10th International Conference on Fracture Mechanics of Concrete and Concrete Structures

FraMCoS-X
G. Pijaudier-Cabot, P. Grassl and C. La Borderie (Eds)
https://doi.org/10.21012/FC10.239783

A LITERATURE REVIEW OF THE HEAD-SIZE EFFECT ON THE CAPACITY OF

CAST-IN ANCHORS

* †
G. DI NUNZIO , G. MUCIACCIA
*
Politecnico di Milano
Department of Civil and Environmental Engineering, Bulding n.4
Piazza Leonardo da Vinci 32, 20133 Milan, Italy
e-mail: [email protected], www.dica.polimi.it
† Politecnico di Milano
Department of Civil and Environmental Engineering, Bulding n.3
Piazza Leonardo da Vinci 32, 20133 Milan, Italy
e-mail: [email protected], www.dica.polimi.it

Key words: Cast-in anchors, Concrete cone, Head-size, Fasteners, Literature review.

Abstract: The design and the verification of anchorage to concrete are currently covered by the
so called “Concrete Capacity Design” approach, which is adopted in the most recent codes and
national regulations. Although it is quite advanced, such an approach is built on a wide range
dataset, which encompasses both cast-in and post-installed anchors with very different geometry
and tested under different boundary conditions. As a result, the CCD method was developed in a
consistent way adopting the simplifying assumption that some parameters, such as the head-size,
have a negligible influence on the pull-out capacity of the anchoring system. As the head-size
increases, some authors found that the method could be rather conservative, while others found that
the method is still accurate. Within this context, a literature review of the research studies and of the
available formulas for predicting the capacity of cast-in anchors is addressed in this paper, focusing
on the effect of the head-size.

anchors. The research started in Germany in

1 INTRODUCTION the mid ‘80s and at the beginning of ‘90s an
Several solutions for fastening structural extended version of the k-method was firstly
and non-structural elements to concrete have proposed by Hilti. This version was the first
been developed in the last decades. The application-oriented method, since it
request for more flexibility in the design and distinguished among different direction of
rehabilitation of civil structures focused the loading. The first design method for cast-in
attention of research on the behavior of post- anchors was developed in United States (US)
installed systems. However, several in the mid ‘70s and it was adopted by ACI
applications require the use of cast-in place Committee 349. Due to the lack of design aids
solutions. for cast-in parts, this method was often used
The first approval for a post-installed by designed also in Europe [1].
system was released in 1975. From this point The need of design reliable steel to concrete
on, a lot of efforts have been dedicated to the connections drove the improvements of the
reach a comprehensive knowledge on existing design procedure available at the time.
fastening behavior, focusing on mechanical These improvements were finally summarized
interlock and the mechanics of expansion in the so-called Concrete Capacity Design

1
 G. Di Nunzio and G. Muciaccia

method (CC method hereinafter), which different geometry and tested under different
coupled the accuracy of the k-method with the boundary conditions. As a result, the CC
transparency of ACI 349 approach. The main method was developed in a consistent way
improvements introduced by this approach adopting the simplifying assumption that some
were the distinction between different parameters, such as the head-size, have a
directions of loading, modes of failure and negligible influence on the pull-out capacity of
condition of the base material [1]. The above- the anchoring system. Therefore, the design
mentioned design approach is currently and the verification of cast-in solutions often
adopted in the most advanced design codes concerns simple detailing rules and the
and national regulations, for both cast-in place designer’s experience using anchors of various
and post-installed anchors. sizes under the assumption that the structural
Fasteners loaded in tension may exhibit behavior is unaffected by the size of the head.
four different failure mechanism [2]. From the comparison between tested and
Assuming the embedment depth not sufficient predicted failure loads, some authors found
to ensure ductile steel failure and the concrete how the method could be rather conservative
member properly designed to resist the effects for large head-size anchors [3] [4]. Among
induced by the fastener, concrete cone failure them, Nilforoush [5] extensively analyzed the
is the decisive failure mechanism. As it will be database of CC method as function of the
discussed, pull-out for cast-in anchors is bearing stress (Figure 2). From the analysis of
mainly a fictitious failure mechanism with the data, he concluded that CC method has not
increasing displacements than a real failure been developed in a consistent way with
mode. respect to the size of anchor heads over the
range of the studied anchor embedment
depths, because the bearing stresses under the
heads of the tested anchors varied significantly
[5]. As a matter of facts, Fuchs [1] reported
that the CC method was developed by
assembling the European and the North
American test data from previous decades.
Nevertheless, as reported in some projects,
cast-in anchors geometry, and particularly the
head-size, could be well beyond the range of
applicability of CC method [6] [7].

Figure 1: Failure modes associated with tensile

loading [2].

Within this context, a literature review on

the behavior of cast-in place anchors when
concrete cone breakout is the dominant failure
mechanism is presented hereinafter. Although Figure 2: Ratio of the bearing stress under the heads
of the tested headed anchors at peak load to the
it is quite advanced, CC approach is built on a concrete compressive strength as a function of
wide range dataset, which encompasses both anchor embedment depth[5].
cast-in and post-installed anchors with very

2
 First A. Author, Second B. Author and Third C. Coauthor

achieves its maximum value according to the

Bažant’s law [10] and the failure load
2 SCOPE increases proportionally to . Nevertheless,
The main objective of the present paper is from experimental tests and numerical
to review existing literature review on the analyses on large size anchors, it was found
available models for predicting the concrete that the size-effect decreases [3] [4], thus the
cone capacity of cast-in place anchors, with ACI 318 provides an alternative approach as it
particular emphasis on the influence of the follows:
head-size. The most relevant studies from
literature are reported and discussed. (2)
The plane of the paper is as it follows: in
the section titled “National codes and
regulations” the current design provisions as Where: Nu,m is mean concrete cone capacity
adopted by different codes are described; in in (N), fc is the concrete compressive strength
the section titled “Research studies” the most measured on cylinders in (N/mm2) and hef is
relevant studies on anchorage behavior are the effective embedment depth in (mm).
reported and discussed, highlighting still open
critical issues; finally, some conclusions are 3.3 EN 1992-4: 2018
provided placing this work in a larger context. EN 1992-4 [11] adopts the formulas from
Sub-sections are introduced and titled as useful CCD approach. Following the equations are
support for the reader. reported with reference to mean and not to
characteristic values:
3 NATIONAL CODES AND
REGULATIONS
As a premise for the literature review, the (3)
most advanced codes and regulations are
summarized as it follows. Some of the Where: Nu,m is mean concrete cone capacity
equations reported herinafter are, indeed, in (N), fc is the concrete compressive strength
adopted in most of the research studies measured on cylinders in (N/mm2) and hef is
presented in the following section. the effective embedment depth in (mm).
3.1 ACI 349-90 3.4 Other design-oriented documents
ACI 349-90 [8] provided a model based on The CC approach was integrated in other
the 45° degrees concrete cone approach, as it design-oriented documents in Europe and,
follows: particulary, in CEB Design Guide [12], in fib
bulletin 58 [13]. Since these documents
(1)
adopted an unaltered version of the method,
Where: Nu,m is mean concrete cone capacity they are not adressed in detail.
in (N) the fcc is the concrete compressive
strength measured on cubes in (N/mm2), hef is 4 RESEARCH STUDIES
the effective embedment depth in (mm) and dh In the past, literature reviews regarding the
is the head diameter in (mm). behavior of anchorage in concrete were
published by different authors. In 1982,
3.2 ACI 318-14 Klingner and Mendonca [14] reviewed the
ACI 318-14 [9] incorporated the CC available procedure for predicting tensile
method with minor modifications only. The capacity of anchor bolts and welded studs,
CC method is based on the assumption that the with reference to both steel and concrete
size-effect for concrete cone mechanism related failure modes. They performed an

3
 G. Di Nunzio and G. Muciaccia

extensive comparison between the test data Where: Nu,m is mean concrete cone capacity
and the predicted capacity available till the in (lb) the fc is the concrete compressive
time of the publication, providing strength measured on cylinders and AN is the
recommendations for design and verification. projected area of the concrete cone on the
Few years later, Eligehausen and Sawade member surface.
wrote a contribution to the Technical The head-size is directly included in the
Committee 90-FMA RILEM, in which they shape of the extraction cone. The effect of the
summarized the proposed analytical models by head-size parameter is, however, to increase
distinguish among different theoretical the projection of the arbitrary 45° cone only. It
approaches based on theory of elasticity, was demonstrated that this method leads to
theory of plasticity, strength criterion with unconservative predictions of the cone
smeared crack and fracture mechanics [15]. In capacity [17] [6].
the framework of the International RILEM This procedure for predicting the cone
Symposium on Connections between Steel and capacity was adopted in several US references,
Concrete, Fuchs [1] presented a paper with a as PCI Design Handbook, TRW-Nelson, TVA
detailed review of the evolution of design Civil Standard and ACI Committee 349 [14].
approaches for fastening to concrete till the According to Fuchs [1], this design approach
beginning of the 21st century. The most recent was originally adopted by ACI 349 because it
background for the cone capacity is available was conceptually simple and in satisfactory
in Eligehausen, Silva and Mallée [2]. agreement with the limited tests results
The authors would like to expand the available at the time. Fuchs provided also an
above-mentioned literature reviews by additional explanation for the adoption of this
including some comments on the head-size method in ACI 349: since the ACI Committee
effect, which are not provided in the original was concerned with the nuclear-related
versions, and by adding the most recent structures, the philosophy was to design
research projects available. ductile fastening [17]. Sufficient embedment
depth such that the concrete must exceed the
4.1 The 45° model strength of the steel was, indeed, suggested.
One of the first model for the prediction of
concrete cone capacity was developed in the 4.2 Fracture mechanics-based description
framework of the researches on nuclear-related of concrete cone
structures by Cannon et al [16]. In this model, A detailed survey regarding the fracture
the cone is defined assuming a failure surface mechanics-based description of concrete cone
radiating from the anchor head to the surface breakout is available in Eligehausen, Mallée
at an assumed angle of 45°. The capacity is and Silva [2]. Several theoretical models were
then calculated imposing the equilibrium developed from the mid ‘80s till the early ‘90s
between the external load and the projection of to explain concrete cone breakout. Ballarini et
the nominal tensile stress acting perpendicular al. [18] developed an analytical model on the
to the surface of the cone. The distribution of basis of experimental results. Although they
tensile stress along the side of the cone is focused the attention on the stability of the
taken linear with the maximum at the head crack growth, they firstly investigated the
location and zero at the concrete surface. effect of different head-sizes, as depicted in
Seeking for simplicity, an average value for Figure 3
the tensile stress is assumed equal to 4 in
[US] units.
The final equation for the cone capacity is
at it follows:
(4)

4
 First A. Author, Second B. Author and Third C. Coauthor

energy (i.e. the sum of elastic deformation

energy and surface energy along the crack).
Assuming a linear fracture mechanics
approach, they calculated the value of the
ultimate load as it follows:

(5)

Where: Nu,m is mean concrete cone capacity

Figure 3: Mathematical model for an embedded anchor
in (N), Ec is the elastic modulus of concrete in
[18]. (N/mm2), Gf is the I mode fracture energy, a
Since they systematically scaled up the the actual crack length measured along the
ratio of head-size and supports distance with crack path, lB is the assumed crack path (37.5°
respect to the embedment depth, the results with respect to the horizontal direction, hef is
cannot be directly interpreted in terms of head- the effective embedment depth in (mm)
size effect. However, it was observed that the I Calculating the failure load from CC
mode stress-intensity factor tends to decrease method in uncracked concrete, the bearing
as the head-size increases. Elfgren [19] pressure is approximately equal to for
numerically investigated concrete cone all the tests, thus no conclusions can be argued
breakout assuming three different paths for the on the influence of head-size. Nevertheless, it
crack path: i. a straight crack inclined 45° is worth to be noticed how the ratio was
degrees with respect to the vertical direction; found to decrease for larger embedment depth,
ii. a straight crack inclined 67° with respect to thus clearly indicating the presence of size-
the vertical direction; iii. a curved crack with effect.
an initial inclination of 73°. He found that the The size-effect on concrete was further
45° path is unlikely to develop in real investigate by Eligehausen et Ožbolt [21] and
structures, because the maximum stress is by Eligehausen et al. [22] during the so-called
reached first in the shear direction. “Prague tests”. In the first study, Eligehausen
Conversely, for the 67° crack path the and Ožbolt numerically simulated the behavior
maximum stress is reached first in tension. of anchor bolts with three different embedment
Eligehausen and Sawade [20] analyzed the depths: 50 mm, 150 mm and 450 mm. Using
results from experimental tests on headed the Bažant’s size-effect low [10], they
anchors with embedment depth from 130 mm calibrated the constants from linear regression
to 520 mm and different head-sizes (Figure 4). analysis of the obtained failure loads (Figure
5).

(6)

Where: Nu,m is mean concrete cone capacity

in (N), fcc is the concrete compressive strength
Figure 4: Different head-sizes tested by Eligehausen measured on cubes in (N/mm2), hef is the
and Sawade [20]. effective embedment depth in (mm), k and
are the calibration factors according to the
Starting from an energetic model, the solution Bažant’s law.
is defined by the minimization of the free

5
 G. Di Nunzio and G. Muciaccia

prism model, as depicted in Figure 6.

Figure 6: Concrete cone surface idealized according

Figure 5: Comparison between size-effect law and
to CCD approach.
numerical results from Eligehausen et al [21].

On the “Prague tests”, the size-effect of The main difference with the approaches
concrete cone was experimentally investigated from the ‘70s relies on the fracture mechanics
using the same embedment depths of the approach. The main outcome from the models
previous stud but performing two series of of late ‘80s till the middle ‘90s it that the
tests: the former, identified with “A”, concrete cone breakout is affected by the size-
assuming an head diameter 1.90 times the effect. The failure loads, thus, increases less
shaft diameter and the latter, identified with the available failure surface, and is given by
“B”, for which the head diameter was fixed the following equation:
such to obtain a constant head pressure equal
to . The size-effect law was, then,
recalibrated according to the test results as it (8)
follows:
(7)
Where: Nu,m is mean concrete cone capacity
Where: Nu,m is mean concrete cone capacity
in (N), fc is the concrete compressive strength
in (N), fcc is the concrete compressive strength
measured on cylinders in (N/mm2), hef is the
measured on cubes in (N/mm2), hef is the
effective embedment depth in (mm), k1, k2 and
effective embedment depth in (mm).
k3 are calibration factors which can be
From the comparison between test results, it
resumed in the single k factor, k factor is equal
was observed an increase of the failure load
to 16.8 and to 12.5 for uncracked and for
for larger heads (test series A). Such an
cracked conditions, respectively.
increase of the failure load could be hidden in
From the analysis of equation (8), three
the exponent of equation (7), which slightly
different parts can be highlighted:
differs from equation (6) (i.e. assumed
maximum size-effect). - which represents the nominal
tensile strength of concrete;
4.3 The Concrete Capacity approach - which represents the basis of the
The genesis of the CC approach (or CCD in rectangular prism model;
-
US) for fastening to concrete, as published on - which represents the maximum
ACI Structural Journal and later adopted by size effect according to Bažant’s law.
the most recent design codes, was explained in The CC approach seems to be in contrast
detail by Fuchs [1], who is also one of the with the fracture mechanics description of the
author of the method. The method was concrete cone failure, because no dependency
developed by implementing the expanded k- of the failure load from the fracture properties
factor method by a user-friendly rectangular is introduced. However, such a dependency is

6
 First A. Author, Second B. Author and Third C. Coauthor

hidden in the term. In fact, assuming that Figure 7: Schematic concrete cone breakout for
different head-sizes [2].
the elastic modulus of concrete is proportional
to and the fracture energy is proportional
The influence of the head-size was
to , it can be easily demonstrated that is experimentally investigated by fixing the
proportional to the product . Despite it is embedment depth at 80 mm and by varying the
less representative of the concrete matrix shoulder size from 0.5 mm to 4 mm (i.e. the
fracture properties, the use of concrete distance measured from the shaft surface to the
compressive strength was intended for ease of tip of the head). The tests were performed in
utilization. both cracked and uncracked concrete. On this
It worth to be noticed how the method does basis, Furche proposed the following model:
not take into account for different head-sizes.

4.4 Behavior of small head anchors

Lieberum investigated the compressive
punching of concrete by varying the size of
circular and annular bearing areas [23]. He
found that the bearing pressure-displacement
relation is linear till 5 · fc and 2.5 · fc for
circular and annular bearings, respectively.
Although they are not directly related to
anchorages, the results could provide useful
information about the behavior of cast-in (9)
anchors. The structural response under tensile
loads is, indeed, strongly affected by the
bearing pressure both in terms of stiffness and
load capacity. As the pressure increases at the
Where: Nu,m is mean concrete cone capacity
anchor’s head location (i.e. small bearing
in (N), fcc is the concrete compressive strength
heads), the displacements tend to increase
measured on cubes in (N/mm2), hef is the
more than linearly thus yielding a reduction of
effective embedment depth in (mm), u is the
the effective embedment depth and,
displacement at peak due to concrete crushing
consequently, of the load capacity.
in (mm), a is the width of the sholder, ka and kA
The behavior of headed studs subjected to
are factors function of the geometry, bp is the
tensile load was extensively investigated by
bearing pressure in (N/mm2) as function of the
Furche [24]. He described the pull-out failure
applied tensile load, Ah is the head area in
as function the bearing pressure. Specifically,
(mm2), d is the shaft diameter in (mm), dh is
if the bearing area is large enough, the
the head diameter in (mm).
concrete cone resistance is not affected by the
As the bearing pressure increases, it was
local crushing of concrete. Conversely, smaller
observed a reduction of the resistance about
head leads to a reduction of the effective
the 77% with respect to the CC method.
embedment depth as consequence of concrete
Furthermore, limits for the bearing pressure to
crushing under the head, (Figure 6).
avoid pull-out failure without development of
the cone surface were derived (Figure 8):
(1
0)

7
 G. Di Nunzio and G. Muciaccia

(1
1)

Figure 8: Maximum bearing pressure to ensure

concrete cone failure without reduction of the
embedment depth in uncracked concrete [2].

Berger [25] investigated the relation

between load and displacement and the
dimensioning of anchor bolts subjected to
tensile load, with and without additional Figure 9: Influence of compressive stength on the
reinforcement. head displacement [25].
He suggested that the displacement
behavior of a bolt far from free edges and 4.5 Behavior of large size anchors
sufficiently spaced from other anchors is
composed by the following contributions: Ožbolt et al performed an extensive
- Axial deformation of the steel shaft; numerical investigation on anchor bolts with
- Displacement due to concrete crushing; large embedment depths [3]. The embedment
- Uplift of the breakout cone. depth was varied from 150 mm to 1500 mm
He presented, then, a simplified model for and, for each embedment depth, three different
the descending part of the load-displacement head-size were simulated, namely “small”,
curve. Specifically, it was assumed that the “medium” and “large” head-sizes. Non-linear
contribution of the concrete cone breakout is analyses were carried out by pulling out
negligible till the peak load is achieved, while headed anchors from unreinforced concrete
the contribution of concrete crushing is fully blocks. Concrete was modelled adopting the
exploided. After the peak, this contribution micro-plane model proposed by Ožbolt [26].
increases almost linearly till the fracture A general increasing of the resistance and
surface reaches the surface and load drops to of the stiffness was observed as the head-size
zero. The slope of the descending branch is increases. Regarding the crack patterns, it was
affected by the embedment depth and the observed that the crack length at the peak was
compressive strength of concrete. In general, shorter for smaller anchor heads and the cone
brittleness of the curve tends to increase with was steeper than in the case of larger heads.
the compressive strength and the embedment Stress gradients in the neighborhood of the
depth. This behavior agrees with some anchor’s head is believed to be the main
numerical results from literature [3]. The axial responsible for those differences. The size
deformation of the anchor shaft is calculated effect on concrete is smaller for larger heads,
and summed up according to the actual value and relevant changes in the main crack is
of the applied load. observed, which propagates in almost mode-I
From the analysis of load-displacement failure instead of mixed-mode fracture.
curves, he presented a modification for the Therefore, a refinement for the CCD approach
approach originally developed by Furche, but was proposed to take into account for the
assuming a different dependency on the effect of the head-size:
concrete compressive strength as depicted in
Figure 9:

8
 First A. Author, Second B. Author and Third C. Coauthor

cracks and a circular pattern around the

anchor. According to Lee et al., the top and the
bottom of the specimens were slightly
reinforced to minimize shrinkage cracks. In
(1
authors opinion, however, this reinforcement
2)
played non negligible role on the development
of concrete cone. As reported in very recent
research project [7], a global mode of failure
involving the entire concrete slab was
Where: Nu,m is mean concrete cone capacity observed for very low bearing pressure
in (N), fc is the concrete compressive strength Therefore, it is challenging to distinguish
measured on cylinders in (N/mm2), fcc is the among the cracked and uncracked conditions,
concrete compressive strength measured on since the reinforcement takes the
cubes in (N/mm2) hef is the effective bending/splitting cracks that cannot be
embedment depth in (mm), Nu,CCD is the mean avoided.
concrete cone capacity calculated according to A comparison of test results with the most
CCD approach in (N), is a factor that takes used predictors from technical literature was
into account for the effective bearing area and reported in the paper. From the comparison it
it is equal to the ratio of the current bearing arises that the method from ACI 349 (i.e. 45°
area Ah and the area A0 which provides the degrees model) is unconservative even for
capacity from CCD approach, k is an exponent larger anchors. On the other end, CC approach
that accounts for the effect of increasing is conservative for large anchors yielding to
embedment depth. measured failure loads approximately 10%
Although the simulation of several cases, higher than the previsions. As an explanation
the authors would highlight that the anchors as for such an increase, it is argued that the CC
simulated in this research projects are still method is based on linear fracture mechanics,
characterized by bearing pressures higher than which is valid for the smaller head-size
5 · fc. anchors (i.e. high stress gradients in the
Lee et al. [6] carried out an experimental neighborhoods of the head). Finally, it is
campaign on large anchors with embedment suggested to adopt the modified CC approach
depths systematically greater than 635 mm. according to ACI 318 Appendix D when the
Such a research project is quite rare in the bearing pressure is lower to 3 or 2.4 times the
technical literature because large anchors concrete compressive strength in uncracked
require high economical efforts to properly concrete and cracked concrete, respectively.
built and test the concrete specimens. As it In order to investigate the influence of
will be discussed in the next section, 635 mm different parameters on the concrete cone
embedment depth represents the upper bound capacity for cast-in anchors, Nilforoush
for the use of CCD approach according to US recently carried out experimental tests [4] and
regulations. numerical analyses [27].
Specimens were designed such that the From the comparison with available
embedment depth increases with constant models, he found that the current design
bearing pressure, which was fixed methods underestimate the tensile resistance of
approximately equal to . Such a value is the anchors and, particularly, when the head-
within the range of full cone development as size increases. However, a more brittle
suggested by other authors [2]. Although the response and increasing of the critical spacing
specimens were designed to avoid splitting and critical edge distance is observed. Those
failure, the observed crack pattern was results match with the observations by Ožbolt
composed by one major crack centered on the [3]
sides of the blocks, horizontal and transversal Hence, a modified CC method was
proposed for the evaluation of the expected

9
 G. Di Nunzio and G. Muciaccia

cone resistance. In particular, three different while for the very large head size the bearing
coefficients, which multiplies the basic cone pressure dramatically drops under the unity.
resistance, were calibrated on the data from Although these anchor bolts are representative
numerical analyses and experimental tests. The of real applications, no comparable data are
coefficients account for the increasing of the available in technical literature. A limited
thickness of the concrete member (ψH), for the increase in the failure load for the very large
increasing of the head-size (ψAH) and for the head-size only is reported [7]. However,
presence of surface reinforcement (ψSr). depending on the bearing pressure ratio, a
ψ ψ ψ migration from local mode of failure (i.e. pure
concrete cone) to global failure involving the
entire concrete slab was observed, for which
surface reinforcement have been placed on
ψ purpose.

ψ (1 5 CONCLUSIONS
3)
A literature review of the existing
ψ approaches for the prediction of the concrete
cone resistance of cast-in anchors is presented
in this paper, with emphasis on the effect of
head-size.
Starting from the existing codes and
regulations, a review of the main research
Where: Nu,m is mean concrete cone capacity studies on concrete cone capacity is reported.
in (N), Ncodeu,m is the mean concrete cone The background documents behind the so-
capacity calculated according to the reference called CC approach are carefully analyzed
code in (N), fc is the concrete compressive highlighting, if available, information about
strength measured on cylinders in (N/mm2), hef the influence of head-size. Two sections are,
is the effective embedment depth in (mm), Ah finally, dedicated to the behavior under
and the area A0 which provides the capacity extreme pressure conditions: small head
from CC approach, H is the thickness of anchors as well as large head anchors.
concrete member in (mm). The following conclusions can be drawn:
Although the method is refined, the - The approach adopted by the most
applicability of the correction factor for CC advanced codes and regulations is the
merely taking into account the head-size can CC (or CCD) approach. Such a method
be questioned. The above-mentioned provides accurate predictions for a
conclusion was derived in the framework of a rather wide range of cast-in anchors.
recent research projects, in which the authors However, it was found to be
investigated the head-size effect on the conservative for large head-size anchors
behavior of cast-in anchors for real industrial and unconservative for small head-size
applications [7]. In particular, anchor bolts anchors. This may be related to the base
with very different head-sizes were tested reference dataset, which encompass a
under monotonic tensile loading by keeping wide range of anchors with different
fixed the embedment depth. Calculating the geometries and tests with different
ratio between predicted load and load bearing boundary conditions;
area, it follows that the second and the third - Models based on fracture mechanics
investigated solutions are able to develop very rarely assume the influence of head-size
low bearing pressure at the peak. Specifically, as relevant parameter for the behavior
the second solution develops up to , of cast-in anchors. This is probably due

10
 First A. Author, Second B. Author and Third C. Coauthor

to the main interest in solving the size- Anchorage In Concrete Construction.

effect most related to the embedment Berlin: Ernst & Sohn, 2006.
depth, which drove the research in the [3] J. Ožbolt, R. Eligehausen, G. Periškić,
‘80s. Since the gradient in the and U. Mayer, “3D FE analysis of
neighborhoods of the head strongly anchor bolts with large embedment
depends on the shape of the head, the depths,” Eng. Fract. Mech., vol. 74, no.
authors suggest that the size-effect 1–2, pp. 168–178, 2007.
could hide a superimposition of several [4] R. Nilforoush, M. Nilsson, and L.
geometrical parameters, namely the Elfgren, “Experimental evaluation of
embedment depth and the anchor’s tensile behaviour of single cast-in-place
head; anchor bolts in plain and steel fibre-
- Some refinement proposals for the CC reinforced normal- and high-strength
approach have been proposed for the concrete,” Eng. Struct., vol. 147, pp.
inclusion of the head-size of cast-in 195–206, Sep. 2017.
anchors. For small head-size, iterative [5] R. Nilforoush, “Anchorage in Concrete
procedures were presented in order to Strucutres - Numerical and
calculate the reduction of the failure experimental Evaluation of Load-
load due to concrete crushing. Carrying capacity of CI headed anchors
Conversely, for large head-size anchors, and PI adhesive anchors,” Luleå
the refinement proposals were mainly University of Technology, 2017.
based on the calibration of correction [6] N. H. Lee, K. S. Kim, C. J. Bang, and
factors for the basic resistance K. R. Park, “Tensile-headed anchors
calculated according to CC method. It is with large diameter and deep
discussed how the applicability of embedment in concrete,” ACI Struct. J.,
refinement methods based on the vol. 104, no. 4, pp. 479–486, 2007.
introduction of correction factors can be [7] G. Di Nunzio, A. Marchisella, and G.
challenged, because the fracture process Muciaccia, “The effect of very low
of large head size anchors is found to be bearing pressure on the behavior of
very different from the basic cast-in anchors,” in Proceedings of
assumption of the method; CONSEC 2019, 2019, pp. 1–10.
- From the results of a recent research [8] ACI, ACI 349-90 - Code requirements
project on anchors with very large for nuclear safety related concrete
heads, a strong interaction with the structures. 1990.
structural behavior of concrete members [9] ACI Committee 318, Aci 318-14. 2014.
was found, despite the design of [10] Z. P. Bažant, “Size Effect in Blunt
specimen should have ensured Fracture: Concrete, Rock, Metal,” J.
undisturbed conditions. On this basis, it Eng. Mech., vol. 110, no. 4, pp. 518–
is suggested to further investigate the 535, 1984.
influence of head-size jointly with the [11] CEN/TC250, “EN1992-4 - Eurocode 2:
structural response of the concrete Design of concrete structures - Part 4:
member. Design of fastenings for use in
concrete,” 2018.
REFERENCES [12] CEB-FIP, Fastenings to Concrete and
[1] W. Fuchs, “Evolution of fastening Masonry Structures - State of Art
design methods in Europe,” in report. Thomas Telford Services Ltd.,
International Symposium on 1994.
connections between steel and concrete, [13] fib, fib bulletin 58 - Design of
2001, pp. 45–60. anchorages in concrete, no. July. 2011.
[2] R. Eligehausen, R. Mallèe, and J. Silva, [14] R. E. Klingner and J. A. Mendonca,
“Tensile Capacity of Short Anchor

11
 G. Di Nunzio and G. Muciaccia

Bolts and Welded Studs: A Literature [24] J. Furche, “Zum Trag- und
Review,” ACI J., vol. 79, no. 27, 1982. Verschiebungsverhalten von
[15] R. Eligehausen and G. Sawade, Kopfbolzen bei zentrischem Zug,”
“Analysis of anchorage behaviour,” in University of Stuttgart, 1994.
Fracture Mechanics of Concrete [25] W. Berger, “Trag- und
Structures: From theory to applications, Verschiebungsverhalten sowie
L. Elfgren, Ed. London: Chapman and Bemessung von
Hall Ltd, 1989, pp. 263–280. Kopfbolzenverankerungen mit und ohne
[16] R. W. Cannon, E. G. Burdette, and R. Rückhängebewehrung unter Zuglast,”
R. Funk, “Anchorage to Concrete, University of Stuttgart, 2014.
Report No. CEB 75-32,” 1975. [26] J. Ozbolt, Y. J. Li, and I. Kozar,
[17] W. Fuchs, R. Eligehausen, and J. E. “Microplane Model for Concrete with
Breen, “Concrete Capacity Design Relaxed Kinematic Constraint,” Fract.
(CCD) Approach for Fastening to Mech. Concr. Struct., vol. 38, pp. 609–
Concrete,” ACI Struct. J., vol. 92, no. 1, 616, 2001.
p. 23, 1995. [27] R. Nilforoush, M. Nilsson, L. Elfgren, J.
[18] R. Ballarini, S. P. Shah, and L. M. Keer, Ožbolt, J. Hofmann, and R.
“Failure Characteristics of Short Anchor Eligehausen, “Tensile capacity of
Bolts Embedded in a Brittle Material,” anchor bolts in uncracked concrete:
in Proceedings of the Royal Society of Influence of member thickness and
London. Series A, Mathematical and anchor’s head size,” ACI Struct. J., vol.
Physical, 1986, vol. 404, no. 1826, pp. 114, no. 6, pp. 1519–1530, 2017.
35–54.
[19] L. Elfgren, U. Ohlsson, and K. Gylltoft,
“Anchor Bolts Analysed with Fracture
Mechanics,” in Fracture of Concrete
and Rock, New York, NY: Springer
New York, 1989, pp. 269–275.
[20] R. Eligehausen and G. Sawade, “A
fracture mechanics based description of
the pull-out behavior of headed studs
embedded in concrete,” Fract. Mech.
Concr. Struct., no. July, pp. 281–299,
1989.
[21] R. Eligehausen and J. Ozbolt, “Size
effect in anchorage behavior,” Fract.
Behav. Des. Mater. Struct., no. June,
1990.
[22] R. Eligehausen, P. Bouska, V.
Cervenka, and R. Pukl, “Size effect of
the concrete cone failure load of anchor
bolts,” First Int. Conf. Fract. Mech.
Concr. Struct., no. January, pp. 517–
525, 1992.
[23] K.-H. Lieberum, H.-W. Reinhardt, and
H. Weigler, “Das Tragverhalten von
Beton bei extremer
Teilflächenbelastung.,” Beton- und
Stahlbetonbau, vol. 84, no. 1, pp. 1–5,
Jan. 1989.

12