yahoo!

not general

PLAXIS

On the Use of the ShotCrete UDSM

for Modelling Concrete

2016
 Edited by:

R. Witasse
PLAXIS bv, The Netherlands
 TABLE OF CONTENTS

TABLE OF CONTENTS

1 Introduction 5

2 General assumptions 7

3 Model validation for plain concrete and reinforced concrete 9

3.1 Indirect tension in a notched beam 9
3.2 Analysis of a notched beam mixed fracture mode 12
3.3 Analysis of a moderately reinforced concrete beam 15

4 Conclusions 19

5 References 21

PLAXIS 2016 | On the Use of the ShotCrete UDSM for Modelling Concrete 3
ON THE USE OF THE SHOTCRETE UDSM FOR MODELLING CONCRETE

4 On the Use of the ShotCrete UDSM for Modelling Concrete | PLAXIS 2016
 INTRODUCTION

1 INTRODUCTION

The aim of the report is to evaluate the applicability of the recently developed shotcrete
model by Schadlich & Schweiger (2014) for plain and reinforced concrete structures.
We will very briefly explain how the model parameters must be set for dealing with plain
and reinforced concrete structures. Then we will go through three different validation
exercises corresponding to respectively.
• Indirect tension of a notched beam
• Mixed fracture mode of a notched beam
• Three point bending test of a reinforced concrete beam
For each example, the model geometry and chosen material parameters will be provided
and most relevant results will be commented and compared against experimental data.

PLAXIS 2016 | On the Use of the ShotCrete UDSM for Modelling Concrete 5
ON THE USE OF THE SHOTCRETE UDSM FOR MODELLING CONCRETE

6 On the Use of the ShotCrete UDSM for Modelling Concrete | PLAXIS 2016
 GENERAL ASSUMPTIONS

2 GENERAL ASSUMPTIONS

The theoritical formulation of the shotcrete model is given in Schadlich & Schweiger
(2014). An overview of the model parameters is given in Table 2.1. Application to 'mature'
plain concrete and reinforced concrete under static loading can be achieved by getting rid
of all time dependent features in the model.
Table 2.1 Model parameters

Parameter Description Unit

E28 Young's modulus of cured shotcrete at thydr stress
ν Poisson's ratio -
fc,28 Uniaxial compressive strength of cured shotcrete at thydr stress
ft,28 Uniaxial tensile strength of cured shotcrete at thydr stress
◦
ψ Dilatancy angle
E1 /E28 Time dependency of elastic stiffness -
fc,1 /fc,28 Time dependency of strength -
fc0,n Normalised initally mobilised strength -
fcf ,n Normalised failure strength (compression) -
fcu,n Normalised failure strength (compression) -
pcp Uniaxial plastic failure strain at 1h, 8h, 24h -
Gc,28 Compressive fracture energy of cured shotcrete at thydr force/length
ftu,n Ratio of residual vs. peak tensile strength -
Gt,28 Tensile fracture energy of cured shotcrete at thydr force/length
Leq Equivalent length (if no regularisation is used) length
'
a Increase of cp with increase of p length
◦
φmax Maximum friction angle
φcr Ratio between creep and elastic strains -
cr
t50 Time for 50% of creep strains time
shr
∞ Final shrinkage strain -
shr
t50 Time for 50% of shrinkage strains time
γfc Safety factor for compressive strength -
γft Safety factor for tensile strength -
thydr Time for full hydration (usually 28 days) time

PLAXIS 2016 | On the Use of the ShotCrete UDSM for Modelling Concrete 7
ON THE USE OF THE SHOTCRETE UDSM FOR MODELLING CONCRETE

For the example considered in this report, the following parameters have been modified
to remove any time dependency (Table 2.2):
Table 2.2 Model parameters used in this example

E28 = E
fc,28 = fc
ft,28 = ft
E1 /E28 = 1
fc,1 /fc,28 = 1
pcp,1h = pcp,24h = pcp
Gc,28 = Gc
Gt,28 = Gc
shr
∞ =0
a = 18
φmax = 37◦

8 On the Use of the ShotCrete UDSM for Modelling Concrete | PLAXIS 2016
MODEL VALIDATION FOR PLAIN CONCRETE AND REINFORCED CONCRETE

3 MODEL VALIDATION FOR PLAIN CONCRETE AND REINFORCED CONCRETE

3.1 INDIRECT TENSION IN A NOTCHED BEAM

3.1.1 TEST PRESENTATION

To gain further insight of the model capabilities to describe non linear concrete behaviour,
reference is made to a bending problem of indirect tension in a notched beam. The
dimensions have been reported in Figure 3.1 and correspond to a specimen tested by
Kormeling & Reinhardt (1969) whose results have been reported by Rots (1988).
Table 3.1 Adopted material parameters for the indirect tension of a notched beam

Ec (GPa) ν (-) ft (MPa) Gt (N/mm) ftu,n (-)

20 0.2 2.4 0.11 0.0

The mesh is composed of 6-node plain strain elements. The set of material parameters
which have been adopted for the present indirect tension test are summarized in Table
3.1. The crack band width h is automatically detected by PLAXIS.

Figure 3.1 Notched beam presentation (dimension in mm)

Figure 3.2 Adopted mesh

3.1.2 MAIN RESULTS

The computed load-deflection curve has been displayed in Figure 3.3 along with available
experimental data that have been reported by Rots (1988). It can be concluded that there

fcc (MPa) pcp (-) Gc (N/mm) fc0,n (-) fcf ,n (-) fcu,n (-)
35 -1.0 × 10−3 100 0.15 0.0 0.0

PLAXIS 2016 | On the Use of the ShotCrete UDSM for Modelling Concrete 9
ON THE USE OF THE SHOTCRETE UDSM FOR MODELLING CONCRETE

is a good match between numerical results and experimental values with a maximum
load which is well predicted. The post-peak response tends to show a slightly more
ductile behaviour than in reality.

Figure 3.3 Compared load-deflection curve

The analysis results in terms of load-deflection are sensitive to the fracture-related

parameters Gt and ft . In this context the same analysis has been carried out using
different fracture energy values and a perfect match have been obtained using Gt = 0.09
N/mm as shown in Figure 3.4. It was initially done by Rots (1988), hence we use σ
= 0.11.
Figure 3.5 displays the contour plot level of the normalised tension hardening parameter
at a deflection of 1mm. The contour plot shows a localisation of the shear deformation
along a narrow band above the notch. This localised shear band corresponds to the
development of a macro-crack as observed experimentally.

10 On the Use of the ShotCrete UDSM for Modelling Concrete | PLAXIS 2016
MODEL VALIDATION FOR PLAIN CONCRETE AND REINFORCED CONCRETE

Figure 3.4 Load-deflection curve using Gt = 0.09 N/mm

Figure 3.5 Contour plot of normalised tension hardening parameter at 1 mm deflection

PLAXIS 2016 | On the Use of the ShotCrete UDSM for Modelling Concrete 11
ON THE USE OF THE SHOTCRETE UDSM FOR MODELLING CONCRETE

3.2 ANALYSIS OF A NOTCHED BEAM MIXED FRACTURE MODE

This example is a reproduction of the experimental test performed by Arrea & Ingraffea
(1981).

Figure 3.6 Main dimensions of the notched beam in mixed mode failure (dimensions in mm)

Figure 3.7 Adopted mesh for the notched beam in mixed mode failure

The geometry of the notched beam along with loading conditions to induce a mixed
fracture mode are presented in Figure 3.6. The concrete beam is 156 mm wide. As it can
be seen on this figure, a steel beam has been used to transmit the loads to the concrete
beam. However, the beam has not been included in the mesh. Instead the loading has
been applied at the points A and B. The adopted mesh is presented in Figure 3.7. The
material data have been summarized in Table 3.2.
Table 3.2 Adopted material parameters for the notched beam in mixed fracture mode

Ec (GPa) ν (-) ft (MPa) Gt (N/mm) ftu,n (-)

24.8 0.18 2.8 0.07 0.0

3.2.1 MAIN RESULTS

The main results obtained in this context are presented in Figure 3.8, Figure 3.9 and
Figure 3.10. Figure 3.8 presents the load versus CMSD curve obtained (CMSD Crack
Mouth Sliding Displacement being the vertical displacement increment between two
points located on each side of the notch). The numerical results also show good
agreement with available experimental data. Load-displacement curve obtained in this
context seems to be indicating that the Young’s modulus reported by Arrea & Ingraffea

12 On the Use of the ShotCrete UDSM for Modelling Concrete | PLAXIS 2016
MODEL VALIDATION FOR PLAIN CONCRETE AND REINFORCED CONCRETE

fcc (MPa) pcp (-) Gc (N/mm) fc0,n (-) fcf ,n (-) fcu,n (-)
−3
30 -1.3 × 10 70 0.5 0.0 0.0

Figure 3.8 Load F versus CMSD of single-notched shear beam

Figure 3.9 Load F versus CMSD of single-notched shear beam for stiffer concrete material (E = 30
GPa)

(1981) was slightly too low for obtaining a better match regarding the initial slope of the
load-displacement curve. An additional run has therefore been carried out considering a
larger Young's modulus E = 30 GPa. It can be seen in Figure 3.9 that an even better
match between numerical results and experimental values can then be obtained.

PLAXIS 2016 | On the Use of the ShotCrete UDSM for Modelling Concrete 13
ON THE USE OF THE SHOTCRETE UDSM FOR MODELLING CONCRETE

a. Deformed mesh

b. Cracked points

c. Normalized tension hardeing parameters

d. Experimental results of Area and Ingraffea (Arrea & Ingraffea (1981))

Figure 3.10 Fracture localisation at end-of-calculation

Figure 3.10 presents the deformed mesh (a), the cracking status (b) and the shear strain
(c) respectively at the end of the calculation (CMSD roughly equals to 0.35 mm) and give
a good overview on fracture localisation occuring in the notched beam.

14 On the Use of the ShotCrete UDSM for Modelling Concrete | PLAXIS 2016
MODEL VALIDATION FOR PLAIN CONCRETE AND REINFORCED CONCRETE

3.3 ANALYSIS OF A MODERATELY REINFORCED CONCRETE BEAM

A simply-supported reinforced concrete beam tested by (Burns & Siess, 1962) and
reported by Kwak & Filippou (1996) has been investigated. In this example a moderately
reinforced concrete beam under three-point displacement driven bending is analyzed.
The geometry of the reinforced concrete beam is given in Figure 3.11. For symmetry
reasons only half of the beam has been modelled here.

Figure 3.11 Main dimensions of the RC beam (dimensions in mm)

In this case, the concrete was modelled by 6-noded plane strain elements and the
reinforcement was modelled by a 3-node plate element using equivalent properties.
Material properties for the concrete and the steel are summarized in Table 3.3 and Table
3.4 respectively.
Table 3.3 Adopted concrete material parameters for the RC beam

Ec (GPa) ν (-) ft (MPa) Gt (N/mm) ftu,n (-)

24.1 0.1 2.4 0.18 0.0

Figure 3.12 presents a comparison between the experimental and computed

load-displacement curve. The current model provides good agreement with the
experimental data along with closely matching failure load (155 kN for our model against
160 kN for the experimental value). The three characteristic parts of the
load-displacement curve (elastic response up to 40 kN, tension stiffening from 40 kN to
155 kN and failure due to rebar yielding beyond 155 kN) are also properly reproduced
from both qualitative and quantitative points of view.
At failure load, the contour plots of the deviatoric strains (Figure 3.13) provides an insight
of the crack pattern whereas the distribution of normal force within the plate (Figure 3.14)
clearly highlights the rebar yielding at mid-span.
In the framework of this study, another strategy has been investigated for the modelling of
rebars by using embedded beams (which in some cases might be more convenient to be
introduced in a FE mesh with moreover possible consideration of bond-slip behaviour
between concrete and rebar).
Results obtained in this context are presented in Figure 3.15. The consideration of
bond-slip behaviour between concrete and rebar allows a slightly better description of the

fcc (MPa) pcp (-) Gc (N/mm) fc0,n (-) fcf ,n (-) fcu,n (-)
−3
30 -1.4 × 10 70 0.15 0.0 0.0

PLAXIS 2016 | On the Use of the ShotCrete UDSM for Modelling Concrete 15
ON THE USE OF THE SHOTCRETE UDSM FOR MODELLING CONCRETE

Table 3.4 Adopted steel material parameters

ρ (-) Ec (GPa) fy (MPa) EA (N/mm) Np (kN/m)

6
0.99 % 203 310 1.02.10 1557

Figure 3.12 Compared load-deflection curve for the RC beam

Figure 3.13 Deviatoric strain at last converged step

Table 3.5 Adopted EP parameters

ρ (-) Ec (GPa) EA (N/mm) Np (kN/m) Tskin (kN/m)

6
0.99 % 203 1.02.10 1557 800/1000

tension stiffening effect although at the lowest value of the Tskin calculation diverges
prematurely before reaching the rebar yielding phase.

16 On the Use of the ShotCrete UDSM for Modelling Concrete | PLAXIS 2016
MODEL VALIDATION FOR PLAIN CONCRETE AND REINFORCED CONCRETE

Figure 3.14 Axial force in rebar at last converged step

Figure 3.15 Compared load-deflection curve for the RC beam

PLAXIS 2016 | On the Use of the ShotCrete UDSM for Modelling Concrete 17
ON THE USE OF THE SHOTCRETE UDSM FOR MODELLING CONCRETE

18 On the Use of the ShotCrete UDSM for Modelling Concrete | PLAXIS 2016
 CONCLUSIONS

4 CONCLUSIONS

The report is to evaluate the applicability of the recently developed shotcrete model by
Schadlich & Schweiger (2014) for plain and reinforced concrete structures.
In this context the following application examples have been considered:
• Indirect tension of a notched beam
• Mixed fracture mode of a notched beam
• Three point bending test of a reinforced concrete beam
For each example, numerical results have shown very good agreement against available
experimental data which demonstrates the perfect applicability of the Shotcrete
user-defined model for modelling the behaviour of 'mature' plain concrete and reinforced
concrete under static loading in PLAXIS.

PLAXIS 2016 | On the Use of the ShotCrete UDSM for Modelling Concrete 19
ON THE USE OF THE SHOTCRETE UDSM FOR MODELLING CONCRETE

20 On the Use of the ShotCrete UDSM for Modelling Concrete | PLAXIS 2016
 REFERENCES

5 REFERENCES

[1] Arrea, M., Ingraffea, A.R. (1981). Mized mode crack propagation in mortar and
concrete. Rep. 81-13. Department of Structural Engineering, Cornell University.
Ithaca, New York.
[2] Bazant, Z.P., Oh, B.H. (1983). Crack band theory. Materials and Sciences, RILEM
16(93), 155–177.
[3] Bouzaiene, A., Massicotte, B. (1997). Hypoelastic tridimensional model for
nonproportional loading of plain concrete. Journal of Engineering Mechanics, ASCE,
123(11), 1111–1120.
[4] Burns, N.H., Siess, C.P. (1962). Load-deformation characteristics of beam-column
connections in reinforced concrete. Civil Engineering Studies, SRS no. 234.
Universty of Illinois, Urbana.
[5] Feenstra, P.H., Rots, J.G., Arnesen, A., Teigen, J.G., Hoiseth, K.V. (1998). A 3d
constitutive model for concrete based on a co-rotational concept. Computational
Modelling of Concrete Structures. de Borst, Bicanic, Mang and Meschke. Rotterdam,
13–22.
[6] Jirasek, Z.P.B.M. (2002). Nonlocal integral formulations of plasticity and damage:
Survey of progress. Journal of Engineering Mechanics, ASCE, 128(11), 1119–1149.
[7] Kormeling, H.A., Reinhardt, H.W. (1969). Determination of the fracture energy of
normal concrete and epoxy modified concrete. Delft University of Technology.
[8] Kupfer, H., Hilsdorf, H.K., Rusch (1969). Behaviour of concrete under biaxial
stresses. ACI Journal, 66(8), 656–666.
[9] Kwak, H.G., Filippou, F.C. (1996). Nonlinear fe analysis of r/c structures under
monotonic loads. Computer and Structures, 65(1), 1–16.
[10] Rots, J.G. (1988). Computational modelling of concrete fracture. PhD Thesis. Delft
University of Technology.
[11] Schadlich, B., Schweiger, H.F. (2014). Shotcrete model. Internal report:
Implementation, validation and application of the shotcrete model.Computational
Geotechnics Group, Institute for Soil Mechanics and Foundation Engineering, Graz
University of Technology.

PLAXIS 2016 | On the Use of the ShotCrete UDSM for Modelling Concrete 21
ON THE USE OF THE SHOTCRETE UDSM FOR MODELLING CONCRETE

22 On the Use of the ShotCrete UDSM for Modelling Concrete | PLAXIS 2016