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

net/publication/322399966

A mechanical model for failures in shear of members without transverse

reinforcement based on development of a critical shear crack

Article  in  Engineering Structures · February 2018

DOI: 10.1016/j.engstruct.2017.12.004

CITATIONS READS

8 307

3 authors:

Francesco Cavagnis Miguel Fernández Ruiz

École Polytechnique Fédérale de Lausanne École Polytechnique Fédérale de Lausanne
5 PUBLICATIONS   55 CITATIONS    111 PUBLICATIONS   1,457 CITATIONS   

SEE PROFILE SEE PROFILE

Aurelio Muttoni
École Polytechnique Fédérale de Lausanne
193 PUBLICATIONS   2,337 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Research grant AGB 2011-015 View project

Projet de recherche AGB 2002/028 Office fédéral des routes (OFROU - Suisse); Bolsa de Doutoramento FCT/BD/13259/2003 - Portugal View project

All content following this page was uploaded by Francesco Cavagnis on 23 November 2018.

The user has requested enhancement of the downloaded file.

 Engineering Structures 157 (2018) 300–315

Contents lists available at ScienceDirect

Engineering Structures
journal homepage: www.elsevier.com/locate/engstruct

A mechanical model for failures in shear of members without transverse T

reinforcement based on development of a critical shear crack
⁎
Francesco Cavagnis , Miguel Fernández Ruiz, Aurelio Muttoni
École Polytechnique Fédérale de Lausanne, ENAC, Station 18, CH-1015 Lausanne, Switzerland

A R T I C L E I N F O A B S T R A C T

Keywords: In this paper, a mechanical model consistent with the main assumptions of the Critical Shear Crack Theory
Concrete structures (CSCT) is proposed for shear design of slender concrete members without shear reinforcement. To that aim, the
Shear strength shear force that can be transferred through the critical shear crack by aggregate interlock, residual tensile
Critical shear crack theory strength and dowelling action as well as due to the inclination of the compression chord are calculated by
Shear-transfer actions
integration of fundamental constitutive laws accounting for the critical shear crack opening and kinematics at
failure. The pertinence of the assumptions is validated through comparisons to detailed test measurements to
assess their validity. The model allows predicting the failure load, the deformation capacity and the location of
the critical shear crack leading to failure. The results are checked against large datasets and the model is finally
used to discuss on the influence of the various parameters on the governing shear-transfer actions. The results are
eventually used to propose improvements on the CSCT failure criterion for shear, relating the shear strength and
its associated deformation capacity.

1. Introduction with lower strengths associated to larger deformation capacities (and

thus larger crack widths).
The shear strength of reinforced concrete beams without transverse A detailed description of the CSCT and of the development of the
reinforcement has been extensively investigated in the past and many critical shear crack can be found elsewhere [8–11]. These investigations
mechanical approaches have been proposed to address this issue [1–7]. have shown that rather different crack patterns may develop for similar
Yet, these models present significant discrepancies on the mechanical reinforced concrete members and that the contribution of each-shear
parameters and shear-transfer actions governing failure and this is still transfer action may significantly vary during loading (being this
a topic of controversy and open discussion. As established since long, strongly dependent on the shape, location and kinematics of the critical
there are various potential shear-carrying actions that allow transfer- shear crack).
ring shear forces in cracked concrete members. They are usually clas- A theoretical description of the main physical parameters governing
sified as beam shear-transfer actions (where the force in tension chord the shear capacity can be found in Fernández Ruiz et al. [8]. This work
varies)—cantilever action (Fig. 1a), residual tensile strength of concrete shows, by means of an analytical approach based on simple constitutive
(Fig. 1b), dowelling action (Fig. 1c) aggregate interlock (Fig. 1d), which laws, that all shear-transfer actions depend eventually on the same
may also be combined (Fig. 1e)—and the arching action (where the mechanical parameters (concrete compressive strength, effective depth,
force in the tension tie remains constant, Fig. 1f). A complete descrip- maximum aggregate size and crack width). These results confirmed the
tion of each of them can be found elsewhere [8]. validity of the failure criterion of the CSCT relating the shear strength
One mechanical approach which considers the contribution of all and the deformation capacity at failure (Fig. 1g) in terms of its shape
potential shear-carrying actions is the Critical Shear Crack Theory and governing parameters. Since that work [8], a number of detailed
(CSCT) proposed by Muttoni et al. [3]. This theory is based on the as- testing programmes using digital image correlation (DIC) have been
sumption that the shear strength of slender members without stirrups is performed [10–12]. These programmes provide very detailed in-
governed by the development of a critical shear crack that limits the formation on the development of the critical shear crack and on the
strength of the theoretical inclined compression strut carrying shear. associated capacity of each shear-transfer action. In this paper, this new
According to this theory, both the shear strength and the deformation experimental data is used in combination with refined constitutive laws
capacity of a member are related through a failure criterion (Fig. 1g), to perform detailed calculations on the contribution of the various

⁎
Corresponding author.
E-mail address: francesco.cavagnis@epfl.ch (F. Cavagnis).

https://doi.org/10.1016/j.engstruct.2017.12.004
Received 28 February 2017; Received in revised form 20 October 2017; Accepted 4 December 2017
Available online 02 January 2018
0141-0296/ © 2017 Elsevier Ltd. All rights reserved.
F. Cavagnis et al. Engineering Structures 157 (2018) 300–315

Fig. 2. Crack kinematics, centre of rotation and horizontal opening u of the cracks trib-
utory to the opening of the critical shear crack in a region of length lB (specimen SC69
[11]).

2. Load-critical shear crack opening relationship

One of the main assumptions of the CSCT, according to Muttoni

et al. [3], is that the opening w of the critical shear crack can be as-
sumed to be proportional to the product of the longitudinal strain in a
control section times the effective depth of the member (w ∝ ɛ·d, where
the strain ɛ can be calculated according to a cracked sectional analysis).
The validity of this assumption has been confirmed by recent experi-
mental investigations [11] on the basis of DIC measurements. Ac-
cording to these measurements, although cracks at narrow spacing may
develop at the level of the longitudinal reinforcement, they merge
thereafter and the sum of all cracks tributary to the critical shear crack
yields an approximately linear profile of the horizontal component of
the crack width (refer to Fig. 2). This tributary length lB has been ex-
perimentally observed to be fairly constant at peak load [11] and can be
approximated by the expression:
lB = d−c (1)

(average of measured-to-calculated values equal to 1.01 with a Coeffi-

cient of Variation of 7% for 11 specimens presented in [11]), where c is
the depth of the compression zone, calculated by assuming a linear
response of concrete in compression and neglecting concrete in tension:

Es ⎛ 2·Ec
c = d·ρ· ⎜ 1 + −1⎞⎟
Ec ⎝ ρ ·Es ⎠ (2)
with Ec taken as Ec = 10′000fc1/3
in MPa.
Thus, considering that the reinforcement strains can be calculated as
(linear profile of compression stresses):
MF MF
εs = =
As ·Es ·z As ·Es ·(d−c /3) (3)

and assuming that the horizontal crack opening at the level of the re-
inforcement is proportional to the product of the tributary length lB
Fig. 1. Shear-transfer actions described with strut and tie models (tensile forces in red
times the strain at the level of the longitudinal reinforcement
and compressive forces in blue): (a) cantilever action; (b) residual tensile strength of
(uA = εs·lB ) , it results:
concrete; (c) dowelling action; (d) aggregate interlock; (e) combined shear-transfer ac-
tions; (f) arching action; (g) failure criterion of the Critical Shear Crack Theory (CSCT) MF d−c
(adapted from Muttoni et al. [3]). (For interpretation of the references to colour in this uA = ·
As ·Es d−c /3 (4)
figure legend, the reader is referred to the web version of this article.)
In simply supported elements subjected to point load, for instance,
MF = VF ·aF and thus Eq. (4) yields:
potential shear-transfer actions during the process of loading and at VF ·aF d−c
failure. As a result, predictions can be made not only on the strength uA = ·
As ·Es d−c /3 (5)
and deformation capacity, but also on the shape and location of the
critical shear crack as well as on the amount of the shear carried by where aF refers the moment-to-shear ratio of the investigated section,
each shear-transfer action. The results are compared to actual tests to corresponding to the distance xF of the section to the support in this
show the consistency of the approach and to discuss on the role and case (the so called shear span, see Fig. 3a).
significance of the various shear-transfer actions. On that basis, im- It has to be noted that Eq. (3) is an approximation as the con-
provements on the CSCT failure criterion will be presented and dis- tribution of the internal forces acting on the critical shear crack (refer to
cussed. Fig. 3d) to the moment MF at a section located at crack tip are neglected

301
F. Cavagnis et al. Engineering Structures 157 (2018) 300–315

investigated in [11] and approximated by the following expression:

π ⎛ α1/3
βAB = ·⎜1 + A ⎞⎟
4 ⎝ 3 ⎠ (6)
MA
where αA is the shear slenderness ratio defined as αA = with MA
VA·d
,
and VA corresponding to the acting bending moment and shear at the
section where the investigated crack intercepts the flexural reinforce-
ment (section xA).
The length lA of the segment A-B of the crack can be calculated
assuming that it propagates up to the neutral axis [10]:
d−c
lA =
sinβAB (7)
With respect to the quasi-horizontal part of the crack, it was ob-
served that its origin is related to the (quasi-vertical) tensile stresses
developing at the tip of the crack, due to the cantilever action between
the flexural cracks (Kani’s tooth model, Fig. 1a, [3,11]) and that it
propagates at a load level which can be significantly lower than the
failure load. Although the length lF and the angle βBF of the segment B-F
were observed to have some level of scatter [3], in the following, lF is
assumed in a simplified manner to be equal to d/6 and βBF equal to π/8
in agreement to experimental measurements [10,11].
The kinematics of such crack is presented in Fig. 3b and c. Ac-
cording to [8] and to the measurements of [11], it can be assumed that
the centre of rotation is approximately located at the tip of the crack.
The rotation of the crack can thus be calculated as a function of the
reinforcement strain:
uA ε ·l
ψ= = s B
dF dF (8)
where uA and ɛs are respectively the horizontal opening of the crack and
the strain at the level of the longitudinal reinforcement, lB is the length
contributing to the opening of the critical crack according to Eq. (1)
(see Fig. 2) and dF = d−c + lF ·sinβBF defines the distance from the tip of
the crack to the flexural reinforcement (Fig. 3b and c). The associated
relative movements are depicted in Fig. 3a. The shape and kinematics
allow defining the opening and sliding at each point of the critical crack
Fig. 3. (a) Kinematics and displacements of the crack lips according to the adopted crack (opening in Fig. 3b and sliding in Fig. 3c). It can be noted that the top
shape; (b) crack opening w perpendicular to the crack surface and (c) sliding δ; (d) part of the crack is characterized by pure opening (mode I), whereas the
equilibrium of the rigid-body and internal forces.
quasi-vertical part by a combined crack opening and sliding (mixed
mode I-II).
(in accordance to the numerical results of detailed calculations ac-
counting for all potential forces in the critical shear crack [11]). 3.2. Residual tensile strength contribution

3. Contribution of the shear-transfer actions Cracked concrete has a residual capacity to transfer tensile stresses
for low crack openings. The softening behaviour can be modelled using
In the following, the contribution of the various shear-transfer ac- several approaches. In this work, the proposal by Reinhardt [15]
tions to the shear resistance will be investigated by integration of characterized by a simple power law equation will be used:
fundamental constitutive laws. This will be done with reference to a c1
w
potential critical shear crack characterized by a given shape and kine- σres = fct ·⎜⎛1−⎛ ⎞ ⎞⎟ ⩾ 0
⎜ ⎟

w
⎝ ⎝ c⎠ ⎠ (9)
matics (crack opening and sliding). The location of the potential critical
shear crack will then be varied in order to find the governing location where c1 = 0.31 and wc = GF / fct ·(1 + c1)/ c1 represents the maximum
leading to the minimum shear strength. crack width for stress transfer. The fracture energy of ordinary normal
weight concrete GF can be calculated according to fib Model Code 2010
3.1. Crack shape and kinematics [16] as GF = 0.073·f c0.18 (N/mm,MPa) and the tensile strength of con-
crete is assumed equal to fct = 0.3·fc2/3 for fc < 50 MPa [16] and
The shape of the critical shear crack for slender members failing in fct = 0.3·(50·fc )1/3 for fc ≥ 50 MPa (approximating [16]). It shall be
shear has been investigated in detail and described by Cavagnis et al. noted that the fib Model Code 2010 does not explicitly account for the
[10]. The critical crack at failure can be approximated by a bi-linear size of the aggregates on the value of the fracture energy. Nevertheless,
shape comprising a quasi-vertical part (segment A-B in Fig. 3a, whose this dependence is acknowledged by the code in its commentary and
inclination is related to the moment-to-shear ratio [10]) and a quasi- also considered in previous versions of the fib Model Code (as that of
horizontal part (segment B-F in Fig. 3a), geometry also assumed by 1990 [17]), implying that the shear transfer capacity of the residual
other researchers [4,6,13,14]. tensile strength depends eventually upon the aggregate size [8].
A trend for the inclination of the quasi-vertical segment A-B with It is important to note that only the top part of the critical shear
respect to the moment-to-shear ratio αA was experimentally crack (segment B-F) is characterized by a response in mode I and it is

302
F. Cavagnis et al. Engineering Structures 157 (2018) 300–315

Fig. 4. (a) Kinematics of mixed mode tests performed by Jacobsen et al. (fc = 41 MPa, dg = 8 mm) [23]; (b–d) comparisons of aggregate interlock stresses according to the model of
Walraven [18] for kinematics analogue to that of Guidotti [22] and the aggregate interlock stresses calculated according to Eqs. (15) and (16) but neglecting the term σres; (e–g)
comparison of mixed mode test results [23] with shear and normal stresses calculated according to Eqs. (15) and (16).

then governed by residual tensile strength of concrete. The quasi-ver- (9)). In addition, Eq. (14) clearly shows the hyperbolic decay of the
tical part (segment A-B) is characterized by a mixed mode response so contribution VRes with increasing crack opening uA (in agreement to
that the residual concrete tensile strength in that part is considered [8]).
together with the aggregate interlock contribution (see following sub-
section). By integration of the stresses along segment B-F, the shear 3.3. Aggregate interlock contribution
force can be determined as:
lF1 Many approaches based on the opening and sliding between the
VRes = ∫0 σres·b·cosβBF ·dη (10) rough surfaces of the crack have been proposed in the literature to
where η is the integration variable, lF1 is the integration limit, b is the calculate the aggregate interlock stresses [18–20]. A consistent ap-
width of the member, σres refers to the residual stress normal to the proach to this issue was developed by Walraven [18], as a two-phase
crack and βBF to the angle of segment B-F. model, whose application has been generalized by Ulaga [21] and
The resulting shear force carried by the residual tensile strength of Guidotti [22] accounting for different kinematical paths. A detail de-
concrete according to Eq. (10) thus results: scription of them can be found in [9,22]. However, the integration of
Walraven’s equations requires numerical procedures and it cannot be
c1
lF1
⎛1−⎛ ψ·η ⎞ ⎞⎟·dη = f ·b·cosβ ·l · solved in a closed-form manner. In order to avoid the use of numerical
VRes = fct ·b·cosβBF ∫0 ⎜ ⎜ ⎟
ct BF F 1
procedures in this work, two analytical equations have been calibrated
⎝ ⎝ wc ⎠ ⎠
1 c by the authors of this paper on the basis of the Walraven’s model for
⎛⎜1− 1 ·⎛ uA ·lF 1 ⎞ ⎞⎟
⎜ ⎟ aggregate interlock, but considering the kinematics of Guidotti which is
⎝ 1 + c1 ⎝ dF ·wc ⎠ ⎠ (11) more representative of the actual case (see Fig. 2). These assumptions
and the associated normal force is: allow calculating the transferred shear stresses (τ) for a given opening
(w) and sliding (δ) as:
NRes = VRes·tanβBF (12)
c3·δ 4/3
The integration of stresses leads to two possible regimes: (i) cases τagg = τagg,0 = fc ·
where the normal stresses develop along the whole length of segment B-
(c2 w )1.8 + c2·δ (15)
F (ψ·lF ⩽ wc ) so that lF1 = lF and (ii) cases where the normal stresses as well as the normal stresses (σ):
develop only close to the tip of the crack (ψ·lF > wc ), with lF1 equal to:
wc w c4·δ 7/3
lF 1 = ·lF = c ·dF σagg = σres−σagg,0 = σres− fc ·
ψ·lF uA (13) (c2 w )3 + c2·δ (16)

It is interesting to note that for the latter case, Eqs. (11) and (13) where σres is defined according to Eq. (9), c2 = 40, c3 = 35 and
give a simple expression: c4 = 400 are constants; δ = δ / ddg and w = w / ddg are the normalized
crack sliding and crack opening and ddg is an average roughness whose
GF ·b·cosβBF ·dF
VRes = value can be calculated as follows:
uA (14)
ddg = min(40 mm,16 + dg ) for fc ⩽ 60 MPa
where the residual strength contribution depends on the fracture energy
of concrete GF and not on the distribution of the tensile stresses (Eq. ddg = min(40 mm,16 + dg ·(60/ fc ) 2) for fc > 60 MPa (17)

303
F. Cavagnis et al. Engineering Structures 157 (2018) 300–315

where dg refers to the maximum aggregate size. Justification of Eq. (17) shear crack (segment A-B) is characterized by a mixed mode I and II
is provided later in this section. behaviour. On the basis of the aggregate interlock laws and the relative
Fig. 4b–d shows comparisons between the aggregate interlock displacements between the lips of the crack, the aggregate interlock
stresses according to the model of Walraven [18] for a kinematics forces at the critical crack can be determined as:
analogue to that of Guidotti [22] (initial opening w0 followed by a
combined opening and sliding, Fig. 4a) and the normal and shear ( l
VAgg = b· ∫l 2 τagg,0·sinβAB dξ − ∫l 2 σagg,0·cosβAB dξ
1 1
l
)
stresses calculated according to Eqs. (15) and (16), but neglecting the l3
+ b· ∫l σres·cosβAB dξ = VAgg,0 + VAgg,res
term σres which is not accounted for in the theoretical model of Wal- 1 (18)
raven. The same comparisons are shown in Fig. 4e–g to aggregate in-
terlock stresses measured by Jacobsen et al. [23] for specimens tested
with the same kinematics, but accounting for the term σres (which is
( l
NAgg = b· ∫l 2 τagg,0·cosβAB dξ + ∫l 2 σagg,0·sinβAB dξ
1 1
l
)
l3
actually present in the tests). It can be noted that the peak value of the − b· ∫l σres·sinβAB dξ = NAgg,0 + NAgg,res (19)
1
calculated shear and normal stresses is in good agreement with the
theoretical and experimental ones, as well as the stiffness and softening where ξ is the integration variable, l1, l2 and l3 are the integration limit,
properties. In addition, the simple superposition of the residual tensile b is the width of the member, τagg,0, σagg,0 and σres are defined in Eqs.
stresses (σres) and aggregate interlock stresses (σagg,0) gives reasonable (15) and (16) and βAB refers to the inclination of segment A-B. Note that
results compared to the test measurements (Fig. 4e–g): it can be ob- the integration of τagg,0 and σagg,0 leads to VAgg,0 and NAgg,0, whereas the
served that the normal stresses vary from tension to compression during integration of σres leads to VAgg,res and NAgg,res.
the combined opening w and sliding δ, but remain in tension for low In a general manner, the resulting shear force that can be trans-
values of δ. ferred by aggregate interlock can be written as7:
It is important to mention that Eqs. (15) and (16) have been cali-
VAgg l2 sinβAB ·c3·δA 4/3 cosβAB ·c4·δA 7/3
brated to be applied to calculate the aggregate interlock stresses for
fc ·b
= ∫l 1.8 + c 2·δA
−
3 + c 2·δA
dξ
mixed mode crack openings, with initial openings w0 and secant mixed
mode angle γ (variable along the quasi-vertical segment of the crack,
1
( ξ
c2·ψ· d
dg ) ( c2·ψ· d
ξ
dg )
c
according to the kinematics of Fig. 3b and c) which are typical for l3 ψ·ξ ⎞ 1 ⎞
+ ∫l cosβAB ·⎜⎛1−⎛ ⎜⎟ dξ ⎟

critical shear cracks of slender members (γ > 45°, [10]). With respect 1
⎝ ⎝ wc ⎠ ⎠
to the average roughness (ddg), this term accounts for two issues:
c3·δA 4/3 l 20.8 + c2 δA −l10.8 + c2 δA
= sinβAB · ·
1.8 + c 2 δA (l2·l1)0.8 + c2 δA
– The first is that concrete cracks, present an undulated (rough) sur-
face (Fig. 5b and d), contrary to Walraven’s approach which as-
(c2 δA + 0.8)· ( uA c 2
·
dF ddg )
sumes cracks as perfect planar surfaces with protruding aggregates c4·δA 7/3 l 22 + c2 δA −l12 + c2 δA
−cosβAB · ·
(Fig. 5a and c). This roughness, ensuring the transfer of shear forces 3 + c 2 δA (l2·l1)2 + c2 δA
by interlocking, is referred to as meso-roughness [8]. The value of
(c2 δA + 2)· ( uA c 2
·
dF ddg )
the meso-roughness depends on the surface properties after crack 1 1 c c
1 ⎛ uA ·l3 ⎞ ⎞ 1 ⎛ uA ·l1 ⎞ ⎞ ⎤
development, but can be assumed as 16 mm [3] for normal cases + cosβAB ·⎡ ⎛ ⎛
⎢l3·⎜1− 1 + c1 · dF ·wc ⎟−l1·⎜1− 1 + c1 · dF ·wc ⎟ ⎥·
⎜ ⎟ ⎜ ⎟

(refer to Eq. (17)). ⎣ ⎝ ⎝ ⎠ ⎠ ⎝ ⎝ ⎠ ⎠⎦

– The second is that, as observed by Sherwood et al. [24], the increase fct
of the interlock capacity is limited for large aggregates. In addition, fc (20)
for high-strength concrete (fc > 60 MPa), a reduction of the ag-
2 + v 2 ·sinθ
uA
δA A uA
gregate size shall be considered [3], since cracks develop through where δA = = A
, vA = ·(lF ·cosβBF + lA·cosβAB ) and
ddg ddg dF
the aggregates, resulting in crack surfaces that are relatively smooth d
θA = βAB−atan l ·cosβ +F l ·cosβ
(refer to Fig. 3a).
[25]. F BF A AB
The integration limits l1 and l2 can be calculated on the basis of the
As shown in Fig. 3a, only the quasi-vertical branch of the critical geometry of the crack (refer to Fig. 3a): l1 = lF ·cos(βAB−βBF ) and
l2 = l1 + lA . With respect to l3, the integration of aggregate interlock
stresses leads to three potential regimes:
(i) cases where the residual stresses σres develop through the whole
length of segment A-B (wA < wc corresponding to low crack openings),
hence l3 is equal to l2; (ii) cases where the residual stresses develop only
on the top region of the segment A-B (wB < wc < wA) so that l3 results:
wc
l3 = ·l2
wA (21)

and (iii) cases where no residual tensile stresses develop through the
quasi-vertical branch of the crack since the opening of the crack along
the whole segment A-B exceeds wc (corresponding to large crack
openings), hence l3 is equal to l1 and VAgg,res and NAgg,res are equal to
zero.

3.4. Dowelling action

Dowelling forces can be activated due to relative vertical displace-

ment between the crack surfaces at the level of the longitudinal re-
Fig. 5. (a) Idealized crack plane according to the model of Walraven [18] and (b) actual inforcement [9]. The capacity of dowelling action to transfer shear is
crack plane; crack surface near the aggregates: (c) idealized, according to the Walraven governed by the effective area of the concrete in tension near the bars
[18] and (d) actual.
and by its effective tensile strength [26]:

304
F. Cavagnis et al. Engineering Structures 157 (2018) 300–315

VDowel = n·fct ,ef ·bef ·lef (22)

where n is the number of bars activated, fct,ef is the effective tensile

strength, bef and lef are the effective width and length in which the
tensile strength develops (Fig. 6a and b). The effective length is esti-
mated as lef = 2db [26] (where db is the diameter of the reinforcing bars,
Fig. 6a) and the effective width bef (Fig. 6b) can be calculated as [26]:
bef = min[b/ n−db,4·cb] (23)

where b is the width of the member and cb is the concrete cover.

With respect to the effective tensile strength of concrete (fct,ef), it
should be noted that its value is strongly influenced by the state of
strains in the flexural reinforcement (due to its interaction with bond
[26]). This effect can be estimated as [27]:
fct ,ef = kb·fct (24)

where kb is a strength reduction factor and it follows a decay for in-

creasing strains in the flexural reinforcement (Fig. 6c). When the re-
inforcement is not strained, kb can be assumed equal to 1. In the fol-
lowing, the reduction factor for increasing value of strains in the
longitudinal reinforcement is calibrated by fitting the experimental
data presented in [27] as (refer to Fig. 6c):

kb = 0.063·εs−1/4 ⩽ 1 (25)

By replacing the steel strain ɛs with the value εs = uA / lB , the coeffi-

Fig. 6. (a) Development of transverse stresses at the cover region along the bar and de- cient kb becomes:
finition of effective length lef; (b) distribution of transverse tensile stresses (perpendicular
to the bar) and definition of effective width bef (adapted from [26]; (c) reduction of the 1/4 1/4
l d−c ⎞
effective tensile strength as a function of the longitudinal strains in the bar (experimental kb = 0.063·⎛ B ⎞
⎜ ⎟ = 0.063·⎛
⎜ ⎟ ⩽1
data from [27]). ⎝ uA ⎠ ⎝ uA ⎠ (26)

The shear resistance due to dowelling action of one-ways slabs with

one layer of reinforcement bars is thus:

Fig. 7. Crack pattern and principal compressive strains at failure load for (a) a slender beam (specimen SC70) and (b) a squat beam (specimen SC67) (adapted from Cavagnis et al. [11]);
(c) hyphothesis of a stress field used to determine the contribution to the shear strength of the compression zone; (d) calculated contribution of the compression zone with respect to the
calculated shear capacity as a function of the distance between the tip of the crack and the axis where the load is applied (tests included in the database by Reineck et al. [31], assuming
xA = 0.5a).

305
F. Cavagnis et al. Engineering Structures 157 (2018) 300–315

VDowel = kb·fct ·n·(b/ n−db)·2·db (27) relatively low for slender beams compared to other shear-transfer ac-
tions, a simplified expression can be used despite the non-negligible
For comparison with test data, when no information is available on
scatter:
the bar spacing or for multiple layers, a value of 3db will be assumed as
bar spacing. With this assumption, the last part of Eq. (27) becomes VCompr kc ·hF
= <1
4n·db2 which can be expressed as a function of the reinforcement ratio Vc rF (31)
(16ρ ·b·d/ π ) so that the shear strength due to the dowel contribution can
where Vc is the total shear capacity
be approximated by:
(Vc = VRes + VAgg + VDowel + VCompr) and kc is a constant obtained by
VDowel ≈ 5·kb·fct ·ρ ·b·d (28) fitting of the calculated contribution of the compression zone and that
can be assumed equal to 0.5. It is important to mention that when a
direct strut can develop, this approach is no longer valid since the
3.5. Contribution of the compression zone theoretical strut carries almost the total shear force (refer to Fig. 7b).
This is for instance the case of members with low shear-to-span ratios
Shear can also be transferred by means of the inclination of the (a/d < 2.5), with high pre-stressing forces (associated to low effective
compression chord [8]. For slender beams, this action is significant slenderness [28]), or with no or limited bond strength [3].
mostly before the propagation of the segment B-F of the critical crack The shear carried by the inclination of the compression chord yields
within the compression zone and then it decreases progressively as the thus from Eq. (31):
inclination of the compression chord is rather flat (Figs. 1e and 7a)
kc ·hF / rF
[11]. For short-span beams (associated to low slenderness), with cracks VCompr = ·(VRes + VAgg + VDowel )
1−kc ·hF / rF (32)
developing without disturbing the theoretical direct strut (typically
cracks whose tip is located close to the acting load), the arching action It can be noted that according to this approach, the contribution of
can develop almost undisturbed and the shear strength is mostly con- the compression zone is determined by the geometry and the location of
trolled by this action (refer to Figs. 1f and 7b). the critical shear crack and it is governed by the same mechanical
A realistic assumption for the inclination of the compression zone is parameters as the other shear-transfer actions.
derived based on the detailed observations of the principal strain di-
rections shown in Fig. 7a and b and in [11] (assuming that the in- 4. Evaluation of the shear capacity
clination of principal strains and stresses is parallel). According to these
measurements, it can be assumed that the resultant of the forces of the The total shear strength Vc can be calculated by summing the con-
compression zone at the section corresponding to point F acts at a tribution of the various shear-transfer actions:
distance cn = 1/3 hF from the top compressive fibre, where hF is the VRes + VAgg + VDowel
thickness of the compression zone above the tip of the crack (hF = d - Vc = VRes + VAgg + VDowel + VCompr =
1−kc ·hF / rF (33)
dF). In addition, at the edge of the loading plate, a stress block of
thickness 2 cm with compressive stresses equal to the full compressive The three components VRes, VAgg and VDowel contained in Eq. (33) are
strength fc can be assumed (refer to Fig. 7c). This can be presumed to be strain-dependent, which means that they can be expressed as a function
a lower-bound of the actual contribution, as the tensile stresses per- of the crack opening uA (see Eqs. (11), (20) and (22)). The crack
pendicular to the strut are neglected. The horizontal and vertical opening uA can be calculated using Eq. (4) which means that an itera-
component acting in the compression zone can thus be calculated tion is required (as the crack opening depends on the shear force V).
iteratively by moment equilibrium of the rigid body with respect to Calculating the shear strength requires thus assuming a location of the
point of contraflexure (point P in Fig. 3d). The iterative procedure in- critical crack and then iteratively increasing the crack opening and
volves assuming an initial distance cm between the top compressive checking that the acting shear force V is equal to the shear capacity Vc.
fibre and the centre of the theoretical strut at the edge of the loading The solution V = Vc is the intersection between the load-deformation
plate, which allows defining the inclination αc of the compression zone: relationship of Eq. (4) and the failure criterion of Eq. (33) as shown in
cn−cm Fig. 8a. The governing location of the critical shear crack can therefore
tanα c = be calculated as the one leading to the minimum strength of all po-
rF ,edge (29)
tential locations.
where rF,edge is the distance between the tip of the crack and the edge of The iterative procedure for calculation of the strength can be sum-
the loading plate (Fig. 7c). marized in the following steps:
Varying the value of cm, the iterative procedure ends when σ reaches
the compressive strength of concrete fc and the vertical component (1) Choose a location of the critical shear crack xA.
results: (2) Calculate the angle βAB as a function of αA = MA/(VA·d) according
cn−cm c −c to Eq. (6) (refer to Fig. 3a).
VCompr = NCompr · = fc ·b·2·cm· n m (3) Assume an initial crack opening uAi (as a first step uA0 = 0.001 mm
rF ,edge rF ,edge (30)
can be assumed).
Since the location of the critical shear crack is unknown, the itera- (4) Calculate, as a function of the shape of the crack and its kinematics,
tive procedure that allows calculating the contribution of the com- the residual tensile strength force (VRes, Eq. (11) in Section 3.2), the
pression zone shall be performed for any potential shear crack in the aggregate interlock force (VAgg, Eq. (20) in Section 3.3), the dow-
span of the member. It can be noted that an increase of the moment-to- elling action (VDowel, Eq. (22) in Section 3.4), the contribution of the
shear ratio of the critical shear crack leads to an increase of the normal compression zone (VCompr, Eq. (32) in Section 3.5) and the shear
force in the compression zone and thus to an increase also of the con- capacity as the sum of these contributions (Vc, Eq. (33)).
tribution of the compression chord [7]. In Fig. 7d, the contribution of (5) Calculate the crack opening uA as a function of the acting bending
the compression zone is calculated for 629 slender beams included in moment at the section corresponding to the tip of the crack (MF),
the database by Reineck et al. [31], assuming xA = 0.5a for all tests. It (refer to Eq. (4) in Section 2), where MF is proportional to the shear
can be observed that the percentage of the total shear carried by the force (V).
compression zone depends significantly on the ratio rF/hF, where rF is (6) Iterate the crack opening uAi at step 3 and repeat from step 3 to 6,
the distance between the tip of the crack and the axis where the load is until uA is equal to uAi.
applied. As the contribution of the compression zone shown in Fig. 7d is

306
F. Cavagnis et al. Engineering Structures 157 (2018) 300–315

Fig. 8. (a) Failure criterion and load-deformation

curve; shear-transfer actions: aggregate interlock,
residual tensile strength of concrete, dowelling
action and contribution of the compression zone;
(b) normal and (c) shear stresses developing along
the critical shear crack for three different values
of εs d/ ddg (0.01, 0.03 and 0.05).

One of the main advantages of this approach is that it can be applied calculated according to Eq. (6) and correspond well with the observed
to general or more complex cases and to different loading conditions. cracks (Fig. 9a). From Fig. 9b–d it can be noted that the strains at the
This can be done by accounting for the influence of the main governing level of the longitudinal reinforcement and the contribution of the
mechanical parameters (moment-to-shear ratio aF, crack opening, re- different shear-transfer actions vary as a function of the location of the
inforcement ratio, aggregate size, compressive strength) and evaluating critical shear crack. It can be observed that the contribution of ag-
the contribution of the different load-carrying actions at the peak load. gregate interlock is dominant in this case for all investigated positions
Fig. 8a shows an instance of the contribution of the different shear- of the critical shear crack. Moreover, when the crack is located close to
transfer actions to the shear strength as a function of the strains at the support (in a general case, the point of contraflexure with zero bending
level of the longitudinal reinforcement. The example refers to a simply moment), the contribution of the compression zone is very limited and
supported beam without transverse reinforcement subjected to point the tensile strength of concrete plays a role, whereas when the critical
loading. The failure criterion in Fig. 8a is expressed as a function of the crack develops closer to the load introduction plate (section of max-
parameter εs·d/ ddg , where the product εs·d is assumed to be proportional imum bending moment), the contribution of the compression zone in-
to the crack opening. Such normalization is consistent to that used by creases and the shear carried by the tensile strength of concrete de-
the CSCT [3] (Fig. 1g) and allows for a direct comparison. It can be creases. In a general manner, the governing theoretical position of the
observed that in this example the aggregate interlock is the governing critical crack is defined as the location where the sum of the con-
shear-transfer action and that the contribution of the concrete in ten- tributions of the different shear-transfer actions reaches its minimum
sion is significant for low openings of the critical shear crack. On the value. In Fig. 9b–d, it can be observed that the shear capacity (sum of
contrary, the contribution of the compression zone and the longitudinal all shear-transfer actions) does not significantly vary between the in-
bars (dowelling action) to the shear strength are very limited in this vestigated sections xA and that the location of the critical crack has thus
case. Moreover, it can be noted that the strength of each shear-transfer a limited influence on the shear strength of the member. This explains
action decreases for increasing opening of the critical crack and that why for this type of members, the experimentally observed position of
their decay follows a similar trend (in agreement to [8]). The reason for the failure crack can present a large scatter and different shear-transfer
this strength decay with increasing crack opening uA is illustrated in actions may eventually be governing [8,11].
Fig. 8b and c, where the residual tensile stresses (σres) and the aggregate The governing location of the critical crack is also investigated in
interlock stresses (σagg and τagg) are shown for three different values of Fig. 10 for different values of slenderness ratio (a/d), reinforcement
the opening of the shear crack (εs·d/ ddg = 0.01, 0.03 and 0.05). It can be ratio (ρ) and effective depth (d). In Fig. 10a, the contribution of the
noted that the normal stress σagg is in tension in a part of segment A-B different shear-transfer actions to the shear capacity is shown for each
for low openings of the critical shear crack (εs·d/ ddg = 0.01), whereas for position xA of the critical shear crack. It can be observed that for cracks
larger openings it is always in compression (consistently to experi- developing within a region between 0.4a and 0.6a, the total shear ca-
mental observations of other researchers [23,29]). With respect to the pacity is almost constant but the amount of shear carried by each shear-
shear stress τagg, it reaches a maximum value of 5 MPa in the upper part transfer actions shows some level of variation. In Fig. 10b and c it can
of the segment A-B and it decreases for increasing openings of the cri- be observed that an increase of the slenderness ratio leads to a variation
tical shear crack. of the governing position of the critical crack, with the governing crack
closer to mid-span for less slender members. In Fig. 10d it can be noted
that the distance rF between the tip of the critical shear crack and the
5. Discussion on the significance of shear-transfer actions and axis where the load is applied varies between d and 2d. A similar trend,
their dependence on the critical shear crack development for concerning the location of the governing critical shear crack has been
members subjected to point load experimentally observed in the tests performed by Leonhardt and
Walther [30]. The influence of the reinforcement ratio is also illustrated
In Fig. 9 the location of the critical crack and the associated capa- in Fig. 10e. An increase in the reinforcement ratio leads to an increase
cities of the shear-transfer actions are investigated with reference to an of the height of the compression zone, which therefore plays a more
actual test (specimen SC61, presented in detail in [10]). The specimen significant contribution in carrying the shear. Consequently, for in-
corresponds to a simply supported beam tested under concentrated load creasing values of the reinforcement ratio, the governing location of the
(a/d = 4.41, d = 556 mm, ρ = 0.89%, fc = 35.3 MPa, dg = 16 mm). critical crack shifts towards mid-span. It can be shown that almost all
Three different potential positions of the critical shear crack (xA = d, other parameters have low influence on the governing position of the
xA = 0.5a and xA = a − d) are investigated. Their inclinations βAB are

307
F. Cavagnis et al. Engineering Structures 157 (2018) 300–315

Fig. 9. (a) Specimen SC61 [10]: crack pattern

(cracks in black for increasing load until max-
imum load, cracks in red after failure) and se-
lected potential locations of the critical shear
crack xA; (b–c–d) contribution of the various
shear-transfer actions at the different locations of
the critical shear crack (xA = d, xA = 0.5a and
xA = a – d), intersection of the load-deformation
relationship with the failure criterion and com-
parison with the experimental result.

critical shear crack (Fig. 10f). completed with the tests by Cavagnis et al. [11]) is plotted against the
An interesting fact to be noted is that the curves representing the normalized crack width parameters. The plot is normalized to account
shear capacity are very flat around the minimum (at a distance rA be- for the effective depth, the width, the compressive strength and the
tween 1.5d and 2.25d from the load introduction plate and at a value of aggregate size of the member [8]. The black points in Fig. 11 represent
xA between 0.4a and 0.6a). For these cases, adopting a fixed control the intersection between the failure criteria calculated according to Eq.
section within this region is even sufficient for calculating the shear (33) at the control section xA = 0.5a and the load-deformation re-
strength (although the relative significance of the shear-transfer actions lationship (refer to Eq. (5)). In addition, the direction of the failure
may not be accurately assessed). To that aim, for instance, a value envelopes in the vicinity of the intersection is plotted (grey lines in
xA = 0.5a in agreement with the considerations of Reineck et al. [31] is Fig. 11). It is interesting to note that failures occur actually in a narrow
reasonable. band despite the large range of cases considered (effective depth d
In Fig. 11, the calculated shear capacity of 635 rectangular concrete ranging 50–2000 mm, flexural reinforcement ratio ρ ranging 0.4–7%,
beams without shear reinforcement (data from Reineck et al. [31] concrete strength fc ranging 10–110 MPa, aggregate size dg ranging

Fig. 10. Normalized shear strength as a function of the location of the critical crack (points in black refer to the location with the minimum shear strength): (a) contribution of the
different shear-transfer actions at different critical crack locations xA; (b–d) influence on the location of the critical shear crack of the slenderness ratio a/d, (e) of the longitudinal
reinforcement ratio ρ and (f) of the effective depth d.

308
F. Cavagnis et al. Engineering Structures 157 (2018) 300–315

where vc is the shear capacity per unit length, k is a constant that can be
obtained by fitting of the calculated shear strengths (k = 0.019 in
Fig. 11) and vc0 refers to the maximum shear strength per unit length
(not investigated in this study).

6. Validation of the approach with test results and improvement

of the CSCT

Fig. 12 shows the comparison between the shear strengths calcu-

lated according to the general procedure (Eq. (5) + Eq. (33)), the
simplified approach (Eq. (3) + Eq. (34)) and the original procedure of
the CSCT [3] with some selected experimental test series. The main
parameters governing the shear strength are investigated: the shear-to-
span ratio a/d, the longitudinal reinforcement ratio ρ, the effective
depth d, the compressive strength fc, the aggregate size dg and the
elastic modulus of the longitudinal bars Es. The comparison shows that
the calculated shear strengths are in very good agreement when com-
Fig. 11. Calculated shear strengths of 635 rectangular beams failing in shear (a/d ≥ 2.5)
pared with experimental results and the three approaches follow similar
(data from Reineck et al. [31] completed with the tests by Cavagnis et al. [11]) as a
function of the reinforcement strains at the critical section (xA = 0.5a).
trends.
Fig. 13 presents a systematic comparison of the shear strengths
calculated solving the set of Eqs. (5) and (33) against 635 tests on
0–32 mm, shear span a ranging 2.5d – 8.5d). simply supported beams or cantilevers subjected to point loading (see
For design purposes, instead of calculating the failure criteria by also Appendix B). The database used is that of Reineck et al. [31]
integration of the stresses along the critical shear crack, Muttoni et al. completed with the tests by Cavagnis et al. [11], where only rectangular
[3] proposed a simple hyperbolic failure criterion (Fig. 1g) that ap- beams with a/d ≥ 2.5 have been considered. It can be observed that
proximates the failure band. Although the CSCT failure criterion pro- there are no clear trends for the main mechanical and geometrical
vides a reasonable estimate of the shear capacity, Fig. 11 shows that its parameters. The average measured-to-calculated shear strength is 1.01
accuracy can be enhanced for low values of the term εs·d/ ddg with a and the Coefficient of Variation is 13.6% (Appendix B and Table 1).
power low expression (red line in Fig. 11) where the exponent of the The accuracy of the model is comparable to that obtained using the
term εs·d/ ddg is −1/2: original formulation of the CSCT [3] (average measured-to-calculated
shear strength 1.02, CoV 15.6%, refer to Table 1). Based on the ob-
k ·d· fc servations of the present study and recent experimental investigations
Vc
vc = = ⩽ vc0 [24], the CSCT [3] can be improved accounting for ddg defined ac-
b εs·d/ ddg (34) cording to Eq. (17) (the ratio between measured-to-calculated shear

Fig. 12. Comparison of the refined proposed approach (black lines). the simplified approach (light blue lines) and the CSCT [3] (red lines) to test series investigating: (a) the slenderness
ratio a/d [11], (b) the reinforcement ratio ρ [32], (c) the effective depth d [33], (d) the compressive strength fc [34], (e) the aggregate size dg [35] and (f) the reinforcement elastic
modulus Es (steel and non-metallic reinforcements) [36]. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

309
F. Cavagnis et al. Engineering Structures 157 (2018) 300–315

Fig. 13. Ratio Vexp/Vcalc (Vcalc calculated ac-

cording to the presented procedure: Eqs. (5) and
(33), assuming xA = 0.5a) as a function of the
different mechanical and geometrical parameters
for the cases of slender simply supported beams
and cantilevers subjected to concentrated load
(data from Reineck et al. [31] completed with the
tests by Cavagnis et al. [11]): (a) shear span-to-
effective depth ratio a/d; (b) longitudinal re-
inforcement ratio ρ; (c) effective depth d and (d)
compressive strength fc.

strength results equal to 1.01 with a CoV of 14.5%, refer to Table 1). (opening and sliding) and the crack roughness properties.
Moreover, as shown in Section 5, the accuracy of the hyperbolic (2) The cantilever action (as described in the Kani’s tooth model) is
failure criterion of the CSCT can be enhanced assuming a power law governing for load levels which are generally lower than the actual
failure criterion (refer to Fig. 11). The shear capacity can thus be cal- failure load. This shear-carrying action is disabled by the develop-
culated in a simple manner by combining the load-deformation re- ment of a horizontal branch of the critical shear crack generated by
lationship of Eq. (3) and the simplified power law failure criterion of the tensile stresses near the crack tip. The load can further be in-
Eq. (34), assuming the control section at xA = 0.5a. For this simplified creased above the capacity of the cantilever action as other shear-
approach, a good agreement is also found between the measured shear carrying actions develop.
strength in the tests and the calculated one, with an average ratio of (3) For slender members, the failure load is eventually governed by
1.03 and with a value of CoV of 14.2% (Table 1). beam shear-transfer actions (aggregate interlock, residual tensile
strength, dowelling action and inclination of the compression
7. Conclusions chord):
– For low crack widths in slender members, both aggregate inter-
This paper presents an overview of the shear-transfer actions and lock and tensile strength of concrete play an important role. For
their contribution to the shear strength of slender reinforced concrete large crack widths, aggregate interlock becomes more dominant.
members without transverse reinforcement. The contribution of the – The dowelling action of the longitudinal reinforcement exhibits a
various shear-transfer actions is quantified on the basis of fundamental more limited contribution than the other beam shear-carrying
constitutive laws and accounting for a realistic crack shape and kine- actions. Yet, its contribution is not necessarily negligible.
matics (based on detailed measurements on tests). The main conclu- – The contribution of the inclined compression chord for slender
sions are listed below: members depends significantly on the location of the tip of the
critical shear crack. For critical shear cracks located at a certain
(1) Shear can be carried by a number of potential shear transfer ac- distance from the load introduction, its contribution is rather low.
tions. In a general manner, there is not a governing shear transfer
action, and the amount of shear force carried by each action de- For squat members, arching action is governing.
pends on the location of the critical shear crack, its kinematics

Table 1
Comparison of calculated and experimental shear strengths: average value Vexp/Vcalc and coefficient of variation (CoV).

Vexp/Vcalc [−] CoV [−]

2008
CSCT [3], xcr = a-d/2 1.02 0.156
Proposed approach Eq. (5) + Eq. (33), assuming xA =0.5a 1.01 0.136
Improvement of the CSCT2008 [3] assuming ddg according to Eq. (17) 1.01 0.145
Simplified power law failure criterion Eq. (3) + Eq. (34), assuming xA = 0.5a 1.03 0.142

310
F. Cavagnis et al. Engineering Structures 157 (2018) 300–315

(4) Despite the fact that the relative amount of each shear-transfer agreement to the previous conclusions and allow describing shear
action may differ depending on the location and shape of the cri- failures in a general manner. The shape of its failure criterion can be
tical shear crack, the total shear capacity (sum of the various con- derived by analytical and numerical considerations, allowing re-
tributions) is relatively constant independently of the location of lating the shear strength to the deformation capacity at failure, and
the critical shear crack. This explains why the location of the failure it can be enhanced for low values of crack opening with a power
crack may vary significantly even for similar specimens. law equation.
(5) The force transferred by the different shear-transfer actions decay
for increasing openings of the critical shear crack and they are Acknowledgement
governed by the same mechanical parameters. Accounting for this
fact and for the relatively constant sum of the various shear-transfer The authors would like to gratefully acknowledge the support and
actions, shear failures can be described by a single failure criterion. funding of the Swiss Federal Road Authority, through the project AGB-
(6) The main assumptions of the Critical Shear Crack Theory are in 2011-0.15.

Appendix A: Notation

a shear span (defined for specimens subjected to concentrated loads as the distance between the centre of the load and the centre
of the support)
aA M/V at section A
aF M/V at section F
b width of the beam
bef effective width of concrete in tension
c depth of the compression zone
c1, c2, c3, c4 constants of the model
cb concrete cover
cm distance from the top compression fibre to the centre of the inclined compression strut at the edge of the loading plate
cn distance from the top compression fibre to the axis where inclined force of the compression zone acts
d effective flexural depth
db diameter of reinforcing bar
ddg dg+16 [mm]
dF vertical position of the tip of the crack
dg maximum aggregate size
fc concrete compressive strength measured in cylinder
fct concrete tensile strength
fct,ef effective tensile strength of concrete
h beam height
hF distance from the top compression fibre to the tip of the shear crack
k coefficient power-law failure criterion
kb reduction factor tensile strength of concrete
kc coefficient compression zone
l span length
l1 l2 l3 lF1 integration limits
lef effective length
lA length of segment A-B of the critical shear crack
lB length of the region of the beam contributing to the opening of the critical crack
lF length of segment B-F of the critical shear crack
n number of longitudinal bars
rA horizontal distance from the axis of the load introduction to the onset of the critical shear crack
rF horizontal distance from the axis of the load introduction to the tip of the critical shear crack
rF,edge horizontal distance from the edge of the loading plate to the tip of the critical shear crack
u opening of the crack measured along the horizontal direction
uA horizontal opening of the critical shear crack at point A
v opening of the crack measured along the vertical direction
vA vertical opening of the critical shear crack at point A
vc shear capacity per unit length
vc0 maximum shear strength per unit length
w crack width perpendicular to the crack surface
w normalized crack opening w/ddg
wA opening perpendicular to the crack surface at point A
wB opening perpendicular to the crack surface at point B along segment A-B
wc maximum crack width allowing tensile stresses to develop in cracked concrete
xA horizontal distance from the support to point A
xF horizontal distance from the support to point F
z inner level arm
As area of longitudinal bars
Ec modulus of elasticity of concrete
Es modulus of elasticity of steel

311
F. Cavagnis et al. Engineering Structures 157 (2018) 300–315

GF fracture energy
M bending moment
MF bending moment at the section corresponding to the tip of the critical shear crack
NAgg horizontal component of the aggregate interlock action
NCompr horizontal component of the inclined compression chord
NRes horizontal component of the residual tensile strength of concrete
V acting shear force
Vc shear capacity
Vmax shear force at failure (maximum value)
VAgg shear force carried by aggregate interlock action
VCompr shear force carried by inclined compression chord
VDowel shear force carried by dowelling action
VRes shear force carried by residual tensile strength of concrete
αc inclination of the compression chord
αA MA/(VA·d) in section A
βAB angle of quasi-vertical segment of the critical shear crack (segment AB)
βBF angle of quasi-horizontal segment of the critical shear crack (segment BF)
γ secant mixed mode angle
γT tangent mixed mode angle
δ crack sliding
δA crack sliding at point A
δ normalized crack sliding δ/ddg
ɛs steel strain
η integration variable for the residual tensile stresses
ξ integration variable for the aggregate interlock stresses
ρ reinforcement ratio of tension reinforcement
σagg aggregate interlock normal stress
σres residual tensile stress
τagg aggregate interlock shear stress
ψ rotation of the critical shear crack

Appendix B

Test series considered in this study and comparison with the proposed expressions: Eq. (5) + Eq. (33).

Researchers No. of specimens fc [MPa] b [mm] d [mm] a/d [−] ρ [%] Vexp/Vcalc (COV)

Adebar et al. [37] 5 49.3–58.9 290–360 178–278 2.92–4.56 0.99–3.04 0.84 (0.068)
Ahmad et al. [38] 17 63.4–68.7 127 184–208 2.70–4.00 1.77–6.64 1.03 (0.210)
Ahmad et al. [39] 3 43.6–80.8 102 178 3.70 1.40 0.95 (0.070)
Angelakos et al. [34] 7 21–80 300 895–925 2.88–2.97 0.50–2.09 0.96 (0.146)
Aster et al. [40] 5 24.6–30.4 1000 250–750 3.65–3.68 0.42–0.91 1.01 (0.072)
Lubell et al. [41] 11 37.1–64.6 250–3005 306–916 2.87–3.27 0.76–0.93 1.12 (0.089)
Bernander [42] 6 27.6–29.1 100 168 4.17 0.97–1.17 1.07 (0.062)
Bhal [33] 8 23.2–29.6 240 300–1200 2.94 0.63–1.26 1.03 (0.114)
Bresler et al. [43] 3 22.6–37.6 305–310 461–466 3.80–6.77 1.81–2.73 1.24 (0.046)
Cladera et al. [44] 6 49.9–87 200 360 2.90 2.23 1.00 (0.135)
Cao [45] 2 27.5–30.1 300 1845–1925 2.77–2.89 0.36–1.52 1.23 (0.042)
Cederwall et al. [46] 1 29.3 135 234 3.42 1.07 1.24 (−)
Chana [47] 23 20.8–38.9 60–203 106–356 3.00 1.74–1.78 1.02 (0.073)
Chang et al. [48] 15 17.7–38.6 102 137 2.60–4.09 1.86–2.89 0.91 (0.115)
Collins et al. [25] 7 36–98 300 925 2.88 1.01 1.07 (0.156)
Diaz de Cossio et al. [49] 5 19.5–31.5 152 254 3.30–5.30 0.98–3.33 0.99 (0.061)
Thorenfeldt et al. [50] 14 54–97.7 150–300 207–442 3.00–4.00 1.82–3.24 1.15 (0.119)
Elzanaty et al. [51] 11 20.7–79.3 177.8 267–273 4.00–6.00 0.93–3.21 0.88 (0.060)
Feldman et al. [52] 8 21.5–36.7 152 252 2.87–6.04 3.35 1.11 (0.141)
Ferguson [53] 1 29.3 101 189 3.23 2.08 0.82 (−)
Ghannoum [54] 22 34.2–58.6 400 65–889 2.50 1.15–2.00 0.98 (0.110)
Grimm [55] 11 90.1–110.9 300 146–746 3.53–3.90 0.83–4.22 1.00 (0.130)
Hallgreen [56] 23 31.1–92.4 150–337 191–211 2.61–3.66 0.57–4.11 1.21 (0.084)
Hamadi [57] 4 21.0–30.4 100 370–372 3.37–5.90 1.08–1.79 1.17 (0.082)
Hanson [58] 4 20.9–31.0 152 267 4.95 1.25–3.50 0.97 (0.073)
Hedman et al. [59] 4 19.9–29.5 152 267 4.95 1.25–2.22 1.03 (0.051)
Islam et al. [60] 5 26.6–83.3 150 203–207 3.86–3.94 2.03–3.22 1.02 (0.105)
Johnson et al. [61] 1 55.9 305 539 3.10 2.49 0.82 (−)

312
F. Cavagnis et al. Engineering Structures 157 (2018) 300–315

Kani [62] 37 24.8–30.8 150–612 132–1097 2.50–8.03 2.59–2.87 1.02 (0.106)

Kani et al. [32] 52 15.4–36.7 149–157 264–287 2.50–5.97 0.49–2.83 0.95 (0.108)
Kawano et al. [63] 8 20.6–27.3 105–600 300–2000 3.00 1.18–1.37 1.06 (0.099)
Kim et al. [64] 18 52 170–300 142–915 3.00–6.00 1.01–4.68 0.94 (0.087)
Krefeld et al. [65] 50 12.9–39.0 152–254 238–483 2.67–8.10 0.80–4.92 0.95 (0.092)
Küng [66] 7 18.4–21.7 140 200 2.50 0.36–1.82 1.20 (0.154)
Kuhlmann et al. [67] 2 49.6 402 250 2.86–4.86 1.56 1.23 (0.121)
Kuhlmann et al. [68] 4 43.4–47.6 400 250 2.86–4.86 1.57–1.92 1.05 (0.033)
Kulkarni et al. [69] 4 40.6–43.6 102 152 3.50–5.00 1.38 1.04 (0.053)
Laupa et al. [70] 6 14.8–32.3 152 262–269 4.82–4.95 1.91–3.97 0.89 (0.062)
Leonhardt et al. [30] 26 13.3–38.3 50–500 70–600 2.78–5.91 1.33–2.40 1.01 (0.113)
Marti et al. [71] 2 29.3–29.5 400 167 3.83 1.38–1.84 1.18 (0.029)
Mathey et al. [72] 9 23.5–30.6 203 403 2.84–3.78 0.47–2.54 0.93 (0.085)
Moody et al. [73] 23 15.4–41.2 152–178 262–272 2.85–3.41 1.60–2.37 0.95 (0.078)
Morrow et al. [74] 12 14.7–45.7 305–308 356–375 3.00–8.10 1.27–3.91 1.00 (0.045)
Mphonde et al. [75] 9 21.3–96.1 152.4 298 3.49 2.33–3.34 1.07 (0.043)
Niwa et al. [76] 3 24.6–27.1 300–600 1000–2000 2.98 0.14–0.28 1.25 (0.135)
Podgomiak [77] 8 37–99 300 225–925 2.88–2.97 0.51–3.14 0.98 (0.101)
Rajagopalan et al. [78] 10 23.7–36.6 151–154 259–268 3.83–4.27 0.25–1.73 0.99 (0.170)
Regan [79] 5 24.5–29.9 152 272 3.27 1.46 1.06 (0.076)
Rehm et al. [80] 1 23.7 900 313 3.19 1.21 1.18 (−)
Reineck et al. [81] 3 24.6–25.8 500 225–226 2.50–3.50 0.79–1.39 1.02 (0.032)
Remmel [82] 4 84.5–85.1 150 160–165 3.06–4.00 1.87–4.09 0.97 (0.053)
Rüsch et al. [83] 3 23–24.2 90–180 111–262 3.60 zo 3.62 2.64–2.65 0.87 (0.014)
Salandra et al. [84] 4 53–70.1 102 171 2.59–3.63 1.45 0.83 (0.127)
Scholz [85] 7 80.6–96.8 200 362–372 3.00–4.00 0.81–3.36 1.01 (0.132)
Taylor [86] 6 27.8–33.2 203 370 3.02–3.50 1.03–1.55 1.09 (0.104)
Taylor [87] 5 22–28.7 200–400 465–930 3.01 1.35 1.12 (0.090)
Walraven [88] 3 24.1–24.4 200 125–720 3.00 0.74–0.83 1.07 (0.077)
Xie et al. [89] 2 38.5–100.9 127 216 2.93 2.08 0.83 (0.022)
Yoon et al. [90] 3 36–87 375 655 3.23 2.88 0.83 (0.052)
Yoshida [91] 1 33.6 300 1890 2.82 0.74 0.95 (−)
Lubell [92] 7 36.9–41 249–1170 287–538 2.91–3.57 0.33–1.73 1.19 (0.079)
Sherwood [35] 21 28.1–77.3 122–300 280–1450 2.77–2.87 0.25–0.83 1.11 (0.111)
Thiele [93] 5 24.6–42.5 400 167–297 3.47–4.97 0.93–2.41 1.05 (0.092)
Winkler [94] 5 35.1 150–450 200–900 3.91–3.94 1.18–1.20 1.09 (0.067)
Rosenbusch [95] 2 43.4 200 260 3.37 0.65–3.55 0.98 (0.077)
Tureyen et al. [96] 3 40.9–43.7 457 360 3.38 0.36–1.92 1.07 (0.121)
Bentz et al. [97] 9 34.3–36.1 101–106 84–333 2.80–2.94 1.55–1.63 1.05 (0.091)
Sneed et al. [98] 8 64.8–74.8 203–613 233–822 2.87–2.92 1.20–1.60 0.79 (0.290)
Cavagnis et al. [11] 6 32.6–35.6 250 556–559 2.52–6.92 0.54–0.89 1.04 (0.078)
635 1.01 (0.136)

Appendix C. Supplementary material

Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.engstruct.2017.12.004.

References [9] Campana S, Anastasi A, Fernández Ruiz M, Muttoni A. Analysis of shear-transfer

actions on one-way RC members based on measured cracking pattern and failure
kinematics. Mag Concr Res 2013;65:386–404.
[1] Collins MP, Mitchell D, Adebar P, Vecchio F. A general shear design method. ACI [10] Cavagnis F, Fernández Ruiz M, Muttoni A. Shear failures in reinforced concrete
Struct J 1996;93:36–45. members without transverse reinforcement: An analysis of the critical shear crack
[2] Park H-G, Choi K-K, Wight JK. Strain-based shear strength model for slender beams development on the basis of test results. Eng Struct 2015;103:157–73.
without web reinforcement. ACI Struct J 2006;103:783–93. [11] Cavagnis F, Fernández Ruiz M, Muttoni A. An analysis of the shear transfer actions
[3] Muttoni A, Fernández Ruiz M. Shear strength of members without transverse re- in reinforced concrete members without transverse reinforcement. Struct Conc
inforcement as function of critical shear crack width. ACI Struct J 2008;105:163–72. 2017. http://dx.doi.org/10.1002/suco.201700145.
[4] Yang Y, Walraven J, den Uijl J. Shear behavior of reinforced concrete beams [12] Huber P, Huber T, Kollegger J. Investigation of the shear behavior of RC beams on
without transverse reinforcement based on critical shear displacement. J Struct Eng the basis of measured crack kinematics. Eng Struct 2016;113:41–58.
2016;143:4016146. [13] Reineck K-H. Ultimate shear force of structual concrete members without transverse
[5] Tung ND, Tue NV. A new approach to shear design of slender reinforced concrete reinforcement derived from a mchanical model. ACI Struct J 1991;88:592–602.
members without transverse reinforcement. Eng Struct 2016;107:180–94. [14] Zararis PD, Papadakis GC. Diagonal shear failure and size effect in RC beams
[6] Fisker J, Hagsten LG. Mechanical model for the shear capacity of R/C beams without web reinforcement. J Struct Eng 2001;127:733–42.
without stirrups: A proposal based on limit analysis. Eng Struct 2016;115:220–31. [15] Reinhardt HW. Fracture mechanics of an elastic softening material like concrete.
[7] Marí A, Bairán J, Cladera A, Oller E, Ribas C. Shear-flexural strength mechanical Heron 1984;29:42.
model for the design and assessment of reinforced concrete beams. Struct [16] Fédération Internationale du Béton (fib), fib Model Code for Concrete Structures
Infrastructure Eng: Maintenance, Manage, Life-Cycle Des Perform 2014;8:337–53. 2010, Ernst & Sohn 2013:434.
[8] Fernández Ruiz M, Muttoni A, Sagaseta J. Shear strength of concrete members [17] CEB-FIP Model Code. Design code. London: Thomas Telford; 1990. p. 1993.
without transverse reinforcement: a mechanical approach to consistently account [18] Walraven JC. Fundamental analysis of aggregate interlock. J Struct Div
for size and strain effects. Eng Struct 2015;99:360–72. 1981;107:2245–70.

313
F. Cavagnis et al. Engineering Structures 157 (2018) 300–315

[19] Gambarova PG, Karakoc C. A new approach to the analysis of the confinement role Schubtragverhalten hochfester Betone. Deutscher Ausschuss Fuer Stahlbeton; 1997.
in regularly cracked concrete elements. In: Transactions of the 7 international p. 477.
conference on structural mechanics in reactor technology 1983; H. [56] Hallgren M. Shear tests on reinforced high and normal strength concrete beams
[20] Li B. Contact density model for stress transfer across cracks in concrete. J Faculty without stirrups. Royal Institute of Technology, Stockholm, Sweden: Dept. of
Eng, the Univ Tokyo 1989;1:9–52. Structural Engineering; 1994.
[21] Ulaga T. Betonbauteile mit Stab- und Lamellenbewehrung: Verbund- und [57] Hamadi YD. Force transfer across cracks in concrete structures, PhD Thesis, London,
Zuggliedmodellierung. PhD Thesis. Thesis no. 15062 (in German). Zurich, UK: Polytechnic of Central London; 1976.
Switzerland: ETHZ; 2003:160. [58] Hanson JA. Tensile strength and diagonal tension resistance of structural light-
[22] Guidotti R. Poinçonnement des planchers-dalles avec colonnes superposées forte- weight concrete. ACI J 1961;58:1–40.
ment sollicitées. PhD Thesis. Thesis no. 4812 (in French). Lausanne, Switzerland: [59] Hedman O, Losberg A. Design of concrete structures with regard to shear forces.
Ecole Polytechnique Fédérale de Lausanne; 2010:416. CEB Bulletin D’Information No. 126; 1978.
[23] Jacobsen JS, Poulsen PN, Olesen JF. Characterization of mixed mode crack opening [60] Islam MS, Pam HJ, Kwan AKH. Shear capacity of high-strength concrete beams with
in concrete. Mater Struct 2012;45:107–22. their point of inflection within the shear span. Proc Inst Civ Eng Struct Build
[24] Sherwood EG, Bentz EC, Collins MP. Effect of aggregate size on beam-shear strength 1998;128:91–9.
of thick slabs. ACI Struct J 2007;104:180–90. [61] Johnson MK, Ramirez JA. Minimum shear reinforcement in beams with higher
[25] Collins MP, Kuchma D. How safe are our large, lightly reinforced concrete beams, strength concrete. ACI Struct J 1989;86:376–82.
slabs, and footings? ACI Struct J 1999;96:482–90. [62] Kani GNJ. How safe are our large reinforced concrete beams? ACI J
[26] Fernández Ruiz M, Mirzaei Y, Muttoni A. Post-punching behavior of flat slabs. ACI 1967;64:128–41.
Struct J 2013;110:801–12. [63] Kawano H, Watanabe H. Shear strength of reinforced concrete columns – Effect of
[27] Fernández Ruiz M, Plumey S, Muttoni A. Interaction between bond and deviation specimen size and load reversal. Proceedings of the Second Italy-Japan Workshop
forces in spalling failures of arch-shaped members without transverse reinforce- on Seismic Design and Retrofit of Bridges 1997:141–54.
ment. ACI Struct J 2010;107:346–54. [64] Kim J-K, Park Y-D. Shear strength of reinforced high strength concrete beam
[28] Fernández Ruiz M, Campana S, Muttoni A. Discussion of paper “Influence of without web reinforcement. Mag Concr Res 1994;46:7–16.
Flexural Reinforcement on Shear Strength of Prestressed Concrete Beams” by E. I. [65] Krefeld WJ, Thurston CW. Studies of the shear and diagonal tension strength of
Saquan and R.J. Frosch. ACI Struct J 2009;106:907–8. simply supported R/C-beams. ACI J 1966;63:451–76.
[29] Calvi PM, Bentz EC, Collins MP. Reversed cyclic experiments on shear stress transfer [66] Küng R. Ein Beitrag zur Schubsicherung im Stahlbetonbau. Betonstahl in
across cracks in reinforced concrete elements. ACI Struct J 2016;113:851–9. Entwicklung, H.33. Luxemburg: TOR-ISTEG Steel Corp; 1985. p. 8–25.
[30] Leonhardt F, Walther R. Schubversuche an einfeldrigen Stahlbetonbalken mit und [67] Kuhlmann U, Ehmann J. Versuche zur Ermittlung der Querkrafttragfähigkeit von
ohne Schubbewehrung. DAfStb H.151, Berlin 1962. Verbundplatten unter Längszug ohne Schubbewehrung. Versuchsbericht 2001-6X:
[31] Reineck K, Bentz EC, Fitik B, Kuchma DA, Bayrak O. ACI-DAfStb database of shear Mitteilung; 2001.
tests on slender reinforced concrete beams without stirrups. ACI Struct J [68] Kuhlmann U, Zilch K, Ehmann J, Jähring A, Spitra F. Querkraftabtragung in
2013;110:867–76. Verbundträgern mit schlaff bewehrter und aus Zugbeanspruchung gerissener
[32] Kani MW, Huggins MW, Wittkopp RR. Kani on shear in reinforced concrete. Stahlbetonplatte ohne Schubbewehrung- Mitteilungen. Institut für Konstruktion
Toronto, Canada: Deptartment of Civil Engineering, University of Toronto; 1979. und Entwurf Stahl-, Holz-, und Verbundbau, Universität Stuttgart, Germany; 2002.
[33] Bhal NS. Über den Einfluß der Balkenhöhe auf die Schubtragfähigkeit von einfel- p. 109.
drigen Stahlbetonbalken mit und ohne Schubbewehrung. Otto-Graf-Institut, H.35, [69] Kulkarni SM, Shah SP. Response of reinforced concrete beams at high strain rates.
Stuttgart, Germany; 1968. ACI Struct J 1998;95:705–14.
[34] Angelakos D, Bentz EC, Collins MP. Effect of concrete strength and minimum stir- [70] Laupa A, Siess CP, Newmark NM. The shear strength of simple-span reinforced
rups on shear strength of large members. ACI Struct J 2001;98:290–300. concrete beams without web reinforcement. Research Series No. 52, University of
[35] Sherwood EG. One-way shear behaviour of large, lightly-reinforced concrete beams Illinois at Urbana-Champaign, USA; 1953.
and slabs. PhD Thesis, Toronto, Canada: University of Toronto; 2008. p. 547. [71] Marti P, Pralong J, Thürlimann B. Schubversuche an Stahlbeton-Platten. IBK-
[36] El-Sayed AK, El-Salakawy EF, Benmokrane B. Shear strength of FRP-reinforced Bericht No. 7305–2, ETH Zürich, Zürich, Switzerland; 1977.
concrete beams without transverse reinforcement. ACI Struct J 2006;103:235–43. [72] Mathey RG, Watstein D. Shear strength of beams without web reinforcement con-
[37] Adebar P, Collins MP. Shear strength of members without transverse reinforcement. taining deformed bars of different yield strengths. ACI J 1963;60:183–207.
Can J Civ Eng 1996;23:30–41. [73] Moody KG, Viest IM, Elstner RC, Hognestad E. Shear strength of R.C. beams: Part 1
[38] Ahmad SH, Khaloo AR, Poveda A. Shear capacity of reinforced high-strength con- – tests of simple beams. ACI J 1954;26:317–32.
crete beams. ACI Struct J 1986;83:297–305. [74] Morrow J, Viest IM. Shear strength of reinforced concrete frame members without
[39] Ahmad SH, Park F, El-Dash K. Web reinforcement effects on shear capacity of re- web reinforcement. ACI J 1957;53:833–69.
inforced highstrength concrete beams. Mag Concr Res 1995;47:227–33. [75] Mphonde AG, Frantz GC. Shear tests of high- and low- strength concrete beams
[40] Aster H, Koch R. Schubtragfähigkeit dicker Stahlbetonplatten. Beton- Und without stirrups. ACI J 1984;81:350–7.
Stahlbetonbau 1974;69:266–70. [76] Niwa J, Yamada K, Yokozawa K, Okamura M. Revaluation of the equation for shear
[41] Lubell A, Sherwood T, Bentz E, Collins M. Safe shear design of large wide beams. strength of R.C.-beams without web reinforcement. Proceedings JSCE No. 372/V-5
Concr Int 2004;26:62–78. 1986–8 Translation. In: Concrete Library of JSCE, No. 9; 1987.
[42] Bernander KG. An investigation of the shear strength of concrete beams without [77] Podgorniak-Stanik BA. He influence of concrete strength, distribution of long-
stirrups or diagonal bars, reinforced with high tensile steel with various rib pat- itudinal reinforcement, amount of transverse reinforcement and member size on
terns. In: Proc., RILEM symposium on bond and crack formation in reinforced shear strength of reinforced concrete members. PhD Thesis, Toronto, Canada:
concrete; 1957. p. 211–4. University of Toronto; 1998.
[43] Bresler B, Scordelis AC. Shear strength of reinforced concrete beams. ACI J [78] Rajagopalan KS, Ferguson PM. Exploratory shear tests emphasizing percentage of
1963;60:51–74. longitudinal reinforcement. ACI J 1968;65:634–8.
[44] Cladera A, Marí A. Experimental study on high-strength concrete beams failing in [79] Regan PE. Shear in reinforced concrete – an experimental study. CIRIA-report, a
shear. Eng Struct 2005;27:1519–27. report to the construction research and information association. Imperial College of
[45] Cao S. Size effect and the influence of longitudinal reinforcement on the shear re- Science and Technology, Dep. of Civil Engineering, Concrete Section; 1971.
sponse of large reinforcement concrete members. PhD Thesis, Toronto, Canada: [80] Rehm G, Eligehausen R, Neubert B. Rationalisierung der Bewehrungstechnik im
Univerisity of Toronto; 2001. Stahlbetonbau. Betonwerk + Fertigteil-Technik 1978;222–7.
[46] Cederwall K, Hedman O, Losberg A. Shear strength of partially prestressed beams [81] Reineck K-H, Koch R, Schlaich J. Shear tests on reinforced concrete beams with
with pretensioned reinforcement of high grade deformed bars. Special Publ axial compression for offshore structures. Stuttgart, Germany: Institut für
1974;42:215–30. Massivbau, Universität Stuttgart; 1978.
[47] Chana PS. Some aspects of modelling the behaviour of reinforced concrete under [82] Remmel G. Zum Zugtragverhalten hochfester Betone und seinem Einfluß auf die
shear loading, Technical Report 43. Cement and Concrete Association; 1981. p. 22. Querkrafttragfähigkeit von schlanken Bauteilen ohne Schubbewehrung. PhD
[48] Chang TS, Kesler CE. Static and fatigue strength in shear of beams with tensile Thesis, Darmstadt, Germany: TH Darmstadt; 1991.
reinforcement. ACI J 1958;54:1033–57. [83] Rüsch H, Haugli FR, Mayer H. Schubversuche an Stahlbeton Rechteckbalken mit
[49] Diaz de Cossio R, Siess CP. Behavior and strength in shear of beams and frames Gleichmassig Verteilter Belastung, Deutscher Ausschuss fur Stahlbeton, Heft 145.
without web reinforcement. ACI J 1960;31:695–735. West Berlin: W. Ernst und Sohn; 1962.
[50] Thorenfeldt E, Drangsholt G. Shear capacity of reinforced high-strength concrete [84] Salandra MA, Ahmad SH. Shear capacity of reinforced lightweight high-strength
beams. ACI Spec Publ 1990;121:129–54. concrete beams. ACI Struct J 1989;86:697–704.
[51] Elzanaty AH, Nilson A, Slate FO. Shear capacity of reinforced concrete beams using [85] Scholz H. Ein Querkrafttragmodell für Bauteile ohne Schubbewehrung im
high-strength concrete. ACI J 1986;83:290–6. Bruchzustand aus normalfestem und hochfestem Beton. Berichte aus dem
[52] Feldman A, Siess CP. Effect of moment shear ratio on diagonal tension cracking and Konstruktiven Ingenieurbau Heft 21, Technische Universität Berlin, Berlin,
strength in shear of reinforced concrete beams. Univ. of Illinois Civil Eng. Studies, Germany; 1994.
Struct. Research Series No. 107 1955. [86] Taylor HPJ. Shear stress in reinforced concrete beams without shear reinforcement.
[53] Ferguson PM. Some implications of recent diagonal tension tests. ACI J Proc Cement and Concrete Ass., Techn. Rep. No. 407; 1968.
1956;53:157–72. [87] Taylor HPJ. Shear strength of large beams. ASCE J Struct Div 1972;98:2473–90.
[54] Ghannoum WM. Size effect on shear strength of reinforced concrete beams. McGill [88] Walraven JC. The influence of depth on the shear strength of lightweight concrete
University Montréal, Canada: Department of Civil Engineering and Applied beams without shear reinforcement. Report S-78-4, Stevin Lab., Delft University,
Mechanics; 1995. Delft, Netherlands; 1978.
[55] Grimm R. Einfluß bruchmechanischer Kenngrößen auf das Biege-und [89] Xie Y, Ahmad SH, Yu T, Hino S, Chung W. Shear ductility of reinforced concrete

314
F. Cavagnis et al. Engineering Structures 157 (2018) 300–315

beams of normal and high-strength concrete. ACI Struct J 1994;91:140–9. Stahlbetonbalken. Versuchsbericht 2011–1. Lehrstuhl für Massivbau, Ruhr
[90] Yoon Y-S, Cook WD, Mitchell D. Minimum shear reinforcement in normal, medium Universität Bochum, Bochum, Germany; 2011. p. 105.
and high- strength concrete beams. ACI Struct J 1996;93:576–84. [95] Rosenbusch J. Zur Querkrafttragfähigkeit von Balken aus stahlfaserverstärktem
[91] Yoshida Y. Shear reinforcement for large lightly reinforced concrete members. PhD Stahlbeton. Germany: TU Braunschweig, Braunschweig; 2004.
Thesis, Toronto, Canada: University of Toronto; 2000. [96] Tureyen AK, Frosch RJ. Shear tests of FRP-reinforced concrete beams without
[92] Lubell A. Shear in wide reinforced concrete beams. PhD Thesis, Toronto, Canada: stirrups. ACI Struct J 2002;99:427–34.
University of Toronto; 2006. p. 455. [97] Bentz EC, Buckley S. Repeating a classic set of experiments on size effect in shear of
[93] Thiele C. Zum Tragverhalten von Stahlbetonplatten ohne Querkraftbewehrung mit members without stirrups. ACI Struct J 2005;102:832–8.
integrierten Leitungsführungen. PhD Thesis, Kaiserslautern, Germany: TU [98] Sneed LH, Ramirez JA. Influence of effective depth on shear strength of concrete
Kaiserslautern; 2010. beams - experimental study. ACI Struct J 2010;107:554–62.
[94] Winkler K. Querkraftversuche an maßstäblich skalierten, schubunbewehrten

315

View publication stats