BEHAVIOUR OF HIGH STRENGTH CONCRETE UNDER COMBINED BENDING AND COMPRESSION

billowycookieΠολεοδομικά Έργα

29 Νοε 2013 (πριν από 3 χρόνια και 11 μήνες)

89 εμφανίσεις

587
BEHAVIOUR OF HIGH STRENGTH CONCRETE UNDER COMBINED
BENDING AND COMPRESSION
Dr. Miguel A. Vicente
School of Civil Engineering
University of Burgos
Spain
Dr. Dorys C. González
Structural Technology Laboratory of Cantabria
Lecturer, School of Civil Engineering of Burgos
University of Burgos
Spain
Dr. Germán Gutiérrez
Professor at the School of Civil Engineering and
Director of the Structural Technology Laboratory of Cantabria
Spain
Engr. Achim Lichtenfels
Research Assistant,
University of Stuttgart
Germany
CBM-CI International Workshop, Karachi, Pakistan
Dr. M. A. Vicente et.al
ABSTRACTThis paper describes the development of a computerized closed-loop testing facility forexperimentally obtaining the stress-strain behaviour of high strength concrete specimensunder combined compression and bending. Using this closed-loop computerized testing
facility, 240 test specimens were tested to failure. The test variables were geometry or
shape of specimens, age at testing and the concrete strength. The computerized closed-loop
testing facility assures that the vertical load is at the edge of the central core during the test,
thus ensuring zero strain at one extreme face. This mimics the strain distribution in the
compression zone. The effect of cross-section shape and the strength of concrete on the
compression zone is the most significant result obtained during the tests.
1.INTRODUCTION
Several studies have focused on the stress distribution of concrete in the compression zonefor concrete with normal as well as high strengths [1-14]. It is widely accepted that theultimate resistance of reinforced or prestressed structural conc rete members depends largely
588
on the concrete compression zone as it generates the compression force to satisfy force
equilibrium. The knowledge of the shape of the concrete compression zone is essential for
non-linear sectional analysis based on strain compatibility and force equilibrium. A more
realistic representation of concrete compression dependent on the shape of the cross-section,the strength of concrete would lead to an improved sectional analysis and better understanding
of the failure mechanisms. This in turn results in accurate assessment of the structural safety
of concrete structures.
A computerized closed-loop testing facility was developed for experimentally obtaining the
stress-strain behavior of high strength concrete specimens under combined compression
and bending. Two hundred and forty test specimens were tested to failure. The test variables
were geometry or shape of test specimens, age at testing, and the concrete strength. Concretewith strengths of up to 105 MPa (15,000 psi) were tested.
The computerized closed-loop testing facility assures that the vertical load is at the edgeof the central core during the test, thus ensuring zero strain a t one extreme face. This mimics
the strain distribution in the compression zone. The test specimens are subjected to an
eccentrically applied compression load. The position of the vertical load is determined by
the location of extreme fiber in tension that is maintained to be at zero strain during the
testing. The maintenance of the triangular strain distribution with increasing load, gives
the depth of the neutral axis for a section under bending (assuming linear strain distribution).
The testing facility is different from conventional compression testing machines normally
used for testing concrete specimens under axial compression. This facility was designed
to facilitate the testing of concrete under combined compression and bending. These tests
provided a better understanding of the failure mechanisms in concrete specimens subjected
to combined compression and bending. The effects of cross-section shape, the magnitude
of the compression load and the strength of concrete on the concrete compression zone aresignificant results obtained during the tests.
The results show that for rectangular compression zone, the maximum usable compressivestrain ( cu) at ultimate decreases with the increase in compressive strength of concrete andthe magnitude of the axial load and ranges from 00035 to 0.0023. For non-rectangular(circular) compressed zone, the maximum usable compressive strain ( cu) at ultimatedecreases with the increase in compressive strength of concrete and the magnitude of theaxial load and ranges from 0.0038 to 0.0030.
2.EXPERIMENTAL INVESTIGATION
To study the behaviour of concrete specimens under combined compression and bending,
a computerized closed-loop testing facility was designed and developed. The testing facility
is unlike the conventional test machines used for testing concrete specimens under compression.
CBM-CI International Workshop, Karachi, Pakistan
Dr. M. A. Vicente et.al
589
The computer-controlled testing facility increases the vertical load over time whileautomatically adjusting the horizontal position of the test specimen to maintain zero strainat the extremely tension fiber under increasing vertical load ( Figures 2.1 and 2.2). The
computer-controlled test facility can be divided into three major parts:
· Loading region
· Displacement region
· Data acquisition system
In the “loading region” a single-effect vertical jack with a capacity of 2.5 MN (562.5 kip)
increases the vertical load over time. A rocker located between this jack and the testing
facility allows the test specimen to move horizontally. The test specimens are placed
between two special steel plates and two pairs of special steel rollers.
The “displacement region” is composed of a double-effect horizontal jack and a complex
system of struts, attached to the two special steel plates. It allows the specimens to movehorizontally during the application of the vertical load.
With this computerized closed-loop testing facility the strain at the extreme fiber is maintained
at zero under an increasing axial load. The vertical load is also computer-controlled. The
loading speed was constant at 0.5 kN/sec (0.1125 kip/sec).
CBM-CI International Workshop, Karachi, Pakistan
Dr. M. A. Vicente et.al
Figure 2.1 Schematic diagram of computerized close loop test facility.
590
Figure 2.2 View of the computerized closed-loop test facility.
The various cross-section shapes tested are show in Figure 2.3. The four values recordedcontinuously were the magnitude of the vertical load (P), eccentricity of the vertical load,defined as distance between the loading application point and the most compressed fiber(e), the maximum compressive strain at extreme fiber ( ) and zero (near zero) strain at thetensile fiber ( 0).
The data acquisition system samples every 2 seconds, except for the tensile fiber value,
which is sampled at 6 data points per second. Figure 2.4 outlines the testing procedure.
CBM-CI International Workshop, Karachi, Pakistan
Dr. M. A. Vicente et.al
591
Figure 2.3 Cross-section shapes of test specimens.
CBM-CI International Workshop, Karachi, Pakistan
Dr. M. A. Vicente et.al
592
Figure 2.4 Outline of the testing procedure.
A newly developed special extensometer (Figure 2.5) was used for measuring thedisplacements at the extreme compressed fiber and the extreme te nsion fiber. These
displacements measured over a gage length were used to compute the strains.
CBM-CI International Workshop, Karachi, Pakistan
Dr. M. A. Vicente et.al
Figure 2.5 The extensometer used for deformation measurements.
593
3.TESTING PROGRAM
Two hundred and forty test specimens were tested using the computer-controlled testing
facility. In addition, 60 cylindrical test specimens (100x200 mm) (3.94x7.88 in) were tested
under axial compression to obtain the strength.
The variables of the test program were compression strength, shape of the cross-section and
age at testing. Five different compressive strengths ranging between 55 and 105 MPa (7850
and 15000 psi) were tested. The shape of cross-section included rectangular, semicircular,
triangular with zero strain at the side and triangular with zero strain at the edge (Figure
2.3). The tests were conducted at ages of 3, 7, 28 and 90 days.
Table 1 shows the number of test replicates, the strength obtained fck, the shape of cross-section, age at testing, the maximum load applied, the corresponding eccentricity (e) and
the corresponding maximum compressive strain at extreme fiber ().
4.ANALYSIS OF THE TESTING RESULTS
The test specimens are divided into two main groups:
· Group 1: The rectangular section and the triangular section with zero strain at the edge.
· Group 2: The semicircular section and the triangular section with zero strain at the side.
The results of these two groups of specimens are described next.
4.1 TEST RESULTS – GROUP 1 SPECIMENS
Figures 4.6 to 4.8 show the typical results of testing group 1 specimens. At the beginning
of the test the vertical load remains static at the theoretical border of the central core until
the maximum strain reaches approximately 0.001. Then, the vertical load starts to move
towards the centre of gravity. The value of the global horizontal movements depends on:
· Concrete compression strength: The higher the compression strength, the smaller
the global horizontal movement.
· Testing age: The more mature the concrete, the smaller the global horizontal movement.
· Shape of cross-section: The rectangular section exhibits a greater global horizontal
movement than the triangular section with zero strain at the edge.
This concurs with the results obtained by Rüsch and Hognestad in the 1960’s [1-5]. Under
a small vertical load, concrete exhibits elastic behaviour. A three-dimensional stress field
develops within the test specimen. The elastic behaviour continues until the vertical loads
reach a certain value. Then two damage regions develop, in this case at the top and at the
bottom of the most compressed side (Figure 4.9).
An increasing vertical load leads to an expansion of the damage zone. The damage moves
towards the inside of he test specimen, thus weakening the compressed region and causing the
CBM-CI International Workshop, Karachi, Pakistan
Dr. M. A. Vicente et.al
594
Table 1 Test Results
CBM-CI International Workshop, Karachi, Pakistan
Dr. M. A. Vicente et.al
595
vertical load to move towards the centre of gravity. Failure occurs at the interface between
the two damage zones. This concurs with Bazant’s investigations [8].
Figure 4.6 shows that for rectangular compression zone, the maximum usable compressivestrain ( cu) at ultimate decreases with the increase in compressive strength of concrete andthe magnitude of the axial load and ranges between 00035 to 0.0023.
CBM-CI International Workshop, Karachi, Pakistan
Dr. M. A. Vicente et.al
fck (MPa)
Semicircular
596
Figure 4.6 Variation of strain at extreme compressive fiber with magnitude of vertical load.
CBM-CI International Workshop, Karachi, Pakistan
Dr. M. A. Vicente et.al
Figure 4.7 Variation of strain at extreme compressive fiber with eccentricity of load.
Strain at extreme compressive fiber
Strain at extreme compressive fiber
597
Figure 4.9 Progression of damage at various load stages for specimens of group 1.
4.2 TEST RESULTS – GROUP 2 SPECIMENS
Figures 4.10 to 4.12 show the typical results from testing group 2 specimens. As withgroup 1 specimens, initially the vertical load remains static at the theoretical edge of thecentral core up to a maximum deformation of approximately 0.001.
Then the vertical load starts to move, however, in this case the load towards the most
compressed side instead of the centre of gravity. The value of the global horizontal
movements depends on the same parameters. No previous studies of this behaviour havebeen reported.
CBM-CI International Workshop, Karachi, Pakistan
Dr. M. A. Vicente et.al
Figure 4.8 Variation of eccentricity with magnitude of vertical load
598
Figure 4.10 Variation of strain at extreme compressive fiber with magnitude of
vertical load.
CBM-CI International Workshop, Karachi, Pakistan
Dr. M. A. Vicente et.al
Figure 4.11 Variation of strain at extreme compressive fiber with eccentricity of load.
599
Figure 4.12 Variation of eccentricity with magnitude of vertical load.
Figure 4.10 shows that for non-rectangular (circular) compressed zone, the maximum usablecompressive strain ( cu ) at ultimate decreases with the increase in compressive strength ofconcrete and the magnitude of the axial load and ranges between 0.0038 to 0.0030.
At the onset of the test, the concrete exhibits an elastic behaviour until the vertical loads
reach a certain value. Then, a damage region develops inside the test specimen (Figure
4.13).
CBM-CI International Workshop, Karachi, Pakistan
Dr. M. A. Vicente et.al
Figure 4.13 Progression of damage at various load stages for specimens of group 2.
600
An increasing vertical load causes the damage zone to expand towards the most compressedside, thus weakening the internal region and causing the vertical load to move towards themost compressed side of the test specimen. Failure occurs at the interface between the twodamage zones and the most compressed side.
Tests results of group 1 and 2 test specimens show that an increase in the concrete's ductility
slows the expansion of the damage zone and leads to an increase in the global horizontal
movement of the vertical load. That is more brittle the concrete, the quicker the damage
zone expands and the smaller the global horizontal movement of the vertical load. Detailsof the test results are reported in [16 and 17].
5. CONCLUSIONS
The results of this study lead to the following conclusions:
1.A new computerized closed-loop testing facility was developed to experimentally obtain
the stress strain distribution in concretes of different strength, including high strength
concretes and different shapes of the compressed zones.
2.Depending on the shape of the cross-section, the eccentrically loaded concrete specimens
exhibit two different types of behaviour. For the Group 1 specimens: the vertical load
moves towards the center of the cross-section. For the Group 2 specimens: the vertical
load moves towards the most compressed fiber.
3.For rectangular compression zone, the maximum usable compressive strain (cu) atultimate decreases with the increase in compressive strength of concrete and themagnitude of the axial load and ranges between 00035 to 0.0023. In group 1 specimensthe damage regions occur at the top and at the bottom of the most compressed fibersand move towards the inside of the specimens. Failure occurs at the interface betweenthe two damage zones.
4.For non-rectangular (circular) compressed zone, the maximum usable compressive strain( cu) at ultimate decreases with the increase in compressive strength of concrete and the magnitude of the axial load and ranges between 0.0038 to 0.0030. In group 2 specimens,the damage regions appear inside the test specimens and move towards the mostcompressed fibers. Failure occurs when the two damage zones reach the most compressedside.
ACKNOWLEDGMENTS
The authors would like to acknowledge the assistance received from the Universities of
Cantabria and Burgos, where the testing procedure was developed. From the many Spanish
companies who collaborated with the authors in this work, special mention is to be given
to PRECOM S.A., BETTOR-MBT, SIKA, etc. This work has been financed by the European
Union.
CBM-CI International Workshop, Karachi, Pakistan
Dr. M. A. Vicente et.al
601
REFERENCES
1.McHenry. D.; Hanson, N.W.; Hognestad, E. “Concrete stress distribution in ultimate strength
design”. Journal of the American Concrete Institute. December 1955.
2.Mattock, A.H.; Kriz, L.B. Hognestad, E. “Rectangular concrete stress distribution in ultimate
strength design”. Journal of the ACI. February 1961.
3.Rüsch, H. “Versuche zur Festigkeit der Biegedruckzone. Festigkeit und Verformung des
exzentrisch gedrückten Rechteckquerschnittes aus unbewehrtem Beton bei kurzzeitiger
Lasteinwirkung”. Deutscher Ausschuss für Stahlbeton. 1955.
4.Rüsch, H.; Kordina, K.; Stöckl, S. “Festigkeit der Biegedruckzone. Vergleich von Prismen- und
Balkenversuchen”. Deutscher Ausschuss für Stahlbeton. 1967.
5.Rasch, C. “Spannungs – Dehnungs – Linien des Beton und Spannungsverteilung in der
Biegedruckzone bei konstanter Dehngeschwindigkeit”. Deutscher Ausschuss für Stahlbeton.
1962.
6.Griffith, A.A. “The theory of rupture”. Proceedings of the First International Conference of
Applied Mechanics. 1924.
7.Hillerborg, A.; Modéer, M.; Petersson, P.E. “Analysis of crack formation and crack growth in
concrete by means of fracture mechanics and finite elements”. Cement Concrete Res. 1976.
8.Bazant, J.P; Planas, J. “Fracture and size effect in concrete and other quasibrittle materials”.
Ed. CRC Press. 1998.
9.Markeset, G. “Failure of concrete under compressive strain gradients”. NTH Universitetet i
Trondheim. 1993.
10.Markeset, G. “Eccentrically loaded prisms of high strength concrete”. 4th International Symposium
on Utilization of High Strength/High Performance Concrete. Paris 1996.
11.Meyer, J. “Improving the ductility of high performance concrete”. Institüt für Massivbau und
Baustofftechnologie, Universität Leipzig. LACER Nº 3, 1998.
12.Meyer, J.; Sint, A. “Calculation of the load – deformation behaviour of over-reinforced concrete
beams with the BDZ Model”. , Universität Leipzig. LACER Nº 4, 1999.
13.Foster, S.J.; Attard, M.M. “Experimental tests on eccentrically loaded high -strength concrete
columns”. ACI Structural Journal. March 1997.
14.Ahmad, S.H.; Shah, S.P. “High Performance Concretes and Applications”. Edward Arnold
Publications. 1994.
15.Shah, S.P.; Swartz, S.E.; Ouyang C. “Fracture Mechanics of Concrete”. Wiley Interscience.
1995.
16.Vicente, M.A. “Estudio del comportamiento en flexo-compresión de los hormigones de alta
resistencia.”. Doctoral Thesis. Departamento de Ingeniería Estructural y Mecánica. University
of Cantabria. Spain. February 2001.
17.González, D.C. “Desarrollo de nuevos diagramas tensión-deformación para el dimensionamientoen E.L.U. de elementos de hormigón estructural de alta resistenc ia”. Doctoral Thesis. Departamento
de Ingeniería Estructural y Mecánica. University of Cantabria. Spain. November 2002.
CBM-CI International Workshop, Karachi, Pakistan
Dr. M. A. Vicente et.al