45 h.
(f) The distance of the first stirrup to the end face of the dapped end beam is recommended to be
less than 40 mm and as close to the end face of the dapped end beam as possible. The
effective range of the stirrups is proposed to be ao = 0.5 h .
(g) Some relatively simple semi-empirical equations for shear strength are proposed in this paper
based on the test results of 16 dapped end beams. The results predicted by these equations are
in good agreement with test results presented in this paper. The predicted shear strength by
Eq. (9) can be used for different types of web reinforcement.
(h) The minimum amount of longitudinal reinforcement at the nib of a dapped end beam can be
determined from Eq. (10) to prevent the flexural failure of the nib.
o

Acknowledgements
The work described in this paper was carried out in the Huaqiao University. The assistances of the
staff are gratefully acknowledged.

References
American Concrete Institute (ACI) (1999), Building code requirements for structural concrete and
commentary, ACI 318-99, Detroit.

Experimental investigation on the shear capacity

235

Blaauwendraad, J. and Hoogenboom, P.C.J. (2000), Design tool for structural concrete D-regions, Proc. of the
Int. Symp. on Modern Concrete Composites & Infrastructures (MCCI2000), Northern Jiaotong University ,
Beijing, China, , 3-10.
European Committee for Standardization (CEN) (1992), Design of concrete structures, Part 1: General rules and
rules for buildings, Eurocode 2, Brussels.
Leonhardt, F. (1984), Grunlagen zur Bemessung im Stahlbetonbau (Teil 1, Dritte Auflage), Springer-Verlag, UK.
Mottaock, A.H. and Chan, T.C. (1979), Design and behavior of dapped end beams, Prestressed Concrete
Institute Journal, (6), 28-45.
Mottaock, A.H. (1986), Behavior and design of dapped end members, Proceedings, Seminar of Precast
Concrete Construction in Seismic Zones, Tokyo, 81-100.
Schlaich, J., Schafer, K. and Jnnewein, M. (1987), Toward a consistent design of structural concrete,
Prestressed Concrete Institute Journal, (3), 74-150.
Thomas, T.C. (1998), Unified approach to shear analysis and design, J. Cement and Concrete Composites, ,
419-435.
Wang, Quanfeng, et al. (1996), Study on detailing at the ends of notched concrete beams, Proc. on Studies and
Applications of Joint Connections of Concrete Structures and of Structural Earthquake Resistant Detailing,
115-121, Qingdao, China (in Chinese).
Zararis, P.D. and Papadakis, G.C. (2001), Diagonal shear failure and size effect in RC beams without web
reinforcement, J. Struct. Eng., ASCE, (7), 733-742.
1

24

32

20

127