5th International Symposium on Advances in Civil and Environmental Engineering

Practises for Sustainable Development (ACEPS-2017)

sage of otal rack train odel to redict the hear

trength of einforced oncrete eams

N.M.S.H. Bandara1, I.M.L.R. Bandara1, K.K. Wijesundara1

1
Department of Civil Engineering
Faculty of Engineering
University of Peradeniya
Peradeniya
SRI LANKA
E-mail: [email protected]

Abstract: The shear strength of concrete beams is an important criterion for designing of structures
because shear failures are brittle and very hazardous. Throughout the world significant efforts have
been made to predict the shear strength of concrete members which are critically affecting the
structural performance. But still there is a great contradiction concerning the mechanisms which
govern shear in concrete. With the development of concrete construction industry, now it is common
that in some cases shorter shear span to depth ratio to beams are resulted. In the design stage of
such elements, as a consequence that less attention paid in predicting the shear capacity than
moment capacity, the brittle failures mode of beams in shear is observed before the ductile failure
mode in moment. This actually violates the concept of ultimate limit state design. Therefore, the
objective of the research study is to predict the shear strength of reinforced concretebeams using
the total crack strain constitutive model and to validate the prediction with availableexperimental
data in the literature. Simply supported beams are modelled with Midas FEA using Totalcrack strain
model and their results are compared with the experimental results. Then the validatedmodel is
used to predict the shear strength of beams in monolithic construction. It was concluded that when
predicting the shear failure of reinforced concrete members by total crack strain model, resultswere
very sensitive to the defined shear stress strain relationship.

Keywords: shear strength, reinforced concrete, stresses, stiffness

1. INTRODUCTION

386
 5th International Symposium on Advances in Civil and Environmental Engineering
Practises for Sustainable Development (ACEPS-2017)

2. TOTAL CRACK STRAIN MODEL

387
 5th International Symposium on Advances in Civil and Environmental Engineering
Practises for Sustainable Development (ACEPS-2017)

Analysis models for

concrete cracking can be classified into a discrete crack model [discontinuum model] and a smeared
crack model [continuum model]. The discrete crack model uses finite elements at which concrete
cracks are separately represented as boundaries. In the smeared crack model, concrete cracks are
assumed to be scattered and distributed, such that discrete elements are not used at the crack
locations. The smeared crack model assumes that locally generated cracks are evenly scattered over
a wide surface. This model is known to be suitable for reinforced concrete structures with reasonable
amount of reinforcement, and its finite element modelling is relatively simple (Vechchio and
Collin,1986). The smeared crack model can be classified into orthogonal and non-orthogonal
crack models depending on the assumption of angles of crack development. The orthogonal crack
model assumes orthogonal crackdirections, whereas the non-orthogonal crack model assumes
non-orthogonal directions of cracks.Also, depending on the numerical analysis methods for cracks,
the smeared crack model is furtherclassified into various models such as a decomposed-strain model
and a total strain model.

The decomposed-strain model in the smeared crack model calculates the total strain in terms of
material strain and crack strain.

The total strain model in the smeared crack model can be rather simply formulated using total strain
without having to decompose it into the strain components. MIDAS, 2015 uses the total strain
crackmodel classified under the smeared crack model. It provides two methods, which are
separated intothe fixed crack model and the rotating crack model depending on the reference
crack axes. Theformer assumes that the axes of cracks remain unchanged once the crack axes are
defined. On thecontrary, the latter is a method in which the directions of the cracks are assumed
to continuouslyrotate depending on the changes in the axes of principle strains.

An iterative scheme is used for concrete crack analysis because of its nonlinearity. In order to satisfy
the equilibrium between external and internal force vectors, one of the incremental iterative
procedures such as the Newton-Raphson method is used. To this end, the constitutive model needs to
be defined by a proper stiffness matrix. MIDAS,2015 uses the secant stiffness and tangent
stiffnessapproaches to determine the stiffness matrix. The secant stiffness approach is especially
suitable forfinding excellent and stable solutions to analyses of reinforced concrete structures,
which widelydevelop cracks. On the contrary, the tangent stiffness approach is known to be very
appropriate foranalyses of local cracking or crack propagation. The secant approach is used
according to the

In this study, Total Crack strain model with configuration of fixed crack model including secant
stiffness, lateral crack effect and confinement effect was used.

388
 5th International Symposium on Advances in Civil and Environmental Engineering
Practises for Sustainable Development (ACEPS-2017)

The compression behavior and tension softening of reinforced concrete material are represented by
Thorenfeldt and Hordijk models as shown in figures 01 and 02 respectively. It is important to note that
the shear stress-strain relationship used in this study is derived by using the Modified Compression
Field Theory for a membrane element.

Shear stress

Shear
strain

Figure 1 Thorenfeldt model Figure 2 Hordijk model Figure 3 Trilinear shear curve

3. FINITE ELEMENT MODEL DEVELOPMENT

3.1 Modelling of simply supported beams

Two simply supported beams called specimen 01 and 02 having span to depth ratio of 2.1 and
longitudinal reinforcement percentage of 1.8 were selected from an experimental programme carried
out at structural laboratory. The percentage of shear reinforcement is 0.3 for specimen 01 and it is 0.8
for specimen 02. The dimensions of the two specimens were same having 1700 mm in length, 350
mm in depth and 200 mm in width.

In this study, the finite element models of the selected beams were developed in MIDAS FEA
programme. Solid elements were used to model the beam and reinforcement elements were used to
model longitudinal and transverse reinforcement as shown in Figure 04. Von-mises yield criteria was
used for steel material and Total Crack strain model was used for concrete material. The total crack
strain model uses fixed crack model with secant stiffness matrix considering lateral crack effect and
confinement effect.

The tension softening and compression behaviour of reinforced concrete material are represented by
Hordijk and Thorenfeldt models, respectively. It is important to note that the shear stress-strain
relationship used in this study is derived by using the Modified Compression Field Theory for a
membrane element. Nonlinear static analysis is performed to obtain the load deflection response of
the selected beams by using displacement based load control with 0.1 mm increments rather than the
force based load control. The displacement increment is applied at the middle of the beam span as in
the experiments.

389
 5th International Symposium on Advances in Civil and Environmental Engineering
Practises for Sustainable Development (ACEPS-2017)

Figure 4 Model of simply supported beam

3.2 Modelling of concrete frame structure

After validating the model, it was used to investigate the effect of support conditions on shear strength
of beams in monolithic construction. Same beams were used to model frames selected for this study.
But the column dimensions and the reinforcement details were selected such that the moment
capacity of reinforced concrete columns are greater than the moment capacity of reinforced concrete
beam section. Each beam section was modeled with columns having shear reinforcement percentage
of 0.31. Longitudinal reinforcement percentage of column specimen was 1.4 with a section of
300*300mm and having 4*T20 bars. Height of columns were taken as 3000mm.

Figure 5 Model of frame structure

4. RESULTS AND DISCUSSION

4.1. Shear stress-strain curve

Shear stress-strain curve is derived for each specimen used in study by using Modified Compression
Field Theory. It is identified that the grade of concrete, the percentage of longitudinal and transverse
reinforcement are the key factors affecting the shear stress-strain relationship. Figure 06 illustrates the

390
 5th International Symposium on Advances in Civil and Environmental Engineering
Practises for Sustainable Development (ACEPS-2017)

resultant shear stress-strain curve for specimen 01. It is a trilinear curve indicating that the elastic limit
of shear stress is 2.3 MPa.

3.5
3
2.5
2
1.5
1 Tri-linear
shear curve
0.5
0
0 0.005 0.01 0.015 0.02
Shear strain
Figure 6 Shear stress strain relationship for specimen 01
4.2. Comparison of experimental and numerical predictions of simply
supported beams
Figure 07 illustrates the resultant distribution of principle tensile strain at the peak resisting loads in
specimen 01 and 02. Highly concentrated principle strain at the mid depth near the support as shown
in Figure 07 indicates the initiation of shear crack and its direction of propagation under the peak
resisting load.

Figure 7 Principal strains at failure Figure 8 Experimental crack pattern

The angle of inclination of the diagonal cracks in the specimens 01 and 02 are about 450, in referring
the principle strain directions. Figure 09 and 10 illustrates the comparison of experimental and
numerical load deflection curves for specimen 01 and 02 respectively. It is clear from the load
deflection curves that shear strength of the two specimens are predicted accurately.

Figure 9 Comparison of experimental and Figure 10 Comparison of experimental and

numerical results of specimen 01 numerical results of specimen 02

When considering specimen 01, peak resisting load and the displacement at peak load are almost the
same in numerical and experimental results. Even though peak resisting load is predicted well for
specimen 2 post peak response of load deflection is significantly different.
Figure 11 shows the reinforcement stresses at failure of the beam. Initially bottom reinforcement of the
simply supported beam gets subjected to tensile stresses. But with the increment of load, shear
reinforcement yields and beam fails in shear.

391
 5th International Symposium on Advances in Civil and Environmental Engineering
Practises for Sustainable Development (ACEPS-2017)

Figure 11: Reinforcement stresses at failure

4.3 Effects of support condition on the shear strength of a beam
Figure 12 and 13 illustrates the comparison of load deflection curves for beam specimen 01 with
simply supported condition and in frame with monolithic connection with column. It is clear from the
load displacement curves that there is an increment of peak resisting load in joints when comparing
with the corresponding beams.

Figure 12: Result comparison of joint Figure 13: Result comparison of joint
specimen 01 specimen 02

When comparing the direction of principle strains of the beam and corresponding joint, variation of
crack angle can be observed.

Figure 14: Principal strains of the joint at failure

392
 5th International Symposium on Advances in Civil and Environmental Engineering
Practises for Sustainable Development (ACEPS-2017)

Figure 15: Reinforcement stresses at failure

Figure 15 illustrates yielding of shear reinforcement before yielding the longitudinal reinforcement.

5. CONCLUSIONS

When predicting the shear failure of reinforced concrete members by total crack strain model, results
were very sensitive for the defined shear stress strain relationship.
However, by comparing with experimental data, it can be concluded that Total Crack strain model with
configuration of fixed crack model including secant stiffness, lateral crack effect and confinement
effect and shear stress strain curve derived by modified compression field theory can accurately
predict the shear strength, location of cracks and initial stiffness of reinforced concrete beams.
There is about 20% increment of shear strength when comparing the load displacement curve of
beam column joint with the load displacement curve of corresponding beam. The main reason for this
increase is the moment shear interaction in the beam column joint

6. REFERENCES
Pang X., Hsu T.T.C., 1995. "Behavior of Reinforced Concrete Members in Shear." ACI
Structural Journal 665-679.
Collins M.P., Mitchell D., Adebar P., Vecchio F.J. 1996. "A General Shear Design Method."
ACI Structural Journal 36-45.
M.Hawkins, N., & A.Kuchma, D. (2005). Simplified Shear Design of Structural Concrete
Members. Urbana: American Association of State Highway and Transportation Officials.

Ritter W.,1899. Construction Techniques of Hennebique. Zurich: Schweizerische Bauzeitung

Vecchio F.J., Collins M.P., 1986. "The modified compression field theory for reinforcedconcrete
elements subjected to shear." ACI journal 219-231.
Selby R.G., Vecchio F.J. 1993. "Three Dimentional Constitutive Relations for Reinforced
Concrete." Toronto, Canada.
midas FEA (Advanced

393