ArticlePDF Available

Experimental Exploration of Metal Cable as Reinforcement in 3D Printed Concrete

Authors:

Abstract and Figures

The Material Deposition Method (MDM) is enjoying increasing attention as an additive method to create concrete mortar structures characterised by a high degree of form-freedom, a lack of geometrical repetition, and automated construction. Several small-scale structures have been realised around the world, or are under preparation. However, the nature of this construction method is unsuitable for conventional reinforcement methods to achieve ductile failure behaviour. Sometimes, this is solved by combining printing with conventional casting and reinforcing techniques. This study, however, explores an alternative strategy, namely to directly entrain a metal cable in the concrete filament during printing to serve as reinforcement. A device is introduced to apply the reinforcement. Several options for online reinforcement media are compared for printability. Considerations specific to the manufacturing process are discussed. Subsequently, pull-out tests on cast and printed specimens provide an initial characterisation of bond behaviour. Bending tests furthermore show the potential of this reinforcement method. The bond stress of cables in printed concrete was comparable to values reported for smooth rebar but lower than that of the same cables in cast concrete. The scatter in experimental results was high. When sufficient bond length is available, ductile failure behaviour for tension parallel to the filament direction can be achieved, even though cable slip occurs. Further improvements to the process should pave the way to achieve better post-crack resistance, as the concept in itself is feasible.
Content may be subject to copyright.
materials
Article
Experimental Exploration of Metal Cable as
Reinforcement in 3D Printed Concrete
Freek P. Bos 1, *ID , Zeeshan Y. Ahmed 1, Evgeniy R. Jutinov 1and Theo A. M. Salet 1,2
1Department of the Built Environment, Eindhoven University of Technology, P.O. Box 513,
5600 MB Eindhoven, The Netherlands; z.y.ahmed@tue.nl (Z.Y.A.); evgeniy.jutinov@bam.com (E.R.J.);
t.a.m.salet@tue.nl (T.A.M.S.)
2Witteveen+Bos, P.O. Box 233, 7400 AE Deventer, The Netherlands
*Correspondence: f.p.bos@tue.nl; Tel.: +31-(0)40-247-2168
Received: 11 October 2017; Accepted: 8 November 2017; Published: 16 November 2017
Abstract:
The Material Deposition Method (MDM) is enjoying increasing attention as an additive
method to create concrete mortar structures characterised by a high degree of form-freedom, a lack
of geometrical repetition, and automated construction. Several small-scale structures have been
realised around the world, or are under preparation. However, the nature of this construction
method is unsuitable for conventional reinforcement methods to achieve ductile failure behaviour.
Sometimes, this is solved by combining printing with conventional casting and reinforcing techniques.
This study, however, explores an alternative strategy, namely to directly entrain a metal cable in
the concrete filament during printing to serve as reinforcement. A device is introduced to apply
the reinforcement. Several options for online reinforcement media are compared for printability.
Considerations specific to the manufacturing process are discussed. Subsequently, pull-out tests
on cast and printed specimens provide an initial characterisation of bond behaviour. Bending tests
furthermore show the potential of this reinforcement method. The bond stress of cables in printed
concrete was comparable to values reported for smooth rebar but lower than that of the same cables
in cast concrete. The scatter in experimental results was high. When sufficient bond length is
available, ductile failure behaviour for tension parallel to the filament direction can be achieved,
even though cable slip occurs. Further improvements to the process should pave the way to achieve
better post-crack resistance, as the concept in itself is feasible.
Keywords: 3D concrete printing; reinforcement; entrainment; cable; chain
1. Introduction
In the last few years, 3D printing in the construction industry has enjoyed rapid growth.
Additive manufacturing of concrete and cementitious materials (AMoC), in particular, is expanding
briskly. After a phase in which showcase objects have been presented that were not intended
for structural applications or actual load regimes (for instance: various objects [
1
], a free-shaped
bench [2], a children’s castle [3]), and exotic studies were initiated to explore additive manufacturing
for extra-terrestrial construction [
4
8
], we have now entered a phase of (announcements of) the first
case-study projects in actual use (a hotel extension [
9
], an office [
10
,
11
], a pedestrian bridge [
12
],
a laboratory [
13
], a bridge for bicycles and pedestrians [
14
], a house [
15
], a motor traffic bridge [
16
],
an office–hotel [
17
]). Meanwhile, Tay et al. [
18
] report a steep growth in academic publications in the
field since around 2013, not only in number but also in the topics that are being covered. Where these
were initially focused on the printing technique, later followed by material analysis, the range of topics
has now expanded to include architectural design, literature reviews, data analysis, and so on.
Materials 2017,10, 1314; doi:10.3390/ma10111314 www.mdpi.com/journal/materials
Materials 2017,10, 1314 2 of 22
A recurring issue that needs to be resolved when AMoC is used in structural applications is
the need to achieve ductility and (flexural) tensile capacity, as the processes that are being applied
generally do not yet provide that inherently [19].
In the limited number of projects that have been realised up to now, the strategy generally is to
use the printed concrete as lost formwork for conventional reinforced concrete [
20
,
21
]. Alternatively,
external pre-stressing tendons have been applied to obtain tensile capacity and ductility [
14
,
22
].
This strategy, which comes down to avoiding tensile stresses in the concrete, can also be applied in a
different way, namely by designing pure compression structures, such as a dome presented by [23].
An altogether different approach has been adopted by Chinese contractor HuaShang Tengda [
24
]
in a project realised in 2016. The reinforced concrete walls of this two-storey house were made by first
erecting the conventional reinforcement mesh and subsequently depositing concrete with a Material
Deposition Method (MDM)-type machine, with two large forked print nozzles that can simultaneously
print on each side of the reinforcement.
However, all these strategies have an important drawback: they seriously limit the level of
form-freedom and automation, which are the key selling points of 3D printing. Khoshnevis [
25
]
already recognised that an alternative to conventional reinforcement would be required, and could be
incorporated in an automated process without limiting shaping possibilities. Therefore, an embedded
coil was suggested that would not only provide longitudinal tensile strength, but also ductility though
the layer interfaces, as half of the coil sticks out of the preceding layer. Extensive results on the
structural performance of this alternative, however, were not reported.
Another possibility is to include fibers to achieve ductility in the print material itself. Hambach
and Volkmer [
26
] reported significantly improved tensile strengths in mortar samples of several
centimeters in size, reinforced with 1 vol % 3–6 mm carbon, glass, and basalt fibers. The 4TU Federation
website [27] shows that this option has been explored by others, too.
Hack and Lauer [
28
] presented the Mesh Mould method, an approach that bears similarities to the
one adopted by HuaShang Tengda in the sense that the reinforcement is constructed first and then used
as a carrier for the matrix material that is added in a subsequent production step. In the Mesh Mould
process, however, the reinforcement is also 3D-printed, thus eliminating the principal problems of
form-freedom and automation associated with the HuaShang Tengda process. Initially, the process was
developed to print a polymer reinforcement, but as this would have limited functionality
due to
the low
stiffness, the process was further developed to allow automated construction of metal reinforcement
as well [29].
One of the techniques that is being developed at the TU Eindhoven (TU/e) builds on the idea
of embedding a coil. Bos et al. [
30
] published the first results using entrained steel cable wires as
a reinforcement medium. The current paper presents a more comprehensive study of this concept,
intending to:
develop a suitable entrainment device and determine a suitable medium to act as reinforcement,
globally establish pull-out characteristics such as bond strength and anchorage length of this
medium in printed and cast concrete,
achieve ductile failure behaviour of printed concrete beams in bending,
explore any process-characteristic behaviour, aspects, and issues to be considered in further
development of this concept.
This paper will go on to describe the reinforcement entrainment device and discuss print process
considerations and reinforcement medium selection (Section 2). Some background theory is discussed
(Section 3). Then, in two separate sections, the experiments of this study are presented and discussed.
Section 4treats the tensile pull-out test in cast and printed concrete with which initial anchorage
characteristics are being established. Subsequently, Section 5elaborates on the design, manufacturing,
and testing of reinforced printed beam elements subjected to four-point bending to assess their
Materials 2017,10, 1314 3 of 22
structural performance. The study shows entrained steel cables are a feasible concept to achieve
ductility in the print filament direction in printed concrete.
2. Technique and Process of Entraining Reinforcement Cable in 3D Printed Concrete
The 3D concrete printing system adopted by the TU/e has been described extensively by
Bos et al. [31]
. It consists of an M-Tec Duomix 2000 mixer-pump with a linear displacement pump that
feeds concrete by a Ø 25 mm hose to a 9.0
×
4.5
×
2.8 m
3
4-DOF gantry robot (Figure 1). For this project,
the printer head, which previously consisted of a simple stainless steel print nozzle, was expanded
with a ‘reinforcement entraining device’ (RED) that allows the introduction of a reinforcement medium
to the concrete filament.
Materials 2017, 10, 1314 3 of 22
2. Technique and Process of Entraining Reinforcement Cable in 3D Printed Concrete
The 3D concrete printing system adopted by the TU/e has been described extensively by
Bos et al. [31]. It consists of an M-Tec Duomix 2000 mixer-pump with a linear displacement pump
that feeds concrete by a Ø 25 mm hose to a 9.0 × 4.5 × 2.8 m3 4-DOF gantry robot (Figure 1). For this
project, the printer head, which previously consisted of a simple stainless steel print nozzle,
was expanded with a ‘reinforcement entraining device’ (RED) that allows the introduction of a
reinforcement medium to the concrete filament.
Figure 1. 3D concrete printing facility at the TU/e.
The concept of the RED consists of a rotating spool feeding the reinforcement into the printing
head where it is introduced in the concrete filament so that an integrated concrete-with-reinforcement
filament leaves the print nozzle. This concept was developed into a working prototype following a
trial-and-error type process.
In order to succeed, it was clear from the onset that the reinforcement medium would need to
meet specific printability requirements, in addition to conventional reinforcing properties. Most
importantly, the cross section bending stiffness needed to be as low as possible. This high flexibility
was required both for the reinforcement to be able to pass through the print devices and to allow it
to follow all 3D freeform lines that can be produced with the concrete filament.
An active feed-system was opted for. A spring-loaded feeder driven by a stepper motor pulls
the reinforcement from a spool to direct it into the concrete filament. By incorporating a controllable
potentiometer, the rotation frequency can be adjusted to match the reinforcement deposition with the
movement speed of the print head so that it equals the length of the print path, straight or curved.
The spool used in this study can hold up to 100 m of cable wire. The resulting printer head, equipped
with a Reinforcement Entrainment Device (RED), is shown in Figure 2. Recently, an updated version
has been developed that can hold more than 2000 m of cable wire, divided over two spools.
Initial tests were performed with Ø 0.35 mm fishing line and Ø 0.4 mm steel wire. These proved
to have insufficient flexibility and to be too thin, causing them to cut through and pull out of the
concrete. Subsequent tests were done with several types of chain (Figure 3). The stepper-motor was
equipped with a wildcat chain wheel to feed the chain from the chain locker to the concrete filament,
a principle borrowed from the maritime industry to reel in heavy anchor chain. Using chain is
promising from a printability perspective: the chain could easily follow any print path. However, the
chain geometry and alternating link orientation was expected to have a detrimental effect on the
filament quality as the no-slump concrete is not able to properly fill all the gaps formed by the links.
Furthermore, the irregular geometry would probably lead to stress concentrations. These combined
effects would likely induce early concrete failure. Chain was thus abandoned as a reinforcement
alternative.
Finally, three different high-strength steel Bekaert Syncrocord® cables (Table 1) have been tested
for printability. These cables are specially developed to combine high tensile strength with great
Figure 1. 3D concrete printing facility at the TU/e.
The concept of the RED consists of a rotating spool feeding the reinforcement into the printing
head where it is introduced in the concrete filament so that an integrated concrete-with-reinforcement
filament leaves the print nozzle. This concept was developed into a working prototype following a
trial-and-error type process.
In order to succeed, it was clear from the onset that the reinforcement medium would need to meet
specific printability requirements, in addition to conventional reinforcing properties. Most importantly,
the cross section bending stiffness needed to be as low as possible. This high flexibility was required
both for the reinforcement to be able to pass through the print devices and to allow it to follow all 3D
freeform lines that can be produced with the concrete filament.
An active feed-system was opted for. A spring-loaded feeder driven by a stepper motor pulls
the reinforcement from a spool to direct it into the concrete filament. By incorporating a controllable
potentiometer, the rotation frequency can be adjusted to match the reinforcement deposition with the
movement speed of the print head so that it equals the length of the print path, straight or curved.
The spool used in this study can hold up to 100 m of cable wire. The resulting printer head, equipped
with a Reinforcement Entrainment Device (RED), is shown in Figure 2. Recently, an updated version
has been developed that can hold more than 2000 m of cable wire, divided over two spools.
Initial tests were performed with Ø 0.35 mm fishing line and Ø 0.4 mm steel wire. These proved to
have insufficient flexibility and to be too thin, causing them to cut through and pull out of the concrete.
Subsequent tests were done with several types of chain (Figure 3). The stepper-motor was equipped
with a wildcat chain wheel to feed the chain from the chain locker to the concrete filament, a principle
borrowed from the maritime industry to reel in heavy anchor chain. Using chain is promising from a
printability perspective: the chain could easily follow any print path. However, the chain geometry
and alternating link orientation was expected to have a detrimental effect on the filament quality
as the no-slump concrete is not able to properly fill all the gaps formed by the links. Furthermore,
the irregular geometry would probably lead to stress concentrations. These combined effects would
likely induce early concrete failure. Chain was thus abandoned as a reinforcement alternative.
Materials 2017,10, 1314 4 of 22
Materials 2017, 10, 1314 4 of 22
lateral flexibility and are normally used to reinforce synchronous belts. Compared to normal
reinforcement steel, their ductility is limited, as the ratio between failure strain and 0.2% offset yield
strain is between 1.22 and 1.44. As the cables are constructed out of sets of smaller strands, the actual
cable perimeter is larger than the product of the diameter times pi (p > π·d). The bond stresses in this
study are calculated based on the actual perimeter.
Figure 2. Active reinforcement entrainment device (RED) for cable reinforcement of printed concrete.
Figure 3. Early version of RED, equipped with chain reinforcement.
Table 1. Cables used as reinforcement.
Property Symbol Value
ID A B C
Name - Bekaert Syncrocord
Force 0.6
Bekaert Syncrocord
Flex 0.9
Bekaert Syncrocord
Flex 1.2
Coating - Galvanised Galvanised Galvanised
Diamete
r
d [mm]
0.63 0.97 1.20
Perimeter p [mm]
2.24 5.11 5.87
Linear Density ρ
lin
[g/m] 2.30 3.46 5.84
0.2% offset yield strain ε
0.2
1.61% 1.72% 1.90%
0.2% offset yield stress f
0.2
[N] 381 1140 1800
Characteristic ultimate
tensile strain ε
uk
2.32% 2.10% 2.40%
Characteristic ultimate
tensile strength f
uk
[N] 420 1190 1925
Axial Tensile Modulus of
Elasticity E
axi
[GPa] 181.6 178.3 156.8
Figure 2.
Active reinforcement entrainment device (RED) for cable reinforcement of printed concrete.
Materials 2017, 10, 1314 4 of 22
lateral flexibility and are normally used to reinforce synchronous belts. Compared to normal
reinforcement steel, their ductility is limited, as the ratio between failure strain and 0.2% offset yield
strain is between 1.22 and 1.44. As the cables are constructed out of sets of smaller strands, the actual
cable perimeter is larger than the product of the diameter times pi (p > π·d). The bond stresses in this
study are calculated based on the actual perimeter.
Figure 2. Active reinforcement entrainment device (RED) for cable reinforcement of printed concrete.
Figure 3. Early version of RED, equipped with chain reinforcement.
Table 1. Cables used as reinforcement.
Property Symbol Value
ID A B C
Name - Bekaert Syncrocord
Force 0.6
Bekaert Syncrocord
Flex 0.9
Bekaert Syncrocord
Flex 1.2
Coating - Galvanised Galvanised Galvanised
Diamete
r
d [mm]
0.63 0.97 1.20
Perimeter p [mm]
2.24 5.11 5.87
Linear Density ρ
lin
[g/m] 2.30 3.46 5.84
0.2% offset yield strain ε
0.2
1.61% 1.72% 1.90%
0.2% offset yield stress f
0.2
[N] 381 1140 1800
Characteristic ultimate
tensile strain ε
uk
2.32% 2.10% 2.40%
Characteristic ultimate
tensile strength f
uk
[N] 420 1190 1925
Axial Tensile Modulus of
Elasticity E
axi
[GPa] 181.6 178.3 156.8
Figure 3. Early version of RED, equipped with chain reinforcement.
Finally, three different high-strength steel Bekaert Syncrocord
®
cables (Table 1) have been tested
for printability. These cables are specially developed to combine high tensile strength with great lateral
flexibility and are normally used to reinforce synchronous belts. Compared to normal reinforcement
steel, their ductility is limited, as the ratio between failure strain and 0.2% offset yield strain is between
1.22 and 1.44. As the cables are constructed out of sets of smaller strands, the actual cable perimeter is
larger than the product of the diameter times pi (p >
π·
d). The bond stresses in this study are calculated
based on the actual perimeter.
Table 1. Cables used as reinforcement.
Property Symbol Value
ID A B C
Name -
Bekaert Syncrocord
Force 0.6
Bekaert Syncrocord
Flex 0.9
Bekaert Syncrocord
Flex 1.2
Coating - Galvanised Galvanised Galvanised
Diameter d [mm] 0.63 0.97 1.20
Perimeter p [mm] 2.24 5.11 5.87
Linear Density ρlin [g/m] 2.30 3.46 5.84
0.2% offset yield strain ε0.2 1.61% 1.72% 1.90%
0.2% offset yield stress f0.2 [N] 381 1140 1800
Characteristic ultimate
tensile strain εuk 2.32% 2.10% 2.40%
Characteristic ultimate
tensile strength fuk [N] 420 1190 1925
Axial Tensile Modulus of
Elasticity Eaxi [GPa] 181.6 178.3 156.8
Materials 2017,10, 1314 5 of 22
Using the active feed system, it proved to be possible to print any curvature allowed by the process,
with all three cables. Therefore, further research was conducted to establish bond characteristics and
the capacity to act as a reinforcement.
3. Cable Reinforcement
To be able to analyse the structural behaviour of steel cables as reinforcement and establish
a method to calculate their performance, it is essential to establish their pull-out behaviour.
Cables are commonly used as anchoring in grout and concrete in tunneling and mining construction.
Geometrically and mechanically, cable bolts are not completely comparable to continuous cables as
they feature a bulb near the end, which significantly increases their bearing capacity. Furthermore,
with tensile strengths in the order of magnitude of several hundred kN, they are more than 100 times
stronger than the cables considered in this study. Their diameters are also much larger, in the order
of 2.5–3.5 cm. Hagan et al. [
32
] note that the influence of critical parameters such as rock mass
confinement, cable surface geometry, water:cement ratio, and embedment length on the pull-out
resistance are well understood, but that a common testing methodology still needs to be developed
for a proper comparison between available cable bolt systems. Chen et al. [
33
] recently found that
in an unconfined condition, the pull-out resistance is proportional to the cable diameter (and thus
to the cable perimeter), whereas under confined boundary conditions (i.e., in which radial pressure
can develop), this is only partially the case. They also note that the failure mode between confined
and unconfined conditions is different (as would be expected): in the former it is characterised by
debonding, whereas in the latter failure occurs through sample split. Stress-slippage models have
also been developed, such as the tri-linear model proposed by [
34
]. However, due to the intended
function of the cables as reinforcement and the geometrical and mechanical deviations mentioned,
their performance is preferable in comparison to conventional bar reinforcement.
The resistance to pull-out behaviour of conventional ribbed reinforcement bars out of NSC
(Normal Strength Concrete) may be understood as consisting of a complex combination of at least three
phenomena: adhesion, dilatancy, and friction, with the former two, adhesion and dilatancy, occurring
before failure and together constituting the bond resistance, and the latter happening after failure
and determining the post-failure resistance [
35
]. The resistance of these three components depends
on a number of parameters such as global stress distribution, concrete quality, compaction or bond
quality around the reinforcement, and level of confinement [
36
,
37
]. Commonly applied models for
bond strength, however, are usually primarily based on the concrete quality, with some additional
parameters accounting for various conditions. For instance, Eurocode 2 (EC2) [38] defines the design
bond strength as:
fbd =2.25η1η2fctd (1)
where
η1
depends on the embedment quality
η2
on the bar diameter, while f
ctd
is the design tensile
strength of the concrete.
The conditions of the concrete applied in 3D concrete printing are rather different from normal cast
concrete. Up to and including the current project, the 3DCP research program has used a mortar with
strength properties approximately equal to C10/12 grade concrete, but with a maximum grain size of
1 mm, i.e., significantly lower than is usually applied in structural concrete in buildings. Furthermore,
the density of printed concrete is lower (
ρ
= 2000
±
50 kg/m
3
) than NSC because the material cannot
be compacted after printing.
The EC2 equation for bond stress, however, assumes a number of properties of the reinforcement
that are not necessarily valid for the applied cables.
First, the stress-strain behaviour of these cables differs from normal reinforcement steel. According
to Eurocode 2, B-grade reinforcement steel, which is commonly used for instance in the Netherlands,
yields at
εyk
= 0.25%, and fails at
εuk
> 5.0%, whereas for the applied cables approximately
ε0.2k = 1.6–1.9%
and
εuk
= 2.1–2.4%. In other words, the linear elastic strength limit is much higher,
and the stiffness is considerably lower than in conventional reinforcement bars (as given by the axial
Materials 2017,10, 1314 6 of 22
modulus of elasticity between 157 and 182 GPa, compared to 200 GPa for reinforcement steel bars).
Besides the steel quality, this is due to the woven structure of the cable.
Furthermore, the cable surface is smooth. Thus, their pull-out behaviour is likely more comparable
to the smooth reinforcement bars that have been used in concrete structures in the past. Since the
1970s, the use of smooth bars has been discontinued in Europe in favour of ribbed bars because of
their far superior bond properties caused by the much higher dilatancy resistance. The bond resistance
of ribbed bars is found to be approximately 6.6 times as high as for smooth bars [
39
]. This, naturally,
also influences
the proportion between adhesive and dilatancy resistance. Since adhesion is found
to cause 60% of the bond strength of smooth bars [
40
], it would only be about 10% for ribbed bars
(assuming the bar geometry does not change the concrete-steel adhesion). Experimental research shows
bond strengths for smooth rebars of between 1.5 and 2.5 MPa [
41
,
42
], which globally corresponds with
the results of [39].
Contrary to conventional reinforcement, the cables applied in this research are galvanised (i.e.,
zinc-coated). Galvanising reinforcement is a technique that has been commonly applied to increase the
corrosion resistance of the reinforcement (and thus the structural durability) since the 1930s, especially
in saline environments [
43
]. Galvanised reinforcement steel is governed by various codes, such as
ISO 14657 [
44
] and the ASTM A767 [
45
]. When galvanised steel comes into contact with uncured
concrete, a passivation reaction occurs that consumes about 10
µ
m of zinc [
43
]. However, several
studies conclude that for all general structural and construction purposes, galvanised coating can
be treated as normal reinforcement, as it performance is equal or better [
46
,
47
]. Hamid & Mike [
48
]
refer to contradictory results in the literature, but conclude galvanisation has no significant impact
on bonding in NSC. A dissenting opinion is voiced by Pernicova et al. [
49
] who found that the
formation of hydrogen during the passivation reaction causes porosity in the surrounding cement
thereby deteriorating the bond strength. The severity of this effect was found to be dependent on the
pH value of the concrete.
Finally, common bond stress methods assume the concrete matrix can be considered a
homogenous material. Applicability often features limits, such as minimal reinforcement diameters
and maximum embedment lengths, e.g., 5 times the diameter (5
·
d [mm]) for the pull out tests of
NEN-EN 100080 [
50
]. However, the applied cables are much thinner while on the other hand the
embedment length of 5
·
d can practically not be achieved in testing, and would also result in a
disproportionate influence of discrete effects such as aggregates.
In many ways, the pull-out behaviour of the cables is likely comparable to that of straight
steel fibres in terms of proportions between adhesion, dilatancy and friction, considering their
small diameter and smooth surface. The parameters that effect these values are mostly similar
to those effecting conventional reinforcement, and include concrete matrix composition and strength,
water:cement ratio, fiber geometry, length and orientation, fiber surface treatment, and load rate [
51
55
].
However, even though the pull-out behaviour may be similar, the parameters of interest are usually
principally different. For FRC (Fiber Reinforced Concrete), ductility is based the actual pull-out of the
fibers (which results in the total debonding energy from adhesion, dilatancy, friction and, depending
on the fiber geometry, plastic fiber deformation) therefore the total pull-out energy is an important
parameter. In conventional reinforced concrete, on the other hand, ductility is provided by plastic
deformation of the rebars while they remain bonded in the concrete on either side of a crack. Hence,
the maximum pull-out strength and anchorage length are the primary parameters to determine.
It may be concluded that existing experimental data and models for reinforcement behaviour
provide a frame of reference to analyse the cable reinforcement behaviour in 3DCP. However,
with regard to geometry, material behaviour of both concrete and steel, as well as the manufacturing
method, significant differences exist that call for caution when interpreting results.
Materials 2017,10, 1314 7 of 22
4. Pull-Out Test
The bond behaviour of three types of cable reinforcement was investigated by pull-out tests on
cast and printed concrete samples with different embedment lengths.
4.1. Method
4.1.1. Specimen Preparation-Cast
For the pull-out tests on cast concrete specimens, rectangular blocks were cast around cables,
that were slightly pre-stressed in order to guarantee their straightness, and compacted on a vibrating
table for 10 s. The same 3DCP print mortar, described by [
31
], was used for the cast and the printed
specimens. Unpublished test results on compressive strength, tensile strength, and modulus of
elasticity have shown these properties to be in the range of C20/25 when cast and printed, loaded in
compression, but more comparable to C10/12 or C12/15 when printed, loaded in tension. It should
be noted, however, that the material is not completely comparable to conventional concrete, among
others because the maximum particle size is only 1 mm.
After casting, the specimens were wrapped in foil to prevent dehydration, and left to cure for
14 days at ambient lab temperature. Several series of specimens were produced, with three different
cables (A, B, and C—see Table 1) and 2 different embedment lengths l
cs
= 15 and 35 mm. The results of
an initial test series with l
cs
= 25 mm were abandoned because of deviating concrete quality, which led
to incomparable results. Throughout the experimental part of the research, each series consisted of
five specimens. Consequently, 30 cast specimens were prepared and tested.
4.1.2. Specimen Preparation—Printed
The printed specimens were taken from three subsequently printed objects (Figure 4a). In each
object, the reinforcement cable (A, B, and C respectively) is in the bottom layer (Figure 4b). Still in
the wet state, the printed concrete filament was cut with a custom designed U-shape device to obtain
each specimen (Figure 4c). The excess material on each side of the cut was removed in order to
obtain protruding cables that could be used to apply loading and measurement equipment. Besides
three different cables, specimens of three different embedment lengths l
cs
= 15, 25 and 35 mm were
produced (resulting in a total of 45 specimens). The specimens were left to cure under foil on the
print table for one day, and then submerged in water until testing at 14 days old. Table 2provides an
overview of the pull-out test specimen series.
Materials 2017, 10, 1314 7 of 22
4.1. Method
4.1.1. Specimen Preparation-Cast
For the pull-out tests on cast concrete specimens, rectangular blocks were cast around cables,
that were slightly pre-stressed in order to guarantee their straightness, and compacted on a vibrating
table for 10 s. The same 3DCP print mortar, described by [31], was used for the cast and the printed
specimens. Unpublished test results on compressive strength, tensile strength, and modulus of
elasticity have shown these properties to be in the range of C20/25 when cast and printed, loaded in
compression, but more comparable to C10/12 or C12/15 when printed, loaded in tension. It should be
noted, however, that the material is not completely comparable to conventional concrete, among
others because the maximum particle size is only 1 mm.
After casting, the specimens were wrapped in foil to prevent dehydration, and left to cure for
14 days at ambient lab temperature. Several series of specimens were produced, with three different
cables (A, B, and C—see Table 1) and 2 different embedment lengths lcs = 15 and 35 mm. The results
of an initial test series with lcs = 25 mm were abandoned because of deviating concrete quality, which
led to incomparable results. Throughout the experimental part of the research, each series consisted
of five specimens. Consequently, 30 cast specimens were prepared and tested.
4.1.2. Specimen Preparation—Printed
The printed specimens were taken from three subsequently printed objects (Figure 4a). In each
object, the reinforcement cable (A, B, and C respectively) is in the bottom layer (Figure 4b). Still in the
wet state, the printed concrete filament was cut with a custom designed U-shape device to obtain
each specimen (Figure 4c). The excess material on each side of the cut was removed in order to obtain
protruding cables that could be used to apply loading and measurement equipment. Besides three
different cables, specimens of three different embedment lengths lcs = 15, 25 and 35 mm were
produced (resulting in a total of 45 specimens). The specimens were left to cure under foil on the print
table for one day, and then submerged in water until testing at 14 days old. Table 2 provides an
overview of the pull-out test specimen series.
(a)
(b) (c)
Figure 4. (a) Print path; (b) and transverse section of printed objects from which the respective
specimens were obtained; (c) cutting and stripping of printed specimens.
Figure 4.
(
a
) Print path; (
b
) and transverse section of printed objects from which the respective
specimens were obtained; (c) cutting and stripping of printed specimens.
Materials 2017,10, 1314 8 of 22
Table 2. Specimen series overview.
Series No. of Specimens Concrete Manufacturing Cable lcs [mm]
C15A 5 Cast A 15
C15B 5 Cast B 15
C15C 5 Cast C 15
C35A 5 Cast A 35
C35B 5 Cast B 35
C35C 5 Cast C 35
P15A 5 Printed A 15
P15B 5 Printed B 15
P15C 5 Printed C 15
P25A 5 Printed A 25
P25B 5 Printed B 25
P25C 5 Printed C 25
P35A 5 Printed A 35
P35B 5 Printed B 35
P35C 5 Printed C 35
4.1.3. Experimental Set-Up
The specimens were subjected to a displacement controlled pull-out test at 0.5 mm/min,
performed in an Instron universal test rig equipped with a 5 kN load cell. The cable slip was recorded
by averaging measurement results of 2 LVDT’s fitted to the cables at the bottom side of the specimen
as shown in Figure 5a–d. The test rig grip was positioned as close to the specimen as possible and its
displacement was also recorded.
Materials 2017, 10, 1314 8 of 22
Table 2. Specimen series overview.
Series No. of Specimens Concrete Manufacturing Cable lcs [mm]
C15A 5 Cast A 15
C15B 5 Cast B 15
C15C 5 Cast C 15
C35A 5 Cast A 35
C35B 5 Cast B 35
C35C 5 Cast C 35
P15A 5 Printed A 15
P15B 5 Printed B 15
P15C 5 Printed C 15
P25A 5 Printed A 25
P25B 5 Printed B 25
P25C 5 Printed C 25
P35A 5 Printed A 35
P35B 5 Printed B 35
P35C 5 Printed C 35
4.1.3. Experimental Set-Up
The specimens were subjected to a displacement controlled pull-out test at 0.5 mm/min,
performed in an Instron universal test rig equipped with a 5 kN load cell. The cable slip was recorded
by averaging measurement results of 2 LVDT’s fitted to the cables at the bottom side of the specimen
as shown in Figure 5a–d. The test rig grip was positioned as close to the specimen as possible and its
displacement was also recorded.
(a) (b)
Figure 5. Cont.
Materials 2017,10, 1314 9 of 22
Materials 2017, 10, 1314 9 of 22
(c)
(d)
Figure 5. Test set-up for the cast (a,b) and printed (c,d) specimens.
4.2. Results
Figures 6a,b and 7a–c show load-slip curves per embedment length of representative specimens
from each series (chosen to illustrate their typical performance) of the cast and printed specimens
respectively. Of each specimen series, the adhesive bond strength (taken as the end of the initial linear
force-displacement path; F
adh
, δ
adh
) and the ultimate bond strength (F
u
, δ
u
) are listed in Table 3, the
former assumed to be determined solely by the cable-matrix adhesion, and the latter by a combination
of adhesive bond, dilatancy, and/or friction. The corresponding shear stresses have been calculated
from the cable perimeters listed in Table 1. These are average shear stresses, i.e., τ
u
= F
u
/pl. Although
it is known that the shear stress distribution will not be distributed equally along the cable length,
but rather peak at the top side where the load is introduced, this approach is commonly applied in
reinforcement-matrix pull-out analyses and therefore maintained here as well. All specimens in both
the cast and printed series failed on cable pull-out, except for the C35A series, in which gradual
breakage of individual strands within the cable introduces failure.
(a)
Figure 5. Test set-up for the cast (a,b) and printed (c,d) specimens.
4.2. Results
Figures 6a,b and 7a–c show load-slip curves per embedment length of representative specimens
from each series (chosen to illustrate their typical performance) of the cast and printed specimens
respectively. Of each specimen series, the adhesive bond strength (taken as the end of the initial
linear force-displacement path; F
adh
,
δadh
) and the ultimate bond strength (F
u
,
δu
) are listed in
Table 3, the former assumed to be determined solely by the cable-matrix adhesion, and the latter
by a combination of adhesive bond, dilatancy, and/or friction. The corresponding shear stresses
have been calculated from the cable perimeters listed in Table 1. These are average shear stresses, i.e.,
τu
= F
u
/pl. Although it is known that the shear stress distribution will not be distributed equally
along the cable length, but rather peak at the top side where the load is introduced, this approach is
commonly applied in reinforcement-matrix pull-out analyses and therefore maintained here as well.
All specimens in both the cast and printed series failed on cable pull-out, except for the C35A series,
in which gradual breakage of individual strands within the cable introduces failure.
Materials 2017, 10, 1314 9 of 22
(c)
(d)
Figure 5. Test set-up for the cast (a,b) and printed (c,d) specimens.
4.2. Results
Figures 6a,b and 7a–c show load-slip curves per embedment length of representative specimens
from each series (chosen to illustrate their typical performance) of the cast and printed specimens
respectively. Of each specimen series, the adhesive bond strength (taken as the end of the initial linear
force-displacement path; F
adh
, δ
adh
) and the ultimate bond strength (F
u
, δ
u
) are listed in Table 3, the
former assumed to be determined solely by the cable-matrix adhesion, and the latter by a combination
of adhesive bond, dilatancy, and/or friction. The corresponding shear stresses have been calculated
from the cable perimeters listed in Table 1. These are average shear stresses, i.e., τ
u
= F
u
/pl. Although
it is known that the shear stress distribution will not be distributed equally along the cable length,
but rather peak at the top side where the load is introduced, this approach is commonly applied in
reinforcement-matrix pull-out analyses and therefore maintained here as well. All specimens in both
the cast and printed series failed on cable pull-out, except for the C35A series, in which gradual
breakage of individual strands within the cable introduces failure.
(a)
Figure 6. Cont.
Materials 2017,10, 1314 10 of 22
Materials 2017, 10, 1314 10 of 22
(b)
Figure 6. (a) Load-slip curves of pull-out tests of cast samples with l
cs
= 15 mm. For each series, one
curve is shown (for reasons of clarity), considered representative for that particular series. It should
be noted that the individual behaviour can differ significantly, as is also clear from the coefficients of
variation. (b) Load-slip curves of pull-out tests of representative cast samples with l
cs
= 35 mm.
(a)
Figure 6.
(
a
) Load-slip curves of pull-out tests of cast samples with l
cs
= 15 mm. For each series,
one curve is shown (for reasons of clarity), considered representative for that particular series. It should
be noted that the individual behaviour can differ significantly, as is also clear from the coefficients of
variation. (b) Load-slip curves of pull-out tests of representative cast samples with lcs = 35 mm.
Materials 2017, 10, 1314 10 of 22
(b)
Figure 6. (a) Load-slip curves of pull-out tests of cast samples with l
cs
= 15 mm. For each series, one
curve is shown (for reasons of clarity), considered representative for that particular series. It should
be noted that the individual behaviour can differ significantly, as is also clear from the coefficients of
variation. (b) Load-slip curves of pull-out tests of representative cast samples with l
cs
= 35 mm.
(a)
Figure 7. Cont.
Materials 2017,10, 1314 11 of 22
Materials 2017, 10, 1314 11 of 22
(b)
(c)
Figure 7. (a) Load-slip curves of pull-out tests of printed samples with l
cs
= 15 mm. For each series,
one curve is shown (for reasons of clarity), considered representative for that particular series.
It should be noted that the individual behaviour can differ significantly, as is also clear from the
coefficients of variation. (b) Load-slip curves of pull-out tests of representative printed samples with
l
cs
= 25 mm; (c) Lload-slip curves of pull-out tests of representative printed samples with l
cs
= 35 mm.
Figure 7.
(
a
) Load-slip curves of pull-out tests of printed samples with l
cs
= 15 mm. For each series, one
curve is shown (for reasons of clarity), considered representative for that particular series. It should
be noted that the individual behaviour can differ significantly, as is also clear from the coefficients of
variation. (
b
) Load-slip curves of pull-out tests of representative printed samples with l
cs
= 25 mm;
(c) Lload-slip curves of pull-out tests of representative printed samples with lcs = 35 mm.
Materials 2017,10, 1314 12 of 22
Table 3. Specimen series results.
Series Fadh [N]; CoV τadh [MPa] Fu[N]; CoV τu[MPa] τadh/τu
C15A 139.03; 17% 4.15 238.83; 27% 7.12 0.58
C15B 330.92; 10% 4.31 417.50; 10% 5.44 0.79
C15C 327.12; 17% 3.72 396.95; 14% 4.51 0.82
C35A 245.30; 6% 3.13 347.76; 9% 4.44 0.70
C35B 688.66; 10% 3.85 826.44; 11% 4.62 0.83
C35C 505.08; 28% 2.46 995.87; 18% 4.85 0.51
P15A 48.62; 17% 1.45 79.57; 22% 2.37 0.61
P15B 80.80; 30% 1.05 102.02; 29% 1.33 0.79
P15C 100.54; 24% 1.14 113.62; 39% 1.29 0.88
P25A 83.41; 19% 1.49 173.24; 31% 3.10 0.48
P25B 142.67; 34% 1.12 177.49; 38% 1.39 0.81
P25C 188.45; 13% 1.28 321.70; 36% 2.19 0.58
P35A 143.20; 18% 1.83 242.40; 27% 3.10 0.59
P35B 163.40; 11% 0.91 253.88; 8% 1.42 0.64
P35C 354.74; 4% 1.73 409.63; 6% 1.99 0.87
Figure 8shows a typical cast sample after testing. In Figure 9, a part of the printed object is shown,
transversely broken open after curing. This was not a specimen, but rather a part of the object not
used for the samples. It shows that the irregularities (voids, bubbles, etc.) around the cable positioning
are much more severe than in the cast specimens. This difference in matrix quality is caused by the
printing process, which, unlike the casting process, does not include compaction on a vibrating table.
Materials 2017, 10, 1314 12 of 22
Table 3. Specimen series results.
Series Fadh [N]; CoV τadh [MPa] Fu[N]; CoV τu[MPa] τadh
/
τu
C15A 139.03; 17% 4.15 238.83; 27% 7.12 0.58
C15B 330.92; 10% 4.31 417.50; 10% 5.44 0.79
C15C 327.12; 17% 3.72 396.95; 14% 4.51 0.82
C35A 245.30; 6% 3.13 347.76; 9% 4.44 0.70
C35B 688.66; 10% 3.85 826.44; 11% 4.62 0.83
C35C 505.08; 28% 2.46 995.87; 18% 4.85 0.51
P15A 48.62; 17% 1.45 79.57; 22% 2.37 0.61
P15B 80.80; 30% 1.05 102.02; 29% 1.33 0.79
P15C 100.54; 24% 1.14 113.62; 39% 1.29 0.88
P25A 83.41; 19% 1.49 173.24; 31% 3.10 0.48
P25B 142.67; 34% 1.12 177.49; 38% 1.39 0.81
P25C 188.45; 13% 1.28 321.70; 36% 2.19 0.58
P35A 143.20; 18% 1.83 242.40; 27% 3.10 0.59
P35B 163.40; 11% 0.91 253.88; 8% 1.42 0.64
P35C 354.74; 4% 1.73 409.63; 6% 1.99 0.87
Figure 8 shows a typical cast sample after testing. In Figure 9, a part of the printed object is
shown, transversely broken open after curing. This was not a specimen, but rather a part of the object
not used for the samples. It shows that the irregularities (voids, bubbles, etc.) around the cable
positioning are much more severe than in the cast specimens. This difference in matrix quality is
caused by the printing process, which, unlike the casting process, does not include compaction on a
vibrating table.
Figure 8. Representative cast specimen after testing.
Figure 9. Transversally broken part of print object (not specimen) to show voids around cable shaft.
Figure 8. Representative cast specimen after testing.
Materials 2017, 10, 1314 12 of 22
Table 3. Specimen series results.
Series Fadh [N]; CoV τadh [MPa] Fu[N]; CoV τu[MPa] τadh
/
τu
C15A 139.03; 17% 4.15 238.83; 27% 7.12 0.58
C15B 330.92; 10% 4.31 417.50; 10% 5.44 0.79
C15C 327.12; 17% 3.72 396.95; 14% 4.51 0.82
C35A 245.30; 6% 3.13 347.76; 9% 4.44 0.70
C35B 688.66; 10% 3.85 826.44; 11% 4.62 0.83
C35C 505.08; 28% 2.46 995.87; 18% 4.85 0.51
P15A 48.62; 17% 1.45 79.57; 22% 2.37 0.61
P15B 80.80; 30% 1.05 102.02; 29% 1.33 0.79
P15C 100.54; 24% 1.14 113.62; 39% 1.29 0.88
P25A 83.41; 19% 1.49 173.24; 31% 3.10 0.48
P25B 142.67; 34% 1.12 177.49; 38% 1.39 0.81
P25C 188.45; 13% 1.28 321.70; 36% 2.19 0.58
P35A 143.20; 18% 1.83 242.40; 27% 3.10 0.59
P35B 163.40; 11% 0.91 253.88; 8% 1.42 0.64
P35C 354.74; 4% 1.73 409.63; 6% 1.99 0.87
Figure 8 shows a typical cast sample after testing. In Figure 9, a part of the printed object is
shown, transversely broken open after curing. This was not a specimen, but rather a part of the object
not used for the samples. It shows that the irregularities (voids, bubbles, etc.) around the cable
positioning are much more severe than in the cast specimens. This difference in matrix quality is
caused by the printing process, which, unlike the casting process, does not include compaction on a
vibrating table.
Figure 8. Representative cast specimen after testing.
Figure 9. Transversally broken part of print object (not specimen) to show voids around cable shaft.
Figure 9. Transversally broken part of print object (not specimen) to show voids around cable shaft.
Materials 2017,10, 1314 13 of 22
Figure 10 shows the development of average adhesive bond stress over embedment length for
cast and printed specimens. In Figure 11, the average adhesive bond stresses are compared.
Materials 2017, 10, 1314 13 of 22
Figure 10 shows the development of average adhesive bond stress over embedment length for
cast and printed specimens. In Figure 11, the average adhesive bond stresses are compared.
Figure 10. Development of average adhesive bond stress over increasing embedment length.
Figure 11. Comparison of average ultimate bond stress in cast and printed specimen.
As failure in all printed specimens occurs through cable pull-out and the results show no clear
trend in bond strength development over embedment length, basic anchorage lengths can be
calculated from the cable data (Table 1) and the bond stress, averaged per cable type over the three
investigated embedment lengths: l
anchorage
= F/τ
max
p. These are listed in Table 4. Anchorage lengths
have been calculated based both on adhesive and ultimate bond stress.
Table 4. Basic anchorage lengths for applied cables in printed concrete, based on adhesive (l
anchorage,adh
)
and ultimate (l
anchorage,u
) bond stress.
Cable Type F
u
[N] P [mm] τ
max,ave 15/25/35
[MPa] l
anchorage,adh
[mm] l
anchorage,u
[mm]
A 420 2.24 2.86 117.9 65.6
B 1190 5.11 1.38 217.3 161.7
C 1925 5.87 1.82 237.1 179.9
Figure 10. Development of average adhesive bond stress over increasing embedment length.
Materials 2017, 10, 1314 13 of 22
Figure 10 shows the development of average adhesive bond stress over embedment length for
cast and printed specimens. In Figure 11, the average adhesive bond stresses are compared.
Figure 10. Development of average adhesive bond stress over increasing embedment length.
Figure 11. Comparison of average ultimate bond stress in cast and printed specimen.
As failure in all printed specimens occurs through cable pull-out and the results show no clear
trend in bond strength development over embedment length, basic anchorage lengths can be
calculated from the cable data (Table 1) and the bond stress, averaged per cable type over the three
investigated embedment lengths: l
anchorage
= F/τ
max
p. These are listed in Table 4. Anchorage lengths
have been calculated based both on adhesive and ultimate bond stress.
Table 4. Basic anchorage lengths for applied cables in printed concrete, based on adhesive (l
anchorage,adh
)
and ultimate (l
anchorage,u
) bond stress.
Cable Type F
u
[N] P [mm] τ
max,ave 15/25/35
[MPa] l
anchorage,adh
[mm] l
anchorage,u
[mm]
A 420 2.24 2.86 117.9 65.6
B 1190 5.11 1.38 217.3 161.7
C 1925 5.87 1.82 237.1 179.9
Figure 11. Comparison of average ultimate bond stress in cast and printed specimen.
As failure in all printed specimens occurs through cable pull-out and the results show no clear
trend in bond strength development over embedment length, basic anchorage lengths can be calculated
from the cable data (Table 1) and the bond stress, averaged per cable type over the three investigated
embedment lengths: l
anchorage
= F/
τmax
p. These are listed in Table 4. Anchorage lengths have been
calculated based both on adhesive and ultimate bond stress.
Table 4.
Basic anchorage lengths for applied cables in printed concrete, based on adhesive (l
anchorage,adh
)
and ultimate (lanchorage,u) bond stress.
Cable Type Fu[N] P [mm] τmax,ave 15/25/35 [MPa] lanchorage,adh [mm] lanchorage,u [mm]
A 420 2.24 2.86 117.9 65.6
B 1190 5.11 1.38 217.3 161.7
C 1925 5.87 1.82 237.1 179.9
Materials 2017,10, 1314 14 of 22
4.3. Discussion
4.3.1. Cast Specimens
The load-slip curves of the cast specimens show a clear steep branch representing the adhesive
bond (the displacement of the cable at the bottom side of the specimens is almost 0), and a consecutive
branch with partial bonding and dilatancy. After the maximum bond strength is reached, a gradual
degrading branch starts caused by friction. It should be noted that the adhesive branch for the C cables
is less steep. This is caused by the structure of that cable, with a core and a mantle.
In neither of the cast specimen series does the embedded length influence the ultimate bond
stress. Although at first glance there seems to be an influence in the A-cable specimens (Table 3),
this difference is caused by a change in failure mechanism (pull-out versus gradual strand breakage in
the cable), not by a change in cable-to-matrix bond strength.
The adhesive bond stress on the other hand, seems to decrease slightly with increasing embedded
length (Figure 10). This might be explained from the real stress distribution. The difference in adhesive
stress on the load side of the cable and the non-load side, will be higher for longer embedment lengths,
resulting in a decreased average stress. However, the effect is minor over the investigated length,
whereas the scatter is considerable, and should therefore be treated cautiously.
The ratio
τadh
/
τu
ranges from 0.6 to 0.8. As discussed previously, this is comparable to values
found for smooth rebar, where the adhesive bond accounts for around 60% of the ultimate bond
strength. In absolute terms, the maximum bond stress is somewhat higher than expected for
smooth rebar.
The differences in adhesive and maximum bond stresses between the series are considerable and
do not show a consequent pattern. An explanation, therefore, remains elusive. The scatter in results is
also substantial, but does not seem to depend on embedded length or cable type.
4.3.2. Printed Specimens
Like the cast specimens, the printed specimens show a clear adhesive branch that is less steep with
the C cables. After the adhesive strength is reached though, a drop in load occurs in about 40% of the
specimens, indicating a sudden release of the cable that is less counteracted by dilatancy. The inferior
concrete matrix quality, as shown in Figure 9, is likely the culprit. In the other cases, the behaviour
after the adhesive branch is comparable to that of the cast specimens.
Contrary to the cast specimens, no clear trend (decreasing or increasing) in the adhesive bond
stress over increasing embedment length can be noticed. Again, the reduced matrix quality is likely to
blame: any increase in shorter embedment lengths due to an altered stress distribution is annulled by
a relatively increasing effect of matrix defects.
The
τadh
/
τu
ratio is comparable to the cast specimens, but can run to slightly higher values from
0.6 to 0.9 (in the latter case the contribution of dilatancy to the overall bond resistance has become
almost negligible). Both the adhesive and ultimate bond stress are significantly lower than for the cast
specimens—again caused by the inferior matrix quality. Nevertheless, the ultimate bond strength is
still close to what could be expected of smooth rebar. The scatter in results for cable type A is clearly
larger than for cable types B and C. The reason for this is unclear.
4.3.3. Previous Research
Finally, it is noted that the anchorage lengths calculated in Table 4can, globally, be matched with
the experimental results presented by [
30
] of reinforced printed beams, loaded in four-point bending.
In those tests, the distance between the load and support points on either side of a beam were 160 mm,
while the cantilever over the support point was 35
±
5 mm, resulting in a total anchorage length of
195 ±5 mm
when calculating from one support point. All specimens with C-cables failed by cable slip
usually from a crack that originated in the four-point bending center span within several centimeters of
a support point, i.e., <240 mm from the edge. The calculated anchorage length for C cables is 180 mm
Materials 2017,10, 1314 15 of 22
based on maximum bond stress, but 237 mm based on adhesive bond stress. This seems to suggest the
adhesive bond stress should be maintained to determine the anchorage length. Exceeding the adhesive
strength likely causes damage (e.g., gradual debonding) that will over longer lengths result in a lower
maximum bond stress (in such a case the bond strength would be independent of embedment length if
the adhesive bond stress is exceeded). On the other hand, none of the A-cable specimens in that study
failed by slip, which matches with the anchorage length of 118 mm (based on adhesive bond) found in
the current study.
4.4. Conclusions on Pull-Out Test
The pull-out tests have shown that considerable ultimate bond strengths (better than smooth
rebar) can be achieved between the cables and cast concrete. Additionally, reasonable ultimate bond
strengths (comparable or slightly worse than smooth rebar) can be achieved between the cables and
printed concrete. However, the difference between the bond strength with cast and printed concrete
is significant. This is likely caused by differences in the concrete matrix quality. Improvement of the
bond strength in printed concrete should, therefore, be pursued. An additional advantage is likely to
be a reduced scatter in strength.
5. Four-Point Bending Test
5.1. Method
5.1.1. Specimen Design
The beam specimen design included cables in each layer, as that is a likely situation in printed
objects in which this technique is applied. Furthermore, the dimensions were chosen so that, from a
simple analytical calculation, the failure moment M
u
exceeds the crack moment M
cr
. For the concept
to be comparable to conventional reinforcement in terms of performance, M
u
> M
cr
is a precondition.
Bos et al. [
30
] show that simple mechanical section calculations for conventional reinforcement to
determine the failure moment, at least globa lly also apply to cable reinforced concrete. The failure
moment was thus calculated from:
Mu=Fu,cable ×n×zave, where (2)
Fu,cable = the ultimate force in a cable, taken from Table 1.
n= the number of cables
z
ave
= the average internal lever arm, taken as z
ave
= 0.9
×
d
ave
, with d
ave
= the average distance
from the reinforcement heart line to the beam top. Based on [
30
], each reinforcement cable is assumed
to be positioned 8 mm from the bottom of the respective layer. The layer thickness itself considered to
be hlayer = 11 mm, the width b= 50 mm.
The crack moment was determined by:
Mcr =fcm W=fcm b h2/6. (3)
This simplified equation is applicable as the influence of the cables on the crack resistance is
negligible (<1%). For the concrete flexural strength, f
cm
= 1.9 MPa was applied. This value is based
on unpublished tensile tests and has been known to deviate. It should therefore be considered
an approximation.
The beam length and span proportions were selected based on the test rig dimensions and
calculated anchorage lengths from the pull-out tests. Figure 12 presents the four-point bending
scheme. The sum of the cantilever cand the load-to-support distance ashould exceed the anchorage
length: a+c>lanchorage.
Materials 2017,10, 1314 16 of 22
Materials 2017, 10, 1314 16 of 22
Figure 12. Four-point bending test scheme.
With a single cable A in each layer and the available print nozzle, it proved difficult to achieve
dimensions in which the failure moment would be higher than the crack moment. This requirement
was therefore abandoned for cable type A (not for the other two). Obviously, this indicates that cable
A is less suitable than cables B and C as replacement of conventional reinforcement in printed
concrete.
These considerations led to a beam section of three layers in height (Figure 13), 1000 mm overall
length, a support span of 900 mm and a load span of 450 mm. The analytically calculated crack and
failure moments are given in Table 5 with the experimental results.
Figure 13. Beam specimen section design.
Table 5. Four-point bending test results.
Beam Specimen Mcr [103 Nmm] Mu [103 Nmm] Mu/Mcr No. Cracks Failure Mode
Analytical estimate 17.1 13.6 80% - Cable break
A1 19.0 14.1 74% 1
Cable break
A2 21.0 18.0 86% 1
Cable break
A3 17.6 15.3 87% 1
Cable break
Average 19.2 15.8 82% - -
Coeff. of variation 9% 13% 9% - -
Analytical estimate 17.1 38.6 304% Cable break
B1 18.8 17.6 93% 2 Cable slip
B2 23.9 21.1 88% 1 Cable slip
B3 18.1 15.0 83% 2 Max. defl.
B4 16.8 29.8 177% 3 Cable slip
B5 15.2 15.1 99% 2 Max. defl.
B6 17.4 11.5 66% 2 Max. defl.
Figure 12. Four-point bending test scheme.
With a single cable A in each layer and the available print nozzle, it proved difficult to achieve
dimensions in which the failure moment would be higher than the crack moment. This requirement
was therefore abandoned for cable type A (not for the other two). Obviously, this indicates that cable A
is less suitable than cables B and C as replacement of conventional reinforcement in printed concrete.
These considerations led to a beam section of three layers in height (Figure 13), 1000 mm overall
length, a support span of 900 mm and a load span of 450 mm. The analytically calculated crack and
failure moments are given in Table 5with the experimental results.
Materials 2017, 10, 1314 16 of 22
Figure 12. Four-point bending test scheme.
With a single cable A in each layer and the available print nozzle, it proved difficult to achieve
dimensions in which the failure moment would be higher than the crack moment. This requirement
was therefore abandoned for cable type A (not for the other two). Obviously, this indicates that cable
A is less suitable than cables B and C as replacement of conventional reinforcement in printed
concrete.
These considerations led to a beam section of three layers in height (Figure 13), 1000 mm overall
length, a support span of 900 mm and a load span of 450 mm. The analytically calculated crack and
failure moments are given in Table 5 with the experimental results.
Figure 13. Beam specimen section design.
Table 5. Four-point bending test results.
Beam Specimen Mcr [103 Nmm] Mu [103 Nmm] Mu/Mcr No. Cracks Failure Mode
Analytical estimate 17.1 13.6 80% - Cable break
A1 19.0 14.1 74% 1
Cable break
A2 21.0 18.0 86% 1
Cable break
A3 17.6 15.3 87% 1
Cable break
Average 19.2 15.8 82% - -
Coeff. of variation 9% 13% 9% - -
Analytical estimate 17.1 38.6 304% Cable break
B1 18.8 17.6 93% 2 Cable slip
B2 23.9 21.1 88% 1 Cable slip
B3 18.1 15.0 83% 2 Max. defl.
B4 16.8 29.8 177% 3 Cable slip
B5 15.2 15.1 99% 2 Max. defl.
B6 17.4 11.5 66% 2 Max. defl.
Figure 13. Beam specimen section design.
Table 5. Four-point bending test results.
Beam Specimen Mcr [103Nmm] Mu[103Nmm] Mu/Mcr No. Cracks Failure Mode
Analytical estimate 17.1 13.6 80% - Cable break
A1 19.0 14.1 74% 1 Cable break
A2 21.0 18.0 86% 1 Cable break
A3 17.6 15.3 87% 1 Cable break
Average 19.2 15.8 82% - -
Coeff. of variation 9% 13% 9% - -
Analytical estimate 17.1 38.6 304% Cable break
B1 18.8 17.6 93% 2 Cable slip
B2 23.9 21.1 88% 1 Cable slip
B3 18.1 15.0 83% 2 Max. defl.
Materials 2017,10, 1314 17 of 22
Table 5. Cont.
Beam Specimen Mcr [103Nmm] Mu[103Nmm] Mu/Mcr No. Cracks Failure Mode
B4 16.8 29.8 177% 3 Cable slip
B5 15.2 15.1 99% 2 Max. defl.
B6 17.4 11.5 66% 2 Max. defl.
Average 20.3 18.4 101% - -
Coeff. of variation 16% 35% 39% - -
Analytical estimate 17.1 62.4 491% Cable break
C1 15.3 27.5 179% 1 Cable slip
C2 14.9 21.4 144% 2 Cable slip
C3 16.0 27.5 172% 2 Cable slip
C4 16.9 22.6 134% 2 Cable slip
C5 12.8 15.1 118% 2 Cable slip
Average 15.2 22.8 149% - -
Coeff. of variation 10% 23% 17% - -
5.1.2. Specimen Preparation
The four-point bending specimens were taken from three subsequently printed objects.
Each object
is three layers high; reinforcement cable (A, B, and C, respectively) are entrained in
each layer. After printing, the objects were covered in foil for one day, subsequently submerged in
water, and left to cure for 28 days. The long straight sides were then sawed into beams with a diamond
saw. Three specimens with cable A were tested, six with cable B and five with cable C.
5.2. Results
The test results of the four-point bending tests are listed in Table 5. The corresponding
load-displacement curves are shown in Figures 1416.
Materials 2017, 10, 1314 17 of 22
Average 20.3 18.4 101% - -
Coeff. of variation 16% 35% 39% - -
Analytical estimate 17.1 62.4 491% Cable break
C1 15.3 27.5 179% 1 Cable slip
C2 14.9 21.4 144% 2 Cable slip
C3 16.0 27.5 172% 2 Cable slip
C4 16.9 22.6 134% 2 Cable slip
C5 12.8 15.1 118% 2 Cable slip
Average 15.2 22.8 149% - -
Coeff. of variation 10% 23% 17% - -
5.1.2. Specimen Preparation
The four-point bending specimens were taken from three subsequently printed objects.
Each object is three layers high; reinforcement cable (A, B, and C, respectively) are entrained in each
layer. After printing, the objects were covered in foil for one day, subsequently submerged in water,
and left to cure for 28 days. The long straight sides were then sawed into beams with a diamond saw.
Three specimens with cable A were tested, six with cable B and five with cable C.
5.2. Results
The test results of the four-point bending tests are listed in Table 5. The corresponding load-
displacement curves are shown in Figures 15–17.
Figure 15. Load-displacement curves of printed beams with A-type cable.
Figure 14. Load-displacement curves of printed beams with A-type cable.
Materials 2017,10, 1314 18 of 22
Materials 2017, 10, 1314 18 of 22
Figure 16. Load-displacement curves of printed beams with B-type cable.
Figure 17. Load-displacement curves of printed beams with C-type cable.
5.3. Discussion
The crack moment of the beams with cable type A (19.2 Nm) is slightly higher than the
analytically calculated value (17.1 Nm). This could be caused by geometrical deviations or the actual
concrete flexural strength.
Failure occurs by cable breakage, as desired. The failure moment is practically equal to the
calculated value. This confirms the suitability of analytical analysis methods for conventional
concrete also for printed concrete with cable reinforcement. As expected, the failure moment does
not exceed the crack moment. This fits with the single localised crack that occurs.
Figure 15. Load-displacement curves of printed beams with B-type cable.
Materials 2017, 10, 1314 18 of 22
Figure 16. Load-displacement curves of printed beams with B-type cable.
Figure 17. Load-displacement curves of printed beams with C-type cable.
5.3. Discussion
The crack moment of the beams with cable type A (19.2 Nm) is slightly higher than the
analytically calculated value (17.1 Nm). This could be caused by geometrical deviations or the actual
concrete flexural strength.
Failure occurs by cable breakage, as desired. The failure moment is practically equal to the
calculated value. This confirms the suitability of analytical analysis methods for conventional
concrete also for printed concrete with cable reinforcement. As expected, the failure moment does
not exceed the crack moment. This fits with the single localised crack that occurs.
Figure 16. Load-displacement curves of printed beams with C-type cable.
5.3. Discussion
The crack moment of the beams with cable type A (19.2 Nm) is slightly higher than the analytically
calculated value (17.1 Nm). This could be caused by geometrical deviations or the actual concrete
flexural strength.
Materials 2017,10, 1314 19 of 22
Failure occurs by cable breakage, as desired. The failure moment is practically equal to the
calculated value. This confirms the suitability of analytical analysis methods for conventional concrete
also for printed concrete with cable reinforcement. As expected, the failure moment does not exceed
the crack moment. This fits with the single localised crack that occurs.
The beams with cable B also crack at a moment that is somewhat higher (20.1 Nm) than the
analytical value. Contrary to the A-cable beams, however, the failure moment is significantly lower
than the analytical result. On average, it barely exceeds the crack moment, is this entirely due to one
specimen outperforming the rest (B4, reason unclear). The cause for this discrepancy is the failure
mode. Instead of cable breakage, cable slip occurs. This also results in a higher scatter for M
u
(35%,
or 22% without specimen B4). Considering the previous results ([
30
]; in this study on cable reinforced
beams, cable breakage of B-type cables occurred) and the calculated anchorage length (l
adh
= 217;
Table 4), this was unexpected. The inferior bond quality could be caused by the low number of layers
in this specimen design. In the pull-out specimens, as well as the beam specimens of [
30
], the internal
pressure from self-weight was larger due to the higher number of layers printed. This likely resulted
in better compaction of the concrete around the reinforcement cables. Together, these results show
the potential of cable reinforcement, but also the need to better control the cable printing process to
achieve more constant results.
The crack moment of the C-cable specimens is significantly lower than that of the A- and B-cable
beams, and lower than the analytical value as well. Again, deviations in concrete geometry and
strength are the likely cause. The failure moment, on the other hand, is significantly higher than
both the crack moment and the failure moments of the other beams, due to the higher reinforcement
strength. Still, it does not come near the analytical value as here, too, failure is induced by cable slip
instead of cable breakage. In the study of [
30
], this was also experienced, even though the pull-out
tests indicate that cable breakage should occur as the anchorage length exceeds ladh (237 mm).
In the B- and C-cable beams unexpected failure caused by cable slip occurs. The internal pressure
may influence the adhesion. An additional cause could be the peak stresses at the load side of the
cable. They will reach a certain value that may cause gradual debonding that will induce eventual
failure regardless of the anchorage length. The low concrete strength and the matrix defects that occur
particularly to the underside of the cable, may cause this effect to occur before the cable has had the
chance to activate sufficient anchorage length to break itself. The pull-out tests did not signal such an
effect, but the forces in those tests stayed well below the strengths of cables B and C. This effect might
only occur at larger embedment lengths. Further research is required.
The four-point bending tests have shown that the presented in-print reinforcement method is
feasible and can result in considerable post-crack resistance, similar to conventionally reinforced
concrete. However, premature bond failure results in underachievement of the cables and high scatter.
Thus, improvement of the concept is necessary.
In comparing cables A, B, and C, it is apparent that cable A currently provides the most predictable
structural behaviour, but is nevertheless less suitable as its absolute strength is limited. Only when the
concrete filament size is significantly reduced or the number of cables per layer is increased, can this
cable provide sufficient strength. Cables B and C, on the other hand, provide sufficient strength but
suffer from adhesion problems. Process improvements should eliminate this in the future.
6. Summary and Conclusions
Directly entrained cables have been introduced as a reinforcement method for 3D-printed concrete.
A device was presented to entrain these cables directly during printing, resulting in a single automated
manufacturing process. Two experiments have been conducted: a pull-out test on cast and printed
concrete with different embedment lengths and three types of reinforcement cables, and a four-point
bending test on printed beams with the same three cables.
The pull-out tests showed the bond strength of cables in cast concrete is low compared to
conventional ribbed rebar, but somewhat higher than that of smooth rebar. In printed concrete,
Materials 2017,10, 1314 20 of 22
the bond strength was considerably lower than in cast concrete, and towards the lower end of what
would be expected of smooth rebar. The concrete matrix showed substantial defects, particularly
underneath the reinforcement cable.
Subsequently, the four-point bending tests showed significant post-crack resistance can be
achieved with the B- and C-type cables. However, failure of the respective specimens was governed by
cable slip, which was not expected based on the pull-out tests. This increases the scatter, and results
in failure moments (far) below the analytically determined potential based on cable breakage. In the
A-cable beams, cable slip did not occur. Rather, failure was induced by cable breakage resulting in
failure moments close to the analytically determined values. Nevertheless, these were below the crack
moment due to the limited cable strength.
The bending tests confirmed the suitability of analytical analysis methods to determine the
resistance of conventional concrete also for printed concrete with cable reinforcement, at least to obtain
global estimates on moment resistance.
The concept of directly entrained cable reinforcement has been shown to be a feasible
reinforcement method that can achieve performances similar to conventional reinforcement in cast
concrete. However, effort needs to be taken to improve the bond strength and scatter on the B- and
C-type cables, so that their failure strengths can be reached before cable slip occurs. The A-type cable
works well, but does not seem to be strong enough for practical applications.
Acknowledgments:
The assistance in the 3DCP research of Master track students Structural Design at the TU/e
Department of the Built Environment is highly valued. The TU/e research program on 3DCP is co-funded
by a partner group of enterprises and associations, which as of the date of writing consisted of (alphabetical
order) Ballast Nedam, BAM Infraconsult bv, Bekaert, Concrete Valley, CRH, Cybe, Saint-Gobain Weber Beamix,
SGS Intron, SKKB, Van Wijnen, Verhoeven Timmerfabriek, and Witteveen+Bos. Their support is gratefully
acknowledged. In particular, the authors would like thank SG Weber Beamix and Bekaert NV for supplying the
printing concrete and reinforcement cables, respectively.
Author Contributions:
The research was conceived by and performed under supervision of Theo A. M. Salet and
Freek P. Bos. The reinforcement entrainment device was developed by Zeeshan Y. Ahmed. The experimental work
was performed by Evgeniy R. Jutinov. The data were analyzed by Freek P. Bos, who also prepared the manuscript.
Conflicts of Interest: The authors declare no conflict of interest.
References
1.
Khoshnevis, B.; Russel, R.; Kwon, H.; Bukkapatnam, S. Contour Crafting Large Prototypes. IEEE Robot.
Autom. Mag. 2001,8, 33–42. [CrossRef]
2.
Lim, S.; Le, T.; Webster, J.; Buswell, R.; Austin, S.; Gibb, A.; Thorpe, T. Fabricating Construction Components
Using Layer Manufacturing Technology. In Proceedings of the Global Innovation in Construction Conference
(GICC’09), Loughborough, UK, 13–16 September 2009.
3.
Totalkustom. Available online: http://www.totalkustom.com/3d-castle-completed.html (accessed on
January 2016).
4.
Khoshnevis, B.; Bodiford, M.; Burks, K.; Ethridge, E.; Tucker, D.; Kim, W.; Toutanji, H.; Fiske, M. Lunar
contour crafting—A novel technique for ISRU-based habitat development. In Proceedings of the 43rd AIAA
Aerospace Sciences Meeting and Exhibit, Aerospace Sciences Meetings, Reno, NV, USA, 10–13 January 2005.
[CrossRef]
5.
Khoshnevis, B.; Zhang, J. Extraterrestrial construction using contour crafting. In Solid Freeform Fabrication
Proceedings; University of Texas: Austin, TX, USA, 2012; pp. 250–259.
6.
Colla, V.; Dini, E. Large scale 3D printing: From deep sea to the moon. In Low-Cost 3D Printing, for Science,
Education & Sustainable Development; Canessa, E., Fonda, C., Zennaro, M., Eds.; ICTP: Trieste, Italy, 2013;
pp. 127–132.
7.
Cesaretti, G.; Dini, E.; De Kestelier, X.; Colla, V.; Pambaguian, L. Building components for an outpost on the
Lunar soil by means of a novel 3D printing technology. Acta Astronaut. 2014,93, 430–450. [CrossRef]
8.
Joshi, S.C.; Sheikh, A.A. 3D printing in aerospace and its long-term sustainability. Virtual Phys. Prototyp.
2015,10, 175–185. [CrossRef]
Materials 2017,10, 1314 21 of 22
9.
3ders. Available online: http://www.3ders.org/articles/20150909-lewis- grand-hotel- andrey-rudenko-to-
develop-worlds-first-3d-printed-hotel.html (accessed on 26 September 2017).
10.
Cnet. Available online: http://www.cnet.com/news/dubai-unveils-worlds-first-3d-printed-office-building
(accessed on January 2016).
11.
Mediaoffice. Available online: http://mediaoffice.ae/en/media-center/news/23/5/2016/3d-printed-office-
building.aspx (accessed on January 2016).
12.
3ders. Available online: http://www.3ders.org/articles/20161214-spain- unveils-worlds- first-3d- printed-
pedestrian-bridge-made-of-concrete.html (accessed on 26 September 2017).
13.
3ders. Available online: http://www.3ders.org/articles/20170602-cybe-construction-completes-3d-
printing-of-168-sq-m-rdrone-laboratory- in-dubai.htm (accessed on 26 September 2017).
14.
De Ingenieur. Available online: https://www.deingenieur.nl/artikel/betonnen-fietsbrug-uit-de-printer
(accessed on 26 September 2017).
15.
3ders. Available online: http://www.3ders.org/articles/20170213-3d-printing-construction-company-apis-
cor-prints-37-m2-house-near-moscow- plans-global- expansion.html (accessed on 26 September 2017).
16.
Cementonline. Available online: https://www.cementonline.nl/proefstuk-eerste- 3d-geprinte- autobrug-
sterker-dan-verwacht (accessed on 24 September 2017).
17.
3ders. Available online: http://www.3ders.org/articles/20170907-3d-printed-concrete-office-hotel-coming-
to-copenhagen.html (accessed on 24 September 2017).
18.
Tay, Y.W.D.; Panda, B.; Paul, S.C.; Mohamed, N.A.N.; Tan, M.J.; Leong, K.F. 3D printing trends in building
and construction industry: A review. Virtual Phys. Prototyp. 2017,12, 261–276. [CrossRef]
19.
Salet, T.A.M.; Bos, F.P.; Wolfs, R.J.M.; Ahmed, Z.Y. 3D concrete printing—A structural engineering perspective.
In High Tech Concrete: Where Technology and Engineering Meet, Proceedings of the 2017 Fib Symposium, Maastricht,
The Netherlands, 12–17 June 2017; Lukovi´c, M., Hordijk,D.A., Eds.; Springer: Berlin, Germany, 2017. [CrossRef]
20.
Youtube. Available online: https://www.youtube.com/watch?v=8_m-fmkuuUA (accessed on 17 July 2017).
21.
Wu, P.; Wang, J.; Wang, X. A critical review of the use of 3-D printing in the construction industry.
Autom. Constr. 2016,68, 21–31. [CrossRef]
22.
Lim, S.; Buswell, R.; Le, T.; Austin, S.; Gibb, A.; Thorpe, T. Developments in construction-scale additive
manufacturing processes. Autom. Constr. 2012,21, 262–268. [CrossRef]
23.
Borg Costanzi, C.; Ahmed, Z.; Schipper, R.; Bos, F.; Knaack, U.; Wolfs, R. 3D Printing concrete on temporary
surfaces: The design and fabrication of a concrete shell structure. Autom. Constr. 2017, under review.
24. Huashangluhai. Available online: www.hstdgm.com (accessed on 14 July 2017).
25.
Khoshnevis, B.; Hwang, D.; Yao, K.-T.; Yeh, Z. Mega-scale fabrication by contour crafting. Int. J. Ind. Syst. Eng.
2006,1, 301–320. [CrossRef]
26.
Hambach, M.; Volkmer, D. Properties of 3D-printed fiber-reinforced Portland cement paste. Cem. Concr.
Compos. 2017,79, 62–70. [CrossRef]
27.
4TU. Available online: https://www.4tu.nl/bouw/en/lighthouse2017/3D%20Concrete%20Printing%
20for%20Structural%20Applications/ (accessed on 26 September 2017).
28.
Hack, N.; Lauer, W.V. Mesh-mould: Robotically fabricated spatial meshes as reinforced concrete formwork.
Archit. Des. 2014,84, 44–53. [CrossRef]
29.
Hack, N.; Lauer, W.V.; Gramazio, F.; Kohler, M. Mesh Mould: Robotically fabricated metal meshes as concrete
formwork and reinforcement. In Proceedings of the 11th International Symposium on Ferrocement and
3rd ICTRC International Conference on Textile Reinforced Concrete, Aachen, Germany, 7–10 June 2015.
30.
Bos, F.P.; Ahmed, Z.Y.; Wolfs, R.J.M.; Salet, T.A.M. 3D printing concrete with reinforcement. In High
Tech Concrete: Where Technology and Engineering Meet, Proceedings of the 2017 Fib Symposium, Maastricht,
The Netherlands, 12–14 June 2017; Lukovi ´c, M., Hordijk, D.A., Eds.; Springer: Berlin, Germany, 2017;
pp. 2484–2493. [CrossRef]
31.
Bos, F.; Wolfs, R.; Ahmed, Z.; Salet, T. Additive manufacturing of concrete in construction: Potentials and
challenges of 3D concrete printing. Virtual Phys. Prototyp. 2016,11, 209–225. [CrossRef]
32.
Hagan, P.; Chen, J.; Saydam, S. The Load Transfer Mechanism of Fully Grouted Cable Bolts under Laboratory
Tests. In Proceedings of the Coal Operators’ Conference, Wollongong, Australia, 12–14 February 2014.
33.
Chen, J.; Hagan, P.C.; Saydam, S. Sample Diameter Effect on Bonding Capacity of Fully Grouted Cable Bolts.
Tunn. Undergr. Space Technol. 2017,68, 238–243. [CrossRef]
Materials 2017,10, 1314 22 of 22
34.
Benmokrane, B.; Chennouf, A.; Mitri, H.S. Laboratory evaluation of cement-based grouts and grouted rock
anchors. Int. J. Rock Mech. Min. Sci. Geomech. Abstr. 1995,32, 633–642. [CrossRef]
35.
Haskett, M.; Oehlers, D.J.; Mohamed Ali, M.S. Local and global bond characteristics of steel reinforcing bars.
Eng. Struct. 2008,30, 376–383. [CrossRef]
36.
Fédération Internationale du Béton. Bond of Reinforcement in Concrete, State-of-Art-Report; International
Federation for Structural Concrete: Lausanne, Switzerland, 2000; ISBN 978-2-88-394050-5.
37.
Cairns, J. Bond and anchorage of embedded steel reinforcement in Fib Model Code 2010. Struct. Concr.
2015
,
45–55. [CrossRef]
38.
NEN-EN 1992-1-1 Eurocode 2. Ontwerp en Berekening van Betonconstructies, Deel 1-1: Algemene Regels en Regels
Voor Gebouwen; Nederlands Normalisatie Instituut: Delft, The Netherlands, 2011.
39.
Melo, J.; Rossetto, T.; Varum, H. Experimental, Experimental study of bond–slip in RC structural elements
with plain bars. Mater. Struct. 2015,48, 2367–2381. [CrossRef]
40.
Abrams, D. Tests of bond between concrete and steel. In Bulletin No. 71; Engineering Experiment Station,
University Illinois Bull: Carbondale, IL, USA, 1913.
41.
Fabbrocino, G.; Verderame, G.M.; Manfredi, G. Experimental behaviour of anchored smooth rebars in old
type reinforced concrete buildings. Eng. Struct. 2005,27, 1575–1585. [CrossRef]
42.
Verderame, G.M.; Ricci, P.; De Carlo, G.; Manfredi, G. Cyclic bond behaviour of plain bars. Part I:
Experimental investigation. Constr. Build. Mater. 2009,23, 3499–3511. [CrossRef]
43.
Yeomans, S.R. Galvanized Steel Reinforcement. In Corrosion of Steel in Concrete Structures; Poursaee, A., Ed.;
Elsevier: Amsterdam, The Netherlands, 2016; ISBN 978-1-78-242381-2. [CrossRef]
44.
ISO 14657. Zinc-Coated Steel for the Reinforcement of Concrete; International Organization for Standardization:
Geneva, Switzerland, 2005.
45.
ASTM A767/A767M-16. Standard Specification for Zinc-Coated (Galvanized) Steel Bars for Concrete Reinforcement;
ASTM International: West Conshohocken, PA, USA, 2016. [CrossRef]
46.
Swamy, R. Design for durability with galvanized reinforcement. In Galvanized Steel Reinforcement in Concrete;
Yeomans, S., Ed.; Elsevier: Oxford, UK, 2004; ISBN 978-0-08-044511-3.
47.
Kayali, O.; Yeomans, S. Bond of ribbed galvanized reinforcing steel in concrete. Cem. Concr. Compos.
2000
,22,
459–467. [CrossRef]
48.
Hamad, B.; Mike, J. Bond strength of hot-dip galvanized reinforcement in normal strength concrete structures.
Constr. Build. Mater. 2005,19, 275–283. [CrossRef]
49.
Pernicova, R.; Dobias, D.; Pokorny, P. Problems connected with use of hot-dip galvanized reinforcement in
concrete elements. Procedia Eng. 2017,172, 859–866. [CrossRef]
50.
NEN-EN 10080:2005. Steel for the Reinforcement of Concrete—Weldable Reinforcing Steel—General; Nederlands
Normalisatie Instituut: Delft, The Netherlands, 2005.
51.
Banthia, N. A study of some factors affecting the fiber–matrix bond in steel fiber reinforced concrete. Can. J.
Civ. Eng. 1990,17, 610–620. [CrossRef]
52.
Chanvillard, G.; Aïtcin, P.-C. Pull-out behavior of corrugated steel fibers: Qualitative and statistical analysis.
Adv. Cem. Based Mater. 1996,4, 28–41. [CrossRef]
53.
Robins, P.; Austin, S.; Jones, P. Pull-out behaviour of hooked steel fibres. Mater. Struct.
2002
,35, 434–442.
[CrossRef]
54.
Soetens, T.; van Gysel, A.; Matthys, S.; Taerwe, L. A semi-analytical model to predict the pull-out behaviour
of inclined hooked-end steel fibres. Constr. Build. Mater. 2013,43, 253–265. [CrossRef]
55.
He, Q.; Liu, C.; Yu, X. Improving steel fiber reinforced concrete pull-out strength with nanoscale iron oxide
coating. Constr. Build. Mater. 2015,79, 311–317. [CrossRef]
©
2017 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).
... Over the past few years, a few structural reinforcing methods have been proposed to incorporate the reinforcement during the extrusion-based 3DCP process including automated entrainment of continuous reinforcement microcable [12][13][14] or continuous fibre [15][16][17] in the filament, barpenetrated reinforcement with lap joints or concurrent deposition of U-nails to strengthen the interface bonding [18][19][20][21], and short-fibre reinforcement for self-reinforced cementitious composites [22][23][24][25]. ...
... Fig. 12 shows the uniaxial tensile stress-strain curves of all ECC mixtures at 28 d, indicating that all mixtures exhibited tensile strain-hardening behaviour and high ductility along with clear multiple micro-cracking (Fig. 13). It should be mentioned that the tensile stress-strain curve with ductility close to the average one of different curves for each mixture was selected as the typic curve for comparison between different mixtures, as shown Fig. 12f. ...
... where δ0 is the crack opening at the maximum bridging stress (σ0), δss is the crack opening corresponding to the steady state bridging stress (σss), Km is the fracture toughness of matrix that is equal to 0.118 MPa•m 1/2 for normal ECC and printed ECC derived by fitting the experimental data of direct tensile tests on ECC without fibre to Eq. (12), and Em is the elastic modulus of matrix obtained from the stress-strain curve under uniaxial compression (18 GPa for both normal and printed ECC in this study). ...
Article
Full-text available
The tensile behaviour of engineered cementitious composites (ECC) is highly dependent on their microstructure characteristics. To date, the strain-hardening behaviour of printed ECC in relation to its microstructure is not yet fully understood. This study presents a systematic investigation on the macroscopic mechanical properties of normal and printed ECC with various polyethylene (PE) fibre lengths (6 and 12 mm) in relation to their microstructural features in terms of pore structure characteristics, fibre orientation and fibre dispersion through a series of mechanical tests and X-ray computed tomography (CT) and backscattered electron (BSE) image acquisition, processing and analysis. Results indicate that it is desirable to use block specimens for mould-casting fabrication as contrast to printed ECC samples. The printed ECC containing 1.5 vol% 6 mm and 0.5 vol% 12 mm PE fibres by extrusion-based 3D printing exhibits unique tensile ductility of over 5% and average crack width of less than 100 μm. Regarding pore structure, normal ECC has a higher probability of large pores (over 1 mm³) than printed ECC, which would increase the risk of damage localization and lead to a significant variation in tensile properties. Besides, normal ECC with thickness of 30 mm and printed ECC possess a similar fist cracking strength as indicated by similar pore size and fracture toughness. Compared to normal ECC, printed ECC has a more uniform dispersion of PE fibres, the orientation of which is more perpendicular to the loading direction, resulting in a higher average tensile strength and strain capacity than normal ECC.
... However, while using conventional rebars for reinforcement in DFC benefits from the use of available lowcost construction components, they do not leverage the high degree of design freedom afforded by digital fabrication. Some research has therefore been conducted on digital fabrication and 3D printing of the reinforcing steel itself [19][20][21]. Mechtcherine et al. [19] suggest 3D printing steel reinforcement for use with 3D printed concrete. This technique demonstrates a steel-concrete bonding quality similar to that of conventional steel reinforcement. ...
... Hack et al. [20] robotically assemble an external reinforcing mesh into which concrete is poured, for use as a curved wall with complex geometry. Bos et al. [21] use reinforcing steel wire that is placed simultaneously during a concrete 3D printing process, and tests the bonding efficacy. The pull-out strength of the reinforcing wire is found to be comparable to that of smooth rebar manufactured without surface deformations. ...
Article
Full-text available
With the rise of interest in digital fabrication of reinforced concrete structures, a great number of structural concrete designs that depart from standard prismatic shapes are being suggested. This has prompted an exploration of steel reinforcement strategies that are alternative to the classical deformed or “ribbed” rebars. One such is to cut internal reinforcement from steel plates using a waterjet cutting machine. Advantages of automated waterjet cutting steel reinforcement include high precision and accuracy, and minimal expense for increasing the complexity of (2D) reinforcement layouts. However, it is not known how the application of ribbing patterns along the cut edge of reinforcing bars affects the steel–concrete bond. This work conducts experimental pullout tests of waterjet-cut steel plate reinforcement with three different ribbing patterns and compares the bond strength with equivalent classic rebars. Two of the tested geometries averaged within 90–91% of the pull-out force of conventional rebar, demonstrating viability of this alternative reinforcement method.
... In conventional methods of concrete reinforcement, rebar is used, but in AM, special fibers are required which should be sufficiently small in diameter. [64], (c) contour crafting combined with vertical steel reinforcement [65], and (d) reinforcement in contour crafting [66]. ...
... Advances in cement-based printing materials are one of the essential pillars tied to the application of AM in the construction industry. [64], (c) contour crafting combined with vertical steel reinforcement [65], and (d) reinforcement in contour crafting [66]. ...
Article
Full-text available
Additive manufacturing has drawn significant attention in both academia and industry due to its capabilities and promising potential in various sectors. However, the adoption of this technology in large-scale construction is still limited due to the numerous existing challenges. In this work, a comprehensive review of large-scale automated additive construction, its challenges, and emerging advances with a focus on robotic solutions and environmental sustainability is presented. The potential interrelations of the two topics are also discussed. A new classification scheme of available and emerging robotic solutions in automated additive construction is presented. Moreover, the vision of environmental sustainability is explored through three lenses: process, material, and printed large-scale structures/buildings. Finally, the current challenges and potential future directions are highlighted. The provided state of the art and challenges can be used as a guideline for future research on large-scale automated additive construction.
... CNC milling process for fabricating the customized formwork led to extra cost and time, thus rendering the fabrication process not ideal for making components with unrepeated geometries [10]. However, construction 3D-printing was especially suitable for fabricating this type of component [10,12,26,27]. Because 3D-printing was a formwork-free process for automatic material deposition [28,29], using such technology could avoid the high cost for fabricating customized formwork and eliminate the waste generated by formwork usage [30]. ...
... In the most prevalent form of extrusion-based 3D printing, concrete was extruded from a nozzle that was attached either to a 3 or 4-axis Cartesian robots [26] or to a 6 axis robotic arm [12,27]. The nozzle extruded concrete filaments along defined paths. ...
Article
Full-text available
Existing approaches from design to concrete 3D-printing fabrication can customize the shapes of compression-dominated concrete arches and vaults but has limited applications due to high facility requirements such as a robotic arm and a reconfigurable print bed for fabricating overhanging geometries. Therefore, there is a need to develop an alternative design-to-fabrication approach for 3D printers without such facility requirements. In this paper, concrete blocks were designed as prismatic shapes which could be customized by a most basic, gantry-based 3D printer with a flat print bed and could be assembled to a larger 3D arch structure designed based on stability and strength analyses. The feasibility of such approach was demonstrated by lab prototyping. Reduced facility requirements in this approach allow 3D-printing to be more widely applied for customizing compression-dominated structures. With further design method innovation in the future, this design-to-fabrication approach can be extended for compression-dominated structures with more complex geometries.
... Perhaps one of the most promising reinforcement concepts in the udirection is the cable reinforcement, where a high strength steel cable runs off a spool into the concrete layer, without limiting the form freedom [16]. Similarly, a flow based pultrusion method has been used to insert multiple lines of basalt fibres into a printed layer [17]. ...
Article
Full-text available
The incompatibility of 3D concrete printing (3DCP) with conventional reinforcement methods is well known. Recently, solutions have suggested the insertion of helical reinforcement rods through a screwing motion into the freshly printed material. The current study focuses on the bond properties of such reinforcement and its relation to placement time relative to the 3D printed concrete age, of which until now hardly any data exists. Confined pull-out tests and micro-computed tomography (μCT) scans were performed to characterize the time-dependent bond properties for automatically placed screw-type reinforcement in 3D printed concrete in the range of 0-200 min after material deposition. An experimental program was carried out using a gantry type 3D concrete printer and a robotic hand with the Automated Screwing Device to automate the reinforcement placement process. In total 200 specimens were produced and tested in pull-out. μCT scans were done on the specimens to quantify air content in the vicinity of the reinforcement, for every other time stamp. Two different screw geometries were used. A high mechanical interlock was achieved resulting in a high bond strength in confined pull-out tests. It was concluded from the confined pull-out tests that the pull-out performance is not influenced significantly by the time of application after mortar deposition in a time frame of up to 200 min. This firmly positions automatically applied helical reinforcement as a viable method to reinforce 3DCP structures.
... The reinforcement added between extruded concrete layers typically consists of conventional reinforcing bars (e.g. [10]) or flexible filaments, such as carbon filaments [11] or steel wires [12]. Determining the bond-slip behaviour between the interlayer reinforcement and the surrounding concrete is essential since it influences the mechanical performance and the reinforcement effectiveness. ...
Article
Full-text available
This study investigates the influence of different printing set-ups and materials on the bond strength of reinforcement manually placed between 3D printed concrete layers. An identical experimental campaign was performed at two institutes in Switzerland and Italy. Each institute used its own printing process, which consisted either of a set on-demand mix or a high yield stress material. Two types of reinforcement (reinforcing bar Ø8 mm and a high strength steel wire Ø1 mm) and three production configurations (casting and printing with the reinforcement parallel or perpendicular to the printing direction) were investigated. The results showed high bond strength with only a limited influence of the fabrication method for the reinforcing bars. An increased bond was observed for the set on-demand approach compared to the high yield strength material. The comparison with existing models showed that the reinforcing bar bond strength of printed concrete varies slightly from conventional concrete. The high strength wires exhibited poor bond. Based on the gathered experiences, insights into the standardisation of testing for 3D concrete printing are discussed.
Article
One of the major limitations of the current 3D-concrete-printing technology is the incorporation of reinforcement. Furthermore, there is a need to decrease the ecological footprint of printable concrete. As a possible solution for these challenges, this paper presents a 3D-printable strain-hardening alkali-activated composite (3DP-SHAAC) that shows pseudo-ductile behaviour under direct tension. The developed 3DP-SHAAC is composed of a one-part (just-add-water) alkali-activated binder made of slag (GGBFS), fly ash (FA) and solid activators. The one-part alkali-activated binder eliminates the need for elevated temperature curing and handling of corrosive alkaline liquids. At first, an optimum matrix was identified by studying the effects of FA to GGBFS ratio on the rheological properties and compressive strength. Subsequently, the optimum matrix was reinforced by PVA fibres to make the 3DP-SHAAC, and printing performance and rheological properties were evaluated. In addition, the influences of curing temperature on the compressive, flexural and tensile performances of the printed specimens were also investigated. The results were compared with those obtained for the mould-cast specimens. The 3DP-SHAAC exhibited superior flexural performance, higher tensile strength, and comparable tensile strain capacity to the mould-cast counterpart. Further, the curing temperature had influence on the mechanical properties of both 3D-printed and mould-cast SHAACs. The underlying reasons for the differences are discussed.
Article
Full-text available
Additive manufacturing methods1–4 using static and mobile robots are being developed for both on-site construction5–8 and off-site prefabrication9,10. Here we introduce a method of additive manufacturing, referred to as aerial additive manufacturing (Aerial-AM), that utilizes a team of aerial robots inspired by natural builders¹¹ such as wasps who use collective building methods12,13. We present a scalable multi-robot three-dimensional (3D) printing and path-planning framework that enables robot tasks and population size to be adapted to variations in print geometry throughout a building mission. The multi-robot manufacturing framework allows for autonomous three-dimensional printing under human supervision, real-time assessment of printed geometry and robot behavioural adaptation. To validate autonomous Aerial-AM based on the framework, we develop BuilDrones for depositing materials during flight and ScanDrones for measuring the print quality, and integrate a generic real-time model-predictive-control scheme with the Aerial-AM robots. In addition, we integrate a dynamically self-aligning delta manipulator with the BuilDrone to further improve the manufacturing accuracy to five millimetres for printing geometry with precise trajectory requirements, and develop four cementitious–polymeric composite mixtures suitable for continuous material deposition. We demonstrate proof-of-concept prints including a cylinder 2.05 metres high consisting of 72 layers of a rapid-curing insulation foam material and a cylinder 0.18 metres high consisting of 28 layers of structural pseudoplastic cementitious material, a light-trail virtual print of a dome-like geometry, and multi-robot simulations. Aerial-AM allows manufacturing in-flight and offers future possibilities for building in unbounded, at-height or hard-to-access locations.
Chapter
This chapter introduces the advanced cement‐based composites, including fiber‐reinforced concrete, (ultra‐) high‐strength and ultra‐high‐performance concretes, polymer‐modified concrete, shrinkage compensating concrete and self‐compacting concrete. The composites also include engineered cementitious composites, confined concrete, high‐volume fly ash concrete, structural lightweight concrete, and sea water and sea sand concrete. These materials have been developed to achieve unique advantages regarding performance or sustainability. The chapter reviews the state of the art of 3D printed concrete, covering materials and fresh properties, the construction process including reinforcement installation, and hardened‐state properties. Additive manufacturing is a disruptive technology that can integrate digitalization and automation in construction. Its application greatly compensates for the shortage of skilled laborers, resources, and construction efficiency. The process of 3D printing in the construction industry can generally be divided into two categories based on how the solid and liquid materials are fed into the printing process: extrusion‐based 3D printing; and powder‐based 3D printing.
Thesis
Full-text available
The ongoing development of 3D Concrete Printing technology is broadly associated with the boons of mass-customization industry, with the potential of productivity, time and cost optimization, as well as with the sustainable potential of the practice, leaning on a largely discussed capacity of the smart and rational material deposition offered by additive manufacturing. This latter is investigated in the present work by the means of the Life-Cycle Assessment (LCA) method applied to an extrusion-based printing technology built upon the 6-axis robotic arm.The manuscript begins with a critical review of environmental performance in construction, putting it into life-cycle perspective and introducing the theoretical framework of the LCA discipline. Then, the environmental analysis of 3D Concrete Printing technology is accomplished with a particular focus on its process-related impact. Finally, a series of case-studies is evaluated on different constructive scales: from the elementary brick of a construction procedure to the full-bodied architectural system. In closing, some aspects of the industry-wide deployment of the practice are outlooked in light of the rebound effect phenomenon. A general discussion on the sustainable development of the practice concludes this work.Mainly, the study puts forward the environmental characterization of material and process-related impacts allowing to calculate the environmental impact of any printed object. Furtherly, those environmental data can be integrated into the design and optimization studies of building elements. The outcomes of the case studies point out a fair significance of the process-related impact, that despite the material savings, prompts some important impact transfers and pollution shifts into the life-cycle balance of a concrete element.Precisely, on the basic scale of the construction procedure, the impact of the 1m3 of printed concrete is almost twice bigger than the impact of 1m3 of casted concrete. The process-related impact represents around 13% in Climate Change category and can vary significantly in function of the process and material settings.The environmental gains on the scale of a structure were shown to be insignificant in the masonry framework but go up to 50% in the reinforced concrete perspective. The process-related impact remains significant but can be compensated by the material savings if properly taken into account during the design phase. Worth noting, the material quantity was shown to be an inappropriate metrics for environmental question in structural design.The integration of the optimized 3D Concrete Printed structure into the full-bodied architectural system of building envelope has demonstrated the impact reduction potential of around 20% in Climate Change category. Finally, regarding the global deployment of the practice, the further studies quantifying the magnitude of the rebound effect are necessary in order to understand the industry-wide environmental profile of 3D Concrete Printing technology.In larger outline, the 3D Concrete Printing technology in particular and the construction automation in general has shown a real capacity for the overall environmental improvement of structural and architectural systems.
Article
Full-text available
2017): 3D printing trends in building and construction industry: a review, Virtual and Physical Prototyping To link to this article: http://dx. ABSTRACT Three-dimensional (3D) printing (also known as additive manufacturing) is an advanced manufacturing process that can produce complex shape geometries automatically from a 3D computer-aided design model without any tooling, dies and fixtures. This automated manufacturing process has been applied to many diverse fields of industries today due to significant advantages of creating functional prototypes in reasonable build time with less human intervention and minimum material wastage. However, a more recent application of this technology towards the built environment seems to improve our traditional building strategies while reducing the need for human resources, high capital investments and additional formworks. Research interest in employing 3D printing for building and construction has increased exponentially in the past few years. This paper reviews the latest research trends in the discipline by analysing publications from 1997 to 2016. Some recent developments for 3D concrete printing at the Singapore Centre for 3D Printing are also discussed here. Finally, this paper gives a brief description of future work that can be done to improve both the capability and printing quality of the current systems. ARTICLE HISTORY
Article
Full-text available
Three-dimensional (3D) printing (also known as additive manufacturing) is an advanced manufacturing process that can produce complex shape geometries automatically from a 3D computer-aided design model without any tooling, dies and fixtures. This automated manufacturing process has been applied to many diverse fields of industries today due to significant advantages of creating functional prototypes in reasonable build time with less human intervention and minimum material wastage. However, a more recent application of this technology towards the built environment seems to improve our traditional building strategies while reducing the need for human resources, high capital investments and additional formworks. Research interest in employing 3D printing for building and construction has increased exponentially in the past few years. This paper reviews the latest research trends in the discipline by analysing publications from 1997 to 2016. Some recent developments for 3D concrete printing at the Singapore Centre for 3D Printing are also discussed here. Finally, this paper gives a brief description of future work that can be done to improve both the capability and printing quality of the current systems.
Article
Full-text available
The goal of this article is to evaluate risks arising from using hot-dip galvanized reinforcement in concrete elements. The article provides detailed summary of current experimental activities but also earlier positive remarks about applicability of hot-dip galvanized reinforcement, mainly from the perspective of corrosion and bond strength with concrete. Based on previously obtained data, the article disproves barrier effect of zinc coating on base steel. The reason is the initial corrosion reaction of zinc coating in fresh concrete producing hydrogen. Further data prove that forming hydrogen irreversibly increases porosity of cement on reinforcement/concrete interface which can significantly reduce bond strength. Sufficient filling of pores by zinc corrosion product could not be confirmed.
Chapter
Full-text available
This chapter reviews the characteristics, performance and use of galvanized steel reinforcement in concrete construction. The traditional method of galvanizing by hot dipping and the recent development of continuous coating of reinforcement will be discussed, and differences in the morphology of the zinc coating will be explained. This chapter discusses the behaviour of zinc in concrete, including coating behaviour in alkaline environments, passivation, carbonation resistance and chloride tolerance. There will also be brief coverage of design considerations for galvanized reinforced concrete, including bond-slip considerations and typical applications of galvanized reinforcement.
Article
One of the geometrical restrictions associated with printed paste materials such as concrete, is that material must be self-supporting during printing. In this research paper a new methodology for 3D Printing Concrete onto a temporary freeform surface is presented. This is achieved by setting up a workflow for combining a Flexible Mould developed at TU Delft with a 4-degrees-of-freedom gantry printer (4 DOF) provided at TU Eindhoven. A number of hypothetical cases are studied, namely fully-printing geometries or combining 3D printing with casting concrete. The final outcome is a 5 m 2 partially-printed and partially-cast shell structure, combined with a CNC-milled mould simulating a Flexible Mould.
Chapter
Recent years have seen a rapid growth of additive manufacturing methods for concrete construction. A recurring issue associated with these methods, however, is the lack of ductility in the resulting product. In cases this is solved by combining printing with conventional casting and reinforcing techniques. Alternatively, this paper presents first findings on the development of a system to directly entrain a suitable form of reinforcement during printing. A device is introduced to apply the reinforcement. Several options for online reinforcement medium are compared for printability and structural performance, based printing test runs and 4-point bending tests respectively. It is shown that high-performance steel cables can provide suitable reinforcement characteristics, although improved bond would allow better use of the cable capabilities. Significant post-cracking deformations and post-cracking strength can be achieved. Further research into optimal reinforcement placement and configuration is recommended.
Book
This book contains the proceedings of the fib Symposium “High Tech Concrete: Where Technology and Engineering Meet”, that was held in Maastricht, The Netherlands, in June 2017. This annual symposium was organised by the Dutch Concrete Association and the Belgian Concrete Association. Topics addressed include: materials technology, modelling, testing and design, special loadings, safety, reliability and codes, existing concrete structures, durability and life time, sustainability, innovative building concepts, challenging projects and historic concrete, amongst others. The fib (International Federation for Structural Concrete) is a not-for-profit association committed to advancing the technical, economic, aesthetic and environmental performance of concrete structures worldwide.
Article
First insights into a 3D-printed composite of Portland cement paste and reinforcing short fibers (carbon, glass and basalt fibers, 3–6 mm) are presented, resulting in novel materials that exhibit high flexural (up to 30 MPa) and compressive strength (up to 80 MPa). Alignment of the fibers, caused by the 3D-printing process is observed, opening up the possibility to use the print path direction as a means to control fiber orientation within the printed structures. Apart from completely dense cementitious bodies, hierarchically structured bodies, displaying precisely adjusted macroporosity, are presented, the latter exhibiting a unique combination of strength and materials efficiency.
Article
Additive manufacturing is gaining ground in the construction industry. The potential to improve on current construction methods is significant. One of such methods being explored currently, both in academia and in construction practice, is the additive manufacturing of concrete (AMoC). Albeit a steadily growing number of researchers and private enterprises active in this field, AMoC is still in its infancy. Different variants in this family of manufacturing methods are being developed and improved continuously. Fundamental scientific understanding of the relations between design, material, process, and product is being explored. The collective body of work in that area is still very limited. After sketching the potential of AMoC for construction, this paper introduces the variants of AMoC under development around the globe and goes on to describe one of these in detail, the 3D Concrete Printing (3DCP) facility of the Eindhoven University of Technology. It is compared to other AMoC methods as well as to 3D printing in general. Subsequently, the paper will address the characteristics of 3DCP product geometry and structure, and discuss issues on parameter relations and experimental research. Finally, it will present the primary obstacles that stand between the potential of 3DCP and large-scale application in practice, and discuss the expected evolution of AMoC in general.