Upload
elfreda-knight
View
219
Download
4
Tags:
Embed Size (px)
Citation preview
Louis Demilecamps
COMPORTEMENT ET EVALUATION DES OUVRAGES SPECIAUX« Behaviour and Assessment of Special R.C. Works
Cracking & Shrinkage » www.ceosfr.org
RELIABLE SHRINKAGE AND CRACK DESIGN:Ceos French National Research Program
Experimental Aspects
Reliable Shrinkage and Crack Design: CEOS Experimental Aspects
• CEOS.fr, a French National Research Program (2008-2011 : – Budget 8M€, 40 members:
• Companies: EDF, Vinci, Eiffage, Bouygues, Areva, Solétanche-Bachy, Italcementi, Iosis, Setec, Co&B, Arcadis,
Oxand, Necs, Advitam, Sites, Chryso, Rincent, Saipem, PX-Dam… • Lab. & Institutions: LCPC, CSTB, CEA, IRSN, ANDRA, ATHIL, CERIB, CEBTP, LERM, CETU…• Universities: ENS Cachan (LMT), INP Grenoble (3S-R), INSA Toulouse (LMDC), EC Nantes (GEM), U. Pau
(Lasagec), Polytech’ Lille (LML)…
• Aims: – at dealing with the control of cracking (a major concern for durability and sustainability), in particular for
special works (specific use, specific shape and size, specific requirements for loading or durability...)
• Subjects: – Cracking under monotonic loadings – Behavior under coupling loadings (THM) – Cracking under cyclic and seismic loadings
• Approaches: – Experiments on large specimens– Numerical modeling – Elaborated by engineers-practitioners
• International connection:– International benchmark (registration until Jun.15/10)
– European network under construction
May 29, 2010 23rd fib International Congress - 2010
Objectives of the experiment
• Trials are intended to provide basic data for the conventional calculation and the scientific models:
– additional data in order to finely tune numeric simulators
– additional data on scale effects and their incidence on the validity of the digital models
– data fitting the 3 targeted fields : static, cyclic, thermo-hydro-mechanical
The experiment does not aim at bringing out or characterizing some new phenomena or behaviors
CEOS.FR Experimental Part
May 29, 2010 33rd fib International Congress - 2010
Project Contents
Trials on Test Bodies
Program:
It will be performed through three
different series :- walls (shearing tests)- blocks (bending tests)- I-shaped blocks (shrinkage
test)
May 29, 2010 3rd fib International Congress - 2010 4
DESIGN TB
FABRICATE TB
TEST TB
MODELIZE TB
GOOD ACCORDANCE
PERFORM VIRTUAL TESTING
IMPLEMENT NEW CRITERIA IN ENGINEERING CODES
IMPROVE MODEL
NO
YES
COMPARE TEST AND MODEL RESULTS
Series 1: Cyclic Loading Tests on Walls, 1/3 Scale
• Shearing of thick reinforced concrete walls under cyclic solicitation up to cracking
• Evaluation of the cracking stage during and after seism
• Result data = crack opening + crack spacing
• The experiment results must:– Be usable as a reference data base for the improvements expected in
modelling– Be added to the other available data in order to allow in fine establishing
and/or verifying the cracking formulas
May 29, 2010 3rd fib International Congress - 2010 5
Cyclic Loading Tests on the WallsConstraints of Experimentation
• The test bodies must be representative of highly reinforced thick walls
• The behaviour must be tested beyond the ULS scale reduction
• The behaviour of reduced scaled test bodies must be similar to the one it would be on a full-sized scale concrete and reduced reinforcement bars with similar behaviour
• The reinforcement ratio must be considerably superior to the non-fragility limit
• The stress state and the cracking mode due to the cyclic solicitation must be perfectly identified
May 29, 2010 3rd fib International Congress - 2010 6
• Reinforcement ratio: from 0.5% to 1%, displayed in two layers
• Maximum jacking capacity = 4 MN limited resistance of reinforcement– Test body: 0.15m thick 1.05m high 4.2m long (H/L ratio = ¼)– Representing an “actual” wall: 0.45m 3.15m 12.6m
Cyclic Loading Tests on the Walls - Proposed Choices
May 29, 2010 73rd fib International Congress - 2010
Test N° Loading method Test body
1 Cyclic and alternate up to the ruin Concrete C25 - Dmax = 8 mm
One wire mesh per face HA10 sq. mesh 100x100mm
2 Monotonous and increasing up to the
ruin
Concrete C40 - Dmax = 8 mm One wire mesh per face
HA10 sq. mesh 100x100mm
3
4 Cyclic and alternate up to the ruin
Concrete C40 - Dmax = 8 mm One wire mesh per face HA8 sq. mesh 80x80mm
• Measurement of the flexion areas by image correlation• Extensometer (vibrating cords, Bragg nets optical fibres)• Determination of the thermo-hydric parameters of concrete• Verification of the overall displacements and deformations
Cyclic Tests – Instrumentation and Loading Process
May 29, 2010 83rd fib International Congress - 2010
Chargement quasi-statique alterné
-30
-20
-10
0
10
20
30
Nombre de cycles
Dép
lace
men
t im
po
sé
(mm
)
A B C D E F G H I J
Quasi-static alternate loading
Con
trol
led
disp
lace
men
t
Quantity of cycles
Series 2 & 3: ”Scale 1” Block Tests
• General arrangement:– Making of the purpose design
test bench– Making of 7 parallelepiped test
bodies - 6.1mx1.60mx0.80m(7.8m³) for free shrinkage tests
– Making of 3 I-shaped test bodies 7.7mx0.50mx0.80m(5.5m³) for restrained shrinkage tests
• Concreting and aging (28 days) of the blocks
• Setting up and tightening to the test slab
• 4 points bending up to cracking (upwards jacking in two rows)
May 29, 2010 3rd fib International Congress - 2010 9
Free shrinkage (large block: 6,10x1,6x0,8)• RL1: reference block : reference concrete, reference reinforcement
(percentage and diameter),• RL2: reference concrete, reduced reinforcement percentage, same
diameter,• RL3: reference concrete, same reinforcement percentage, reduced
diameter,• RL4: reference concrete, changed reinforcement arrangement (different
coating: 3 and 7cm),• RL5: reference reinforcement, changed concrete (/tensile strength),• RL6: study of the variation of one block to the other => reference block
must be made once again• RL7: reference block with local and controlled defects (PVC pipes for
instance)
Restrained shrinkage (reduced size tension member ~6.10x0.80x0.50)• THM8: same parameters as for RL1• THM9: reference concrete, minimum reinforcement• THM10: reference concrete, maximum reinforcement
“Scale 1” Block Tests - Experimental Program
May 29, 2010 103rd fib International Congress - 2010
Series 3: Block Tests - Restrained Shrinkage Block
May 29, 2010 3rd fib International Congress - 2010 11
Series 2: Block Tests - Free Shrinkage Blocks
May 29, 2010 123rd fib International Congress - 2010
RL Blocks:Shape / Reinforcement
• 1st RL block cast in October 2009
Block Tests (RL1 block)
May 29, 2010 133rd fib International Congress - 2010
• 9 temperature sensors• 3 hygrometry sensors
“Scale 1” Block Tests - Instrumentation of the Blocks
May 29, 2010 143rd fib International Congress - 2010
• 16+2 vibrating cord sensors
• 9 optical cord extensometers
(3 internal, 6 external)
• 12 reinforcement strain gauges
“Scale 1” Block Tests - Instrumentation of the Blocks
May 29, 2010 153rd fib International Congress - 2010
Measurement of forces and overall deformations
“Scale 1” Block Tests - Instrumentation of the Blocks
May 29, 2010 163rd fib International Congress - 2010
4 jacks + pressure sensors
15 displacement sensors
Deformation Measurement by Image Correlation
• Instrumentation based on the techniques of image correlation. Quantities to be observed:
– The crack layout during test
– The displacement field
– The limit conditions
– The crack openings
• Size of the test body => 4 cameras for 1 face, with partial image superposition
• Camera: Canon EOS 450D type, 4272x2848 pixels, 12 bit depth(1 pixel = 5m x 3 x 4272 = 0.39mm)
• Result accuracy: approx. 1/10th pixel (pessimistic estimation)=> measurement down to 39µ
Testing bench
Test body(different according to tests)
Picture 4Picture 3 Picture 2Picture 1
4 points bending test
Principle scheme of picture superposition
May 29, 2010 173rd fib International Congress - 2010
• Specific paint pattern applied on concrete surface
Deformation Measurement by Image Correlation
May 29, 2010 183rd fib International Congress - 2010
SN4060 SN4059 SN4057 SN4051 SN4053 SN4067 SN4068 SN4069 SN4145
12/14, 12h 12/14, 14h 12/14, 16h 12/14, 18h 12/14, 20h 12/14, 22h 12/15, 0h2009
-1.00
-0.75
-0.50
-0.25
0.00
0.25
0.50
0.75
1.00
1.25
1.50
1.75
2.00
2.25
2.50
2.75
3.00
3.25
3.50
3.75
4.00mm
15ème palier - Allongement maximum
Resserrage
Correlation between cracking pattern (ULS stage) and reinforcement arrangement
Deformation Measurement by Image Correlation
May 29, 2010 193rd fib International Congress - 2010
• Approximate position of the sensors
Detection and Localization of Crack by Acoustic Analysis
May 29, 2010 203rd fib International Congress - 2010
• Classification and localization for the sensors for acoustic events
Acoustic Analysis Data and Reports
May 29, 2010 213rd fib International Congress - 2010
• Calorimetric test on cement with Langavant bottle• QAB test on concrete• Young’s modulus and Poisson’s coefficient, compression strength at
short terms (6 terms at least for a maturometric follow up)• Tensile strength (same terms)• Cracking energy • Endogenous shrinkage• Drying shrinkage• Measurement of the weight loss for one 16*32 test body under
laboratory conditions (50% relative humidity)• Creeping tests at early age under tension and compression conditions• Ring tests under endogenous conditions• Thermal dilatation coefficient• Calorific capacity of concrete or aggregates and paste
All Series: Characterization of Concrete (Scheduled Program)
May 29, 2010 223rd fib International Congress - 2010
• CEOS.fr:– End of experiment: December 2010– End of project: June 2012– More information on (mainly in French!)
http://www.ceosfr.org
• International benchmark CONCRACK:– Open to research or engineering groups– Blind stage (June 15 to December 15, 2010)– Feed-back stage (January to April 2011)– Restitution workshop in Paris (May or June 2011)
http://www.concrack.org
Conclusion
May 29, 2010 233rd fib International Congress - 2010