Conference PaperPDF Available

COMPREHENSIVE ANALYTICAL MODEL FOR ANCHORAGES WITH SUPPLEMENTARY REINFORCEMENT

Authors:

Abstract and Figures

The models included in the current standards and guidelines 1,2,3 to evaluate the failure loads for anchorages with supplementary reinforcement subjected to either tension or shear forces are very conservative. The existing models do not explicitly consider all three major components that provide resistance to applied forces for anchorages with supplementary reinforcement namely the concrete struts, the tension ties and the nodes. The models in EN1992-4 1 and fib Bulletin 58 3 consider the failure of tension tie (stirrups) due to bond or yielding while giving an indirect and very conservative consideration to the node (hooks) and ignore the possibility of strut failure. ACI 318 2 considers only yielding of the stirrups provided sufficient anchorage length is available within and outside the breakout body, while not considering strut failure. For anchorages under shear forces, the model in EN1992-4 considers the failure crack to originate from the front row of the anchors, which leads to a small reinforcement resistance making the model very conservative. The model in ACI and fib consider the failure crack to originate from the back anchor row however in that case the anchor steel failure (as well as the pryout failure) is calculated considering that the shear load is taken up by only the anchors in the back anchor row as well. Nevertheless, in all the models, the resistance for the concrete breakout and the reinforcement are not combined but only the higher of the two resistances is considered as the failure load of the anchorage. These assumptions make the existing models quite conservative for many cases. However, on the other hand, due to the fact that the possibility of the strut failure is not explicitly considered, for the cases of anchorages with high amounts of supplementary reinforcement, the existing models tend to get unconservative. In this work, a new analytical model is developed based on detailed experimental investigations performed on anchorages with supplementary reinforcement to take up tension forces subjected to tension loads and anchorages with supplementary reinforcement to take up shear forces close to an edge loaded in shear perpendicular and towards the edge. The model gives due consideration to and explicitly considers all three components of resistance provided by the anchorage with supplementary reinforcement. The model combines and synergizes the strength of the models proposed based on research performed at the University of Stuttgart 4,5. The mean failure loads calculated by the new model are shown to be in very good agreement with the experimentally obtained mean failure loads. The model is objective and is applicable equally well for anchorages with supplementary reinforcement under either tension or shear forces.
Content may be subject to copyright.
COMPREHENSIVE ANALYTICAL MODEL FOR ANCHORAGES
WITH SUPPLEMENTARY REINFORCEMENT
Akanshu Sharma1*, Rolf Eligehausen1,2, Jörg Asmus2
1 Institute of Construction Materials, University of Stuttgart, 70569 Stuttgart, Germany
2 IEA, Engineering Office, Eligehausen-Asmus-Hofmann, 70563 Stuttgart
*Corresponding Author Email: akanshu.sharma@iwb.uni-stuttgart.de
ABSTRACT
The models included in the current standards and guidelines1,2,3 to evaluate the failure loads for
anchorages with supplementary reinforcement subjected to either tension or shear forces are very
conservative. The existing models do not explicitly consider all three major components that provide
resistance to applied forces for anchorages with supplementary reinforcement namely the concrete
struts, the tension ties and the nodes. The models in EN1992-41 and fib Bulletin 583 consider the
failure of tension tie (stirrups) due to bond or yielding while giving an indirect and very conservative
consideration to the node (hooks) and ignore the possibility of strut failure. ACI 3182 considers only
yielding of the stirrups provided sufficient anchorage length is available within and outside the
breakout body, while not considering strut failure. For anchorages under shear forces, the model in
EN1992-4 considers the failure crack to originate from the front row of the anchors, which leads to a
small reinforcement resistance making the model very conservative. The model in ACI and fib
consider the failure crack to originate from the back anchor row however in that case the anchor steel
failure (as well as the pryout failure) is calculated considering that the shear load is taken up by only
the anchors in the back anchor row as well. Nevertheless, in all the models, the resistance for the
concrete breakout and the reinforcement are not combined but only the higher of the two resistances
is considered as the failure load of the anchorage. These assumptions make the existing models quite
conservative for many cases. However, on the other hand, due to the fact that the possibility of the
strut failure is not explicitly considered, for the cases of anchorages with high amounts of
supplementary reinforcement, the existing models tend to get unconservative.
In this work, a new analytical model is developed based on detailed experimental investigations
performed on anchorages with supplementary reinforcement to take up tension forces subjected to
tension loads and anchorages with supplementary reinforcement to take up shear forces close to an
edge loaded in shear perpendicular and towards the edge. The model gives due consideration to and
explicitly considers all three components of resistance provided by the anchorage with
supplementary reinforcement. The model combines and synergizes the strength of the models
proposed based on research performed at the University of Stuttgart4,5. The mean failure loads
calculated by the new model are shown to be in very good agreement with the experimentally
obtained mean failure loads. The model is objective and is applicable equally well for anchorages
with supplementary reinforcement under either tension or shear forces.
3rd International Symposium on
Connections between Steel and Concrete
Stuttgart, Germany, September 27th -29th , 2017
Akanshu Sharma, Rolf Eligehausen and Jörg Asmus
1 Introduction
In an accompanying paper6, systematic investigations made on quadruple anchorages (2x2 anchor
groups), in the form of WELDA® anchor plates from Peikko Group Corporation, without and with
supplementary reinforcement under tension loads and under shear loads close and towards the edge
are presented. Different amounts of supplementary reinforcement were used in the tests targeting the
reinforcement and concrete failure modes. It was shown that the failure load for anchorages under
tension or shear close and towards the edge can be significantly increased by using supplementary
reinforcement. In case of anchor groups with supplementary reinforcement, once the concrete cracks,
the stirrups get activated and provide resistance to the applied tension (or shear) loads until
reinforcement yielding or bond failure occurs. Thus, the tension or shear strength of the anchorage
can be increased by increasing the amount of supplementary reinforcement. However, this increase is
limited by concrete based failure modes such as strut failure or pryout failure in case of anchorages
provided with high amount of supplementary reinforcement.
In the current standards, such as EN1992-41, ACI 3182 and fib bulletin 583, a very conservative
approach is given to consider the influence of the supplementary reinforcement on failure loads of
the anchorages. In this paper, the test results are compared with the existing models as given in
EN1992-41 for anchorages with supplementary reinforcement loaded in tension or shear towards the
edge. Further, a new model is proposed for predicting the failure loads for anchorages with
supplementary reinforcement undergoing reinforcement failure by modifying the model proposed by
Schmid4 for anchorages with supplementary reinforcement loaded in shear towards the edge and the
model proposed by Berger5 to consider the possible strut failure in case of anchorages with high
amount of supplementary reinforcement. Additionally, a new approach is presented to calculate
pryout failure loads for anchor groups with more than one anchor row when the failure crack for
concrete edge failure is assumed from the back anchor row. It is shown that with the proposed
model, the failure loads for the low to high amount of reinforcement (where reinforcement failure
dominates) can be predicted very well.
2 Model given in EN1992-4
According to EN1992-41, for anchorages with supplementary reinforcement, the load corresponding
to failure of reinforcement in the concrete breakout body can be obtained on the basis of the strut-
and-tie model (Figure 1). In EN1992-4, only bars with a distance ≤ 0.75 times the embedment depth
(for tension loads, see Figure 1a) or the edge distance (for shear loads, see Figure 1b) from the
fastener are assumed as effective.
(a) Tension loads (b) Shear loads
Figure 1: Strut-and-tie Model in EN1992-41 for anchorages with supplementary reinforcement
Akanshu Sharma, Rolf Eligehausen and Jörg Asmus
The stirrups are considered effective provided the anchorage length l1 (Figure 1) in the concrete
breakout body is at least equal to 10 times the rebar diameter (straight bars) or at least equal to 4
times the rebar diameter (bars with a hook, bend or loop).
As per EN1992-4 (2015), the characteristic resistance, NRk,re of the supplementary reinforcement
provided for one fastener associated with anchorage failure in the concrete breakout body is given
by:
,
0
,
Rk re
Rk re n
NN
(1)
With, n = number of legs of the anchor reinforcement effective for one fastener
1,
0, , ,
12
s re bk
Rk re yk re s re
l d f
N f A

 
 
(2)
Where,
l1 = anchorage length = distance from the intersection of theoretical crack and the rebar to the stirrup
end (Figure 1)
ds,re = diameter of the stirrup
fbk = characteristic bond strength = 1.5 fbd
fbd = design bond strength according to EN1992-1-17
fyk = characteristic yield strength of reinforcing bars
As,re = Area of reinforcing bar used as stirrup
α1 = influencing factor that assumes a value of 0.7 for hooked rebar if cd < 3dsre or 1.0 if cd 3dsre
and 1.0 for straight rebar7
α2 = factor to consider the influence of cover on the bond strength defined for hooked bars as7
 
2 , , 2
1 0.15 3 ; with 0.7 1.0
d s re s re
c d d

 
cd = clear cover to the reinforcing bar in any direction or half the clear distance to the adjacent rebar,
whichever is smaller
For tension loads, Eq (1) gives the load corresponding to supplementary reinforcement failure. For
shear loads, this load is converted to shear loads considering the eccentricity between the applied
shear force and stirrups as
,
,
Rk re
Rk re N
Vx
(3)
Where, x is the factor to consider for the eccentricity between the reinforcement and the applied
shear load (compare Figure 1b)
1s
e
xz




es = distance between reinforcement and shear force acting on a fixture
z = internal lever arm of the concrete member that is approx. equal to 0,85d
d = min(depth of concrete member, 2hef, 2c1)
Akanshu Sharma, Rolf Eligehausen and Jörg Asmus
The characteristic failure load for the anchorage with supplementary reinforcement under tension (or
shear) loads, is given as
 
,,
max ;
Rk Rk c Rk re
N N N
;
 
,,
max ;
Rk Rk c Rk re
V V V
Where,
NRk,c = Characteristic resistance corresponding to concrete cone failure (under tension) of an
anchorage without supplementary reinforcement
VRk,c = Characteristic resistance corresponding to concrete edge failure (under shear) of an anchorage
without supplementary reinforcement
The mean resistances can be obtained by multiplying Eq. (2) by 1.338.
From Eq. (2), it implies that if sufficient cover is available, the hook resistance is considered as
around 40% of the bond resistance of the stirrups. Thus, longer the bond length, higher is the hook
resistance as well, which does not seem logical intuitively. This aspect was highlighted and targeted
in the model by Schmid4 as will be discussed later.
3 Comparison of test results with the model of EN1992-4
3.1 Tension tests
The comparison of the mean failure loads obtained from the experiments with the mean failure loads
calculated as per the model of EN1992-41 for the tension tests are given in Table 1. As described in
the accompanying paper6, two different configurations of supplementary reinforcement against
tension forces were tested, with 2 stirrups only outside the anchorage (4 stirrup legs in total) and with
2 stirrups outside and two within the anchorage (8 stirrup legs in total). These configurations were
named as Type 1 and Type 2 reinforcement configuration respectively.
Table 1: Comparison of mean failure loads for tension tests
Diameter of
supplementary
reinforcement
Type of
reinforcement
Total c.s. area of
stirrup legs [mm2]
Mean failure loads [kN]
Experiment
EN1992-4
Ratio Nu,exp/Nu,calc
0 (Unreinforced)
NA
0
256.9
295.2
0.87
10 mm
Type 1
314
361.2
295.2
1.22
Type 2
628
460.0
295.2
1.56
16 mm
Type 1
804
447.6
295.2
1.52
Type 2
1608
589.6
295.2
2.00
The mean tension failure predicted by the model of EN1992-41 for the anchorage in unreinforced
concrete is slightly on the unconservative side, which can be attributed to the scatter of the failure
loads. However, for the cases with supplementary reinforcement, while the tests show a clear
increase in the failure loads but the model in EN1992-41 predicts no increase in the failure loads.
This is due to the fact that the reinforcement resistance never exceeds the concrete cone resistance in
unreinforced concrete for the tested cases. The results clearly bring out the conservatism in the
existing model of EN1992-41.
Akanshu Sharma, Rolf Eligehausen and Jörg Asmus
3.2 Shear tests
Table 2 presents the comparison of the mean failure loads obtained from the experiments with the
mean failure loads calculated as per the model of EN1992-41 for the shear tests. It can be seen that
the model for concrete edge failure under shear loads is even more conservative than the model for
concrete cone failure under tension loads. This is due to the fact that EN1992-4 considers the failure
load to be evaluating considering the failure crack from front anchors. Even for the case of
anchorage without supplementary reinforcement, the model given in EN1992-4 is quite conservative.
Again, as per the model the reinforcement resistance does not exceed the concrete edge resistance in
unreinforced concrete for the tested cases.
Table 2: Comparison of mean failure loads for shear tests
Diameter of
supplementary
reinforcement
C.S. area of one
stirrup leg [mm2]
Mean shear failure loads [kN]
Experiment
EN1992-4
Ratio Vu,exp/Vu,calc
0 (Unreinforced)
0
161.7
60.2
2.69
12 mm
113
263.4
60.2
4.38
20 mm
314
328.8
60.2
5.46
4 New model
4.1 Reinforcement failure
The model for reinforcement failure proposed here is a modified version on the model developed by
Schmid4 for anchorages with supplementary reinforcement loaded under shear loads. Schmid4
proposed that the anchorage capacity of the stirrup leg is given by the sum of the hook capacity and
the bond capacity. As per Schmid4, the hook capacity is relatively independent of the bond length of
the stirrup in the breakout body. The original model by Schmid4 has been modified considering the
observations made in the tests as well as to make the model suitable for tension loads as well.
For anchorages in unreinforced concrete, in general, the new model uses the same formulations as
given in EN1992-4 to calculate the failure loads. However, under shear loads close and towards the
edge, it is assumed in the proposed model that the failure loads are always calculated from the back
anchor row both in case of anchorage without and with supplementary reinforcement. Further, the
load carrying capacity of the anchor reinforcement consists of two parts: the contribution of hook
and the contribution of bond.
The mean anchorage capacity of one stirrup leg,
0,
Rm re
N
is given as
0 0 0
, , , ,
Rm re Rm hook Rm bond s re ym
N N N A f  
(4)
Where, As,re = area of one stirrup leg and fym = mean yield strength of stirrup
The stirrups that enclose the surface reinforcement (for tension loads) or the edge reinforcement (for
shear loads) are considered as effective provided they have a bond length of at least 4ds within the
breakout body. The mean value of hook contribution for a particular stirrup leg,
0,
Rm hook
N
is given as
Akanshu Sharma, Rolf Eligehausen and Jörg Asmus
0.1
,
0, 1 2 3
30
cm cube
um hook s ym f
N A f
 

  

(5)
Where, the factor
1
considers the influence of the position of the stirrup. A value of
10.95
is
assumed for stirrups that either lie between the outermost anchors or, if they lie outside the
anchorage, are first intercepted by the crack. If yielding of the stirrups first intercepted by the crack
takes place, the next stirrup is assigned a value of effectiveness factor,
1,2 0.95
else,
1,2 0.16
.
The factor
2
considers the influence of the diameter of the surface reinforcement (for tension) or
edge reinforcement (for shear), ds,L with respect to the diameter of the stirrup, ds,re:
23
,
2,
1.2
sL
s re
d
d





(6)
The factor
3
considers the influence of the bond length, l1, of the stirrup in the breakout body and
its diameter given as:
0.4 0.25
1
3,
10
ef s re
l
hd









(for tension loads) (7)
0.4 0.25
1
31, ,
10
n s re
l
cd
 

 
 
 
(for shear loads) (8)
Where, hef is the effective embedment depth of the anchors (in case of tension loads) and c1,n is the
edge distance of the back anchor row (in case of shear loads).
The contribution of the bond of one stirrup leg is given as:
0, , 1 1,min 2
( ) /
Rm bond s re bm
N d l l f

 
(9)
Where, l1,min is the minimum anchorage length required (=4ds,re), fbm is the mean bond strength
(=1.33fbk)
α2 = factor to consider the influence of cover on bond strength defined as
2 , ,
1 0.15( ) /
d s re s re
c d d
 
cd = clear cover to the stirrup leg in any direction or half the clear distance to the adjacent stirrup,
whichever is smaller
The total capacity of the anchor reinforcement is calculated by summing up the capacities of all
effective stirrup legs
0
,,
Rm re Rm re
n
NN
(10)
Akanshu Sharma, Rolf Eligehausen and Jörg Asmus
n = number of effective stirrup legs of the anchorage. Effective are stirrups with an anchorage length,
l1 ≥ 4ds,re in the theoretical breakout body
Under tension, Eq. (10) gives the contribution of the supplementary reinforcement under tension
loads. However, to obtain the contribution of the supplementary reinforcement towards the shear
loads, Eq. (3) is used to convert the tension resistance of the reinforcement to the shear resistance.
The evaluation of the test results6 highlighted that the concrete carries a significant percentage of the
failure load corresponding to concrete cone (or edge) failure in unreinforced concrete when the peak
load of the anchorage is reached. Based on the evaluation of the test results and to keep the model
simple, it seems reasonable to assume that at peak load, the load taken up by the concrete is approx.
50% of the concrete cone (or edge) failure load of the anchorage without supplementary
reinforcement. Therefore, it is proposed that the peak failure load of an anchorage with supplemental
reinforcement is given as the failure load corresponding to 50% of the concrete cone (or edge) failure
load in unreinforced concrete plus the load corresponding to reinforcement failure calculated in
accordance with the new model. Thus the mean tension resistance of an anchorage with
supplementary reinforcement is given by
Rm, Rm, ,
0.5
Rm c re Rm c
N N N N 
(11)
Similarly, the mean shear resistance of an anchorage loaded towards the edge with supplementary
reinforcement is given by
Rm, Rm, ,
0.5
Rm c re Rm c
V V V V 
(12)
4.2 Strut failure
An upper limit to the tension failure load of the anchorage with supplementary reinforcement applies
due to a possible strut failure. This failure mode was investigated by Berger5 by performing
equivalent tests on anchorages in unreinforced concrete with varying the position of the supports.
According to the model by Berger5, the maximum possible failure load of an anchorage with anchor
reinforcement compared to the same anchorage in unreinforced concrete is given by Eq. (13).
,max ,
um strut Rm c
NN

(13)
where,
strut
is the strut factor which depends on the anchorage and stirrup configuration.
The basic strut factor is defined as
02.75 1.17 1.0
strut ef
x
h
 
(14)
x = distance between the secondary failure cone on the concrete surface and the anchor axis (see
Figure 2). Eq. (14) is valid for stirrups that enclose the surface reinforcement.
Akanshu Sharma, Rolf Eligehausen and Jörg Asmus
Figure 2: Strut-and-tie Model according to Berger5 for anchorage with supplementary reinforcement
For single anchor with one stirrup each at both sides of the anchor, the strut factor,
strut
is equal to
the basic strut factor,
0
strut
. For the case of anchor groups with supplementary reinforcement,
Berger5 proposes that the factor for strut failure should be calculated considering the contributory
area of the anchorages.
(a) Type 1 (b) Type 2
Figure 3: Consideration of strut formation in case of anchor groups with supplementary
reinforcement under tension forces
For anchorages with stirrup legs arranged symmetrically outside the anchorage (Type 1), the
maximum failure load considering strut failure is given as
, ,1 , ,2 , ,3
00
, , ,
( ) ( )
c N c N c N
strut strut strut
c N c N c N
A A A
xx
A A A
 
 
(15)
Struts
Secondary failure cone
Primary failure cone
x
Akanshu Sharma, Rolf Eligehausen and Jörg Asmus
For the case of anchorages with supplementary reinforcement also lying within the outermost anchor
rows and arranged symmetrically to the anchors (Type 2), the factor ψstrut is given as:
)()()()(
,
3,,2,,
2
0
,
41,,
1
0
Nc
NcNc
strut
Nc
Nc
strutstrut A
AA
x
A
AA
x
(16)
A model analogous to the model shown here is also applicable for anchorages with supplementary
reinforcement under shear loads. However, due to the space limitations and the fact that in the tests
under shear loads, strut failure did not occur due to the precedence by pryout failure, it is not
discussed here.
4.3 Pryout failure
As described in the accompanying paper6, in case of tests with high amount of supplementary
reinforcement under shear loads, the reinforcement failure was preceded by the pryout failure. Please
note that this failure could be avoided by using a restraint against uplift of the base plate as was used
in another similar test program by the authors9. In such cases the concrete contribution is not 50% of
its capacity in unreinforced concrete as used in Eq. (12) but 100%, even at the point of reinforcement
yielding. Nevertheless, it is impractical or unreliable to have such a constraint in reality, therefore in
this test program, the uplift restraint was not used.
The current approach to evaluate the failure load of an anchorage corresponding to pryout failure,
VRm,cp involves multiplying the tension failure load of the anchorage by a factor, k3. This model
considers the resistance associated with concrete breakout as well as the ratio between the tension
force in the anchor and the applied shear force8.
, 3 ,
Rm cp Rm c
V k N
(17)
Where,
VRm,cp is the mean pryout failure load for the anchorage
NRm,c is the mean tension failure load for the anchorage
k3 is the factor that assumes the value
3
3
1.0 for 60mm
2.0 for 60mm
ef
ef
kh
kh


As per the current standards1,3, for the anchorages close to an edge and loaded in shear towards the
edge, if the concrete edge failure load is evaluated assuming the failure crack from the front anchor
row, all anchors are considered to evaluate the pryout failure load (approach given in EN1992-41).
However, if the load corresponding to concrete edge failure is evaluated assuming the failure crack
from back anchor row then only the back anchors are considered to evaluate the pryout failure load
(approach given in fib3).
Akanshu Sharma, Rolf Eligehausen and Jörg Asmus
In the new model, the concrete edge failure loads are calculated assuming the failure crack from back
anchor row. However, if only the back anchor row is considered to calculate pryout failure load, the
failure loads of the anchorages with two or four anchor rows would be essentially equal. This
concept is explained in Figure 4a. However, the results of a few tests additionally performed on
anchorages with four anchor rows (not presented here) clearly show that the failure loads for
anchorages with four anchor rows could reach much higher values than the loads which induced
pryout failure in anchorages with two anchor rows with supplementary reinforcement. Therefore, the
approach of considering only the anchors in the back row to calculate pryout failure loads is
unobjective.
(a) Current approach (b) Proposed approach
Figure 4: Approach to evaluate pryout failure loads of anchorage
The test results6 displayed that in the cases where pryout failure occurred, a major crack appeared
from the front anchor row. Therefore, this anchor could not take up major shear load. Therefore, it is
proposed that the pryout failure load for an anchorage with supplementary reinforcement is evaluated
considering the anchor group formed by all anchor rows except the front anchor row and considering
a free edge at the line of the front anchor row. This method is explained in Figure 4b.
Akanshu Sharma, Rolf Eligehausen and Jörg Asmus
5 Comparison of test results with the new model
Using the formulations given above, the mean failure loads for the anchorages as obtained following
the new model were compared with the experimentally obtained mean failure loads for the
anchorages tested.
5.1 Tension loads
The comparison of the mean failure loads obtained from the experiments with the mean failure loads
calculated as per the new model for the tension tests are given in Table 3. For comparison, the failure
loads calculated as per the model of EN1992-41 are also included in Table 3.
Table 3: Comparison of mean failure loads for tension tests with new model
Diameter of
supplementary
reinforcement
Type of
reinforcement
Total c.s. area
of stirrup legs
[mm2]
Mean tension failure loads [kN]
Experiment
EN1992-4
New
model
Ratio
Nu,exp/Nu,calc,EN1992-4
Ratio
Nu,exp/Nu,calc,NewModel
0 (Unreinforced)
NA
0
256.9
295.2
295.2
0.87
0.87
10 mm
Type 1
314
361.2
295.2
317.2*
1.22
1.14
Type 2
628
460.0
295.2
486.9*
1.56
0.94
16 mm
Type 1
804
447.6
295.2
500.4#
1.52
0.89
Type 2
1608
589.6
295.2
566.6#
2.00
1.04
* The failure load is controlled by reinforcement failure. # The failure load is controlled by strut failure
It can be seen that for the cases with supplementary reinforcement, the new model gives much better
predictions compared to the model in EN1992-41. The new model predicts an increase in the failure
loads even with relatively low amounts of supplementary reinforcement, same as in the experiments.
Furthermore, due to the consideration of strut failure, the failure loads calculated by the new model
are in close agreement with the experiments also for high amounts of supplementary reinforcement.
The failure modes predicted by the new model are the same as observed in the experiments6.
5.2 Shear loads
Table 4 presents the comparison of the mean failure loads obtained from the experiments with the
mean failure loads calculated as per the new model for the shear tests.
Table 4: Comparison of mean failure loads for shear tests with the new model
Diameter of
supplementary
reinforcement
C.S. area of one
stirrup leg [mm2]
Mean shear failure loads [kN]
Experiment
EN1992-
4
New
Model
Ratio
Vu,exp/Vu,calc,EN1992-4
Ratio
Vu,exp/Vu,calc,NewModel
0 (Unreinforced)
0
161.7
60.2
142.5
2.69
1.13
12 mm
113
263.4
60.2
240.0*
4.38
1.10
20 mm
314
328.8
60.2
315.6#
5.46
1.04
* The failure load is controlled by reinforcement failure # The failure load is controlled by pryout failure
The failure loads predicted by the new model for concrete edge failure under shear loads in
unreinforced concrete are much more close to the experimentally obtained mean failure loads due to
the consideration of failure crack from back anchors. This approach is also recommended by fib
Bulletin 583 and ACI 3182 but not in EN1992-41.
Akanshu Sharma, Rolf Eligehausen and Jörg Asmus
For the anchorages with supplementary reinforcement, the predicted failure loads are in very good
agreement with the experimentally obtained mean failure loads. The new model predicts an increase
in the failure load even with relatively small amount of supplementary reinforcement, the same as
observed in the experiments6. For relatively high amounts of supplementary reinforcement, the
failure load is controlled by pryout failure in the new model, which is also the failure mode observed
in the tests6.
Thus, the new model is able to predict the mean failure loads as well as the failure modes which are
in good agreement with the experimentally obtained mean failure loads and the failure modes, both
under tension loads as well as shear loads.
6 Conclusions
Based on the evaluation of test results presented in the accompanying paper, a new model is
presented in this work to calculate the failure loads for anchorages with supplementary
reinforcement. The model considers both the reinforcement and concrete failure modes. The model
for the failure of supplementary reinforcement is based on the model proposed by Schmid4 for
anchorages under shear loads and is modified to consider the realistic behavior of the anchorages
with supplementary reinforcement as observed from the tests. The model was also modified to make
it applicable for anchorages with supplementary reinforcement under tension loads as well. The
model for strut failure is based on the model proposed by Berger5. A new approach is proposed to
calculate the pryout failure loads for the anchorages with more than one anchor row when the failure
crack is assumed to appear from the back anchor row. The new model is able to predict the mean
failure loads in excellent agreement with the experimentally obtained mean failure loads as well as
correctly predict the dominant failure mode, both under tension loads as well as shear loads.
7 Acknowledgements
The tests reported in this work were funded by Peikko Group Corporation. The support and efforts of
Mr. Jan Bujnak, Peikko is greatly appreciated.
References:
1. FprEN 1992-4. 2015. Eurocode 2: Design of concrete structures - Part 4 Design of fastenings
for use in concrete, European committee for standardization, CEN/TC 250, Brussels.
2. American Concrete Institute. 2014. ACI 318: Building Code Requirements for Structural
Concrete (ACI 318-14).
3. International federation for structural concrete (fib). 2011. fib Bulletin 58. Design of
anchorages in concrete - Guide to good practice, fib Special Activity Group 4.
4. Schmid, K. 2010. Behavior and design of fastenings at the edge with anchor reinforcement
under shear loads towards the edge. PhD Thesis, Institute of Construction Materials, University of
Stuttgart (In German).
Akanshu Sharma, Rolf Eligehausen and Jörg Asmus
5. Berger, W. 2015. Load-displacement behavior and design of anchorages with headed studs
with and without supplementary reinforcement under tension load). PhD Thesis, Institute of
Construction Materials, University of Stuttgart (In German).
6. Sharma, A., Eligehausen, R., Asmus, J., “Comprehensive experimental investigations on
anchorages with supplementary reinforcement, Proceedings, 3rd International Symposium on
Connections between Steel and Concrete (ConSC2017), September 27-29, 2017, Stuttgart, Germany.
7. EN1992-1-1: Eurocode 2: Design of concrete structures - Part 1-1: General rules and rules for
buildings, December 2004
8. Eligehausen, R., Mallée, R., Silva J.F.: Anchorage in Concrete Construction, Berlin: Ernst &
Sohn, 2006.
9. Sharma, A., Eligehausen, R., Asmus, J., “Experimental investigations on concrete edge
failure of multiple row anchorages with supplementary reinforcement”, Structural Concrete. 2017;
18: 153163.
10. Sharma, A., Eligehausen, R., Asmus, J., “A new model for concrete edge failure of multiple
row anchorages with supplementary reinforcement reinforcement failure”, Structural Concrete,
2017; 0: 19. https://doi.org/10.1002/suco.201700002
... However, the ACI 318 [1] and fib Bulletin 58 [2] sets the maximum distance between anchor and anchor reinforcement at 0.5•c1 and thus is more conservative. The positive influence of the anchor reinforcement on the load-bearing capacity of fastenings under shear loading perpendicular to the edge have been shown by various researchers in the past [4][5][6][7][8][9][10][11][12][13][14][15][16]. In particular, Randl and Kunz [4] investigated the influence of cast-in reinforcement on the shear loading behaviour of post-installed (metal expansion and bonded anchor) fastenings. ...
... They found significantly higher resistance in case of bonded anchors again concrete edge breakout failure with cast-in reinforcement, while in case of metal expansion anchor the increase in the shear capacity was marginal. Schmid [5] [11][12][13][14][15][16]. Furthermore, the beneficial influence of supplementary reinforcement on the behavior and resistance of cast-in anchorages has been demonstrated by [17][18][19]. ...
... However, in all previous works listed before [4][5][6][7][8][9][10][11][12][13][14][15][16], only the influence of cast-in supplementary reinforcement were investigated. Nevertheless, due to recent advancements in adhesive product developments and design methods for post-installed reinforcement it is possible now to use in many applications post-installed reinforcing bars instead of cast-in-place reinforcement. ...
Conference Paper
Full-text available
Existing anchorages in concrete structures may need strengthening due to an increase in the applied load during its lifetime. In certain cases, due to limited dimensions of the structural member, the size of anchorages is also limited, and the standard design may not be enough to provide required load-carrying capacity. In such cases, a method to strengthen the anchorage may be needed. In the present work, it is attempted to develop a method for strengthening of anchorages against concrete edge failure under shear loads by using post-installed reinforcing bars. Shear tests were performed on bonded anchors (single and group) close to an edge without and with post-installed reinforcing bars. The main objective of the research was to investigate the influence of the post-installed reinforcement on the load bearing and load-displacement behaviour of the anchorages in case of concrete edge failure. Furthermore, the effect of the arrangement of rebars on the possible load increasing was also studied. The test parameters including anchor size, embedment depth, edge distance etc. were chosen in a way that in case of reference tests, concrete edge failure occurred. The reinforcement (ribbed steel) was installed in various positions. The test results show clearly that post-installed reinforcement can strongly enhance not only the load-carrying capacity but also the displacement behaviour of the anchorages. Furthermore, it is shown that not only the amount of reinforcement, but also its position or geometry has a strong influence on its effectiveness and thus on the increasing the load-carrying capacity of the anchorages. For low amount of reinforcement spaced close to the anchors, reinforcement yielding could occur. In case of relatively high amount of reinforcement or in case of large distance between the reinforcement and the anchor, strut failure could be observed.
... Furthermore, Sharma et al. [18][19][20][21] studied the load-bearing behavior of anchor groups consisting headed-studs with supplementary reinforcement in detail. A comprehensive research campaign on headed stud anchorages with supplementary reinforcement under tension and shear towards the edge was studied and on the basis of the results a new model for predict the load-carrying capacities of anchorages with supplementary reinforcement was developed [22][23][24][25][26]. Furthermore, the beneficial influence of supplementary reinforcement on the behavior and resistance of cast-in anchorages has been demonstrated by [27][28][29]. ...
... It can be seen that beyond a certain amount of reinforcement, the increase in failure load with the amount of reinforcement is relatively low, since the strut failure limits the maximum achievable increase in the load carrying capacity of the anchorage. Similar behavior was shown valid for cast-in anchorages with cast-in supplementary reinforcement [15,24]. With the highest amount of reinforcement of 6x12 (A s,re = 678 mm 2 ), the mean failure load reached was 2.18 times the mean failure load of the reference test without reinforcement. ...
... This is partly due to indirectly avoiding strut failure and partly due to the fact that the mandrel diameter of bars bigger than 16 mm diameter becomes rather large according to EN1992-1-1. However, it has been shown by the previous works by Sharma et al. [19,24] as well as in this work that the strut failure may be reached even with smaller bar diameters depending on the configuration (placement) and amount of supplementary reinforcement. In test Series T-S_3.4, ...
Article
This paper presents the results of comprehensive and systematic experimental investigations carried out on anchorages (single anchors and groups) of bonded anchors supplemented with post-installed reinforcement subjected to tension loading. Until now, the use of supplementary reinforcement is allowed in the codes only for anchorages with cast-in headed anchors supplemented with cast-in reinforcement. In this work, tension tests are performed on single anchors and anchor groups with different amounts and arrangements of post-installed reinforcement. Epoxy based mortar was used for installation of both anchors and the reinforcement. It was found that the post-installed supplementary reinforcement can significantly increase the capacity of anchorages undergoing concrete cone breakout failure. Due to high bond strength of the mortar, even relatively short bond lengths within the breakout body were sufficient to activate the reinforcement and contribute significantly towards the resistance. The ultimate load was limited either by reinforcement yielding (for low amounts of reinforcement) or strut failure (for moderate to high amounts of reinforcement). Not just the amount but also the placement and arrangement of reinforcement plays an important role in the tension behavior of anchorages with post-installed supplementary reinforcement. At peak load, both the concrete as well as reinforcement contribute significantly to the resistance of the anchorage.
... Although design assumptions and detailing rules like those from ACI 318 are given, it is necessary to advance the guidelines on the use of stirrups as supplementary reinforcement, e.g., to assist the decision-making process when it is not feasible to fully anchor the stirrups within the concrete cone nor avoid the concrete cone failure. Besides, the models included in design codes consider only failures associated with loss of bonding or yielding of the supplementary reinforcement but ignore the possibility of concrete strut failure, as pointed by Sharma et al. [22], in cases where significant amounts of steel are added. ...
... The experimental results presented in this paper were summed to data available in the literature. They were used to discuss the theoretical estimates obtained following the provisions given by ACI 318 [2], EN [23], and Sharma et al. [22], providing valuable inclinations to improve the methods of calculation adopted today in design standards. ...
... Sharma et al. [22] simplified the method proposed by INFASO [23] but keeping the assumption that the concrete cone resistance can be calculated by the sum of the contributions given by the supplementary reinforcement and the concrete. In their proposal, the concrete contribution is assumed as 50% of the concrete cone resistance of headed bars without supplementary reinforcement. ...
Article
This paper presents the experimental response and resistance of sixteen axial tensile tests on headed deformed bars embedded in reinforced concrete members, used as cast-in anchors, under concrete cone failure. Nine of these tests investigated the influence of the flexural reinforcement ratio, which affects the concrete cracking state in the vicinity of the anchor. The other seven tests measured the shear reinforcement contribution, adjusted to work as supplementary reinforcement, distributed in the tests in different amounts and arrangements. Furthermore, design and theoretical methods were used to discuss the authors’ experimental results compared to other literature results. The flexural reinforcement ratio signifcantly influenced the concrete cone resistance, as it controls the crack width. Well-detailed stirrups, placed following the design codes’ spacing limitations for supplementary reinforcement, can substantially increase the concrete cone resistance. The assumptions underlying the design methods are conservative, which is justifed by the simplicity of their equations, but more accurate calculation methods are required
... 18 On this basis, some authors suggested to rely on specific strut-and-tie models, including near-surface reinforcement against splitting forces, while dealing with fasteners in presence of supplementary reinforcement. Within this context, Sharma et al. 20 proposed a new model which assumes that (a) the stirrups are loaded according to their position, (b) the concrete contribution is non-negligible when the stirrups are activated, and (c) the available resistance is limited by the capacity of the concrete struts. ...
... The contribution of the supplementary reinforcement to the load-bearing capacity of cast-in-place fasteners is evaluated by comparing the current design approach (Equation (6)) with the Sharma's approach 20 (Equation (7)). Since the latter approach assumes the stirrups to work as function of their position with respect to fastener bearing zone, only the first row of the stirrups is assumed to be fully intercepted by the crack (ψ 1 = 0.95). ...
... Finally, it is recommended to account for the above-mentioned effects also in the calculation of expected crack width at the anchor location; • In case the installation is performed with supplementary reinforcement, to calculate the load-bearing capacity using reliable strut and tie models. For instance, the approach recently proposed by Sharma et al. 20 can be adopted. ...
Article
Screwed‐head fastener is the common fabricated hold‐down bolt for steel structures and machine foundations. Although different models are available for evaluating its structural behavior, there are still aspects that need to be investigated. In particular, conflicting approaches for the design can be found among the European design‐oriented documents. Within this context, a comprehensive experimental study on screwed‐head fasteners under tensile loading was recently carried out at Milan Polytechnic. In this paper, some results are presented and commented which include (a) the presence of cracks and (b) the presence of supplementary reinforcement. In the discussion, predictive models are recalled demonstrating the need for a specific design approach, which should consider the geometry and the resistance of the fastening system, including that of the concrete member. Some design recommendations end the paper as useful guidance for the structural designer.
... Within this context, Sharma et al. (2017a) proposed a new analytical model based on the results of a comprehensive experimental campaign comprising both pull-out tests and tests under an applied shear force. They demonstrated that both concrete and supplementary reinforcement contribute to the pull-out resistance (Eq. ...
... Moreover, some concrete spalling in the cover close to the anchor and corresponding to the first two stirrups was clearly observed. Such a mechanism may be related to bond failure along the stirrups closest to the anchor, which are firstly intercepted by the crack (Sharma et al., 2017a). ...
... The contribution of supplementary reinforcement to the bearing capacity can be evaluated by comparing two different approaches: (i) current approach from the most advanced design codes (i.e. EN 1992-4 in Europe and ACI 318-14 in US); and (ii) Sharma's approach (Sharma et al., 2017a). ...
Conference Paper
The most recent design codes and guidelines for anchors are based on the rather conservative hypothesis that the load-bearing capacity be the greater between the resistance corresponding to concrete-cone breakout and that corresponding to supplementary reinforcement failure, because of either steel yielding or bond collapse. Among the few studies available in the literature on the behavior of anchors placed inside locally-reinforced concrete, a new analytical model was proposed by Sharma, who demonstrated that the local reinforcement contributes to anchor resistance. Within such a framework, some results are presented and commented from an experimental campaign on cast-in place anchors embedded inside concrete blocks heavily reinforced with closed stirrups to prevent concrete-cone breakout: Predictive models are recalled as well.
... As for the concrete strut, Berger [43] developed a new model where strut capacity is evaluated as an equivalent concrete cone resistance increased as function of the effectiveness of stirrups. Such models were later discussed and incorporated in a new comprehensive model for fasteners with supplementary reinforcement [81]. ...
... Sharma et al. [81] proposed a new analytical model (Eq. (24)) based on the results of a comprehensive experimental campaign comprising both pull-out tests and tests under an applied shear force. ...
Article
The correct detailing of discontinuity regions (D-regions), particularly of fastenings to concrete, is becoming a crucial task in modern structural engineering. In the past, a lot of efforts have been dedicated to study metal fasteners in concrete thus resulting in various mechanical descriptions of their structural behaviour. On this basis, semi-empirical models were proposed in which geometrical and mechanical parameters were individually evaluated. However, looking at all the factors in a larger picture, potential critical issues arise, particularly regarding cast-in-place fasteners. This paper presents a critical review on those aspects which may strongly influence the response under tensile loading. The influence of bearing pressure, the presence of reinforcement, as well as the interaction with the concrete member and in groups, are presented and commented.
Article
Full-text available
A bridge bearing anchor transmits various loads of a superstructure to a substructure. Most anchors are generally designed without consideration of characteristics such as concrete pedestal, grout bedding, and anchor socket. This study investigated the shear behavior of anchors in accordance with the edge distance, embedment depth, compressive strength of concrete, and height of the concrete pedestal in order to simulate the practical characteristics of the bridge bearing anchors. The actual shear capacity of the anchor differs from the shear strengths calculated by the ACI 318 and EN 1992-4; especially, the importance of the embedment depth is underestimated in these codes. An increase in the height of the concrete pedestal has a negative effect on the shear capacity because of the stress concentration. The grout is fractured prior to the occurrence of local damages in concrete, resulting in a secondary moment. As a result, the effect of the level arm is observed. An equation, which can predict the relative cracking degree of concrete, is proposed by analyzing the displacement of grout and concrete. High strain occurs in the stirrups close to the anchor, and the behavior of the strain is more influenced by the embedment depth than the edge distance. The comparison of obtained and analytically evaluated failure loads by calculations according to EN 1992-4, Schmid model and Sharma model was conducted to consider the effect of supplementary reinforcement. Finally, the design equation of concrete breakout strength is modified to predict the more precise shear resistance of a bridge bearing anchor.
Article
In recent years, the use of post-installed anchors to connect the haunch elements with the structural members has been investigated proficiently. It represents a promising alternative for strengthening of joints of existing reinforce concrete (RC) structures (moment resisting frames) with low invasion. The efficacy of the fully-fastened-haunch-retrofit (FFHR) solution has been proven by past works for two-dimensional RC beam-to-column joints, subjected to cyclic loading, without transverse beam and slab. In these cases, the presence of the haunch, providing a suitable design of the anchoring system, modifies the strength hierarchy shifting the mode of failure from joint’s shear breakout to the formation of the plastic hinge in the beam. However, due to the presence of the slab and transverse beam, as the authors discussed elsewhere, an increase both (i) in the joint resistance and (ii) in the beam flexural resistance must be expected, but particular care must be taken to the non-symmetric behavior. In this regard, results obtained for two RC beam-to-column sub-assemblies with transverse beam and slab retrofitted using FFHR are presented. The structural behavior under cyclic load is compared with the as-built identical specimen. The applicability of the FFHR to existing RC structures, is confirmed but, at high level of ductility demand, both anchorage break-down and reduced displacement capacity are observed.
Conference Paper
Full-text available
The behavior of anchorages with multiple headed studs is significantly influenced by the presence of supplementary reinforcement. The supplementary reinforcement to resist tension forces on the anchorage consists of surface reinforcement and the stirrups, usually enclosing the surface reinforcement. For the anchorages placed close to the edge and loaded in shear towards the edge, the supplementary reinforcement in the form of edge reinforcement and the stirrups enclosing edge reinforcement can be provided. The behavior of anchorages with supplementary reinforcement can be best described using a strut-and-tie model in which the forces applied to the anchorage are resisted by a network of concrete struts taking up the compression forces and tension ties formed by the surface (or edge) reinforcement and the stirrups. Thus, in principle, there are three major components in this strut-and-tie model, namely the concrete struts, the tension ties and the nodes. Increasing the amount of supplementary reinforcement leads to an increase in the failure load of the tension ties. However, beyond a certain level of reinforcement, the failure of concrete struts can limit the failure load for the anchorage. In this work, detailed experimental investigations were carried out on anchorages (WELDA® anchor plates from Peikko Group Corporation) consisting of multiple headed studs with supplementary reinforcement to take up tension forces subjected to tension loads and anchorages with supplementary reinforcement to take up shear forces close to an edge loaded in shear perpendicular and towards the edge. The experimental program was designed to capture the behavior of the different components and the forces taken up by concrete and reinforcement were segregated using the data obtained from strain gauges applied on the stirrups. It was clearly brought out that several assumptions made in the existing models e.g. the models of EN1992-4 1 , ACI 318 2 or fib Bulletin 58 3 are not entirely in accordance with the real behavior and therefore the models are either overly conservative or tend to become unconservative depending on the configuration and the amount of supplementary reinforcement. Based on the detailed evaluation of the experimental results, a new model is developed that can rationally consider and capture the realistic behavior of the anchorages with supplementary reinforcement. The model is presented in accompanying paper.
Article
The paper presents a new model for predicting the resistance of anchorages with supplementary reinforcement loaded in shear toward and perpendicular to the edge in case of reinforcement failure. The model is based on the evaluation of the results of a comprehensive test program performed on anchorages with multiple anchor rows and supplementary reinforcement, reported in an earlier paper. It is shown that the existing models available in the codes and standards are conservative for low to medium amounts of supplementary reinforcement but tend to be unconservative for high amounts of reinforcement. The new model is able to predict the failure loads corresponding to reinforcement failure under shear loads very well. The approach to incorporate other failure modes into the design model for anchorages with supplementary reinforcement under shear loads toward the edge will be presented in another paper.
Article
The presence of supplementary reinforcement, in the form of edge reinforcement and stirrups, has a significant influence on the load-bearing capacity of anchorage groups with multiple anchor rows loaded in shear perpendicular to the edge. The current models available in the codes and standards are conservative for low to medium amounts of supplementary reinforcement but tend to be unsafe for high amounts of reinforcement. This paper presents the results of a comprehensive test program carried out to investigate the behavior of anchor groups with supplementary reinforcement loaded in shear toward the edge. The test results are discussed in detail to highlight the influence of supplementary reinforcement on the load-bearing capacity of the anchorages. Based on the evaluation of these test results, a realistic and rational model has been developed to predict concrete edge failure loads for anchorages with supplementary reinforcement that will be presented in another paper.
Article
A comprehensive treatment of current fastening technology using inserts (anchor channels, headed stud), anchors (metal expansion anchor, undercut anchor, bonded anchor, concrete screw and plastic anchor) as well as power actuated fasteners in concrete. It describes in detail the fastening elements as well as their effects and load-bearing capacities in cracked and non-cracked concrete. It further focuses on corrosion behaviour, fire resistance and characteristics with earthquakes and shocks. It finishes off with the design of fastenings according to the European Technical Approval Guideline (ETAG 001), the Final Draft of the CEN Technical Specification 'Design of fastenings for use in concrete' and the American Standards ACI 318-05, Appendix D and ACI 349-01, Appendix B. © 2006 Ernst & Sohn Verlag für Architektur und technische Wissenschaften GmbH & Co. KG, Berlin.
Building Code Requirements for Structural Concrete
  • American Concrete Institute
American Concrete Institute. 2014. ACI 318: Building Code Requirements for Structural Concrete (ACI 318-14).
Behavior and design of fastenings at the edge with anchor reinforcement under shear loads towards the edge
  • K Schmid
Schmid, K. 2010. Behavior and design of fastenings at the edge with anchor reinforcement under shear loads towards the edge. PhD Thesis, Institute of Construction Materials, University of Stuttgart (In German).
Load-displacement behavior and design of anchorages with headed studs with and without supplementary reinforcement under tension load)
  • W Berger
Berger, W. 2015. Load-displacement behavior and design of anchorages with headed studs with and without supplementary reinforcement under tension load). PhD Thesis, Institute of Construction Materials, University of Stuttgart (In German).
Comprehensive experimental investigations on anchorages with supplementary reinforcement EN1992-1-1: Eurocode 2: Design of concrete structures-Part 1-1: General rules and rules for buildings
  • A Sharma
  • R Eligehausen
  • J Asmus
Sharma, A., Eligehausen, R., Asmus, J., " Comprehensive experimental investigations on anchorages with supplementary reinforcement ", Proceedings, 3 rd International Symposium on Connections between Steel and Concrete (ConSC2017), September 27-29, 2017, Stuttgart, Germany. 7. EN1992-1-1: Eurocode 2: Design of concrete structures-Part 1-1: General rules and rules for buildings, December 2004