pression exercee sur le coffrage par le beton auto-placant formwork pressure exerted ... ·...

rjn UNIVERSITY DE

blJ SHERBROOKE

Faculte de genie Departement de genie civil

PRESSION EXERCEE SUR LE COFFRAGE PAR LE BETON AUTO-PLACANT

FORMWORK PRESSURE EXERTED BY SELF-CONSOLIDATING CONCRETE

These de doctorat es sciences appliquees

Speciality : genie civil

Membres du jury :

Kamal Khayat, directeur de these

Nicolas Roussel

Olafur Wallevik

Emmanuel Attiogbe

Joseph Assaad

Brahim Benmokrane

Ahmed Fathy OMRAN

Sherbrooke (Quebec), Canada Juillet 2009

1*1 Library and Archives Canada

Published Heritage Branch

395 Wellington Street OttawaONK1A0N4 Canada

Bibliotheque et Archives Canada

Direction du Patrimoine de I'edition

395, rue Wellington OttawaONK1A0N4 Canada

Your file Votre reference ISBN: 978-0-494-52850-1 Our file Notre reference ISBN: 978-0-494-52850-1

NOTICE: AVIS:

The author has granted a nonexclusive license allowing Library and Archives Canada to reproduce, publish, archive, preserve, conserve, communicate to the public by telecommunication or on the Internet, loan, distribute and sell theses worldwide, for commercial or noncommercial purposes, in microform, paper, electronic and/or any other formats.

L'auteur a accorde une licence non exclusive permettant a la Bibliotheque et Archives Canada de reproduire, publier, archiver, sauvegarder, conserver, transmettre au public par telecommunication ou par Plnternet, preter, distribuer et vendre des theses partout dans le monde, a des fins commerciales ou autres, sur support microforme, papier, electronique et/ou autres formats.

The author retains copyright ownership and moral rights in this thesis. Neither the thesis nor substantial extracts from it may be printed or otherwise reproduced without the author's permission.

L'auteur conserve la propriete du droit d'auteur et des droits moraux qui protege cette these. Ni la these ni des extraits substantiels de celle-ci ne doivent etre imprimes ou autrement reproduits sans son autorisation.

In compliance with the Canadian Privacy Act some supporting forms may have been removed from this thesis.

Conformement a la loi canadienne sur la protection de la vie privee, quelques formulaires secondaires ont ete enleves de cette these.

While these forms may be included in the document page count, their removal does not represent any loss of content from the thesis.

Bien que ces formulaires aient inclus dans la pagination, il n'y aura aucun contenu manquant.

1*1

Canada

ABSTRACT

Self-consolidating concrete (SCC) is an emerging technology that utilizes flowable concrete

that eliminates the need for consolidation. The advantages of SCC lie in a remarkable reduction of

the casting time, facilitating the casting of congested and complex structural elements, possibility to

reduce labor demand, elimination of mechanical vibrations and noise, improvement of surface

appearance, producing a better and premium concrete product.

While SCC has been successfully used in North America in the precast industry, a certain

number of technical issues have slowed down its use in cast-in-place applications, in particular due

to the lack of knowledge on the lateral pressure that such concrete can exert on formwork systems.

This prompts contractors and engineers, as recommended by ACI 347R-03 (Guide to Formwork for

Concrete), to design for full hydrostatic pressure, which increases drastically the cost of

construction made of SCC. This adverse effect compromises profitability, due to the need to design

for robust formwork construction and detailed joint sealing.

The research focussed on capturing existing knowledge and making recommendations for

current practice. An experimental program was undertaken at the Universite de Sherbrooke to

evaluate the lateral pressure developed by SCC mixtures. A portable devise (UofS2 pressure

column) for measuring and predicting lateral pressure and its rate of decay of SCC was developed

and validated. The UofS2 pressure column is cast with 0.5 m high fresh concrete and air pressure is

introduced from the top to simulate casting depth up to 13 m. Then, develop and implement test

method for field evaluation of relevant plastic and thixotropic properties of SCC that affect

formwork pressure were done. Portable vane (PV) test based on the hand-held vane test method

used to determine the undrained shear strength property of clay soil was the first setup as well as

the inclined plane (IP) test. The IP device involves slumping a small concrete cylinder on a

horizontal plate and then lifting up the plate at different durations of rest until the slumped sample

starts to move. Identifying role of material constituents, mix design, concrete placement

characteristics (casting rate, waiting periods between lifts, and casting depth), temperature, and

formwork characteristics that have major influence on formwork pressure exerted by SCC were

evaluated in laboratory and validated by actual field measurements. Relating the maximum lateral

pressure and its rate of decay to the plastic properties of SCC were established. In the analytical

i

part of the research, effective ways to reduce lateral pressure by developing formulation expertise

and practical guidelines to lower lateral pressure of SCC were proposed. Various design equations

as well as chart diagrams to predict formwork pressure that can be exerted by SCC on column and

wall elements were derived and reported.

In general, the results obtained show that measured lateral pressure is lower than

corresponding hydrostatic pressure. The study has shown that lateral pressure exerted by SCC is

closely related to the structural build-up at rest (or thixotropy) of SCC. The latter can be controlled

using different mixture proportionings, material constituents, and chemical admixtures. SCC

mixture with a high rate of structural build-up at rest can develop low lateral pressure on formwork.

Increased rate of structural build-up at rest can be ensured by incorporating a greater volume of

coarse aggregate, lower paste volume, and/or lower sand-to-total aggregate ratio. Incorporating

coarse aggregate of larger maximum size could also increase the thixotropy and hence reduce the

lateral pressure. This can also be achieved by reducing the workability of SCC using less HRWRA

concentration. Indeed, all mixture factors have been replaced by measuring the rate of structural

build-up at rest (or thixotropy) using the developed portable vane and inclined plane field-oriented

test as well as the modified Tattersall MK-III concrete rheometer. On the other hand, increasing or

maintaining the concrete temperature at a certain level plays an important role to reduce the lateral

pressure. The higher concrete temperature can accelerate the heat of hydration of cement with water

and increase the internal friction leading to higher thixotropy.

Controlling the placement rate has a great impact on the resultant lateral pressure of SCC.

The lateral pressure can be reduced by slowing down the casting rate, as concrete has more time to

build-up. However, this can slow down the rate of construction. The casting rate should be

optimized to yield a cost effective formwork system. Pausing the continuous casting by a waiting

period can reduce the exerted lateral pressure.

The research investigation could accelerate the acceptance and implementation of SCC

technology in cast-in-place applications, which is the preponderate business of the ready mixed

concrete suppliers. The research findings could also contribute to the removal of some of the major

barriers hindering the acceptance of SCC in cast-in-place applications and provide the industry with

much needed guidelines on formwork pressure.

ii

RESUME

Le beton auto-placant (BAP) est une technologie emergente qui utilise un beton fluide qui

elimine le besoin de consolidation. Les avantages du BAP sont la reduction remarquable du temps

de mise en place, la facilite de mise en place dans les elements structuraux encombres et

complexes, la possibility de reduire la demande de main-d'ceuvre, l'elimination des vibrations

mecaniques et du bruit et 1'amelioration de l'apparence exterieure, en produisant un beton de

meilleure qualite.

Tandis que le BAP a ete employe avec succes en Amerique du Nord dans l'industrie

prefabriquee, un certain nombre de preoccupations techniques ont ralenti son utilisation dans les

applications coulees sur place, en particulier dues au manque de connaissance sur la pression

laterale qu'un tel beton peut exercer sur des systemes de coffrage. Cela incite les entrepreneurs et

les ingenieurs, tel que recommande par l'ACI 347R-03 (Guide de coffrage pour le beton), a

concevoir les coffrages pour une pression hydrostatique pleine, ce qui augmente radicalement le

cout des constructions fait en BAP. Cet effet nuisible compromet la rentabilite, due a la necessite de

concevoir des coffrages robustes et des joints d'etancheite detailles.

La recherche s'est concentree sur le rassemblement des connaissances existantes et a emis des

recommandations pour la pratique actuelle. Un programme experimental a ete entrepris a

l'Universite de Sherbrooke afin d'evaluer la pression laterale developpee par des melanges de BAP.

Un appareil portatif (colonne de pression UdeS2) pour mesurer et prevoir la pression laterale et le

taux de diminution de la pression d'un BAP a ete developpe et valide. La colonne de pression

UdeS2 est coulee avec 0.5 m de haut de beton frais et de l'air est introduit sous pression a partir du

dessus pour simuler une profondeur jusqu'a 13 m de beton frais. Par la suite, le developpement et

l'application d'une methode d'essai pour 1'evaluation en chantier des proprietes plastiques et

thixotropiques d'un BAP, qui affectent la pression de coffrage, ont ete faits. Le premier montage

retenu est l'essai portatif de la palette, base sur l'essai du scissometre employe pour determiner la

resistance au cisaillement non drainee d'un sol d'argile. Le deuxieme montage retenu est l'essai du

plan incline (PI). Le dispositif du PI consiste a affaisser un petit cylindre de beton frais sur la

surface horizontale et ensuite de soulever cette surface vers le haut pour differents temps de repos,

jusqu'a ce que l'echantillon affaisse commence a s'ecouler. L'identification du role des materiaux,

i i i

de la conception du melange, des caracteristiques de mise en place (taux de mise en place, periodes

d'attente entre les levees et profondeur de coulee), de la temperature et des caracteristiques du

coffrage, qui ont une influence majeure sur la pression de coffrage exercee par les BAP, ont ete

evaluees au laboratoire et valides par des mesures reelles sur le terrain. Les relations entre la

pression laterale maximum et son taux de diminution par rapport aux proprietes plastiques du BAP

ont ete etablies. Dans la partie de Panalyse de la recherche, des facons efficaces de reduire la

pression laterale ont ete propose en developpant une expertise dans la formulation de BAP et en

dormant des directives pratiques. Diverses equations ainsi que des diagrammes pour predire la

pression qui peut etre exercee par des BAP sur des elements de colonne et de mur de coffrage ont

ete derivees et rapportees.

En general, les resultats obtenus demontrent que la pression laterale mesuree est inferieure a

la pression hydrostatique correspondante. L'etude a demontre que de la pression laterale exercee par

le BAP est etroitement liee a la restrucruration au repos (ou thixotropie) du BAP. Ce dernier peut

etre controle en utilisant differentes proportions de melange, differents materiaux et des adjuvants

chimiques. Un melange de BAP avec un taux eleve de restrucruration au repos peut developper une

pression laterale tres faible sur le coffrage. Un plus grand taux de restrucruration au repos peut etre

obtenu en incorporant un plus grand volume de gros granulats, en diminuant le volume de pate

et/ou en abaissant le ratio global de sable-granulat. Incorporer des granulats de plus grande

dimension permettrait aussi d'augmenter la thixotropie et done reduire la pression laterale. Ceci

peut egalement etre realise en reduisant l'ouvrabilite du BAP en utilisant moins de superplastifiant

(S.P.). En effet, tous les facteurs du melange ont ete remplaces en mesurant le taux de

restrucruration au repos (ou thixotropie) avec les essais empiriques de palette portative et de plan

incline developpes ainsi qu'avec le rheometre de beton modifie Tattersall MK-III. D'un autre cote,

l'augmentation ou le maintien de la temperature du beton a un certain niveau joue un role important

pour reduire la pression laterale. Une temperature du beton plus elevee peut accelerer l'hydratation

du ciment avec l'eau et augmenter la friction interne, menant a une thixotropie plus elevee.

Le controle du taux de placement a un grand impact sur la pression laterale du BAP. La

pression laterale peut etre reduite en reduisant le taux de mise en place, ce qui fait en sorte que le

beton a plus de temps pour se structures Cependant, ceci peut ralentir le taux de construction. Le

taux de mise en place devrait etre optimise pour arriver a un systeme de coffrage rentable. Faire une

IV

pause lors d'une coulee en continu par une periode d'attente peut reduire la pression laterale

exercee.

La presente recherche pourrait accelerer l'acceptation et 1'application de la technologie de

BAP dans les applications coulees sur place, qui est une affaire encourageante des fournisseurs de

beton pret a l'emploi. Les resultats de la recherche pourraient aussi contribuer a l'elimination des

barrieres empechant l'acceptation des BAP dans les applications coulee en place et dormer a

rindustrie des guides sur les pressions laterales.

Mots de cles

Beton auto-pla9ant (BAP), pression laterale, coffrage, pression lateral sur colonne et mur,

thixotropie, restructuration au repos, et modeles de prediction

v

DEDICATION

&Tj>?wckm< ^JzM!tity& a?&//?w afataifa&p <JMwtizm wA& mme/& #fe e#%Afen<^ dmifrw mp t^J^afe

cmt/Aa#e made a- && a^^^e/^^seJfr m^^S.

VI

ACKNOWLEDGMENTS

Foremost, praise and thanks go to my Creator and Provider (Allah) for his uncounted grace undeservingly bestowed upon me. "Glory be to You, we have no knowledge except what You have taught us. Verily, it is You, the All-Knower, the All-Wise."

The author would like to express his greatest appreciation and thanks to his supervisor Prof. Kamal Khayat for offering him the opportunity to work on such a challenging subject. The author is proud to work with him. All of the guidance, insight, helpful advice, and encouragement provided throughout the course of the thesis are greatly appreciated. The author has learned a lot from his thorough knowledge and experience as well as from his nice and decent personality. Despite his busy schedule, Prof. Khayat was always available to follow up and suggest new and bright ideas for improving the final quality and impact of the work. The Author can hardly find the right words to express the extent of his gratitude and thanks.

The Author is greatly indebted to Drs. Ammar Yahia, Trimbak Pavate, Peter Billberg, and Ahmed El-Sayed for their cooperation, consulting, and friendship during the research. The author also thanks his colleague Yacine Elaguab for his cooperation and assistance. Eng. Elaguab, during his entire master program, had worked with the author in the same field of study.

The Author also thanks the professors of Civil Engineering Department and all the faculty staff he has interacted with. Special thanks to Mrs. Christine Couture, the research secretary and Mrs. Marielle Beaudry, the department secretary. Appreciation is expressed to Mr. Claude Faucher who helped in manufacturing and solving any technical problems concerning the test setups used in entire research. The kind cooperation of all the professors, research assistants, technicians, and graduate students of the Cement and Concrete Research Group at the Universite de Sherbrooke is also acknowledged. The Author cannot forget to acknowledge all friends he has gained during his study who made the stay at Sherbrooke enjoyable. Great acknowledgement is forwarded to his colleagues at University of Northwestern and the members of CTLGroup who participated in part of the study.

The Author also wishes to thank Lafarge Laboratory Research Center, France for funding part of his research. Also, thanks to Ready-Mix Concrete Research Foundation (RMC) and American Concrete Institute-Strategic Development Council (ACI-SDC) for funding part of the research.

The Author is thankful to the members of the dissertation committee; Prof. Nicolas Roussel, Prof. Olafur Wallevik, Dr. Emmanuel Attiogbe, Dr. Joseph Assaad, and Prof. Brahim Benmokrane for their valuable comments and advices.

The author wishes to express his greatest and profound gratitude to his parents, brother, sisters, and mother-in-law for their praying to him. Finally, the author wishes to thank his wife Nancy for her useful discussions, love, and happiness she brought to him. The life at Sherbrooke would not have been so full without his wife and his smiley kids Mustafa and Mariam.

OMRAN, Ahmed

vii

CONTENTS

ABSTRACT i RESUME iii DEDICATION vi ACKNOWLEDGMENTS vii CONTENTS vui SYMBOLS AND NOTATIONS xiii

CHAPTER 1 INTRODUCTION 1 1.1 Introduction 1 1.2 Objectives 4 1.3 Thesis outline 5

CHAPTER 2 REVIEW ON FORMWORK PRESSURE AND FUNDAMENTALS OF RHEOLOGY 8

2.1 Introduction 8 2.2 Review of various recommendations for formwork design 8

2.2.1 Models proposed to evaluate formwork pressure 10 A. Rodin's models [1952] 10 B. ACI models 11 C. Models of German Standard [DIN 18218, 1980] 13 D. CIRIA 108 design models [1965 - 1978] 14 E. Gardner's models [1980 -1984] 15 F. Models of French Standard [NF P93-350, 1995] 16 G. Comparison between models 16

2.2.2 Theoretical models to predict formwork pressure 17 A. Vanhove and co-authors' model [2004] 17 B. Roussel and Ovarlez's model [2005] 18 C. Graubner and Proske's model [2005A] 23 D. Khayat and Assaad's model [2005A] 27

2.3 Relationship between form pressure and rheology of SCC 32 2.3.1 Thixotropy of cement-based materials 32 2.3.2 Concrete rheometer 35 2.3.3 Dynamic yield stress and plastic viscosity 37 2.3.4 Approaches to quantify thixotropy of concrete 38

A. Hysteresis curves 39 B. Structural breakdown curves 40 C. Apparent viscosity 43 D. Static yield stress at rest 43 E. Structure build-up at rest minus irreversible structure 45

2.3.5 Relationships between lateral pressure and rheological properties 46 2.4 Parameters affecting formwork pressure and thixotropy 46

2.4.1 Material properties 47 A. Composition and content of binder 47 B. Characteristics of coarse aggregate 49 C. Water content and w/cm 51

2.4.2 Consistency level 51 2.4.3 Placement conditions 52

A. Placement rate 52 B. Placement method 54 C. Ambient and concrete temperature 55 D. Relationship of pressure cancellation time and setting time of concrete 55

viii

2.4.4 Formwork characteristics 57 A. Formwork dimension 57 B. Type of formwork surface material 58

2.5 Lateral pressure measuring systems 59 2.5.1 Instruments and devices to monitor lateral pressure 59 2.5.2 Pore waterpressure measurements to determine lateral pressure 62

2.6 Case studies for formwork pressure exerted by SCC 67 2.7 Concluding remarks 75

CHAPTER 3 MATERIALS AND MIX DESIGNS 76 3.1 Materials 76

3.1.1 Cement 76 3.1.2 Limestone filler 77 3.1.3 Fly ash 78 3.1.4 Aggregate 79 3.1.5 Chemical admixtures 81

3.2 Mixture composition 82 3.3 Mixing and testing sequence 82

CHAPTER 4 METHODOLOGY FOR LATERAL PRESSURE MEASUREMENTS 90 4.1 Introduction 90

4.1.1 Evaluation of initial maximum formwork pressure 90 4.1.2 Evaluation of lateral pressure decay 91 4.1.3 Evaluation of formwork width on lateral pressure 91

4.2 Data acquisition 93 4.3 Measuring systems 93 4.4 Calibration of pressure sensors 94 4.5 Pressure sensor configurations (19 mm vs. 38 mm) 95

CHAPTER 5 DEVELOPMENT PORTABLE DEVICE TO MEASURE SCC FORMWORK PRESSURE 96

5.1 Introduction 96 5.2 Research significance and objectives 96 5.3 Testing program 97 5.4 Fresh concrete properties 97 5.5 Reduce concrete height in the UofSl pressure column 99 5.6 Design of new pressure device 99 5.7 Successive reduction of concrete plug in the UofS2 pressure column 101 5.8 The UofS2 pressure column vs. 3-m free standing PVC column 104 5.9 Lateral pressure decay 106 5.10 Repeatability responses of the UofS2 pressure column 108 5.11 Evaluation of key parameters affecting form pressure 109

5.11.1 Effect of casting rate 109 5.11.2 Effect of slump flow 110 5.11.3 Effect of VMA and HRWRA contents 112 5.11.4 Effect ofpaste volume 113 5.11.5 Effect of maximum-size of aggregate 115

5.12 Conclusions 116

CHAPTER 6 FIELD-ORIENTED TEST METHODS TO EVALUATE STRUCTURAL BUILDUP AT REST OF SCC 117

6.1 Introduction H7 6.2 Research significance and objectives 118 6.3 Testing program H 8

6.4 Field-oriented test methods to evaluate structural build-up at rest of SCC 119 6.4.1 Portable vane H 9

IX

A. Background 119 B. Development of the PV test 121 C. Calculation of static yield stress from the PV test 122

6.4.2 Inclined plane 123 A. Background 123 B. Development of the IP test for SCC 124 C. Calculation of static yield stress from the IP test 126

6.4.3 Evaluating repeatability of field-oriented test methods 126 6.5 Correlating field-oriented method responses to rheometric measurements 127

6.5.1 Correlation between initial responses 128 A. Correlations between initial static yield stress determined from PV and IP tests 128 B. Correlations between initial static yield stress determined from field-oriented and

rheometric test methods 129 C. Correlations between initial static yield stress determined from field-oriented tests and

initial drop in apparent viscosity at 0.7 rps determined from concrete rheometer 130 D. Correlations between initial static yield stress determined from field-oriented tests and

initial breakdown area determined from rheometric test method 131 6.5.2 Correlation between time-dependent responses 132

A. Correlating time-dependent static yield stress determined from field-oriented tests 132 B. Correlating time-dependent static yield stress determined from field-oriented tests and

rheometric test 132 C. Correlating time-dependent static yield stress determined from field-oriented tests to

time-dependent drop in apparent viscosity at 0.7 rps determined from rheometric test 133 6.6 Conclusions 135

CHAPTER 7 EFFECT OF SCC MIX DESIGN ON FORMWORK PRESSURE CHARACTERISTICS 136

7.1 Objectives 136 7.2 Testing program 136 7.3 Test results of Phase I 139

7.3.1 Effect of<)),S/A and Vca on SCC formwork pressure 139 7.3.2 Effect of <j), S/A, and V a on thixotropy of SCC 142 7.3.3 Relationship between ICo and structural build-up at rest 142 7.3.4 Correlation between decay of lateral pressure and thixotropy of SCC 145 7.3.5 Models to simulate effect of (|), S/A, and V^ on SCC lateral pressure and thixotropy 148

A. Model derivation 148 B. Relative errors of derived models 153 C. Validation of derived models 153 D. Correlations between modeled responses 155 E. Contour diagrams for the statistical models 155

7.4 Test results of Phase II 162 7.4.1 Effect ofVp on SCC formwork pressure 162 7.4.2 Effect of Vp on thixotropy (breakdown area) 164 7.4.3 Relationships between breakdown area and Ko 164

7.5 Test results of Phase III 164 7.6 Conclusions 167

CHAPTER 8 EFFECT OF PLACEMENT CHARACTERISTICS AND FORMWORK DIMENSIONS ON LATERAL PRESSURE OF SCC 170

8.1 Introduction 170 8.2 Testing program 171 8.3 Test results of Phase I 173

8.3.1 Effect of concrete temperature on BCo 173 8.3.2 Effect of concrete temperature on pressure cancellation time 175 8.3.3 Effect of concrete temperature on pressure decay 175

8.4 Test results of Phase II 1 76

X

8.4.1 Effect of casting rate on Ko 176 8.4.2 Effect of thixotropy on Ko 177 8.4.3 Abacus between Ko and thixotropic indices 177

8.5 Test results of Phase III 180 8.6 Test results of Phase IV 182

8.6.1 Effect of D,,^ on KQ 182 8.6.2 Effect ofDmin on pressure decay 184 8.6.3 Establishing a correction factor for Ko as function of D^ 185 8.6.4 Establishing a correction factor for AK(t) as function of Dmin 186

8.7 Conclusions 188

CHAPTER 9 PREDICTION MODELS FOR LATERAL PRESSURE CHARACTERISTICS 190 9.1 Introduction 190 9.2 Testing program 190 9.3 Correlations between thixotropic indices 191 9.4 Prediction models for maximum lateral pressure 193

9.4.1 Models of P ^ as function of H, R, T, D,^, and thixotropy index at T=22±2°C 193 9.4.2 Models of P , ^ as function of H, R, D,^, and thixotropy index at various temperature 194 9.4.3 Introducing effect of MSA in the prediction models of P , ^ 196 9.4.4 Introducing effect of WP in the prediction models of P1Bax 196 9.4.5 Relationships between measured and predicted Pmax 196 9.4.6 Abacuses for prediction of Ko values 198 9.4.7 Maximum lateral pressure of SCC on formwork system 203

9.5 Prediction models for lateral pressure decay 206 9.5.1 Models of AK(t) as function of thixotropy index 206 9.5.2 Introducing effect of D,^ in the AK(t) prediction models 206 9.5.3 Relationships between measured and predicted AK(t) 207

9.6 Comparison between predicted lateral pressure values determined using UofS model and other published guidelines 209

9.7 Conclusions 212

CHAPTER 10 FIELD MEASUREMENTS AND VALIDATION OF LATERAL PRESSURE MODELS 215

10.1 Objectives 215 10.2 Field testing of wall and column elements 215 10.3 Test results of casting wall elements and discussion 224

10.3.1 Typical results 224 10.3.2 Effect of casting rate 224 10.3.3 Effect of casting depth 225 10.3.4 Effect of mix design approach 226

10.4 Test results of casting column elements and discussion 227 10.5 Validation of formwork pressure models using field measurement results 229

10.5.1 Validation of P ,^ and Ko models 229 10.5.2 Validation of AK(t) models 231

10.6 Conclusions 232

CHAPTER 11 SUMMARY AND CONCLUSIONS 234 11.1 Introduction 234 11.2 Measuring formwork pressure 234

11.2.1 Portable device to measure maximum formwork pressure 234 11.2.2 Evaluation of lateral pressure decay 235

11.3 Field-oriented test methods to evaluate structural build-up at rest of SCC 236 11.3.1 Portable vane test 236 11.3.2 Inclined plane test 236 11.3.3 Thixotropic indices from field-oriented test methods 237 11.3.4 Validation ofPV and IP field-oriented test methods 237

XI

11.4 Factors affecting SCC formworkpressure and thixotropy 238 11.4.1 Investigated parameters 238 11.4.2 Comparison between SCC and conventional concrete of normal slump consistency 239 11.4.3 Influence of parameters on lateral pressure characteristics and thixotropy of SCC 239

A. Factors affecting Ko 239 B. Factors affecting AK(t) 240 C. Factors affecting tc 241 D. Factors affecting thixotropy 241

11.4.4 Statistical models for lateral pressure characteristics and thixotropy of SCC to simulate effect of (KV^andS/A 242

11.5 Model for lateral pressure prediction 242 11.5.1 Models for Pmax prediction 242 11.5.2 Models for the prediction of lateral pressure decay 243 11.5.3 Validation of the UofS prediction model with the published guidelines 244

11.6 Field measurements and validation of UofS models 244 11.7 Further work 245

Appendix A: CHAPTER 4 METHODOLOGY FOR LATERAL PRESSURE MEASUREMENTS 247 Appendix Al: Sensor calibration 247

A. Mechanical calibration 247 B. Hydrostatic calibration 248 C. Water calibration prior to each use 250

APPENDIX B: CHAPTER 6 FIELD-ORIENTED TEST METHODS TO EVALUATE STRUCTURAL BUILD-UP AT REST OF SCC 252

Appendix Bl: Fresh concrete properties , 252 Appendix B2: Protocols for the field-oriented test methods 253

A. Protocol ofthePV test 253 B. Protocol of the IP test 254 C. Precautions for field-oriented test methods 255

Appendix C: CHAPTER 7 EFFECT OF SCC MLX DESIGN ON FORM PRESSURE 256 Appendix CI Fresh concrete properties 256 Appendix C2 Relationship between Ko and thixotropy 258 Appendix C3 Correlation between AK(t) and thixotropic indices 261 Appendix C4 Correlations between various modeled responses (virtual points) 264 Appendix C5 Contour diagrams for the derived statistical models 269

Appendix D: CHAPTER 8 EFFECT OF PLACEMENT CHARACTERISTICS AND FORMWORK DIMENSIONS ON LATERAL PRESSURE OF SCC 278

Appendix Dl: Fresh concrete properties 278 Appendix D2: Effect of casting rate on Ko 282 Appendix D3: Effect of thixotropy on Ko 287

Appendix E: CHAPTER 9 PREDICTION MODELS FOR LATERAL PRESSURE CHARACTERISTICS 289

Appendix El: Abacuses for Ko prediction 289

REFERENCES 299 LIST OF FIGURES 307 LIST OF TABLES 314

xii

SYMBOLS AND NOTATIONS

x + Abi

Ab2

AEA

Athix

b

CON.

CC

CEM

CH

cm

r c

/"Dmin

/^Dmin

FA

fk

./MSA

fwp

g

H

h

HRWRA

IP i P W t ) IPx, 0rest@15min

Mean value of number of observations

Slump flow diameter of SCC (mm)

Initial breakdown area carried out between time of 0 and 30 min from end of casting (J/m3.sec)

Initial breakdown area carried out between time of 120 and 150 min from end of casting (J/m3.sec)

Air-entraining agent

Thixotropy factor [Roussel and Ovarlez, 2005] (Pa/sec)

Binder

Coefficient of variation

Conventional concrete

Concrete-equivalent mortar

Coefficient of hydrostatic calibration

Cementitious material

Coefficient of mechanical calibration

Coefficient of water calibration

Minimum cross-sectional dimension of formwork/ formwork width (mm)

Correction factor for initial maximum lateral pressure or relative initial maximum lateral pressure (Pmax or Ko) accounting for the effect of different minimum formwork dimension (Dmin) (dimensionless)

Correction factor for lateral pressure decay indices [AK(t)(0-60 min) or AK(t)(0-tc) ] accounting for the effect of different minimum formwork dimension (Dmin) (dimensionless)

Fly ash

Frictional force [Oremus and Richard, 2006]

Correction factor for initial maximum lateral pressure or relative initial maximum lateral pressure (Pmax or Ko) accounting for the effect of different maximum-size of aggregate (MSA) (dimensionless)

Correction factor for initial maximum lateral pressure or relative initial maximum lateral pressure (Pmax or Ko) accounting for the effect of waiting period between lifts (dimensionless)

Gravitational acceleration (= 9.806 m/ses2)

Concrete depth / concrete height / casting depth / casting height (m)

Filling height of the concrete sample in the portable vane test

High-range water-reducing agent

Inclined plane field-oriented test method

Time-dependant change of static yield stress, obtained using inclined plane field-oriented test method (Pa/min)

Static yield stress determined at 15 min time of rest, obtained using inclined plane field-oriented test method (Pa)

xiii

Ko

K()@H=x

KOB

Koi

m.g.sin

MSA

n

N/A

P

£ 1 raw data

* ^corr.M

P3COrr.H

P4Corr.W

Phyd

Phyd@Hi

P 1 max

"max@Hi

PV

PVTorest(t)

0rest@15min PVT,

R

r

R2

RE

RheometerTorest(t)

5min

R;@ 15-60 min

Ri@15min

S/A

sec

Initial relative maximum lateral pressure (dimensionless or %)

Initial relative maximum lateral pressure at concrete depth, H; = "x" (dimensionless or %) Initial relative maximum lateral pressure determined from the pressure sensors fixed in the short lateral dimension, width, (A sensor) of the plywood formwork (dimensionless or %)

Initial relative maximum lateral pressure determined from the pressure sensors fixed in the long lateral dimension, length, (B sensor) of the plywood formwork (dimensionless or %)

Initial relative maximum lateral pressure at given concrete depth, H; (dimensionless or %)

Gravitational force [Oremus and Richard, 2006]

Maximum-size of aggregate (mm)

Number of observations

Not applicable

powder

Output milli-volt signal from the pressure transducer

Output kPa-pressure values corrected by the mechanical coefficient (Cm)

Output kPa-pressure values corrected by the mechanical and hydrostatic coefficients (Cm

and CH, respectively)

Output kPa-pressure values corrected by the mechanical, hydrostatic, and water coefficients (Cm, CH, and Cw, respectively)

Equivalent hydrostatic pressure (kPa)

Equivalent hydrostatic pressure at a given concrete depth, Hj (kPa)

Initial maximum lateral pressure (kPa)

Initial maximum lateral pressure at a given concrete depth, Hj (kPa)

Portable vane field-oriented test method

Time-dependant change of static yield stress, obtained using portable vane field-oriented test method (Pa/min)

Static yield stress determined at 15 min time of rest, obtained using portable vane field-oriented test method (Pa)

Casting rate (m/hr)

Radius of the vane used in the portable vane test

Coefficient of correlation

Relative error

Time-dependant change of static yield stress, obtained using modified Tattersall MK-III concrete rheometer (Pa/min)

Static yield stress determined at 15 min time of rest, obtained using modified Tattersall MK-III concrete rheometer (Pa)

Initial response determined between 15 and 60 min time of rest, obtained using the concrete rheometer or the field-oriented test method

Initial response determined at 15 min time of rest, obtained using the concrete rheometer or the field-oriented test method

Sand-to-total aggregate

Self-Consolidating concrete

xiv

SE

SRA

SSD

T

T

T.I.

T.I.

T.I. •(t)

@15min

T.I.@T=22±2°C

T.I.@Ti

T50

tc

UofSl pressure column

UofS2 pressure column


US

V

VMA

VP

w/b

w/cm

w/p

WAT

WP

a

Tc

AK(t)

AK(t)(0-60 min)

AK(t)(0-te)

AT|app@N=0.7rps

ATI; !app@N=0.7ips' ;(0

An, app@N=0.7rps@15min

e

p

T0(t)

T00

^Orest

Standard error

Set-retarding agent

Saturated-surface dry unit weight (dimensionless)

Measured torque (N.m)

Concrete temperature (°C)

Thixotropy index

Time-dependant change of the thixotropy index

Initial thixotropy index determined at 15 min time of rest

Thixotropy index determined at laboratory temperature (T) of 22 ± 2 °C

Thixotropy index determined at various concrete temperature (T;)

Time corresponding to 50 mm spread of SCC (sec)

Time to pressure cancellation (min)

Pressure column version 1, developed at University of Sherbrooke



Undisturbed spread field-oriented test method

Volume of coarse aggregate (1/m3)

Viscosity-modifying agent

Paste volume (1/m3)

Water-to-binder

Water-to-cementitious material

Water-to-powder

Water-adding time of cement and water

Waiting period between successive lifts (min)

Critical angle of inclination in the inclined plane test (degree)

Unit weight (dimensionless)

Relative lateral pressure decay (%/min)

Relative lateral pressure decay during the first 60 min after the end of casting (%/min)

Relative lateral pressure decay during the entire time to pressure cancellation (%/min)

Drop in apparent viscosity at rotational frequency of 0.7 rps, obtained using the modified Tattersall MK-III concrete rheometer (Pa.s)

Time-dependant change of drop in apparent viscosity at rotational frequency of 0.7 rps, obtained using the modified Tattersall MK-III concrete rheometer (Pa.s/min)

Initial drop in apparent viscosity at rotational frequency of 0.7 rps, determined at 15 min time of rest, obtained using the modified Tattersall MK-III concrete rheometer (Pa.s)

Critical angle of inclination in the inclined plane test (degree)

Density of material (kg/m3)

Apparent static yield stress [Roussel and Cossigh, 2008] (Pa/sec)

Initial static yield stress [Roussel and Cossigh, 2008] (Pa)

Static yield stress, obtained using modified Tattersall MK-III concrete rheometer or PV and IP field-oriented test methods (Pa)

Time-dependant change of static yield stress at rest [Billberg, 2006] (Pa/min)

XV

CHAPTER 1

INTRODUCTION

1.1 Introduction

Over the years, concrete technology has advanced at a relatively slow pace that has been

associated with a labor-intensive industry and tedious placing in the formwork. The two milestones

that have probably had the greatest impact on propelling this low-skill industry to a technology-

driven one are the introduction of superplasticizer and the development of self-consolidating

concrete (SCC). The SCC is a new class of high-performance concrete (HPC) that flows readily

under its own weight and consolidates without the use of mechanical vibrations and with minimum

risk of segregation. The SCC is a complex system that is usually proportioned with a number of

chemical admixtures and supplementary cementitious materials. Such concrete exhibits low

resistance to flow and moderate plastic viscosity necessary to maintain homogeneous deformation

during placement and thereafter until the onset of hardening. The first SCC prototype was

successfully produced by Ozawa et al. [1989] in the late 1980s. Since then, the market share of

SCC has rapidly increased in precast applications or for ready-mix concrete applications, due to a

number of economic opportunities and the improvement in the work environment associated with

its use. SCC has been successfully used in North America in the precast industry. A recent

overview on SCC types, test methods, and properties are given by Khayat [1999], Khayat et al.,

[1999], and Bonen and Shah [2004, 2005].

The benefits obtained from using SCC can be summarized as follows:

•a Decrease in construction cost due to labor reduction;

<9 Reduction in construction time;

3 Simplification of the casting process as no vibration is needed;

% Improvement of working conditions through less noise hazards;

3 Ability to cast congested and complex structural elements in various shapes and

dimensions that are not achievable by any other conventional techniques;

# Ability to cast hard-to-reach areas for placement, and consolidation;

4» Improving appearance and quality of the finished surfaces and reduction in the

occurrence of bug holes, honeycombing, and other surface imperfections;

1

Chapter 1 Introduction

# Producing a better and premium concrete product;

# Larger variety of architectural shapes by using any form shape. This is one of the major

advantages of SCC where it is possible to cast heavily reinforced elements and structures

with a complicated geometry that otherwise are not attainable by any other conventional

techniques [Khayat et al., 2001, Walraven, 2002, Okamura and Ouchi, 2003, Mullarky

and Vaniker, 2002];

# Higher durability of concrete structures;

® Lowering pumping pressures, and as a consequence, reducing wear and tear on pumps,

i.e. extends their service life; and

® Lowering the need for cranes to deliver concrete in buckets at the job site by facilitating

concrete delivery through pumping.

Despite the various benefits that can be gained by using SCC, there are some limitations that

should be taken into consideration when using such new material, including:

# Raw materials cost of SCC can be 13% to 30% higher than the cost of conventional

mixtures with similar mechanical properties [Schlagbaum, 2002, Martin, 2002].

Nonetheless, cost analysis shows that even if the selling cost of SCC is reduced by only a

few percent because of the decrease in labor and construction time, the profitability can

be increased by about 10% [Szecsy et al., 2002];

*§ SCC requires greater quality control and quality assurance measures to ensure proper

workability, including high resistance to segregation and stability of entrained air voids;

# SCC has greater potential for shrinkage and creep, and care should be considered in

designing the concrete elements. Greater risk of shrinkage and creep arise from the large

volume of fine materials in use, particularly in the case of SCC without any VMA, and

the lower content of coarse aggregate; and

« Lack of knowledge on the relative lateral pressure that SCC could exert on formwork

systems. This adverse effect may compromise profitability, due to the need to design for

robust formwork construction and detailed joint sealing.

Of the benefits listed above, the greatest incentives for the industry to adopt this technology

are related to the potential profitability brought about by shortening of the casting time, reduced

labor, minimized logistics due to elimination of the need for vibrations, and production of esthetic

surfaces with high quality. In turn, a rapid rate of casting of concrete in a formwork system leads to

2


an increase in lateral pressure exerted by the concrete, which could reach full hydrostatic pressure.

Such high pressure is attributed to two factors:

d The low initial shear stress of the plastic SCC; and

® The rate of vertical placing in the formwork that exceeds the rate of stiffening of the

concrete in the formwork.

Formwork systems for wall and column elements can contribute up to 40% of the overall cost

of construction projects [Rodin, 1952]. This was recently reported to be up to 60% of the total cost

of completed concrete structures in place in the USA [ACI Guide to Formwork for Concrete,

2004]. Any savings in the cost of formwork, for example by reducing the design loads affecting

lateral pressure exerted by plastic concrete, would be of great interest. The relatively high lateral

pressure exerted by SCC is considered the main technical hindrance that slows down the

widespread use of SCC in cast-in-place applications. Lateral pressure exerted by concrete is of

concern to construction engineers because its overestimation results in expensive formwork, while

its underestimation can lead to bulging of the formwork or, in extreme cases, failure of the

formwork system. Additionally, high formwork pressure presents a major safety issue. As the

lateral pressure of the concrete increases, so does the potential risk for liability in the event of a

failure. According to the current provisions, responsibility for the safe construction of formwork

rests on the contractor or the engineer hired by the contractor to design the formwork. Provisions in

ACI 347R-7 document stipulate that when working with mixtures with high slump characteristics,

such as SCC, the presumed lateral pressure should be equal to the hydrostatic pressure of the fresh

concrete "until the effect on formwork pressure is understood." Similarly, the European Federation

of Producers and Contractors of Specialist Products for Structures (EFNARC) recommends that

forms higher than 3 m are designed for full hydrostatic head, [EFNARC, 2002]. In that case, either

the total cost of the formwork has to be increased or the rate of placing should be decreased.

To date, limited information exists regarding the magnitude of the lateral pressure that can be

developed by SCC on vertical wall or column elements. Contractors and engineers recognize

design recommendations elaborated with the use of normal-consistency concrete, which cannot be

fully applied to SCC due to the higher fluidity level of the SCC that could result in the lateral

pressure reaching full fluid pressure. Therefore, existing equations for estimating lateral pressure

that are necessary for the design of formwork need to be modified to account for the high

flowability of SCC. So far, formwork is designed prudently by assuming that the SCC exerts full

3

hydrostatic pressure until setting time. Such pressure is expressed as: Pmax = p x g x H where: p, g,

and H correspond to the concrete unit weight, gravity, and head of concrete, respectively. This

approach can result in increased construction costs and can limit the rate of rise of the concrete in

the formwork. Designing for high values of hydrostatic pressure requires a robust formwork

construction and detailed joint sealing, which could adversely affect profitability.

A comprehensive research program was undertaken at "Universite de Sherbrooke" to evaluate

the formwork lateral pressure exerted by SCC. The proposed research program aimed at developing

a portable devise to measure and predict the lateral pressure of SCC, in addition to developing

field-oriented test methods to evaluate the plastic properties of concrete. The program also aimed at

evaluating the role of the major influencing parameters on formwork pressure, proposing design

equations to predict formwork pressure that could be exerted by SCC on column and wall elements,

and finding effective ways to reduce lateral pressure by developing formulation expertise and

practical guidelines to lower lateral pressure of SCC.

1.2 Objectives

In view of the complexity of the problem, our goal is to make a linkage between the plastic

properties of SCC and lateral pressure and to find effective ways to evaluate and reduce the lateral

pressure. Accordingly, the objectives of this investigation are nine-fold:

# Literature review to capture existing knowledge and make recommendations for current

practice;

•a Develop portable apparatus for measuring and predicting the lateral pressure and its rate

of decay of SCC;

<M Develop test methods for field evaluation of relevant plastic properties of SCC that affect

formwork pressure;

# Evaluate in laboratory the lateral pressure characteristics (the initial maximum lateral

pressure measured right after casting at different heights and the variation of lateral

pressure with time and up to the pressure cancellation) exerted by SCC on the formwork

system as a function of the most influencing key parameters affecting lateral pressure

including: material properties, mixture composition, admixtures, consistency level,

placement characteristics (rate of casting rise, temperature, waiting periods between

successive lefts,...), and formwork geometry and materials.

4


a Relate the maximum lateral pressure and its initial rate of decay to the initial rheological

properties (including static yield value and thixotropy). Relate the initial lateral pressure

and its variations in time to the rate of increase in shear strength properties, namely

structural build-up at rest of the plastic concrete.

« Propose design equations to predict formwork pressure that could be exerted by SCC on

column and wall elements.

# Carry out field measurements to validate laboratory observations.

a Propose effective ways to reduce lateral pressure by developing formulation expertise

and practical guidelines to lower lateral pressure of SCC.

1.3 Thesis outline

The thesis is divided into 11 chapters that might be summarized as follows:

Chapter 1 gives an introduction about the use of SCC worldwide and North America, the

advantages gained from the using SCC and limitations restricting its wide spread, objectives, and

brief summary on the contents of the thesis.

Chapter 2 presents literature review on formwork pressure and fundamentals of rheology.

This chapter presents the theoretical models and the existing equations proposed by many

researchers and specifications to evaluate and predict lateral pressure. The relationships between

SCC formwork pressure and rheological properties are also presented. The assessment of

thixotropy can provide some indication of the degree of restructuring of the concrete after

placement once left at rest in the formwork. So, the assessment of thixotropy for cement-based

materials and SCC, the testing methods and protocols that used to evaluate the degree of thixotropy

of the various SCC mixtures, are highlighted. The influence of the various parameters affecting

lateral pressure developed by plastic concrete is discussed. The available measuring systems for

lateral pressure and pore water pressure determination are brought to light. Case studies for lateral

pressure evaluation collected from different projects worldwide are discussed. Finally, some

concluding remarks are drawn.

In Chapter 3, all material properties and mixture compositions used throughout the laboratory

experimental work and the field investigations are introduced. Fabrication of concrete and sequence

of material mixing are also presented in this chapter.

In Chapter 4, the methodologies used for determining concrete lateral pressure are discussed.

The measuring devices adopted for evaluating concrete lateral pressure resulting from the fluid

5


phase and up to cancellation time including: metallic pressure column, UofSl pressure device,

experimental 1.2-m and 3-m high PVC columns, and instrumented 1.5-m high plywood formwork

of variable minimum cross-sectional dimensions are presented in this chapter. The measurement

systems (pressure sensors) adopted for evaluating concrete lateral pressure also are involved in this

chapter. Mechanical and hydraulic calibrations for these sensors are elaborated. Pressure

transducers of various contact areas with the concrete surface are compared.

Devise portable and field test method for lateral pressure determination is focused on in

Chapter 5. The devising started with trials to reduce the free concrete head in the UofSl pressure

column to fit the field application. Second version (UofS2 pressure column) was proposed. Further

reduction of the free concrete head in the UofS2 pressure column was attempted. The pressure

characteristics resulted from the UofS2 pressure column were compared to the equivalent responses

determined using 3-m and 1.2-m high PVC columns. The repetition of the UofS2 pressure column

was evaluated. Validation of the UofS2 pressure column with various mixture compositions and

casting characteristics was evaluated.

Chapter 6 presents development of two field-oriented test methods to measure structure

build-up at rest of SCC. The repetitions of the two tests are performed using SCC mixtures of low

and high thixotropy levels. Using various SCC mixtures, the two field-oriented tests are validated

using the modified Tattersall MK-III concrete rheometer.

Throughout Chapters 7 and 8, several parameters that are likely to affect lateral pressure

development of SCC are evaluated. The parameters include:

# Effect of mixture compositions of SCC on formwork pressure is evaluated in

Chapter 7. The work achieved in this chapter are classified under three main Phases.

Phase I consists of experimental design to investigate effect of slump flow ((j)), volume of

coarse aggregate (Vca), and sand-to-total aggregate ratio (S/A). Phase II is parametric

study to investigate effect of paste volume (Vp). Phase III is proposed to investigate

effect of maximum-size of aggregates (MSA).

a Effect of concrete temperature (T), casting rate (R), waiting period between lifts (WP),

and minimum formwork lateral dimensions are presented in Chapter 8.

In Chapter 9, all tested mixtures are grouped together and analyzed to establish prediction

models for various lateral pressure characteristics. These models are function of thixotropic

properties and several key parameters affecting SCC lateral pressure.

6


In order to validate the results obtained in the laboratory investigation obtained using the

developed devices in Chapters 5 and 6, field measurements for lateral pressure and rheological

properties are conducted through two main projects. The first project consists of casting wall panels

measuring 3.6 and 4.4 m in height during the construction of new material laboratory at Universite

de Sherbrooke, Canada. The second project is casting instrumented reinforced concrete columns

having 3.6 m in height and 600 mm in diameter at CTLGroup, Chicago, USA. Different SCC

mixtures prepared with various mixture compositions are tested in the two projects and reported in

Chapter 10. Validations of the developed models using the obtained field results are discussed in

the same chapter.

In the final chapter (Chapter 11), an overview of the major findings obtained throughout the

thesis is presented. Summary and conclusions are also presented. Structure of the thesis is

presented in Fig. 1.1.

Structure of the thesis

Guidelines for lateral pressure estimation

' -itfaJJItfflJisi»l~

ML Conclusionsand

future work

Fig. 1.1 Structure of the thesis

CHAPTER 2

REVIEW ON FORMWORK PRESSURE AND FUNDAMENTALS OF RHEOLOGY

2.1 Introduction

An extensive literature review was undertaken to capture existing knowledge of formwork

pressure exerted by flowable concrete and SCC. The literature survey addressed five major

topics. In the first part of this review, the design recommendations and theoretical models

proposed by various code regulations and researchers to predict formwork pressure, including

recent recommendations for SCC were addressed. Relationship between form pressure and

rheology of SCC was reviewed in the second part. The parameters affecting thixotropy and

structural build-up of cement-based materials, evaluation methods of thixotropy, and relationship

to the initial development of lateral pressure and its variation in time were the most topics

discussed in the second part. The third part of the literature review focused on the influence of

various parameters affecting formwork pressure and thixotropy. These parameters were divided

according to material properties, consistency level, placement conditions, and formwork

characteristics. Lateral pressure measurement systems were surveyed in the following part.

Instruments and devices that have been used to monitor lateral pressure, including pressure

transducers and pore-water pressure sensors were discussed. Well-documented case studies

highlighting observations of formwork pressure measurements exerted by SCC were reviewed in

the last part.

2.2 Review of various recommendations for formwork design

In conventional construction practice, concrete is cast into wall or column forms in lifts,

which are vibrated to be consolidated. The concrete is usually consolidated using poker-type

vibrators, which are immersed into the concrete at the top layer (about 1.0 m). The vibration

causes the development of full fluid pressure at the top layer.

In considering formwork pressure, two main items should be considered to ensure safely

designed and cost-effective formwork systems. The first item is the initial maximum lateral

pressure developed by the plastic concrete immediately after casting. The relative lateral pressure

(Ko) is defined as the maximum lateral pressure divided by the hydrostatic liquid head at the

same level (Ko = Pmax. / Phydrostatic)- Such value is the most critical because it directly

8

Cha

pter

2:

Rev

iew

on

form

wor

k pr

essu

re a

nd fu

ndam

enta

ls

ofrh

eolo

gy

Lite

ratu

re r

evie

w d

iscu

sses

the

fo

llow

ing

topi

cs:

Var

ious

rec

omm

enda

tions

for

for

mw

ork

desi

gn

Mod

els

prop

osed

to

eval

uate

for

mw

ork

pres

sure

The

oret

ical

mod

els

to

pred

ict

form

wor

k pr

essu

re

Rel

atio

nshi

p be

twee

n fo

rm

pres

sure

and

rh

eolo

gyof

SC

C

>

Rod

in's

mod

els

[195

2]

>

Scho

jdt's

mod

els

[195

5]

>

AC

I m

odel

s >

A

dam

et a

l.'s

mod

els

[196

3]

>

Mod

els

of G

erm

an

Stan

dard

[D

IN 1

8218

, 19

80]

>

CIR

IA m

odel

s [1

965-

1978

] >

G

ardn

er's

mod

els

[198

0-19

84]

>

Mod

els

of F

renc

h St

anda

rd [

NFP

93-

350,

19

95]

>

Van

hove

and

co

auth

ors'

mod

el [

2004

] >

R

ouss

el a

nd O

varl

ez's

m

odel

[20

05]

>

Gra

ubne

r an

dPro

ske'

s m

odel

[20

05A

] >

K

haya

t an

d A

ssaa

d's

mod

el [

2005

A]

>

Thi

xotr

opyo

f ce

men

t-ba

sed

mat

eria

ls

>

App

roac

hes

to

quan

tify

thix

otro

py o

f co

ncre

te

>

Rel

atio

nshi

ps

betw

een

late

ral

pres

sure

and

rh

eolo

gica

l pr

oper

ties

Para

met

ers

affe

ctin

g fo

rmw

ork

pres

sure

an

d th

ixot

ropy

Lat

eral

pre

ssur

e m

easu

ring

sy

stem

s

;The

mai

n [p

aram

eter

s ca

n be

[c

lass

ified

und

er t

he

jfollo

win

g m

ain

[cat

egor

ies:

>

Mat

eria

l pr

oper

ties

>

Con

sist

ency

lev

el

>

Plac

emen

t co

nditi

ons

>

Form

wor

k ch

arac

teri

stic

s

Inst

rum

ents

and

; de

vice

s to

m

onito

r la

tera

l pr

essu

re

!> P

ore-

wat

er

pres

sure

m

easu

rem

ents

to

det

erm

ine

late

ral p

ress

ure

Cas

e st

udie

s fo

r fo

rmw

ork

pres

sure

exe

rted

by

SC

C

>

Rev

iew

ing

the

exis

ting

fiel

d re

sults

of

SCC

la

tera

l pr

essu

re

9

Chapter 2: Review on formwork pressure and fundamentals ofrheology

affect the design of formwork systems. The rate of pressure drop with time [AK0(t)] is also of

special interest in designing formwork systems. In most lateral pressure investigations carried out

using normal-consistency concrete, the pressure can be found to decrease slowly before dropping

to zero approximately 3 hr after casting. However, this is not always applicable for cast-in-place

SCC where the set can be delayed. Better knowledge of the rate of pressure drop enables better

scheduling of the placement of subsequent concrete lifts. This is particularly true in case of

casting into deep and large elements requiring considerable volume of concrete. The elapsed time

before pressure cancellation is also important for better schedule of the re-use of formwork.

2.2.1 Models proposed to evaluate formwork pressure

Several equations were proposed in the literature to evaluate the magnitude and shape of

the lateral pressure envelope. Some of these models elaborated to estimate formwork pressure for

conventional concrete and few recent studies targeting formwork pressure of SCC are

summarizes below.

A. Rodin's models [1952]

Rodin [1952] reviewed published experimental data on lateral pressure of fresh concrete

against formwork. Rodin concluded that the major factors influencing lateral pressure are rate of

pour, vibration, mixture consistency and mixture proportion, concrete temperature, concrete

setting time, size and shape of the form. Rodin reported that the formwork should be designed

according to two cases: externally vibrated and non-externally vibrated concrete. The latter case

was consequently divided into two categories: internally vibrated concrete and hand-placed

concrete. The details of the two cases can be expressed as follows:

For externally vibrated concrete

The formwork should be designed for full hydrostatic pressure of a liquid having the same

density as concrete.

For non-externally vibrated concrete

For internally vibrated concrete Pmax = 23.4 Hmax Eq. 2.1

For hand-placed concrete Pmax= 17.2 Hmax Eq. 2.2

where, Hmax : head at which the maximum pressure occurs,

Hmax=1.63R1/3 Eq.2.3

Pmax : maximum lateral pressure, kPa

10

R : rate of placing, m/hr

These equations are for concrete having 1:2:4 cement:sand:coarse aggregate mass fractions, a

unit weight of 2,400 kg/m3, a slump consistency of 150 mm, and a temperature of 21 °C. The

concrete pressure distribution on the formwork as proposed by Rodin [1952] is shown in Fig. 2.1.

, Max.pressure used \J lor design of

sheathing

Simplified distribution uud v I for design

Actual pressure distribution

^Hydrostatic • pressure

„:•:• ; . . . i r '.. • t

,£-.: FORMWORK

• &->•:

ISO H

. » . •

S E IF ".'-• • ? •• . Vn *. • vp. f .

5 * r ' . , Q ' V -

Pressure Intensity P™

Fig. 2.1 Concrete pressure distribution on formwork [Rodin, 1952]

B. ACI models

The American Concrete Institute (ACI) Committee 622 [1958] (currently designated as

ACI 347) "Formwork for Concrete" [2001 and 2004] proposed that the lateral pressure diagram

is assumed to be trapezoidal in shape: the diagram is presumed to be a triangular distribution

from the upper free surface of the casting down to some limiting depth, beyond which the value

of pressure reached is considered constant until the bottom of the formwork. The significant

variables considered in the ACI recommendations are the placement rate and method, consistency

of concrete, coarse aggregate concentration, aggregate nominal size, concrete temperature,

smoothness and permeability of the formwork material, size and shape of the formwork,

consolidation method, pore-water pressure, content and type of cement, as well as the depth of

the concrete placement, or concrete head. The ACI equations are reported along with the

limitation of use, in the following paragraphs.

For wall element:

Eq. 2.4 R<2.14m/hr Pmax = 7 . 1 9 + 1 77 ^ ?

r < 95.8 or 23.5 H

11

244 R 2.14<R<3m/hr Pm =36 + Ea 2 5

r+i7.78 q-

P =7.19+ U 5 6 + 244R < 95.8 or 23.5 H Ea 2 6 " 7+17.78 T+17.78 q

R>3m/hr Pmax = 23.5H < 95.8 Eq. 2.7 For column element

/>ma* = 7 . 1 9 + 1 77 ^ r < 143.7or 23.5H Eq. 2.8

For wall and column elements

Pmax = 7c-H Eq. 2.9

where Pmax: maximum lateral pressure, kPa;

R : rate of placement, m/hr;

T : concrete temperature, °C; and

H : head of concrete, m.

Notes: 1- The formulas are used only for normal internal vibration, immersion of vibrator in

concrete < 1.2 m, any re-vibration is allowed only in plastic stage, Type GU cement,

no pozzolans or admixtures, yc = 2,400 kg/m3, and slump at time of casting < 100 mm,;

2- Eq. 2.6 and term of [23.5 H] were added in 1963;

3- Eq. 2.7 was added in 1978; and

4- Eq. 2.9 was added in 1988 for all types of concrete.

In 2002, Hurd recognized that such equations are too conservative to be adopted nowadays,

thus resulting in more expensive formwork. This is due to evolution in the composition of

concrete mixtures, mainly with the introduction of chemical admixtures and Portland cement

replacements. Consolidation and placement techniques have also undergone significant changes

with the use of fluid and highly fluid concrete. Hurd [2002] proposed applying some coefficients

to the ACI equations [1958] in order to take into account different unit weights that can be

encountered on the job-site, as well as the chemical admixtures and supplementary cementitious

materials.

For wall and column elements

S "*" 3° C" (m * "- ' 15° C" C' ^ E,. 2.10 P < y H

max I c

P = C C max w c

7.19+-v

where yc : unit weight of concrete, kg/m ;

12

max

3 .

H : head of concrete, m;

Pmax : maximum lateral pressure, kPa;

R : rate of casting, m/hr

T : concrete temperature, °C;

Cw : Unit weight coefficient calculated as follows:

Cw = 0.5(l + — but>0.$ for yc< 2240 kglm1

Cw = 1.0 for 2240 kg I m1 < yc< 2400 kg IV

Cw= J ^ j for yc> 2400 kg Im'

Cc : chemistry coefficient calculated as follows:

Cc = 1.0 for cement Type GU or HE without retarder

Cc = 1.2 for blended cement without retarder (blended means: Type GU cement

with < 70% slag or < 40 % fly ash replacements).

Cc = 1.4 for blended cement with retarder (retarder refers to set retarder, water-

reducing agent, or superplasticizer).

C. Models of German Standard [DIN 18218,1980]

DIN 18218 presented a series of equations to calculate the limiting lateral pressures of

internally vibrated concrete made with various consistency levels and temperature of 15 °C [Eq.

2.11 or Eq. 2.12]. In order to adjust for variable concrete temperatures, it is recommended to

decrease the limiting pressure (developed for concrete at 15 °C) by 3% for every degree above

15 °C and to increase it by 3% for every degree below 15 °C.

For concrete cast at T = 15 °C:

Pmax = rc C2Kt (0.48 R + 0.74) Eq. 2.11

^ = 21 + 5 R for stiff mixtures ^

P m = 19 + 10 R for soft mixtures

v. p™» = 18 + 14 R f o r fluid mixtures \. max •

For concrete cast at T<15 °C: 3 % increase in Pmax for each degree below 15 °C.

For concrete cast at T>15 °C: 3 % decrease in Pmax for each degree above 15 °C.

where Pmax: maximum lateral pressure, kPa;

yc: unit weight of concrete, kg/m :

C2 : added coefficient;

13

Eq. 2.12

K t : temperature coefficient = (145 - 3 T)/l 00

R : rate of placement, m/hr; and

T : concrete temperature, °C.

D. CIRIA 108 design models [1965 - 1978]

The Construction Industry Research and Information Association (CIRIA) sponsored a

large-scale field investigation of formwork pressures carried out by the Cement and Concrete

Association and published in 1965. The CIRIA study proposed a lateral pressure design method

that involved consideration of the rate of placement, concrete temperature, slump, concrete

constituent materials, concrete unit weight, formwork dimensions and shape, and continuity of

vibration. The CIRIA design procedure considered that lateral pressure envelope is hydrostatic up

to a maximum value (Pmax) limited by the lesser of concrete stiffening and arching effects, as

given by the two equations below. In narrow sections, it was found that the wall friction can

significantly limit the maximum exerted pressure [CIRIA, 1965]. In 1978, CIRIA published a

two-page design chart to replace these equations.

For arching criterion

Pmax =14.37 + 0.094 d + 3.14 R Pmax < 24 H or 143.7 Eq.2.13

For stiffening criterion or concrete hardening

ycRT

General formula

/>°« = i . - V l / , , 4 + ( 4 - 6 - L 8 9 i ? ) Pmax < 24 H or 143.7 Eq. 2.14

P =y max / c

CX4R+C2K^H-CX4R ?yc.h (kPa) Eq.2.15

where, Pmax : maximum lateral pressure, kPa;

d : minimum formwork dimension, mm;

R : rate of placing, m/hr;

T : fresh concrete temperature, °C;

t : time after start of placing, hr;

tmax : stiffening or hardening time, hr;

c : vibrating time;

yc : unit weight of concrete, kg/m3;

H : vertical formwork height, m;

14

h : height of fresh concrete above the point considered, m;

Ci : coefficient depends on size and shape of formwork (= 1.0 for wall);

C2 : coefficient depends on constituents of concrete (= 0.3 - 0.6);

K : temperature coefficient taken as (36/T+16))2; and

(c and tmax are defined in empirically derived charts).

E. Gardner's models [1980 - 1984]

Gardner [1980] carried out series of laboratory studies using a large instrumented form.

The variables considered by Gardner [1980] were the depth of vibration, power of vibrator,

casting rate, concrete temperature, member dimension, and concrete slump. For formwork design

purposes, Gardner [1980] considered that the lateral pressure envelope is bilinear. The envelope

is hydrostatic from the free surface to a maximum value and becomes constant thereafter until the

bottom. The proposed equations are as follows:

. . . 3000HP d 400JR S-75 „ A T T „ . . , Pn**=Hh+ + — + + <2AH Eq.2.16

" ' d 40 18 + 7 10

where Pmax : maximum lateral pressure, kPa;

H : total height of formwork, m;

hi : immersed depth of vibrator not to be less than 1 meter, m;

d : minimum formwork dimension, mm;

HP : horsepower of vibrator;

R : rate of placement, m/hr;

T : concrete temperature, °C; and

S : slump after application of superplasticizer, mm.

A subsequent investigation using the same apparatus, Gardner [1982, 1984] investigated the

effect of incorporating superplasticizers and supplementary cementitious materials; Class F fly ash,

and blast furnace slag on lateral pressure. It was found that partial replacement of Portland cement

by fly ash or blast furnace slag can increase the mobility of the concrete and decreases the rate of

strength gain at early age, thus resulting in an increase in formwork pressure. An additional factor

was introduced in Eq. 2.16 to account for fly ash and slag substitutions. 3000HP d 400y[R f if\r\ \

P =24/z.+ + — + d 40 18 + 77

100

100 -%F + - ^ <24H Eq.2.17

10

15

where; F: percent substitution of cement by Class F fly ash or blast furnace slag.

The equation was shown to give conservative design values for fly ash concrete.

F. Models of French Standard [NF P93-350,1995]

The French Standard [NF P93-350, 1995] reported that the formwork must be designed to

withstand forces in the elastic domain due to the placing of ordinary concrete with a density of

2,400 kg/m3. The exerted maximum lateral pressure is given by the following equation:

Pmax = 2.400 g H < 72 kPa at the bottom of a 3 m high form Eq. 2.18

where, g: gravitational acceleration; and H: formwork height in m.

G. Comparison between models

A comparison of the lateral pressure envelopes that can be obtained from design equations

offered by DIN 18218, CIRIA 108, and NF P93-350 models for conventional vibrated concrete

with flowable consistency is made in Fig. 2.2. These results are plotted for fresh concrete having

a unit weight (yc) of 2,500 kg/m3, temperature (T) of 15 °C, and an end of solidification time (tE)

of 5 hr cast at placement rates of 1 to 25 m/hr in a wall element measuring 20 m in height. The

data show the influence of the casting rate on the pressure envelope. According to the DIN

18218 model, the increase in casting rate from 1 to 12.5 and 25 m/hr would lead to linear design

pressure envelope equal to hydrostatic pressure in the upper 1.5, 9, and 18 m portions of the wall,

respectively. These values were approximately 1.5, 6, and 8 m, respectively, for the CIRIA 108

design model.

u a VI VI

a u Q.

E u

£ C3

500

400

200

100

• DIN 18218; v=1 m/h -DIN 18218; v= 12.5 m/h -DIN 18218; v = 25 m/h -CIRIA 108; v=1nVh -CIRIA 108; v= 12,5 m * -CIRIA 108; v = 25 m/h . NFP 93-350/ hydrostatic

0 1

—~-ffi>^fc.'

• • •

0.0 5,0 10.0 15.0 20,0

Height (m)

Fig. 2.2 Formwork pressure - DIN 18218 (D), CIRIA 108 (GB), and NF P93-350 (F) [Proske and

Graubner, 2002]

16

2.2.2 Theoretical models to predict formwork pressure

A. Vanhove and co-authors' model [2004]

Vanhove et al. [2004B] selected the silo geometry from Janssen models [Janssen, 1885] for

soil mechanics and applied it in a model aimed at predicting formwork pressure of fresh concrete.

The approach treats the granular medium as a continuous and assumes perfect frictional contact

to the wall. The lateral pressure P'(h) (Fig. 2.3) is proportioned to the vertical pressure P (h), as

follows:

P'(h)=K.P(h) Eq.2.19

where K is a phenomenological coefficient, which depends on the internal friction angle (p of the

material [Ritchie, 1962]. As lateral pressure is measured on site at the end of casting, the at-rest

state can be applied. In this case, the phenomenological coefficient K may be expressed by:

K = 1 - sin <p Eq. 2.20

p' MMM ptmtum r BcfcnsSesa ft twi^tf

A. area«xi

Fig. 2.3 Schematic representation of stress in a formwork system [Vanhove et al., 2001]

Triaxial tests established that <p for SCC was equal to approximately 5°. The wall friction,

according to Janssen's models, can be expressed as follows:

T(K) = M-P'(h) Eq. 2.21

Janssen assumes that, at all points, pressure is at the slip threshold, which is taken in its

Coulomb form, and for a general approach, Eq. 2.21 should be rewritten as follows:

T(h)= jU.P'(h) + ro Eq.2.22

where x is the friction stress (or tangential stress), x0 is the threshold friction stress, and JU is the

friction coefficient, which is assumed to be constant in Janssen's model. A tribometer was

designed to find /*, [Djelal, 2001; Vanhove et al., 2004A]. It is therefore possible to write the

equilibrium equation between forces exerted by material on walls and vertical forces as follows:

17


A(P+dP)+ju.K.P'.(2e + 2L)dh = p.g.A.dh + A.P Eq. 2.23

where A is the area, e is thickness of the structure, L is width, p is density of the granular

material, and g is the gravity acceleration. According to Eq. 2.19 and Eq. 2.23:

P'(h) = - p-gA \-e f(2e+2L).M.K\ ^

Eq. 2.24 (2e + 2L)./u.K

This calculation concerns totally relaxed stacking and represents a long-term model. Eq.

2.24 was found to underestimate real lateral pressure exerted by fresh concrete on formwork.

Unlike the materials studied by Janssen, fresh concrete has a shear threshold [Tattersall and

Bloomer, 1979] and [Hobbs, 1976].

Tests carried out in the granular medium have shown that it is often necessary to apply a

coefficient taking into account physical phenomena, which are difficult to quantify. A friction

coefficient a is placed in front of the parameters describing grain-to-grain or concrete-to-wall

friction.

= p.g.A-ax0.{2e + 2L)y \ -A J* c ™ a(2e + 2L).ju.K

( (a(2e+2L).ft.K\\

Eq. 2.25 V

By analyzing the site results obtained at the end of the concrete casting, it was possible to

estimate the coefficient a as 0.15 for SCC cast from the top of the formwork and 0.34 for SCC

pumped from the bottom [Vanhove et al., 2004B]. Eq. 2.25 can then be used to determine the

lateral pressure exerted by fresh concrete on formwork by knowing the friction coefficient JU of

the concrete against the wall. The authors obtained a good fit between the on-site tests and the

tribometer results.

B. Roussel and Ovarlez's model [2005]

The model proposed by Roussel and Ovarlez [2005] considered that SCC is characterized

by a yield stress (T0), which grows with rest time. For sake of simplicity, the authors considered

that the yield criterion is a Tresca criterion, i.e. T0 is the maximum shear stress sustainable by an

internal plane. Moreover, below this yield stress, it behaves as an elastic material. The elastic

theory gives, for an isotropic elastic material and in the limit of small deformations, a linear

relationship between the stress tensor components ajj and the strain tensor components e :

ESy = (l + vp )av - VpSyCTu Eq. 2.26

18

where E is the Young's modulus, and vp the Poisson's ratio. Roussel and Ovarlez [2005] used

the coordinates x, y, and z in the directions shown in Fig. 2.4; the walls are the planes x = ± L/2

and y = ± e/2. The authors recalled a general framework of homogeneous isotropic linear

elasticity, and then predicted the behavior of an elastic material of density p confined in a rigid

rectangular formwork, i.e. displacements:

ux (±L 12, y, z) = uy (x, ±e/2,z) = 0 Eq. 2.27

The authors imposed a Tresca boundary condition at all points at the walls:

<Jxz(LI2,y,z) =<jyz(x,e/2,z) = ro Eq. 2.28

l r i / X

:• Length (L):

' . t •

- :Density = p-; •; • •

%

:tif.-'.

: Width©:

Fig. 2.4 Sketch for the formwork wall

Using the stress-strain relationship (Eq. 2.26) and internal equilibrium relationship, 3jajj= - p gj;,

the authors found:

°J?)= -Pg+^o (\ iY\

•+-

f C7(Z) = <7(Z)=K

yy\ •pg+2T0\-+-

<r„(x,y,z)=-T0\

<?yz(x,y,z)=-T0

X

JJ2,

Veil.

Eq. 2.29

Eq. 2.30

Eq. 2.31

Comments on the model

(1) Eq. 2.30 was found to be like Janssen model [Janssen, 1885] expressed by CTXX = cjyy = Kazz-

In the Janssen model, the parameter K is defined as the ratio of horizontal to vertical stresses

and is related to Poisson's ratio, vp: K = vp/ (l-vp).

19

(2) Shear stress at walls can take any value between 0 and to, and the normal stresses at the walls

correspond to a value between the profile described by Eq. 2.30 and hydrostatic pressure.

(3) Ovarlez et al., [2003] have shown that Janssen's model agrees well with experiments

performed in granular columns. However, there is no need for force chains, as in granular

materials, for writing such proportionality between vertical and radial stresses.

(4) There is a limitation of the model for low values of K; however, in the case of SCC with

standard air contents, this is not the case.

(5) The Tresca condition at z = 0 is not compatible with a free top surface for which T0(L/2,y,z) =

Xo(x,e/2,z)

(6) The vertical displacement is parabolic, which is not compatible with a flat displacement

imposed by a rigid base. The solution may be slightly modified near the column bottom.

(7) The Tresca condition may be satisfied somewhere in the bulk if the material is compressible

(i.e. K ^ 1), e.g., at the center of the formwork where x = y = 0. The maximum shear stress

on an internal plane is:

1 l-2vnf m

2V2 l-v

General case

Pg-2T0

f\ 1 ^ — + - Eq. 2.32

Boundaries on stresses at the walls may be computed out of any mechanical model (the

material may be elastic or not), simply by considering it as a Tresca material.

(1) The Tresca criterion imposes that the material cannot lean too much against the walls:

^{LI2,y,z)>-r0 and o^(r,g/2,z)^-T0

(2) Moreover, the material should not yield in the bulk, i.e.

o-xx(x,y,z)-azz(x,y,z)<yj2x0 and ayy (x,y,z)-crZ! (x ,y , z )< V2r0

here it is supposed that \axx |, \ayy < \ozz |

(3) The equilibrium equation of horizontal slices maybe written as:

Z . / 2 e / 2 / - 3 / \ \ ell Lll

J J ' dxdy=-pgLe-2Jaxz(L/2,y,z)dy-2\ axz(x,eI2,z)dx -LI2-el2\ & J -ell -LI2

(4) The authors assumed that axx(x,y,z) and o-yy(x,y,z) don't vary much with x and y so:

Lll ell L/2 ell

\ \ -ayy(*'y>z^xdy = 'ayy&Le and I I " C T « ( x ' y > Z ^ d x d y = ~ ° « ^ L e

-LI2-el2 -L/l-e/1

20

(5) Combining Eqs. given in a, b, and c, the authors obtained a lower boundary on normal

stresses at the walls close to the one obtained with elasticity, without any assumption on the

mechanical modeling of the material.

-cjyy{z)>pgZ-2T0z{\IL + \le)-sl2r0 Eq. 2.33 -aIX(z)>pgz-2T0z(^L + \/e)-y/2T0

The main difference with lateral pressure prediction given by Eq. 2.30 is that for sections

where the depth is on order or less than that of the width, the boundary could be lower than in the

case of the elastic model. In such an approach, the shear stress at walls can vary between 0 and x0,

depending on the local deformation. Even if the yield stress of the material is evolving, there

should not be any change at the walls as there is no deformation. In case of concrete, this

deformation can occur as the material slightly consolidates under its own weight [Khayat and

Assaad, 2005B], and a surface settlement can be obtained. This means that the assumption that

the yield shear stress is fully mobilized at the wall is licit and remains licit as the yield stress

increases with rest time.

Comparison with experimental measurements

The above approach was compared to a field experiment where concrete was used to fill

formwork measuring 10 m in height, 5.44 m in length, and 0.20 m in width. The concrete had a

unit weight (p) of 2,265 kg/m3 and was pumped from the bottom of the formwork at high casting

rate (R) of 43.5 m/hr. Pressure sensors mounted at 0.55, 1.95, and 3.36 m from the bottom were

used to determine lateral pressure. From Eq. 2.30, the shear stress r, between depths zt and zi+i

can be computed as follows:

1 T< = 2

^ ( zM ) - ^ ( z i ) , p g 1 Eq. 2.34

l(l/L + l/e)

The calculated and the measured yield stress evolutions are compared in Fig. 2.5 and show

quantitative agreement despite the uncertainty of yield stress measurement.

Practical application of the proposed model

(1) Evolution of apparent yield stress at rest

It can be assumed that the evolution of yield stress at rest is linear with time (at least at early

age), and it may be described using Eq. 2.35 as:

r0{t) = AthJ Eq.2.35

where t is the rest time and At/,a is a flocculation coefficient.

21

4 0 0 -

50Q-,

<u u s I/)

« g <8 I- & • 300. ® «

£ "g 200-« -a es .2 j£ ^100-

10

height frcrn She fbraswork bottom. •• 0,5 t o Z.Q m

2.0 so 3,4 m — 1 4 to. 5.5 ta —»—yie ld stress measuted with the BT-RHEOM

20 40 53 70

Time (min)

Fig. 2.5 Comparison between calculated shear stress and measured yield shear stress in terms of

resting time [Roussel and Ovarlez, 2005]

(2) Computation of the maximum pressure during casting

It was assumed that (i) casting rate (R) is constant (at a time t after the beginning of casting, H

= R.t); (ii) vertical deformation of the concrete under its own weight is always sufficient for

the shear stress to reach T#; and (iii) at the bottom of the zone where concrete is at rest

(everywhere in the formwork except in the upper layer of thickness e), the rest time is

maximum = (H - e)/R. Therefore, the yield stress of the concrete varies with depth and needs

to be integrated to compute the lateral stress at the bottom of formwork, as follows:

K f \ a =cr = —

" *" Le pgHLe - 2{L + e) j T0 (z)dz Eq. 2.36

Using Eq. 2.35, H = R.t, and with L » e: Eq. 2.36 becomes for:

Rectangular formwork with width, e

a„=a„ = K pgH-

V

{H-efAtl

eR f o r H » e <T = o- = KH

yy Pg

Circular column with radius, r

a = G„ = K pgH-rR

thix f o r H » r o- = o\„, = KH yy Pg-

eR

HAh*

Eq. 2.37

rR Eq. 2.38

The only parameter that is fitted in the predictions of the model is Athix. This factor is estimated

at 0.6 Pa/s for SCC tested by Billberg [2003]. This value is an average value as the additives

22

were different from one concrete to another. Athix. is equal to 0.2 Pa/s for the SCC tested by

Khayat and Assaad [2005B]. These values were in the same order of magnitude as those

obtained for the SCC tested by the authors (0.1 - 0.2 Pa/s), which confirms the quantitative

validity of the proposed approach [Roussel and Ovarlez, 2005].

The relative lateral stress a is defined as the ratio between the lateral stress and the

associated hydrostatic pressure at that depth.

a = axx _ axx =K

1 hyd. pgH 1 HA, hix

\

pgeR Eq. 2.39

This critical casting rate fulfills the following condition:

dH = K Pg-

2HAhlx

eR = 0

and thus, the critical rate can be expressed as follows:

2HAthlx Rcrit ~ '

epg

Eq. 2.40

Eq. 2.41

From Eq. 2.41, it becomes possible to decrease the lateral stress exerted on the formwork by

reducing the casting rate to the critical value.

C. Graubner and Proske's model [2005A]

Graubner and Proske [2005A] established a model to predict SCC lateral pressure against

the formwork based on the Silo Theory by Janssen. An experimental column measuring 4.3 x 0.3

x 0.3 m with measurement points Ml, M2, M3, and M4 at heights of 0.3, 1.3, 2.3, and 3.3 m

from the base, respectively, was used for monitoring lateral pressure (Fig. 2.6). The assumed

pressure distribution is shown in Fig. 2.7. The steps followed to construct the model are as

follows:

Equilibrium equation A. av + yc.A.dh = A.(crv +d<rv) + rw .U.dh

Pressure ratio

Friction coefficient

Differential equation

Vertical pressure

Horizontal pressure

CTh/oV

/ / ( t ) = rw / o-h

dh —dtv da U dt A

7cy

-JMl).M(t).v.^dt U .—c A Jrc.v

$m).M(t).v.jdt dt+C

o-.=crv.A(t)

Eq. 2.42

Eq. 2.43

Eq. 2.44

Eq. 2.45

Eq. 2.46

Eq. 2.47

Eq. 2.48

23


V s ; / / " " c n n r . r B t e S ? 1 / ! ^ / l e a s t from

Concrete

30 cm Stress in the silo

A Area of the mould cross-section H Height of the formwork U Perimeter of the mould cross-section yc Fresh concret volume weight oh Horizontal pressure o„ Vertical pressure X Pressure ratio u Friction coefficient xw Friction stress t, tA, tE, Time, initial setting time,

and final setting time

Fig. 2.6 Experimental column and sketch for pressure calculations

[Graubner and Proske, 2005B]

Concrete level

F«$h eoncretej

Hardened c,

_ Horizontal pressure ah

Hydrostatic pressure a,,'ma'until the max. value

Reduction aftertiA

Fig. 2.7 Pressure distribution [Graubner and Proske, 2005A]

Graubner and Proske used the testing machine shown in Fig. 2.8 to determine the model

parameters. The pressure ratio (X) and the friction coefficient (u) for the calculation can be

obtained from Eq. 2.49 and Eq. 2.50, respectively, or from Fig. 2.9.

^ = 1-0.18 +2.43 •9.14 + 7.84 •1.79 a =80kN/m2

u=0.23 -1.093 +1.72 — -0.636 V * E /

>o =80-170kN/m2

Eq. 2.49

Eq. 2.50

24

Jh1

m

cv VM !

""' ~!

••jr2**-**'- ^

•-'-£!? r , ""J

1 : : • ' . •

' " ' • ' - ; ' " " " • >

-*- ' fjj

Formwork 25/25/25 cm and steel frame

Load cell

Metal blade / Coated wooden panel

jf TW ; v = 0 - 5 mm/min =lp

Fig. 2.8 Testing machine used to determine model parameters [Graubner and Proske, 2005A]

1.00 'tjM m- 4 0 i M « 1.0 • • • • * * .

'% \

I I; t ft « h .1 ©,£ -ji MelsJWidt J 11 Horizontal pr«f*w« 0 • 30C kfj*rt*

" sm ftsOt -,. ?5 cm b 0,4

*& ,«

* Pressure ratio •• Coeffietnf of friction

t ,« 8 S |V4MM«H«Sti

9S

§:4

mm *0,51

0:3 1

1:201 4

0,108 0 35 | .

» 1 0.2-I • - •

0.036 # ( y i i i 0,008 0,013^ 0;;01% 0,0'% w s 9 * - — — — -

4.157 h 0,22

0.0

B . l l 0,1 6

6.0 0,2 ft* 0,8 Q;8 1,0 StondaiKf se«! taw M( H

Fig. 2.9 Pressure ratio X and friction coefficient |j, for the calculation of formwork pressure

[Graubner and Proske, 2005A]

An example for maximum lateral pressure calculated using the Graubner and Proske's

model given in Eq. 2.47 and Eq. 2.48 which use the relationships of pressure ratio X and friction

coefficient [i of Eq. 2.49 and Eq. 2.52, respectively, is illustrated in Fig. 2.10. In this figure, b =

(bsi. bS2) / (bsi + bS2 ), where bsi and bS2 represent the dimensions of the formwork cross-section.

For wall elements and small cross sections: b is taken as the smallest width of the section. A

comparison between data measured to the calculated from the model is shown in Fig. 2.11.

Consideration of setting time

^ a x ( lA ) = °"h,max (chart) Eq. 2 .51 tA<4h -»

tA>4h -> o-h>max (tA ) = o-h;max (chart). 4/?

Eq. 2.52

Consideration for pumping from bottom: hydrostatic pressure should be assumed.

25

From the results shown in Graubner and Proske's model, it can be concluded that:

1. Considering the time-dependent behavior of the concrete, the Silo theory describes

approximately the real stress state.

2. In general, the assumption of hydrostatic pressure is not warranted.

3. The maximum formwork pressure depends significantly on the casting rate, setting time of

the concrete, and formwork width. The smaller the formwork width, the lower the formwork

pressure would be.

. "r * r< - s . Z '.'. T t .. .,

• f '

* z - t :z - % ' ~ - z> a ' o r n

: c - ' J

* " £ . 1 . - i * *s . *

b = 2 00 m e = inf-Hfe '. L.4U <6 Jt 3 coftrete 'Ho.vao a'»

DiN 18 2t8Max. wafe

8 7 S - 1 1

!/:

50 IS I- 7 5 +? — - "

itm*iv i»i f Veieeiyaf fishqgv |m#il

Fig. 2.10 Calculated maximum pressure using Graubner and Proske's model [2005A]

§

5

I 1

Rising velocity: (Colour)

• v £ 1 m/h

» v <, 2 m/h

• v <, 5 m/h

O v i 12,5 m/h

9 v < 25 m/h

• v>25m/h

Assumtioiv. b„,„ = b

Fig. 2

Large symbols: Setting time is known 0 50 JQ0 i50 Small symbols: Setting time is estimated

Calculated Formwork Pressure [kN/m-J

11 Comparison between the calculated data using Graubner and Proske's model and the

measured data - Influence of reinforcement [Graubner and Proske, 2005B]

26

-ss**-*aSif>e!MK#, 6 * JO es

• - * -S.vs25:Wl i« ts1 ' i ,g .» -•~f«et»!ess#), 6 = J< en

»•• • mmmmM., 6 * «l m f i e s t a s * *

*• >'.* .- y.tMr w-> ".-Oaf. <;

. ^ v ' 1^ ** -*-;„" *.r - SM Ik's j ;

°<ofc>'

M«3IS0ss5 i s I l l s

»• s 23 J i f fs?

-X t i l OaBislsfl x-. I l?j !###

3S

\ , = 1WM4*rf

2s *3 m m

| a 3? £»

% * g h

* 2 3 , O M * * f

580 m # «t i e t i »

."..^ft* "NK

-Xs

>* ~ 2s « k fsyrea -| " v ~;^§??#i ?^ci^- f "1

s§ i » ' » 2c :

0

\ . »~ *» v ^ . l*?*

Is» ,84 « * | .' !,.'» Mils j ^

•;, * KtS WW*' I * MesTfl-?6#m f

\ X

'%

4# 60

• - v = 2 m/h v = 2 tn/h, rechnerisch

—*- v = 10 m/h —er-v =10 m/h, rechnerisch

hydrostatisch

Wande - RWTH Aachen

b = 24 cm tE' = 3,0 h

yt = 23,5 kN/mJ

sm = 70-75 cm

20 40 60 Frischbetondruck [kNAna]

80

Fig. 2.11 (cont'd) Comparison between the calculated data using Graubner and Proske's model

and the measured data - Influence of reinforcement [Graubner and Proske, 2005B]

D. Khayat and Assaad's model [2005A]

An extensive investigation was carried out by the authors to determine the key factors

affecting SCC formwork pressure. The thixotropy of SCC has been shown to have considerable

influence on both initial lateral pressure and pressure decay. SCC with high degree of thixotropy

is shown to exert lower initial lateral pressure and higher rate of pressure drop in time compared

to those with low thixotropy. Khayat and Assaad [2005A] used an instrumented PVC column

27


measuring 0.20 m in diameter and 2.8 m in height to monitor the variations in lateral pressure of

SCC soon after casting. A similar column measuring 1.1 m in height was used to monitor

pressure decay until pressure cancellation. The columns were discharged the SCC continuously

from top at a rate of rise of 10 m/hr, for the most part, and without any mechanical consolidation.

In total, 70 SCCs of different mix designs and material constituents were prepared to derive

the pressure prediction models. The mixtures had slump flow, temperatures, unit weights, and air

volumes of 650 ± 15 mm, 20 ± 3 °C, 2,200 ± 200 kg/m3, and 7% ± 2%, respectively. The

breakdown area (At,) determined from structural breakdown curves (see Section 2.3.2-B) was

used to determine thixotropy at time intervals. The values AM, Ab2, and Ab3 were determined at

time intervals of Tl (0-30 min), T2 (60-90 min), and T3 (120-150 min) [Assaad et al., 2003A].

Relationship between thixotropy and relative lateral pressure

The relationship between thixotropy (At,) and relative lateral pressure (K=Pmax/Phyd)

obtained near the bottom of the 2.8-m-high column is illustrated in Fig. 2.12 for all tested SCC

mixtures. The relative lateral stress K is defined as the ratio between the lateral stress and the

associated hydrostatic pressure at that depth. The equations enabling the estimate of K at different

time intervals with respect to thixotropy can be expressed as follows: Ko (%) = - 0.047 Abl + 105.8 (R2 = 0.89) Eq. 2.53

Kioo (%) = - 0.099 Ab2 + 112.2 (R2 = 0.85) Eq. 2.54

K2oo (%) = - 0.125 Ab3 + 116.8 (R2 = 0.84) Eq. 2.55

where: Ko, Kioo, and K2oo : relative pressures determined at elapsed times of 0, 100, and 200 min

after the end of casting, respectively, kPa; and

Abi, Ab2, and Ab3: structural breakdown areas determined at Tl, T2, and T3 time

intervals, respectively, J.m /sec.

The Kioo and K2oo values can be estimated from the AM index since Ab2 and Ab3 values

showed good correlations to the AM value, as demonstrated in Fig. 2.13.

A M = 1.14 Aw (R2 = 0.96) Eq.2.56

Ab3 = 1.29 Abi (R2 = 0.90) Eq. 2.57

Substituting Eq. 2.56 and Eq. 2.57 into Eq. 2.54 and Eq. 2.55, Kioo and K200 values can be

estimated from breakdown area determined initially at time interval Tl (0-30 min), as follows:

Kioo (%) = - 0.113 Abi + 112.2 (R2 = 0-83) E* 2 - 5 8

K200 (o/0) = - 0.161 Abi + 116.8 (R2 = 0.81) Eq. 2.59

28


100

80

60

> 40

20 H

K 0 = -0.047 x A b 1 + 105.8

R2 = 0.89

K200 = -0 .125xAb3+116.8 a

FT = 0.84

x Ab1 @ 0 to 30 m in

a Ab2 @ 60 to 90 min

A Ab3@ 120 to 150 min

Kioo = -0 .099xA b 2 + 112.2

50 800 200 350 500 650

Breakdown area (J/m3.s)

Fig. 2.12 Breakdown area (Ab) vs. relative lateral pressure measured initially, and after 100 and

200 min [Khayat and Assaad, 2005 A]

800

0 100 200 300 400 500 600

Breakdown area at Tj (J/m3.s)

Fig. 2.13 Relationship between breakdown areas determined at various time intervals

[Khayat and Assaad, 2005A]

Relationship between drop in apparent viscosity and Kn values

The drop in apparent viscosity was also determined to estimate lateral pressure. This value

is calculated as: Anapp = (x; - xe) If , where y refers to the shear rate (s1) corresponding to a

given rotational velocity. Variations in An app determined at 0.3 and 0.9 rps during (0-30 min)

with respect to Kn are plotted in Fig. 2.14.

The equations relating K0 to An app at 0.3 and 0.9 rps during (0-30 min) were as follows:

29


For N = 0.3 rps, Ko (%) = - 0.112 An app + 103.9 (R2 = 0.74)

For N = 0.9 rps, KQ (%) = - 0.168 An app + 106.7 (R2 = 0.82)

Eq. 2.60

Eq. 2.61

100

_ 90

£ o • * *

es % 8 0 -

>

70

• Drop in app. vis. @ 0.3 rps

° Drop in app. vis. @ 0.9 rps

K 0 = - 0.112 X A T I app+ 103.9

R2 = 0.74

K 0 = -0 .168 x ATI app + 1 ° 6 - 7

R2 = 0.82

50 100 150 200 250 300

Ailapp (Pa-s)

Fig. 2.14 Drop in apparent viscosity vs. initial lateral pressure [Khayat and Assaad, 2005A]

Effect of concrete head on lateral pressure

In order to evaluate effect of concrete head on initial pressure, Ko values were determined

at various heights along experimental columns filled with concrete to heights of 1, 1.3 2, 2.4, and

2.8 m. Fig. 2.15 presents the relationships between concrete head and Ko values for SCC with

different thixotropy values for mixtures cast at 10 m/hr. As expected, the Ko values decreases

with the increase in height given the longer time duration needed to fill the column sections.

100

^ 96

s o SB 4>

>

92

88

84

80

76

;

•

.

Rate

:

of casting = 1 Om/h

^ — ~ 1 . 0 m

\ 2 . 0 m

^ ^ ^ 2.4 m

2.8 m

100 600 200 300 400 500

Breakdown area at Ti (J/m3.s)

Fig. 2.15 Effect of concrete head on variations of Ko values for mixtures having various degrees

of breakdown areas [Khayat and Assaad, 2005 A]

30


Effect of casting rate on proposed models

Casting rate (R) used to derive the above pressure models were set to 10 m/hr. In order to

evaluate the effect of changes in R on the prediction models, a reference SCC mixture was cast in

the 2800-mm high column at R values of 5, 15, 25, and 30 m/hr. Based on the results, the

following model was proposed to account for variations in casting rate:

Ko (%) = 9.254 Ln (R) + 66.5 (R2 = 0.95) Eq. 2.62

The K models established earlier for R of 10 m/hr can then be modified as follows:

Ko at any given R (m/hr) = Ko determined at 10 m/hr + AK0 Eq. 2.63

where AKo is the spread of Ko from values predicted for R = 10 m/hr. A number of A Ko values

are given in Table 2.1 for various casting rates.

Table 2.1 Spread in pressure from relative pressure determined at casting rate of 10 m/hr

Casting rate (R), m/hr

A Ko (%) from Eq. 2.63

5

-6.1

7

-3.1

10

0

13

2.3

15

3.6

17

4.7

20

6.1

23

7.3

25

8.0

30

9.6

Relationship between measured and predicted K values

The relationships between K values measured directly from the 2800-mm experimental

column (KM) and those predicted (Kp) using Eq. 2.53 and Eq. 2.55 are plotted in Fig. 2.16 for the

tested SCC. Good correlation can be established between the two values.

0 20 40 60 80 100

Measured K values (%)

Fig. 2.16 Relationship between predicted and measured K [Khayat and Assaad, 2005A]

31


2.3 Relationship between form pressure and rheology of SCC

2.3.1 Thixotropy of cement-based materials

As illustrated above, form pressure developed by SCC can be related to the degree of

structural build-up of the material after a given period of rest. In cementitious materials, this

structural build-up is a function of both the reversible structural changes from the thixotropic

phenomena and the irreversible structural changes due to hydration mechanisms altering the

resulting microstructure. Barnes et al. [1989] defined thixotropy as a decrease in time of

viscosity under constant shear stress or shear rate, followed by a gradual recovery when the stress

or shear rate is removed. The transition of the material from the at-rest state to shearing

conditions, and vice versa, is illustrated in Fig. 2.17 [Barnes, 1997].

Oftentimes, thixotropy is confused with shear-thinning behavior of non-Newtonian fluids.

It is important to point out that thixotropy is related to Non-Newtonian time-dependent changes,

whereas shear-thinning refers to Non-Newtonian time-independent changes. In a plot of viscosity

versus time (for a given shear rate), a shear-thinning fluid has a constant viscosity for any given

time, whereas a thixotropic fluid displays a decrease in viscosity. In actuality, nearly all shear-

thinning materials are thixotropic because it takes time for the microstructure to realign itself.

Fig. 2.17 Breakdown and build-up of a 3-D thixotropic structure [Barnes, 1997]

Thixotropy of cement-based systems is strongly dependent on mixture composition and

processing parameters, such as mixing and vibration. Tattersall and Banfill [1983] reported that

cement characteristics, such as packing density, fineness, and chemical composition can

significantly affect thixotropy. Struble [1991] suggested that thixotropy and the yield stress are

dependent on both particle packing and interparticle links responsible for flocculation, while

viscosity depends primarily on particle packing.

32


Thixotropic behavior typically occurs in heterogeneous materials, and it occurs due to the

finite time that it takes for the microstructure to change from one state to another. The specific

causation of thixotropy depends on interactions at the molecular level and unfortunately, these

mechanisms are poorly understood. The decrease in apparent viscosity that is accompanied with

thixotropy is believed to be due to the resulting flow altering the microstructure. When a specific

microstructure is agitated, the viscosity will decrease with the shearing time until an equilibrium

state (the lowest energetically possible state) is achieved. Thus, the time-scale in which the

microstructural changes take place is an important parameter in the consideration of thixotropy

[Ferron et al., 2006]. According to Barnes [1997], such structural changes can be attributed to

two simultaneous processes: shear induced breakdown of the structure and build-up of the

structure whereby the yield stress increases with increasing recovery time. Helmuth (as reported

by Struble [1991]) suggested that mixing breaks down the flocculent structure responsible for

thixotropic behavior. Thus, from a microstructural perspective, thixotropy is a result of structural

degradation due to the rupturing of floes of linked particles [Saak, 2000]. In cement paste, it is

likely that thixotropy is governed by a combination of reversible coagulation, dispersion, and

then re-coagulation of the cement particles [Wallevik, 2003]. When a cementitious suspension is

sheared, its network structure is broken into smaller agglomerates and with continued shearing

there is eventually an equilibrium state in which the agglomerates cannot be broken into smaller

fragments. When the suspension is at rest, the particles can form weak physical bonds and

agglomerate into a network. The rheological behavior of the suspension is related to this network

structure and the rate at which it can form.

The phenomenon demonstrating the effect of rest time on thixotropy for highly flowable

concrete containing VMA is plotted in Fig. 2.18 [Assaad, 2004]. After 2 min of rest following

placement of the concrete in the bowl of a rheometer, the concrete was subjected to constant

rotational speed of 0.9 rps resulting in a continuous breakdown of the flocculated structure. Such

breakdown with time under the imposed shear rate indicates the origin of thixotropy. An

equilibrium structure is achieved after approximately 10 sec where a balance between

flocculation and deflocculation is reached. This results in a constant viscosity (r| = xj y, where xe

is the shear stress at equilibrium (Pa), and y is the shear rate in (s"1). In contrast, under static

conditions, individual particles of the concrete begin to collide and flocculate causing progressive

change in the microstructure through the formation of a gel structure and interparticle links. In

33


other words, when the shear is stopped, and the material is allowed to rest for 4 min, the

experiment conducted using the same rotational speed of 0.9 rps shows that the measured

viscosity is initially higher, but that it decreases again to the same equilibrium shear stress noted

after the 2-min rest period.

The increased flocculation in cement-based materials can also be caused by hydrogen and

ionic bonds that can develop between adjacent molecules leading to a rise in cohesiveness

[Khayat et al., 2002A]. The longer the material is maintained at rest, the more the thixotropic

structural build-up becomes significant, thus requiring higher initial yield stress to breakdown the

structure. Once shearing occurs, the particle spatial distribution and alignment become

asymmetrical in the flow direction, and the number of entanglements or associations decreases to

the minimum. This leads to similar values of viscosity at equilibrium [Khayat et al., 2002A].

ouu -i VMA concrete

A 4-minutes rest period

x 2-minutes rest period

N = 0.9 rps

0 5 10 15

Shearing time (sec)

Fig. 2.18 Variation of viscosity with time for VMA concrete following 2 and 4 min of rest,

[Assaad, 2004]

Several attempts have been made to relate the experimental observations, such as those

given in Fig. 2.18, to physical processes taking place in flocculated suspensions in order to

propose physical models accounting for the breakdown phenomenon. The most quantitative

theory was elaborated by Tattersall (summarized in [Tattersall and Banfill, 1983]) for describing

the thixotropic behavior of cement paste. The basis of the structural breakdown theory is related

to work done in a rotational viscometer to overcome normal viscous forces, to break existing

linkages, and to maintain broken ones. The equation used to describe the decay in stress with

time for a given shear rate can be expressed as:

34


x = xe + (TO - xe) e Eq. 2.64

where xe: equilibrium stress, To: initial stress, 0: breakdown constant, and t: time. The breakdown

constant was shown to depend on the number and strength of interparticle links, as follows:

(2 n K) w (w - w,) P = ~ " Eq.2.65

where K and w\\ constants, w: angular velocity (equivalent to strain rate), n0 : number of links at

the beginning of the experiment, and yr. work required to breakdown each link.

2.3.2 Concrete rheometer

There are many types of concrete rheometers in the concrete market such as: BML

rheometer, BTRHEOM rheometer, IBB rheometer, Two-Point rheometer, and UIUC rheometer

(used only for oil tests). The MK-III concrete rheometer (commercially known as IBB rheometer)

modified by Tattersall and Bloomer [1979] (Fig. 2.19 left) was used initially to determine time-

dependent viscosity changes necessary to assess thixotropy [Beaure, 1994]. This rheometer is

capable of measuring the rheological parameters of low to very high workability concrete. The

apparatus uses a data acquisition system to drive an H-shaped impeller rotating in a planetary

motion at a constant rotational speed. The measures taken by this rheometer are torque (Nm) as

function of time at rotational frequencies relative to deformation-rate in (rps). The mathematical

equations that help determining rheological parameters in fundamental units are described in

following sections. The H-shaped impeller measures 100 mm in height and 130 mm in width

(Fig. 2.20 left). The concrete bowl leaves a 50 mm gap between the impeller and the bowl. The

recommended maximum-size of aggregate for this geometry is 25 mm. The sample size is 21 L.

The first reading obtained at a given rotational speed is considered as the initial maximum

torque (Ti), and the mean of the five smallest measurements at the same speed is taken as the

equilibrium torque (Te). The general testing procedure consists of varying the rotational speed

between 0.3 and 0.6 rps and noting AT (AT = T; - Te). AT value is considered as an indication of

the structural breakdown of the plastic concrete. The assessment of thixotropy is, however,

problematic with the H-shaped impeller given a long waiting period necessary prior to develop an

appreciable AT value. Slip flow at the interface of the H-shaped impeller with the surrounding

concrete and the reduced sheared surface of the concrete during rotation are the main reasons for

the delay before appreciable differences in T; and Te values could be measured.

35

In order to facilitate the evaluation of thixotropy at early age, an alternative form of the

impeller was designed by Assaad [2004]. The new impeller functions on the same principle of the

IBB rheometer except the H-shaped impeller is substituted by four 1.5-mm thick blades arranged

at equal angles around an 8-mm diameter shaft. A schematic illustration of the new impeller is

given in Fig. 2.20 (right).

Fig. 2.19 Tattersall MK-III concrete rheometer in its commercial version (left) and with the

modified vane (right)

Gears for planetary motion

Main shaft

(O = 5 mm)

Level of concrete,

.£ Bowl dimensions:

Diameter =360 mm

Height =280 mm

100 mm

130 mm

Concrete level

i Main axis (0 = 8 mrnl

h

r

—>

r

1

*N(rps)

>

* D '

< >

i

' •

j

1

Zi >

H

L

Z2

DT

Schematic of H-shaped impeller rotating in Schematic of four-bladed vane [Assaad, 2004] a planetary motion Fig. 2.20 Geometry of vane used to measure the rheology of concrete.

36

The vane geometry offers an important advantage in the rheological measurements as it

enables the shearing to take place along a cylindrical surface circumscribed by the vane. Such

surface can be considerably greater than the one resulting from the use of the H-shaped impeller.

The material yields within itself, so that all problems associated with slip flow against smooth

surfaces are absent. In addition, the introduction of the vane into the suspension does not cause

significant disturbance to the sample prior to measurement even at high impeller speeds of 0.9

rps, which is particularly important when evaluating thixotropic suspensions [Liddell and Boger,

1996]. To prevent progressive migration of large particles away from the center during rotation, a

slot was cut through each of the four blades of the vane. Such design can enable the impeller to

remain in contact with "new" concrete during motion, as materials displaced from the center of

the bowl can be immediately replaced by fresh concrete coming from the outer part of the bowl.

The vane dimensions were selected to satisfy the following criteria: H/D < 3.5, DT/D > 2, Zi/D >

1, Z2/D > 1, Zj/H > 0.5, and Z2/H > 0.5 [Nguyen, 1983]. Definitions of symbols are offered in

Fig. 2.20 (right). The selected vane measures 130 mm in height and 90 mm in diameter. The

apparatus was calibrated using same braking system employed for the IBB rheometer.

2.3.3 Dynamic yield stress and plastic viscosity

The torque readings (7) are calculated into shear stress (r in Pa) while the rotational

frequency (TV) is calculated into shear rate ( y in s"1). Determinations of r and / enable evaluating

a so-called flow curve (r vs. / ) . The (r vs. y) curve is obtained by increasing the (N) from low to

high (0.3 to 0.9 rps) and then decreasing stepwise from high downwards to low (0.9 to 0.3 rps).

The equilibrium stresses (torque) corresponding to each deformation rate (rotational frequency)

are then used for a regression to obtain yield stress (to in Pa) and plastic viscosity (nPi in Pa.s).

Assuming the concrete behaves according to the linear Bingham model (Eq. 2.66), To is the stress

corresponding to the intercept of the (r vs. y) curve with stress axis, i.e., the value obtained by

extrapolation of the regression to zero y, and JLIPI is the slope of the (T VS. / ) curve (Fig. 2.21).

T = To+Vpl-r Eq.2.66

T and y are calculated using measured values of T and N as follows:

r=T/K Eq.2.67

37


with, K = 27rR2 (H +Ah) + 4/3 7iR3 = 0.0019393425 m3; and H, Ah, and R are height, correction

for height, and radius of the vane, respectively. For the vane used in studies of Assaad [2004], H

= 130 mm, Ah =1.85 mm, and R = 45 mm, respectively.

N r Ln(S)

T , » (n-Ln(S))2 (n-Ln(S))4

3 45 Eq. 2.68

where: S = — = — = 2.889 and n = that can determined as slope of the curve (Fig. 2.22). R 4.5 dLnr

a.

t/>

Equilibrium stresses corresponding to each shear rate level

Shear rate (1/s)

Fig. 2.21 Bingham model

Lnr

-1 y= 1.2404 x- 7.8411

R2= 0.99

Fig. 2.22 Calculation method for the parameter (n)

2.3.4 Approaches to quantify thixotropy of concrete

There exist a number of test protocols that can be used to evaluate structural build-up or

thixotropy of concrete. Measuring the rheology of cement paste gives an indication of the

colloidal state and interactions that are occurring. There are no standard methods to measure

38


thixotropy, but typical thixotropic experiments often consist of either rheological tests conducted

at a constant shear rate (equilibrium flow curves) or using varied sheared rates (hysteresis curves)

[Lapasin et al. 1983, Barnes, 1997, and Mewis, 1979].

A. Hysteresis curves

Thixotropic materials have typical hysteresis loops that are readily plotted from (T VS. / )

experiments. Ish-Shalom and Greenberg [1962] used hysteresis measurements to characterize the

flow properties of cement paste. In such tests, y is increased from zero to some pre-defined

maximum value and then decreased back to zero. When x is plotted as a function of y, the up

(loading) and down (unloading) curves can be obtained, as illustrated in Fig. 2.23. The enclosed

area between the up and down curves (i.e. hysteresis) provides a measure of degree of thixotropy

in the sample [Tattersall and Banfill, 1979]. The down curve is normally linear and fits the

Bingham model (Eq. 2.66).

Shear rate (y)

Fig. 2.23 Hysteresis loop flow curve [Ish-Shalom and Greenberg, 1962]

Although the hysteresis loop testing procedure has been used for measuring the flow

characteristics of cement paste, the shape of the hysteresis loop was criticized by several

researchers. For example, Worral et al. (reported in [Banfill and Saunders, 1981]) showed

experimentally that two suspensions of quite different thixotropic properties could give similar

hysteresis loops. Barnes [1997] also reported that the hysteresis loop tests are not recommended

for two main reasons. Firstly, the loop test is often carried out too quickly, and inertia effects due

to the measuring head are introduced but not always accounted for. Secondly, a test where both

shear rate and time are changed simultaneously on a material where the response itself is a

39


function of both shear rate and time is not quite adequate. This is because the response cannot

then be resolved into the separate effects arising from both variables. When carefully run and

interpreted, the use of the hysteresis loop can be useful for evaluating the structural build-up of

cement-based materials. For example, Douglas et al. [2005] used this approach to show that

structural build-up within the induction period of hydration is time dependent (Fig. 2.24) and is

adversely related to the superplasticizer content in the mixture. The higher the superplasticizer

dosage is, the lower the thixotropy and the lower the rate of structural build-up are.

aoo

~LOQf»iobrt«g!o equilibrium

«4rest - TO intra

-Isest = SO min

»t»«t = SO in in

} 200 300 400 500

Shear rate (s"1)

Fig. 2.24 Effect of superplasticizer content and rest time on thixotropy and hysteresis loops

[Douglas et al., 2005]

B. Structural breakdown curves

The other method that can be used to characterize thixotropy of cement-based systems is

the steady state, or equilibrium, approach illustrated in Fig. 2.18 and schematically in Fig. 2.25.

The steady state approach consists of measuring the behavior of shear stress versus time while

fixing a constant shear rate. The initial shear stress necessary to breakdown the structure (n) is

generally considered to correspond to the initial structural condition, [Shaughnessy and Clark,

1988]. On the other hand, the shear stress decay with time towards an equilibrium value (xe)

corresponds to an equilibrium condition that is independent of the shear history.

The steady state approach has been widely used for interpreting the effect of chemical

admixtures as well as different forces existing in cement-based systems on the flow conditions

([Tattersall and Banfill, 1983] and [Ghezal et al., 2002]). In addition to its simplicity, this

approach provides an advantage compared to the hysteresis loop tests as it enables measuring the

40


entire shear stress range as a function of time for a given shear rate. Conversely, hysteresis loops

normally measure transient flow properties somewhere in between the peak and equilibrium

stresses for a given shear rate [Saak et al, 2001].

7 R " T peak

Ax

Time

Fig. 2.25 Steady state flow curve [Shaughnessy and Clark, 1988]

Assaad et al. [2003 A] used structural breakdown approach to evaluate thixotropy of SCC.

The concrete was subjected to constant N of 0.3, 0.5, 0.7, 0.9 rps. A typical set of structural

breakdown curves is plotted in Fig. 2.26. The corresponding shear rates were calculated

according to Legrand's approach [Legrand, 1971] by considering that the specially designed

bladed vane rotates in an infinite medium. In this approach, immediately after the vane drive

mechanism is started, readings of the torque are noted as function of time without delay. The first

reading is considered as the initial maximum torque (Ti) necessary to initiate the flow of the vane.

The (TO is used to calculate the peak yield stress (n), which corresponds to the initial structural

condition. The mean of the five smallest measurements, over 25 sec duration at each N, is taken

as the equilibrium torque (Te). The Te is used to calculate the equilibrium shear stress (xe), which

corresponds to an equilibrium condition that is independent of the shear history, for that speed.

The rest period during which the concrete was not subjected to any shearing action prior to

conducting each of the four structural breakdown tests was 5 min for SCC. This period was found

necessary to obtain sufficient spread between Ti and Te values. In total, the time required to

perform all of the thixotropy tests at the four rotational speeds was approximately 28 min.

The spread between xj and xe gives, for a given N, a measurement of the amplitude of the

structural modifications inside the tested material. Lapasin et al. [1983] suggested that the area

41

1


comprised between the initial flow curves [x; (N)] and the equilibrium flow curves [xe (N)] can be

used to quantify thixotropy. This area "breakdown area (Ab)" provides a measure of the energy

done per unit time and volume of concrete necessary to break the initial linkages and internal

friction in order to pass from the initial state into equilibrium state. The Ab can be calculated as:

Eq. 2.69 0.3

Example of the variations in shear stress with shearing time and the calculation of the

structural breakdown area between x; vs. N and xe vs. N plots are shown in Fig. 2.26 and Fig.

2.27, respectively. These data are determined for SCC made with ternary cement containing silica

fume (SF) and fly ash (FA) and a set-retarding admixture (RET).

800

• N = 0.3rps

- N = 0.5rps

- N = 0.7rps

x N = 0.9rps

10 15

Time (sec)

Fig. 2.26 Typical example of structural breakdown curves for SCC [Assaad et al., 2003 A]

C3

800

600 H

u 400 (A 1. « -C (/} 200

SCC-SF+FA-RET

• Initial shear stress

a Shear stress at equilibrium y = 237 2 x + 316 3 x. +

R* = 0.9986

Breakdown area «Abw

y = -7,9 + 235.9 * + 4 3 t x* R2 is 0,9943

0.2 0.4 0.6 0.8

Rotational speed (rps)

Fig. 2.27 Typical example of structural breakdown area calculation [Assaad et al., 2003 A]

42

C. Apparent viscosity

For a given y, the apparent viscosity (/7app) is calculated as the ratio between rand the y:

>?, app

T

r Eq. 2.70

The drop in apparent viscosity (Ar|app) at a constant shear rate can also be considered as an

index to evaluate the degree of thixotropy. The value of (At|app) was used by Assaad et al.

[2003 A] at various shear rates to characterize thixotropy of SCC. At a given rotational speed, the

concrete is sheared causing destruction of the links between the internal colloidal particles until

equilibrium state is reached. The difference between the initial shearing stress representing initial

destruction and the equilibrium shearing stress measured over the applied shear rate can be

defined as the drop in apparent viscosity (Ar|app) (Fig. 2.28) and can be calculated as follows:

T- — T Magnitudeof thixotropic structure= — -

f

T, T • • ~ 'fappj Tlapp,e — ", app

1-

2xco

[ T° 1 U-0

X2 7T

x Ln ro"

Eq. 2.71

Eq. 2.72

where; Xi : initial shear stress (Pa); xe : equilibrium shear stress (Pa); y : applied shear rate (s"1); and

co : rotational speed (rps).

! s I

Tbne<9)

| ^ a p p j

" app,c

i - T j

~Tt

r •

Shear rate (1/s)

Fig. 2.28 Principle for the evaluation of drop in apparent viscosity

D. Static yield stress at rest

The static yield stress (also referred to as shear-growth yield stress) can be used as measure

of the strength and number of interparticle bonds that are ruptured due to the applied shear or

stress [Tattersall and Banfill, 1983]. Tattersall and Banfill reported initial static yield stress

43


values for a normal consistency cement paste ranging from 50 - 200 Pa [Tattersall and Banfill,

1983]. Similarly, the static yield stress values for conventional concrete range from 500 Pa to

several thousands Pa, and for SCC these values tend to range from 0 - 60 Pa [Wallevik, 2003].

The protocol adopted for the determination of static yield stress at rest (xo rest) consisted of

applying a minute and constant N to a vane immersed in undisturbed material following a certain

period of rest and recording the resulting torque as a function of time. In the case of concrete

shown in Fig. 2.29, the SCC was placed in the bowl of the rheometer and allowed to rest for 5

min and N was set at 0.03 rps. This was chosen so that the maximum torque (Tmax) is not affected

by the N of the vane. As the blades start rotating only when the applied shear stress exceeds the

resistance created by the friction forces and bonds between the particles. The profile shows a

linear elastic region followed by a yielding moment where the torque exerted on the vane shaft

reaches a maximum value corresponding to the beginning of the microscopic destruction of the

bonds between the particles and the suspension allowing the material to flow. Beyond this value,

the torque decays towards a steady state region. It constitutes a new dynamic arrangement of the

particles, offering an internal friction less than that resisted the first destruction. Therefore, the

peak shear stress value is considered as the (to rest)- The calculation of (xo rest) from the measured

(Tmax) requires knowledge of the geometry of the yield surface and shear stress distribution on the

surface. Dzuy and Boger [1985] assumed that the material is sheared along a localized cylindrical

surface circumscribed by the vane, and this shear stress is uniformly distributed over this surface.

Hence, good approximation of the shear stress can be calculated as follows:

TO rest ~ Tmax / K E q . 2 . 7 3

Variation of (xo rest) with rest time can also be used to determine another degree of

thixotropy expressed as the rate of gain in static yield stress with time [XQ rest(t) in Pa/s].

0.20 1

Time (sec)

Fig. 2.29 Typical torque-time profile for concrete with 200 mm slump [Assaad et al., 2003A]

44

E. Structure build-up at rest minus irreversible structure

Billberg [2006] used a Physica MCR 300 rheometer to assess thixotropy of micro mortar

where only sand fines sieved passing 0.125 mm were used. Five micro mortars made with

Cemll/A-LL 42.5 R (Swedish cement for housing) and w/c ranging between 0.34 and 0.42 were

tested. As shown in Fig. 2.30, the testing protocol was carried out to measure the dynamic

rheology before and after a given rest period to find the magnitude of the irreversible structural

change with time. After the rest period and before the last dynamic rheology measurement, the

structure is broken down with relatively large shear rate loop.

The Bingham model is used to determine (to) and (jxpi) at / o f 10 to 30 s"1 using the down-

curves of the first and last loops. During the rest period between the first and last dynamic tests,

the [xorest(t)] is determined. After successive rest periods of 10 min, the micro mortar is subjected

to an increasing stress from 0 to 300 Pa at a rate of 3.33 Pa/s until the structure breaks and the

measurement is cancelled. The criterion for this cancellation is when the y reaches 0.5 s"1, i.e.,

when the deformation due to the broken structure increases. The results of each are shown in Fig.

2.31. The area between the static and dynamic curves represents the reversible structure, i.e., total

structural build-up minus the irreversible structure over time.

it) mini, resl fuiiowed by a .sir-ess mcKune m 3.33 PH/S

f 0-300 Pa in *>0 »} Criteria: shear rate < 0.5 5>"!..

Repealed 4 times

H l«0

0 {£~

Time,

Fig. 2.30 Configuration of rheometer for tests on micro mortar [Billberg, 2006]

•5- -'*> 1

7. JOO •

i 550-

*• 200 •

1 150 -

1 1(H)-

I 50

(

\v

o - -

)

•C 0.36

10

v-5.uii-ru _ -->* K'-0.9»4 ^^*^

v «'i>7:5J"«.o»

20 JO 40

Time [miuut«s]

»

i

50

— 350-

7»oo-8

-. 200 •

| 150-

| 100

M 50

7. o-(

\V C 0.42

10

o -

flMZx

- o

-8.5 R!-9.«7«

20 JO

Time [inimil«s]

Staric yield stress

Dynamic yield sti

y-(Ufl*lt»

40

|

50

Fig. 2.31 Results of dynamic yield stress development and structural build-up of micro mortars

made with w/c ranging from 0.34 to 0.42 [Billberg, 2006]

45


2.3.5 Relationships between lateral pressure and theological properties

It is reasonable to presume that lateral pressure of SCC can be attributed to same physical

factors and mixture characteristics that apply to normal-consistency concrete. Rate of

restructuring of concrete should have a significant influence on changes in pressure distribution

with time. The restructuring phenomenon is considered to be mainly due to the development of

internal friction and attractive forces among the solid particles at rest and to the increase in

degree of physico-chemical bonds during cement hydration. Evolution of restructuring rate of can

be indirectly assessed by observing thixotropy, which describes the destructuring phenomenon.

Concrete exhibits faster degree of restructuring can develop greater cohesiveness soon after

casting, thus acting as a cohesive body exerting less pressure than its full hydrostatic pressure.

Khayat and Assaad [2005B] showed that lateral pressure exerted by SCC could be directly

related to magnitude of thixotropy. The greater the degree of thixotropy, the lower the initial

lateral pressure can be and the faster is the rate of pressure drop in time. This is attributed to the

stiffening effect, which enables the material to re-gain its shear strength when left at rest without

any shearing action [Khayat and Assaad, 2005B]. Changes in both the initial lateral pressure and

pressure determined after 100 and 200 minutes after concrete casting in 2.8 m high PVC column

of 200 mm in diameter are shown in Fig. 2.12 with the thixotropy of SCC. The thixotropy was

determined using the breakdown area (At,) approach evaluated at different time intervals (Section

2.3.2-B). As expected, the most thixotropic mixtures were shown to exhibit the lowest relative

lateral pressure (Ko) and the highest rate of pressure decay. For example, SCC with high Ab of

450 J/m3.s can develop approximately Ko of 60% near the bottom of the 2.8 m column after 200

min, whereas a less thixotropic mixture with low Ab of 180 J/m3.s can maintain a high Ko of 95%.

Billberg et al. [2006] showed that the increase in structural build-up at rest of SCC can

result in shaper rate of drop in formwork pressure that was determined during the first hour after

casting (Fig. 2.32). The formwork pressure was determined at a height of 50 mm from the bottom

of tube measuring 1.5 m in height and 200 mm in diameter. The structural build-up at rest

monitored at 10 min intervals was evaluated as per the procedure described in Section 2.3.2-D.

2.4 Parameters affecting formwork pressure and thixotropy

Since the early 1900s, numerous laboratory and field investigations were carried out to

provide broad understanding of the variables that can affect lateral pressure of fresh concrete.

ACI Committee 622 [1958] studied all published field and laboratory investigations of lateral

46

pressure developed on formwork. From these results and those reported latter in the literature

[Rodin 1952, Schojdt 1955, Adam et al., 1963, Gardner 1980, CIRIA 1985, Assaad 2004, etc.],

key factors influencing lateral pressure of concrete are summarized in Table 2.2. These factors

are discussed in below with special emphasis, whenever possible, on formwork pressure exerted

by SCC as well as thixotropy and structural build-up characteristics.

+

I 0.2

0.0

<G>

50 100 150 200

Structural build-up, rs(t) (Pa/min) 250

Fig. 2.32 Effect of structural build-up on pressure loss in 1st hr after casting [Billberg et al., 2006]

Table 2.2 Factors influencing formwork pressure

Material properties

- Cement type and content

- Use of supplementary cementitious materials and fillers

- Size, type, and content of coarse

aggregate - Water content

- Chemical admixtures - Unit weight of

concrete

Consistency level

-Slump for normal concrete or slump flow values for SCC

Placement conditions

- Placement rate - Height of concrete

- Placement method T T ' 1 i J?

- Height of pouring - Consolidation method - Vibration magnitude and

Hiirjitinn U U l d l l U l l

- Impact during placing - Ambient and concrete

temperatures

- Setting time - Time needed for form removal

Formwork characteristics

- Formwork dimension

- Case of reinforcement

- Formwork surface J. V X 1 1 1 f t V / l l V L7l>U A.14-VV

material - Permeability and

drainage of formwork

- Formwork surface roughness

- Demolding agent characteristics

2.4.1 Material properties

A. Composition and content of binder

Cement content, type, and use of supplementary cementitious materials and fillers have

significant influence on development of lateral pressure exerted by SCC. Assaad and Khayat

[2005A, 2005B] investigated formwork pressure exerted by SCC prepared with different binder

47


contents. The binder content varied between 400 and 550 kg/m3. A fixed content of 450 kg/m3

was employed for the mixtures made with T GU and T HE cements. The w/c remained constant

at 0.40 for all tested mixtures, and the sand-to-total aggregate ratio was set at 0.46, by mass. The

dosage of liquid-based viscosity-modifying admixture (VMA) was set at 260 mL/100 kg of

binder. The dosage rates of the high-range water reducing admixture (HRWRA) and air-

entraining agent (AEA) were adjusted to secure initial slump flow of 650 ± 1 5 mm and air

content of 6% ± 2%. An instrumented PVC column measuring 2.8 m in height and 0.2 m in

diameter was used to determine lateral pressure. The authors concluded that for a given binder

type, higher initial pressure can be obtained for mixtures containing greater binder contents; this

is due to the lower coarse aggregate volume. For longer elapsed times, lateral pressure variations

depend significantly on the development of cohesion resulting from the binder phase. Therefore,

the increase in binder content leads to sharper pressure decay, as illustrated in Fig. 2.33.

0 50 100 150 200 250 300 350

Time after casting (min) Fig. 2.33 Variations in relative pressure of SCC made with various contents of ternary binder

[Assaad and Khayat, 2005A]

Andreas et al. [2005] studied influence of binder type on SCC lateral pressure. Two

different types of ordinary Portland cements (CEM I 42.5 N according to European Standard EN

197-1) were used. SCC mixtures were also prepared with either FA or limestone filler (LF) at

33% volume replacement of cement. The SCC was cast from top of a formwork measuring

0.20x0.20x0.975 m. The partial substitution of ordinary Portland cement (OPC 2) with FA or LF

can accelerate the pressure decrease. The acceleration was more pronounced when using FA.

This is attributed to decrease in grain size leading to higher surface area and smaller inter-particle

distance. This causes greater build-up of internal structure and accelerated pressure decay.

48


B. Characteristics of coarse aggregate

Amziane and Baudeau [2000] reported that the use of discontinuously-graded aggregate

with MSA of 30 mm can lead to higher lateral pressure for conventional vibrated concrete than

continuously-graded aggregate with MSA of 7 mm. The authors considered that concrete is a

two-phase heterogeneous material composed of cement paste and coarse aggregate. The paste

possesses a rheological behavior that is exclusively viscous, whereas the granular phase

contributes to the resistance to shear stress through aggregate friction. Conventional concrete

mixtures made with w/c of 0.5, various aggregate contents, MSA, slump values of 50 to 250 mm

were evaluated. The lateral pressure was determined using steel formwork measuring 1650 mm

in height, 1350 mm in length, and 200 mm in width. Full hydrostatic pressure was obtained in the

case of cement paste. The maximum lateral pressure divided by the pressure obtained for the

cement paste mixture (Pm/Pp) was shown to decrease with the increase in coarse aggregate

volume until the volumetric ratio of the paste-to-coarse aggregate (Vp/Vagg) approached one. As

indicated in Fig. 2.34, the variation in Pm/Pp with Vp/Vagg is linear between A and B that

correspond to mixtures having aggregate concentrations lower than 40%. Only a slight reduction

in the Pm/PP value was obtained despite considerably reduction in the Vp/Vagg value. The authors

suggested that as long as the volume of mortar is dominant, the magnitude of internal friction

remains limited resulting in considerably higher lateral pressure. The second slope (BC)

corresponds to mixtures with aggregate concentration greater than 40%, and indicates that a

significant decrease in the relative pressure can be obtained for a small decrease in the Vp/Vagg

value. This suggested that the concrete tends to have a granular behavior enabling the

development of shear strength mainly through aggregate inter-particle friction, hence leading to

significant reduction in lateral pressure.

Fig. 2.34 Variations of Pm/Pp values vs. coarse aggregate content [Amziane and Baudeau, 2000]

49

In case of highly flowable mixtures with slump flow of 650 ± 15 mm, Assaad and Khayat

[2005C] found that the increase in coarse aggregate volume could reduce the lateral pressure and

increase the rate of pressure drop after casting. This was attributed to the increases of the degree

of internal friction resulting from greater coarse aggregate content, which reduces the mobility of

the concrete and the resulting lateral pressure. As illustrated in Fig. 2.35, the initial lateral

pressure can decrease from 99% to 77% of hydrostatic pressure when the sand-to-total aggregate

ratio (S/A) decreases from 1.0 to 0.30, respectively. The lateral pressure was determined near the

bottom of a pressure column measuring 2.8 m in height in which concrete is cast at 10 m/hr. The

rate of drop in pressure with time was also found to be influenced by the S/A value. For example,

for the 0.75-SCC and 0.30-SCC mixtures, the time required to reduce lateral pressure by 10% of

hydrostatic value was 225 and 80 min, respectively (Fig. 2.35). In addition to lateral pressure, the

decrease in S/A resulted in greater thixotropy, as illustrated in Fig. 2.36. For example, the

breakdown area (At,) is shown to increase from 130 to 340 J/m3.s and then to 550 J/m3.s during

the first series of measurements (Tl time period) when the S/A value decreased from 1.0 to 0.46

and then to 0.30, respectively.

The MSA was also found to affect formwork pressure. The initial relative pressure was

shown to decrease from 92% to 85%, and pressure decay was more pronounced when using

coarse aggregates with 14 mm MSA compared to 10 mm MSA. Further increase of MSA to

20 mm had limited effect on pressure drop compared to the SCC with 14 mm MSA [Assaad and

Khayat, 2005C].

* - » ^ , . t < 4 ^ 4 L0.SOC

2 09 L "D"°-- _. ~ - - . - _ . s u a 1v »*. " • • • • • • . . _

^ ^ ^ w • - ,r- "•'!™v'

I OH

0.7

0.6

0.5

0.75-SOC aump = 85

<^4A. " " " " • • • • I . . ,£• as 4. **«*>*„ A^4s

<x**>c, " & ^ . 050-10-SCC

" f/""".. *"*^Z?^X ./— °-46-soc

- < , _ ' . . . K x ' ! . ^ K " - ^ . ^ 3ump=140mnn

"y^u ""'•>.. ^ ^ ^ 04OSCC 030-SOC / " t" f-f< . 036-SCC Slimp=110mm Surrp = 220tmi **4- t +^ SLmp =160mm

0 100 200 300 400

Time after casting (min) Fig. 2.35 Variations of relative pressure with elapsed time after casting for mixtures made with

10 mm MSA [Assaad and Khayat, 2005C]

50

_ 600 -<*>

S SOD

S '

V 400 -&•

M

| 300-O

•a •i£ 200 u

« 100

1.0- 0.75- 0.50-10- 0.46- 0.40- 0.36- 0.30-

scc sec sec sec sec sec sec

Fig. 2.36 Variations of breakdown area for mixtures made with various coarse aggregate

concentrations (MSA =10 mm) [Assaad and Khayat, 2005C]

C. Water content and w/cm

Khayat and Assaad [2006] reported that changes in w/cm have an effect on lateral pressure

and thixotropy. For a given slump flow of 550 mm, SCC proportioned with 0.46 w/cm exhibited

lower thixotropy and slightly greater initial pressure compared to SCC made with 0.40 or 0.36

w/cm (Fig. 2.37). This was related to the increased water and paste contents and reduction in

coarse aggregate volume, which lead to lower shear strength properties of the plastic concrete.

Furthermore, the rate of drop in lateral pressure and gain in thixotropy with time were found to be

considerably greater in SCC made with w/cm of 0.46 (Fig. 2.37). The elapsed time to reduce the

relative pressure by 25% decreased from 200 to 150 min with the decrease in w/cm from 0.40 to

0.36 for SCC made with PNS-based HRWRA. This is attributed to the lower HRWRA demand

of the SCC made with the higher w/cm, which can present less interference with the rate of

structural build-up and development of cohesiveness than SCC with lower w/cm [Khayat and

Assaad, 2006].

2.4.2 Consistency level

Most investigators reported that lateral pressure is expected to increase for mixtures having

higher consistency level. The greater lateral pressure is due to the reduced degree of shear

strength resistance that causes the material to behave like a fluid.

Gardner [1980] found that for concrete with 170 mm slump cast at 6 m/hr, the lateral

pressure is equal, in some cases, or typically exceeds that of a concrete with 50 mm slump cast at

much higher rate of 46 m/hr. A 100-mm slump concrete cast at 46 m/hr was found to develop

35% higher lateral pressure compared to a mixture with 50 mm slump, also cast at 6 m/hr.

51

• T1 = 0 to 30 min

• T2 = 60 to 90 min

• T3 = 120 to 150 min

1.0 i

'« 0.9

I £ 0 . 8

| 0.7 i

Le 0< 0.5

\ * * • •

* * A * * * • • • S A , * • • • « » 0.36-PNS

^S^. * • • • * « , Slump = 180 mm

^ . A 4 & A

° i A 4 4 0.40-PNS Slump = 150 mm A Slump =160 mm

0.46-PNS — ' °B i i ' " A A i

0 50 100 150 200 250 300

Time after casting (min)

Fig. 2.37 Effect of w/cm on relative pressure variations of SCC made with PNS-based HRWRA

[Khayat and Assaad, 2006]

Assaad and Khayat [2006] investigated SCC mixtures prepared with 450 kg/m3 of binder

and w/cm of 0.40. The mixtures had slump flow values of 550, 650, and 750 ± 1 5 mm, and

incorporated VMA concentrations corresponding to 200, 260, and 350 mL/100 kg of binder,

respectively. The sand-to-total aggregate ratio was held constant at 0.46 for all mixtures, and the

air-entraining agent dosage was adjusted to ensure a fresh air content of 6 ± 2%. Variations of

lateral pressure with time determined near the bottom of a PVC column of 2.8 m in height and

200 mm in diameter are shown in Fig. 2.38 for concrete cast at 10 m/hr. The results clearly

indicate that the development of lateral pressure of highly flowable concrete (FC) and SCC are

significantly affected by the initial consistency level. For a given mixture proportion, concrete

with higher dosage of HRWRA and consistency level was found to exert higher initial pressure

and lower rate of pressure drop with time. For example, depending on the mixture consistency,

the initial pressure can vary from 75% to 98% of hydrostatic and the time to reduce the pressure

by 10% can range from 45 to 167 minutes.

Ferron et al, [2006] showed that the degree of thixotropy is strongly related to the initial

fluidity of the paste. As the fluidity of the system increases, thixotropy decreases due to a

reduction in the coagulation forces and inter-particle collisions.

2.4.3 Placement conditions

A. Placement rate

Several studies established that the casting rate could have marked effect on formwork

pressure exerted by SCC (Vanhove et al., [2001], Khayat et al., [2002B], Leemann and Hoffmann

52


[2003], Assaad [2004], Fedroff et al., [2004], Beitzel et al., [2004], and Tejeda-Dominguez et al.,

[2005]). When the pouring rate is so fast that no stiffening is allowed, (such as in small volume

pours that can be completed in one single lift), SCC formwork pressure could well reach

hydrostatic. However, when formwork pressure was measured in larger structures where the

pouring rate was indeed slower, the maximum pressure was considerably less than hydrostatic.

SCC-T10-650 Slump = 180 mm

SCC-TER-750 Slump = 230 mm

100 150 200 250

Time after casting (min) 350

Fig. 2.38 Effect of mixture consistency on relative pressure (consistency values are noted at the

end of each test) [Assaad and Khayat, 2006]

Billberg [2003] evaluated the formwork pressure exerted by SCC cast at relatively low

placement rates of approximately 1 to 2.5 m/hr. Two different mix designs were employed SCC1

and SCC2 with w/c of 0.40 and 0.45, respectively, in addition to a conventional concrete (CC).

The slump flow at the time of casting was 730 ± 50, 700 ± 50 for the SCC1 and SCC2,

respectively. The concrete was dropped from 1 ± 0.5 m height over the concrete surface in a

formwork measuring 3-m tall. The correlation between casting rate and form pressure was found

linear for SCC (Fig. 2.39).

W"%

If"' 11,,

OsMifi^No. 5 rut

Y CUSX-t t .H

Oa;*Un£ No. $ w.

Casting rate (m/hr)

Fig. 2.39 Variation of relative form pressure with casting rate for SCC [Billberg, 2003]

53


For SCC placed at relatively moderate-to-high casting rates, Assaad and Khayat [2006]

evaluated the effect of casting rate of SCC using a PVC column measuring 2.8 m in height and

200 mm in diameter. As noted in Fig. 2.40, the decrease in casting rate from 25 to 5 m/hr can

reduce the relative initial pressure by 15%; however, no significant effect was noted on the rate of

pressure drop with time. The interruption of casting for 10 or 20 min between subsequent lifts at

the middle of the placement was reported to lead to considerable reduction in formwork pressure

despite the fact that the casting rate was maintained at 10 m/hr when placement was occurring

[Assaad and Khayat, 2006].

1.0-

1 0.8 -1 1_

•§, .c

^ 0.6 -s

J E

eT 0.4 -

0.2 0 50 100 150 200 250 300

Time following the beginning of casting (min)

Fig. 2.40 Effect of casting rate on relative pressure of SCC [Assaad and Khayat, 2006]

B. Placement method

Wolfgang and Stephan [2003] conducted an investigation on a wall model with h x w x d

of 3.30 x 3.51 x 0.24 m, respectively. Four walls were cast using SCC: two cast with buckets

from top and two being pumped from the bottom at placement rates of 2 and 10 m/hr. The fifth

wall was cast using conventional vibrated concrete (VC) placed from top with buckets with a

placement rate of 7.5 m/hr. As noted in Fig. 2.41, much lower lateral pressure distribution was

obtained with the SCC cast from the top with a bucket. Hydrostatic pressure was obtained when

placement was carried out from the bottom with concrete pump. The resulting lateral pressure

was approximately twice that of SCC cast from the top at the same casting rate. Slowing down

the casting rate at 2 m/hr in the case of bottom injection was found to require higher pumping

pressure.

Slump flow "= 640 mm

TER-20 mixture

Slump values are around 140 mm

Without stoppage, at 5 m/h Without stoppage, at 10 m/h

-Without stoppage, at 25 m/h - Two resting periods of 10 min each

Two resting periods of 20 min each

54


0 0.50 1.00 1.50 2.00 2.50 3.00 3.50 Filling level (m)

Fig. 2.41 Force in the lower anchor versus filling level [Wolfgang and Stephan, 2003]

C. Ambient and concrete temperature

Gardner [1984] evaluated the effect of concrete temperature on lateral pressure on vibrated

mixtures with slump values ranging between 65 and 115 mm. Concrete temperatures were

varying from 2 to 27 °C. The lateral pressure was found to increase with the decrease in concrete

temperature. The author found that for lower temperatures, the hydration of cement can be

slowed down, and mechanical properties can develop at a slower rate resulting in higher lateral

pressure. The author reported that lateral pressure is rather controlled by the concrete

temperature, and not by the ambient temperature.

The effect of SCC temperature on lateral pressure variations was evaluated by Assaad and

Khayat [2006]. The mixtures had a ternary binder content of 450 kg/m3, w/cm of 0.40, slump

flow of 650 ±15 mm, and air content of 6 ± 2%. The mixtures were prepared at 10, 20, and 30 ±

2 °C; are referred to as TER-10, TER-20, and TER-30 mixtures, respectively, in Fig. 2.42. The

variations of relative pressure of these mixtures cast at 10 m/hr in a PVC column measuring 2.8

m in height and 200 mm in diameter are plotted in Fig. 2.42. The mixtures prepared at initial

temperatures of 10, 20, and 30 °C develop similar relative pressures of 91% at the end of casting.

On the other hand, the rate of pressure drop with time was significantly affected by the concrete

temperature. For example, the elapsed time to reduce the relative pressure by 25% decreased

from 400 to 250 and 160 min for the TER-10, TER-20, and TER-30 mixtures, respectively.

D. Relationship of pressure cancellation time and setting time of concrete

The initial and final setting times of concrete are defined as the onset of solidification of

fresh concrete mixture, and the transition from a fluid to rigid state, respectively, [Mehta and

55

Monteiro, 2006]. According to ASTM C-403, the initial set represents approximately the time at

which concrete can no longer be properly mixed, placed, and compacted. In addition, the final set

represents approximately the time of start of mechanical strength gain. The setting rate of

concrete is affected by the characteristics of the concrete mixture (cement composition, w/c,

admixture, and temperature) as well as the prevailing conditions at the project site, including

ambient temperature, humidity, and wind.

1.0

^ 5 4 4 ,

1 0.8 -IT **xt>*. •a ••xj

6^0.7

I °-6

0.5

0.4

Casting rate= 10 m/h

TER-10 Slump = 170 mm

<?**, ^ ^

' * * •

**» <*X*

'%

'AAA. 'AAA

<** He ' ^ * A A A

*x* AAA,

«Xx_

* * * * * * x x * * *x ,

'AA&

TER-20-ACC Jf* Slump= 125 mm

T30-20 Slump = 105 mm

\

TER-20 Slump= 140 mm

A&AA

"THR-30

Slump = 180 mm

50 100 150 200

Time after casting (min) 250 300

Fig. 2.42 Effect of SCC temperature on variation of relative pressure [Assaad and Khayat, 2006]

The times corresponding to initial and final settings of concrete are typically determined

using mortars extracted from concrete in compliance with ASTM C 403. These setting times are

considered as suitable references to indicate the time at which the concrete can no longer be

properly handled or placed. Even though they are based on purely arbitrary measurements, their

evaluation can be important to estimate beginning of the stiffening phase [Jiang and Roy, 1991].

Khayat and Assaad [2005A] attempted to relate the standard ASTM C 403 setting times to

the time required to cancel lateral pressure exerted by SCC. Relationships between the times

corresponding to canceling of pressure (tc) and the initial and final setting times of mortars

extracted from numerous SCC mixtures prepared with various material characteristics and

mixture proportions are plotted in Fig. 2.43. As expected, mixtures exhibiting longer setting

times necessitated longer periods after concrete placement prior to lateral pressure cancellation.

Pressure cancellation was determined using a pressure sensor attached near the bottom of a 1.1 -m

high PVC column measuring 200 mm in diameter.

56


1500 -

1200 "

s "t 900-S

M .5 600 *

300

0

0 300 600 900 1200 1500

Time for pressure cancellation (min)

Fig. 2.43 Relationship between initial and final setting times and elapsed time for lateral pressure

cancellation [Khayat and Assaad, 2005A]

2.4.4 Formwork characteristics

A. Formwork dimension

Limited data exist regarding the effect of size and shape of the formwork on lateral pressure

characteristics. Rodin [1952] reported that the general tendency indicates that the maximum

pressure appears to be lower in formwork systems of smaller cross-sections. This can be

attributed to the increased degree of the arching effect, which limits lateral pressure. Gardner

[1980] demonstrated that the larger the dimension of the formwork, the larger the lateral pressure

could be for conventional vibrated concrete.

Khayat et al. [2005A] studied the effect of column diameter on changes in lateral pressure.

Two experimental formwork systems were used. A PVC tube of 2.1-m high and 200-mm

diameter was used for first system. The second column consisted of a sonotube measuring 3.6-m

high and 920-mm diameter. The sonotube had an impermeable plastic liner and was adequately

braced and reinforced. Lateral pressures were determined using pressure sensors located at

various locations along the heights of the experimental columns. Both columns were cast at the

same casting rate of 10 m/hr. Fig. 2.44 illustrates the variations of relative lateral pressure

determined at approximately 2 m from the top of the formwork systems. Initially, the mixture

placed in the larger column exhibited slightly greater relative pressure of 99% compared to 96%

for the 200-mm diameter column. This can be due to an arching effect in the relatively restricted

section. However, the rates of drop in pressure were significantly different. In the case of the

57

Pressure cancellation = 1.01 x final set

Pressure cancellation = 1.16 x initi

• Initial set; R2 = 0.93

° Final set; R2 = 0.95


concrete placed in the 920-mm diameter column, the time required to reduce lateral pressure by

5% of the hydrostatic value was 20 min, resulting in a rate of decay of 5.3 kPa/hr. Conversely, for

the 200-mm diameter column, this period was 38 min, or an initial rate of decay of 3.3 kPa/hr.

1.0

0.9

-0.8

5-0.7

0.6

Casting rate = 10 m/h

Comparisonevaluatedat 2050 mm from the top

• Column of 920-mm diameter • Column of 200-mm diameter

0 20 40 60 80 100 120 140 160 180


Fig. 2.44 Effect of section width on lateral pressure [Khayat et al., 2005A]

B. Type of formwork surface material

Tejeda-Dominguez and Lange [2005] evaluated the effect of formwork material on SCC

lateral pressure. Two different sonotube configurations were used (one without any modification

[ST-1] and the second with an impermeable plastic liner to cover the interior wall [ST-2]). Also,

two different formwork configurations of PVC columns were used (one without any modification

or reinforcement [PVC-1] and the second cut along one side from top to bottom and reinforced

with steel straps to keep it tightly closed [PVC-2]). All the columns have a diameter of 0.25 m

and a total height of 3 m. For each column, two sensors were placed at 0.15 m from the bottom.

All columns were filled from the top at an approximate rate of 27 m/hr. The concrete was not

vibrated or compacted by any means. The four tests showed that the lateral pressure

characteristics of SCC were close to hydrostatic immediately after casting. However, as shown in

Fig. 2.45, the decrease in pressure after casting was dependant on the forming material. The study

illustrates the importance of using rigid, self-supporting column apparatus to monitoring lateral

pressure variations, especially when pressure sensors are attached to the formwork material. If the

column material is not rigid enough, as was the case of the plain sonotube that can imbibe water,

the rate of lateral pressure drop would seem to be sharp given the swelling of the column material

and separation of the pressure sensor that was initially set flush with the plastic concrete.

58


Fig. 2.45 Variation of SCC pressure cast in formworks of different materials [Tejeda-Dominguez

and Lange, 2005]

2.5 Lateral pressure measuring systems

2.5.1 Instruments and devices to monitor lateral pressure

A number of methods have been adopted to measure lateral pressure exerted by plastic

concrete on formwork. Roby [1935] measured the concrete pressure by deflection of a steel plate

extending to the full width of the form and resting on movable edges 700 mm apart. The steel

plate located near the bottom of the formwork measured 150 mm in width and 10 mm in

thickness. By means of a pivoting bar connected to a lever to give a ratio of 10:1, the deflection at

the center of the plate was determined on a scale graduated in increments of 0.4 mm. By using a

magnifying glass, it was possible to read within 0.1 mm, which corresponds to a plate deflection

of 0.01 mm. The thickness of the timber sheathing above and below the steel plate was designed

to give the same deflection under load as the steel plate. For a maximum pressure of 7,000 kPa,

the deflection at the center of the plate was 6 mm. Such deflection can be considered quite

appreciable and might have the effect of relieving the steel plate of some pressure.

Stanton [1937] used pressure sensors consist of a metal disk. A sheet-rubber diaphragm

was clamped to one side of the sensors in a manner similar to that of a drumhead. The shallow

space between the rubber diaphragm and the disk was filled with liquid that would operate as

ordinary pressure gage mounted on the back of the disk. The pressure sensors of 150 or 300 mm

in diameter were inserted into the form wall such that the rubber diaphragm was flush to inner

surface of the wall. All air from the pressure sensors was extracted. No indication was given to

the volume change undergone by the sensors when the concrete pressure was applied.

59


Macklin [1946] determined the pressure of concrete against the formwork by measuring the

deflection of the wood sheathing. The deflection of the sheathing relative to the supporting studs

was measured by means of a dial-type micrometer mounted on a bridge arrangement.

Gardner and Ho [1979] employed Cambridge-type load cells to determine the form

pressure. The total load capacity of the cell was around 2500 N. The vertical and horizontal

measuring strips were 65 mm in width and 8 mm in thickness. The two gauges on each

measuring strip were wired in series of eight pairs of two pairs per load cell.

Khayat and co-workers [Assaad et al., 2003B] selected pressure sensors from Honeywell

(Model AB MP) to monitor formwork pressure. The strain-gage based pressure sensors shown in

Fig. 2.46 are known as flush diaphragm mill-volt output type pressure transducers. Typically, the

diameter of the sensor used by the authors for SCC was 19 mm, though greater diameters can be

used when the concrete has large nominal size aggregate. The sensors are calibrated against an

analog pressure gage using an oil pump and are also calibrated using a given water head. During

the experiments, the sensor is connected to a data acquisition system with a scanning voltage of 5

mV. The pressure sensors are set flush with the inner side of the formwork through drilled holes.

A thin film of grease is applied to the sensors to protect them from the concrete.

Fig. 2.46 Honeywell pressure sensor of 19 mm diameter [Assaad et al., 2003B]

Andreas and Cathleen [2003] used sensor (Fig. 2.47) located on the inner surface of the

formwork to measure the concrete pressure. The sensor used was manufactured by Baumer

Electric AG and were built to record a maximum pressure of 1.6 bars. The area of the sensor

exposed to pressure is 7.5 cm2. The principle of the sensor is based on a change in electrical

resistance of thin-film metal wire strain gages when they are deformed due to pressure.

Billberg [2003] measured lateral pressure through the determination of stresses exerted on

formwork tie rods. This method requires that the base of the formwork can move freely on its

60


foundation in order to prevent friction leading to unreliable results. Such base friction could be

minimized using rollers [Brameshuber and Uebachs, 2003].

Fig. 2.47 Pressure sensor used by Andreas and Cathleen [2003]

Andriamanantsilavo and Amziane [2004] measured the lateral pressure using a diaphragm

pressure transducers mounted flush with the forming material. Under the pressure of the fluid or

gas materials, the membrane deforms and produces a variation in both the sensor wire resistance

and the output voltage. When a material is setting, the deformation of the transducer diaphragm

due to flow of freshly mixed paste is not reversible, although no pressure is being applied by the

transducer [Fig. 2.48(a)]. The transducer operating principle is based on the implementation of a

controlled air backpressure, which is continuously balanced with the pressure exerted by the

tested medium [Fig. 2.48(b)]. The device is composed of two interconnected measuring chambers

[Fig. 2.48(c)]. The first chamber is equipped with an absolute pressure transducer connected to a

compressed air control valve. The second chamber is equipped with an inductive standard

displacement transducer attached to a thin elastometric latex membrane. The other side of the

membrane is in direct contact with the material tested. During the test, the pressure in both

chambers is controlled so that the membrane is kept in a vertical position. Indeed, this position is

indicative of the pressure equilibrium on both sides of the membrane. Consequently, the pressure

exerted by the material on the formwork is equal to the pressure measured in the chambers.

Arslan et al., [2005] measured the formwork lateral pressure exerted by fresh concrete

using two strain gage plates (Fig. 2.49). Full bridge (Wheatstone bridge) with 10-mm long gages,

-10% transverse sensitivity, and 120 ± 03 Q resistance, were set up on every strain gage plate, as

seen in Fig. 2.49. Strain gage plates were calibrated by applying known forces. For each strain

gage plate, a regression formula between the applied forces and corresponding strain values was

developed. Strain gage plates were then mounted at each bottom side of formwork. The foil

bridge circuits are connected to a computer-based data logger via a switching box to monitor

form pressure variations with time.

61

Compisswd air

<a) (he M»»esii*Bi problem of t k SoSal pressure

<s«ttil b>- a « in»s ptm during its setting CfllKBI pSSlC

Canesil paste

(b l she opemihm of Slit* ikvicc ss taui! on rtw

vertical position control of the nwmbraise.

Pft&UK transducer

CcroeM paste

(C)) The lirsi cd! is provided *ifh an riwalum

(iressuB iraisAjccr wrcscclai to a itoaribul«r of

totnpreucd air

ftfswire iKaataw

(e«) The $ect>nd celt is equipped with as

irsteln-f tlSKbril feplsimni tasduetr

irsserdepsrsffan (» a !s(« inembraw

Fig. 2.48 Design of lateral pressure device [Andriamanantsilavo and Amziane, 2004]

V« ( i i i j

Plate &f strain gauge I Ps) Vc (in)

Full bridge circuit

Fig. 2.49 Strain gage plate and strain measurement system [Arslan et al., 2005]

2.5.2 Pore water pressure measurements to determine lateral pressure

Soil mechanics principles consider that lateral pressure exerted by soil-like material to be

the sum of pore water pressure and pressure exerted by the submerged solid skeleton:

o =& + [ U x ( A - A c ) / A ] Eq.2.74

where a is the total lateral pressure, a' is the effective pressure resulting from the solid particles,

U is the pore water pressure, and Ac and A denote the area of contact points on a given plane and

the total area of the plane, respectively. The above equation, developed initially by Terzaghi and

Peck [1967], assumes that the value of Ac can be neglected compared to that of A, thus leading

to: a = a' + U. Furthermore, the lateral pressure exerted by a dry granular material on a frictional

surface is proportional to the vertical pressure, as follows:

62

Chapter 2: Review on formwork pressure and fundamentals of r neology

Ph = K x Pv Eq. 2.75

where Ph is the lateral pressure and Pv = y g h is the vertical pressure (y, g, and h being the unit

weight, gravity, and head of granular material, respectively). K value corresponds to the lateral

earth pressure coefficient, which depends on internal friction of the material and on whether the

lateral pressure is active (Ka), passive (Kb), or at-rest (Ko).

During tests carried out by Alexanridis and Gardner [1981], it was considered that the

undrained cohesion and coefficient of internal friction values are both pore-water-pressure-

dependent for concrete. This assumption was based on the fact that pore pressure developed

under field conditions can differ significantly from those in the laboratory due to different

drainage conditions. Therefore, it is difficult to appreciate the applicability of any results obtained

without taking into account pore water pressure. The authors reported that at very early age and

low vertical stress, the fresh concrete behaves as a fluid with vertical stresses transformed into

lateral stresses, and the at-rest Ko coefficient for the solid phase approaches unity. However, as

the fresh concrete starts to gain shear strength, Ko decreases rapidly and approaches that of

Poisson's ratio for cured concrete. The authors noted that this result is in disagreement with

conventional soil mechanics unless the pore fluid has a density close to that of fresh concrete.

Nonetheless, the density of the fluid phase corresponds to that of concrete during vibration, and

eventually decreases to that of the density of water.

Radocea [1994] studied the evolution of pore water pressure with time until setting of

cement paste (Fig. 2.50). The pore water pressure is shown to decrease from PI to P2 due to

settlement of the cement grains after placing [Radocea, 1994]. In the same period, bleeding can

also occur at the surface. The initial pressure PI depends on the paste density and the

measurement depth. The surface is covered with bleeding water during the period from tl to t2

and the pore water pressure will remain stable. At t2, the surface starts to dry out because of free

evaporation, and the pore water pressure will decrease. This is due to the formation of meniscus

at the surface and the hydration of the cement. The effect of cement hydration can be seen at t3

where the pore pressure is decreasing in a sealed sample. If the specimens are water cured, the

rate of pore water pressure decrease will be reduced because the water will be transported into the

specimens due to suction caused be the lower pore water pressure [Radocea, 1994].

Assaad and Khayat [2004] compared the lateral pressure to measurements of pore water

pressure exerted by SCC using an experimental column measuring 1 m in height and 200 mm in

63


internal diameter. The lateral pressure was measured using three pressure sensors mounted at 50,

150, and 350 mm from the base. At the same heights, pore water pressure sensors were used to

determine the pressure resulting from the fluid phase. Variations of lateral pressures and pore

water with time measured for the sensors located at 50 and 150 mm from the base are plotted in

Fig. 2.51 for SCC made with 0.46 sand-to-total aggregate ratio. Fig. 2.52 shows the variations of

both lateral pressure and pore water at 50 mm from the base of the experimental column as well

as the temperature variation measured at the center of the 200-mm diameter column for SCC

made with 10 mm MSA and 0.50 sand-to-total aggregate ratio. As shown here, a depression of

approximately -10 kPa was measured from the pore water pressure sensors; this corresponds to

the limit of the pore water pressure sensor. Such sensors require continuously water saturation to

function. As the hardening process takes place, concrete can cause some suction of the water in

the sensor, hence interrupting any further measurements.

Pore water pressure

Forces acting on the structure: (cavitation Capillary forces .

(hydration)

Atmosplwiic pressure

Legend: P I —initial pore pressure P 2 « hydrostatic pressure t l — total time of bleeding t2 = time when water menisci

develop at the top surface t 3 = time when water is

absorbed from the surface

curing

Fig. 2.50 Variations of pore water pressure in cement paste with time [Radocea, 1994]

Fossa [2001] suggested that the mechanism of pore water pressure drop is governed by

chemical shrinkage caused by cement hydration that starts as soon as the cement begins to react

with water. However, it is after the end of the plastic stage that progressive formation of

hydration products can cause the creation of a network of connections and development of empty

capillary pores. The largest capillary pores will begin gradually to dry, and gel pores formed

during the hydration will start to drain water from the coarsest capillary pores, as free water is

held by forces that are inversely proportional to the apparent diameter of the capillary pore (self-

64


desiccation process) [Aiitcin, 1999]. The consequence of such process is the formation of

meniscus at the water/vapor interface, resulting in a decrease in relative humidity and drop in

pore water pressure towards negative values [Fossa, 2001].

Lateral pressure at 50 m m

Lateral pressure at 150 mm


Fig. 2.51 Variations of pore water and lateral pressures with respect to time for the 0.46-SCC

mixture [Assaad and Khayat, 2004]

Time after casting (hour)

Fig. 2.52 Variations of pore water and lateral pressures and concrete temperature with time for

the 0.50-10-SCC mixture [Assaad and Khayat, 2004]

Andriamanansilav and Amziane [2004] tried also to relate the kinetics of variations in

lateral pressure to that of pore water pressure and stiffening of cement paste. An experimental

65


setup made of a tubular glass column measuring 1.1 m in height and 110 mm in diameter was

used (Fig. 2.53). The column is connected to two pressure measuring devices positioned at a

height of 50 mm from the base. In order to simulate the equivalent hydrostatic pressure of fresh

cement paste at heights of 5 and 10 m, an equivalent pressure is applied by an air actuator on the

surface of the material inside the column. In addition to the total lateral pressure, pore water

pressure, temperature, and setting of cement paste using the Vicat test were determined.

Cement pastes with w/c of 0.30, 0.36, and 0.45 were used. The paste was cast into the

tubular column in two layers, each vibrated for 15 sec. The results presented in Fig. 2.54 describe

the evolution of pore water pressure and total lateral pressure. The pore water pressure kinetics of

the fresh cement pastes with w/c equal to 0.30, 0.36, and 0.45 are shown in Fig. 2.55. The authors

concluded that the time of lateral pressure cancellation was delayed with the increase in w/c. For

a given cement paste, the profiles of the pore water and the total lateral pressures remained

hydrostatic from the initial state until pressure cancellation. During this period, the kinetics of

evolution of both pressure measurements was identical. Once the total lateral pressure was nil,

the pore water pressure passes to a depression state. The w/c, depth of casting, and vibration

frequency period were shown to have considerable impact on the kinetics of the lateral and pore

water pressures.

A-Air actuator

«* -B-Tubular column

C-Pore water

^ pressure transducer

_D-Total lateral pressure transducer

Fig. 2.53 View of the set-up device [Andriamanansilav and Amziane, 2004]

66

2S» •

200

^?^m^*^.^®&&&^&KW®mM%'^*.®^:?v,-®JZ:%:®:<?<%.f<

4 6 8

Time (hours) 10 12

Fig. 2.54 Diagram of the evolution of pore water pressure and total lateral pressure

[Andriamanansilav and Amziane, 2004]

gj ISM-u

« 15(1

I, 3

^ j ^ e H a t e f pr*¥*»fe<M 1ttt

fKMr? water ijsmMire at 10 m

; (a) 1

7 I

Time (hours)

— f^-fK wafer prtsw** m I «s

pore wafer pressure ai 5 m

purv wafer pressure *tl 10 m

to

Time (hours)

— pare water pressure at I w

<• pare water pressure at § m

pyre water pressure at t (t m

. 1 4 5

Time (hours)

Fig. 2.55 Kinetics of pore water pressure of

fresh cement paste (a) w/c = 0.30, (b) w/c =

0.36, and (c) w/c = 0.45 [Andriamanansilav and

Amziane, 2004]

2.6 Case studies for formwork pressure exerted by SCC

The French research center on buildings and infrastructures (CEBTP) reported in 1999 on

the results of an experimental study to determine lateral pressure exerted by SCC on high

experimental wall measuring 12.5 m in height [Fig. 2.56(a)]. The pressure of the concrete was

measured using seven sensors placed at various heights. The concrete contained 350 kg/m of

cement and 100 kg/m3 of limestone filler, a water-to-powder ratio (w/p) of 0.46, and HRWRA.

The slump flow at the time of casting was 700 mm. The SCC was cast from the top using a

bucket at a rate of 18 m/hr. This rate was 25 m/hr for the second wall where the concrete was

67


pumped from the bottom. The resulting lateral pressure envelopes are shown in [Fig. 2.56(b)] and

indicate that the pressure distribution is not linear and is lower than hydrostatic. A net deviation

from the theoretical hydrostatic pressure was observed beyond 2.5 m from the top of the wall. At

the base of the formwork, the deviation from the hydrostatic distribution was 30% in the case of

SCC pumped from the bottom and 35% for concrete cast using a bucket from the top. The

difference between hydrostatic and measured pressures was attributed to the reduction in

hydraulic head due to friction between the rising concrete during placement and the surface of the

steel formwork [Vie et al., 2001].

(a) (b)

Fig. 2.56 Pressure sensors and formwork (a) and pressure envelope (b) [CEBTP, 1999]

Similar studies were carried out in Sweden [Skarendahl, 1999] during casting of bridge

piers measuring 5 m in height. The pier was cast in successive layers of approximately 0.5 m in

height with rest periods that were unspecified, resulting in low placement rate. The lateral

pressure exerted by SCC at the base of the formwork was approximately half of that resulting

from normal-consistency concrete consolidated by internal vibration.

In 2000, three industrials including the GTM Construction (France), NCC AB (Sweden),

and Sika Admixtures (Spain) conducted a large field-experimental program dealing with the

68


surface quality and lateral pressure of SCC. The report was published in 2000 as part of the Brite-

EuRam Project entitled "Rational Production and Improved Working Environment through Using

Self-Compacting Concrete" [Tejeda-Dominguez et al., 2005]. Special emphasis was placed on

the effect of wall geometry and casting rate on lateral pressure exerted by SCC.

The GTM Construction tested two types of SCC prepared with and without VMA that are

intended for cast-in place civil engineering construction. The cement (CPA CEM I 52.5 R) and

filler (Limestone Picketty Type A) contents were fixed at 290 and 174 kg/m3, respectively. The

sand-to-total aggregate ratio was 0.46, by mass. A fixed dosage of 9 kg/m3 of HRWRA

(Sikament 10) was used. The water content was then adjusted to secure slump flow value

between 700 to 880 mm. Different formwork dimensions were used: the length was set to 1.25

and 2.5 m, the height to 2.8 and 5.6 m, and the width to 0.25 and 0.40 m. A mesh of reinforcing

bars was placed in the formwork corresponding to 50 or 80 kg/m3 for walls having a width of

0.40 or 0.25 m, respectively. A large range of concrete casting rates varying between 10 and 150

m/hr was tested during the trials. Different concrete placement methods were studied, which

included pumping the concrete from the bottom of the formwork and placement by pump or

bucket from the top. Lateral pressure measurements were made using two different systems. The

principle of the first system (GTM pressure equipment) consisted of casting water into a system

made of stiff pipes and rubber bags and measuring SCC pressure directly with the manometer.

The second system consisted of electronic pressure sensors of 200 mm in diameter that enables

instantaneous lateral pressure measurements. The experimental data indicated that both systems

led to equivalent results. For most of the tested SCC mixtures, the measured lateral pressure was

close to the hydrostatic pressure. This can be due to the very high casting rates and high

deformability of the concrete. In general, it was reported that for a given slump flow and casting

rate, mixtures containing VMA exhibited higher initial pressure and lower rate of pressure drop

with time; this may well be because of the greater water content of these mixtures since the

HRWRA dosage was held constant, and the water content was adjusted to secure the required

slump flow consistency. Casting the concrete using bucket from the top was reported to reduce

slightly the maximum pressure, despite the relatively high casting rate of 50 m/hr. When

pumping from the bottom was used, the lateral pressure was observed to significantly increase.

In the case of the NCC AB Company, 10 full-scale trial castings were conducted on wall

elements in Billeberga, Sweden. Eight of the trials were carried out on a wall element measuring

69


2.65 m in height, 5.7 m in length, and 0.16 m in width. The last two trials were conducted on

walls having higher heights of 5.3 and 8 m. A single SCC mixture was used to fill all walls. The

cement and limestone filler (Ignaberg 500) contents were set at 330 and 125 kg/m3, respectively.

The w/c was 0.55. The contents of sand (0-8 mm) and coarse aggregate (8-16 mm) were 1,047

and 702 kg/m3, respectively. A polycarboxylate-based HRWRA (Sika Viscocrete 2) was used at a

dosage of 1.7%, by cement mass. The slump flow of the tested mixtures varied from 620 to 780

mm. The lateral pressure was measured using hydraulic jacks located at various elevations.

Different techniques were used to place the concrete in the formworks. A pump was used to place

the concrete in seven walls, while buckets were used to place the remaining three walls. All

placements were from the top of the formwork. When the pump was used, a fireman's hose was

added at the end of the pump line to limit the freefall of the concrete to approximately 750 mm.

The casting rates varied from 6 to 120 m/hr. Despite these high casting rates, NCC AB engineers

reported that the developed pressure right after the end of casting was considerably lower than

hydrostatic pressure. For example, for the wall measuring 8 m in height and cast at rate of rise of

120 m/hr, the pressure measured at 550 mm from the bottom was 29% of the hydrostatic value.

In the case of the wall measuring 5.3 m in height, this pressure was 59% of the hydrostatic limit.

Proske and Graubner [2002] evaluated the influence of casting rate and slump flow value

on formwork pressure developed by SCC. Eleven experimental tests were conducted on columns

measuring 4 m in height and 0.3 m x 0.3 m in cross-sectional dimensions. Ten columns were

reinforced. The casting rate and slump flow value varied from 12.5 to 160 m/hr and 550 to 750

mm, respectively. The SCC mixture used for casting the columns had a w/c of 0.43 and a sand-to-

coarse aggregate ratio of 0.48. The author reported that the SCC mixtures of 750 mm slump flow

placed at casting rates of 25, 40, 80, and 160 m/hr., developed almost hydrostatic pressure. For

the casting rate of 12.5 m/hr, the SCC was shown to exhibit a pressure reduction of 23% from the

hydrostatic limit. On the other hand, for a fixed casting rate of 25 m/hr, the decrease in slump

flow from 750 to 550 mm was reported to reduce the maximum pressure by approximately 40%

of the hydrostatic pressure.

Andreas and Cathleen [2003] compared lateral pressure characteristics of SCC of various

workability levels to that of conventional vibrated concrete. Experimental wall elements

measuring 2.7 m in height, 0.75 m in length, and 0.20 m in width were used (Fig. 2.57). The

walls were reinforced with 10-mm diameter bars employed at a density of 50 kg/m3. Five

70


pressure sensors were used to determine the lateral pressure distribution. The walls were cast at

approximate rate of 8 m/hr. The walls were filled with two batches, each taking 1.5 min to cast

with 20-min break between the lifts. SCC cast from the top displayed between 87% and 90% of

the hydrostatic pressure. The conventional concrete developed about 55% of the hydrostatic

pressure. Within the first 20 min, the pressure decay at the lowest sensor ranged between 7% and

20%. Within two hours, the concrete pressure reached about 50% of the maximum initial

pressure. Andreas and Cathleen [2003] also reported that the slump flow of SCC had no great

influence on the maximum pressure when the casting rate was about 8 m/hr. This was, however,

not the case a slower casting rate of 4 m/hr was tried. The incorporation of VMA was shown to

lead to lower lateral pressure. In the case of SCC pumped into the formwork from its base, the

pressure was shown to rise locally above hydrostatic values, and that the maximum pressure

might be dependent on the pump pressure.

famivrofk

(270*90)

-Sonde 2 5ensor2

r Fig. 2.57 Test set-up for pressure measurements in laboratory (wall 2.70 x 0.75 x 0.20 m)

[Andreas and Cathleen, 2003]

Billberg [2003] used two SCC mixtures (SCC1 and SCC2 with 0.40 and 0.45 w/c,

respectively) and one conventional concrete to cast eight experimental wall elements measuring 3

m in height, 3.5 m in length, and 0.3 m in width. The slump flow at the time of casting was 730 ±

50, 700 ± 50 for the SCC1 and SCC2, respectively. The formwork surface material was timber on

one side and plywood on the other. A form-releasing agent was applied on half of each form-side.

The concrete was dropped from a height of 1 ± 0.5 m above the concrete surface. Relatively low

casting rates were used in casting the wall elements, as indicated in Table 2.3. The pressure

71


envelopes for the eight tested mixtures are shown in Fig. 2.58. The lateral pressure for the

vibrated concrete (NC) was slightly lower than that of SCC placed at the same casting rate. The

increase in the rate of casting led to greater pressure. A linear relationship was established

between the maximum initial lateral pressure and the casting rate; this relationship was presented

earlier in the report in Fig. 2.39.

Table 2.3 Variation of mixture designs and casting rates [Billberg, 2003]

Casting no. Mix type Casting rate (m/hr) 1 2 3 4 5 6 7 8

SCC1 SCC2 SCC2 SCC1 SCC1 SCC2 SCC1 NC

1.4 1.3 1.0 0.8 1.5 2.2 2.3 1.5

19 20 3ft Form pressure (kPa)

40 50

Fig. 2.58 Final form pressure envelopes for fully cast walls using SCC (casting no. 1-7) and

conventionally vibrated concrete (casting no. 8) [Billberg, 2003]

Khayat and co-workers [2005B] evaluated formwork pressure characteristics exerted by

SCC used for the repair of an underpass retaining wall in Montreal, Canada that was carried out

in 2003. The repair consisted in removing the outer layer of concrete to a depth of 0.15 to 0.20 m

and the first layer of steel, and replacing it with high-performance SCC. New vertical and

horizontal reinforcing bars were provided. The SCC was cast through a pump line from the top of

the wall panels measuring 6.7 m in width. In the work reported here, the lateral pressure was

monitored along 5.5-m high repair sections. Seven pressure sensors were mounted at various

72


heights along the plywood formwork to monitor the pressure. The mixture compositions and

fresh characteristics of the tested mixtures are given in Table 2.4.

Table 2.4 Mixture proportion and workability of SCC used in repair [Khayat et al., 2005B]

Identification BIN-0.50-Nap

TER-0.50-Nap

TER-0.44-Nap

TER-0.54-

Nap-Fib

TER-0.54-

Poly-Fib

Cement T10, kg/m3

Binary cement T10E-SF, kg/m3

Ternary cement T10EF-SF, kg/m3

Tuf-Strand synthetic fibers, kg/m3

237

238

475 475 475 2.3

475 2.3

w/cm

Sand, kg/m3

Coarse aggregate (2.5-10 mm), kg/m3

0.40

780 790

0.40

770 775

0.40

670 880

0.40

850 675

0.40

850 650

PNS HRWRA, L/m3

PC HRWRA, L/m3

Water reducer, mL/100 kg of binder VMA, mL/100 L of water AEA mL/100 kg of binder

7.3 -

200 1100 80

6.8 -

200 1100 85

8.6 -

200 1100 110

8.2 -

200 1000 100

-

4.3 -

260 50

Casting rate (rn/hr)

Initial temperature of fresh concrete (°Q

Initial slump flow after pumping (mm)

Initial slump flow through J-Ring (mm)

Initial L-box blocking ratio (h2/hl)

Initial filling capacity (%)

Initial air volume (%)

Unit weight (kg/m3)

Maximum surface settlement (%)

6.0

8.0

665

650

0.63

87.5

7.1

2,160

6.5

20.4

740

810

0.97

93.2

7.9

2,250

7.5

17.0

590

470

0.58

76.0

5.1

2,225

0.49-

9.5

22.8

640

540

0.47

63.6

7.1

2,200

0.54

8.0

14.0

600

585

0.52

67.9

5.7

2,230

Comparing variations of normalized lateral pressures developed by the five SCC mixtures

in time, as determined from the first sensor placed at 0.3 m from the bottom where the maximum

pressure was measured are shown in Fig. 2.59. The SCC TER-0.44-Nap mixture made using a

higher concentration of coarse aggregate (S/A of 0.44) developed the lowest maximum pressure;

this mixture also had the highest thixotropy value. The maximum initial pressure was limited to

40% of the hydrostatic limit. This was due to the lower initial slump flow and greater internal

73


friction caused by the higher coarse aggregate concentration of this concrete. The TER-0.44-Nap

mixture exhibited relatively long time before pressure cancellation. This is believed to be due to

the retarding effect of the naphthalene-based HRWRA that was incorporated at relatively high

dosage. The TER-0.50-Nap and BIN-0.50-Nap mixtures developed the highest lateral pressure of

approximately 60% of hydrostatic pressure. Contrary to the BIN-0.50-Nap SCC, the former SCC

exhibited sharp pressure decay. This was related to the differences in concrete temperature (8 °C

vs. 20 °C) of the fresh concrete as well as the ambient temperatures during the early stages of

hardening, which were higher in the case of the TER-0.50-Nap SCC.

TER-0.44-Nap BIN-0.50-Nap

TER-0.50-Nap

-TER-0.54-Poly-Fib

- TER-0.54-Nap-Fib

i _ i n I • • — • •

0 200 400 600 800 1000 1200 0 200 400 600 800 1000 1200

Time afler pouring (min) Time afte r pouring (min)

Fig. 2.59 Variations in relative lateral pressure at 0.3 m from the bottom of 5.5-m high repair wall

sections determined for different repair SCC mixtures [Khayat et al., 2005B]

Tejeda-Dominguez et al. [2005] evaluated the lateral pressure exerted by SCC during the

casting of a massive reaction wall at the Civil Engineering lab. Facility, University of Illinois at

Urbana-Champaign. The wall was heavy reinforced and measured 24.4 m in length, 1.5 m in

width, and 8.5 m in height (Fig. 2.60). The concrete was cast continuously from the top using a

pump at a mean casting rate of 1.22 m/hr; the rate of rise of the SCC in the formwork fluctuated

between 0.61 to 1.68 m/hr. The targeted slump flow was 710 mm with occasional variations from

600 mm to 735 mm. The concrete temperature at the time of casting was 16 °C on the average

and peaked at 60 °C by the second day after placement. The initial and final setting times,

determined by penetration resistance (ASTM 403), were 5.9 and 7.8 hours, respectively.

The distribution of the lateral pressure was determined using seven sensors placed along the

wall. As shown in Fig. 2.61, hydrostatic pressure was not achieved throughout the entire height

of the wall. The Pmax recorded was only 23% of the Phyd. The highest pressures were not

74


measured at the bottom of the wall, where the head of concrete was larger, but corresponded to

the periods in time where the casting rates were faster. The maximum pressure reached in the

third wall was only 32% of the maximum hydrostatic pressure.

Fig. 2.60 Formwork of 8.5-m high strong Fig. 2.61 Envelope of maximum pressure

reaction wall exerted by the SCC in the reaction wall

[Tejeda-Dominguez et al., 2005] [Tejeda-Dominguez et al., 2005]

2.7 Concluding remarks

Several methods to characterise thixotropy are treated in the survey and merely all of them

have been used for cement paste and some also for concrete. Regarding the specific field of

cementitious materials, this literature survey has shown that these materials are true thixotropic

and also that the structural build-up at rest is significant and could be used as another way to

express the thixotropy rather than the breakdown area. It has also been shown that the various

parameters influencing the formwork pressure exerted by flowable concrete such as SCC could

affect also the degree of thixotropy. Recently the interest for thixotropy of SCC has grown

considerably and several attempts to model the form pressure related to its thixotropy have been

reported. A need to portable and compatible field test devices to monitor lateral pressure and

measure the rheological properties of SCC are mandatory. Propose effective ways to reduce

lateral pressure of SCC by providing models compromising the most affecting parameters,

guidelines, charts, to the engineer designers and contractors dealing with SCC.

75

CHAPTER 3

MATERIALS AND MIX DESIGNS

3.1 Materials

A wide diversity of materials was used in the course of this research. A focus on the

specific material characteristics used in the laboratory investigations as well as the mix design for

the evaluated mixtures are elaborated in the following sections.

3.1.1 Cement

Four types of cements were used. General type (Type GU cement), moderately sulfate-

resistant hydraulic cement (Type MS) and high initial-strength hydraulic cement (Type HE) were

used from St-laurent. Ternary tercem 3000 cement (GU-bS/F) from Lafarge with approximately

22% granulated blast-furnace slag, 6% silica fume, and 72% Type GU cement was also used. The

particle-size distributions for these cements are presented in Fig. 3.1, while physico-chemical

properties as well as mineralogical compositions of these cements are given in Table 3.1.

/ ~ v

£ <^s bl) a v> <n « a > •** «s 3 s s U

100

80

60

40

20

0 1000 100 10 1 0

Particle-size distribution (jim)

Fig. 3.1 Grain-size distributions of GU, HE, and GU-bS/F cements

Two additional cement types were used in the field study (construction of the CFI

laboratory): Type GU blended with fly ash (GU-bF), and Type GU blended with fly ash and

silica fume (GU-bF/S).

76

Chapter 3: Materials and mix designs

Table 3.1 Characteristics of GU, HE, MS, and GU-bS/F cements

Identification

Silicium (Si02), %

Alumina (A1203), %

Iron oxide (Fe203), %

Total calcium oxide (CaO)

Magnesium oxide (MgO),

Sulfur trioxide (S03), %

Equivalent alkali (Na20eq).

Loss on ignition (LOI), %

C3S, %

C2S, %

C3A, %

C4AF, %

Blaine surface, m2/kg

Specific gravity

Autoclave expansion, %

Water expansion, %

Sulfate resistant, %

Percentage passing 45 urn,

Air content, %

Compressive strength at 1



, %

%

, %

%

day, MPa

days,

days,

Compressive strength at 28 days

Initial setting time (Vicat),

Final setting time (Vicat),:

min

min

MPa

MPa

,MPa

TypeGU

19.9

4.7

3

62.7

2.1

3.4

0.89

2.4

58.2

13.1

7.4

9.3

384

3.15

0.03

-

-

94

6.4

-

27.8

31.9

39

125

210

Type HE

19.5

5.9

2.3

62.7

1.8

3.8

0.86

3.1

59.8

10.9

9.2

7

476

3.14

0.101

-

-

98

7.1

21.2

35.3

39.7

46.2

120

205

Type MS

21.2

4.6

2.9

63.3

1.9

3.4

0.92

0.8

51.6

21.9

7.3

8.8

389

3.14

0.038

0.013

0.035

94

7.8

-

27.3

33.4

42.3

120

250

GU-bS/F

33.8

10.4

3.4

44

1.4

2.6

0.68

2.1

-

-

-

-

443

2.88

0.022

0.004

0.042

83

5.7

-

17.3

22.5

38.9

140

230

3.1.2 Limestone filler

Two types of limestone fillers were used: Betocarb 3 and Betocarb 8 provided by the Omya

Company of St-Armand, Quebec. The physico-chemical characteristics of these fillers are

presented in Table 3.2. The particle-size distributions of the fillers are shown in Fig. 3.2.

The limestone filler is characterized by high calcium carbonate content (CaC03). This

mineral filler is inert and is mainly used to increase paste content to improve rheological

77


properties of SCC without increasing the heat of hydration. Generally, addition of fine particles

improves the granular skeleton and contributes in reducing water and superplasticizer demands.

Table 3.2 Physico-chemical properties of limestone fillers

Identification Betocarb 3 Betocarb 8

Calcium carbonate (CaCO"3), % Magnesium carbonate (MgCC^), % Y Brightness Retained on sieve (325 urn), % Humidity loss at 110 °C, % Specific gravity Average diameter, um Blaine surface, m2/kg

97 0.5 95.5 0.01 0.07 2.71 2.7 600

95 2 89

0.01 0.05 2.71 7.5 490

100 n"7TrT~!—•*

Betocarb 3

1000 100 10 1

Particle-size distribution (um)

Fig. 3.2 Grain-size distributions of limestone fillers

3.1.3 Fly ash

Fly ash (FA), a by-product of coal and pulverized lignite combustion, was provided by the

St. Lawrence Cement Company. According to ASTM C618-84, the FA Class F can be used up to

20% partial cement replacement, by mass of total binder content. The physico-chemical analysis

of the FA is presented in Table 3.3. The particle-size distribution of the FA is shown in Fig. 3.3.

78


Table 3.3 Physico-chemical properties of Class F fly ash

Chemical analysis

Oxides

Silicium (Si02), %

Alumina (A1203), %

Iron oxide (Fe2C>3), %

Total calcium oxide (CaO), %

Magnesium oxide (MgO), %

Sodium oxide (Na20), %

Potassium oxide (K2O), %

Sulfur trioxide (S03), %

Equivalent alkali (Na20eq), %

Value

52.4

27.2

8.3

4.5

0.96

0.20

2.33

0.05

1.74

Physical analysis

Designation

Blain specific surface, m /kg

Passing 45 urn, %

Mean apparent diameter, urn

Specific gravity

Bulk unit weight, kg/m3

Loss on ignition (LOI), %

Value

410

90

13

2.53

2160

2.2

100 T;2

? 80 £ CUD -i-T-i

1 60 ;rM W ..:?...!

^> £ 4 0 .11,

I 20 ;ti 3 u

0 L -

1000

Fig. 3.3 Particle-size distributions of Class F fly ash

3.1.4 Aggregate

Well-graded natural siliceous sand was used. The sand was provided by Betons Aime Cote

Inc. It has saturated-surface dry (SSD) specific gravity of 2.70, absorption of 1.12%, and fineness

modulus of 2.4. The particle-size distribution of this sand is given in Table 3.4. As presented in

Fig. 3.4, the particle-size distribution of the sand lies within CSA A23.2-5A recommended limits.

In order to avoid contamination with external agents, the sand was stored in air-tight barrels.

Crushed metamorphic calcareous aggregate, provided by Betons Aime Cote Inc. was used.

Three different maximum-size of aggregate (MSA) of 10, 14, and 20 mm were used. Some

100 10 1 Particle-size distribution (urn)

79

combinations between the different sizes of coarse aggregates were employed to enhance packing

density of the aggregate. The coarse aggregates had SSD specific gravities of 2.72, 2.71, and 2.72

and absorption values of 0.57% 0.38%, and 0.26%, respectively. The physical characteristics of

the aggregates are given in Table 3.5 and the particle-size distributions are presented in Table 3.6.

Table 3.4 Particle-size distribution of sand

Sieve size (mm) Passing (%) Specification limits

10 5

2.5

1.25

0.630

0.315

0.160

pan

^100 £ o w 80 a X X

2 60 P * <

Q»

>£ * 40 « s § 20

\ji

n

100

96.48

83.38

72.72

58.44

30.62

7.88

0.02

/ / / J?

/ / if *f

if 4 Jf /

/ft / f ,"•••• s? ~'*

meS-t-^.".....:, i i

**'

J at* n^r t

Ja / j r / t

/ t

t / / /

-CSAlowei

- Cumulate

100-100

100-94

100-80

90-50

66-24

34-10

10-2

-

JgigiPV*^

SP^ *

limit limit

epassing(%)

0 0,16 0315 0,63 1,25 2,5 5 10 Sieve size (mm)

Fig. 3.4 Grain-size distribution of sand

Table 3.5 Physical characteristics of

Test description 2.5-10 mm

Specific gravity (SSD) 2.72 Bulk specific gravity (dry) 2.70 Absorption (%) 0.57

coarse aggregates

5-14 mm

2.71 2.70 0.38

5-20 mm

2.72 2.71 0.36

80


Table 3.6 Grain-size distributions of coarse aggregates

Sieve size (mm) -

25 19

12.5 9.5

4.75 2.36 1.18 pan

Cumulative passing

2.5-10 mm

100 100 100 100 13 2 2 0

5-14 mm

100 100 95 69 18 4 3 0

(%)

5-20 mm

100 99 68 40 7 1 1 0

3.1.5 Chemical admixtures

Several chemical admixtures were used in this study. Glenium 3030 NS, Plastol 5000 SCC,

and Optima 203 from BASF, Euclid, and Chryso, respectively, were used as high-range water-

reducing agents (HRWRA). The HRWRA was blended with the mixing water in order to

homogenize the solution. The water contained in the HRWRA was taken into account in mixture

proportioning to maintain a fixed w/b. Pozzolith 100 XR from BASF as set-retarding agent

(SRA) was introduced in specific SCC mixtures.

In order to increase the stability of fluid concrete, viscosity-modifying agent (VMA) and

air-entraining agent (AEA) were used. Three VMA were used: Rheomac VMA 362 and Rheomac

VMA 358 from BASF and Visctrol from Euclid. Air extra from Euclid was used as AEA. The

values for specific gravity, solid contents, and recommended dosages are presented in Table 3.7.

Table 3.7 Characteristics of chemical admixtures used

Admixture Identification Company Specific % Solid gravity content

Recommended dosage

HRWRA

VMA

Optima 203

Plastol 5000 SCC

Glenium 3030 NS

Rheomac VMA 362

Rheomac VMA 358

Visctrol

Chryso

Euclid

BASF

BASF

BASF

Euclid

1.040

1.070

1.047

1.002

1.100

1.210

21.5

32.0

20.3

~

39.1

43.5

300-3000 mL/100 kg of binder





1100-2700 mL/100 kg of water

AEA Air extra Euclid 1.000 10.5 30-100 mL/100 kg of binder

SRA Pozzolith 100 XR BASF 1.260 46.0 80-450 mL/100 kg of binder

81


3.2 Mixture composition

In total, 64 mixtures including 59 SCC mixtures, three conventional vibrated concretes

(CC), and two concrete-equivalent mortars (CEM) were prepared and tested in this study. The

mixture compositions are given in Table 3.8. It is worth to note that six mixtures had slump flow

values of 560 mm, which classify them as semi-SCC. They were named as SCC mixtures for

simplicity. In the CEM mixtures, the coarse aggregate was replaced by sand of an equivalent

surface area.

3.3 Mixing and testing sequence

All mixtures were prepared using 110-L capacity horizontal bowl mixer. In case of large

volume of concrete was needed, two mixers were prepared simultaneously. Mixing speed, along

with the combined action of blades and internal rotor, enable good dispersion of particles and

good homogenization of the mixture. The batching sequence was as follows:

1. Sand was added to the mixer and homogenized for 30 seconds before taking 500-g sample to

determine the humidity. Humidity and absorption of sand are required to make necessary

corrections in the previously determined quantity of mixing water.

2. The quantity of sand determined for the humidity correction was discharged in the mixer.

3. Coarse aggregate corrected for the humidity was then introduced and mixed for one minute.

4. The AEA, if any, dissolved in one third of mixing water was then introduced and the materials

were mixed for one minute.

5. The total powder (cement, limestone filler, and fly ash) and total quantity of HRWRA diluted

in one third of mixing water were then introduced and mixing was resumed for 3 minutes.

6. The VMA and SRA liquefied in the last third of the mixing water were then discharged and

the mixing was continued for two minutes.

7. The mixture was then left at rest for two minutes, in which the concrete temperature was

recorded, and after that it was agitated for 30 sec.

8. The slump or slump flow was measured. For some mixtures, additional dosages of HRWRA

were added to achieve the target slump. Two-minutes mixing were recommenced for each

increase in the HRWRA dosage before measuring the slump again.

Once the target slump was obtained, the characterization tests for the concrete were proceeded.

82

Cha

pter

3: M

ater

ials

and

mix

des

igns

Mat

eria

ls

Slum

p fl

ow (

targ

et v

alue

)

Tab

le 3

.8 M

ixtu

re c

ompo

sitio

ns

SCC

1 SC

C2

SCC

3 SC

C4

SCC

5 SC

C6

SCC

7 SC

C8

SCC

9 SC

C10

mm

56

0±13

* 66

0±13

76

0±13

56

0±13

66

0±13

76

0±13

66

0±13

66

0±13

66

0±13

66

0±13

Cem

ent T

ype

GU

Cla

ss F

Fly

ash

Tot

al b

inde

r: b

(c

emen

titio

us m

ater

ials

: cm

)

Tot

al p

owde

r: p

Wat

er

Sand

(0-

5 m

m)

Coa

rse

agg.

(5-

14 m

m)

Coa

rse

agg.

(5-

20 m

m)

kg/m

3

365

122

487

487

200

365

122

487

487

200

365

122

487

487

200

365

122

487

487

200

365

122

487

487

200

365

122

487

487

200

365

122

487

487

200

335

112

447

447

185

365

122

487

487

202

394

131

525

525

219

w/b

(w

/cm

)

w/p

Past

e vo

lum

e

Sand

/Tot

al a

ggre

gate

, by

vol.

1/m

3

0.42

0.42

370

0.5

0.42

0.42

370

0.5

0.42

0.42

370

0.5

0.42

0.42

370

0.5

0.42

0.42

370

0.5

0.42

0.42

370

0.5

0.42

0.42

370

0.5

0.42

0.42

340

0.5

0.42

0.42

370

0.5

0.42

0.42

400

0.5

kg/m

3

804

825 -

2.18

_

804

825 -

2.73

_

804

825 -

3.42

.

804

825 -

2.39

av

2.8

804

825 -

2.93

av

2.8

804

825 -

3.44

av

2.8

804

825 -

2.18

5.1

818

867 -

3.16

av

2.8

791 -

812

2.94

2.8

752

772 -

2.29

2.8

HR

WR

A (

BSA

F G

leni

um 3

030

NS)

,,

3 1/m

V

MA

(R

heom

ac V

MA

362

) 13

mm

slu

mp

flow

var

iatio

n w

as c

hose

n to

cor

resp

ond

0.5

in.

av: a

vera

ge o

f tw

o m

ixtu

res

or m

ore

83

Cha

pter

3: M

ater

ials

and

mix

des

igns

Tab

le 3

.8 (

cont

'd)

Mix

ture

com

posi

tions

Mat

eria

ls

Slum

p fl

ow (

targ

et v

alue

)

Cem

ent

Typ

e G

U-b

S/F

Cem

ent

Typ

e G

U

Cla

ss F

Fly

ash

Bet

ocar

b 8

limes

tone

fill

er

Tot

al b

inde

r: b

(c

emen

titio

us m

ater

ials

: cm

)

Tot

al p

owde

r: p

Wat

er

w/b

(w

/cm

)

w/p

Past

e vo

lum

e

Sand

/Tot

al a

ggre

gate

, by

vol.

Sand

(0-

5 m

m)

Coa

rse

agg.

(2.

5-10

mm

)

Coa

rse

agg.

(5-

14 m

m)

HR

WR

A (

BSA

F G

leni

um 3

030

NS)

HR

WR

A (

Opt

ima

203)

VM

A (

Rhe

omac

VM

A 3

62)

mm

kg/m

3

1/m

3

kg/m

3

1/m

3

seen

SC

C12

SC

C 1

3

660±

13 6

60±1

3 66

0±13

- 390

130

520

520

189

0.37

0.37

370

0.5

791 - 812

4.04

- 2.8

-

342

114

456

456

213

0.47

0.47

370

0.5

792 - 813

2.13

- 2.8

-

365

122

487

487

202

0.42

0.42

370

0.5

797

812 -

3.37

- 2.8

CE

M14

CE

M15

275±

20 2

75±2

0

560 - - 187

560

747

271

0.49

0.37

542

1.00

1212

- - - 7.5 -

500 - - 167

500

667

237

0.49

0.37

486

1.00

1383

- - - 14

3.76

SCC

16

660±

13

-

370

92

462

462

192

0.42

0.42

350

0.5

818 -

840 3 - 2.9

SCC

17

560±

13

-

542

136

678

678

248

0.37

0.37

479

0.45

583 -

732 2 - 4.1

SCC

18

560±

13

-

378

94

472

472

172

0.37

0.37

333

0.45

756 -

949

3.1 - 4.1

SCC

19

560±

13

-

419

105

524

524

192

0.37

0.37

370

0.55

872 -

732

2.8 - 4.1

SCC

20

600±

13

- 398

133

531

531

210

0.40

0.40

393

0.48

731 -

813

2.0 -

1.75

84

Cha

pter

3: M

ater

ials

and

mix

des

igns

Tab

le 3

.8 (

cont

'd)

Mix

ture

com

posi

tions

Mat

eria

ls

Slum

p fl

ow (

targ

et v

alue

)

Cem

ent

Typ

e G

U

Cla

ss F

Fly

ash

Tot

al b

inde

r: b

(c

emen

titio

us m

ater

ials

: cm

)

Tot

al p

owde

r: p

Wat

er

w/b

(w

/cm

)

w/p

Past

e vo

lum

e

Sand

/Tot

al a

ggre

gate

, by

vol.

Sand

(0-

5 m

m)

Coa

rse

agg.

(5-

14 m

m)

HR

WR

A (

BSA

F G

leni

um 3

030

NS)

VM

A (

Rhe

omac

VM

A 3

62)

mm

kg/m

3

1/m

3

kg/m

3

1/m

3

SCC

21

600±

13

296

99

397

397

156

0.40

0.40

292

0.50

944

894

3.78

1.75

SCC

22

600±

13

422

141

563

563

223

0.40

0.40

417

0.52

772

732

2.5

1.75

SCC

23

600±

13

395

132

527

527

209

0.40

0.40

390

0.44

685

894

2.1

1.75

SCC

24

720±

13

504

168

672

672

266

0.40

0.40

497

0.44

560

732

2.5

1.75

SCC

25

600±

13

504

168

672

672

267

0.40

0.40

498

0.44

560

732

1.71

1.75

SCC

26

600±

13

395

132

527

527

209

0.40

0.40

391

0.44

685

894

1.89

1.75

SCC

27

600±

13

422

141

563

563

223

0.40

0.40

417

0.52

772

732

2.20

1.75

SCC

28

600±

13

296

99

395

395

156

0.4

0.4

295

0.52

944

894

4

1.75

SCC

29

720±

13

504

168

672

672

267

0.40

0.40

498

0.44

560

732

2.27

1.75

SCC

30

720±

13

395

132

527

527

209

0.40

0.40

391

0.44

685

894

2.29

1.75

85

Cha

pter

3: M

ater

ials

and

mix

des

igns

Tab

le 3

.8 (

cont

'd)

Mix

ture

com

posi

tions

Mat

eria

ls

Slum

p fl

ow (

targ

et v

alue

)

Cem

ent

Typ

e G

U-b

S/F

Cem

ent T

ype

GU

Cla

ss F

Fly

ash

Bet

ocar

b 8

limes

tone

fill

er

Tot

al b

inde

r: b

(c

emen

titio

us m

ater

ials

: cm

)

Tot

al p

owde

r: p

Wat

er

w/b

(w

/cm

)

w/p

Past

e vo

lum

e

Sand

/Tot

al a

ggre

gate

, by

vol.

Sand

(0-

5 m

m)

Coa

rse

agg.

(2.

5-10

mm

)

Coa

rse

agg.

(5-

14 m

m)

HR

WR

A (

BSA

F G

leni

um 3

030

NS)

HR

WR

A (

Opt

ima

203)

VM

A (

Rhe

omac

VM

A 3

62)

mm

kg/m

3

1/m

3

kg/m

3

1/m

3

SCC

31

720±

13

-

422

141

563

563

223

0.40

0.40

418

0.52

772 -

732

2.96

-

1.75

SCC

32

660±

13

-

295

98

393

393

154

0.40

0.40

294

0.52

944 - 894

6.27

-

1.75

SCC

33*

660±

13

-

408

136

544

544

216

0.40

0.40

404

0.48

731 -

813

2.50

.- 1.75

CC

34+

180±

20

-

408

136

544

544

216

0.40

0.40

401

0.48

731 -

813 - -

1.75

CC

35

200±

20

384 - - 128

384

512

188

0.49

0.37

370

0.5

811

165

661 - 2.8 -

SCC

36

700±

20

353 - - 118

353

471

167

0.49

0.37

340

0.5

849

174

692 - 8.5 -

SCC

37

700±

20

374 - - 125

374

499

179

0.49

0.37

360

0.5

823

168

671 - 6.4 -

SCC

38

700±

20

384 - - 128

384

512

185

0.49

0.37

370

0.5

809

165

661 - 5.8 -

SCC

39

700±

20

405 - - 135

405

540

195

0.49

0.37

390

0.5

783

160

639 - 5.7 -

SCC

40

700±

20

415 - - 138

415

553

201

0.49

0.37

400

0.5

770

157

628 - 5.0 -

* T

este

d si

x tim

es in

Chp

t. 6

(SC

C33

-A, B

, C, D

, E, F

) an

d 4

times

in C

hpt.

7 (S

CC

33-C

,D, E

, F)

+ C

onve

ntio

nal

conc

rete

86

Cha

pter

3: M

ater

ials

and

mix

des

igns

Tab

le 3

.8 (

cont

'd)

Mix

ture

com

posi

tions

Mat

eria

ls

Slum

p fl

ow (

targ

et v

alue

)

Cem

ent

Typ

e G

U-b

S/F

Bet

ocar

b 8

limes

tone

fill

er

Tot

al b

inde

r: b

(c

emen

titio

us m

ater

ials

: cm

)

Tot

al p

owde

r: p

Wat

er

w/b

(w

/cm

)

w/p

Past

e vo

lum

e

Sand

/Tot

al a

ggre

gate

, by

vol.

Sand

(0-

5 m

m)

Coa

rse

agg.

(2.

5-10

mm

)

Coa

rse

agg.

(5-

14 m

m)

HR

WR

A (

Opt

ima

203)

VM

A (

Rhe

omac

VM

A 3

62)

mm

kg/m

3

1/m

3

kg/m

3

1/m

3

SCC

41

700±

20

353

118

353

471

167

0.49

0.37

340

0.5

849

173

692

9.2

1.2

SCC

42

700±

20

374

125

374

499

179

0.49

0.37

360

0.5

823

168

671

6.7

1.3

SCC

43

700±

20

384

128

384

512

184

0.49

0.37

370

0.5

809

165

659

6.1

1.3

SCC

44

700±

20

405

135

405

540

195

0.49

0.37

390

0.5

783

160

638

6.1

1.4

SCC

45

700±

20

415

138

415

553

200

0.49

0.37

400

0.5

770

157

628

5.8

1.5

SCC

46

700±

20

353

118

353

471

167

0.49

0.37

340

0.5

849

173

692

8.5

2.6

SCC

47

700±

20

374

125

374

499

179

0.49

0.37

360

0.5

823

168

671

6.7

2.8

SCC

48

700±

20

384

128

384

512

185

0.49

0.37

370

0.5

810

165

660

6.5

2.9

SCC

49

700±

20

405

135

405

540

195

0.49

0.37

390

0.5

783

160

638

6.1

3.0

SCC

50

700±

20

415

138

415

553

200

0.49

0.37

400

0.5

770

157

628

6.2

3.1

87

Cha

pter

3: M

ater

ials

and

mix

des

igns

Tab

le 3

.8 (

cont

'd)

Mix

ture

com

posi

tions

Mat

eria

ls

Slum

p flo

w (

targ

et v

alue

) m

m

Cem

ent

Typ

e G

U-b

S/F

Cem

ent

Typ

e G

U

Cem

ent T

ype

MS

Cem

ent

Typ

e H

E

Cla

ss F

Fly

ash

kg

/m3

Bet

ocar

b 8

lim

esto

ne f

iller

T

otal

bin

der:

b

Tot

al p

owde

r: p

W

ater

w

/b (

w/c

m)

w/p

•5

Past

e vo

lum

e 1/m

Sa

nd/T

otal

agg

rega

te, b

y vo

l. Sa

nd (

0-5

mm

) C

oars

e ag

g. (

2.5-

10 m

m)

, ,

, kg

/m3

Coa

rse

agg.

(5-

14 m

m)

Coa

rse

agg.

(5-

20 m

m)

HR

WR

A (

BSA

F G

leni

um 3

030

NS)

H

RW

RA

(O

ptim

a 20

3)

HR

WR

A (

Euc

lid P

last

ol 5

000)

V

MA

(R

heom

ac V

MA

362

) ^

3

VM

A (

Rhe

omac

VM

A 3

58)

m

VM

A (

Euc

lid V

istr

ol V

MA

)

Set R

etar

der

(Poz

zolit

h 10

0 X

R)

Euc

lid A

EA

(A

ir e

xtra

)

SCC

51

SCC

52

SCC

53

SCC

54

SCC

55

SCC

56

700±

20

700±

20

700±

20

700±

20

700±

20

700±

20

475

- 47

5 -

340

- -

- 44

0 50

0 35

2 85

-

88

99

-

475

425

475

440

500

440

475

425

475

440

500

440

197

175

183

139

163

141

0.42

0.

42

0.39

0.

34

0.34

0.

34

0.42

0.

42

0.39

0.

34

0.34

0.

34

365

323

351

296

373

290

0.5

0.5

0.5

0.46

0.

54

0.54

78

3 82

4 80

4 83

9 93

6 99

5 81

0 -

830

- -

-84

6 -

985

798

870

2.6

3.3

2.6

9.6

- 10

10

1.

0 2

- -

- -

-0.

5 0.

75

..

..

0.9

0.9

0.11

.

..

.

SCC

57

640±

20

500

500

500

162

0.34

0.

34

330

0.54

93

6

798

9.15

av

0.5 -

SCC

58

SCC

59

SCC

60

600±

13

600±

13

600±

13

384

384

384

128

128

128

384

384

384

512

512

512

186

185

186

0.49

0.

49

0.49

0.

37

0.37

0.

37

370

370

370

0.5

0.5

0.5

815

814

815

827

825

828

4.8

4.4av

4.

5

. -

88

Cha

pter

3: M

ater

ials

and

mix

des

igns

Tab

le 3

.8 (

cont

'd)

Mix

ture

com

posi

tions

Mat

eria

ls

CC

61

SCC

62

SCC

63

SCC

64

Slum

p fl

ow (

targ

et v

alue

) m

m

120±

30

650±

25

650±

25

650±

25

Cem

ent

Typ

e G

U-b

S 41

0

Cem

ent

Typ

e G

U-b

F/S

Tot

al b

inde

r: b

(c

emen

titio

us m

ater

ials

: cm

)

Tot

al p

owde

r: p

Wat

er

kg/m

3

-

410

410

165

450

450

450

157

500

500

500

185

420

420

420

175

w/b

(w

/cm

) 0.

40

0.35

0.

37

0.42

w/p

0.

40

0.35

0.

37

0.42

Past

e vo

lum

e 1/

m3

301

330

370

330

Sand

/Tot

al a

ggre

gate

, by

vol.

0.44

0.

48

0.49

0.

5

Sand

(0-

5 m

m)

810

872

840

910

Coa

rse

agg.

(2.

5-10

mm

) kg

/m3

829

755

685

717

Coa

rse

agg.

(5-

20 m

m)

211

189

175

183

HR

WR

A (

Supe

r E

ucon

37-

Euc

lid (

PNS)

) 3.

675

- -

5.9

HR

WR

A (

Plas

tol

5000

PC

P) (

Dss

s=1.

07)

VM

A (

Vis

trol

VM

A)

(Dss

s=l .

21)

1/m

Set r

etar

der,

Euc

on 7

27 (

Dss

s=1.

16)

3

- -

0.20

5.11

0.86

0.54

5.5

1

0.62

- 1.5

-

Wat

er r

educ

er, E

ucon

DX

(Dss

s=l.

14)

0.

94

0.82

0.

91

0.96

89

CHAPTER 4

METHODOLOGY FOR LATERAL PRESSURE MEASUREMENTS

4.1 Introduction

Various instrumentations were developed for monitoring lateral pressure variation exerted

by SCC. A metallic pressure column developed at Universite de Sherbrooke in 2003 for

monitoring the variation of maximum initial lateral pressure [Khayat and Assaad, 2008] was

initially modified using the same geometry but with PVC material to reduce weight (UofSl

pressure column). As elaborated in Chapter 5, the device was again modified to develop a

portable pressure column enabling the simulation of concrete castings up to 13 m in height

(UofS2 pressure column). An experimental PVC column measuring 3 m in height and 0.2 m in

diameter to monitor variation of lateral pressure with time during the plastic stage was used to

validate the UofS2 pressure column in Chapter 5.

The variation of lateral pressure with time until pressure cancelling was monitored using

PVC column measuring 1.2 in height and 0.2 m in internal diameter. Plywood formwork of 1.5 m

in height, 0.4 m in length and variable widths of 0.2, 0.25, 0.3, and 0.35 m was also prepared to

investigate the effect of formwork geometry on lateral pressure variation until cancellation.

Pressure cells of 19 mm in diameters were used for the sensors. They were calibrated frequently

using hydraulic machine and with water head prior each use. The devices used in this project to

monitor form pressure are elaborated in the sections that follow.

4.1.1 Evaluation of initial maximum formwork pressure

A metallic pressure column was developed in the Universite de Sherbrooke for monitoring

the development of formwork pressure shortly after casting and the variation in pressure envelope

for plastic concrete. This apparatus was proposed in 2003 by Khayat and Assaad and has been

used in a series of experiments since, both in the laboratory and in the field [Khayat and Assaad,

2008]. The metallic column is 1.2 m in height and 0.2 m in internal diameter and can be used to

simulate concrete heads of up to 9 m given overhead pressure introduced at the top of the column

(Fig. 4.1). The column consists of a metallic tube of 3 mm in thickness and has a smooth inner

surface to minimize friction with concrete. Prior to each use, the inner surface of the column is

coated lightly with formwork release oil. SCC is continuously cast to a height of 1.0 m with the

90

Chapter 4: Methodology for lateral pressure measurements

desired placement rate without vibration. On the other hand, conventional concrete is placed in

four layers and rodded 10 times per layer using the standard rod. The top end of the column is

then sealed, and the air pressure is gradually introduced from the top to simulate concrete casting

up to 9 m at a given rate of rise. Dial gauge of 200 kPa capacity and 2 kPa precision is mounted

on a regulator to adjust and control passing of air to the column. The rate of rise in pressure is

adjusted to maintain the desired casting rate. The stress induced by concrete pressure in the

metallic pressure column is determined using four pressure sensors (Fig. 2.46) fixed at 0.92, 0.77,

0.62, and 0.42 m from the top surface of the free concrete head.

For field-compatible, the material of the column was changed to light and rigid PVC of 10

mm in thickness (Fig. 4.2). The column has similar dimensions and instrumentations as the

metallic column. This modified column is identified here as UofSl pressure column.

A rigid PVC tube (Fig. 4.3) measuring 3 m in height and 0.2 m in diameter with wall

thickness of 10 mm was used to monitor lateral pressure variations during the plastic stage. The

pressure is measured using a pressure transducer placed at the bottom of the tube and flushed

with the inner surface.

4.1.2 Evaluation of lateral pressure decay

In order to determine concrete pressure variations until cancellation, an experimental PVC

column measuring 1.2 m in height and 0.2 m in diameter was used (Fig. 4.4). The column is

made of rigid PVC with 10 mm wall thickness and has a smooth inner surface to minimize

friction with concrete during placement. The PVC column has a seam along its height to facilitate

demolding and is tightened along the height by a number of radial ties. The inner surface of the

column is coated lightly with formwork release agent prior to each use. The SCC mixtures are

cast continuously at the desired rate from the top without any vibration; whereas the conventional

concrete is cast in four layers to total height of 1 m with each layer compacted using the standard

rod (10 blows per layer). The concrete pressure is monitored using three pressure sensors (Fig.

4.6) mounted at 1.0, 0.8, and 0.6 m from the top surface.

4.1.3 Evaluation of formwork width on lateral pressure

As illustrated in Fig. 4.5, a rectangular plywood formwork measuring 1.5 m in height, 0.4

m in length, and variable widths of 0.2, 0.25, 0.3, and 0.35 m was used to determine the effect of

91


formwork geometry on initial lateral pressure and its decay. The inner surface of the formwork is

sprayed with one layer of release oil prior to each use.

Three pressure sensors were set flush to the inner surface of the concrete in the short lateral

dimension (of the variable dimensions) and named "A sensors". In addition, three sensors were

attached in the same manner to the long lateral dimension (0.4 m) and named "B sensors". As

shown in Fig. 4.5, the A sensors were fixed at 0.1 m from the edge at depths of 1.45, 1.25, and

0.95 m from the top concrete surface. The B sensors were fixed at the same depths, but in the

middle of the 0.4 m.

In .--it

:iii;; £-1

Fig. 4.1 Metallic

pressure column of 1.2

m in height and 0.2 m in

diameter to monitor

initial pressure of

simulated formwork

height up to 9 m

.•tfttIM

Fig.4.2PVCUofSl Fig. 4.3 3-mhigh

pressure column of 1.2 PVC column of 0.2 m

m in height and 0.2 m in diameter to monitor

in diameter to monitor pressure variation

initial pressure of during the plastic

simulated formwork stage

height up to 9 m

Fig. 4.4 1.2-m

high PVC column

of 0.2 m in

diameter to

measure pressure

decay

92


•

B-sensors J '- • '\

1 A-sensors

Fig. 4.5 Rectangular plywood formwork instrumented with pressure sensors in

longitudinal and transverse direction (all dimensions in m)

4.2 Data acquisition

A data acquisition system was used to pick up output milli-volt signals representing

pressure variation at 90-sec intervals. The pressure data were collected 2 hr after end of casting

the pressure columns and 24 hr for the 1.2-m high PVC column and plywood formwork.

4.3 Measuring systems

The lateral stress induced by the concrete in the pressure columns, plywood formwork, as

well as the formworks constructed in the field were determined using pressure sensors from

Honeywell (Flush Diaphragm Millivolt Output Type Pressure Transducer) (shown in Fig 2.47).

93


The Honeywell sensor works using semiconductor gages on bending beams isolated by stainless

steel media. This sensor is AB-high-performance pressure transducer and is extremely accurate

down to 0.25% over a wide compensated temperature range. The sensor is 19 mm in diameter

and can operate over a temperature range varying from -54 to +93 °C. The sensor is connected to

data acquisition system, which excites the sensor by 5 Vdc. This sensor has a capacity of 344 kPa

(50 psi). The transducer's body is made in a configuration permitting its use as a "flush-mounted"

device to the inner surface of the wall element through drilled holes. These sensors were fixed at

the pressure columns using special adaptors, while they mounted in the plywood formwork and

field formworks using steel plates and riveted screws. A film of grease was applied on the sensor

contact surface for protection. The sensor was sealed to prevent leakage during the overhead

pressure application. The pressure cells were periodically calibrated, as discussed below.

Lateral pressure for the 1.2 and 3-m high PVC columns was determined using 19-mm

pressure sensors having 100-kPa capacities (Fig. 4.6). In order to compare the effect of sensor

diameter on the lateral formwork pressure characteristics, the 19-mm diameter sensor described

above was compared to a 38-mm diameter sensor (GP:50) shown in Fig. 4.7. The GP:50 sensor

can sustain pressure up to 6900 kPa (1000 psi) with 0.1%-0.5% accuracy. The enclosure of this

sensor is stainless steel, and is excited by 12 Vdc.

/"

(a -.

Fig. 4.6 19-mm pressure sensor used in

PVC Columns

Fig. 4.7 GP:50 flush diaphragm pressure

transmitter of 38-mm diameter

4.4 Calibration of pressure sensors

Pressure sensors were occasionally calibrated by mechanical pressure using special

mechanical device. A coefficient of mechanical calibration (Cm) was deducted to convert the

output milli-volt signal of the pressure sensor collected by the data acquisition system (Plraw data

in mV) to kPa-pressure value (P2corrM in kPa). The mechanical calibration was monthly validated

94


hydrostatically using given water column and overhead air pressure. The hydrostatic calibration

was carried out directly on the sensors that are already instrumented in the pressure column. A

hydrostatic calibration factor (CH) is employed to adjust P2corr.M value and produce a new

pressure value corrected hydrostatically (P3COrr.H)- Prior to each use, fast check up of the Cm and

CH calibration factors with water head was carried out. Water calibration factor (Cw) is

determined to modify the P3corr.H value to a new pressure value (P4corr.w)- More details about the

calibration methods are found in Appendix A.

4.5 Pressure sensor configurations (19 mm vs. 38 mm)

An experiment was conducted to show the effect of the exposed surface area of the pressure

sensor on the pressure readings. The Honeywell pressure sensor of 19-mm diameter (shown in

Fig. 2.46) and the GP:50 pressure transducer of the 38-mm diameter (shown in Fig. 4.7) were set

flush to the inner surface of the plywood formwork at depth of 0.8 m from the top concrete

surface. Typical SCC formulation with initial slump flow of 700 mm and unit weight of 2.324

t/m was cast at 21.5 m/hr. The variations of maximum lateral pressure over time of the two

sensors were found to be identical, as shown in Fig. 4.8. Therefore, for SCC made with coarse

aggregate of nominal maximum size of 20 mm, the use of the smaller pressure sensor does not

seem to present a problem with pressure measurements.

60

R = 21.5 m/hr 4 = 700 mm Effective H = 0.80 m

38-mm diameter pressure sensor

120 180 Time (min)

240 300

Fig. 4.8 Variations of lateral pressure determined from 19 and 38-mm diameter pressure sensors

95

CHAPTER 5

DEVELOPMENT PORTABLE DEVICE TO MEASURE SCC FORMWORK PRESSURE

5.1 Introduction

Use of a reliable and accurate measuring system to evaluate lateral pressure exerted on the

formwork by the plastic concrete is essential for the proper design of the formwork, both from

economic and safety points of views. In the first stage of the research program, the UofSl

pressure column was developed based on the earlier metallic pressure column [Khayat and

Assaad, 2008] and successfully employed in laboratory investigations to monitor lateral pressure

exerted by SCC; however, it could not satisfy the requirements for being a field-compatible

apparatus due to the heavy weight of the concrete plug. The modified version of the pressure

column is therefore designed to be portable and to receive a relatively limited volume of concrete

for the determination of lateral pressure envelope versus concrete placement height. The modified

pressure column was also made to enable the simulation of greater concrete height (13 m vs. 9 m

for the first version) and to provide more accurately control of the overhead pressure exerted on

the concrete.

5.2 Research significance and objectives

Special tools for accurate measurements of the lateral pressure exerted by SCC on

formwork systems are necessary given the rapid spread of SCC in ready-mix applications. This

chapter presents the development of a field-compatible apparatus for monitoring SCC formwork

pressure and its variation in time after casting. This device can be used selecting raw material,

developing mix designs, and monitoring actual pressure envelop for a given concrete during

casting, which it could be of special interest to design engineers of the formwork and contractors.

The main objective of this chapter is to validate the new pressure column. Series of

validations started by comparing the lateral pressure characteristics monitored using the new

device to those obtained from a 3-m high PVC column, followed by comparing its early decay in

lateral pressure to that obtained from a sacrificial PVC column measuring 1.2 m in height over 24

hr. The validation process was extended to monitor lateral pressure exerted by SCC of various

compositions and consistency levels cast at different placement rates.

96

Chapter 5: Development portable device to measure SCC formwork pressure

5.3 Testing program

A summary of the testing program presented in this chapter is described in Table 5.1. In

total, 13 SCC mixtures covering wide range of SCC compositions were used: SCC1 to SCC13.

The mixture proportionings are presented in Table 3.8. Work was undertaken first to investigate

the possibility of reducing the free concrete height in the UofSl pressure column from one meter

to 0.5 m and below. Based on results of the first experiment, fabrication of a new design for the

pressure device was employed (UofS2 pressure column). Further reduction of the free concrete

height in the UofS2 pressure column was investigated. For this purpose, free concrete heights of

0.35 and 0.2 m were proposed to be investigated.

Validation of the UofS2 pressure column was investigated through four main steps. Based

on the effective concrete plug recommended from the former stage, tests were carried out to

compare the initial lateral pressure and its early decay obtained using the UofS2 pressure column

filled with 0.5 m concrete height and subjected to an overhead pressure to simulate a 3-m high of

concrete to that obtained from a PVC column of similar diameter filled with 3-m high free

standing concrete. The pressure characteristics were determined while applying the overhead

pressure at rates of rise of 10 m/hr in the case of the pressure column. Three SCC mixtures were

used in this comparison. The second validation involved the comparison of the early pressure

decay monitored using the UofS2 pressure column to that obtained from the sacrificial PVC

column of 1.2 m in height during 24 hr. The third validation involved the evaluation of the

repeatability of the UofS2 pressure column through the measurements of the pressure responses

from four repetitions carried out on one SCC mixture. Finally, the capability of the UofS2

pressure column to capture the effect of key parameters affecting form pressure was evaluated.

These factors included placement rate (R), slump flow ((()), VMA dosage, paste volume, and

maximum-size of aggregate (MSA).

5.4 Fresh concrete properties

Once concrete mixing had finished, the concrete temperature, slump spread, T50, unit

weight, and air content of evaluated SCC mixtures were determined. The fresh properties of

evaluated mixtures are summarized in Table 5.2. The final HRWRA demands to produce the

target slump flow are presented in Table 3.8.

97

Table 5.1 Experimental work used to validate the UofS2 pressure column

Activity Variables SCC mixture* Reduce concrete height (H) in the UofSl pressure column

H= 1.0 and 0.50 m Typical SCC mixture

Design new pressure device (UofS2 pressure column) More reduction of H in new pressure column H = 0.5, 0.35, 0.2 m SCC4, SCC5, SCC6

S3

a o

03 

1-Compare pressure characteristics to a PVC column of 3-m high free standing concrete

SCC2, SCC3, SCC6

2-Compare early pressure decay to that of 1.2-m high PVC column

SCC5, SCC6, SCC8, SCC 10

3 - Repeatability of the pressure column Four repetitions SCC5 Effect of casting rate R = 2 and 10 m/hr SCC5

4-Evaluation of key parameters affecting form pressure

Effect of slump flow <|> = 560, 660, 760 mm SCC1.SCC2, SCC4, SCC5, SCC8, SCC9,

SCC3 SCC6 SCC 10

Effect of VMA and HRWRA contents

VMA =0,2.8, 5.1 1/m3 SCC1.SCC4 SCC2, SCC5, SCC7 SCC3, SCC6

Effect of paste volume Vp = 340, 370, 400 l/mJ SCC5, SCC8, SCC 10 Effect of MS A MSA = 10, 14, 20 mm SCC5, SCC9, SCC13

* Details of mixture proportioning are given in Table 3.8.

Table 5.2 Fresh concrete properties

Mixture Free H in UofS2 R Temp

pressure column (m) (m/hr) (°C)

Slump flow (mm)

T5o (sec)

Unit weight

Air content

(kg/mQ (%) SCC1 SCC2 SCC2* SCC3 SCC3* SCC4 SCC4 SCC5 rept. 1 SCC5 rept. 2 SCC5 rept. 3 SCC5 rept. 4 SCC5 SCC5 SCC6 SCC6 SCC6*

0.5 0.5 0.5 0.5 0.5 0.5

0.35 0.5 0.5 0.5 0.5 0.5

0.35 0.5 0.35 0.5

10 10 10 10 10 10 10 10 10 10 10 2 10 10 10 10

23.6

23.6

23.1

23.1

23.1

24.7

~

23.4

23.6

22.6

22.2

25.7

25.7

22.8

25.9

24.4

570 670 670 740 760 560 550 660 660 660 660 670 660 770 750 760

—

~

~

—

—

2.85

2.16

1.34

1.90

2.29

3.03

3.37

1.75

1.75

2.57

2.16

2,300

2,322

2,301

2,348

2,321

2,282

2,318

2,325

2,321

2,336

2,325

2,330

2,327

2,284

2,290

2,307

3.5 ~

~

1.5 —

4.3 3.6 —

2.5 2.3 ~

2.4 3.3 2.5 3.2 ~

98

Chapter 5: Development portable device to measure SCC formworkpressure

Table 5.2 (Cont'd) Fresh concrete properties

0.5 0.5 0.35

0.5 0.5 0.5 0.5 0.5

10 10 10 10 10 10 10 10

22.9

23.7

24.2

—

24.5

23.6

24.5

22.4

670 660 660 660 660 685 660 680

2.29

4.00

3.40

1.92

2.87

3.56

2.50

1.50

2,298

2,315

2,326

2,370

2,315

2,327

—

2,347

3.7 3.5 3.2 1.8 2.8 2.2 ~

2.5

Mixture Free H (m) R T § T50 p Air content (%) _ _ _ _ _

SCC8 SCC8 SCC9 SCC 10 seen SCC12 SCC13

Used in comparing the UofS2 pressure column to the 3-m high PVC column

5.5 Reduce concrete height in the UofSl pressure column

The results of decreasing the free concrete height from 1 to 0.5 m in the UofSl pressure

column were carried out. Typical SCC mixture was prepared to fill the UofSl pressure column with

0.5 and 1 m consequently before applying the overhead pressure to simulate 9-m high formwork at

10 m/hr. The pressure envelops for the two concrete heights are compared in Fig. 5.1 and variations

of their lateral pressure in time are shown in Fig. 5.2.

The results indicate that the UofSl pressure column of 0.5 m concrete height produced

maximum pressure versus apparent concrete height and variation of lateral pressure over time

identical to that cast up to 1 m height. These promising results led to develop a shorter pressure

column with that can receive only 0.5 m concrete plug.

5.6 Design of new pressure device

A new design was made and is referred to as UofS2 pressure column (Fig. 5.3). The UofS2

pressure column is circular in cross-section and measuring 0.7 m in height, 0.2 m in internal

diameter, and 10 mm in thickness. A pressure transducer set flush with the fresh concrete is

employed at a height of 63.5 mm from the base. Another transducer is fixed above the concrete at

625 mm from the column base to determine the overhead pressure inside the column. The lateral

pressure is monitored using pressure transducers having similar characteristics to those used in

the UofSl pressure column, except for greater capacity of 1380 kPa (200 psi). Numerical dial-

gauge (manometer) of high precision (Fig. 5.4) was attached via control chamber to the column

for better management of the air pressure flow on the top surface of the concrete.

99

Lateral pressure (kPa) 100 200 300

Hydrostatic |\ / pressure

Concrete \ height = 1.0 m

Fig. 5.1 Lateral pressure envelop

obtained using the UofSl pressure

column filled with SCC at 0.5 and 1 m

200

« 160

Concrete height = 0.5 m

Concrete height = 1.0 m

2 s 120 en

8 2 80 S3

i-

2 40 0

0 10 20 30 40 50 60 70 80 Time (min)

Fig. 5.2 Variations of lateral pressure in time using

the UofSl pressure column filled with SCC at 0.5 and

1.0m

Sensor <t> 19 mm

704 mm

Sensor <)> 19mm

T 190 mm

63.5 mm

577 mm

500 mm

63.5 mm

-I*—190 —4 !

Fig. 5.3 The UofS2 pressure column of 0.7-m high and 0.2-m

diameter to evaluate lateral pressure envelop and early pressure

decay of SCC mixtures

Fig. 5.4 Digital

manometer to control

overhead air pressure

100


Various methods for implementing the compressed air on the top concrete surface were

evaluated. Applying the pressurized air directly on the top of the concrete, inflating the air in a

balloon that pushes on the concrete, and use an isolating latex membrane separating the air from the

concrete were compared. Example of these comparisons between the balloon and latex membrane

is shown in Fig. 5.5. The results showed that all the configurations came out with similar lateral

pressure. Hence, the simplest configuration providing an overhead pressure was retained.

Lateral pressure (kPa)

0 40 80 120 160 200

Hydrostatic pressure

Blowing up air in a balloon

Fig. 5.5 Comparison between two methods of applying the overhead air pressure on the lateral

pressure envelops obtained using the UofS2 pressure column

5.7 Successive reduction of concrete plug in the UofS2 pressure column

Attempts to reduce the free concrete head in the UofS2 pressure column from 0.5 to 0.35 m

were conducted Three SCC mixtures (SCC4, SCC5, and SCC6) with initial slump flows of 560,

660, and 760 were tested with concrete heights of 0.5 to 0.35 m. The variations of the lateral

pressure with time for the tested mixtures are shown in Fig. 5.6. The column filled with 0.35 m

concrete produced higher lateral pressure at the conclusion of the pressure rise than that of 0.5 m.

However, both systems resulted in similar pressure decay up to a certain time, after which the

column of 0.35 m concrete exhibited sudden increase in lateral pressure. The lateral pressure

peaked to that of the overhead air pressure. This might be attributed to a "blow-up" at sensor

located at relatively small distance from the free surface. For the more flowable mixtures, such

101

sudden increase in lateral pressure was observed at later ages; 94, 108, and 118 min from the start

of casting for SCC mixtures with slump flows of 560, 660, and 760 mm, respectively.

/-" « Si 22 s i*i m

C a ^ 05 i . a>

• * *

« -J

240

200

160

170

80

40

0

280

§ 240

IT 200 | 160

£ 120 ss

I 80 J 40

0

280

J§ 240 £ 200 I 160

- 120

t 80 -** J 40

SCC4 <|> = 560 mm R= lOm/hr

,"W JN'

H = 0.35 m

H = 0.5 m

decrease in pressure reflects material restructuring

Sudden increase in pressure reflects "blow-up" of overhead pressure at sensor location

Time (min)

20 40

SCC5 <|) = 660 mm R=10m/hr

60 80 100 120 140 160

H = 0.35 m

H = 0.50 m

Time (min)

20


40 60 80 100 120 140 160

H = 0.35 m

H = 0.50 m

1 Time (min) 0 20 40 60 80 100 120 140 160

Fig. 5.6 Variation of lateral pressure in time obtained using the UofS2 pressure column filled

with 0.5 and 0.35 m concrete plugs for SCC mixtures of various slump flow values

102

The initial maximum lateral pressure (Pmax) along the form work height of 12 m, using the

UofS2 pressure column filled with 0.5 and 0.35 m free concrete heads are presented in Fig. 5.7.

The variations of Pmax with height are approximately hydrostatic at top three to four meters. At

greater depths, greater spread from hydrostatic pressure was noted. This spread was greater for

the less fluid SCC. Again, the UofS2 pressure column filled with 0.35 m concrete resulted in

higher lateral pressure at the end of the pressure rise compared at the 0.5 m.

0

0 '

2 ? r 4

igh

£ 6 u 8 a u i o

12

14

-

H =

-

-


= 0.50

100 200

1 1

SCC4 (j> = 560 mm

s R=10m/hr

\ \ x

A \ \ Hydrostatic x \ \ \ \ \ pressure

\ \ * \ \ * X \ * \ \ * \ \ * L \ * T \ * 1 I * * \ N \ \ %

* H = 0.35 m

300

i

\ »

0

2

w 4

JSP "3 6 •S

« 8

Con

cr

o c

12

14

Lateral pressure (kPa) 0 100 200 300

- ^y

-

-

H

-

i

V. **>

\ \ \

= 0.5m

SCC5 § = 660 mm R=10m/hr

V\

\ \ Hydrostatic y \ \Dressure \ \ * \ \ * \ i *

k 1 * \ 1 x

\ \ * \ \ * X x

H = 0.35 m


0

2

? r 4 .c W)

1 6

E 8 a

U 10

12 14

Fig. 5.7 Lateral pressure profiles obtained using the UofS2 pressure column having 0.5 and

0.35 m concrete plugs for SCC mixtures of various slump flow values

103

- \

-

-

' H =

-

SCC6 <* <|> = 760 mm * \ R=10m/hr

\N> \ \> V V

\ \* Hydrostatic \ \ \ pressure

= 0.50 m k ] \ / \ / \ / \ / % i \

H = 0.35 m


Based on these results, it was concluded that the UofS2 pressure column filled with 0.5 m of

concrete is more adequate to avoid the formation of "pressure blow-up" when simulating the

lateral pressure exerted by 13 m of concrete head.

5.8 The UofS2 pressure column vs. 3-m free standing PVC column

SCC2, SCC3, and SCC6 were cast in the UofS2 pressure column up to 0.5 m. Then the

column was sealed, and air pressure was gradually applied on the top surface of the concrete to

simulate a 3-m high concrete head. The concrete casting rate and rate of applying the overhead

air pressure were set to 10 m/hr. In parallel, a 3-m high PVC tube (Fig. 4.3) was cast

continuously with the same SCC mixtures at 10 m/hr. Variations of the lateral pressure over time

determined at the base of the two columns are shown in Fig. 5.8, while variations of the lateral

pressure along the concrete height for the two column systems are compared in Fig. 5.9.

Both pressure systems resulted in similar lateral pressure variations and relative values to

the equivalent hydrostatic pressure (Ko). The differences between Ko values at the 3-m casting

heights measured using the UofS2 pressure column and the 3-m free standing PVC column was

found to be 0.8% for SCC2 of slump flow of 660 mm that corresponded to major sector of SCC

in the concrete market. These differences in Ko values increased to 5.5% and 7.6% for SCC3 and

SCC6, respectively, of high slump flow values of 760 mm. In all cases, the UofS2 pressure device

resulted in the slightly higher Ko values compared to the PVC tube. This slight increase in Ko

obtained with the pressure device is not danger since it gives the engineers more safety when

designing the formwork systems for SCC.

More validations of the new pressure device with the real concrete casting are conducted by

the field measurements on wall and column elements of 4.4 m and 3.6 m in heights, respectively,

as will be discussed in Chapter 10. The field observations show 1:1 relationship between the

measured lateral pressure values in field to the predicted values obtained using the new pressure

column. It could then be concluded that the UofS2 pressure column can be used to determine

lateral pressure characteristics since it resulted in approximately similar values as those obtained

from the free standing concrete.

104


70 r Hydrostatic pressure

a V i-s V i-a a

• * *

« -

uu

50

40

30

70

10

9

a

s-



3-m PVC column

20 40 60 Time (min)

80 100

80

70

60

50

40

30

20

10

0

SCC3 § = 760 mm

"R=10m/hr


3-m PVC column

z^mzmk^Rmmnvw^


i i i

RRwVWA

i

20 40

'5s

% V u s t/>

a u a ^m

CU • * *

70

60

M)

40

30

20

10

0

60 80

Time (min)


100 120


3-m PVC column

SCC6 (j) = 760 mm R= lOm/hr

20 40 100 120 140 60 80 Time (min)

Fig. 5.8 Lateral pressure variation with time determined using the UofS2 pressure column versus

3-m free standing PVC column for different SCC mixtures

105

SCC2 (j) = 660± 13 mm R= lOm/hr


2-5 TJ0fS2 pressure column

3.0


20 40 60 80

SCC6 <|> = 760± 13 mm R=10m/hr




20 40 60 80

SCC3 (|> = 760±13mm R=10m/hr


Fig. 5.9 Lateral pressure envelops from UofS2 pressure column vs. 3-m free stand PVC column

5.9 Lateral pressure decay

Decay of relative lateral pressure over time Ko(t) until lateral pressure cancellation is

determined using the 1.2-m PVC column for SCC5, SCC6, SCC8, and SCC10. An example of

the pressure decay of SCC8 is presented in Fig. 5.10. The early pressure decays for the same SCC

mixtures were determined using the UofS2 pressure column after the conclusion of the simulation

of 13-m high concrete casting.

106


The early decay in Ko monitored using the UofS2 pressure column was compared to that

obtained from the 1.2-m high PVC column. An example for this comparison is shown in Fig. 5.11

for SCC8. The two pressure devices showed similar rates of pressure decay. Therefore, the 1.2-m

high PVC column can be effectively employed to determine the rate of pressure drop of SCC all the

way to pressure cancellation. The UofS2 column can also serve to evaluate the early pressure decay.

SCC8 (|) = 660± 13 mm R=10m/hr

0 60 120 180 240 300 360 420 480 540 600

Time (min)

Fig. 5.10 Decay of relative lateral pressure monitored using 1.2-m PVC column

SCC8 (|> = 660± 13 mm R=10m/hr

0 20 40 80 100 120 60 Time (min)

Fig. 5.11 Decay of relative lateral pressure monitored using 1.2-m high PVC column and the UofS2

pressure column; the latter monitored for one hour after the end of casting

107


5.10 Repeatability responses of the UofS2 pressure column

To determine the repeatability of the UofS2 pressure column, the lateral pressure

characteristics exerted by SCC5 tested four times were determined. The variations of lateral

pressure over time and the lateral pressure envelops are presented in Figs. 5.12 and 5.13,

respectively. The repeatability was evaluated statistically by relative error (RE). The RE is

calculated according to a 95% confidence interval using the Student's distribution (Eq. 5.1).

# £ = ^ . 1 0 0 (%)=3.1824z^=.100 (%) Eq. 5.1

where, 3.1824 = the coefficient representing 95% confidence interval for the Student's

distribution for n equal to four observations

SE: Standard error representing 95% confidence limit

a : standard deviation

n : number of observations

x : the mean value of the observations.

The RE estimated for the lateral pressure readings at various concrete heights are indicated

in Table 5.3. The RE was limited to ± 4% and are shown to increase with concrete head.

200 Experiment #4 Experiment

#1

/

SCC5 (|) = 660±13mm R=10m/hr

0 20 40 60 80 100 120 140 160

Time (min)

Fig. 5.12 Variations of lateral pressure in time for SCC5 used to determine the experimental error

of the UofS2 pressure column

108

0

2

S 4 .£f °3

•S «

= 8 o U

10

12

14

SCC5 <|> = 660 mm R=10m/hr

\ Hydrostatic vpressure

Experiment # 1

Experiment #2 'V

Experiment #3 l"*

Experiment #4

Fig. 5.13 Lateral pressure envelops of SCC5 used to determine the experimental error of the


Table 5.3 Repeatability of lateral pressure at various casting heights

H(m) Relative error (%)

1

4

8

12

±0.7

±2.4

±2.3

±4.0

5.11 Evaluation of key parameters affecting form pressure

The UofS2 pressure column was employed in a series of experiments to evaluate the effect

of six parameters affecting greatly the lateral pressure characteristics exerted by SCC on

formwork. The evaluation is elaborated below.

5.11.1 Effect of casting rate

The variation of lateral pressure of SCC5 cast at casting rate (R) of 2 and 10 m/hr using the

UofS2 pressure column is shown in Fig. 5.14 and the lateral pressure envelop along the simulated

12-m high formwork is presented in Fig. 5.15. Beyond the top 4 m height where the lateral

pressure corresponded approximately to hydrostatic pressure, the pressure column reflects well

the increase in lateral pressure with placement rate. For example, lowering the rate of rise from

109

10 to 2 m/hr decreased the relative lateral pressure (Ko) at 12 m from 61% to 49%. This is due to

the fact that concrete can have longer time to coagulate and build-up and hence increase its

cohesiveness and ability to support some of its mass with reduced lateral pressure on formwork.

R= 10 m/hr

R = 2 m/hr

SCC5 (|) = 660±13mm

0 60 120 180 240 300 360 Time (min)

Fig. 5.14 Variation of lateral pressure with time of SCC6 cast at 2 and 10 m/hr


SCC5 <(> = 660±13mm

R= lOm/hr

Fig. 5.15 Lateral pressure envelops of SCC5 cast at 2 and 10 m/hr

5.11.2 Effect of slump flow

Variations of lateral pressure during pressure increase in the column device to simulate

casting of 13 m of SCC with various slump flows (§) are presented in Fig. 5.16. Two groups of

110


SCC mixtures proportioned with and without VMA were tested. The lateral pressure envelops of

the tested mixtures are shown in Fig. 5.17.

As anticipated, SCC1 and SCC4 of 560-mm slump flows produced lower lateral pressure

compared to those of 660 mm and in their turn are less than that of 760 mm. Similar results were

reported by Assaad and Khayat [2006]. Unexpected lower lateral pressure for SCC3 of 760-mm

slump flow and no VMA than SCC2 of 660 mm was obtained. The HRWRA demand for SCC3

was increased to produce high fluidity of 760 mm. As a result, this mixture exhibited some

segregation especially it was not contain VMA. At the demolding of the pressure column, high

concentration of coarse aggregate was observed, which would increase internal friction and the

thixotropy of the concrete near the bottom of the pressure column where the pressure is determined.

200

« S3 N — '

ti u 9 i/>

!/> CL> U C

^ • 4

« s-

ate

-

160

120

SO

40

R= lOm/hr noVMA

SCC3 4 = 760 ±13 mm

SCC2 c|) = 660±13mm

10 20 30 Time (min)

40 50

200

2a i6o 2 i 120 u

u

-J

80

40

0

R=10m/hr VMA = 2.8 L/m3

SCC6 (|) = 760 mm

SCC4 (j) = 560 mm

SCC5 (() = 660 mm

0 10 20 30 Time (min)

40 50

Fig. 5.16 Increase of lateral pressure during pressure simulation of 8-m high column for SCC

mixtures of various slump flows (slump flow was increased by addition of HRWRA

111

Chapter 5: Development portable device to measure SCCformwork pressure


R= lOm/hr noVMA

Hydrostatic \ y pressure

660 mm

Lateral pressure (kPa) 0 50 100 150 200

0 «

1

12

-** .Sf 3

« 4

o -> U

6

7

8

- ^ R=10m/hr

^ VMA = 2.8 L/m3

^ Hydrostatic \ x pressure

SCC6 ^ ^ \ \ \ -e- 7 6 0 m m V * V \ SCC5 V \ x4> = 660 mm

_SCC4 ^-=\ \,)K 4> = = 560 mm \ \ \

\ \ * \ \ \

Fig. 5.17 Variation of lateral pressure envelops for SCC mixtures of various slump flows (slump

flow was increased by addition of HRWRA

5.11.3 Effect of VMA and HRWRA contents

The rise in lateral pressure with concrete head of the SCC2, SCC5, and SCC7 mixtures

containing VMA at contents of 0, 2.8, and 5.1 l/m3, respectively, that had initial slump flow of

660 ± 13 mm, are presented in Fig. 5.18. The lateral pressure envelops are presented in Fig. 5.19.

As expected, the SCC5 with 2.8 l/m3 VMA exerted lower lateral pressure compared to that

without any VMA. Adding more VMA from 2.8 to 5.1 l/m exhibited in the inverse action with

increased lateral pressure. This can be attributed to the coupled effect of VMA and HRWRA

demands where the SCC7 made with higher VMA concentration resulted in greater HRWRA

demand. The HRWRA demand for the SCC7 mixture was 1.63 times greater than that of SCC5.

The higher demand of HRWRA can lead to lower rate of restructuring at rest, thus maintaining

greater pressure in time. This was compatible with findings reported by Khayat and Assaad

[2006] who found that SCC with low VMA dosage can develop less lateral pressure than those

without any VMA, or those with relatively medium or high VMA concentrations.

112

250

200

s -

I 150 u a « 100

- 50

0

{))= 660 ±13 mm R=10m/hr

Wv,

M 1

SCC2 *^ VMA = none

SCC5 VMA = 2.8 1/m3

i i

SCC7 VMA = 5.1 1/m3

i i

0 30 120 150 60 90

Time (min)

Fig. 5.18 Increase of lateral pressure during pressure simulation of 13 m height cast with SCC

mixtures with different VMA concentrations

J3 4

.SP

v 6

S 8 U


\ >

-

SCC5

i i i

<|> = 660 ± 13 mm R=10m/hr

^ Hydrostatic ^KN pressure

\ \s>

\ \ \ x scc2 ^ \ \ \ \ no VMA

„VMA = 2.8 l/m3\ \ \ v ^ T \ / * I \ / x

SCC7 VMA =

\ \ J *

= 5.1 1/m3

10

12

14

Fig. 5.19 Lateral pressure envelops for SCC mixtures containing different VMA concentrations

5.11.4 Effect of paste volume

SCC5, SCC8, and SCC 10 mixtures with paste volumes (Vp) of 340, 370, and 400 1/m3,

respectively, were selected to evaluate the ability of the UofS2 pressure column to reflect the

influence of paste content on form pressure. The variation of lateral pressure with casting time is

presented Fig. 5.20 for the three SCC mixtures, while profiles of maximum initial pressure along

113

the casting height are presented in Fig. 5.21. SCC with higher Vp was shown to exert greater

lateral pressure. This was due to the increase in Vp, which is accompanied with a reduction in

aggregate content. As a result, internal friction and shear strength of the plastic concrete are

reduced with the increase in Vp, as reported by Assaad and Khayat [2005 C].

320

cs Sa ^ • s

M)

?40

200

160

120

80

40

0

§ = 660 mm - R= lOm/hr _

_

r^ f™Tv *

Q***

SCC 10 j / Vp = 400 1/m3

SCC8 Vp = 340 1/m3

1 l I

SCC5 Vp = 370 1/m3

i i i

0 20 40 60 80 100 Time (min)

120 140 160

Fig. 5.20 Increase of lateral pressure during the simulation of casting of 13-m high column using

SCC mixtures of different paste volumes

0

2

ja ^ •SP ' 3 •a 6 <u

a 8

o y i o

12

14


(|) = 660 mm R=10m/hr


SCC 10 \ V p = 400 1/m3

ISCC8 Vp = 340 1/m3

~SCC5 Vp = 3701/m3

Fig. 5.21 Variation of initial lateral pressure with concrete height for SCC mixtures containing

various paste volumes

114


5.11.5 Effect of maximum-size of aggregate

Variations of lateral pressure with time for SCC13, SCC5, and SCC9 mixtures

proportioned with various maximum-size of aggregate (MSA) of 10, 14, and 20 mm,

respectively, using the UofS2 pressure column are presented in Fig. 5.22. The lateral pressure

variations are shown in Fig. 5.23. The increase of MSA from 10 to 14 and further to 20 mm

reduced the lateral pressure at the end of simulation of 13-m height casting from 209 to 173, and

then to 148 kPa, respectively. Similar reductions in lateral pressure envelop were observed when

the MSA increased from 10 to 14, and then to 20 mm. Packing density of coarse aggregate and

HRWRA demand relating to the each MSA are the two keys responsible for the reduction in

lateral pressure. Higher packing densities of 63% for 20-mm MSA vs. 62% for 14-mm MSA, and

of 56% for 10-mm MSA, might impedance the mobility of the concrete and increase the inter-

particle friction, hence resulting in lower lateral pressure. Also, larger MSA can cause arching

inside the concrete structure, which reduces the lateral pressure. It is worth noting that the

HRWRA concentrations were adjusted to secure the target slump flow of 660 mm, and this could

affect the resultant lateral pressure. Therefore, the obtained lateral pressure results reflect not only

the effect of MSA, but also the HRWRA demand.

0 20 40 60 80 100 120 140 160 Time (min)

Fig. 5.22 Variation of lateral pressure with time for SCC mixtures proportioned with different

MSA (slump flow of 660 ± 13 mm)

115

0

2 ? •M A JA * .Sf '3 •= 6 <u

2J = 8 o V

10

12

14

(|) = 660 ±13 mm R=10m/hr

Hydrostatic v\ pressure

SCC9 ^ \ \ \ SCC 13 (MSA = 20 \ \ V (MSA = 10

mm) ~ ^J \ \? \ mm)

SCC5 _(MSA=14mm)

Fig. 5.23 Variation of initial lateral pressure with concrete height for SCC mixtures proportioned

with different MSA (slump flow of 660 ±13 mm)

5.12 Conclusions

Based on the above results, the following conclusions can be attained:

• A suitable and field-compatible pressure device to evaluate the lateral pressure development of

SCC was developed. The UofS2 pressure column measures 0.19 m in internal diameter and 0.7

m in height can be filled with 0.5-m high concrete sample then pressurized with an overhead

air pressure to simulate equivalent concrete head of up to 13 m.

• The lateral pressure characteristics of the UofS2 pressure column filled with 0.5 m and

overhead pressure to simulate a head of concrete of 3 m are found to be comparable to those

obtained from a PVC column of the same diameter filled with 3 m of fresh concrete.

• The UofS2 column can be used to evaluate early decay in lateral pressure, up to two hours after

the end of casting simulation.

• Good repetition can be obtained with the UofS2 pressure column.

• The UofS2 pressure column was validated using SCC mixtures made with various material

characteristics, mix design parameters, and cast at different placement rates. The findings from

the pressure device are comparable to those reported in the literature.

116

CHAPTER 6

FIELD-ORIENTED TEST METHODS TO EVALUATE STRUCTURAL BUILD-UP AT

REST OF SCC

6.1 Introduction

The rate of restructuring of concrete has considerable influence on lateral pressure

characteristics. Concrete exhibiting faster degree of restructuring will develop greater

cohesiveness after casting, thus exerting less lateral pressure on the formwork system. The

restructuring phenomenon is considered to be due to the development of internal friction and

attractive forces among solid particles at rest as well as to an increase in the degree of physical

and chemical bond during cement hydration [Assaad et al., 2003A]. The longer the concrete is

left at rest, the higher the build-up of internal structure and shear strength would be. The

restructuring was indirectly assessed by determining the breakdown area: the energy needed to

breakdown the interior structure of the material after a certain period of rest to reach an

equilibrium state [Assaad et al., 2003A]. The restructuring at rest was evaluated by Roussel and

Ovarlez [2005], where the authors defined the structural build-up at rest as "thixotropy factor

(Athix, Pa/s)", which was the rate of variation of static yield stress with resting time. Billberg

[2006] also utilized the time-dependant change of static yield stress at rest after a given resting

period [xs(t), Pa/min] to estimate the structural build-up at rest of SCC.

The structural build-up at rest can be assessed using concrete rheometers. However, it is

necessary to develop economical, portable and field-oriented test methods to assess structural build

up at rest of concrete. Six field-oriented test methods were initially developed and employed by a

team of researchers at Universite de Sherbrooke to evaluate the structural build-up at rest of SCC

[Roby et al., 2006]. The selected tests include portable vane (PV) based on the hand-held shear-

vane test used to determine the undrained shear strength of soil, the inclined plane test (IP), the

undisturbed spread test (US), a standard cone penetration test (Swedish cone penetration test), the

K-slump tester, and a sinking-ball test. This chapter will focus on the PV and IP tests. These tests

were recommended for SCC based on the repeatability results on thixotropic and non-thixotropic

mixtures that used to establish the relative error for each test method [Khayat et al., 2008].

117

Chapter 6 Field-oriented test methods to evaluate structural build-up at rest of SCC

6.2 Research significance and objectives

Appropriate determination of structural build-up at rest of freshly cast SCC in field

environment before placement is essential for quality control of SCC deliveries, predict the

lateral pressure exerted on formwork, optimize formwork use and cost, and ensure fast placement

associated with cast in-place using SCC. Simple field-oriented test methods reported in this

chapter can enable the evaluation of structural build-up at rest of SCC, and consequently

determine the lateral pressure developed by SCC. This can be beneficial to contractors and

engineers to select proper mixtures that exhibit lower pressure.

This chapter details the PV and IP test methods and elaborates the test protocol for each test

method to quantify structural build-up at rest. The repeatability of these tests is also determined

for SCC mixtures of various thixotropy levels. The ultimate goal of this chapter is to validate the

results of the PV and IP test methods with those obtained from the modified Tattersall MK-III

concrete rheometer using different SCC mix designs.

6.3 Testing program

Summary of the testing program undertaken in this chapter is shown in Table 6.1. Full

description of the PV and IP test methods are introduced. The repeatability characterizations of

the test methods were evaluated using SCC mixtures of different degrees of thixotropy: SCC40

(low thixotropy) and SCC46 (high thixotropy) (Table 3.8). Each SCC was repeated five times.

The resting time between measurements was 15 min at 15, 30, 45, and 60 min for the PV test.

This time was only 5 min for the IP test (measurements performed at 15, 20, 25, and 30 min).

In total, 42 measurements on SCC9, SCC12, SCC16 to SCC32, six repetitions of SCC33,

SCC40, SCC46, SCC51 to SCC56, five repetitions of SCC62, two repetitions of SCC63, and two

repetitions of SCC64 (Table 3.8) were used to compare the field-oriented test methods with the

rheological values obtained with modified Tattersall MK-III concrete rheometer. The mixtures were

chosen to cover wide range of thixotropy levels and had initial slump flow values varying from 560

to 720 mm. The initial and the time-dependant responses determined simultaneously from the field-

oriented test methods and concrete rheometer were correlated.

Concrete temperature, slump flow, T50, unit weight, and air volume were determined for the

tested mixtures and reported in Appendix Bl. The required HRWRA concentrations to produce

the target slump flow are shown in Table 3.8.

118

Chapter 6 Field-oriented test methods to evaluate structural build-up at rest ofSCC

Activity

Repeatability on mixtures

sec

Table 6.1

Variables

Summary of experimental

five repetitions on each SCC

program

Mixtures

SCC40 (low thixotropy) SCC46 (high thixotropy)

1. Correlations between initial responses 4 0 c r i r , • . _,. , 1 , ~ t , , fz J L L mixtures i

A. Field-oriented vs. iield-onented n i •• u • B. Rheometerxorest vs. field-oriented Z,Z,Z?' „}„' Correlations between *• IJ • tA SCC16-SCC32; r- , , • . , . . C. AriaDD(a) o 7 ms vs. iield-onented . . . ' „ field-oriented test }apm°'^s

A • . , six repetitions of SCC33; *u A A t D- Abi vs. iield-onented „^„„\v „ ^ ^ . ^ methods and concrete ~ n \ t- u . ,. , . 4 SCC40, SCC46; , 2. Correlation between time-dependent „^,„ ' „ , , „ '

rheometer K SCC51-SCC56; change of the responses _ . . ^„^,^,^^

A. Field-oriented vs. field-oriented Ave repetitions of SCC62; B. Rheometerx0rest vs. field-oriented t w o ™V««]°™ ° "CC63; and ^ A r- , , • . , two repetitions of SCC64 C. Ar|[email protected] vs. field-oriented ^

6.4 Field-oriented test methods to evaluate structural build-up at rest of SCC

Detailed description of the PV and IP test methods are discussed below.

6.4.1 Portable vane

A. Background

Bauer et al. [2007] reported that the vane test method (Fig. 6.1) has been developed in soil

mechanics to determine a parameter defined as "undrained shear stress soils" (in particular clay

soils). Seed and Riemer [lecture] used a vane shear tip of 75 mm in diameter and 112 mm in

height in field tests; typically the vane's height is double its diameter. They introduced the vane

into the borehole to the depth where the measurement of the undrained shear strength is required.

Then, the vane is rotated at a specified rate that should not exceed 0.1 degree per second

(practically 1 degree every 10 sec) using a protractor fixed on the ground surface, and the

torsional force required to cause shearing is measured using a torque-meter. The shear strength of

the soil was calculated as the measured torque divided by a constant K, which depends on vane's

dimensions and shape. The test procedure and equipments are described in ASTM D2573-72.

Bauer et al. [2007] reported that the vane test has been used in rheology studies of material in

different fields. The vane test has proven to be a simple and effective method in measuring non-

Newtonian fluid properties since they have a flow on smooth surfaces and are common in devices

used in different types of rheometers (parallel disc rheometers or coaxial cylinders). The value of

the yield stress obtained by the vane test coincides to a great extent with the majority of currently

available rheological methods [Nguyen and Boger, 1985] and [Barnes and Nguyen, 2001].

119


Fig. 6.1 Vane shear test used in to determine undrained shear strength of clay soils

Roussel and Cussigh [2008] carried out two rheological tests on concretes: slump flow test

and scissometer or vane shear test. The geometry of the scissometer used in this work [Fig.

6.2(left)] was designed according to the recommendations of Nguyen and Boger [1985] considering

size of the constitutive particles of the mixture [Fig. 6.2(right)]. The apparatus records the highest

torque needed to initiate flow after a time of rest. The vane and protocol of measurements are

similar to the apparatus used by Assaad et al., [2003A] or Billberg [2005] to measure the apparent

(or static) yield stress. The highest torque measured is thus either proportional to the initial yield

stress of the material (too) if the test is carried out just after mixing or to the apparent yield stress of

the material [xo(t)] after a time of rest t. The repeatability of the static yield stress measurement was

estimated to be around 15% [Roussel and Cussigh, 2008]. The experimental measurements were

carried out on four SCC mixtures shown in Table 6.2 and Fig. 6.3. It has to be noted that all

scissometer measurements were carried out on "virgin" concrete samples.

Fig. 6.2 Scissometer or vane shear test: (left) geometry and (right) measurement in circular

bucket filled with concrete [Roussel and Cussigh, 2008]

120


8000 X 9CC M-1

a sec NU2

0 SOC ST-1

60 83

Time (mill)

140

Fig. 6.3 Static or apparent yield stress as a function of resting time [Roussel and Cussigh, 2008]

Table 6.2 Rheological measurements [Roussel and Cussigh, 2008]

Measurements SCC No. 1 SCCNo.2 SCCNo.3 SCCNo.4

Initial slump flow (mm)

Slump flow after remixing at 80 min (Pa)

Initial yield stress, too (Pa)

Structural rate, Athix (Pa/s)

Structural rate, Athix (Pa/min)

630 630

54

0.12

7

670 —

40

0.36

22

700 450

50

0.47

28

630 ~

70

1.14

68

B. Development of the PV test

In 2006 [Roby et al.], the standard hand-held vane test for soils was used to determined

static yield stress on mortar. Four-blade vanes of different sizes (Table 6.3) were designed to

enable the shearing of material at different rest times. The blades were designed then modified

using stainless steel to enable the use of a torque-meter to capture shear strength of the plastic

concrete (Figs. 6.4 and 6.5). The largest vane is used for the weakest structure, i.e., shortest

resting time, and the smallest vane for the strongest structure, i.e., the longest resting time. Four

square-shaped moulds were employed to avoid full rotation of relatively stiff SCC in the mould,

or a plug flow. Three torque-meters of different configurations and precisions were tried to

capture the torque values needed for SCC (Fig. 6.6) during the development of the PV test. The

one of high precision shown in Fig. 6.6 (c) was selected.

Immediately after mixing, the four vanes are centered vertically in the containers. Then, the

containers are filled with the tested mixture to a given height (h) indicated in Table 6.3 and Fig. 6.5.

The rested material is protected from evaporation with plastic cover. Middle hole of a diameter of 2

121


mm greater than the vane's shaft diameter is drilled in the plastic cover to help maintaining the

vane vertically. It is worth noting that the conditions to obtain accurate measurements are a flat

surface of vanes, vertical placing of the vanes in tested samples, and turning the torque-meter with

a constant speed. The development was achieved by the collaboration of Ms Siwar Naji, Master

student, Concrete Group, Universite de Sherbrooke, [Naji, MS thesis 2009].

Table 6.3 Vane dimensions

Vane

Vane # 1 (large vane)

Vane #2

Vane # 3

Vane # 4 (small vane)

Time at rest (min)

15

30

45

60

Vane dimensions (mm)

R H

37.5

37.0

37.5

37.5

250

200

149

100

h (filling height)

Varies from 50 mm for highly thixotropic SCC mixture to total vane height (H) for relatively low thixotropic mixture

(a) cylinderical (b) squared bucket buckets

Fig. 6.4 Four buckets and four vanes used in the PV test Fig. 6.5 Schematic of PV test

(a) (b) (c)

Fig. 6.6 Torque-meter used in the PV test

C. Calculation of static yield stress from the PV test

The test protocol is presented in Appendix B2. The maximum torque needed to breakdown the

structure is noted. This value is converted to static shear stress (torest) using Eqs. 6.1 and 6.2.

122


^Orest 1 G

where: G=27tr2 h + 1

[Khayat and Yahia, 2006]

Eq. 6.1

Eq. 6.2

T = measured torque (N.m), h and r are defined in Fig. 6.5 (in m).

Variations of static yield stress (xorest) determined using the PV test [PVxorest] with resting

time for typical SCC mixtures designed with different thixotropic characteristics are shown in

Fig. 6.7. The torest obtained from the PV at 15 min (PVxorest@i5rain) corresponding to the initial

response was used as a structural build-up index [T.I.@i5rajn]. Similarly, the rate of change of

static yield stress with time [PVxorest(t)] or [T.I.(t)] and the coupled effect of [[email protected].(t)] can

also be used as structural build-up indices.

4000

£ 3000

o2000

> * 1000

0 15 30 45 60 75 90 105 Resting time (min)

Fig. 6.7 Variations of static yield stress at rest with time obtained with portable vane (PV) test

6.4.2 Inclined plane

A. Background

Evaluation of viscoelastic flow on an inclined plane (IP) is an important idealization in a

variety of natural situations (avalanches, mud slides, liquefaction, etc.) and other viscous

materials such as commercial gels, bentonite slurries, and cementitious mixtures [Coussot et al.,

1996]. The principle of the IP test can be explained by the one-dimensional analysis. When a

mass 'm' is kept on a plane lifted by an angle '6 ' , as shown in the free body sketch shown in Fig.

6.8, two external forces affect downward momentum of the system: (i) gravitational force (m g

sin 9) and (ii) frictional force (fk) are existed. The force initiating the flow is the difference

between these two forces (m g sin 0 - fk), where g sin 9 is an acceleration due to the gravity

123


resolved along the IP. The fk shown in Fig. 6.8 exists between the bottom of the mass 'm' and the

top surface of the IP when the entire mass begins to slide downwards. However, when using an

IP of roughened surface, the physical sliding of the material is avoided in order for the material

could be shear. Thus, the mobility of the flowing mass sheared at a particular inclination, only the

gravitational force (mg sin9) activates the downward flow. The downward movement of the

sheared mass of cementitious material on the IP at an angle 9, and an effective height h due to

gravitational acceleration is shown in Fig. 6.9 [Oremus and Richard, 2006].

Fig. 6.8 Free-body diagram of mass 'm' kept on an inclined plane with slope angle '0 ' [Oremus

and Richard, 2006]

Fig. 6.9 Sketch showing the movement of cementitious material along the inclined plane at the

critical angle '9 ' [Oremus and Richard, 2006]

B. Development of the IP test for SCC

The development of the IP test was carried out with the collaboration of Dr. Pavate, a

research professor, Concrete Group, Universite de Sherbrooke. The tested IP model was an

advanced version of the test protocol elaborated by Pavate and Khayat in 2006 [personal

communication] As mentioned earlier, the surface of the inclined plane needed to be treated to

avoid any slipping or sliding. Initially, in one of the trials the inclined plane was roughened on

the entire surface of the plane with commercial resin impregnated with fine Ottawa sand.

124


Assuming uniform roughness was difficult given the quick drying and hardening of resin.

Another approach was taken to use commercial water-proof sand papers. After a number of

attempts, superfine sand papers having grit number of 600' was selected to ensure proper

performance. The grit is made of aluminum-oxide abrasive material that is water proof. The

sheets of sand papers are changed after 10-15 tests depending on the surface condition. The sand

paper is fixed on the top of PVC surfaces of inclined planes using adhesive tape, as shown in Fig.

6.10. Schematic detail of the IP giving all dimensions is shown in Fig. 6.11. A small convenient

protractor can be held or fixed between the joint of the horizontal and inclined plates to measure

the inclination angle of the plane.

SCC concrete specimen

Sand paper covered

inclined plans

ISOMETRIC VIEW

FRONT VIEW

Fig. 6.10 Inclined plane test at

different rest times: increase in flow

distance and critical angle

610 mm

r 400 m|m

1

PVC Plate 10 mm thick

C top & lower plate

Surface covered by sand

paper (No. 600) to avoid

slipping and sliding

Fig. 6.11 Schematic for the IP test

PLAN

Several attempts were made to select a sample holder to ensure proper spread and height of

the spread. Finally, transparent plexiglass cylinder with 60-mm diameter, 120 mm in height, and

5-mm thick, opened at both ends was selected. Mortar is filled to a specified height of 100 mm,

while concrete is filled up to the top of cylinder (i.e. 120 mm height). The use of different heights

for fluid mortar and SCC is for practical convenience so that the spread of flowing mortar and

concrete after lifting the cylinders would be within the width of sand paper on the inclined plane.

' Grit is defined with reference to the number of abrasive particles per inch (25.4 mm) of sand paper. Lower grit number indicates higher roughness of the sand paper and conversely higher the grit number the smoother is the sand paper.

125

Chapter 6 Field-oriented test methods to evaluate structural build-up at rest o/SCC

C. Calculation of static yield stress from the IP test

The corresponding angles necessary to initiate flow is used to determine the static yield

stress, IPtorest (Pa), as follows:

IPtorest = p g h sin 0 Eq. 6.3

where: p: density of mixture (concrete or mortar) (g/cm3);

g: gravitational acceleration (=9.81 m/s );

h: characteristic mean height of spread for tested mixture at horizontal position (mm) (step 5); and

6: critical angle of the plane when the sample starts to flow (degree).

Variations of xorest obtained from the IP test (IPxorest) with respect to the time of resting for

typical SCC mixtures of various rates of structural build-up are shown in Fig. 6.12. Similar to the

PV test, three structural build-up indices can be obtained using the IP method: [T.I.@i5min],

[T.I.(t)], and [[email protected].(t)].

1000 r

i? 800 L y*"

0 I _ _ . i i i i — - — _ _ — i

0 10 20 30 40 50 Resting time (min)

Fig. 6.12 Variations of static yield stress at rest with time obtained with inclined plane test

6.4.3 Evaluating repeatability of field-oriented test methods

Two SCC mixtures were prepared and tested five times each in two series: SCC40 (low-

thixotropic mixture) and SCC46 (high-thixotropic mixture), Table 3.8. The resting time between

measurements was 15 min (i.e., rest periods of 15, 30, 45, and 60 min) for the PV test. The

resting time for the IP test was only 5 min (at 15, 20, 25, and 30 min) in order to enable sufficient

readings before the stiffening of the thixotropic mixture.

The repeatability was evaluated by calculating the coefficient of variation (COV) calculated

according to Eq. 6.4 and relative error (RE) calculated according to a 95% confidence interval

using the Student's distribution (Eq. 5.1).

126


COV = - 1 0 0 x

Eq. 6.4

where, a= standard deviation and 3c = the mean value of the observations.

The statistical values (3c, a, COV, and RE) for SCC46 and SCC40 are presented in Table

6.4. The RE for the two field-oriented tests, are also shown in Fig. 6.13. For the repeatability series

using the thixotropic mixture, the RE for the PV and IP methods were 19% and 8.5%, respectively.

These values increased to 28% and 9%%, respectively, for the non-thixotropic mixture.

Table 6.4 Repeatability of field-oriented tests on thixotropic concrete mixture: SCC46

X

a

COV (%)

RE (%)

3c

a

COV (%)

RE (%)

3c

a

COV (%)

RE (%)

3c

a

COV (%)

RE (%)

Values could not

Rest time (min)

15

30

45

60

PVtores, (Pa)

SCC46

1567

189.5

12.1

19

2974

221

7.4

12

4337

189

4.4

9

NA**

SCC40

411

67.6

16.5

20

706

74.3

10.5

13

1007

128.7

12.8

16

1573

350.3

22.3

28

oe determined due to stiffening o

Rest time (min)

15

20

25

30

• concrete

IPtOrest

SCC46

476

32.4

6.8

8.5

535

34.5

6.4

8

594

40

6.7

8

654

25

3.8

4.7

(Pa)

SCC40

285

15.8

5.6

9

321

9.2

2.9

4.6

357

3.0

0.8

1.3

394

4.7

1.2

2

** Only three vanes vv

6.5 Correlating field-oriented method responses to rheometric measurements

The values of static yield stress at rest were measured for 42 SCC mixtures covering a wide

range of thixotropy levels using the PV and IP field-oriented test methods (PViorest and IPxorest,

respectively). At the same time, static yield stress, drop in apparent viscosity at rotational frequency

of 0.7 rps (determined using Eqs. 2.71 and 2.72.), and initial breakdown area (described in section

2.3.4) (Rheometerxorest, Ar|app@N=o.7rps, and Abi, respectively) were determined using the concrete

127


rheometer. These properties were measured initially at 15 min time of resting as well as with

resting time. The initial measurements and the time-dependent change of these rheological

properties are referred to as thixotropy indices (T.I.) or structural build-up at rest indices. The time-

dependent change of a rheological property is important, since two SCC mixtures can have similar

initial rheological properties, but have different rates of internal structural build-up at rest. The

time-dependent change of a rheological response is determined as the slope of variation of this

response with rest time.

100

Co 80 0 s

2 60 u

'•C cs

§ 2 0

0

Thixotropic mixture: 0 Portable vane

A Inclined plane

1 1 !

SCC46

i i

100

~ 80

£ 60 Lm i -V

« 40 «

I 2 0

0

Non-thixotropic mixture: SCC40 0 Portable vane A Inclined plane

10 20 30 40 Resting time (min)

50 20 40 60 80 Resting time (min)

Fig. 6.13 Relative error of field-oriented tests on concrete mixture

The initial and time-dependant change of static yield stress obtained from the PV and IP test

methods were compared to those obtained from the modified Tattersall MK-III concrete rheometer.

As the rheological responses are highly dependent on shear history, only responses determined

simultaneously were considered in this comparison. For highly thixotropic mixtures, the IP test was

terminated after the second or third measurement due to reaching the maximum inclination (9 = 90

degrees). Thus, the data points available for comparison were reduced for this test. Also, the Abi

was measured once as it requires 30 min to carry out the test at four rotational velocities compared to

a minute to perform the field-oriented tests. The Abi was measured only for 15 SCC mixtures.

6.5.1 Correlation between initial responses

A. Correlations between initial static yield stress determined from PV and IP tests

The static yield stress at rest measured at 15 min time of rest (R @15 min) obtained using the

portable vane test (PVTorest@i5min) was compared to that determined from the inclined plane test

128


(IPxorest@i5min) in Fig. 6.14. Similar correlation for the responses measured between 15 and 60 min

resting times (Rj @15-60 min) are also shown in the figure. The values of the coefficient of

correlation (R2) determined from the regression line was found to be 0.71 for the Rj @15-60 min

were taken into account, but 0.82 for the Rj @15 min measurements were considered. The static

yield stress resulted from the IP test is lower than that obtained from the PV test. This difference

became less when considering only the R, @15 min measurements. This difference can be referred

to the basic change in the shear phenomena undergoing in the two tests, since two different

instruments can lead to different shear stress values [Moller et al., 2006]. However, there is a good

consistency in the results using the two methods.

2400

2000

§,1600

J1200 > * 800

400

0

(

Measurements from 15 • o

oo

y = 1.62x #° R2 = 0.71 8 o y

o \ ° / * 0 o / » b o

" ^P^° i i i

) 300 600 900 IPT0rest(Pa)

61

o

o o o

) min

3 °

O

1

1200

1800

1500

^1200

~ 900

> 600 CM

300

0

Measurements at 15 min o

y = 1.23x R2 = 0.82

300 600 900 1200 IPT0rest(Pa)

Fig. 6.14 Correlation of static yield stress determined from portable vane and inclined plane tests

B. Correlations between initial static yield stress determined from field-oriented and

rheometric test methods

Two correlations between PVxorest and Rheometerxorest determined initially at 15 min time of

rest and between 15 and 60 min time of rest are presented in Fig. 6.15. The two correlations

resulted in R values of 0.82 and 0.93, respectively. Similar correlations between the IPxorest and

Rheometerxorest are shown in Fig. 6.16 that had R values of 0.82 and 0.78, respectively. The

relationship between the PVxorest and Rheometerxorest was 1:1 when comparing the measurements

taken place at 15 min time of rest. This relationship deviated from unity when the responses at

later resting times were introduced, given the fact that the shear history of the two tests were not

the same. In the PV test, the measurements were determined on four virgin, undisturbed samples,

129


while for the concrete rheometer, the measurements were carried out on the same sample sheared

after four resting times over 60 min. The samples were manually rehomogenized for one minute

after each measurement and left to rest until the following measurement (typically 15 min of

rest). The IPxorest is slightly lower than that obtained from the concrete rheometer. The difference

was larger when the measurements at all resting times (between 15 and 60 min) were regarded.

9000

7500

£ 6 0 0 0

J 4500

£ 3000

1500

0

easurements from 15 •

y = 1.57x o R2 = 0.93 ojr

o°T

°Jr

o

-60 min 2500

2000

CM

^ 1500

Ore

> 1000

500

Measurements at 15 min

y = l.OOx R2 = 0.82

2000 4000 6000

Rheometer T0rest (Pa) 0 1000 2000 3000

Rheometer T0rest (Pa) Fig. 6.15 Correlation between static yield stress determined from PV and concrete rheometer

Measurements from 15 - 60 min Measurements at 15 min

p.

1200

900

|600

300

0

y = 0.85x y R2 = 0.78 / °

«°

i i i

/ ° o

1 1

0 400 800 1200 1600 2000

Rheometer T0rest (Pa)

0 300 600 900 1200 1500 Rheometer x0rest (Pa)

Fig. 6.16 Correlation of static yield stress determined from inclined plane and concrete rheometer

C. Correlations between initial static yield stress determined from field-oriented tests and

initial drop in apparent viscosity at 0.7 rps determined from concrete rheometer

RheometerAr|[email protected] values are compared to PViorest (Fig. 6.17) and to the IPxorest (Fig.

6.18). For the initial measurements undertaken at 15 min time of rest and those obtained between

130

15 and 60 min time of rest, the comparison between the RheometerAr|[email protected] and the PVxorest

values resulted in R2 values of 0.81 and 0.98 for, respectively. These values were found to be 0.67

and 0.74 for the comparisons between the RheometerAr|[email protected] and IPxorest, respectively. The two

regression lines for the two correlations betweenAnapp@N=o.7rps and PVxorest were forced to origin.

8000 Measurements from 15-60 min 2500 Measurements at 15 min

J 6000

J4000 > CM

2000

0

(

"O

)

AT|

/ y = 3.78x $r R2 = 0.98

1 1

1000 2000

app@N=0.7rps ( " a - s )

1

3000

1? ft*,

a u

0 H

> PM

2000

1500

1000

500

0

-

8 0 /

0 /

O /

/ ° f O

y = 3.71x R2 = 0.81

1 1

0 500 1000

'^tlapp@N=0.7rps (P a -S )

Fig. 6.17 Relationship between static yield stress determined from the portable vane test and

drop in apparent viscosity at 0.7 rps obtained using concrete rheometer

1200

1^900 MM

Bre

st

H — -

300

0

Measurements from 15 - 60 min r ° °y

0 0 s^

0 0 ?S° ig9£¥) y = 1.18x + 251

mS8g° R2 = 0.74 fF°

1 1 1 1 1

1200

1000

J 800

S 600

& 400

200

0

0 200 400 600 800 1000 A^appfffi^OJrps ( P a . s )

Measurements at 15 min

= 1.23x + 263 R2 = 0.67

0 200 400 600 Anapp®N=0.7rps ( P a - S )

Fig. 6.18 Relationship between static yield stress determined from inclined plane test and drop in

apparent viscosity at 0.7 rps obtained using concrete rheometer

D. Correlations between initial static yield stress determined from field-oriented tests and

initial breakdown area determined from rheometric test method

For 16 SCC mixtures, the PV"Uorest@i5min values are compared to the Abi values in Fig. 6.19,

while the IPxorest@i5min values are compared to the Abi values as shown in Fig. 6.20. Good

131


correlations are obtained between the Abj and PVTorest@i5min and IPTorest@i5min with R2 values of

0.87 and 0.70, respectively.

2500

2000

CM 1500

£ 1000 CM

500

0 300 600 900 1200 1500 Abi (J/m3.s)

Fig. 6.19 Relationship between static yield

stress determined from portable vane test and

breakdown area obtained using concrete

rheometer

1200

1000

/^. « 800

est

h» ~ " " o H & 400

200

0

-

-

o

-°oY °s

!

0 200

O y

!

400 Abx

o /

s o 'y = 1.35x R2 = 0.70

i I I

600 800 1000 (J/m3.s)

Fig. 6.20 Relationship between static yield

stress determined from inclined plane test and

breakdown area obtained using concrete

rheometer

6.5.2 Correlation between time-dependent responses

A. Correlating time-dependent static yield stress determined from field-oriented tests

The correlation between the time-dependent static yield stress obtained from the PV and IP

test methods [PVxorest(t) and IPxorest(t), respectively] had high R value of 0.85, as shown in Fig. 6.21.

B. Correlating time-dependent static yield stress determined from field-oriented tests and

rheometric test

The relationships between PVTorest(t) and IPtOrestO) versus that obtained using the concrete

rheometer [Rheometerxorest(t)] are shown in Figs. 6.22 and 6.23, respectively. High R2 values

were obtained for the PV and IP tests of 0.96 and 0.93, respectively. These relationships are not

1:1 due to the differences in the shear histories of these tests. The PV test resulted in higher rate

of change in static yield stress than those obtained with the concrete rheometer. This is due to the

fact that the SCC in the concrete rheometer was disturbed between readings, which was not the

case for the PV test.

132

Chapter 6 Field-oriented test methods to evaluate structural build-up at rest o/SCC

0 10 20 30 40 IPT0rest(t) (Pa/min)

Fig. 6.21 Relationship between time-dependent static yield stress determined from portable vane

and inclined plane tests

140

^ a • mm

170

S 100 « 0H

N—•

& s—^

Ore

H > CU

XO

60

40

70

0

0 20 40 60 Rheometer T0rest (t) (Pa/min)

Fig. 6.22 Relationship between time-

dependent static yield stress determined from

portable vane and concrete rheometer tests

c g OS

PH

^ * • * - '

rest

e H

OH

35

30

25

20

15

10

5

y = 0.60x R2 = 0.93

Y.S

*yt x fir

r*

i i 0

0 20 40 60 Rheometer T0rest (t) (Pa/min)



inclined plane and concrete rheometer tests

C. Correlating time-dependent static yield stress determined from field-oriented tests to

time-dependent drop in apparent viscosity at 0.7 rps determined from rheometric test

The correlations between the time-dependent change of the drop in apparent viscosity at

rotational frequency of 0.7 rps obtained using the concrete rheometer [Ar)[email protected](t)] versus the

PVxorest(t) and IPTorest(t) are shown in Figs. 6.24 and 6.25, respectively. For the PV and IP field-

oriented methods, the R2 values for the correlations with the rheometeric measurements were

0.96 and 0.93, respectively.

133

140

o

a/m

fe • * *

~ u 1 -

©

> OH

120

100

80

60

40

20

-

"x

y

y = R2

3.49x >X = 0.96 J ^

{\

i i i 0

0 10 20 30 40 [email protected](t) (Pa.s/min)



portable vane test and time-dependent drop in

apparent viscosity at 0.7 rps obtained using

concrete rheometer

/ • s

s 93

& /—"s

^ w

a. M

35

30

25

20

15

10

5

-

-

is X

y = R2 =

X

X

o.82x y1

= 0.93 A

*yr

i i i

0 10 20 30 40 [email protected](t) (Pa.s/min)



inclined plane test and time-dependent drop in

apparent viscosity at 0.7 rps obtained using

concrete rheometer

Based on the above results, the PV and IP field-oriented test methods enable the evaluation

of the magnitude of structural build-up knowing the initial static yield stress or the time-

dependant change of static yield stress with respect to the rest time (slope). A third index can

correspond to the couple effect of initial rheological response and slope. This index is calculated

as the multiplication of initial response and slope. A summary of structural build-up indices

determined from the PV and IP field-oriented tests as well as the modified Tattersall MK-III

concrete rheometer are listed in Table 6.5. These indices will be used to predict the lateral

pressure characteristics exerted by SCC on formwork as will be elaborated in Chapters 7 to 10.

Table 6.5 Thixotropic indices obtained from the field-oriented test methods

Initial response at 15 min time of resting

Time-dependent change of the response (slope)

Couple effect of initial and slope

Breakdown area

Ab,

~

~

Concrete rheometer

Static yield stress

Rheometercorest@i5min

Rheometerrorest(t)

Rheometerr0res,@i5min XT0rest(t)

drop in apparent viscosity at 0.7 rps

AT|app@N=0.7rps@15min

AT|app@N=0.7rps(t)

AT| app@N=0.7rps@ 15min x

AT|app@N=0.7rps(t)

PV test

Static yield stress

PVTorest@15min

PVT0rest(t)

PVTorest@15minx

PVT 0 r e s t ( t )

IP test

Static yield stress

IPT0rest@15min

I P t o rest(t)

IP't0rest@15min><

IPT0rest(t)

134


6.6 Conclusions

Based on the results presented in this chapter that aimed at developing field-oriented test

methods to determine the rate of structural build-up of SCC and correlate the responses to scientific

rheometric measurements, the following conclusions are warranted:

1. The portable vane and inclined plane test methods show good repeatability when used to assess

thixotropy of SCC. These tests are simple and can easily be used in laboratory and in the field.

2. Static yield stress and its changes with respect to resting time obtained using the portable vane

and inclined plane tests can be used to reflect the change in the rate of structural build-up at rest

of SCC mixtures.

3. The results of the field-oriented test methods are validated using up to 42 SCC mixtures of

different compositions. Good correlations are obtained using the PV and IP field-oriented tests

in terms of static yield stress at rest and rheometric concrete measurements in terms of static

yield stress at rest, drop in apparent viscosity at rotational frequency of 0.7 rps, and breakdown

area. The compared parameters are measured initially and with respect to time of rest. The R2

values for these correlations are summarized in the following table.

Table 6.6 R2 values for initial and time-dependent responses

PVlOrest VS. IPx0 r e s t

PViorest vs. Rheometerxorest IPTorest vs. Rheometerxorest PVxorest VS. Ar|app@N=0.7rps

IPTorest VS. AnapP(2jN=0.7rps

PVxorest VS. A b j

IPXOrest VS. A b i

Initial responses Measurements from Measurements

15-60 min at 15 min 0.71 0.93 0.78 0.98 0.74

—

0.82 0.82 0.82 0.81 0.67 0.87 0.70

- Time-dependent responses

0.85 0.96 0.93 0.96 0.93

—

135

CHAPTER 7

EFFECT OF SCC MIX DESIGN ON FORMWORK PRESSURE

CHARACTERISTICS

7.1 Objectives

As presented in Chapter 2, several mix design parameters and concrete rheological

parameters affect lateral pressure exerted by SCC. These parameters elaborated in Table 2.2

also affect thixotropy of SCC. The main objective of this chapter is to evaluate the effect of

slump flow ((|)) of SCC, relative content of coarse aggregate (Vca), sand-to-total aggregate

(S/A), paste volume (Vp), and maximum-size aggregate (MSA) on the development of

thixotropy and lateral pressure characteristics of SCC. Both parametric approach and full-

factorial design were prepared for this purpose. The chapter also aims at developing statistical

prediction models to be used as guidelines for evaluating thixotropy and formwork pressure

characteristics of SCC.

7.2 Testing program

The testing program was divided into three main phases, as indicated in Fig. 7.1. The

experimental program of each phase is described below.

Phase I: Effect of slump flow, sand-to-total aggregate ratio, and coarse aggregate

content on SCC formwork pressure and thixotropy

The influence of <|>, S/A, and Vca on SCC formwork pressure characteristics and relevant

rheological properties were evaluated using a full-factorial design. The (|), S/A, and Vca were

considered for the mix design parameters in the experimental design. As shown in Table 7.1,

three distinguished ranges were considered for each mixture parameter to produce different

stiffening rates. The 23 experimental design necessitated eight mixtures (SCC25 to SCC32).

Four central points (SCC33) were also elaborated to estimate the experimental errors. The SCC

mixtures were compared to a reference concrete of normal slump consistency (CC34). The

compositions of the 10 mixtures are given in Table 3.8. Several responses were considered in

the experimental design. Lateral pressure characteristics (K0 at three different casting heights

and decays of lateral pressure during two periods: first hour after end of casting and the total

time of pressure cancellation) and 10 thixotropic indices form the concrete rheometer and field-

oriented test methods.

136

Chapter 7 Effect of SCC mix design on formworkpressure characteristics

Fig. 7.1 Flow chart for Chapter 7

Evaluate effect of mix design parameters on the lateral pressure characteristics and thixotropy of SCC

Phase I Investigate effect o/<j>, S/A, Vc

Phase II Effect of Vp

Phase III Effect of MSA

• Eight SCC mixtures + one SCC mixtures repeated four times.

• Three ranges for each parameter were incorporated in the 9 SCC: 4 = 600, 660, and 720 mm S/A = 0.44, 0.48, and 0.52 Vca = 0.27, 0.30, and 0.33

• One conventional concrete was used as reference.

• Five SCC mixtures paste volumes of 340, 360, 370, 390, and 400 1/m3.

• One conventional concrete was used as reference.

• Two series of SCC mixtures with different thixotropy were investigated.

• Each series has 3 SCC mixtures proportioned with MSA of 10, 14, and 20 mm.

The fresh concrete properties including slump flow or slump consistency, T5o, unit weight,

air content, and concrete temperature were determined for each tested mixture and the results are

presented in Table C.l (Appendix CI). The UofS2 pressure column was used to monitor lateral

pressure profiles for 13-m high placement. The PVC column of 1.2-m height and 0.2-m diameter

was also used to monitor variation of lateral pressure over 24 hr. The SCC was continuously cast

at a rate of rise (R) of 10 m/hr and at an approximate temperature (T) of 22 ± 2 °C. The CC35 was

placed at the same rate as that of SCC, but in layers of about 0.25 m in height. Each layer was

consolidated with standard rod (10 strokes per layer). The various thixotropic indices described in

Chapter 6 obtained using the modified Tattersall MK-III concrete rheometer and the selected

field-oriented test methods were also determined.

Table 7.1 Ranges : of selected parameters for the factorial design

Low

Stiffening rate(1)

Medium High

<|> (mm)

Vca, by volume

S/A, by volume

720±13 (2 )

0.33

0.52

660 ±13

0.30

0.48

600 ±13

0.27

0.44 ( , ) Refers to level of thixotropy controlled by varying Vca and <j) (2) 13 mm approximately equals to Vz inch

Phase II: Effect of paste volume on SCC formwork pressure and thixotropy

Five SCC mixtures (SCC36 to SCC40) proportioned with five paste volumes (Vp) varying

between 340 and 400 1/m3, were used (Table 7.2). The SCC mixtures were compared to a normal

concrete (CC35) containing 370 1/m3 of paste. The SCC and CC mixtures were proportioned with

137

Chapter 7 Effect ofSCC mix design on formwork pressure characteristics

cement Type GU-bS/F with the compositions given in Table 3.8. The SCC mixtures were

designed with initial slump flow of 700 ± 20 mm, where CC35 with 180 ± 20 mm slump.

After mixing, slump flow (or slump consistency), J-Ring spread, concrete temperature,

air content, and unit weight were measured. The J-Ring spread was repeated after 40 min and

the slump flow after 120 min both from end of mixing. The obtained properties are

summarized in Table C.2 (Appendix CI). The UofSl pressure column was employed to

determine the lateral pressure profile. The PVC column measuring 1.2 m in height and 0.2 m

in diameter was used to monitor the variations of lateral pressure with time until pressure

cancellation (tc). The SCC and conventional concrete mixtures were placed in a similar

method as that in Phase I. The thixotropic properties of the tested mixtures were determined

using the modified Tattersall MK-III concrete rheometer.

Table 7.2 Mixtures used to evaluate effect of paste volume on lateral pressure and thixotropy

Mixture Vp (1/m3) <j) (mm)

CC35(ref) 370 180 ± 20 (slump)

SCC36 340

SCC37 360

SCC38 370 700 ± 20 (slump flow)

SCC39 390

SCC40 400

R=10m/hr, T = 22±2°C

Phase III: Effect of maximum-size of aggregate on SCC formwork pressure

A parametric study was undertaken to investigate effect of MSA on the relevant lateral

pressure characteristics using two series of SCC of different thixotropic levels. In each series,

three SCC mixtures proportioned with MSA of 10, 14, and 20 mm were tested. The SCC13,

SCC5, and SCC9 mixtures of the first series were designed to produce relatively low

thixotropy. On the other hand, the SCC58, SCC59, and SCC60 used for the second series had

relatively high thixotropy. Summary of the experimental work undertaken in Phase III is

presented in Table 7.3. The mix designs of these mixtures are given in Table 3.8. The SCC

was placed at R of 10 m/hr and T of 22 ± 2°C. The concrete was tested to determine, slump

flow, T5o, unit weight, air content, and temperature. The test results are summarized in Table

C.3 (Appendix CI). The UofS2 pressure column was used to monitor the variation of lateral

pressure profile when simulating 13-m high formwork. The structural build-up at rest was

determined using the PV field-oriented test.

138

Chapter 7 Effect ofSCC mix design on formworkpressure characteristics

Table 7.3 Experimental work to evaluate effect of MSA on SCC lateral pressure

Maximum-size of aggregate (MSA) <|) (mm)

10 mm 14 mm 20 mm Series # 1: low thixotropy SCC 13 SCC5 SCC9 660 Series # 2: high thixotropy SCC58 SCC59 SCC60 600 Note: R=10m/hr, T = 22±2°C

7.3 Test results of Phase I

The effects of mixture parameters {§, S/A, and Vca) on lateral pressure characteristics and

thixotropic properties are discussed below. Correlations between formwork pressure and

relevant structural build-up at rest of SCC are established. Statistical models derived for 16

different responses for the plastic concrete that were established from the experimental design

are presented.

7.3.1 Effect of <|>, S/A and Vca on SCC formwork pressure

The maximum lateral pressure at different concrete heights (Pmax@Hi) relative to the

local equivalent hydrostatic pressure (Phyd@H/) was calculated to determine the relative lateral

pressure (Ko,). Typical variations of KG, with height are illustrated in Fig. 7.2. The lateral

pressure profile for SCC27 (<)) = 600 mm) is compared to SCC31 (§ = 720 mm); similarly

SCC33E (<|> = 600 mm) is compared to CC34 (d> = 180 mm) in Fig. 7.2(a). As expected, SCC

with higher <|) develops greater Koi values. Each pair of the above mixtures has the same

mixture composition, except for the FIRWRA dosage to produce the target slump flow value.

Changes in Vca and S/A play an important role on lateral pressure profile, as illustrated in

Figs. 7.2(b) and (c), respectively. The increase in Vca from 270 to 330 1/m3 leads to a

reduction in lateral pressure profile. Increase the Vca value reduces the paste volume and

consequently leads to increase in internal friction, thus resulting in lower lateral pressure.

The decay in lateral pressure until pressure cancellation [K(t)] monitored using the 1.2-m

high PVC column is reported here in terms of the pressure decay during the first 60 min after

the end of casting [K(t)(0-60min)] as well as during the entire period necessary for pressure

cancellation time [K(t)(0-tc)]. The pressure decay characteristics of SCC27 and SCC31 are

compared in Fig. 7.3 (a). These mixtures have the same mixture compositions, except for the

HRWRA dosage, had (j) values of 600 and 720 mm, respectively. In the same figure, SCC33E is

compared to the conventional concrete (CC34) that had § values of 600 and 180 mm,

respectively. The results show that concrete proportioned with higher concentration of

HWRWA results in slower pressure decay and longer time before pressure cancellation. The

139

increase of Vca from 270 and 330 1/m (SCC31 to SCC32) led to sharper lateral pressure decay

and shorter time to pressure cancellation, as shown in Fig. 7.3 (b). In addition, increasing the

S/A from 44% to 52% (SCC30 to SCC32) resulted in slightly slow pressure decay and longer

time for pressure cancellation, as indicated in Fig. 7.3 (c). The SCC32 showed about 80 min

longer in the time of pressure cancelling than that noted for SCC30.

Relative lateral pressure K^ (%)

0 20 40 60 80 100

JSP "3

2 u a ©

§4

10

12

14

(a):effect of (|> R = lOm/hr T = 22 ± 2°C

CC34 (slump = 180 mm)

SCC33E (4> = 660 mm)

SCC31 %. (<|> = 720 mm)

SCC27 (<|> = 600 mm)

JSP ' 3

B O

U

Relative lateral Iressure Koi (%)

20 40 60 80 100

0

2

4

6

8

10

12

14

(b):effectofV, R=10m/hr T = 22 ± 2°C

SCC32 (Vca=0.33)

SCC31 (Vca = 0.27)

"3

2 w a o U

Relative lateral Iressure K^ (%) 20 40 60 80 100

0

2 Z>

4

6

8

0

2

A

(c):effectofS/A w R=10m/hr VY

' T = 22 ± 2°C Jr

P SCC30 A (S/A = 0.44) / /

X / SCC32 / (S/A = 0.52)

/ /

/ /

/ /

Fig. 7.2 Variation of relative lateral pressure with concrete height

140

100 £ **mS

i» 3 VI fi

0* u a, « u • * *

CS 4»

PS

80

60

40

?0

r SCC31 ®fei fc (()> = 720 mm)

> THilliifrili /

/ - * \ CC34 \

(slump = 180 mm) \ i i i i \

(cC3. = 660

i

5E mm)

(a): Effect of <|> R=10m/hr T = 22 ± 2°C

SCC27 (<|> = 600 mm)

| 4 * T V . I « • I I

0 60 120 180 240 300 360 420 480 540 600 660 720 780 Time (min)

(b): Effect of Vca

R=10m/hr T = 22 ± 2°C

60 120 180 240 300 360 420 480 540 600 660 720 780 Time (min)

^ 100

80 g 4> U a « u 4>

4J

^ "3 ^ PS

60

40

20

-

- ^HjLt ^S%. SCC32

^ % ^ ^ (S / A = 0-52)

SCC30 ^ ^ W M f f l l ^ f r ^ (S/A = 0.44) ^vmwg^

1 1 1 1 1 1 1

(c): Effect of S/A R=10m/hr T = 22 ± 2°C

? § I 4» I i i

60 120 180 240 300 360 420 480 540 600 660 720 780

Time (min)

Fig. 7.3 Pressure decay characteristics

141

Chapter 7 Effect of'SCC mix design on formworkpressure characteristics

7.3.2 Effect of <|>, S/A, and Vca on thixotropy of SCC

The structural build-up at rest of the tested mixtures was determined using the modified

Tattersall MK-III concrete rheometer and the field-oriented test methods (PV and IP tests)

evaluated in Chapter 6. The effect of SCC mixture parameters (()), S/A, and Vca) on the

variations of the PVTorest@i5min are indicated in Fig. 7.4. For given values for Vca and S/A of

0.52 and 0.27 (SCC27 and SCC31), increasing <|> from 600 to 720 mm led to a reduction in the

PVTO rest@i5min from 327 to 210 Pa. Also, proportioning SCC with higher Vca leads to increase

in thixotropy. For example, the increase of 60 1/m3 of Vca of mixture SCC32 compared to

SCC31 resulted in 258 Pa increase in the PVTorest(2>i5min value.

1600

b 1200

® 800

o

PL, 400

o CO

Effect of <)> on thixotropy

<W fr

Effect of V. on thixotropy

4J 4" 4" 4"

Effect of S/A on thixotropy

4" 4"

Fig. 7.4 Effect of mix design parameters of SCC on PVT0rest@i5min

7.3.3 Relationship between Ko and structural build-up at rest

As presented in Chapter 6, several thixotropic indices (T.I.) were determined using the

modified Tattersall MK-III concrete rheometer (RheometerTorest and Ar|app@N=o.7rps), portable vane

test (PViorest), and inclined plane test (jPiorest)- The initial thixotropy indices at 15 min of resting

time [T.I.@i5min], the time-dependant change of thixotropic indices [T.I.(t)] (slope), and the couple

effect of initial and slope [[email protected].(t)] were established for each test method to produce a

total of 12 T.I. These indices are correlated to Ko as discussed below.

The Ko values at simulated concrete heights (H) of 1, 4, 8, and 12 m were correlated to the

T.I.@i5min, T.I.(t), and [email protected].(t) determined from the rheometer Ar|app@N=o.7lpS and portable

vane test, as shown in Figs. 7.5 and 7.10. In addition, the K0 values at the same casting heights

142

were correlated to the various T.I. determined from the rheometer T0rest and inclined plane test, as

presented in Appendix C2. The increase of the T.I. of SCC mixture led to a reduction in the Ko

values. For example, increasing the PVx0 rest@i5min from 210 to 1455 Pa led to a reduction in the

Ko values at H of 12 m from 66% to 23%. Increasing the casting depth resulted also in a

reduction in the K0 values. For example at the PVTO rest@i5min of 1455 Pa, the Ko values were 62%,

43%, and 23% at H of 4, 8, and 8 m, respectively. The reductions in the Ko values due to the

increase of casting depth for the mixtures of greater T.I. were larger than those obtained for the

mixtures of lower T.I. For example at the PVio rest@i5min of 210 Pa, the reduction in the K0 values

when the casting depth increased from 1 to 4 m and then to 8 m, were 9% and 12%, respectively.

The corresponding reductions at the PVTorest@i5min of 1455 Pa were 15% and 19%, respectively.

^

5

100 90 80 70 60 50 40 30 20 10 0

R= lOm/hr <t> = 600 -720 mm

K0@H=1 m Q R2 = 0.74

» K 0 @H==4 m

R2 = 0.81

K,y:aiH=8 m 'R2 = 0.85

rjK0@H=12m

R2 = 0.86

100 200 300

A1lapp@N=0.7rps@15min (Pa.S)

400

Fig. 7.5 Variations of K0 with Ar|app@N=o.7rps@i5min at different heights of placement

100 90 80 70

~ 60 b 50 ^ 4 0

30 20 10 0

K0@H=1 m R2 = 0.82

• A ^ . - f c ^ K„(tf;H=4 m 89

R= lOm/hr <|> = 600 -720 mm

«. K0@'H=8 m ^ R2 = 0.91

K0@H=12 m

R2 = 0.92

500 1000 1500 * * T0 rest@15min ( P a )

2000

Fig. 7.6 Variations of K0 with PVxorest@i5min at different heights of placement

143

Chapter 7 Effect of SCO mix design on formworkpressure characteristics

100 90 80 70 60 50

^ 4 0 30 20 10 0

^

K0@H=1 m R2 = 0.83

"""**• —-^L K0('fllll=4 m X R2 = 0.88

R= lOm/hr <t» = 600 -720 mm

K0@H=8 m A R2 = 0.89

K0@H=12m R2 = 0.89

10 20 Allapp@N=o.7rpS(

t) (Pa.s/min) 30

Fig. 7.7 Variations of K0 with Ar|app @ N=0.7 rps(t) at different heights of placement

K0@H=1 m R2 = 0.84

y K0(«UI=4 m A R2 = 0.89

20 40 60 80 PVT0rest(t)(Pa/min)

100

Fig. 7.8 Variations of K0 with PVT0rest(t) at different heights of placement

R= 10m/hr <|> = 600 -720 mm

100 90 80 70

C? 60 *% 50 * 40 h

30 20 10 0

0 2000 4000 6000 8000 2 10000

Fig. 7.9 Variations of K0 with Anapp@N=o .7rps@15min xAr|app@N=o 7rps(t) at different heights of

placement

144

K0@H=1 m R2 = 0.86

K0@H=4 m R2 = 0.89

* • » ^ K0(fl>,H=8 m

R2 = 0.90

K0@H=12 m R2 = 0.89

100

80

^ 60

J 40

20

0

K0@H=1 m

R=10m/hr <|> = 600 -720 mm

R2 = 0.91

K0@H=12m R2 = 0.90

50000 100000 150000 P V T 0 rest^lSmin^O rest(0 (Pa 2 /min)

Fig. 7.10 Variations of K0 with PVTorest@i5minxPVTorest(t) at different heights of placement

The R2 values for 12 correlations between Ko values at casting depths of 1, 4, 8, and 12 m

and various thixotropic indices were ranged between 0.74 and 0.92, as shown in

Table 7.4. The correlations between K0 values and [email protected].(t) determined from the

Rheometeriorest and PVrorest are recommended to enable the prediction of K0 values at different

concrete heights as they had the higher R2 values that ranged between 0.88 and 0.91. In addition,

the two relationships include the different structural behaviours as they take into account the

thixotropic behaviour at the initial stage as well as its change with time.

Table 7.4 R2 values for the various correlations between Ko and thixotropy indices

H(m)

Rheometertorest

AT)app@N=0.7 rps

PVlOrest

IPtOrest

T.I.@15min

1

0.81

0.74

0.82

0.77

4 8

0.86 0.89

0.81 0.85

0.89 0.91

0.84 0.88

12

0.89

0.86

0.92

0.90

T.I-(t)

1

0.79

0.83

0.84

0.83

4 8

0.83 0.84

0.88 0.89

0.89 0.91

0.88 0.89

12

0.84

0.89

0.91

0.89

T.I.fglSmin^T.I.ft)

1 4 8 12

0.89 0.91 0.91 0.90*

0.86 0.89 0.90 0.89

0.88 0.91 0.91 0.90*

0.87 0.90 0.91 0.90

* Recommended test methods

7.3.4 Correlation between decay of lateral pressure and thixotropy of SCC

The two pressure decay parameters, [AK(0-60 min) and AK(0-tc)], were correlated to

the various thixotropy indices obtained using the concrete rheometer and field-oriented test

methods. The correlations of the pressure decay and the thixotropy indices of the rheometer

drop in apparent viscosity and portable vane test are shown in Figs. 7.11 to 7.16. In addition,

the correlations between pressure decay and the thixotropy indices determined from the

rheometer Torest and inclined plane test are presented in Appendix C3. The increase of thixotropy

level resulted in faster pressure decay. For example, increasing the PVto rest@i5min values from

145

210 to 1455 Pa led to an increase in the [AK(0-60 min) values from 0.13 to 0.28 %/min,

respectively. The corresponding increase in the AK(0-tc)] values were from 0.14 to 0.25

%/min, respectively. SCC mixtures showed faster decay during the first 60 min compared to

the overall decay until pressure cancellation. At the PVxo rest@i5min value 1455 Pa, 0.28 and

0.25 %/min were recorded for the [AK(0-60 min) and AK(0-tc)] values, respectively.

0.30

0.25

J 0.20

d °-15

° 0.10

0.05

0.00

<

AK(t)(0-60min) y = 0.0005x + 0.1132

R2 = 0.86

AK(t)(0-tc) y = 0.0003x + 0.1297

R2 = 0.77

0 100 200 300 400 ' 1lapp@N=0.7rps@lSmin ( I» .S)

Fig. 7.11 Variations of pressure decay with Ar|app@N=o.7 rpS(t)

0.30 AKftlCO-60 min^ ^

0.25

| 0.20

2,0.15 ~ 0.10 M

0.05

0.00

AK(t)(0-60 min)

y = 0.0001 x + 0.1092 R2 = 0.84

tfC^K

A AK(t)(0-y

y = O.OOOlx + 0.1259 R2 = 0.73

0 500 1000 1500 " V T 0 rest@15min (™a)

2000

Fig. 7.12 Variations of pressure decay with PVT0 rest@i5min

0.30

0.25

.S 0.20 s £0.15 S o.io

< 0.05

0.00

AKYW0-60 min> y = 0.0060x +0.1197

R2pj).86

'e A

AK(t)(0-tJ> y = 0.0036x +0.1326

R2 = 0.79

0 10 15 20 AiLppfcNHUrpitf) (Pa.s/min)

25 30

Fig. 7.13 Variations of pressure decay with Ar|app@ N=o.7rPs(t)

146


0.05 -

0.00 ' ' ' ' ' '

0 20 40 60 80 100 PVTores,(t)(Pa/min)

Fig. 7.14 Variations of pressure decay with PVxorest(t)

0.05 -

0.00 I ' ' ' ' '

0 2000 4000 6000 8000 10000 Allapp@N=0.7rps@15n,inXATlapp@N=0.7rpS(t)[(Pa.S)2/min]

Fig. 7.15 Variations of pressure decay with Anapp@N=o.7rpS@i5minxAr|app@N=o.7rPs(t)

0.05 -

0.00 ' ' ' '

0 50000 100000 150000

P V T 0 rest@15minXTo restW ( P a 2 / m i n )

Fig. 7.16 Variations of pressure decay with PVTorest@i5minxTorest(t)

147

The R2 values for the all 12 relationships between pressure decay and T.I. are presented in

Table 7.5. Based on these findings, the [AK(0-60 min)] can be predicted using Ar|app@N=o.7ips@i5min

and PVTorest@i5min- The RheometeiTorest@i5minxTorest(t) and PVT0rest@i5minxTorest(t) are recommended

to predict [AK(O-tc)] values, as indicated in Table 7.5.

Table 7.5 R2 values of the various correlations between pressure decay and thixotropy indices

T.I.

Rheometerxorest

AT|app@N=0.7 rps

PVtOrest

IPfOrest

AK(t)(0-

T.I.@l5min T.I.(t)

0.71

0.86*

0.84+

0.83

0.73

0.86

0.82

0.80

60 min)

T.I •@15minxT.I.(t)

0.72

0.84

0.78

0.78

T.I.@l5min

0.77

0.77

0.73

0.68

AK(t)(0-tc

T.I.(t) T.I

0.79

0.79

0.78

0.77

)

•@15minxT.I.(t)

0.89*

0.86

0.85+

0.83 + Recommended test method in general

Recommended when the concrete rheometer is available

7.3.5 Models to simulate effect of <|>, S/A, and Vca on SCC lateral pressure and thixotropy

A. Model derivation

Table 7.6 shows the full-factorial experimental plan. The absolute values corresponding

to coded values of-1, 0, and +1 for the <)> parameter in Table 7.6 are 600, 660, and 720 mm.

These values for S/A are 0.44, 0.48, and 0.52, and 0.27, 0.30, and 0.33 for Vca, respectively.

The coded values can be calculated using the formulas in Table 7.7.

Table 7.6 Full-factorial experimental design carried out in Phase II

Main matrix

Central points

Mixture -

SCC25

SCC26

SCC27

SCC28

SCC29

SCC30

SCC31

SCC32

SCC33C

SCC33D

SCC33E

SCC33F

4 -1

-1

-1

-1

1

1

1

1

0

0

0

0

Coded values

S/A

-1

-1

1

1

-1

-1

1

1

0

0

0

0

• ca

-1

1

-1

1

-1

1

-1

1

0

0

0

0

Absolute values

§ (mm)

600

600

600

600

720

720

720

720

660

660

660

660

S/A (ratio) V

0.44

0.44

0.52

0.52

0.44

0.44

0.52

0.52

0.48

0.48

0.48

0.48

ca (ratio)

0.27

0.33

0.27

0.33

0.27

0.33

0.27

0.33

0.30

0.30

0.30

0.30

Reference CC34 of slump consistency of 180 ± 20 mm

Note: R=10m/hr, T = 22±2°C

148

Table 7.7 Formulas used to convert absolute to coded values for parameters considered

Coded (|> value Coded S/A value

Coded Vca value

Absolute <j> value

Absolute S/A value Absolute Vca value

= (Absolute <|> value - 660)/60 = (Absolute S/A value - 0.48)/0.04

= (Absolute Vca value - 0.30)/0.03

= 60 x coded <j> value + 660

= 0.04 x coded S/A value + 0.48

= 0.03 x coded Vca value + 0.30

Eq. 7.1 Eq. 7.2

Eq. 7.3

Eq. 7.4

Eq. 7.5 Eq. 7.6

In total, 14 statistical models were derived as function of the three mixture parameters

(<j), S/A, and Vra). Six models to estimate lateral pressure characteristics. Besides, 10 models

to predict various thixotropic indices from the modified Tattersall MK-III concrete rheometer

and field-oriented test methods (PV and IP tests), as shown in Table 7.8.

The estimate, Prob. >|t|, and coefficient of correlation (R2) values for the 16 derived models

are given in Table 7.8. The estimate for each parameter refers to the coefficients of the model

found by the least square approach. The Prob. >|t| term is the probability of getting an even

greater t statistic, in absolute value, that tests whether the true parameter is zero. Probabilities less

than 0.10 are typically often considered as significant evidence that the parameter is not zero, i.e.

that the contribution of the proposed parameter has a significant influence on the measured

response. The proposed models have high R2 varying from 0.84 to 0.99.

Table 7.9 compares the effect of the three mixture parameters and their interactions on

the modeled responses. The sign of the estimates (+/-) indicates the positive or the negative

effect of variable on the considered response. For example, the increase of <j> and S/A (+) will

lead to an increase in K0@H=4m, while the increase of Vca (-) will reduce K0@H=4nv The models

in Table 7.9 also give an indication of the relative significance of the various parameters and

their interactions. For example, the K0@H=4m is affected mainly by the changes in Vca followed

by ()>, then S/A, etc. The 14 statistical predicting models based on the absolute values of the

tested mixture parameters (<)) in mm, S/A, and Vca as ratios) are shown in Table 7.10.

149

Cha

pter

7 E

ffec

t of

SCC

m

ix d

esig

n on

form

wor

kpre

ssur

e ch

arac

teri

stic

s

Tab

le 7

.8

Para

met

er e

stim

ates

of

deri

ved

mod

els

tics cteris chara essure eral pi 3

sail r.i. fr dified

OUI

sld-t

thefi sd tesl

o u

H

MK-III concrete rheometer

o •+-»

u

Mod

el

Ko@

H=4

m (

%)

Ko@

H=8

m (

%)

Ko@

H=1

2 m

(%

)

AK

(t)(

0-60

min

) (%

/min

)

AK

(t)(

0-tc

) (%

/min

)

t c (

min

)

Rhe

omet

erT

0re S

t@i5

min

(P

a)

Ar| a

pp@

N=0

.7rp

s@15

min

(P

a.S

)

Rhe

omet

erio

res

t(t)

(P

a/m

in)

Ar|a

p P@

N=o

.7rp

s(t)

(Pa.

s/m

in)

PV

Tor

est@

15m

in(P

a)

PVxo

rest

(t)

(Pa/

min

)

IPt0

rest

@15

min

(^

a)

IPxo

rest

(t) (

Pa/m

in)

R2

0.94

0.94

0.91

0.98

0.88

0.98

0.84

0.86

0.91

0.98

0.89

0.99

0.85

0.99

Ter

m

Est

imat

e P

rob»

|t|

Est

imat

e Pr

ob>

|t|

Est

imat

e Pr

ob>

|t|

Est

imat

e Pr

ob>

|t|

Est

imat

e Pr

ob>

|t|

Est

imat

e Pr

ob>

|t|

Est

imat

e Pr

ob>

|t|

Est

imat

e Pr

ob>

|t|

Est

imat

e Pr

ob>

|t|

Est

imat

e Pr

ob>

|t|

Est

imat

e Pr

ob>

|t|

Est

imat

e Pr

ob>

|t|

Est

imat

e Pr

ob>

|t|

Est

imat

e Pr

ob>

|t|

C:

Inte

rcep

t

82

<.0

001

67.7

<

.000

1 53

.5

<.0

001

0.16

83

<.0

001

0.16

<

.000

1 58

7.7

<.0

001

548.

92

<.0

001

113.

83

<.0

001

10.3

892

<.0

001

8.32

64

<.0

001

537.

5 <

.000

1 26

.757

5 <

.000

1 49

4.67

<

.000

1 7.

089

<.0

001

 3.

015

0.00

03

4.06

75

0.00

03

5.11

75

0.00

07

N/A

N

/A

-0.0

0625

0.

0044

38

.062

5 0.

0002

-1

73.5

0.

005

-33.

875

0.00

92

-4.1

663

0.00

16

-3.0

494

0.00

01

-155

.375

0.

003

-12.

2175

<

.000

1 -1

55.8

75

0.00

63

-3.4

025

0.00

04

S/A

1.68

75

0.00

7 1.

96

0.01

44

2.23

25

0.04

85

-0.0

175

<.0

001

-0.0

0625

0.

0044

24

.187

5 0.

0029

-1

02.7

5 0.

0486

-3

1.87

5 0.

0124

-1

.838

8 0.

0648

-2

.603

0.

0003

-1

27.8

75

0.00

87

-9.3

4 0.

0001

-1

27.8

75

0.01

68

N/A

N

/A

v ca

-3.1

75

0.00

02

-4.7

275

0.00

01

-6.2

775

0.00

02

0.03

25

<.0

001

0.00

875

0.00

06

-48.

5625

<

.000

1 14

3.25

0.01

27

52.6

25

0.00

07

5.11

875

0.00

05

4.19

94

<.0

001

220.

125

0.00

03

14.3

675

<.0

001

208.

625

0.00

12

3.76

0.

0002

(j)x

S/A

N/A

N

/A

N/A

N

/A

N/A

N

/A

N/A

N/A

N

/A

N/A

9.

9375

0.

1083

N

/A

N/A

N

/A

N/A

N

/A

N/A

N

/A

N/A

N

/A

N/A

3.

0425

0.

0174

N

/A

N/A

N

/A

N/A

<f>x

Vca

0.9

0.08

44

1.17

75

0.09

34

N/A

N

/A

N/A

N/A

N

/A

N/A

N

/A

N/A

-9

8

0.05

72

N/A

N

/A

-1.7

613

0.07

42

-0.8

886

0.04

74

N/A

N

/A

-4.9

55

0.00

23

N/A

N

/A

-2.2

075

0.00

49

S/A

xV

ca

N/A

N

/A

N/A

N

/A

N/A

N

/A

-0.0

075

0.00

36

N/A

N/A

N

/A

N/A

N

/A

N/A

N

/A

N/A

N

/A

N/A

-1

.143

4 0.

0186

N

/A

N/A

-3

.767

5 0.

0075

N

/A

N/A

N

/A

N/A

150

Cha

pter

7 E

ffec

t of

SCC

m

ix d

esig

n on

form

wor

kpre

ssur

e ch

arac

teri

stic

s

Tab

le 7

.9

Stat

istic

al m

odel

s in

cod

ed v

alue

s (v

alue

s of

(j),

S/A

, an

d V

ca f

rom

-1

to +

1)

l- 3 1/3

11)

Q.

—*

cS

i/>

O

•j2 C/l w

o

?s

K0@

H=4

m (

%)

= 8

2 -

3.17

5 V

ca +

3.0

15 $

+ 1

.687

5 S

/A +

0.9

$. V

ca

Eq.

7.7

KO

@H

=8 m

(%

) =

67.

2 -

4.72

75 V

ca +

4.0

675

<J> -

f-1.

96 S

/A +

1.

1775

4>.

Vca

E

q. 7

.8

K0@

H=i

2 m

(%

) =

53.

5 -

6.27

75 V

ca +

5.1

175

$ +

2.2

325

S/A

E

g. 7

.9

AK

(t)(

0-60

min

) (%

/min

) =

0.1

683

+ 0

.032

5 V

ca -

0.0

175

S/A

- 0

.007

5 S/

A.

Vca

E

q. 7

.10

3 %

A

K(t

)(0-

t c)

(%/m

in)

= 0

.16

- 0.

0062

5 (|)

+ 0

.004

4 S/

A +

0.0

006

Vca

E

q. 7

.11

tc (

min

) =

587

.7 -

48.

5625

Vca

+ 3

8.06

25 (|>

+ 2

4.18

75 S

/A +

9.9

375

fS/A

E

q. 7

.12

_ ^

Rhe

omet

erT 0

rest@

i5m

in (

Pa)

= 5

48.9

-173

.5 ((

) +14

3.25

Vca

-10

2.7

5 S

/A-9

8 (|)

.Vca

E

q. 7

.13

||

| H

|

| M

i app

@N

^ 7rp

s@15

min

(Pa.

s)=

113

.8 +

52.

625

Vc

a-

33.8

75 <

|> -

31.

875

S/A

E

q. 7

.14

~ "°

1

^ I

8 R

heom

eter

xo re

st(t

) (P

a/m

in)

= 1

0.4+

5.

1187

5 V

ca-

4.16

625

$-

1.83

875

S/A

-1.

7612

5 <|>

.Vca

Eq.

7.15

ATi

app @

N=o

. 7lp S

(t) (

Pa.

s/m

in)

= 8

.33

+ 4

.199

375

Vca

- 3

.049

375

4 -

2.60

3125

S/A

- 1

.143

375

S/A

.Vca

- 0

.888

625

<)>.V

ca E

q. 7

.16

| PV

T 0res

t@i5

min

(P

a) =

537

.5 +

220

.125

Vca

-15

5.37

5 4

- 12

7.87

5 S/

A

Eq.

7.1

7

J 2

| "§

PV

To r

est(

t) (

Pa/

min

) =

26.

76 +

14.

3675

Vca

-12

.217

5 (|)

- 9.

34 S

/A -

4.9

55 (

|).V

ca

- 3.

7675

S/A

.Vca

+ 3

.042

5 4>

.S/A

E

g. 7

.18

~ «a

I

I IP

T 0re

st@i5

min

(Pa)

= 4

94.6

7 +

208

.625

Vca

- 1

55.8

75 ^

-12

7.87

5 S/

A

Eq.

7.1

9

o IP

TQ

rest(t

) (P

a/m

in)

= 7

.1 +

3.7

6 V

ca -

3.4

025

<)) -

2.2

075

(|).V

ca

Eq.

7.2

0

151

Cha

pter

7 E

ffec

t of

SCC

m

ix d

esig

n on

form

wor

kpre

ssur

e ch

arac

teri

stic

s

Tab

le 7

.10

Stat

istic

al m

odel

s in

abs

olut

e va

lues

(()>

= 6

00-7

20m

m,

S/A

= 0

.44-

0.52

by

volu

me,

and

Vca

= 0

.27-

0.33

by

volu

me)

Ko@

H=4m

(%)

= 1

59.3

4 -

0.09

975

<|> +

42.

1875

S/A

- 4

35.8

33 V

ca +

0.5

<|>.

Vca

E

q. 7

.21

| J

K0@

H=8

m (

%)

= 1

76.2

7 -

0.12

8458

3 cb

+ 4

9 S/

A -

589

.333

Vca

+ 0

.654

167

<|>.

Vca

E

q. 7

.22

K0@

H=i

2m (

%)

= 3

3.15

+ 0

.085

2916

7 <j>

+ 5

5.81

25 S

/A -

209

.25

Vca

E

q. 7

.23

O

.—

O.

W

•g |

A

K(t

)(0-

60 m

in)

(%/m

in)

= -

0.8

4667

+ 1

.437

5 S/

A +

4.0

8333

Vca

- 6

.249

9 S/

A.

Vca

E

q. 7

.24

H |

A

K(t

)(0-

tc)

(%/m

in)

= 0

.216

25 -

0.0

001

<|> -

0.1

5625

S/A

+ 0

.291

667

Vca

E

q. 7

.25

to (m

in)

= 1

676

-1.3

5312

5 <|»

- 2

128.

125

S/A

- 1

618.

75 V

ca +

4.1

4062

5 <|>

.S/A

E

q. 7

.26

_ ^

Rhe

omet

erto

rest@

i5min

(Pa)

= -

852

2 +

13.

44 <

|> -

256

8.75

S/A

+ 4

0708

.33

Vca

- 5

4.44

c|).

Vca

E

q. 7

.27

| 1

| 3

||

ATi

app@

N=o

.7n«@

i5min

(Pa

.s)

= 3

42.7

- 0

.564

58 <{

> -

796

.875

S/A

+ 1

754.

17 V

ca

Eq.

7.2

8

~ |

1 %

I

8 R

heom

eter

to re

st(t)

(Pa/

min

) =

- 1

66.6

42 +

0.2

241

$ -

45.9

6875

S/A

+ 8

16.4

17 V

ca -

0.9

7847

(t>

.Vca

Eq.

7.2

9 H

° ^

AT

i app @

N^,

7 ^(

t) (

Pa.s

/min

) =

- 2

03.8

+ 0

.097

tfr +

220

.76

S/A

+ 9

23.1

58 V

ca -

0.4

94 (

|).V

ca -

952

.8 S

/A.V

ca

Eq.

7.3

0 tt~

^ PV

x 0re

st@i5

min

(Pa

) =

158

0 -

2.59

<|> -

319

6.88

S/A

+ 7

337.

5 V

ca

Eq.

7.3

1

I -a

•§

"§

PVTo

rest(

t) (P

a/m

in)

= -

465

.98

+ 0

.013

7 $

- 12

8.31

3 S/

A +

380

2.75

Vca

+ 1

.27

<|>.

S/A

- 2

.752

78 (

|).V

ca -

313

9.58

S/A

.Vca

Eq.

7.3

2

H «

a |

I IP

Tores

t@i5m

in (

Pa)

= 1

657.

5 -

2.59

792

<|> -

319

6.87

5 S/

A +

695

4.16

7 V

ca

Eq.

7.3

3

§ IP

xo rest

(t) (

Pa/m

in)

= -

23

5.9

+ 0

.311

2 (()

+93

4.75

Vc

a-

1.22

639

<|).V

ca

Eq.

7.3

4

152

B. Relative errors of derived models

The central mixture (SCC33) was tested four times to evaluate the experimental errors for

the developed models. Table 7.11 shows the mean (x), standard deviation (a), standard error

corresponding to 95% confidence limit (SE), and relative error corresponding to 95% confidence

limit (RE) (Eq. 5.1) for the four measured responses. The RE for the pressure characteristics

varied between 1% and 6%; however, it varied from 3% to 12% for the T.I.

Table 7.11 Repeatability of test results (n = 4)

sure

ra

l pr

es

Lat

e

P

.1. f

roi

H

6 o son

char

acte

risi

~o

todi

fie

fa

ld-

fie

_

atte

rsa

H

<1> -4-»

T3

G

O

MK

-II

lods

m

et]

<L)

oncr

et

o

eom

et

•fi

Predicting model

Ko@H=4 m

Ko@H=8 m

Ko@H=12 m

AK(t)(0-60 min)

AK(t)(0-tc)

tc

Rheometerio rest@i5min

Ar)app@N=0.7 rps@15min

RheometerTo rest(t)

Ar|app@N=0.7 rps(t)

PVxo rest@15min

PVTorest(t)

IPxo rest@15min

IPXOrest(t)

X

81.5

66.7

52

0.17

0.16

586

446

88.8

10.5

8.44 448

25.5

380

6.5

a

1.24

0.95

1.3

0.001

0.005

20

32.7

1.9

0.69

0.37 20

1.56

22

0.26

SE

2

1.5

2.1

0.002

0.008

32.3

52

3

1.09

0.59 32

2.48

35

0.42

R E ( 0 /

2.4

2.3

4

1.4

4.6

5.5

12

3

10

7 7

10

9

6

C. Validation of derived models

Correlations between predicted responses from the derived models and actual

measurements using the four central mixtures are shown in Table 7.12. The ratio between the

measured and predicted responses (x/y) is also presented in this table. The predicting models

of lateral pressure characteristics produced excellent x/y values (0.95 to 1.03). Good x/y

values were found for the various T.I. (0.70 to 1.1), except for the model of IPto rest(t) that

resulted in weak x/y far from one (0.41 to 0.47).

153

Cha

pter

7 E

ffec

t of

SCC

mix

de

sign

on

form

wor

k pr

essu

re c

hara

cter

isti

cs

Tab

le 7

.12

Mea

sure

d ve

rsus

pre

dict

ed r

espo

nses

usi

ng d

eriv

ed s

tatis

tical

mod

els

Stat

istic

al m

odel

s M

easu

red

resp

onse

(x)

Pr

edic

ted

Xl

83

68

52

X2 82

67

52

x 3

80

65

50

X4 81

67

54

resp

ons

82

68

53

xi/y

x 2

/y

x 3/y

xV

y

11)

lH

=3

w

vi

<1> O,

i—H

«J

VH

<D

V) o

V

• »—

1

CLI

c>

(Tt

l-l

C3

Ko@

H=4

m (

%)

Ko@

H=8

m (

%)

Ko@

H=1

2 m

(%

)

1.02

1.00

0.97

0.99

0.98

0.97

0.98

0.97

0.94

0.99

0.99

1.00

a

AK

(t)(

0-60

min

) (%

/min

) 0.

172

0.16

9 0.

169

0.16

9 0.

168

1.02

1.

00

1.00

1.

00

AK

(t)(

0-t c)

(%

/min

) 0.

158

0.15

9 0.

168

0.16

0 0.

160

0.99

0.

99

1.05

1.

00

t c (m

in)

590

608

559

588

588

1.00

1.

03

0.95

1.

00

61

ls^

tM

2c

£7

s s

c o

o

o

Rhe

omet

en0r

est@

i5m

in(P

a)

469

429

478

409

Ar(a

pp@

N=o

.7rps

@i5m

in (

Pa.s

) 89

86

90

90

R

heom

eter

Tor

est(t

) (P

a/m

in)

10.7

11

.4

10.1

9.

9

An a

pp@

N=0

. 7rp

s(t)

(Pa.

s/m

in)

8.29

8.

97

8.12

8.

37

549

114

10.4

8.33

0.85

0.78

1.03

1.00

0.78

0.76

1.10

1.08

0.87

0.79

0.97

0.98

0.75

0.79

0.95

1.00

Vi

SU

vi

ft

'

"*•'

T3

*T3

££

o

PV

T0re

st@

15m

in(P

a)

PVxo

rest(

t) (P

a/m

in)

437

25.2

425

26.9

461

23.5

468

26.5

537

26.8

0.81

0.94

0.79

1.01

0.86

0.88

IPx0

rest

@15

min

(Pa)

IPxo

rest(

t) (P

a/m

in)

397

3.11

387

2.89

388

3.30

347

2.97

495

7.09

0.80

0.44

0.78

0.41

0.78

0.47

0.87

0.99

0.70

0.42

154


D. Correlations between modeled responses

Virtual values generated from the derived models for virtual values of the modeled

parameters (4>, S/A, and Vca). This was done to compare various responses. The responses of

the thixotropy indices were correlated to those of lateral pressure characteristics.

Correlations among various thixotropy indices were also established. For example, the

PVTorest@i5min was correlated to K0@H=4m, K0@H=8m, and K0@H=i2m, as that shown in Fig. 7.17.

The other correlations are included in Appendix C4. Figure 7.17 shows that the increase in

the PVTorest@i5min reduces the Ko values. The increase in the casting depth results also in a

reduction of Ko values. The various relationships reported here and those in the Appendix

C4 yielded R values greater than 0.88. These correlations indicate that the responses of the

lateral pressure characteristics can correlate well to the responses of the thixotropy indices,

and also, there are good correlations between the various thixotropy indices.

£ S.*'

*r •o

u IS

u QH

100

90

80

70

60

50

40

Fig. 7

K0@H=4m(time = 24rnin) >2 :

:0@H=8m(time = 48min) 2 = 0.96

(time = 72 min)

R2 = 0.97

0 200 400 600 800 1000 1200 Predicted PVT0 r£St@15 min (Pa)

. 17 Relationship between predicted Ko and PVTQ rest@i5 min values

E. Contour diagrams for the statistical models

Contour diagrams were established as a simple interpretation for the derived statistical

models. The contour diagrams are used to compare the trade-off between the effects of the

different parameters on the considered responses. The contour diagrams for six statistical models

are presented in Figs. 7.19 to 7.24 (contour diagrams for other models are presented in Appendix

C5). The two dimensional contour charts constructed here show how the response varies with the

variation of two parameters at a time for a given value of the third one. For example, two contour

diagrams for Ko@H=8m model were established by varying § and Vca at given values of S/A of 0.44

155

and 0.48, respectively. Also, two other diagrams for the same response were established by

varying S/A and Vca at given values of <|> of 600 and 720 mm, respectively. In the contour

diagram of Ko@H=8m at S/A of 0.44 [Fig. 7.18 (b: left)], the increase in (j> values results in an

increase in the Ko@H=8m values. On the other hand, increasing the Vca ratio leads to a reduction in

Ko@H=8m values. Now considering the two charts in Fig. 7.18 (b) and for certain values of Vca and

((), the Ko@H=8m values are found to increase when the S/A changed from 0.44 to 0.48.

156

Cha

pter

7 E

ffec

t ofS

CC

mix

des

ign

on fo

rmw

orkp

ress

ure

char

acte

rist

ics

0.3

3

0.27

0.44

0.

45

->-f

ig-

L

0.3

3

0.3

3

a .2

0.32

0.31

0.46

0.

47

0.48

0.

49

0.5

0.51

0.

52

Sa

nd

-to

-to

tal

ag

gre

ga

te r

ati

o (S

/A)

(dim

ensi

on

less

)

(a)

Eff

ect

of §

on

K0@

H=8

m

0.3

3

0.45

0

46

0.47

0.

48

0 49

0.

5 0.

51

Sa

nd

-to

-to

tal

ag

gre

ga

te

rati

o (S

/A)

(dim

en

sio

nle

ss)

0.52

60

0.3

0.29

0.28

o > 0

.27

' j

f

^

^$r

-<&

*S

' *

& s

^ K

0@

H=

8m

,%

S/A

= 0

.44

&

gjh

J&'

-*T

^I

i i

***

&

jS

1

-

y£

y-6

00

62

0 6

40

66

0 6

80

Slu

mp

-flo

w d

iam

eter

(m

m)

700

720

0.27 600

620

640

660

680

Slu

mp

-flo

w

dia

me

ter

(mm

) 7

00

72

0

(b)

Eff

ect

of S

/A o

n K

0@H

=8m

F

ig.

7.18

T

rade

-off

of

diff

eren

t pa

ram

eter

s af

fect

ing

Ko@

H=8

m

157

Cha

pter

7 E

ffec

t ofS

CC

mix

des

ign

on fo

rmw

ork

pres

sure

cha

ract

eris

tics

0.33

1 0.

32

•0.3

1

<"

0 3

0.29

0.28

0.27

0.44

0.3

3

.2

0.3

2

-S

S

0-31

& °3

0.29

g

0.28

3

>.

^

^"

»•>*

-

vO-

0>

*

^'

'''A

K(t

)(0

-tc

min

),

%/m

in

4> =

72

0 m

m

,.-

^*

'

-HO

-N

*>*_

,^

^ ^

-

,^

0.33

AK

(t)(

0-t

cm

in),

%/m

in

0.45

0.

46

0.47

0.

48

0.49

0.

5 0.

51

0.52

S

an

d-t

o-t

ota

l a

gg

reg

ate

ra

tio

(S/A

) (d

imen

sio

nle

ss)

j i_

0.45

0.

46

0.47

0.

48

0.49

0.

5 0

51

0.52

S

an

d-t

o-t

ota

l a

gg

reg

ate

ra

tio

(S/A

) (d

ime

nsi

on

less

)

(a) E

ffec

t of<

J» o

n A

K(t

)(0-

t c)

0.27

^ • ^^

*.«

• ^

' ^

'

^ A

K(t

)(0-

t cm

in),

%/m

in

S/A

= 0

.44

__.«

!&'"

.0-^

0-«

-

^

^s>*

0.3

3

600

620

640

660

680

Slu

mp

-flo

w

dia

met

er

(mm

) 700

720

0.27 600

620

640

660

680

Slu

mp

-flo

w

dia

met

er

(mm

) 7

00

72

0

(b) E

ffec

t of

S/A

on

AK

(t)(

0-t c)

F

ig. 7

.19

Tra

de-o

ff o

f di

ffer

ent

para

met

ers

affe

ctin

g [(

AK

(t)(

0-t c)

]

158

Cha

pter

7 E

ffec

t of

SCC

mix

des

ign

on fo

rmw

orkp

ress

ure

char

acte

rist

ics

0,3

3

1 0-

32

oe

00

0.3

10.2

9

S

0.2

8 3 o > 0

.27

2P°

Po

rta

ble

va

ne

, T

ore

s„

Pa

<t> =

60

0 m

m

"^

3-

lOQ

,^

" eP

°*

-too

^sE

E.

0.3

3

&

0.3

k

0.29

0.2

8 r-

o >

0.4

4 0.

45

0.46

0.

47

0.48

0.

49

0.5

0.51

0.

52

San

d-t

o-to

tal

aggr

egat

e ra

tio

(S/A

) (d

imen

sion

less

)

5?

~

Po

rta

ble

va

ne

, To

r.s„

Pa

<t> =

72

0 m

n «5J°

<$£>

*»°

>$>o

'

a»°

>2

P°'

0

27

' 0

.44

0.3

3

c £ 0

.32

2*

0.3

1 -

If0

3

0.2

9 -

0.2

8 -

o > 0

.27

^^

^$§s

_,**

'

>«J>

°' •&

- ..«*»

""

.

/^

"

,^~-

* .w

-"3"

-tcP

«pp

.^sP

""

^Po

rta

ble

va

ne

.

S/A

= 0

.44 -VC

f*

_#> ^*P

°^^

JS

F

^o

res

t* P

a -^

«*#>

^-

^

-^

f

^-

(a)E

ffec

tof(

|)onP

VT

o re

st@

15m

in

0.3

3

3 0

.32

OS

s •a

S

0.3

1 O

i

CS

120

a>

00

UJ

r

oo

a « 0.

231-

0.2S

-

60

0 6

20

70

0 7

20

0.2

7 60

0 6

20

640

660

680

Slu

mp

-flo

w d

iam

eter

Cm

m)

(b) E

ffec

t of

S/A

on

PVx 0

res

t@15

min

Fig.

7.2

0 T

rade

-off

of

diff

eren

t pa

ram

eter

s af

fect

ing

PVxo

resi

640

660

680

Slu

mp

-flo

w d

iam

eter

(m

m)

70

0

0.45

0.

46

0.47

0.

48

0.49

0.

5 0.

51

0.52

S

and

-to-

tota

l ag

greg

ate

rati

o (S

/A)

(dim

ensi

on

less

)

--'

J$p

'#*>

<§§

>

^r

__-x

^

#>

#$>

, ^

A&

"' P

ort

ab

le v

an

e,

x ora

sU

S/A

= 0

.48 •

^

^^

Pa

&P

.$&

^^

^^

. ^

^ 72

0

: 15

min

159

Volume of coarse aggregate (dimensionless) o

Volume of coarse aggregate (dimensionless)

U a 5

Volume of coarse aggregate (dimensionless) p OJ

o w

o Ui 0)

o -e-o

< H o

Volume of coarse aggregate (diinensionless) Volume of coarse aggregate (diinensionless)

S §

ft 3 O 3. r+ 3 •—»•

CTQ M

H o

Vohune of coarse aggregate (diinensionless) '-a H o

Volume of coarse aggregate (diinensionless)

•S

re*

re

Co'

o s

s o s

s: o

re

re

£5

re <-* re

re Co


1A Test results of Phase II

7.4.1 Effect of V p on S C C formwork pressure

The variations of K0 with casting depth (H) for SCC made with V p of 340, 370, and 400 1/m3

(SCC36, SCC38, and SCC40, respectively) are presented in Fig. 7.23 (left). The CC35 is also

included in the figure. The SCC40 of higher V p of 400 1/m3 showed pressure envelop greater than

that noted for SCC36 of lower V p of 340 1/m3. A 2 6 % to 3 5 % reduction in K0 is obtained for the

CC35 compared to SCC38 mixture of the same composition, except for the dosage of HRWRA.

Between two given H, the reduction in Ko for a mixture made with higher V p was found greater

than that for mixture made with lower Vp . For example, between H of 1 and 7 m, the reduction in

the Ko values for SCC40 of 400 1/m3 Vp was about 24%, but only 1 1 % was recorded for SCC36

made with V p of 340 1/m3.

The Ko at H of 3 and 7 m are shown to increase with V p , as shown in Fig. 7.23 (right). The

reduction in Vp results in an increase in aggregate volume, hence an increase in internal friction

leading to less lateral pressure.

1.0

0.8

^ 0 . 6

0.4

0.2

SCC40 (V. = 400 1/m3)

SCC38 (Vp = 370 1/m3)

CC35 (Vp = 370 1/m3)

R = 10 m/hr

0.9

0.8

0.7 P '0.6

SCC36 (Vp = 340 1/m3) 0.5

0.4

" H R2

-

= 3 m A

= 0.84>>

• L

g^R = R2 =

= 7m = 0.82

I

*A

0 1 2 3 4 5 6 7 8 Concrete height (m)

Fig. 7.23 Variations of Ko with concrete height (left) and paste volume (right)

320 340 360 380 400 420 Paste volume (L/m3)

Variations of Ko with t ime obtained using the 1.2-m high PVC column for SCC made with

various V p values are shown in Fig. 7.24 (left). The two pressure decay indices at 60 min and

until pressure cancellation are presented in Fig. 7.24 (right). The pressure decay is shown to

become sharper with the increase in V p given the effect of cement content on the rate of

restructuring (both reversible and non-reversible). For a given Vp , the pressure decay in the first

162

60 min was found sharper than the average decay during the entire period of pressure

cancellation. The SCC36 made with Vp of 340 1/m3 showed 0.5 and 0.177 %/min for the

[AK(t)(0-60 min)] and [AK(t)(0-tc)], respectively. In the first hour, both physical (restructuring of

the particles) and chemical (cement hydration) actions take place versus chemical hydration at

later age, resulting in faster rate of decay in the early age.

The pressure cancellation time (tc) is practically independent of the concrete height. tc for the

CC35 mixture was the minimum among the tested concretes at approximately 170 min. The longest

tc value was 525 min recorded for the SCC46 mixture. tc decreases with the increase in the Vp , as

shown in Fig. 7.25.

1.0

0.8

0.6

0.4

0.2

0.0

~

wj

-

( V P

SCC40 (Vp = 400 1/m3)

SCC38 ^ ^ t a ^ ^ ^ * * = 3701/m3) ^ ^ u £ >

i i i T

SCC36 (Vp = 340 1/m3)

1.0

0.8

-1 0.6 0 s

£ 0 . 4

^ 0 . 2

0.0

r

AK(t)(0-60 min) Q

R2 = 0.98 J0<^

O ^ ^ AK(t)(0-tc) R2 = 0 .9 jU a " P

•.- if

i i i i

0 60 120 180 240 300 360 420 3 3 ° 3 5 ° 3 7 0 3 9 0 4 1 0

Time (min ) Paste volume (1/m3)

Fig. 7.24 Variations of Ko with time for SCC mixtures of different paste volume (left) and

pressure decay indices with paste volume (right)

500 s

tio

a ,—s

cell

min

a w « u « +* « £ ? B> % '*3

^ OH

400

300

200

100

0

R2 = 0.97

330 340 350 360 370 380 390 400 410 Paste volume (1/m3)

Fig. 7.25 Relationship between pressure cancellation time and paste volume

163

Chapter 7 Effect of SCC mix design on formwork pressure characteristics

7.4.2 Effect of Vp on thixotropy (breakdown area)

The breakdown area determined between 0-30 min (Abi) (calculated as per Eq. 2.69) is

correlated to Vp in Fig. 7.26. As expected, the mixture of high Vp develops low thixotropy as the

relative content of aggregate diminishes with the increase in Vp. For example, increasing Vp from

340 to 400 1/m3 decreased the Abi values by 430 J/m3.s.

800

*? 600

s

^ 400

< 200

0

330 340 350 360 370 380 390 400 410 Paste volume (1/m3)

Fig. 7.26 Effect of paste volume on the breakdown area between 0-30 min

7.4.3 Relationships between breakdown area and Ko

The relationships between breakdown area as a thixotropic index and variation of Ko values

at casting heights of 1, 3, and 7 m are illustrated in Fig. 7.27. Reductions in Ko with the increase

of Abi and (Ab2 - Abi) at various concrete heights were observed. The R values for (Ab2 - Abi)

were greater than those for Abi, and SCC of higher thixotropy exerts then less pressure on the

formwork. The higher thixotropy level can be achieved by proportioning the SCC mixture with

lower paste volume.

7.5 Test results of Phase III

Investigating the effect of MSA on lateral pressure characteristics was carried out using two

series of SCC of different thixotropy levels. In each series, three SCC mixtures proportioned with

MSA of 10, 14, and 20 mm were tested: SCC13, SCC5, and SCC9 in series # 1 (low thixotropic

SCC), SCC58, SCC59, and SCC60 in series # 2 (low thixotropic SCC). The concrete was placed at

R of 10 m/hr and T of 22 ± 2°C. The structural build-up at rest was determined using the PV test at

15 min resting time and was found to be 450 and 763 Pa for SCC5 (series # 1) and SCC59 (series #

2), respectively. Both of these mixtures were made with 5-14 mm coarse aggregate.

^Ab, =-7.13 VMA+3089 R2 = 0.84

164


^ v

rf

100

90

80

70

60

50

40

1(

;

)0

^0@H=lm 1.017- 0.0004 Ab, X r ^ - R2 = 0.61

x-*^^V xx

O "••*•* J£

Ko@H=7m = 0.715-0.0002 Ab, • R2 = 0.34

1 1

300 500

Abj (J /m

K-O@H

'

3.s)

=3m = 0.909 - 0.0003 Ab, — R2 = 0.50

^ ^ £ O

i i

700 900

100

90

80

^ 70

60

50

40

K0@H=lm = -0.072(Ab2-Ab1) + 89 R2 = 0.73

058(Ab2-Ab,) + 83

K0@H=7m = -0.035(Ab2-Ab1) + 68 R2 = 0.61

50 100 150 200 (Abj-Abj) (J /m 3 . s)

250 300

Fig. 7.27 Effect of breakdown area on Ko

Variations of lateral pressure profiles for series # 1 and # 2 using the U01S2 pressure

column are presented in Fig. 7.28. The increase of MSA from 10 to 14 mm and further to 20 mm

reduced the lateral pressure. For the low thixotropic SCC, the value of Ko was hydrostatic up to 4

m in depth. Deviation from hydrostatic values was noted at deeper casting heights in the case of

thixotropic mixtures. It is worth noting that the HRWRA concentrations were adjusted to secure

the target slump flow of 660 and 600 mm for the SCC mixtures in series # 1 and # 2,

respectively. There was not significant change in HRWRA demand for SCC mixtures in each

series, as shown in Table C.3 (Appendix CI). Therefore, the HRWRA does not play a role on the

lateral pressure variation. The coarse aggregate with MSA of 20, 14, and 10 mm had packing

165

density values of 0.63, 0.62, and 0.56, respectively. The increase in packing density can lead to a

reduction in Ko since this can lead to an increase the inter-particle friction.


2 4 M)

a 6

8 § 8

10

12

14

Low thixotropic SCC

(|> = 660±13 mm R=10m/hr

Hydrostatic "\ pressure

SCC13 \ ( M S A = 10 mm)

SCC5 (MSA= 14 mm)

0

2

? r 4 JS b£

J3 6 4*

t 8 a o

° 1 0

12

14


100 200 300

X* X > X. *

-

SCC60 [MSA = 20

>

mrr

SCC59 (MSA =14

• i

High thixotropic SCC

R =

\ \ \ \ \ *

V, \\\ 1 1

Hi 11

mm)

600 ±13 mm = lOm/hr

Hydrostatic k s pressure

\ \ \ \ > \ \ \ \ \ \

"~~"""~~- SCC58 \ (MSA= 10 mm)

(a) Series #1 (b) Series #2

Fig. 7.28 Variation of initial lateral pressure with concrete height for SCC proportioned with

different MSA

From the differences in Ko values versus casting height for SCC made with different MSA

values, correction factors were proposed in Fig. 7.29. For SCC of relatively low and high

thixotropy levels were therefore deduced and are expressed as follows:

[email protected] < 7 0 0 P a

K0(MSA = 10 mm) = K0(MSA = i4mm) + (1-26 H - 5.04) Eq. 7.35

K0(MSA = 20 mm) = K0(MSA = M mm) - (0.35 H + 1.4) Eq. 7.36

P V w ^ ^ ^ O O P a

K()(MSA = 10 mm) = Ko(MSA = 14 mm) + ( 0 . 1 1 H - 0 .46 ) E q . 7 .37

Ko(MSA = 20mm) = Ko(MSA= 14mm) - ( 0 . 1 7 H + 0 .67 ) E q . 7 .38

For deep placement up to 13 m, the spread of Ko due to the changes in MSA was limited to

4% in the case of Eqs. 7.40 to 7.42. This variation was within the experimental error of the UofS2

pressure column prediction of Ko at 12 m (Table 5.3). Therefore, there is no need to consider any

166

Chapter 7 Effect of SCC mix design on formwork pressure characteristics

correction factor for SCC made with different MSA values when the concrete has relatively high

thixotropy (PVxorest@i5min > 700 Pa) or when it is low thixotropy and is proportioned with MSA of

20 mm. On the other hand, the spread in Ko was more than 4% in Eq. 7.35. A correction factor

(fMSA) is then considered for low thixotropic SCC with 10 mm MSA, as follows:

> For relatively low thixotropic SCC [P V XQ rest @is mm 5- 700 Pa]

H < 4 m fMSA = 1

H = 4 - 1 2 m fMSA = \ whenMSA = 20mm

, 1.26 H-5.04 fMSA = 1 + — when MSA = 10 mm

> For high thixotropic SCC [PVr0rest@is mm > 700 Pa]

H = l - 1 2 m fMSA=l when MSA = 10 and 20 mm

20

16

12

8

.o 4

0

-4

-8

-12

-16

-20

£

jSeries # 1: SCC of low thixotropy

y = 1 . 2 6 x - 5 . 0 4 , ^ ' '

** Concrete height (m) yo

20

15

10

5

2 4 6 8 10 y=0.35x+1.4

— »MSA= 10 mm -M—- MSA = 14 mm ——MSA = 20 mm

-5

-10

-15 -

-20 -

Series # 2: SCC of high thixotropy

y = 0.114x Concrete height (m)

4 6 8 10 12 14 y = 0.167x

Fig. 7.29 Recommended modification coefficients of Ko with changes in MSA

7.6 Conclusions

Based on the above results, the following conclusions can be drawn:

1. The maximum lateral pressure exerted by SCC is lower than hydrostatic pressure. Large

deviation from hydrostatic pressure can be observed with the increase in thixotropy. For

example, SCC30 with PVxorest@i5min of 815 Pa, cast at 10 m/hr and 22°C, can have a Koi value

at 12-m depth as low as 30%.

167

2. The lateral pressure decay is sharper initially than the average decay until pressure

cancellation. For example, SCC with PVxorest@i5min of 740 Pa, showed an initial decay over

60 min [AK(t)(0-60 min)] of 0.22 %/min compared to 0.17 %/min for the mean rate of

pressure decay until pressure cancellation [AK(t)(0-tc min)].

3. As expected, conventional concrete exhibited lower Ko values than SCC of similar mix

design, but with high HRWRA dosage. The KO@H=12 m for CC34 and SCC33E mixtures were

23% and 50%, respectively. The conventional concrete also had faster pressure decay than

that of corresponding SCC mixture.

4. The increase in slump flow (<(>) of SCC due to the addition of HRWRA (the same mix

design) is shown to reduce thixotropy and increase lateral pressure. This can be illustrated

from the SCC27 and SCC31 had (|) values of 600 and 720 mm, respectively, as shown below.

Mixture

SCC27

SCC31

-e-

600

720

Ko@H=4m

85

89

Ko@H=8m

71

77

AK(t)(0-tc) (%/min) PVx0

0.15

0.14

Decrease of paste volume (Vp) or increase of coarse aggreg

in thixotropy and decrease in maximum lateral

rate, and pressure cancellation takes place later

Mixture

SCC36

SCC38

SCC39

V p 3 (1/m3)

340

370

390

v c a (1/m3)

319

305

295

Ko@H=3m

(%)

64

75

87

rest@15min

(Pa) 327

210

ate volume (Vca) leads to an incr

pressure. The lateral pressure decays at a sic

on when Vp

Ko@H=7m

(%)

58

65

74

decreased, as shown below.

AK(t)(0-tc) (%/min)

0.177

0.265

0.312

Abj (J/m3.s)

543

440

230

6. For a given paste content, the decrease in S/A produces SCC mixture of high thixotropy and

lower lateral pressure.

7. Increasing the MSA results in SCC mixture of high thixotropy and lower lateral pressure. A

correction factor (/ksx) as function of casting depth (H) was proposed to account for the effect

of MSA other than 14 mm on lateral pressure of SCC. The f^sA is 1.0 for SCC made with

different MSA values when the concrete has relatively high thixotropy (PVxorest@i5min > 700

Pa) or when it is low thixotropy (PVxorest@i5min < 700 Pa) and is proportioned with MSA of

20 mm. On the other hand, for low thixotropic SCC with 10 mm MSA, the fMSA is in order of

[l+(1.26H-5.04)/100].

168


8. The various thixotropic indices [T.I.@i5mjni, T.I.(t), and T.I.@i5minix T.I.(t)] obtained using the

concrete rheometer and field-oriented test methods correlate well to Ko at different depths

and pressure decay. Tables 7.4 and 7.5 summarize the R2 values for the various correlations.

Based on the R values, the following indices are recommended for the prediction of the

various lateral pressure characteristics:

Pressure characteristics

Ko

AK(t)(0-60 min)

AK(t)(0-tc)

Recommended index # 1

RheometerTorest@15minxXOrest(t)

Ar|app@ N=0.7 rps@15min

Rheometeriorest® 15min xT0rest(t)

Recommended index # 2

PVlOrest® 15min xtOrest(t)

PVxorest® 15min

P VXQrest® 15min xtOrest(t)

9. Statistical models to predict lateral pressure characteristics and thixotropic properties as

function of mix design (slump flow, sand-to-total aggregate ratio, and coarse aggregate

content) were established. The derived models had high R2 values of 0.84 to 0.99.

10. Contour diagrams for predicting responses are established to illustrate trade-offs the effect of

different parameters on the tested responses pertaining to thixotropy and formwork pressure.

11. Based on the validation of the models using the central points, the thixotropic indices that

can be used to capture the lateral pressure characteristics are Anapp@N=o.7rps(t) and PVxorest(t).

169

CHAPTER 8

EFFECT OF PLACEMENT CHARACTERISTICS AND FORMWORK DIMENSIONS

ON LATERAL PRESSURE OF SCC

8.1 Introduction

One of the main attributes of SCC in the industrial applications is the acceleration of the

placement process especially in densely reinforced and restricted sections. Casting rates of SCC can

vary with the size of the cast element and placement method. When the pouring rate is so fast that

no stiffening of the cast concrete is allowed (as in small volume pours that can be completed in one

single lift) formwork pressure could reach hydrostatic values. However, in large structures where

the casting rate is slower, the maximum pressure will be considerably less than hydrostatic.

The CEBTP reported that the lateral pressure envelopes showed a reduction of 30% from

hydrostatic distribution for SCC pumped from bottom at a rate of 25 m/hr and 35% for concrete

cast using a bucket from top at 18 m/hr when casting 12.5-m high experimental wall elements

[CEBTP, 1999]. Studies carried out in Sweden [Skarendahl, 1999] during casting of 5-m high

bridge piers with SCC in successive layers of approximately 0.5 m in height with rest periods

resulted in lateral pressure at the base of the formwork to be half of that resulting from normal-

consistency concrete consolidated by internal vibration.

In addition to casting rate, concrete temperature and formwork minimum dimensions can

influence formwork pressure. Assaad and Khayat [2006] reported that the SCC prepared at initial

temperature of 10, 20, and 30 °C developed similar relative pressure on formwork of 91% at the

end of casting. On the other hand, the rate of pressure drop with time was significantly affected

by the concrete temperature. They recorded 400, 250, and 160 min to reduce the relative pressure

by 25% for the SCC cast at temperatures of 10, 20, and 30 °C, respectively. Limited data exist

regarding the effect of formwork width and reinforcement on lateral pressure characteristics.

The objective of the study presented in this chapter is to investigate the effect of casting

rate and fresh concrete temperature on lateral pressure characteristics. The study aims at

determining the effect of placement rate of SCC of various thixotropy levels on the lateral

pressure characteristics. The study is also planned to examine the influence of short interruption

in SCC placement on lateral pressure.

170

Chapter 8 Effect of placement characteristics andformwork dimensions on lateral pressure ofSCC

8.2 Testing program

This chapter consists of four phases dealing with effect of concrete temperature, casting

rate, waiting period between lifts, and minimum form work dimension on lateral pressure of SCC.

Phase I: Effect of temperature on lateral pressure characteristics of SCC

The scope of work done in this phase is summarized in Fig. 8.1. SCC38 mixture was tested at

temperature of 12, 22, and 30 ± 2 °C. The concrete temperature was adjusted by cooling or heating

the various ingredients at the required temperature 24 hr prior to mixing. The ambient temperature

was also adjusted to be similar to the concrete temperature. SCC38-12, SCC38-22, and SCC38-30

refer to SCC38 prepared at 12, 22, and 30°C, respectively. The SCC38 was compared to a normal

concrete (CC35) with similar mixture composition, except for a lower HRWRA dosage to yield a

slump consistency of 200 ± 20 mm. The CC35 was cast at 22 ± 2 °C. All the mixtures were cast at

rising rate of 5 and 10 m/hr. The mixture compositions of SCC38 and CC35 are given in Table 3.8.

Table 8.1 Variables to evaluate effect of temperature on SCC lateral pressure characteristics

Temperature (°C)

12±2

22 ± 2

30 ± 2

* The second number

Concrete mixtures cast at rate of 5 and 10 m/hr

SCC38-12*

SCC38-22 and CC35-22

SCC38-30

in the mixture codification refers to concrete temperature

The values of initial slump flow (or slump consistency) and those at 120 min, initial T50,

J-Ring spread at 10 and 40 min, air content, temperature, and unit weight are given in Appendix

Dl. The UofSl pressure column was used to monitor variations of initial lateral pressure, and the

1.2-m high PVC column was employed to determine the pressure decay. The modified Tattersall

MK-III concrete rheometer was used to determine the rheological properties.

Phase II: Effect of casting rate on SCC lateral pressure characteristics

The experimental work of this phase was done with the collaboration of Mr. Yacine

Elaguab, master student, concrete group, UofS, [Elaguab, MS 2008]. The influence of casting

rate on formwork pressure characteristics exerted by SCC was investigated using eight SCC

mixtures of different thixotropy levels (Table 8.2). The mixture proportions are given in Table

3.8. The initial slump flow values were adjusted at 700 ± 20 mm and the concrete temperature at

22 ± 2 °C. Six casting rates varying from 2 to 30 m/hr were investigated. Therefore each mixture

171

Chapter 8 Effect of placement characteristics and formwork dimensions on lateral pressure qfSCC

was batched seven times; six batches for each casting rate and the seventh batch to determine the

different thixotropic indices of the concrete.

The mixtures were characterized for workability and formwork pressure characteristics

similar to those carried out in Phase I. The test results for workability properties are given in

Appendix Dl. The thixotropic indices were performed using the modified Tattersall MK-III

concrete rheometer and the PV and IP field-oriented tests.

Table 8.2 Parameters to evaluate effect of casting rate on SCC lateral pressure characteristics

SCC mixtures listed in ascending Casting rate, R (m/hr) order according to thixotropic level 2 5 10 17 24 30

1-

2-

3-

4-

5-

6-

7-

8-

SCC40 (low

SCC51

SCC52

SCC53

SCC54

SCC46

SCC55

SCC56 (higr

thixotropy)

1 thixotropy)

V V V V V V V V V V V V V V V V

V V V V V V V V

V V V V V V V V

V V V V V V V V

V V V V V V V V

Temperature 22 ± 2 °C, <|> = 700 ± 20 mm

Phase HI: Effect of waiting period between successive lifts on SCC formwork pressure

SCC57 and SCC59 mixtures of different thixotropy levels were used to evaluate effect of

interruption of casting by a waiting period (WP) between successive lifts. The compositions of the

two mixtures are given in Table 3.8. Summary of the work done in Phase III is shown in Table 8.3.

The two SCC mixtures were continuously placed at rate of 10 m/hr and temperature of 22 ± 2 °C

using the UofS2 pressure column. On the other hand, the two concretes were cast at the same rate

with a WP of 30 min at the middle of casting. After casting one third of the column (UofS2), a 30-

min WP was observed. The same was repeated after casting two thirds of the column.

Table 8.3 Methodology to evaluate effect of waiting period between lifts on SCC lateral pressure

Casting Continuous One WP of 30 min at Two WP of 30 min after one and condition casting middle of casting two thirds of casting height

Notes SCC57 (§ = 640 mm) and SCC59 (<|> = 600 mm) were used in casting R=10m/hr ,T = 22±2°C

172

Chapter 8 Effect of placement characteristics and formwork dimensions on lateral pressure ofSCC

Phase IV: Effect of minimum formwork dimension on lateral pressure characteristics

Plywood formwork (Fig. 4.5) was used to investigate the effect of minimum cross-sectional

dimension (Dmjn) of formwork on initial lateral pressure of concrete and its decay until hardening.

The formwork was configured with four Dmjn, as shown in Table 8.4. For each configuration, the

concrete was continuously cast from top using buckets at 10 m/hr and a temperature of 22 ± 2°C.

Two SCC formulations (SCC38 and SCC62) and one concrete of conventional slump (CC35) were

used. The mixture proportions of the three mixtures are given in Table 3.8.

The fresh concrete properties for the different castings are shown in Appendix Dl.The initial

slump flow values for SCC38 and SCC62 were 700 ± 20 mm and 650 ± 25 mm, respectively. The

slump consistency of CC35 was 200 ± 20 mm. The thixotropic properties were determined using

the modified Tattersall MK-III concrete rheometer and PV and IP field-oriented tests.

Table 8.4 Parameters to evaluate effect of formwork dimension on lateral pressure characteristics

Cross-sectional dimensions (mm)

200 x 400 mm 250 x 400 mm 300 x 400 mm 350 x 400 mm

Three sensors located at depths of 0.95, 1.25, and 1.45 m were employed. SCC38, SCC62, and CC35 were used. R=10m/hr ,T = 22±2°C.

8.3 Test results of Phase I

8.3.1 Effect of concrete temperature on Ko

The increase in the concrete temperature can lead to an increase in the rate of cement

hydration and cohesiveness. The increase in the cohesion enables the plastic concrete to develop

higher shear strength and hence lower lateral pressure. Variations of maximum lateral pressure

(Pmax) with concrete height for the various mixtures cast at placement rates (R) of 5 and 10 m/hr

are illustrated in Fig. 8.1. The increase in the concrete temperature can lead to lower Pmax values.

For example, at R of 5 m/hr and H of 7 m, the change in the temperature from 12 to 22, and then

to 30 °C led to reductions in Pmax from 115 to 96, and then to 72 kPa, respectively. The lateral

pressure profiles for mixtures cast at R of 5 m/hr were lower than those cast at 10 m/hr. For

example, the Pmax values at H of 7 m for SCC38-30 cast at R of 5 and 10 m/hr were 72 and 82

kPa, respectively. The lateral pressure profile of CC35-22 was shown to be lower than that of

corresponding SCC mixture. Pmax values of 56 and 96 kPa were recorded for CC35-22 and

SCC38-22 cast at 5 m/hr and H of 7 m, respectively.

173


The decrease in Ko values with the increase in concrete temperature are presented in Fig.

8.2. At R of 5 m/hr, reductions of 19% and 29% were recorded for Ko at H of 3 and 7 m,

respectively, when the concrete temperature increased from 12 to 30 °C. These reductions were

14% and 23% in case of R of 10 m/hr.


50 100 150 200 o

— . 1 1 1 o

R = 5 m/hr

Phyd A SCC38-12 O SCC38-22 X SCC38-30 • CC35-22

Lateral pressure (kPa) 100 200

R = 10 m/hr Phyd

A SCC38-12

O SCC38-22

X SCC38-30

• CC35-22

Fig. 8.1 Lateral pressure envelops for concrete cast at 5 and 10 m/hr and different temperatures

SCC38, H = 7 m

e-SCC38,H = 3m

• CC35, H = 3m

D CC35,H = 7m

15 25 Temperature, T (°C)

35

1.0

0.9

0.8

^ 0 . 7

0.6

0.5

0.4

0.3 R= 10 m/hr

A-SCC38, H = 3m

B-SCC38,H = 7m

• CC35, H = 3m

= 7m

15 25

Temperature, T (°C)

35

Fig. 8.2 Variations of lateral pressure with concrete temperature at casting rates of 5 and 10 m/hr

174


8.3.2 Effect of concrete temperature on pressure cancellation time

The elapsed time until pressure cancellation (tc) for SCC38 and CC35 at different

concrete temperature are indicated in Fig. 26. The values of tc is not affected by the casting

rate of 5 or 10 m/hr. However, the increase in concrete temperature has a marked effect on tc

values. tc values of 435, 345, and 225 min were obtained for SCC38 mixtures cast at 12, 22,

and 30 °C, respectively, as higher temperature accelerates cement hydration. The tc value for

the CC35-22 was 200 min. The difference between tc values of SCC38-22 and CC35-22 was

145 min. This is mainly due to the retarding effect of the HRWRA incorporated at higher

concentration in the SCC38-22 mixture.

500 E

s o

• * - *

Ha

Ol u a

V u s 03 VI 0* u

CK

^^ B •— E

400

300

200

100

SCC38

- • - R = 5 m/hr -B-R=10m/hr

10 15 20 25 Temperature, T (°C)

30 35

Fig. 8.3 Variations of pressure cancellation time with concrete temperature

8.3.3 Effect of concrete temperature on pressure decay

The pressure decay during the overall period of pressure cancellation [AK(t)(0-tc)] for

SCC38 and CC35 cast at different concrete temperatures is presented in Fig. 8.4. The increase

in concrete temperature can lead to sharper pressure decay. A slight increase in the [AK(t)(0-

tc)] value was observed when the temperature was raised from 12 to 22 °C. However, a

significant increase in the [AK(t)(0-tc)] value was monitored during the increase from 22 to 30

°C. For example and at casting rate of 5 m/hr, the SCC38 showed increases of 0.014 and

0.136 %/min in the [AK(t)(0-tc)] values when the temperature increased from 12 to 22 °C and

then from 22 to 30 °C, respectively. The CC35-22 had higher pressure decay than the

corresponding SCC38-22. It was also observed that the SCC30-30 exhibited in pressure decay

approximately similar to that monitored for CC35-22. The higher pressure decay of CC35 was

referred to the relatively shorter pressure cancellation time associated with the CC35 mixture.

175

0.40

fo.35

£o.30 u

2 0.25

*0.20 <

0.15

CC35 • SCC38

R = 5 m/hr

R= 10 m/hr

10 15 20 25 30 35 Temperature(°C)

Fig. 8.4 Variations of pressure decay with concrete temperature

8.4 Test results of Phase II

8.4.1 Effect of casting rate on Ko

Variations of Ko values with casting rate (R) for SCC54 of thixotropy level # 5 (Abj =

735 J/m3.s) are represented in Fig. 8.5. Similar variations for SCC mixtures of various

thixotropy levels are presented in Appendix D2. The figure shows that K0 values increase

with the increase in the casting rate. Greater increase in K0 was observed at shallow concrete

depths. Based on the data in Fig. 8.5, increasing R from 2 to 30 m/hr resulted in an increase in

Ko@H=7m value from 19% to 52%. This increase ranged from 73% to 99% for K0@H=im-

X

o X

o

X

o A.

o

SCC54 Thixotropy level # 5

X K0 @H = 1 m O K 0 !u;\) ••• 3 m

-*-K.0@H = 7m "•-K.0(tf:H= 10 m

10 15 20 Casting rate (m/hr)

25 30

Fig. 8.5 Variations of Ko with casting rate at various depths [SCC54 of thixotropy level # 5]

The trend of increasing Ko with R is more noticeable for SCC mixtures of low thixotropy

levels, as shown in Fig. 8.6 that plotted at depth of 3 m. Similar figures at concrete depths of 7

and 10 m are presented in Appendix D2. This is because concrete with lower thixotropy

176


develops higher lateral pressure. The advantage of a low K0/R value is enabling the increase in

casting rate and hence resulting in faster casting without exerting more lateral pressure on

formwork. The greater lateral pressure exerted due to the higher casting rate can be explained

by the fact that, the material does not have sufficient time to restructure at rest when a high

casting speed is applied, and therefore, vertical pressure due to concrete weight can lead to a

higher lateral pressure.

0 5 10 15 20 25 30 Casting rate (m/hr)

Fig. 8.6 Variations of Ko at H = 3 m with casting rate for SCC of various thixotropy levels

8.4.2 Effect of thixotropy on Ko

Lateral pressure envelops for SCC mixtures of various thixotropy levels are shown in

Fig. 8.7 for casting rates of 2 and 30 m/hr. Other envelops obtained for casting rates of 5, 10,

17, and 24 m/hr are given in Appendix D2. The pressure profiles of SCC mixtures diverge

from the corresponding hydrostatic values as the depth increases. Small deviation of lateral

pressure was noted from hydrostatic pressure for SCC of relatively low thixotropy. For a very

high casting rate of 30 m/hr, the lateral pressure was close to hydrostatic values at shallow

depths of 1 to 3 m. Reduction in Ko was noted beyond 3 m. Such reduction of Ko was

obtained at shallower depths when the SCC was cast at lower R values (2 to 10 m/hr). This

relative reduction was observed from the initial casting depth in case of low casting rate. For

all tested casting rates, the most thixotropic mixture that resulted in the lowest pressure

profiles was the SCC56 mixture.

8.4.3 Abacus between Ko and thixotropic indices

Samples of the relationships (abacuses) between K0 values at different casting depths

(H) and various thixotropic indices are shown in Fig. 8.8. These abacuses were set for a

177

concrete cast at rate of 2 m/hr at 22 °C. Other abacuses for K0 at different H values and

thixotropic indices are presented in Chapter 9 and Appendix E; Chapter 9 deals with modeling

of K0. The established abacuses enable the prediction of K0 values at any H by using one of

the thixotropic indices and hence can be used as guidelines for the design and construction of

formwork systems for casting SCC.

Maximum lateral pressure ( kPa )

50 100 150 200 250

R = 2m/hr Phyd

— B — SCC40(Low thixotropy) —X— SCC52 —*— SCC53 —•— SCC46 —A— SCC55

V—e— SCC56 (High thixotropy)

-a a 0)

- a <u z u & o U

Maximum lateral pressure ( kPa ) 50 100 150 200 250

R = 30 m/hr Phyd

D SCC40 (Low thixotropy) —*— SCC52 —-*— SCC53 —*— SCC46 —A— SCC55 — « — SCC56 (High thixotropy)

Fig. 8.7 Lateral pressure envelops for SCC mixtures of different thixotropy levels cast at 2

m/hr (top) and 30 m/hr (bottom)

178

Cha

pter

8 E

ffec

t of

pla

cem

ent

char

acte

rist

ics

andf

orm

wor

k di

men

sion

s on

lat

eral

pre

ssur

e of

SCC

100

r

R =

2 m

/hr

T =

22

°_C 30

0 60

0 90

0

Ab

j (J

/m3.s

ec)

R =

2m

/hr

T =

22

°C

K,

K

K(

K, 0@

H=l

m

0@H

=2m

0@

H=4

m

0@H

=6m

^0

@H

=8m

Ko@

H=1

0m

1200

15

00

Ko@

H=l

m

Ko@

H=2

m

Ko@

H=4

m

K()@

H=6

m

Ko@

H=8

m

K-0

@H

=10m

400

800

PV

T,

1200

16

00

0 re

st@15

min

(Pa)

2000

100 r

80

^ 60

5 40

20

0

100

80

^ 60

J 40

R = 2m/hr

T = 22 °C

100

200

300

400

^Tlapp

@N

=0.7

rps@

15m

in (

P»

'S)

20

hR =

2m

/hr

T =

22

°C

0

^0@

H=l

m

^o@

H=2

m

Ko@

H=4

m

Ko@

H=6

m

Ko@

H=8

m

^0@

H=1

0m

500

^-0@

H=l

m

Ko@

H=2

m

K()@

H=4

m

Ko@

H=8

m

K-0

@H

=10m

400

800

IPT

,

1200

16

00

2000

0 re

st@15

min

(P

a)

Fig

. 8.

8 A

bac

use

s of

Ko

valu

es a

nd i

niti

al t

hixo

trop

ic i

ndic

es o

btai

ned

usin

g co

ncre

te r

heom

eter

and

fie

ld-o

rien

ted

test

s

179

8.5 Test results of Phase III

SCC57 (of low thixotropy level: PVT0rest@i5min = 445 Pa) and SCC59 (of relatively high

thixotropy level: PVTorest@i5min = 765 Pa) were continuously cast at rate of 10 m/hr, with one

waiting period (WP) of 30 min in the middle of casting time, and with two WP's of 30 min each

after casting one and two thirds of the column height. The variations of lateral pressure with time

for the three casting conditions for the two SCC mixtures are indicated in Figs. 8.9 and 8.10,

respectively. The lateral pressure envelops for the same casting conditions and for two SCC

mixtures are also shown in Fig. 8.11. In case of interruption of the continuous casting with one

WP, 7% and 11% reductions in K0 were obtained for the SCC57 and SCC59 mixtures,

respectively. These reductions increased to 11% and 15% when two WP's were allowed for the

SCC57 and SCC59 mixtures, respectively.

7% reduction in Kn

% reduction in Kn

SCC57 (<|> = 640 mm) R = 10 m/hr

Continous casting -e—Cast with 1 WP of 30 min

Cast with 2 WP of 30 min each

0 20 40 60 80 100 120 Time (min)

140 160

Fig. 8.9 Variations of lateral pressure with time for SCC57 (low thixotropy level) cast at

continuous rate of 10 m/hr and with one and two waiting periods

Based on these results, a correction factor for Ko accounting for effect of WP between

successive lifts {fwp) was introduced. Values of fwp depend on thixotropy level of the tested

mixture and can be determined as shown in Fig. 8.12. For the continuous casting, the/^/> value is

set at one. The fwp can have a value less than one when a waiting period between consecutive lifts

is allowed. The value of the fwp decreases with the increase in thixotropy. For SCC59 mixture

with PVxorest@i5min of 765 Pa, the values fwp were 1, 0.89, and 0.85 for the continuous casting,

180

casting with 1 WT of 30 min, and with 2 WP's of 30 min each, respectively. Therefore, the first

WP of 30 min resulted in considerable drop in Ko, especially for high thixotropic SCC. Further

increase in WP had limited effect on reducing KQ.

11% reduction in Kn

^ 15% reduction in K0

SCC59 (<|> = 600 mm) R = 10 m/hr

Continuous casting

- e - C a s t with 1 WP of 30 min

- * - Cast wit 2 WP of 30 min each

Fig.

0 20 40 60 80 100 120 140 160 180

Time (min)

8.10 Variations of lateral pressure with time for SCC59 (high thixotropy level) cast at

continuous rate of 10 m/hr and with one and two waiting periods


SCC57 (<|> = 640 mm) Continuous casting

•e— Casting with 1 WP Casting with 2 WP Hydrostatic pressure

0

2 E

a 4 « V

2 6 C ©

U 8

10

12


» 1 I I

\ SCC59 (<() = 600 mm)

sC\ ~ ,-> .,, , „ m

\ x ~—*<r~— ^ast witn z w r V \ Hydrostatic pressure

\ \ * \ 1 \

)

I >

\ \ \ \ \ N \ \ \ \ \ \ \ \

\

Fig. 8.11 Lateral pressure envelops for SCC57 of low thixotropy level (left) and SCC59 of high

thixotropy level (right); both cast continuously and with one and two waiting periods

181


1.1

1.0

0.9 a.

0.7

0.6

200 400 600 800 1000

PVT0rest@15min ( P a )

Fig. 8.12 Relationship between the modification factor for waiting period between successive

l i f t s ifwp) a n d PVT0rest@15min

8.6 Test results of Phase IV

The lateral pressure was monitored using two sets of pressure sensors. Each set consists of

three pressure sensors fixed at depths of 1.45, 1.25, and 0.95 m. The first set of sensors was fixed

at 100 mm from the edge of the transverse direction, or the width referred to as A sensors. The

second set of sensors was attached in the middle of the longitudinal direction, or the length

referred to as B sensors. More details about the plywood configurations are shown in Fig. 4.5.

The width (Dmjn) of the formwork was varied at 200, 250, 300, and 350 mm. The effect of

changing Dmin on lateral pressure was monitored using SCC38, SCC62, and CC35 mixtures.

8.6.1 Effect of Dmin on K0

The Ko values monitored by the A and B sensors (KOA and KOB, respectively) located at a

depth of 1.45 m are correlated to Dmjn in Fig. 8.13 for the three mixtures. The Ko values are

shown to increase with the increase in Dmj„. For example, KOA values for SCC62 increased from

74% to 92% when Dmjn increased from 200 to 350 mm. The degree of increase in KOA with Dmjn

is greater than KOB- For example, for SCC62 the increase in Dmin from 200 to 350 mm led to 18%

increase in KOA, but only 7% increase in the case of KOB-

At a given casting depth, the lateral pressure is theoretically distributed equally in all

directions regardless the cross-sectional dimensions. However, differences in lateral pressure due

to change in Dmjn can still take place due to the arching effect, especially in relatively restricted

H = 1 2 m R= lOm/hr

182

Chapter 8 Effect of placement characteristics and formwork dimensions on lateral pressure o/SCC

sections. The arching action takes place due to the friction between concrete and the walls of the

formwork. In the narrow sections, the friction takes place in a large surface area between the

concrete and the two opposite sides of the formwork. Thus, most of the lateral pressure is carried

out by the wall friction on those opposite sides. Only small portion of the stress is transferred

laterally to the supporting wall, as illustrated in [Fig. 8.14 (thin section)]. This can explain the

lower Ko values that monitored by the A sensors [Fig. 8.14 (left)]. On the other hand, in wide

section, the surface wall friction is small and the portion of the lateral pressure supported by the

formwork wall is relatively large [Fig. 8.14 (right)].

100

90

g J 80

70

60

SCC38 . ^ - ^ r ^

Mf •••. '-.B sensor.-'

* _ - * ! W////A

Depth of 1.45 m

X CC35

T , .200 mm

-400 mm-

100

90

~ 8 0

70

60

CC35 -X -X

150 200 250 300 350 400 Formwork width (mm)

150 200 250 300 350 400 Formwork width (mm)

Fig. 8.13 Variations of Ko values with Dmin determined from pressure sensors fixed in transverse

(top) and longitudinal (bottom) directions

Supported by arching effect and wall friction

Lateral pressure exerted on sensor A

Supported by arching effect and wall friction

4

<?

Lateral pressure exerted on sensor A

B

- O

*Q

i£L 200 mm

400 mm

• 400 mm - 400 mm •

Fig. 8.14 Arching action in thin and wide sections

183


For the tested mixtures, the (KOB-KOA) values were found to decrease with the increase of

Dmjn, as shown in Fig. 8.15. At Dmj„ of 200 mm, the (KOB - KOA) values were 13% and 8% for

SCC62 and SCC38, respectively. These values were 2% and 1% for Dmjn of 350 mm. The

decrease in (KOB-KOA) values with the increase in Dmjn for SCC62 was sharper than that of

SCC38. The reduction in (K0B-KOA) between Dmjn of 200 and 350 mm was 11% and 6% for

SCC62 and SCC38 mixtures, respectively. The extrapolation of the curves for the two SCC

mixtures intersected at Dmin of 400 mm. This means that the lateral pressure exerted by the

concrete on all formwork sides is uniform when the cross section becomes square. It is important

to note that both the C35 and SCC38 mixtures which had same mixture proportioning, except for

the higher dosage of HRWRA of the latter mixture, had similar (KOB-KOA) variations with Dmjn.

^

I 09

14 r

12

10

8

6 h

Depth of 1.45 m ,"'B sensor -,

;

150 200 250 300 Formwork width (mm)

350 400

Fig. 8.15 Influence of Dmjn on lateral pressure distribution

8.6.2 Effect of Dmj„ on pressure decay

Variations of average pressure decay during the time required for pressure cancellation

[AK(t)(0-tc)] determined using the A and B sensors with Dmin are presented in Fig. 8.16. The rate of

pressure decay decreased with the increase in Dmjn. This can be explained by the longer pressure

cancellation time (tc) that monitored when Dmj„ increased. For example, increasing Dmjn from 200

to 350 mm resulted in 108 min delay in tc for SCC38. The [AK(t)(0-tc)] measured using the B

sensors were higher than those measured with the A sensors. It was noted that the A and B sensors

monitored the same tc. Thus, the higher [AK(t)(0-tc)] values monitored with the B sensors can be

184

referred to the higher KOB values. The CC35 decayed at a sharper rate than those of SCC mixtures.

This can be due to the longer tc for the SCC mixtures that incorporated greater HRWRA. m

in)

< • :

2<

shor

t dir

. A

? ©

1

0.50

0.45

0.40

0.35

0.30

0.25

0.20

0.15

0.10

B sensor

SCC62

CC35 H

SCC38

150 200 250 300 350 Formwork width (mm)

1

<

0.50

0.45

0.40

0.35

0.30

0.25

0.20

0.15

0.10

400

B sensor'

CC35

SCC38

SCC62

150 200 250 300 Formwork width (mm)

350 400

Fig. 8.16 Variation of pressure decays with Dmjn of formwork

8.6.3 Establishing a correction factor for K0 as function of Dmin

Ko values obtained throughout the whole thesis were obtained using the UofS pressure of

internal diameter (minimum lateral dimension, Dmin) of 200 mm. A correction factor (flumin) to

account for the effect of Dmjn other than 200 mm on the Ko values is derived. The KOA values at

depths of 0.95, 1.25, and 1.45 m for SCC38 and SCC62 were correlated to the Dmin, as shown in

Fig. 8.17. Ko value of 81.3% monitored at Dmin of 200 mm can correspond to flDmin value of 1.0.

185

Chapter 8 Effect of placement characteristics andformwork dimensions on lateral pressure o/SCC

Proportioning Ko values at Dmjn of 250, 300, and 350 mm to Ko and//Dmin values at Dmjn of 200 mm

(81.3% and 1.0, respectively), produces different/i Dmin values. The relationship between flDmm and

Dmin (in mm) is presented in Fig. 8.18. This relationship can be expressed as follows:

/iomin = 0.000968 Dmm + 0.806332 Eq. 8.1

100

70

60

Ko = 0.0787 Dmin + 65.5333

Dmin = 200 mm

/^Variable depths of 0.95, 1.25, and 1.45 m

SCC38 and SCC62


400

Fig. 8.17 Relationship between Ko values and Dn

1.3 e 6 O

52j 1.2

s. o

I 1.0 G O

S 0.9 i-u e

0.8

/ Dmin = 0.000968 Dmin + 0.806332

For average depth

Dmin = 200 mm SCC38 and SCC62


Fig. 8.18 Relationship between correction factor for Ko and Dn

400

8.6.4 Establishing a correction factor for AK(t) as function of Dmj„

The 1.2-m high PVC column and Dmjn of 200 mm was used to monitor AK(t)(0-tc) values.

In order to include the effect of different Dmjn other than 200 mm, a correction (/2Dmin) was

186

Chapter 8 Effect of placement characteristics and for mwork dimensions on lateral pressure ofSCC

introduced. The AK(t)(0-tc)shOrt dir.A values at depths of 0.95, 1.25, and 1.45 m for SCC38 and

SCC62 versus Dmjn are shown in Fig. 8.19. An average value of 0.246 %/min for the AK(t)(0-

tc)short dir.A that monitored at Dmjn of 200 mm can correspond to/2Dmin value of one. Proportioning

the values of AK(t)(0-tc) at Dmin of 250, 300, 350, and 350 mm to that determined at Dmin of 200

produces different/^Dmin values. The relationship between f2vm\n and Dmjn (in mm) is presented in

Fig. 8.20 that can be expressed as follows:

f2Dmin = 1.260353 - 0.001302 Dmjn Eq. 8.2

0.350

*G • 3

s

%/

^^

r.^

•o

C

1 r-"s

-*-<± >—.' ^-s +* s_x

M <

0.300

0.250

0.200

0.150

0.100

0.050

0.000

x AK(t)(0-tc) = -0.00032 Dmin + 0.30982

0.246 %/min

Variable depths of 0.95, 1.25, and 1.45 m

Dmjn = 200 mm SCC38 and SCC62

150 200 250 300 350

Formwork width (mm)

Fig. 8.19 Variation of pressure decays with Dmjn of formwork

1.1 /?_ . = 1

f2Dmin = 1.260353 - 0.001302 D n

400

150 200 250 300 350

Formwork width (mm)

Fig. 8.20 Relationship between correction factor for AK(t)(0-tc) and D,

187

400


8.7 Conclusions

Based on the above results, the following conclusions can be drawn:

1. The CC mixtures developed lower Ko values, shorter pressure cancellation time (tc), and higher

decay in lateral pressure [AK(t)(0-tc)] compared to SCC of similar mixture composition, except

for higher concentration of HRWRA. The values of Ko@H=7m, [AK(t)(0-tc)], and tc for SCC38-

22 cast at 5 m/hr were 58%, 0.260, and 385 min, respectively. These values were 34%, 0.376

Pa/min, and 185 min, respectively, for the CC35-22 mixture.

A. Factors affecting Ko

2. Increasing concrete temperature (T) results in lower lateral pressure. At T of 12 °C, SCC38

exhibited approximately hydrostatic pressure. However, rising T to 22 °C and then to 30 °C

displayed significant deviation from hydrostatic. At casting rate (R) of 5 m/hr, the Ko values at

depth of 7 m were 71%, 58%, and 42% at T of 12, 22, and 30 °C, respectively.

3. For shallow depths, Ko was close to 100%. However, beyond a depth of 3 m, the pressure

envelop diverged from hydrostatic pressure. The Ko values decreased linearly with the increase

in concrete height. For the SCC38-22 mixture cast at R of 5 m/hr, Ko values at depths of 1, 3,

and 7 m were 81%, 73%, and 58%, respectively.

4. Ko values increases with the increase in R. For a very high R value of 30 m/hr, Ko approaches

100% especially at shallow depths. A significant reduction in Ko, even at shallow castings, is

obtained at slow rate (R = 2 m/hr).

5. The formwork lateral pressure decreases with increase in thixotropy. The lateral pressure at

different concrete heights (KO@HO can be correlated to various thixotropic indices determined

using concrete rheometer or field-oriented tests. Abacuses were established to facilitate the

estimate of Ko vs. thixotropy for various R values.

6. Interruption of concrete casting for a waiting period (WP) of 30 min can reduce Ko by up to 10

%, especially for highly thixotropic SCC.

7. Ko increases with the increase in Dmjn. A correction factor (flamm) for Ko is derived to account

for changes due to variations of Dmin between 200 to 350, as follows: flomm = 0.000968 Dmin +

0.806332, Dmin : mm.

B. Factors affecting AK(t)

8. The pressure decay increases with the increase in concrete temperature. SCC38 cast at 5 m/hr

had AKCtXO-tc) values of 0.20, 0.21, and 0.35 %/min at T of 12, 22, and 30 °C, respectively.

188


9. The AK(t)(0-tc) decreases with the increase in Dmin. For SCC38, the AK(t)(0-tc) values

measured with the pressure sensor mounted at the short lateral dimension at depth of 1.45 m,

were 0.33, 0.30, 0.27, and 0.25 %/min for Dmin of 200, 250, 300, and 350 mm, respectively.

10. Correction factor (f2Dmm) for AK(t)(0-tc) values accounting for the changes in lateral formwork

dimensions can be calculated as:f2umm = 1.260353 - 0.0001302 Dmjn, Dmin in mm.

11. The R and WP do not have any influence on lateral pressure decay.

C. Factors affecting tc

12. The WP and WP do not have any influence on lateral pressure decay.

13. Increasing concrete temperature reduces tc. SCC38 cast at R of 5 m/hr and at T of 12, 22, and

30 ± 2 °C had tc values of 470, 385, and 230 min, respectively.

14. tc increases with the increase in Dmin. For SCC38, increasing Drain from 200 to 350 mm led to

an increase in tc of about 110 min. At a given concrete depth, the tc measured from the two

lateral dimensions were found to be same, regardless of the ratio between the cross-sectional

dimensions.

189

CHAPTER 9

PREDICTION MODELS FOR LATERAL PRESSURE CHARACTERISTICS

9.1 Introduction

Determining the maximum initial lateral pressure (Pmax) developed by the fresh concrete

right after casting is the most important parameter for safe and cost-effective design of formwork

systems. The overestimation of Pmax can lead to higher construction cost, while underestimation

of Pmax could result in deflection of the formwork and, in extreme cases, to formwork opening or

collapse. The second important item for designing formwork is the pressure decay with time

[AK(t)]. The investigations carried out on normal-consistency concrete showed that lateral

pressure decreases slowly and then drops to zero after approximately three hours from end of

casting. This duration corresponds to the initial setting time [Rodin, 1952]. On the other hand,

this can no longer be true in the case of SCC proportioned with chemical admixtures, which can

delay setting to 7 hr or, in some cases, 15 hr after water-adding time (WAT). This makes the rate

of pressure drop an important item in designing the formwork to enable better optimization of the

scheduling of the re-use of the formwork.

This chapter is divided into four parts. The first aims at investigating the relationship between

the various thixotropic indices (T.I.)- The second and third parts of this chapter aim at proposing

equations to enable the prediction of Pmax and AK(t) values, respectively, through the evaluation of

different casting characteristics, formwork dimensions, and thixotropy. In the fourth part,

comparison between the proposed models and various published models was carried out.

9.2 Testing program

A comprehensive testing program was undertaken to propose equations for the prediction

of Pmax and AK(t) values using the key parameters affecting SCC formwork pressure. The

investigated parameters included casting depth (H), placement rate (R), concrete temperature (T),

minimum lateral dimension of formwork (Dm;n), mix design parameters (§, Vca, Vp, and S/A),

maximum-size of aggregate, and waiting period (WP) between successive lifts, as shown in Table

9.1. The thixotropy index was used to express the different mix design parameters in the

prediction equations. In the study, the thixotropy indices were determined at two different

temperatures. The first was carried out at the laboratory temperature of 22 ± 2 °C (T.I.@T=22±2°c)-

190

Chapter 9 Prediction models for lateral pressure characteristics

The second was tested at the various concrete temperatures (T.I.@TD- Based on the method of

determining thixotropy index, two sets of prediction models for Pmax values were developed. In

the prediction models that include the T.I.@T=22±2°c indices, the effect of concrete temperature is

introduced through a separate factor (T) to account for the changes in concrete temperature in the

real casting conditions (in-situ conditions). On the other hand, in the Pmax prediction models

based on the T.I.@xi, the temperature effect is already considered with thixotropy index, and thus

the factor T does not appeared in the equations. The lateral pressure variations were evaluated for

around 63 SCC mixtures (Table 3.8). The UofS pressure columns, the 1.2-m high PVC column,

and the plywood formwork of different Dmin values were employed to evaluate the lateral

pressure characteristics. The various thixotropy indices were determined using the modified

Tattersall MK-III concrete rheometer and the PV and IP field-oriented test methods.

Table 9.1 Modeled parameters in the prediction models of SCC formwork pressure

Prediction models Parameter Range

H l - 1 3 m

R 2, 5, 10, 17, 24, and 30 m/hr

T* 12,22,and30±2°C

Dmin 200, 250, 300, and 350 mm

Pmax T.I.@T=22±2°C or T.I.@Ti

MSA 10, 14, and 20 mm

• Continuous WP • W P o f 3 0 m i n at middle of casting

• Two WPs of 30 min each at middle of casting

AK(t) - ^ Dmin 200, 250, 300, and 350 mm

* Concrete temperature appears only when considering T.I.@T=22±2oc

9.3 Correlations between thixotropic indices

The relationships among the various thixotropy indices determined from the modified

Tattersall MK-III concrete rheometer (Abi, Rheometertorest, and Ar|app@N=o.7rps) and field-oriented

test methods (PVtorest and IPxorest) along with their R2 values are given in Table 9.2. The initial

191

„ , 1. Initial slump flow (<b) To replace - — VT/

mix design 2. Volume of coarse aggregate (Vca) parameters 3. p a s t e volume (Vp) including: ~ - : ; : ~ZTT7~

4. Sand-to-total aggregate ratio (S/A)


thixotropic indices determined at 15 min of resting [T.I.,] and the time-dependant change of

thixotropic indices [T.I.(t)] determined from different test methods were employed in the various

correlations, except for the Ab)-index that was determined only initially. Excellent correlations

were found between the various thixotropy indices with R2 values varying between 0.90 and

1.00. The different thixotropy indices were used in the prediction models of the formwork

pressure characteristics.

Table 9.2 Relationships between different thixotropic indices

Relationship R Equation

Abi = 0.567 x Rheometer xorest@i5min 0.97 Eq. 9.1

Abi = 22.7 x Rheometer x0 rest(t) 0.90 Eq. 9.2

Abi = 2 .7 x Ar|app@N=0.7rps@15min 0.98 Eq. 9.3

Ab, = 32.31 x AnapP@N=o.7 rps(t) 0.95 Eq. 9.4

Ab, = 0.584 x PV To rest@15min 0.98 Eq. 9.5

Ab, = 9 x PV T0 rest(t) 0.92 Eq. 9.6

Ab, = 0.633 x IP x0 rest@15min 0.99 Eq. 9.7

Ab, = 39.4xIPx0 r e s t( t) 0.94 Eq. 9.8

Rheometer To rest@i5min = 39.82 x Rheometer Torest(t) 0.88 Eq. 9.9

Rheometer T0rest@i5min = 4.733 x Anapp@N=0.7rps@i5min 0.98 Eq. 9.10

Rheometer T0 rest@i5min = 56.57 x Anapp@N=o.7rps(t) 0.93 Eq. 9.11

Rheometer T0 rest@i5min = 1 -027 x PV T0 rest@i5min 0.99 Eq. 9.12

Rheometer T0rest@i5min = 15.81 x PV T0rest(t) 0.92 Eq. 9.13

Rheometer T0 rest@15min = 1 . 1 1 1 x IP to rest@15min 0.99 Eq. 9.14

Rheometer Torest@i5min = 69.02 x IP x0rest(t) 091 Eq. 9.15

Rheometer T0 rest(t) = 0.115 x Ar|app@N=o.7rPs@i5min 0.91 Eq. 9.16

Rheometer T0 rest(t) = 1.408 x Ar|app@N=o.7 rps(t) 0.98 Eq. 9.17

Rheometer T0 rest(t) = 0.025 x PV T0 rest@i5min 0.92 Eq. 9.18

Rheometer T0 rest(t) = 0.392 x PV T0 rest(t) 0.95 Eq. 9.19

Rheometer T0 rest(t) = 0.027 x IP To rest@i5min 0.91 Eq. 9.20

Rheometer TQ rest(t) = 1.715 x IP T0rest(t) 0.97 Eq. 9.21

Anapp@N=0.7rps@15min = 1 1 . 9 2 x Ar|app@N=0.7 rps(t) 0.92 Eq. 9.22

Ar|app@N=0.7 rps@15min = 0 . 2 1 6 5 x P V T0 rest@15min 0 . 9 9 E q . 9 .23

Ariapp@N=0.7 rps@15min = 3 . 3 2 7 x P V T0 rest(t) 0 .91 E q . 9 .24

Ar|app@N=0.7 rps@15min = 0 . 2 3 4 x IP x0 rest@15min 0.99 Eq. 9.25

Ar|app@N=o.7rps@i5min = 14.532 x IP T0rest(t) 0.92 Eq. 9.26

192


Atlapp@N=0.7 rps(t) - 0.0178 x PV To rest@15min

ATlapp@N=0.7rps(t) = 0.2779 x PV T0 rest(t)

Ariapp@N=0.7 rps(t) = 0.019255 x IP Torest@15mii

Ariapp@N=o.7 rps(t) = 1.216 x IP Torest(t)

0.95

0.96

0.94

0.98

Eq.

Eq.

Eq.

Eq.

9.27

9.28

9.29

9.30

PVxo rest@15min 15.38 x P V To rest(t)

PV T0rest@15min = 1.08 X IP T()rest@15min

PV T0rest@15min = 67.15 X IP T0 rest(t)

0.92 Eq. 9.31

1.00 Eq. 9.32

0.92 Eq. 9.33

PVx0rest(t) = 0 . 0 6 9 x I P fOrest@15min

PVTorest(t) = 4 .332xIPT 0 r es , ( t )

0.93

0.95

Eq.

Eq.

9.34

9.35

IP TO rest@15min 61.992362 x IP x0 rest(t)

0.92 Eq. 9.36

Note J/m3.s Abi

RheometerT0rest@i5min : Pa Rheometer to rest(t) : Pa/min Ariapp@N=0.7rps@15min : Pa.S

Ar|aPp@N=0.7 rps(0

P V Torest@15min

PVTOrest(t)

IP tOrest@15min

IPXOrest(t) :

: Pa.s/min :Pa : Pa/min :Pa Pa/min

9.4 Prediction models for maximum lateral pressure

Several attempts were done to derive field-oriented models for the Pmax exerted by SCC on

the formwork. These attempts started by deriving models of one parameter at a time. Different

combinations between two parameters at a time were then investigated. In this stage, the

derivation of the models was achieved by the linear regression analysis. The derivation was

extended to include three, four, and then five parameters using special statistical software (NCSS,

2007). The derived models for Pmax are discussed below in terms of T.I.@T=22±2°C or T.I.@xi as well

as H, R, T, and Dmjn.

9.4.1 Models of Pmax as function of H, R, T, Dmjn, and thixotropy index at T=22±2°C

The results obtained from approximately 800 data points were used to establish models to

predict Pmax The derived models were function of H, R, T, Dmjn, and T.L@T=22±2°C- In total, 13

prediction models for 13 different thixotropy indices were established, as shown in Table 9.3.

These models are suitable for predicting Pmax when SCC is evaluated at ambient temperature of

22 ± 2 °C. It is important to note that the prediction models (Eqs. 9.37 to 9.49) are valid for the

ranges of variables shown in Table 9.3.

193

oZ

+-»

Pi

^^

P< X

Table 9.3 Prediction models for Pmax as function of H, R, T, Dmjn,

"max

"max

" max

"max

"max

"max

"max

"max

"max

"max

"max

"max

^ max

= pgH [111 - 3.8 H + 0.6 R - 0.6 T + 0.011 Dmin - 0.0345 Abi]

= pgH [114 - 3.8 H + 0.6 R - 0.6 T - 0.001 Dmin - 0.0203 Rheometerx0rest@i5 minj

= pgH [113 - 3.8 H + 0.6 R - 0.6 T + 0.0027 Dmin - 0.095 AT|app@N=0.7rps@15min]

and T.I.@T=

= pgH [112.5 - 3.8 H + 0.6 R - 0.6 T + 0.01 Dmin - 0.021 PVT0rest@i5min]

= pgH [112 - 3.83 H + 0.6R - 0.6 T + 0.01 Dmin - 0.023 IPT0rest@i5min]

= pgH [109.6 - 3.9 H + 0.6 R - 0.614 T + 0.005 Dmin - 0.67 Rheometerxorest(t)

= pgH [111 - 3.9 H + 0.6 R - 0.612 T + 0.001 Dmin - 0.99 AT|app@N=0.7rps(t)]

= pgH [109.5 - 3.9 H + 0.7 R - 0.6 T + 0.003 Dmin - 0.29 PVT0rest(t)]

= pgH [104.2 - 3.9 H + 0.6 R - 0.6 T + 0.0036 Dmin -1.22 IPiorest(t)]

= pgH [106 - 4 H + 0.6 R - 0.63 T + 0.0107 Dmin - 0.00036 RheometerT0rest@15 minxT0rest(t)]

= pgH [106 - 4H + 0.6R - 0.63 T + 0.011 Dmin - 0.0024 Anapp@N=0.7rps@15minxAr|app@N=0.7rps(t)]

= pgH [106 - 4 H + 0.6 R - 0.63 T + 0.01 Dmin - 0.00015 PVTo r es t@ 15min xXorest(t)]

= pgH [104.7 - 4 H + 0.6 R - 0.63 T + 0.019 Dmin - 0.0007 IPtOrestfffil 5min

=22±2°C

Eq. 9.37

Eq. 9.38

Eq. 9.39

Eq. 9.40

Eq. 9.41

Eq. 9.42

Eq. 9.43

Eq. 9.44

Eq. 9.45

Eq. 9.46

Eq. 9.47

Eq. 9.48

Eq. 9.49

where the variables and their ranges are: H = 1 -13 m Anapp@N=o.7rps@i5min = 0-450 Pa.s R = 2 - 30 m/hr Ar)app@N=o.7rps (0 = 0- 40 Pa.s/min T = 12 tO 3 0 ± 2 °C PVx0rest@15min = 0 - 2 0 0 0 P a

Dmin = 200 - 350 mm PVx0rest(t) = 0-125 Pa/min A b i = 0 - 1 2 0 0 J /m 3 .S IPx0rest@15min = 0 - 1 2 0 0 P a

Rheometerxorest@i5min = 0 - 2000 Pa/min IPtorest(t) = 0-30 Pa/min Rheometerxorest(t) = 0-50 Pa/min fu%k and /wp are dimensionless

9.4.2 Models of Pmax as function of H, R, Dmjn, and thixotropy index at various temperature

The second set of prediction models for Pmax is function of H, R, Dmjn, and T.I.@ii, as

summarized in Table 9.4. These models are suitable for predicting Pmax when the determination of

the thixotropy index is determined at the same temperature of the cast concrete.

Predicted Ko values determined using the models given in Eqs. 9.40 and 9.53 selected to

demonstrate the degree of prediction of Pmax with temperature considered as a variable or in the

194

thixotropy index are compared in Fig. 9.1. A 1:1 relationship with R2 value of 1.0 is obtained,

indicating an excellent agreement between the corresponding prediction models.

Table 9.4 Prediction models for Pmax as function of H, R, Dmjn, and T.I.@xi

(T is considered within T.I.@TJ)

c£

•<-»

ei

^^

P£ X

0?

" max

"max

"max

"max

"max

"max

* max

"max

"max

"max

"max

"max

"max

= pgH [97.4 - 3.81 H + 0.59 R + 0.0122 Dmin - 0.034 Abi

= pgH [100.4 - 3.83 H + 0.6 R + 0.00008 Dmin - 0.02 Rheometerx0rest@i 5min]

= pgH [99.3 - 3.8 H + 0.6 R + 0.004 Dmin - 0.094 Ar|apP@N=o.7rps@i5min]

= pgH [98 - 3.82 H + 0.63 R + 0.011 Dmin - 0.021 PVx0rest@i5min]

= pgH [98.4 - 3.8 H + 0.6 R + 0.011 Dmin - 0.0227 IPiorest@i5min]

= pgH [95.8 - 3.98 H + 0.6 R + 0.0065 Dmin - 0.67 RheometerT0rest(t)]

= pgH [97.4 - 3.9 H + 0.59 R + 0.0015 Dmin - 0.986 Anapp@N=o.7rpS

= pgH [95.9 - 3.84 H + 0.71 R + 0.0041 Dmin - 0.29 PVx0rest(t)]

- pgH [90.4 - 3.9 H + 0.59 R + 0.037 Dmin - 1.22 IPx0rest(t)]

= pgH [92.1 - 4 H + 0.6 R + 0.0112 Dmin - 0.00036 Rheometerx0rest@l 5min xX0rest(t)]

= pgH [92.2 - 3.97 H + 0.59 R + 0.011 Dmi„ - 0.0024 Ariapp

@N=0.7rps@15minxAr|app@N=o.7rps(t)] - pgH [92.1 - 4 H + 0.62 R + 0.011 Dmin - 0.00016

PVxorest@l 5min xt;0rest(t)]

- pgH [90.7 - 3.97 H + 0.59 R + 0.0191 Dmin - 0.00072 IPt0rest@15min

xIPTOrest(t)]

(t)]

Eq. 9.50

Eq.9.51

Eq. 9.52

Eq. 9.53

Eq. 9.54

Eq. 9.55

Eq. 9.56

Eq. 9.57

Eq. 9.58

Eq. 9.59

Eq. 9.60

Eq. 9.61

Eq. 9.62

100

£ 80

a* = P

60

40

•a

3 20 -a

0

y=1.00x R2=1.00

N = 795 points

0 20 40 60 80 100

Predicted KQ from Eq. 9.40 (%)

Fig. 9.1 Relationship between the predicted Pmax values from Eqs. 9.40 and 9.53

195

9.4.3 Introducing effect of MSA in the prediction models of Pmax

The MSA can be incorporated in the prediction models through a correction factor (/MSA)

accounting for different MSA other than 14 mm. The details for determining the/MSA are included

in Chapter 7 (section 7.5). The values of the/MSA can be estimated as follows:

> For relatively low thixotropy SCC \PVxorest@i5 min < 700 Pa]

H < 4 m /MSA = 1

H = 4 - 1 2 m fMsA = \ whenMSA = 20mm

, 1.26 H - 5.04 JMSA = 1 + • • •• when MSA = 10 mm

> For high thixotropy SCC [PVr0rest@i5mm > 700 Pa]

H = l - 1 2 m /MSA=1 when MSA =10 and 20 mm

The Pmax prediction models considering the effect of MSA will be as shown in Eq. 9.63.

Pmax = {Eqs. 9.37 to 9.62} xfMSA Eq. 9.63

9.4.4 Introducing effect of WP in the prediction models of Pmax

The effect of WP between successive lifts on the lateral pressure characteristics was

discussed in Chapter 7 (section 7.5). Based on these results, a correction factor accounting for the

effect of WP between successive lifts ifwp) was obtained. The values of the fwp can be determined

as shown in Fig. 8.12. For the investigated cases, the values offwp were ranged between 0.85 and

1. These values depend on the number and duration of the WP, as well as on the thixotropy level

of the tested concrete. The prediction models that incorporate the effect of WP are considered as

in Eq. 9.68.

Pmax = {Eqs. 9.37 to 9.62} * fMSA x fWP Eq. 9.64

9.4.5 Relationships between measured and predicted Pmax

The predicted Pmax values determined from the models of (H, R, T, Dmjn, T.I.@T=22±2°c,./MS/f,

and fWp) were correlated to the measured Pmax values in 1:1 relationships, as shown in Table 9.5.

Similar correlations for the Pmax values determined from the models of (H, R, Dmjn, ^^-@TI,/MSA,

and fwp) were obtained, as shown in Table 9.6. The relationships were excellent with R2 values

greater than 0.92. Based on the R values and the method of determining the thixotropy index, the

models shown in Eqs. 9.40 and 9.53 can be recommended for the Pmax prediction.

196


Table 9.5 Correlations between measured and predicted Pmax values determined from the models

of (H, R, T, Dmin, T.I.@T=22±2°C>./M£45 and/«//>)

ps

- * - »

Pi

- 4 - *

pt X

PZ

Analytical models

Abi

Rheometer T0rest@i5min

AT|app@N=0.7 rps@15min

PVxo rest@15min

IPxo rest@15min

Rheometerxo rest(t)

Ar|app@N=0.7 rps(t)

PVx0rest(t)

IPlOrest(t)

Rheome te rTo rest@15minxT0 rest(t)

Anapp@N=0.7rps@15min xAr|app@N=0.7rps(t)

PVxo rest@15minxTo rest(t)

IP to rest@15minx1;0 rest(t)

K0 = / H, R, Dmin, and T.I.@T= 22 ± 2 °c)

Measured = 1.00 x predicted













R2

0.93

0.93

0.93

0.94*

0.93

0.92

0.93

0.93

0.93

0.92

0.92

0.93

0.92

Table 9.6 Correlations between the measured and predicted Pmax values determined from the

models of (H, R, Dmin, T.l.@Ti,/MSA, andfWP)

PZ

-+->

Pi

PZ X

Pi

Analytical models

Abi

Rheometer x0 rest@i5min

An a p p @ N=0.7 rps@15min

P V Xo rest@15min

IP T0rest@15min

Rheometer xo rest(t)

AUapp @ N=0.7 rps(0

PVx0rest(t)

IPXOrest(t)

R h e o m e t e r X0 rest@15minxTo rest(t)

At|app@N=0.7rps@15min xAnapp@N=0.7rps(t)

P V Xo rest@15minx'C0 rest(t)

IP To rest@15minx1;0 rest(t)

K0 = / H, R, Dmin, and T.I.@Ti)














R2

0.93

0.93

0.93

0.94*

0.93

0.92

0.93

0.93

0.93

0.92

0.92

0.93

0.92 * Recommended

197


9.4.6 Abacuses for prediction of Ko values

The prediction models for lateral pressure exerted by SCC can be used to propose simple

abacuses for formwork pressure prediction. Abacuses to predict relative initial maximum lateral

pressure (Ko) values calculated from Pmax prediction models (Ko = Pmax/pgH) are shown in Figs.

9.2 to 9.5. These abacuses are determined for thixotropic indices corresponding to values

obtained after 15 min of rest. Other abacuses between Ko and thixotropy indices expressed in

terms of the rate of structural build-up at rest are also given in Appendix El. The abacuses were

constructed for given values of R, T, Dmjn, WP, and MSA. The values of the fixed parameters are

indicated in each figure. In the abacuses, the Ko values are varying with the thixotropy index

determined from various test methods. The Ko values are indicated at different casting depths (H)

from 1 to 12 m. As shown from the abacuses, the increase in the thixotropy index leads to a

decrease in Ko. Ko also decreases with the increase in casting depth.

198

Cha

pter

9

Pre

dict

ion

mod

els

for

late

ral

pres

sure

ch

arac

teri

stic

s

'mm

MSA

= 1

4 m

m

WP

= 0

K()

@H

=lm

K()

@H

=2m

K<)

@H

=4m

^0@

H=

6m

K<)

@H

=8m

Ko@

H=

10m

^0@

H=

12

m

400

800

1200

16

00

2000

Rh

eom

eter

x0

r£S

t@15

min

(P

a)

K(

K(

K(

K,

K, 0@

H=

lm

0@H

=2m

0@H

=4m

0(ffi

H=6

m

MSA

= 1

4 m

m

WP

= 0

0@H

=8m

Ko@

H=

10m

Ko(

5>H

=12m

400

800

1200

16

00

2000

Rh

eom

eter

T0

rest@

15m

in (

Pa)

^

100 80

60

^4

0 20 0

100 80

nun

MSA

= 1

4 m

m

WP

= 0

400

800

Rh

eom

eter

T, 0

rest

@15

min

(Pa)

C?

60

£ 40

20

R =

T

= 2

2 QC

D

min =

200

mm

M

SA =

14

mm

W

P =

0

K-0

@H

=lm

£0

@H

=2

m

K()

@H

=4m

K()

@H

=6m

K-0

@H

=8m

K, 0@

H=1

0m

0@H

=12m

1200

16

00

2000

K-0

@H

=lm

Ko@

H=

2m

K<)

@H

=4m

^o@

H=

6m

^0@

H=

8m

Ko@

H=

10m

^0(f

fiH

=12

m

400

800

1200

Rheometer T,

1600

0 re

st@

15m

in

("a)

2000

Fig

. 9.

2 C

orre

lati

ons

betw

een

KQ

and

Rhe

omet

erT

o res

t@i5

min

199

Cha

pter

9 P

redi

ctio

n m

odel

s for

lat

eral

pre

ssur

e ch

arac

teri

stic

s

mm

MS

A

WP

= 0

100

200

300

400

Ana

pp@

N=0

.7rp

s@15

min

(P

a-S

)

100

200

300

400

ATla

pp@

N=0

.7rp

s@15

min

(P

a.S

)

Ko@

H=

lm

K()

@H

=2m

K-0

@H

=4m

Ko@

H=6

m

K()

@H

=8m

Ko@

H=1

0m

^0@

H=

12m

500

Ko@

H=

lm

K-0

@H

=2m

K-0

@H

=4m

K-0

@H

=6m

K<)

@H

=8m

^0@

H=

K

, 10

m

0@H

=12m

500

100 80

^ 6

0

J 40

20 0

100 80

^6

0

J40 20

100

200

300

400

Atla

pp@

N=0

.7rp

s@15

min

(P

a.s

)

R

T =

22

°C

Dnu

n = 2

00m

m

MSA

= 1

4 m

m

WP

= 0

100

200

300

400

^Tla

pp@

N=0

.7rp

s@15

min

(P

»«

s)

|>0@

H=

lm

N)@

H=2

m

H=4

m

H=6

m

H=8

m

Ko@

H=1

0m

^•0@

H=1

2m

500

K(

K(

K,

K,

K, 0@

H=l

m

0@H

=2m

0@H

=4m

0@H

=6m

0@H

=8m

Ko@

H=1

0m

K()

@H

=12m

500

Fig

. 9.

3 C

orre

lati

ons

bet

wee

n K

0 an

d A

r| app

@N=o

.7 rp S

@i5

min

200

Cha

pter

9 P

redi

ctio

n m

odel

s for

lat

eral

pre

ssur

e ch

arac

teri

stic

s

^

100 80

^60

^4

0 20 0

100

WP

= 0

400

PV

T, 0

rest

@15

min

(Pa)

J0@

H=l

m

^0@

H=

2m

K<)

@H

=4m

^0

@H

=6m

K

<)@

H=8

m

K()

@H

=10m

^0@

H=

12m

800

12

00

16

00

20

00

K<)

@H

=lm

^0@

H=

2m

Ko@

H=4

m

Ko@

H=6

m

K<)

@H

=8m

Ko@

H=1

0m

^K)@

H=1

2m

400

80

0 1

20

0 1

60

0 2

00

0

PV

T 0

rest

@15

min

i„

(Pa)

100 80

5 60

^4

0 20 0

100 80

^6

0

^4

0 20 0

MS

A=

14

mm

W

P =

0 40

0 80

0

PV

T,

1200

1

60

0

0 re

st @

15m

ln("

a)

13

^ =

200

mm

M

SA =

14 m

m

WP

= 0

0 40

0

PV

T,

Fig

. 9

.4

Co

rrel

ati

on

s b

etw

een

Ko

and

PV

TQ

rest

@i5

min

0

rest

@15

min

in

(?

«)

|>0@

H=

lm

£0@

H=

2m

£0@

H=

4m

£0@

H=

6m

*H)@

H=8

m

Ko@

H=1

0m

Ko@

H=1

2m

2000

K()

@H

=lm

^0@

H=

2m

K()

@H

=4m

Ko@

H=6

m

Ko@

H=8

m

Ko@

H=1

0m

^0@

H=

12m

800

1200

1600

2000

201

Cha

pter

9 P

redi

ctio

n m

odel

s for

lat

eral

pre

ssur

e ch

arac

teri

stic

s

100 80

60

^4

0 20 0

(

100 80

C?6

0 0

s )£

40

MSA

W

P =

0 400

800

IPT

,

1200

16

00

0 re

st@

lSm

in (P

a)

20

WP

= 0 40

0 80

0 IP

T,

1200

16

00

0 re

st@

15m

in (P

a)

K<)

@H

=lm

Ko@

H=2

m

*S)@

H=4

m

K()

@H

=6m

^0@

H=8

m

^0@

H=1

0m

Ko@

H=1

2m

2000

K<)

@H

=lm

*M)@

H=2

m

Ko@

H=4

m

*S)@

H=6

m

Ko@

H=8

m

K()@

H=1

0m

^0@

H=1

2m

2000

400

IPT

, 0 re

st @

15m

in

i„(P

a)

100 80

60

i?

^4

0 20

E^n

= 2

00 m

m

MSA

= 1

4 m

m

WP

= 0 40

0

IPT

, 0 re

st@

15m

in (P

a)

Ko@

H=l

m

K{)

@H

=2m

K

<)@

H=4

m

Ko@

H=6

m

^0@

H=

8m

^N)@

H=1

0m

K()

@H

=12m

800

1200

16

00

2000

K()

@H

=lm

*M)@

H=2

m

Ko@

H=4

m

Ko@

H=6

m

Ko@

H=8

tn

K<)

@H

=10m

Ko@

H=1

2m

800

1200

16

00

2000

Fig.

9.5

C

orre

latio

ns b

etw

een

KQ

and

IPT

Q rest@

i5min

202

" Chapter 9 Prediction models for lateral pressure characteristics

9.4.7 Maximum lateral pressure of SCC on formwork system

The rate of structural build-up at rest (thixotropy), placement rate, and casting depth are

found to be the most effective parameters governing the lateral pressure exerted by SCC on

formwork systems. Therefore, the general proposed prediction model of Pmax (Eq. 9.91) was

employed to calculate Pmax at different thixotropy indices, placement rates (R), and casting depths

(H). Samples of Pmax values predicted at different R and H values are shown in Table 9.7. Each

table was established for a given thixotropy level determined using the PV test method

(PVxorest@i5min)- These tables were constructed for typical SCC mixture used frequently in cast-in-

place applications that proportioned with MSA of 14 mm and has unit weight (p) of 2350 kg/m3.

The formwork minimum dimension (Dmjn) was also considered as 200 mm. The concrete was

assumed to be continuously cast without waiting period (WP) at concrete temperature (T) of 22 ±

2 °C. Other tables can be established for different variations in the parameters affecting the

formwork lateral pressure.

Table 9.7 is divided into intervals according to Pmax values. Each interval is characterized

by a certain color. The cells shaded with a white color have low P^x values (< 60 kPa).

Formwork panels of low capacity can be used to support this lateral pressure for the considered

casting conditions. On the other hand, the cells shaded with black color have higher Pmax values

(> 170 kPa), and formwork system of high capacity are needed. The cells of different degree of

grey color are set for different Pmax intervals, as shown in Table 9.7. These tables can enable

formwork design given that a factor of safety should be introduced. Table 9.8 shows different

colors used to distinguish between the various Pmax intervals.

203


Table 9.7 Pmax (in kPa) values for formwork

PVTorest@15min = 200 Pa

H(m)

0

1

2

3

4

5

6

7

8

9

10

11

12

13

p = 2350

R(m/hr)

1

0

22

41

59

76

90

103 ^^SwMS^i

136

14(1

143

kg/m3 ,T

2 5 10 15 20

0 0 0 0 0

22 22 23 23 24

42 42 44 45 47

60 61 63 65 67

76 78 81 83 H6 ^ i » * r , , • • ' » . - i*Hftti> ,- . ' .-> , 'li.i :• . . .

•;.'.' r:%i ;V* u o , M lllJ

115 118 123 127 132

124 128 1 " 130 144

132 130 142 148 154

138 142 149 I5d 163

142 147 154 162 169

144 151 15S loft T 4

145 154 159 168 1 " = 22°C, D^n = 0.2 m, WP = 0, MSA = 14 mm

25

0

25

48

69

89

123

137

15D

161

170

177

1 S3

1X6

30

0

26

49

71

92

1 10

127

142

155

107

177

185

191

195

PVx0rest@15min = 400 Pa

H(m)

0

1

2

3

4

5

6

7

8

9

10

11

12

13

R(

1

0

21

39

56

85

1(17

p = 2350 kg/m3, T

2

0

21

40

57

8(» (>S

108

= 22°C, D

5 10

m/hr)

15

0 0 0

21 22 22

40 42 43

58 60 62

74 77 T9

88 91 o>

KM) 105 100

i * " •

nin = 0.2 m, WP = 0, MSA = 1

20

0

23

45

64

82

•>8

4 m m

25

0

24

46

66

8>

102

| 7

166

171

174

30

0

25

47

68

88

105

• i •

l"4

179

18^

204

Table 9.7 (Cont'd) Pmax (in kPa) values for formwork

PVT0rest@15min = 6 0 0 P a

H(m)

0

1

2

3 4

5

6 7

8

9 10

11

12 13

R(m/hr) 1 2 5 10 15 20 25 30

0 0 0 0 0 0 0 0 20 20 20 21 22 22 23 24 37 38 39 40 41 43 44 45 54 54 55 57 59 61 63 66 68 68 70 73 76 78 81 84 80 81 83 87 90 94 97 100 91 92 95 99 103 107 100 101 104 109 108 109

sr^F* -,£&'' -yA

rS-""'..^rilfe i.-^'^ "L:"-)te-"W " ''|!:;W:-:':-'- = WKSE^M

p = 2350 kg/m3, T = 22°C, D ^ = 0.2 m, WP = 0, MSA = 14 mm

PVTo«st@15miii = 1 2 0 0 P a

H(m)

0

1

2

3

4

5

6

7

8

9

10

11

12

13

R(m/hr)

1 2 5 10 | 15 20 25 30

0 0 0 0 0 0 0 0

17 17 17 18 19 19 20 21

32 32 33 34 35 37 38 40

45 45 46 49 51 53 55 57

56 57 58 61 64 67 70 72

66 <>7 60 72 76 70 82 X6

74 75 77 81 85 90 04 ox

80 81 84 80 04 08 I(H 108

84 86 8*) 04 1(H) 1(1 111 117

8^ 88 92 08 105 11] 117 123

88 80 l>4 luu l«r 114 121 128

87 8^ 03 fill 1<>8 116 124 131

85 86 0"* 1(H) U)S 116 124 133

80 82 01 Od 105 114 12~ 132

p = 2350 kg/m3, T = 22°C, D ^ = 0.2 m, WP = 0, MSA - 14 mm

205

Table 9.8 Color indication for various Pmax intervals

Color

.

R|E1!tl»K4*l

Intervals of Pmax values (kPa)

<50

50-80

80-110

110-140

140-170

>170

9.5 Prediction models for lateral pressure decay

In this section, prediction models for AK(t) as function of thixotropy indices are proposed.

A correction factor for AK(t) values accounting for the effect of various D^n of formwork are

then introduced in the prediction models. Relationships between measured and predicted AK(t)

are established.

9.5.1 Models of AK(t) as function of thixotropy index

Lateral pressure decay during the first 60 min after the end of casting [AK(0-60 min)] and

over the total period of pressure cancellation [AK(0-tc)] were correlated to the Ariapp@N=o.7rps and

PVxorest determined initially at 15 min of rest as well as the rate of change with rest time, as

shown in Figs. 7.11 to 7.16. Similar correlations with the Rheometercorest and IPxorest determined

at 15 min of rest and rate of change with rest time are given in Appendix C3. The linear

regressions for the correlations indicated in these figures were used as prediction models for

AK(t). Table 9.9 summarizes 12 prediction models for [AK(0-60 min)]. Another set of 12 models

for predicting [AK(0-tc)] is given in Table 9.10.

9.5.2 Introducing effect of D,™ in the AK(t) prediction models

The models for determining the AK(t) were obtained using the 1.2-m high PVC column that

has Dmin of 200 mm. To include effect of different D,™ other than 200 mm, a correction factor

(/ Dmin) was introduced, as discussed in Chapter 8 (section 8.6.4). The/2Dmin a can be determined

from Eq. 8.2 [A^min = 1.260353 - 0.001302 D ^ ] . The prediction models for the AK(t) will be as

follows:

AK(t)(0-60 min) = {Eqs. 9.65 to 9.76} x/2Dmin Eq. 9.65

AK(t)(0-tc) - {Eqs. 9.77 to 9.88} xf2Dmin Eq. 9.66

206


£

- 4 - ^

e4

X

Table 9.9 Prediction models for [AK(t)(0-60 min)] as function of thixotropy index

Analytical models

AK(t)(0-60 min) = 0.111 + 0.0001 x RheometerT0rest@i5min

AK(t)(0-60 min) - 0.1132 + 0.0005 x Ariapp@N=0.7ips@i5min

AK(t)(0-60 min) = 0.1092 + 0.000112 x PVx0rest@i5min

AK(t)(0-60 min) = 0.109 + 0.0001 x IPx0rest@i5min

AK(t)(0-60 min) = 0.1228 + 0.0045 x Rheometen0 rest(t)

AK(t)(0-60 min) = 0.1197 + 0.006 x Ar|app@N=o.7rps(t)

AK(t)(0-60 min) = 0.1264 + 0.0016 x PVx0rest(t)

AK(t)(0-60 min) = 0.1306 + 0.0066 x IPx0rest(t)

AK(t)(0-60 min) = 0.148 + 0.00000292 x Rheometerx0 rest@i

AK(t)(0-60 min) = 0.146 + 0.00001782 x Ariapp@N=o.7ips@15min

AK(t)(0-60 min) = 0.1491 + 0.000001021 x PVx0rest@i5minxxc

5minxX0rest(t)

X AT|app@N=o .7rps(t)

rest(t)

AK(t)(0-60 min) = 0.1486 + 0.00000472 x IPx0rest@i5minXxorest(t)

Eq.

Eq. 9.67

Eq. 9.68

Eq. 9.69

Eq. 9.70

Eq. 9.71

Eq. 9.72

Eq. 9.73

Eq. 9.74

Eq. 9.75

Eq. 9.76

Eq. 9.77

Eq. 9.78

Table 9.10 Prediction models for [AK(t)(0-tc)] as function of thixotropy index

dS

•+-»

Pi

X

6Z

AK(t)(0-tc;

AK(t)(0-tc;

AK(t)(0-tc;

AK(t)(0-tcN

AK(t)(0-tc

AK(t)(0-tc;

AK(t)(0-tc;

AK(t)(0-tc;

AK(t)(0-tc;

AK(t)(0-tc;

AK(t)(0-tc:

AK(t)(0-tc]

Analytical models

) = 0.1246 + 7E-05 x Rheometerx0rest@i5min

) = 0.1297 + 0.0003 x Ar|app@N=o.7lps@i5min

) = 0 . 1 2 5 9 + 0 . 0 0 0 1 X PVx0rest@15min

) = 0 . 1 2 6 + 0 . 0 0 0 1 X IPx0rest@15min

) = 0.1324 + 0.0029 x Pvheometerx0rest(t)

) = 0.1326 + 0.0036 x Ar|app@N=o.7lpS(t)

) = 0.1368 + 0.001 X PVxorest(t)

) = 0.1396 + 0.0039 x IPx0rest(t)

) = 0.14726 + 0.000002 x Rheometerx0rest@i5minxxorest(t)

) = 0.14767 + 0.0000112 x Ar|app @N=0.7rps@15min xAr|app@N=o.7ips(t)

I = 0 . 1 4 9 1 + 0 . 0 0 0 0 0 0 6 5 7 X PVx0rest@15minxT0rest(t)

I = 0 . 1 4 8 9 + 0 . 0 0 0 0 0 3 X IPx0rest@15minXTorest(t)

Eq.

Eq. 9.79

Eq. 9.80

Eq. 9.81

Eq. 9.82

Eq. 9.83

Eq. 9.84

Eq. 9.85

Eq. 9.86

Eq. 9.87

Eq. 9.88

Eq. 9.89

Eq. 9.90

9.5.3 Relationships between measured and predicted AK(t)

The correlations between measured and predicted [AK(0-60 min)] and [AK(0-tc)] and their

R2 values are listed in Tables 9.11 and 9.12, respectively. In general, the prediction models for

[AK(0-60 min)] based on the T.I.,- had higher R2 values than those based on T.I.# and T.I.,-xT.I.^.

On the other hand, the prediction models for [AK(0-tc)] derived from the T.L/xT.I.^ resulted in

207


higher R2 values than the models derived from the T.I.,- and T.l.(t). These models were resulted in

1:1 relationships between the measured and predicted responses

Table 9.11 Correlations between measured and predicted AK(t)(0-60 min) values

dS

•4—»

rt

P? X

Analytical models

Rheometerc0rest@i5min Anapp@N=0.7rps@l 5min

P VTorest@15min

IPtOresttglSmin

Rheometerxo rest(t)

A n a p p @ N = o .7rps(t)

PVTorest( t )

IPTOrest(t)

RheometerxorestgisminXTorestCt)

AT|app@N=0.7rps@l 5min x AT|app@N=0.7rps(t)

P Vlores t® 15min x T()rest(t)

IPx0rest@ 15min X ^Orest(t)

AK(t)(0-60

Measured =

Measured =

Measured =

Measured =

Measured =

Measured =

Measured =

Measured =

Measured =

Measured =

Measured =

Measured =

min)

= 1.02 x predicted

= 1.00 x predicted

= 1.04 x predicted

= 1.08 x predicted

= 1.00 x predicted

= 1.00 x predicted

= 1.00 x predicted

= 0.98 x predicted

= 1.00 x predicted

= 1.00 x predicted

= 1.00 x predicted

= 1.00 x predicted

R2

0.71

0.86++

0.84+

0.82

0.73

0.86

0.82

0.80

0.72

0.84

0.78

0.78

Table 9.12 Correlations between measured and predicted AK(t)(0-tc) values

tf

+^

&

Pi X

Analytical models

Rheometercorest® 15min

Ar|app@N=0.7rps@l 5min

PVTorest@15min

IPt0rest@15imn

Rheometerxo rest(t)

Ariapp@N=0.7rps(t)

PVXorest(t)

IPTOrest(t)

Rheometerxorest@i5minxxorest(t)

^rlapp@N=0.7rps@15minxAr|app@N=o.7rps(t)

PVXorest@l 5minxTorest(t)

IPtOrest® 15min x TOrest(t)

AK(t)(0-te)

Measured =

Measured =

Measured =

Measured =

Measured =

Measured =

Measured =

Measured =

Measured =

Measured =

Measured =

Measured =

= 0.99 x predicted

= 0.99 x predicted

= 0.89 x predicted

= 0.91 x predicted

= 1.00 x predicted

= 1.00 x predicted : 0.99 x predicted

= 0.99 x predicted

= 1.00 x predicted

= 1.00 x predicted

= 1.00 x predicted

= 1.00 x predicted

R2

0.77

0.77

0.73

0.68

0.79

0.79

0.78

0.77

0.89++

0.86

0.85+

0.83

Recommended when concrete rheometer is available + Recommended for field application

208


The model proposed in Eq. 9.68 resulted in a 1:1 relationship between measured and

predicted responses should be used for the prediction of [AK(0-60 min)]. When a concrete

rheometer is not available, the PV can be used, and the model suggested in Eq. 9.69 is

recommended for the prediction of [AK(0-60 min)].

The prediction [AK(0-tc)] model in Eq. 9.87 derived from the RheometerTorest@i5minxt;orest(t)

is recommended. This model is suitable for the prediction of [AK(O-tc)] in the laboratory where

the concrete rheometer is available. The model derived from the PVxorest@i5minxTorest(t) (Eq. 9.89)

is recommended for the filed applications using the PV device.

9.6 Comparison between predicted lateral pressure values determined using UofS model

and other published guidelines

Predicted Pmax values from the developed models (UofS model) (Eq. 9.92) is correlated to

the values determined using the ACI 347 model modified by Hurd [2002] (Eq. 2.10) in Fig. 9.6.

These correlations were established for SCC mixtures made with HRWRA and have a unit

weight (yc) of 2360 kg/m3. The mixtures have three thixotropic levels (Abi values of 200, 600,

and 1000 J/m3.sec). The concrete is to be cast in formwork with a minimum lateral dimension

(Dmin) of 200 mm at rate of casting (R) of 2 to 30 m/hr and with concrete temperature between 12

and 30 °C. The comparisons indicate that the ACI 347 model does overestimate Pmax compared to

the UofS model. The overestimation becomes higher with the increase in thixotropy. For Abi of

200 J/m3.sec, the predicted Pmax from the ACI 347 is 1.14 times that predicted from the UofS

model. This increases to 1.91 times for concrete with Abi of 1000 J/m .sec.

The predicted Pmax values determined from the model of the German Standard DIN 18218

[1980] (Eq. 2.12) were correlated to that determined from the UofS model (Eq. 92) in Fig. 9.7.

The correlation was established for SCC mixture with the characteristics shown in the figure. The

German Standard DIN 18218 [1980] model resulted in higher Pmax values (28% on the average).

The predicted Pmax values determined from the Roussel and Ovarlez's model [2005] (Eqs.

2.37 and 2.38 for the wall and column elements, respectively) were correlated to that determined

from the UofS model (Eq. 92) in Fig. 9.8. The correlation was established for SCC mixture with

the material ranges given in the figure. The model of Roussel and Ovarlez [2005] resulted in

higher Pmax values compared to the UofS model (64% on the average).

Khayat and Assaad's model [2005A] prescribed in Chapter 2 (section 2.2.2.-D), was used

to predict the relative initial maximum lateral pressure (Ko values) exerted by SCC mixture with

209

the properties shown in Fig. 9.9. It is to be noted that the casting depth (H) in this model is

limited to 2.8 m. The predicted Ko values from the Khayat and Assaad's model were correlated to

the corresponding values obtained from the UofS model (Eq. 9.92), as shown in Fig. 9.9. Only

5% increase in the Ko values was estimated when using the Khayat and Assaad's model

compared to the UofS model. This correlation was resulted in higher R2 value of 0.94.

«

© a

t -

u < S .2

Abt = 1000 J/m3.sec Ab, = 600 J/m3.sec • y=1.91x \ / y = 1.43x

* v * Ab, = 200 J/m3.sec

y=1.14x

250

200

150

100

50

0 50 100 150 200 250 Pmax from UofS model (kPa)

Fig. 9.6 Comparison of predicted Pmax from ACI 347 and UofS models for SCC mixtures of

different thixotropy levels

yc = 2360 kg/m3

R = 2-30m/hr T = 1 2 - 3 0 ° C D , ^ 200 mm

•§& a _

"*•• T 3

cc "§ s S

o . a i-i © <^

,& 00

E M PH Q

300

250

200

150

100

50

0

0

y = 1.28x

yc = 2360 kg/m3

R = 2 - 30 m/hr T=12-30°C H = 3 - 12m Dmin = 200mm Ab! = 200 -1000 J/m3.s

50 100 150 200 250 Pmax from UofS model (kPa)

Fig. 9.7 Comparison between P^* values predicted from German Standard DIN 18218 and UofS

models

210

-" 300

t ~ 2 5 0

"a «-200

<u o S °, 150 © —

* "§ 100

2 a

n 50 a E

* 0 c

(

- en

)

D O DC n o i i a o q| D to as n o eg a adsE U UUHJll a D p n :

1 II ffllllllM

aqpram'

50 100

JDDD D jam a gamD jnnn m n u n o nan ma'

not*

150

an an

ID C

a •

a /

'y = 1.64x

yc = 2260- 2420 kg/m3

R=5-30m/hr T= 12-30 °C H= 1 - 13 m 0 , ^ = 2 0 0 mm PVT0rest(t)=10-120Pa.s Air content =0.1-0.075

200 250

Pmax from UofS model (kPa) Fig. 9.8 Comparison between Pmax values determined from Roussel and Ovarlez's model [2005]

and UofS model

100 ^

^ "w1

C8 ®

S ° £ -a

o «*

90

80

70

60

y=1.05x R2 = 0.94

T = 22°C R=5-30m/hr H = l - 2 . 8 m 0 ^ = 200 mm Abj = 200 - 500 J/m3.s

60 70 80 90 100 Ko from UofS model (%)

Fig. 9.9 Comparison between Ko values determined from Khayat and Assaad's (2005A) model

and UofS model

It can be concluded from these results that the models of ACI 347, German Standard DIN

18218, and Roussel and Ovarlez [2005] overestimate lateral pressure. Khayat and Assaad's

model [2005A] yields close correlation to the general model proposed in Eq. 9.92 (UofS model).

The lateral pressure envelops determined from the models of ACI 347, German Standard

DIN 18218 [1980], Roussel and Ovarlez [2005], and Khayat and Assaad [2005A] are compared

to that obtained using the UofS model (Eq. 9.92) and to the hydrostatic pressure (Phyd), as shown in

211


Fig. 9.10 for concrete mixture having the characteristics shown in the same figure. The profile of

lateral pressure obtained from the ACI 347 model is equal to the Phyd up to approximately a 9-m

casting depth, and then continue constant with further depths. The German Standard DIN 18218

model resulted in lateral pressure profile similar to Phyd up to casting depth of 4 m, and then with

constant value for deeper casting depths. The lateral pressure predicted using the model of

Roussel and Ovarlez [2005] was close to the hydrostatic all the way with casting depth up to 13

m. On the other hand, the UofS model resulted in lateral pressure profile deviating from Phyd

starting from the top. The Pmax from the UofS model is noted at casting depth of 9 m, and then the

pressure decreases slightly at deeper depths. The Pmax from the UofS model is approximately half of

that noted from the ACI 347 model. The Pmax monitored using the UofS model was slightly higher

than that determined from the German Standard DIN 18218 model at deeper depths. The model

proposed by Khayat and Assaad's model [2005 A] resulted in similar lateral pressure variations as

those determined by the UofS model.


yc = 2360 kg/m3

R=10m/hr T = 22 °C D ^ 200 mm At>! = 500 J/m3.sec (0.93 P/sec)

• Phydrostatic

•UofS model

•ACI 347 model

•German Standard DIN 18218,1980

• Khayat and Assaad, 2005A

• Roussel and Ovarlez [2005]

Fig. 9.10 Comparison between lateral pressure variations with casting depth determined from

UofS model and published models

9.7 Conclusions

Based on the results discussed in this chapter, the following conclusions can be drawn:

212

1. Good correlations can be established (R2 values of 0.9 to 1.0) between the various

thixotropic indices determined from the modified Tattersall MK-III concrete rheometer and

the PV and IP field-oriented test methods.

2. Models to predict maximum lateral pressure (Pmax) and pressure decay [AK(t)] of SCC are

proposed in terms of casting depth (H), placement rate (R), concrete temperature (T),

minimum lateral dimension of formwork (Dmin), and thixotropy index. The latter can be

expressed at the actual concrete temperature (T.I.@xi) or at 22 ± 2 °C (T.I.@T=22±2°c) with a

correction factor for actual temperature. The effect of waiting period between successive lifts

(WP) and maximum-size of aggregate (MSA) were also determined in the prediction models.

3. The recommended prediction models for Pmax that resulted in the best correlation between

the predicted and measured responses and with the highest R2 values are:

(a) for thixotropy index determined at laboratory temperature (T.I.@x=22±2°c):

Pmax = pgH [112.5 - 3.8 H + 0.6 R - 0.6 T + 0.01 D ^ - 0.021 PVT0rest@i5 Eq. 9.91 min@T=22±2°c] x /MSA X fwP

(b) for thixotropy index determined at various concrete temperatures (T.I.@Ti):

Pmax = pgH [98 - 3.82 H + 0.63 R + 0.011 D ^ - 0.021 PVTorest@i5min@Ti] x Eq. 9.92 /MSA xfwp

where: H = 1 -13 m

R = 2-30m/hr

T = 1 2 t o 3 0 ± 2 ° C

Dmin = 200-350 mm

PVTorest@15min = 0 - 2 0 0 0 P a

PVxorest(t) = 0-125 Pa/min

fusA and^wp are dimensionless

/MSA '• correction factor for MSA different than 14 mm, and estimated as follows:

> For relatively low thixotropy SCC [PVx0rest@i5 min < 700 Pa]

H < 4 m /MSA = 1

H = 4 - 1 2 m /MSA = \ whenMSA = 20mm

, 1.26H-5.04 , ,„A , r t

fifiA = 1 + • • •• when MSA = 10 mm J MSA 1 Q 0

> For high thixotropy SCC [PV x0 rest @is min > 700 Pa]

H = l - 1 2 m /MSA=1 when MSA = 10 and 20 mm

213

fwp : correction factor for WP, and can be determined from the following figure.

1.1

1.0

0.9

^ 0.8

0.7

0.6

continuous casting - B E3--

2 w 7 ^ m m e a c h

200 400 600 800 1000 PVT„ l0rest@15min

(Pa)

4. Excellent agreement is found between the two recommended prediction models for Pmax (1:1

with R2= 1.0).

5. Abacuses to provide quick and simple prediction of Ko as function of thixotropy indices are

established for various H and R values.

6. The following prediction models are recommended for estimating lateral pressure decay:

(a) Rheometric measurements:

AK(t)(0-60 min) = [0.1132 + 0.0005 x Ar|app@N=o.7ips@i5min]

AK(t)(0-tc) = [0.14726 + 0.000002 x Rheometerc0rest@i5.™nX<corest(t)] x/2Dmin

(b) Field-oriented devices:

AK( t ) (0 -60 m i n ) = [0 .1092 + 0 . 0 0 0 1 1 2 X PVTorest@15min] X^Dmin

AK(t)(0-te) = [0 .1491 + 0 . 0 0 0 0 0 0 6 5 7 X PVT0rest@15minXX0rest(t)] X ^ D m i n

where: f2Dmin = 1.260353 - 0.001302 D .

7. The results of Pmax determined from the prediction model (Eq. 9.92) (UofS model) correlate

very well to the model proposed by Khayat and Assaad [2005A]. However, the UofS model

considers a wider range of casting conditions and can be applied using field-oriented tests to

estimate thixotropy.

8. Based on the data points for the Pmax values generated using the various prediction equations

published in literature and the UofS model, the models of ACI347, German Standard DIN

18218, and Roussel and Overlez [2005] are leading to an overestimate for the Pmax values

exerted by SCC on formwork compared to those obtained using the UofS model.

Eq.

Eq.

Eq.

Eq.

Ec

9.93

9.94

9.95

9.96

1.8.2

214

CHAPTER 10

FIELD MEASUREMENTS AND VALIDATION OF LATERAL PRESSURE MODELS

10.1 Objectives

The objective of this chapter is to validate the models elaborated in Chapter 9 using actual

field measurements. This was achieved by conducting large-scale measurements on eight wall

panels in 2008 during the construction of the "Integrated Research Laboratory on Materials

Valorization and Innovative and Durable Structures", at the Department of Civil Engineering, at

the Universite de Sherbrooke in Canada. The second validation involved the casting of full-scale

columns at the Material Laboratory of CTLGroup, II, USA. The models developed in the thesis

are then used to predict and validate the magnitude of the form pressure and compare such values

to actual field measurements.

10.2 Field testing of wall and column elements

Eight wall elements of two different heights were cast during the construction of the

"Integrated Research Laboratory on Materials Valorization and Innovative and Durable

Structures", in Sherbrooke in 2008. Walls # 1 to 4 cast in level 1 were 3.7 m in height, and Walls

# 5 to 8 cast in level # 2 were 4.4 m in height, as shown in Figs. 10.1 to 10.5. Summary of the

experimental program undertaken for casting the walls is given in Table 10.1. The walls were

0.20 m in thickness and 5.6 m in length. The CC61, SCC62, SCC63, and SCC64 mixtures (Table

3.8) were used for the casting of the walls. The SCC mixtures were proportioned with two paste

volumes (Vp) of 330 and 370 1/m3. The w/cm was set to 0.35 and 0.42 with the latter SCC

proportioned with VMA to ensure adequate stability. Two types of HRWRA [polcarboxylate

(PCP) and polynaphthalene sulphonate (PNS)] were also employed, as shown in Table 10.1. The

target slump flow (<|)) and slump value for the SCC and CC61 mixtures were 650 ± 25 mm and

120 ± 30 mm, respectively. Concrete temperature (T) at the time of casting varied between 22

and 30 °C. The concrete was cast by pumping from the top at a rate of rise (R) of 5 to 15 m/hr.

The variations of lateral pressure for each wall element were monitored using six pressure

sensors from Honeywell (Fig. 2.46). The sensors were set flush with the inner surface of concrete

using steel plates fixed to the formwork. The sensors were mounted vertically at 0.5-m intervals,

as shown in Fig. 10.1. Two thermo-couples were attached at the center of each wall at 1 m from

215

Chapter 10 Field measurements and validation of lateral pressure

the top to monitor concrete temperature. Data from the pressure sensors and thermo-couples were

collected using acquisition system at 90-sec intervals during 24 hr following casting. The walls of

level 1 cast during winter of low temperature were covered with thermal insulating blankets, and

the enclosure was also heated to prevent freezing.

Casting point 1.64 m

Casting point

5.59 m

E

Wall # 8 Casting point 1 Casting point 2

I_

1.48

column

Pressure sensors

0.9

. D.40,

1" 5.59 m

0.05

Fig. 10.1 Configurations of walls # 5 to 8 constructed on level 2

216


Wi i*-I

. t * Fig. 10.2 Wall panels # 6 to 8: the steel bars used for reinforcement (left) and during casting (right)

iraMU IfiUllll

Wk m Fig. 10.3 Pressure sensor set flush with concrete surface

• • 4k

'• ' 9 - • «

i'

*$•

.,[•*., ; jr.

O'lbS*

;> - - A , - J « * . ' • ; • • . - •£?- .. ita^™* •

; FAT*

y - : *

Fig. 10.4 Casting SCC in formwork

217

Fig. 10.5 Appearance of wall element cast with SCC after formwork removal

Table 10.1 Experimental program for eight walls cast at the Universite de Sherbrooke

Slump (mm)

Slump flow (mm)

Type of HRWRA

Paste volume (1/m3)

Casting rate (m/hr)

w/b (theoretical)

Level 1000 (Height = 3.7 m)

CC

61

wal

l 1

120 ±30

~

~

~

7.5

0.40

SCC

62

wal

l 2

SCC

62*

wal

l 3

SCC

62*

wal

l 4

—

650 ± 25

PCP

Low 330

5 10 15

0.35

Level 2000 (Height = 4.4 m) C

C61

w

all

5

120 ±30

—

—

~

7.5

0.40

SCC

62*

wal

l 6

SCC

63

wal

l 7

SCC

64

wal

l 8

—

650 ± 25

PCP

Low 330

High 370

PNS

Low 330

10

0.37 0.35 0.42+VMA

Air content < 3.5% and T = 22 - 30 °C Casting were pumped from the top

Part of the concrete was delivered at the Concrete Materials laboratory at the Universite de

Sherbrooke (UofS) for full characterization by a team of researchers (Fig. 10.6). The UofS2

pressure column was employed to estimate the maximum lateral pressure for simulating casting

depths similar to those employed in the field. The lateral pressure variations, until pressure

cancellation, were determined using the 1.2-m high PVC column. The fresh concrete properties

are indicated in Table 10.2. Rheological properties and thixotropic indices from various field-

oriented and rheometric test methods were also conducted and are presented in Table 10.3.

218


Fig. 10.6 Full characterization at Material laboratory (UofS) by a team of researchers

Table 10.2 Fresh concrete properties of mixtures used in field investigation at Sherbrooke

Mixture Wall# Concrete temp. (°C) Air content (%) Unit wei

Slump/ slump flow

J-Ring (mm)

ght (kg/m3) Initial (mm)

T50 (sec) @30 min (mm)

T50 (sec) @45 min (mm)

T50 (sec) initial @30 min @45 min

Max. settlement (%) Segregation index (%)

L-box (%) Time (sec)

Filling capacity (%)

CC61 1

13.3 1.8

2389 130

110

100

—

0.46 ~

—

~

SCC62 2 22 2

2388 620 3.14 635 3.39 630 3.03 560 560 540 0.5

2.08 0.38 5.75 84

SCC62* 3

25.3 1.5

2401 670 2.45 680 1.85 635 6.67 630 640 630 0.55 3.09 0.65 2.06 94

SCC62* 4

21.8 1.4

2381 630 3.64 610 3.68 625 3.92 570 585 580 0.52 1.89 0.64 3.3 73

CC61 5

13.6 1.5

2386 90

90

85

~

0.49 —

—

—

SCC62* 6

23.5 2.7

2406 630 2.72 590 5.27 570 3.77 545 530 500 0.55 4.32 0.6 2.7 72

SCC63 7 15 1.7

2385 655 2.86 660 2.25 530 5.25 645 590 420 0.49 1.75 0.73 2.58 91

SCC64 8

25.7 1.6

2391 665 1.45 635 2

620 3.17 640 600 540 0.39 1.22 0.84 2.76 73

* Used in Chapter 8 to cast the plywood formwork

219


Table 10.3 Rheological and thixotropic measurements of mixtures used in the field investigation

Mixture

Wall#

RheometerT0rest@i5min (Pa) RheometerTorest(t) (Pa/min) Ar|app@N=0.7 rps 15min (Pa . s )

Ar|app@N=o.7 rps(t) (Pa.s/min)

PVTorest@15min ( P a )

PVT0rest(t) (Pa/min) IPT0rest@15min (Pa )

IPxorest(t) (Pa/min)

CC61

1 ~

—

1813 21.5

—

SCC62 SCC62*

2 447 0.8 112 1.2 426 8.3 343 4.9

3

291 1.9 54 0.4 219 6.8 254 2.6

CSCC62*

4

309 5.2 69 1.8 323 5.9 355 3.9

CC61

5 2047 18.7 58 3.2

1903 11.4

—

SCC62*

6

450 0.7 72 2.9 411 6.2 403 2.9

SCC63

7 363 2.8 75 1.2 246 4.2 306 4.2

SCC64

8 619

8 131 5.1 503 15

307 4.2

Summary of testing program carried by CTLGroup is given in Table 10.4. In total, eight

circular steel columns of 3.66 m in height and 0.61 m in diameter were used to evaluate the effect of

casting rate and SCC thixotropy on formwork pressure characteristics. Each column was

instrumented with two pressure sensors from Honeywell measuring 20 mm in diameter and fixed at

depths of 2.74 and 3.35 m (Figs. 10.7 to 10.10). The columns were cast with three SCC mixtures of

low, medium, and high thixotropy levels (SCC-L, SCC-M, and SCC-H, respectively). The concrete

was placed at casting rates varying between 2 and 22 m/hr. Column # 7 was cast continuously using

SCC-M at rate of 5 m/hr. and with waiting period of 20 min at 5 m/hr. The same SCC was used to

cast column # 8 at 5 m/hr, except a waiting period of 20 min was introduced at mid of casting. The

mix designs and detailed results for casting these columns are reported in RMC-SDC 2009 final

report. Fresh properties and lateral pressure characteristics determined for the eight columns are

listed in Table 10.5. These results are used to validate our models presented in Chapter 9.

Table 10.4 Test matrix for CTLGroup columns

SCC-L

SCC-M

SCC-H

Relative thixotropy

Low

Medium

High

2

~

—

Col.#5

5

—

Col.#7

Col.#3

Casting rate (m/hr)

5+WP of 20 min

—

Col.#8

—

10

~

—

Col.#4

13

Col.#l

~

15

~

Col.#6

22

Col.#2

—

220


0.61m

Fig. 10.7 Configuration of CTLGroup column elements

Fig. 10.8 Column forms extending from lower level of the laboratory to the upper level (left),

fixation of pressure sensor set flush with the concrete surface in the column form (middle), and

reinforcement of the column shown from the plane view (left)

221


Fig. 10.9 Laser-based monitoring stand for controlling casting rate, pressure monitoring data

acquisition system, fresh concrete testing equipments

Fig. 10.10 Columns at the end of casting that remain in rest for 24 hr before demolding

222

Cha

pter

10

Fie

ld m

easu

rem

ents

and

val

idat

ion

of la

tera

l pre

ssur

e

Tab

le 1

0.5

Res

ults

of

eigh

t co

lum

n el

emen

ts c

ast

at C

TL

Gro

up l

abor

ator

y

Mix

ture

Cas

ting

date

Col

umn

No.

Am

bien

t tem

p.

Slum

p flo

w *

Air

vol

ume*

Uni

t w

eigh

t

Con

cret

e te

mp.

PV

x 0re

st@

l 5m

in

PV

Tor

est(

t)

AK

(t)(

0-60

min

) A

K(t

)(0-

t c)

MSA

W

aitin

g pe

riod

t-'m

in

Cas

ting

rate

Cas

ting

dept

h

"max

Ko

Uni

ts

°C

mm

%

kg/m

3

o

C

Pa

Pa/m

in

%/m

in

%/m

in

mm

m

in

mm

m/h

r

m

kPa

%

SCC

-L

Nov

embe

r 14

th 2

008

1 2

9.6

640

5.5

2330

22.2

345

3.5

0.22

0.

27

SCC

-H

Dec

embe

r 18

th 2

008

3 4

-5.4

600 3

2386

12.8

1082

23

.1

0.12

0.

22

SCC

-H

Janu

ary

19th 2

009

5 6

-4.7

600

3.6

2358

8.1

1222

65

.2

0.17

0.

13

SCC

-M

Mar

ch 1

2th 2

009

7 8

-5.8

620

4.7

2323

16.7

712

18.6

0.

28

0.23

10

0 20

610

12.9

4

3.36

68

89

2.75

57

91

21.8

7

3.34

75

98

2.73

61

97

4.98

3.35

74

95

2.74

63

97

9.85

3.

35

68

87

2.74

54

85

1.94

3.26

46

61

2.65

39

64

13.7

2

3.35

75

97

2.74

63

99

4.97

3.

34

65

86

2.73

55

88

4.99

3.35

59

77

2.74

46

73

*

Det

erm

ined

at c

oncr

ete

deliv

ery

time

from

rea

dy-m

ix p

lant

and

mai

ntai

ned

cons

tant

unt

il ca

stin

g th

e co

lum

ns

223


10.3 Test results of casting wall elements and discussion

10.3.1 Typical results

Lateral pressure variations with time determined from the pressure sensors mounted at

different concrete depths (H) for wall # 6 is shown in Fig. 10.11. The variations of concrete

temperature with time are also indicated in the figure. Wall # 6 was cast using SCC62 at 10 m/hr.

The lateral pressure variations monitored in the field were compared to the pressure decay

obtained from the 1.2-m high PVC column cast in laboratory at the same rate.

The maximum lateral pressure (Pmax) as well as the pressure cancellation time (tc) are

shown to increase with the increase in depth. Values for Pmax of 36, 51, and 58 kPa and tc values

of 790, 820, and 980 min were monitored at H of 1.8, 3.3, and 3.9 m, respectively, for the

concrete cast in wall # 6. Pressure cancellation took place at the same time as concrete

temperature started to increase considerably, corresponding to setting. The profile of the lateral

pressure decay determined using the 1.2-m high PVC column was similar to those determined in

the field from the pressure sensors mounted at different depths.

Wall # 6, SCC62, R = 10 m/hr 45

40 Q

o

3

35 1 a S a> H

30

25

200 400 600 Time (min)

800 1000 1200

Fig. 10.11 Variations of lateral pressure and concrete temperature with time, wall # 6 SCC62)

10.3.2 Effect of casting rate

Variations of Pmax with H for SCC62 cast at R of 5, 10, and 15 m/hr (walls # 2, 3, and 4,

respectively) are shown in Fig. 10.12 left. These variations were found to be similar to the

224


corresponding values obtained using the UofS2 pressure column, presented in Fig. 10.12 right.

The lateral pressure envelops in the two cases were close to hydrostatic values given the shallow

casting depth of 3.7 m and low thixotropy of SCC62. As expected, the increase in R led to

increase in Pmax. However, this increase was not very significant since the thixotropy of SCC62

mixture also varied for the concrete delivered at the job site by the ready-mix supplier. Indeed,

the PVxorest@i5min values for the SCC62 cast at 5, 10, and 15 m/hr received on different days were

425, 220, and 325 Pa, respectively.

(m)

-a DC

a

u

0

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5


1 i i i i

. SCC62 >x Field measurements

\ 3 *

l||k Hydrostatic \ sk \^" pressure

> ^ S \ Wall #4 ^ \ S S S X .R = 15 m/hr

Wall #2 Y ^ f R = 5 m/hr \ V \ \

Wall #3 ' Y \ \ R= 10 m/hr V \

». formwork base

a de

pth

(

Ml (3

U

0

0.0 t

0.5

1.0

1.5

2.0

2.5

3.0

3.5


I

^ U o f S 2

Wall #2 -R =

-

= 5 m/hr

Wall #3

I l 1 1

SCC62 pressure column

Hydrostatic

k z^-^ pressure

\ vs.* \ \ \ \ - - ^ \ \ Wall #4

V % \ R= 15 m/hr \ \ \ ^ / \ \ \ * /

R=10m/hi V \ \ "x* \ \ \ *

formwork base 4.0 4.0

Fig. 10.12 Effect of casting rate on lateral pressure envelops from field observations (left) and using the

UofS2 pressure column (right) (SCC62, walls # 2,3, and 4)

10.3.3 Effect of casting depth

The lateral pressure variations with casting depth using SCC62 cast in walls # 3 and 6

measuring 3.7 and 4.4 m in height, respectively, are shown in Fig. 10.13 (left). The

corresponding variations resulted from the UofS2 pressure column are presented in Fig. 10.13

(right). In case of field measurements, an increase in the pressure envelops was observed for the

two wall elements of the different casting depths. However, the UofS2 pressure column resulted

in similar lateral pressure variations for the two casting depths. The former variations in lateral

pressure can be referred to the differences in PVxorest@i5min values determined for SCC62 used in

the two castings (220 and 410 Pa for walls # 3 and 6, respectively).

225

Lateral pressure (kPa) 0 20 40 60 80 100

(UI

^ ^ s

J3

a « •o Ml a -** SB

9t U

Fig.

0.0

0.5

1.0

1 5

2.0

2.5

3.0

3.5

4.0

10.13

Field measurements SCC62,R=10m/hr

Hydrostatic C\ kf pressure A IX v ^ \

(T

^

ja a

Ml

IX!


— r — — — i i i i 1

UofS2 pressure column SCC62, R = 10 m/hr

Hydrostatic J/ pressure

- <?/

5.0 L

Effect of casting depth on lateral pressure envelops of SCC62: from field observations (left)

and the UofS2 pressure column (right)

10.3.4 Effect of mix design approach

The lateral pressure envelops for SCC62 and SCC63 (walls # 6 and 7) made with paste

volumes (Vp) of 330 and 370 1/m3, respectively, are shown in Fig. 10.14. The two SCC mixtures

were cast at 10 m/hr. The monitored lateral pressure variations with casting depth were similar to

those determined using the UofS2 pressure column, as presented in Fig. 10.14. As expected, the

increase in Vp resulted in higher lateral pressure. The w/cm for SCC62 and SCC63 were 0.37 and

0.35, respectively. This reduction in w/cm between the two concretes reduced the effect of Vp on

lateral pressure variation.

SCC was proportioned with relatively low w/cm of 0.37 with relatively low dosage of

VMA and with a high w/cm of 0.42 and greater VMA content to secure adequate stability. The

lateral pressure envelop of the SCC63 made with 0.42 w/cm and 1.5 1/m VMA concentration

(wall # 6) is compared to that obtained from SCC64 made with w/cm of 0.37 and VMA dosage of

1.0 1/m3 (wall # 8) in Fig. 10.15. Similar results are observed from the UofS2 pressure column.

The increase in the w/cm from 0.37 to 0.42 in SCC63 and SCC64, respectively, led to an increase

in the lateral pressure variation up to 2.3 m depth. Beyond the 2.3 m depth a reduction was

observed for unspecified reason. On the other hand, the UofS2 pressure column resulted in similar

pressure envelops due to the approximate thixotropy level of the two SCC mixtures.

226

o. a -a ex

u

UofS2 pressure column R = 10 m/hr

Hydrostatic ,y pressure

Wall # 6, SCC62, Vp = 330 l/m3

- w/cm =0.35

Wall # 7, SCC63, Vp = 370 1/m3

w/cm = 0.35

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

Fig. 10.14 Effect of paste volume and w/cm on the variations of lateral pressure with casting depth

determined from field measurements (left) and using the UofS2 pressure column (right)

/"~v

E

J3 hfi 4>

J3

o

E i -o to

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

^

- \

\

-

Wall # 6 , SCC62, " Vp = 330 1 w/cm = 0.1

1 i

I

Field measurements R =

w * w > V\ x

m-\ \ 7 *

= 10 m/hr


Wall # 7, \ y SCC63, ^ < Vp = 3701/m3, \ \ w/cm = 0.35 \ \ k \ N

» \ \ \


Field measurements R = 10 m/hr


Wall # 8, SCC64, w/cm = 0.42, VMA = 1.5 l/m3

Wall # 6, SCC62, w/cm = 0.37,

A = 1.0 l/m3

dept

h (m

) C

asti

ng

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5


UofS2 pressure column R = 10 m/hr


Wall # 6, SCC62, w/cm = 0.37, VMA = 1.0 l/m3

Wall # 8, SCC64. w/cm = 0.42 VMA = 1.5 1/m3

Effect of VMA concentration and w/cm on the variations of lateral pressure with casting

depth: from field measurements (left) and the UofS2 pressure column (right)

10.4 Test results of casting column elements and discussion

Typical results obtained from the eight columns cast at the CTLGroup laboratory are shown

in Fig. 10.16 for the concrete used for casting two columns. Other results are given in reference

227

RMC-SDC [2009]. Lateral pressure variations with time using the SCC designed with high

thixotropy (SCC-H) that was employed to cast column elements # 5 and 6 are shown in Fig.

10.16. For each column, the pressure variations are monitored using two pressure sensors fixed at

depths of 3.35 and 2.74 m. The variation in lateral pressure measured from the column elements

are compared to those determined using a 0.9 m in height PVC column measuring 0.25 m in

diameter. The pressure was monitored during 14 hr after the end of casting. The pressure

variations for SCC-H cast at rate of 1.94 m/hr is significantly lower than that determined for the

same SCC cast at 13.7 m/hr (columns # 5 and 6, respectively). Pmax values of 46 and 75 kPa were

determined from the pressure sensor located at depth of 3.35 m for SCC cast at 1.94 and 13.7

m/hr, respectively.

For SCC-H cast at 1.94 m/hr, Pmax values of 39 and 46 kPa were obtained at depths of 2.74

and 3.35 m, respectively. No cancellation time was recorded for these castings due to thermal

expansion of concrete. Once the expansion due to temperature rise is taken into consideration, the

lateral pressure cancellation was achieved. More details about the calculations to modify the

results to account the deformations due to thermal expansion at early age are reported in RMC-

SDC [2009].

The 0.9-m high PVC column resulted in pressure decay similar to that obtained from the

columns elements at early stages; however, the lateral pressure of the PVC column did exhibit

pressure cancellation after about 13 hr.

SCC-H = C = Col #6, R = 13.7 m/hr, H = 3.35 m < = > Col #6, R= 13.7 m/hr, H = 2.74 m ^ ^ C o l #5, R = 1.9 m/hr, H = 3.35 m — C o l #5, R = 1.9 m/hr, H = 2.74 m

Fig.

0

10.16

10 12 14 2 4 6 8 Time (hr)

Variations of lateral pressure with time for columns # 5 and 6 cast using SCC-H

228


As expected, the variations of Pmax with casting depth for the SCC-H mixture cast at 1.94

m/hr is lower than that determined for the same SCC cast at 13.7 m/hr (columns # 5 and 6,

respectively), as shown in Fig. 10.17 (left). The concrete cast at the former rate was close to

hydrostatic. The lateral pressure envelop for SCC mixture with medium thixotropy (SCC-M) cast

continuously at 5 m/hr is compared in Fig. 10.17 (right) to that cast at the same rate, except with

a 20-min waiting period (WP) at mid casting (columns # 7 and. 8, respectively). Interrupting the

casting by 20 min resulted in a reduction in Pmax values at a depth of 3.36 m from 65 to 59 kPa.

More analysis and discussions for the test results obtained from the column elements can be

found in reference RMC-SDC [2009].


20 40 60 80 100


20 40 60 80 100

SCC-M R = 5 m/hr

Col. # 7 without WP

Col. # 6 R= 13.7 m/hr


Fig. 10.17 Effect of casting rate (left) and waiting period (right) on lateral pressure envelop

10.5 Validation of formwork pressure models using field measurement results

10.5.1 Validation of Pmax and Ko models

The measured Pmax values monitored using the pressure sensors mounted at different

casting depths of the six wall panels (walls # 2 to 4 and 6 to 8) and eight column elements

(columns # 1 to 8) were compared to the corresponding values obtained from the prediction

model of Pmax proposed in Chapter 9 (Eq. 9.92), as shown in Fig. 10.18 (top).

229


100

80 -

£ 60

s so «

40

S 20

-

-

/'

/

/

/

/'*}/ /£ / y = 1.01x / JS A / W = 0.97 (J^Y /""

/xi& / /™v

/ y&£ / • CTLGroup 8 columns (results of y ' / columns # 3 and 6 were excluded

/ /' due to overestimate of thixotropy)

' ./ XCFI wall panels cast with SCC /' mixtures (6 walls)

i i i i

100

90

^

I 2 80

u

4>

70 -

60 '

50 50

20 40 60

Predicted Pmax (kPa)

80 100

-

-

-

/ '

s

y\ A X / X X

A . / XX XV^ x x X

/ A XA* / / A / x/

/ x w . / X /

y = 1.02x A / R2 = 0.90 /

/

/ A /

/ A

/

L - ' 1 1 1

60 90 100 70 80

Predicted K„ (%)

Fig. 10.18 Measured to predicted lateral pressure values for CFI walls and CTLGroup columns

230


Pmax = pgH [98 - 3.82 H + 0.63 R + 0.011 Dmjn - 0.021 PViorest® 15min@Ti ] X/MSA X/WP Eq. 9.92

The effect of concrete temperature is included in the PVx0rest@i5min index determined at the

same temperature of concrete cast in the field. The results of columns # 3 and 6 were excluded

from the correlation shown in Fig. 10.18 due to difference in shear histories of concrete samples

used for the PV test and that actually cast in the columns. Indeed, the concrete used in casting the

two columns were re-mixed before placement, which resulted in a breakdown of the internal

structure of the concrete leading to lower structural build-up values. On the other hand, the concrete

sample used in the PV test was kept without agitation and had high structural build-up values.

The comparison of field results (in total six walls and eight columns elements) yield good

correlation, as shown in Fig. 10.18 (top) where the R2 value of 0.97 was obtained. On the other

hand, the correlation between measured and predicted Ko values for the same casting conditions

(Fig. 10.18, bottom) resulted in a lower R value of 0.90. The Ko is the ratio between Pmax and

corresponding equivalent hydrostatic pressure (Phyd)- Introducing the Phyd value to calculate Ko

resulted in greater dispersion of the results. The upper and lower limits corresponding to 90 %

confidence level for the two correlations between the measured and predicted Pmax and Ko are

indicated in Fig. 10.18. These limits were calculated as reported in Assaad [2004].

10.5.2 Validation of AK(t) models

The measured index of pressure decay during the first 60 min [AK(t)(0-60 min)] and that

during the time required for pressure cancellation [AK(t)(0-tc)] of the field-cast SCC were

compared to predicted values calculated from models proposed in Chapter 9 (Eqs. 9.95, 9.96, and

8.2, shown below). These comparisons are given in Figs. 10.19 and 10.20, respectively. Six data

points for six wall elements and four data points for column elements cast at different days were

employed in the comparison of the AK(t)(0-60 min) values. Since no cancellation times were

recorded for the column elements due to the thermal expansion, only the six data points for the

six wall elements were considered in the comparison between the measured and predicted

AK(t)(0-tc) values. The two indices of the pressure decay were determined for the field

measurements using the deepest pressure sensor of the wall and columns elements.

AK(t)(0-60 min) = [0.1092 + 1.12x 10^ x PVTorest@i5min] *^Dmin Eq. 9.95

AK(t)(0-tc) = [0.1491 + 6.57 xlQ"7 PVT0rest@15minXX0reSt(t)] x^Dmin Eq. 9.96

231

where: f2Dmin = 1.260353 - 0.001302 D . •* " • • m i m m

Eq. 8.2

The two correlations shown in Figs. 10.19 and 10.20 indicate that the prediction models

can offer adequate level of estimate of the decay in lateral pressure. This is in exception of wall #

8 that resulted in higher measured-to-predicted AK(t)(0-60 min) value. The actual measurements of

the lateral pressure decay indices, and consequently the predicted values, are found to be in a narrow

range of the data points needed to validate the prediction models. Thus, further field observations

should be carried out to validate further the pressure decay models.

s a

0.4

a 0.3

X0.2

< •a s «

0.1

0.0

ACFI 6 wall elements OCTLGroup 4 casting days

0.0 0.1 0.2 0.3 0.4

Predicted AK(t)(0-60 min) (%/min)

Fig. 10.19 Measured-to-predicted

AK(t)(0-60 min) values

0.2

X 0.1

< a> i_ S in 08

0.0

A FCI 6 wall panels

0.0 0.1 0.2

Predicted AK(0-tc min) (%/min)

Fig. 10.20 Measured-to-predicted

AK(t)(0-tc) values

10.6 Conclusions

Based on the results of the field validations presented in this chapter, the following results

can be made:

1. Pressure decay obtained from the small scale PVC column was similar to that obtained from

wall and column castings. The pressure cancellation times resulted from casting SCC62 in

the PVC column and wall # 6 (from the pressure sensor fixed at 3.85-m depth) were 915 and

975 min, respectively. The PVC column can therefore be used to assess the effect of mixture

composition and concrete temperature on variations in lateral pressure with time.

232


2. The UofS2 pressure column resulted in similar variations of lateral pressure with casting

depth to those obtained from the actual casting in wall and column elements. The pressure

column reflected well the effect of casting rate and other mixture proportioning.

3. Either of the portable vane or the inclined plane test methods can be successfully employed

to determine the structural build-up of SCC at rest. These indices can be used to differentiate

between SCC mixtures of various compositions and help in better selection of the mix design

that exerts lower lateral pressure on formwork.

4. The field-oriented models proposed in Chapter 9 were successfully validated using six wall

and eight column elements. The relationship between the measured and predicted Pmax values

resulted in high R2 value of 0.97.

5. The developed model for lateral pressure decay during the first 60 min following the end of

casting and that over the pressure cancellation period resulted in adequate estimate for the

majority of the wall and column elements.

233

CHAPTER 11

SUMMARY AND CONCLUSIONS

11.1 Introduction

The study presented in this thesis aimed at developing a pressure column device that can be

used to predict lateral pressure exerted by SCC, as well as field-oriented test methods to

determine the structural build-up at rest of SCC. The role of material constituents, mix design,

concrete casting rate, formwork geometry, and temperature on SCC form pressure are

investigated. This study aims also at proposing design equations to predict formwork pressure of

SCC on column and wall.

11.2 Measuring formwork pressure

11.2.1 Portable device to measure maximum formwork pressure

A portable pressure device (referred to here as the UofS2 pressure column) was developed

to evaluate lateral pressure exerted by plastic concrete. The UofS2 pressure column has a circular

cross-section measuring 200 mm in internal diameter, 700 mm in total height and 10 mm in wall

thickness. The column is initially filled to a height of 0.5 m with SCC at a given rate of casting,

without any vibration. The top of the pressure column is then closed, and air pressure is gradually

introduced from the top to simulate pressure increase at a given placement rate. The device

enables the simulation of castings sections up to 13 m in height. A pressure sensor is set flush

with the fresh concrete surface at 63 mm from the base of the column to record the exerted lateral

pressure during casting and to monitor pressure decay. Another transducer is fixed above the

concrete surface at 625 mm from the column base to determine the net overhead pressure inside

the column. The pressure device is demolded before concrete hardening.

An AB-high-performance pressure transducer supplied by Honeywell is used. The sensor is

extremely accurate with relative error of 0.25% over a wide range of temperature. The sensor has

a capacity of 1380 kPa, and measures 19 mm in diameter. It can operate over a temperature range

varying from -54 to +93°C and is excited using 5 V dc current. The sensors are connected to data

acquisition system to monitor pressure variation at 90-s intervals. In addition to the pressure

sensor, a digital dial-gauge (manometer) is fixed in a small regulating chamber attached to the top

of the UofS2 pressure device to control air pressure on the free concrete surface.

234

Chapter 11 Summary and conclusions

The output results from the UofS2 pressure column can be presented in three forms. The first is

the variation of lateral pressure with time, which can be referred to as pressure decay. The second and

third forms are the variations of maximum lateral pressure (Pmax) and relative lateral pressure (Ko)

with casting concrete depths (H), (Pmax vs. H) and (Ko vs. H), respectively.

The repeatability of the UofS2 pressure column was determined using typical SCC mixture

employed for cast-in-place application that was prepared four times. The results of the relative

errors (RE) in the Ko values at various depths are given in Table 11.1 and indicate high precision of

pressure measurements. The lateral pressure characteristics of SCC mixtures determined using the

UofS2 device filled with 0.5 m and overhead pressure to simulate concrete head of 3 m were also

compared to measurements obtained from a free standing PVC column of the same diameter filled

with 3 m of SCC. Good agreement was obtained between both systems in terms of initial lateral

pressure and pressure drop in time. The UofS2 pressure device can also be used to evaluate early

decay in lateral pressure, up to 2 hours after the end of casting. The pressure device was validated

using SCC mixtures made with various material characteristics, mix design parameters, and cast at

different placement rates. The findings from the pressure device are comparable to published

results.

Table 11.1 Relative error in predicting relative lateral pressure value

Concrete height, H (m) Relative error (%)

1 ±0.7

4 ±2.4

8 ±2.3

12 ± 4 ^

11.2.2 Evaluation of lateral pressure decay

A 1.2-m high PVC column measuring 0.2 m in diameter was used to monitor lateral

pressure variations until cancellation (pressure decay). The column is made of rigid PVC with 10

mm wall thickness and has a smooth inner surface to minimize friction with concrete during

placement. The PVC column has a seam along its height to facilitate demolding and is tightened

along the height by radial ties. The inner surface of the column is coated lightly with formwork

release agent prior to each use. The SCC mixtures are cast continuously at the desired rate from

the top without any vibration. The concrete pressure is monitored using three pressure sensors of

100-kPa capacity mounted at 1.0, 0.8, and 0.6 m from the top surface of concrete.

235


11.3 Field-oriented test methods to evaluate structural build-up at rest of SCC

The portable vane and inclined plane field-oriented test methods are simple and can easily

be used in laboratory and in the field to determine the rate of structural build-up at rest of SCC.

11.3.1 Portable vane test

The portable vane test is inspired from a field test for the in-situ measurement of shear

strength of clay soil. Four-blade vanes of different sizes (Table 11.2) were manufactured from

stainless steel to enable use of high precision torque-meter to capture shear strength of the plastic

concrete after various times of rest (typically 15, 30, 45, and 60 min). The largest vane is used for

the weakest structure, i.e., shortest resting time, and vice versa.

Table 11.2 Vane dimensions of the portable vane test

Vane

Vane # 1 (large vane)

Vane #2

Vane #3

Vane # 4 (small vane)

Time at rest (min)

15

30

45

60

Vane dimensions (mm)

R H

37.5

37.0

37.5

37.5

250

200

149

100

h (filling height)

Varies from 50 mm for highly thixotropic SCC mixture to total vane height (H) for relatively low thixotropic mixture

Immediately after mixing, the four vanes are centered vertically in the containers. The

containers are filled with SCC to a given height (h), indicated in Table 11.2. The rested materials

are covered. At the given time of rest, the torque-meter is attached to the axis of the vane and

turned slowly (10 to 15 s for a quarter turn). The maximum torque needed to breakdown the

structure is then noted. The torque values are converted to static shear stress of the portable vane

test (PViorest) using Eqs. 1 and 2 as follows:

G PVr, Orest

where: G=2nr2 1 > 7

J

Eq. 6.1

Eq. 6.2 h+-r V 3

T= measured torque (N.m), h, and r are the filling height and vane diameter (mm), respectively.

11.3.2 Inclined plane test

The inclined plane method involves casting concrete in a cylindrical mould (120 mm in

height and 60 mm in diameter) onto a horizontal plate of a given roughness, then lifting the plate

236


slowly (in 10 sec) to initiate flow of the material at different times of rest. The corresponding

angle necessary to initiate the flow is used to determine the static yield stress of the inclined

plane, IPiorest (Pa), as follows:

Torest = P-g-h.sina (3)

where p is the density of the sample (in kg/m ), g is the gravitation constant (=9.81 m/s ), h is the

characteristic height (in mm) of the slumped sample, and a is the critical angle of the plane (in

degree) when the sample starts to flow. The characteristic height (h) is determined by calculating

the mean value of five heights of the slumped sample near the center of the spread. Four tests are

performed after different periods of rest to evaluate the rate of increase in xorest at rest. For SCC

mixture of relatively low thixotropy level, the time of resting can be considered as: 15, 30, 45, and

60 min, while for highly thixotropic mixtures these times can be decreased to 5, 10,15, and 20 min.

11.3.3 Thixotropic indices from field-oriented test methods

A summary of the various structural build-up indices determined from the PV and IP field-

oriented test methods are presented in Table 11.3.

Table 11.3 Various thixotropic indices obtained from the field-oriented test methods

Static yield stress

Portable vane (PV) test Inclined plane (IP) test

Initial response at 15 min time of resting PVx0rest@i 5min IPto rest@i 5min

Time-dependent change of response (slope) PVrorest(t) IPtorest(t)

C o u p l e effect Of in i t ia l a n d Slope PVT0rest@15minxPVTQrest(t) I P T Q rest@15minxIPTQrest(t)

11.3.4 Validation of PV and IP field-oriented test methods

• The PV and IP test methods show good repeatability and low relative error when used to

assess thixotropy of SCC.

• Static yield stress and its changes with respect to resting time obtained using the portable

vane and inclined plane tests can be used to reflect the change in the rate of structural build

up at rest of SCC mixtures.

• The results of the field-oriented test methods are validated using up to 42 SCC mixtures of

different compositions. Good correlations are obtained using the PV and IP field-oriented

tests in terms of static yield stress at rest and rheometric concrete measurements in terms of

static yield stress at rest, drop in apparent viscosity at rotational frequency of 0.7 rps, and

237

breakdown area. The compared parameters are measured initially and with respect to time of

rest. The R2 values for these correlations are summarized in the following Table 11.4.

Table 11.4 R values for the correlations between PV and IP tests versus concrete rheometer

PVtOrest VS. IPx0rest

PViorest vs. Rheometeriorest IPtorest vs. Rheometerxorest PVxorest VS. Ar|app@N=o.7rps

IPtOrest VS. AtlapP(2)N=0.7rps

PVxorest VS. A b i

IPtOrest VS. A b i

Initial responses Measurements from Measurements

15 - 60 min at 15 min 0.71 0.93 0.78 0.98 0.74

—

0.82 0.82 0.82 0.81 0.67 0.87 0.70

- Time-dependent responses

0.85 0.96 0.93 0.96 0.93

—

11.4 Factors affecting SCC formwork pressure and thixotropy

11.4.1 Investigated parameters

The investigated parameters include mixture proportions, concrete temperature, casting

characteristics, and minimum lateral dimension of formwork, as given in Table 11.5.

Table 11.5 Modeled

Parameter

Casting depth (H)

Placement rate (R)

Concrete temperature (T*)

parameters in the prediction models of SCC formwork pressure

Minimum lateral dimension of formwork Dmjn

Range

l - 1 3 m

2,5, 10, 17, 24, and30m/hr

12, 22, and 30 ± 2°C

200, 250, 300, and 350 mm

• Initial slump flow (<|>) • Volume of coarse aggregate (Vca) • Paste volume (Vp) • Sand-to-total aggregate ratio

(S/A)

are replaced by structural build-up at rest (or thixotropy) indices (T.I.):

• Determined at laboratory temperature of 22±2°C (T.I.@T=22±2°c)

• Determined at various concrete temperature (T.I.@ii)

Maximum-size of aggregate (MSA) 10, 14, and 20 mm

Waiting period between successive lift (WP)

• Continuous • WP of 30 min at middle of casting • Two WPs of 30 min at middle of casting

238

11.4.2 Comparison between SCC and conventional concrete of normal slump consistency

For the investigated cases, the conventional concrete (CC) mixtures developed lower Ko

values, shorter pressure cancellation time (tc), and faster decay in lateral pressure compared to SCC

of similar mixture composition, except for higher concentration of HRWRA.

11.4.3 Influence of parameters on lateral pressure characteristics and thixotropy of SCC

A. Factors affecting Ko

1. The Pmax exerted by SCC is lower than hydrostatic pressure (Phyd)- Large deviation from Phyd

can be observed with the increase in thixotropy. For example, SCC30 with PVxorest@i5min of

815 Pa, cast at 10 m/hr and 22°C, can have a Koi value at 12 m depth as low as 30%.

2. The increase in the <|> of SCC due to the addition of HRWRA (the same mix design) is shown

to increase lateral pressure. For example, increasing $ values from 600 and 720 mm in

SCC27 and SCC31 mixtures, respectively, resulted in Ko@H=8m values of 71% and 77%.

3. Decrease of Vp or increase of Vca in SCC mixture leads to a decrease in Pmax, as indicated in the

table below.

Mixture

SCC36

SCC38

SCC39

Vp(l/m3)

340

370

390

Vca(l/m3)

319

305

295

Ko@H=3m (%)

64

75

87

Ko@H=7m (%)

58

65

74

4. For the same paste volume, increasing the sand-to-total aggregate ratio in SCC mixture leads to

a reduction in coarse aggregate content and thus results in higher lateral pressure.

5. Increasing the MSA produces SCC mixture that exerts lower lateral pressure on formwork. A

correction factor (/MSA) as function of H is proposed to account for the effect of MSA other

than 14 mm on lateral pressure of SCC. The /MSA is 1.0 for relatively high thixotropy SCC

(PVx0rest@i5min > 700 Pa) or for low thixotropy SCC (PVx0rest@i5min < 700 Pa) and

proportioned with MSA of 20 mm. On the other hand, for low thixotropy SCC with 10 mm

MSA, the/MSA is in order of [l+(1.26H-5.04)/100].

6. The increase of concrete temperature results in lower lateral pressure. At T of 12 °C, SCC38

exhibited pressure close to Phyd- However, increasing T to 22 °C and then to 30 °C produced

significant deviation from Phyd- At T of 12, 22, and 30 °C, Ko values, at depth of 7 m for SCC

cast at 5 m/hr, are 71%, 58%, and 42%, respectively.

239


7. At shallow depth, Ko is close to 100%. However, beyond 3-m depth, the pressure envelop

diverged from Phyd. The Ko values decreased linearly with the increase in concrete depth. At H

of 1, 3, and 7 m, SCC38 mixture cast at 5 m/hr and 22°C resulted in K0 values of 81%, 73%,

and 58%, respectively.

8. Ko values increase with the increase in placement rate. For a very high R value of 30 m/hr, Ko

approaches 100% especially at shallow depths. A significant reduction in Ko, even at shallow

castings, is obtained at slow rate (R = 2 m/hr).

9. The maximum formwork lateral pressure decreases with increase in the thixotropy level. The

Ko@Hi can correlate to various thixotropic indices determined using concrete rheometer or

field-oriented tests. Abacuses were established to facilitate the estimate of Ko vs. thixotropy

for various R values.

10. Interruption of concrete casting for a waiting period (WP) of 30 min can reduce Ko by up to

10 %, especially for highly thixotropic SCC.

11. Ko increases with the increase in minimum formwork dimension. A correction factor

(/ Dmin) for Ko is derived to account for changes due to variations of Dmjn between 200 to 350,

as follows C/7Dmin = 0.000968 Dmin + 0.806332).

B. Factors affecting AK(t)

1. The lateral pressure decay is sharper initially than the average decay until pressure

cancellation. For example, SCC with PVxorest@i5min of 740 Pa, showed an initial decay over 60

min [AK(t)(0-60 min)] of 0.22 %/min compared to 0.17 %/min for the mean rate of pressure

decay until pressure cancellation [AK(t)(0-tc min)].

2. The increase in initial slump flow of SCC due to the addition of HRWRA (the same mix

design) is shown to delay the pressure decay.

3. The lateral pressure decays at a slower rate when coarse aggregate volume increased or when

paste volume decreased, as shown in the table below. The reduction in coarse aggregate volume

can be achieved by increasing the sand-to-total aggregate ratio for same paste content.

Mixture

SCC36

SCC38

SCC39

Vp(l/m3)

340

370

390

VCa(l/m3)

319

305

295

AK(t)(0-tc) (%/min)

0.177

0.265

0.312

240

4. The pressure decay increases with the increase in concrete temperature. At T of 12, 22, and

30 °C, the SCC38 cast at 5 m/hr had [AK(t)(0-tc)] values of 0.20, 0.21, and 0.35 %/min,

respectively.

5. The pressure decay decreases with the increase in minimum form work dimension. At Dmjn of

200, 250, 300, and 350 mm, the [AK(t)(0-tc)] values at H of 1.45 m for SCC38 were 0.33, 0.30,

0.27, and 0.25 %/min, respectively. Correction factor (/2Dmin) for AK(t)(0-tc) accounting for the

changes in Dmjn can be calculated as^Dmin = 1.260353 - 0.0001302 Dmjn.

6. The casting rate and waiting period between lifts have no influence on lateral pressure decay.

C. Factors affecting tc

1. Increasing concrete temperature reduces pressure cancellation time. At T of 12, 22, and

30 ± 2 °C, SCC38 cast at 5 m/hr and had tc values of 470, 385, and 230 min, respectively.

2. tc increases with the increase in Dmin. For SCC38, increasing Dmjn from 200 to 350 mm led to

an increase in tc of about 110 min. At a given concrete depth, the tc measured from the two

lateral dimensions were found to be same, regardless of the ratio between the cross-sectional

dimensions.

3. The R and WP do not have any influence on pressure cancellation time.

D. Factors affecting thixotropy

1. The increase in initial slump flow of SCC due to the addition of HRWRA (the same mix

design) is shown to reduce thixotropy. This can be illustrated from SCC27 and SCC31 that

had <|> values of 600 and 720 mm, respectively, as shown in the table below.

Mixture <|> PVxorest@ismin (Pa)

SCC27 600 327

SCC31 720 210 2. Decreasing the paste volume or increasing the coarse aggregate volume leads to an increase in

thixotropy, as shown in the table below.

Mixture

SCC36

SCC38

SCC39

Vp(l/m3)

340

370

390

Vca(l/m3) Abi(J/m3.s)

319 543

305 440

295 230

241

3. Proportioning SCC mixture with coarse aggregate of larger MSA that has higher packing

density increases the internal friction and leads to higher thixotropy level.

11.4.4 Statistical models for lateral pressure characteristics and thixotropy of SCC to

simulate effect of <j>, Vca, and S/A

Six statistical models to predict lateral pressure characteristics and 10 other models to

estimate thixotropic properties as a function of mix design parameters (slump flow, sand-to-total

aggregate ratio, and coarse aggregate content) were established. The derived models had high R2

values varying between 0.84 and 0.99. Contour diagrams for predicting responses are established

to illustrate trade-offs between the effect of different parameters on the tested responses

pertaining to thixotropy and formwork pressure. These diagrams can be used as a guideline for

the determination of lateral pressure and thixotropy of SCC.

11.5 Model for lateral pressure prediction

11.5.1 Models for Pmax prediction

The results obtained from approximately 800 data points were used to establish models to

predict Pmax of SCC (UofS model). The models were designed in terms of casting depth (H),

placement rate (R), concrete temperature (T), minimum lateral dimension of formwork (Dmjn),

and thixotropy index. The latter can be determined from the PV and IP field-oriented test

methods or concrete rheometer and expressed at the actual concrete temperature (T.I.@TD or at

22±2°C (T.I.@T=22±2',C) with a correction factor for actual temperature. The effect of waiting

period between successive lifts (WP) and maximum-size of aggregate (MSA) are also determined

in the prediction models. The recommended prediction models for Pmax that resulted in the best

correlation between the predicted and measured responses and with the highest R values are:

• for thixotropy index determined at laboratory temperature (T.I.@T=22±2°C):

Pmax = pgH [112.5 - 3.8 H + 0.6 R - 0.6 T + 0.01 Dmin - 0.021 PVT0rest@i5 Eq. 9.91 min@T=22±2°c] x /MSA X fwp

• for thixotropy index determined at various concrete temperatures (T.I.@Ti):

Pmax = pgH [98 - 3.82 H + 0.63 R + 0.011 Dmin - 0.021 PVT0rest@i5min@Ti] * Eq. 9.92 /MSA X/WP

where: H - 1 - 13 m

R = 2-30m/hr

242

T = 1 2 t o 3 0 ± 2 ° C

Dmin = 200 - 350 mm

PVT0rest@15min = 0 - 2 0 0 0 P a

PVTorest(t) = 0-125 Pa/min

fush and/wp are dimensionless

/MSA '• correction factor for MSA different than 14 mm, and estimated as follows:

> For relatively low thixotropy SCC [PVtorest@i5 min < 700 Pa]

H < 4 m /MSA = 1

H = 4 - 1 2 m /MSA = 1 when MSA = 20 mm

1.26 H-5.04 A MSA 1 +

100 .... when MSA = 10 mm

> For high thixotropy SCC [PV x0 rest @is min > 700 Pa]

H = l - 1 2 m /MSA = 1 when MSA = 10 and 20 mm

fwp '• correction factor for WP, and can be determined from the following figure.

1.1

1.0

0.9

^ 0.8

0.7

0.6

continuous casting - — £ 3 E3~

m'n each

200 400 PVT,

600 0rest@15min

800 1000 (Pa)

Excellent agreement is found between the two recommended prediction models for Pmax

that include T.I.@T=22±2°C and T.I @j„ respectively. They correlated in a 1:1 relationship with R

value of 1.0. Abacuses to provide quick and simple prediction of Ko as a function of thixotropy

indices are established for various H and R values.

11.5.2 Models for the prediction of lateral pressure decay

Two prediction models are recommended for estimating lateral pressure decay:

• Rheometric measurements:

AK(t)(0-60 min) = [0.1132 + 0.0005 x Ar|app@ Eq. 9.93

243


AK(t)(0-tc) = [0.14726 + 0.000002 x Rheometerx0rest@i5minXTorest(t)] xf2Dmin Eq. 9.94

• Field-oriented devices:

AK(t)(0-60 min) = [0.1092 + 0.000112 x PVT0rest@i5min] x^Dmin Eq. 9.95

AK(t)(0-tc) = [0.1491 + 0.000000657 x PVT0rest@i5minXxorest(t)] x/2Dmin Eq. 9.96

where: f2Dmin = 1.260353 - 0.001302 Dmjn Eq. 8.2

11.5.3 Validation of the UofS prediction model with the published guidelines

The results of Pmax determined from the UofS prediction model correlate very well to the

model proposed by Khayat and Assaad [2005A]. However, the UofS model considers a wider

range of casting conditions and can be applied using field-oriented tests (PV and IP tests) to

estimate thixotropy. The ACI347 and German Standard DIN 18218 models overestimate Ko for

SCC compared to the UofS model.

11.6 Field measurements and validation of UofS models

Actual field measurements on large large-scale elements were measured and used to

validate the models elaborated in Chapter 9 to estimate lateral pressure characteristics. The lateral

pressure characteristics obtained using the UofS2 pressure device and the 1.2-m high PVC

column were compared to those obtained from the field measurements. Eight wall elements cast

during the construction of the "Integrated Research Laboratory on Materials Valorization and

Innovative and Durable Structures", at the Department of Civil Engineering of the Universite de

Sherbrooke, in Canada, are used as first validation. The second validation involved casting of

eight column elements at the Materials Laboratory of CTLGroup, II, USA. This validation

resulted in the following conclusions:

1. Pressure decay obtained from the small scale PVC column was similar to that obtained from

wall and column castings. The pressure cancellation times resulted from casting SCC62 in

the PVC column and wall # 6 (from the pressure sensor fixed at 3.85-m depth) were 915 and

975 min, respectively. The PVC column can therefore be used to assess the effect of mixture

composition and concrete temperature on variations in lateral pressure with time.

2. The UofS2 pressure column resulted in similar variations of lateral pressure with casting

depth to those obtained from the actual casting in wall and column elements. The pressure

column reflected well the effect of casting rate and other mixture proportioning.

244


3. Either of the portable vane or the inclined plane test methods can be successfully employed

to determine the structural build-up of SCC at rest. These indices can be used to differentiate

between SCC mixtures of various compositions and help in better selection of the mix design

that exerts lower lateral pressure on formwork.

4. The field-oriented models proposed in Chapter 9 were successfully validated using six wall

and eight column elements. The relationship between the measured and predicted Pmax values

resulted in high R value of 0.97.

5. The developed model for lateral pressure decay during the first 60 min following the end of

casting and that over the pressure cancellation period resulted in adequate estimate for the

majority of the wall and column elements.

11.7 Further work

• In our study, the effect of Dmin of formwork (caisson effect) on lateral pressure

distribution was determined using plywood formwork of 1.5 m in height, 0.4 m in length,

and different widths varying from 0.2 to 0.35 m. In order to reduce the volume of the cast

concrete and enable the simulation of deeper castings, a new pressure device shown in

Fig. 11.1 (referred to as UofS3 pressure column) should be designed. The UofS3 pressure

device has a rectangular shape and cross-sectional measurements of 0.2 m in width, 0.4 m

in length, and 0.7 m in height. Two pressure sensors are fixed near the bottom: one at each

side (at 63 mm from the base), and a third sensor near the top above the concrete (at 600

mm from the base). The UofS3 pressure device can be filled with 0.5 m-high concrete

head and then pressurized with air to simulate casting depths up to 13 m. Pressure sensors

having the same characteristics as those used in the UofS2 pressure device can be used.

• Other variables affecting formwork pressure of fresh concrete still require investigation.

This includes formwork surface material, formwork releasing agents, and reinforcement

density.

• More field evaluations using larger formwork dimensions, different variations in SCC

mixture compositions, and casting characteristics are still needed to validate the proposed

models of lateral pressure prediction.

• Study the effect of nearby traffic inducing vibration that can reduce degree of structural

build-up and increase lateral pressure can be investigated.

245


• Investigate effect of mechanical consolidation using vibrators on the lateral pressure (0-3)

exerted by semi-SCC mixtures.

• Modeling and simulation of lateral pressure characteristics using software based on the

finite element analysis. The experimental results obtained throughout the thesis will be

used to validate the new models.

r c ^ , >

-ypl

Fig. 11.1 Proposed design for UofS3 pressure column to simulate the caisson effect on lateral

pressure distribution

246

Appendix A: CHAPTER 4

METHODOLOGY FOR LATERAL PRESSURE MEASUREMENTS

Appendix Al: Sensor calibration

A. Mechanical calibration

The pressure sensors were frequently calibrated mechanically. The calibration consists of

applying a certain pressure using mechanical tools (PM in psi) directly on the top surface of the

pressure sensor and subsequently registering the sensor response by data acquisition system

(PIraw data in mV). A coefficient of mechanical calibration (Cm) correlating PM and Plraw data was

then determined from the regression line of the relationship between them. The Cm factor was

used to convert the read milli-volt signal picked up by the pressure transducer (Plraw data) to the

corresponding kPa-pressure value corrected mechanically (P2C0IT.M)- In the example shown in

Table A.l and Fig. A.l, Cm factor of 1.7636 was obtained.

Table A. 1 Mechanical pressure vs. output signal to determine mechanical calibration factor (Cm)

Applied

psi

0

3

6

9

12

15

mechanical pressure (PM)

kPa

0

20.7

41.4

62.1

82.7

103.4

Output signal picked up by pressure sensor (Plraw data in mV)

0

0.011

0.023

0.035

0.047

0.059

247

Appendix A: chapter 4 Methodology for lateral pressure measurements

2 120 r OH

«T 100

1 _80 ££60 •a 40

| 20

0 0 10 20 30 40 50 60 70 Signal picked up by pressure

sensor (Plrawdata in mV)

Fig. A.l Mechanical calibration of pressure sensor to determine mechanical calibration factor (C«)

B. Hydrostatic calibration

Verification of the mechanical calibration using hydrostatic calibration (water column and

overhead air pressure, which have known specific gravities) was conducted monthly. The

hydrostatic calibration was carried out directly on the pressure sensors that are already mounted

on the pressure columns. In this calibration, the pressure tube is filled with water to four

predetermined levels (Fig. A.2) then sealed to exert addition pressure using compressed air. The

total applied hydrostatic pressure (PH) was determined and correlated to the output signals of the

pressure sensor modified by the Cm (P2COIT.M)- An example for calculation of PH and P2COrr.M is

indicated in Table A.2. The relationship between PH and P2C0IT.M used in Table A.2 is shown in

Fig. A.3. A hydrostatic calibration factor (CH) of 1.0004 is deducted from this calibration. The CH

factor is employed to adjust the P2corr.M values and produce a new pressure values that corrected

hydrostatically (P3COIT.H)-

248


yi—^f

500 mm

700 mm

850 mm 1000 mm ¥-

Level 4

Level 3

Level 2

©

O Level 1

O

O

T 250 mm

L-

Sensor # 4

250 mm

250 mm

Sensor # 3

Sensor #2 250 mm

Sensor # 1 —

Fig. A.2 Sketch for UofSl pressure column indicating sensor positions and water levels used in

the hydrostatic calibration

K 180 OH

€ 150 s | 120 S O

.2 £2 90 -** ^^ A « 60 o I .

1, 30 a 0 ^

r

CH = 1.0004 R2 = 1.00

V s ' '

0 30 60 90 120 150 180 (P2corrM)(kPa)

Fig. A. 3 Relationship between hydrostatic pressure (PH) and the pressure corrected mechanically

(P2COIT.M) to determine hydrostatic calibration factor (CR)

249


Table A.2 Calculation of hydrostatic pressure (PH) and P2corr.M to determine hydrostatic

calibration factor (CH)

* S B •§ % B

Fille

le

vels

up

to

937

mm

of

/erh

ead

air

sure

12 ° 2

- 5 03

Applied air

pressure

kPa

0

30

40

60

80

100

120

136

140

Head of water (H)

mm

0

209

371

641

937

937

937

937

937

937

937

937

937

Applied hydrostatic pressure (PH)

Pressure of water column (Pw=p.g.H)

kPa

0

2.04 3.64

6.28

9.2

9.2

9.2

9.2

9.2

9.2

9.2

9.2

9.2

Overhead air pressure

(Pair)

kPa

0

30

40

60

80

100

120

136

140

PH=Pw+Pair

kPa

0

2.04 3.64

6.28

9.2

39.2

49.2

69.2

89.2

109.2

129.2

145.2

149.2

Pressure corrected by *--m V"-^COIT.M)

kPa

0.000

3.55 4.15

5.92

9.48

38.37

46.56

69.63

88.52

109.9

131.96

144.31

147.98

C. Water calibration prior to each use

To fully validate pressure sensor results, daily calibration was conducted using water column.

This type of calibration was undertaken prior each use of the pressure cells that mounted in the

pressure set-ups. Beside its simplicity, water calibration takes a short time to carry out and

simulates real condition of the pressure cells. In the water calibration, the instrumented pressure

tube prepared to cast the concrete is filled with water up to four levels similar to that shown in

Fig. A.2. The hydrostatic pressure related to each water level (Pw) is calculated. The milli-volt

signals picked up by the pressure sensor and corrected by the Cm and CH coefficients to produce

(P3Corr H) were determined. An example for computing Pw and P3corr.H is indicated in

Table A.3. Water calibration factor (Cw) of 0.998 was obtained from the relationship

between Pw and P3COrr.H (Fig. A.4). The Cw is used to modify the P3C0rr.H value to a new pressure

value (P4COrr.w)-

250

Table A.3 Calculation of water pressure and P3corr.H to determine water calibration factor (Cw)

Water pressure

Head of water, H (m) P = pw.g.H (kPa) Pressure corrected by Cm and CH (P3COIT.H)

(kPa)

0

208.5

371.0

640.5

937.0

0

2.04

3.64

6.28

9.19

0.000

3.554

4.146

5.922

9.476

u s so in <U l a a u

10

8

6

4

2

0

c = R2

-

y/6

= 0.998 / = 0 . 9 9 ^ ^

/ v

i i i i

0 2 4 6 8 10 P3corr.H(kPa)

Fig. A.4 Relationship between water pressure and P3corr H to obtain water calibration factor (Cw)

251

OO

3*

3 o

cr

3 13 I o

o

CS

CD

3>

l-t o w

X

CD

OO

O

n ON

l—i > W

O

O

m

00 o o ON

tO

> CO

s 13 00 o o ON

> to

S

CD

C/5

c

r o

t3 3"

H

P cr

CD

t—^

o k>

OO

n o 4^

O

^ OO

O

O

4

^ O

N

GO

n o Ul •—'

c?

OO

o n v/i

ON

fa

i-i

CD

CT

O

lB 5'

> •a

T3 ft)

CJ a.

x" a >""*

00 O

o to

(S\

<-+

•

o

on

o o JO

B D-

00 O

O

U)

U) O

^ O

w

•fl

C/3 cr

o 1 3*

£ %

CD

CS

&*

x"

n K-*

O0

O0

00

00

OO

0

0 0

0

o o

n o

o n

o n

n n

o o

n o

to

w

w

w

w

w

to

4^-

OJ

tO

K^

<ul

UJ

O

DO

>

W

W

(O

to

N)

M

W

U>

4^

tO

IO

tO

tO

4>

> O

N

ON

U

I 0

\ o

\ U

I

—1

ON

O

N

ON

O

N

ON

O

i K

) O

O

O

N

ON

O

N

NO

O

O

O

O

O

O

W

I

O

tO

NO '»

->

Ul

[O

^

H-

si

!t»

ui

lo

oo

to

o u>

JO

JO

JO

JO

tO

tO J

O

"to

t*

j To

"u

> to

L

o t

o

^

i-

OO

O

N

•-

O

^

NO

s

i O

>

-'

4^

sO

tO

LO

©

to

w

to

bo

©

bo

NO

00

00

00

00

no

on

no

on

I—»

I—

* I

—*

I—

»

SO

00

-J

ON

to

to

to

to

4^

tO

tO

tO

w

w

ON

In

Ui

Ui

W

ON

u

i U

I O

N

si

o

o

o

o

U>

4^

l/l

NO

O

N

4^

SO

4^

JO

JO J

O J

O

li)

U)

INJ

W

I-1

tO

si

©

U>

00

NO

O

N

W

W

M

M

Lo

ON

O

N 4

^

SCC9 I

099

_ NO

to

2,370 bo

SCC12 to

4^

isi 099

N>

L

/i i i i i

SCC46E I ( " " ) to

to

to

685 C/i

L» 2,350

SCC46D * to

to

665 to

NO

s

i 2,340 I I

SCC46C 3 to

to

665 to

si

to

2,330 I I

SCC46B ( " " ) to

to

683 I 2,340 i I

SCC46A (high thixo.) to

NO

695 <_o

O

N

00 2,340 I I

SCC40E i

( „ „ )

to

bo

655 4^

NO

4

^ 2,360 I I

SCC40D ( " " ) to

to

NO

663 Ul

4^ o 2,360 I

SCC40C 5 to

to

670 4^ to

to

2,370 I

SCC40B ( " " ) to

to

bo

667 Lti

©

NO

2,380 SCC40A (low thixo.) to

to

655 Ln

to

2,360 I

X

CD o o a o

-—

v •"

*

o C

D

CD 3

C?

^3

GfQ

«

3~

* cr

i! CD

3

•a

H

2 CD*

GO

|-rj

CD £•

con o

1-1

CD

CD

prope 3- CD

l/l

ft s 8"

x*

03

i—*

re

ST

n

one -i

CB

•a

•t

o

13 CD

ties

g

GO

H

O

«5 n n

r o 6 2 2 H

W

O

H

W

GO

H

2 w

H

HODS H

O

• r d ATE GO

H

RUCTU

> •H

X

B:CH, >*»•

PTERi ON

5 el r o

Appendix B: Chapter 6 Empirical test methods to evaluate structural build-up at rest o/SCC

Appendix B2: Protocols for the field-oriented test methods

A. Protocol of the PV test

A brief protocol for the PV testing can be summarized as follows:

1. Prepare four-square plastic buckets of about 2-mm thick, 200-mm length, and 400 mm in

height. Fix a screw bolt of 4-ram long and 2-mm diameter in middle of the bucket base going

from outer to inner direction of the bucket. Tighten a nut of 4-mm thick to the appeared part

of screw bolt inside the bucket. Therefore, 2 mm of the nut thickness is fastened to the bolt,

and the other 2 mm is hollow to hold the vane's shaft centered (Fig. B. 1).

Vane shaft

Nuiof4-Screw bolt of 4-mm lons> and 2-mm diameter

Bucket base

Fig. B.l Schematic of centering the vane to the bucket base in the PV test

2. The four buckets numbers 1 to 4 should be kept in a place with no disturbance or vibration.

Position the vane of large dimension at the center of bucket # 1 with the aid of the nut

described in step 1. The smallest vane is placed in bucket # 4.

3. Fill up the four containers with concrete (or mortar) up to a height (h) which should not

exceed the total vane's height (H). The casting height (h) depends on the thixotropy level of

the tested mixture. Shorter h of about 50 mm for highly thixotropic mixture and maximum

h (i.e. H) for relatively low thixotropic mixture. Cover each bucket by plastic cover of a

hole having a diameter of 2 mm greater than the vane's shaft diameter.

4. Record the resting time starting from the end of bucket filling. The first time of rest under

covered conditions can be about 15 min. The remaining three resting times for the other

three buckets can be 30, 45, and 60 min.

5. At the end of each rest time, remove the plastic cover and attach the torque-meter (the

torque-meter indicator arrows should be coincided at zero) to the top tip of the vane's shaft.

Turn slowly (10 to 15 sec for a quarter turn) until mixture starts to flow.

253

Appendix B: Chapter 6 Empirical test methods to evaluate structural build-up at rest ofSCC

6. Record the torque value and corresponding rest time.

7. Repeat steps (4 to 6) at other rest times.

8. Use Eqs. 6.land 6.2 to obtain the static yield stress.

B. Protocol of the IP test

1. Arrange the four inclined planes with suitable sand papers (of No. 600 grit), fixed properly

on the top surface on a level table. In order to position the sample in the centre of the plane,

mark the centre line and write the distances from one edge of the plane to the other.

2. Lightly spray water on the surface of sand paper prior to testing to avoid any absorption of

moisture by the sand paper from the sample. Fill up the first plexiglass cylinder (60-mm

diameter and 120-mm height) up to 100 mm in case of mortar and up to the top (120 mm)

in case of concrete. The time of filling is noted with respect to initial water-addition time

(WAT) corresponding to cement-water contact.

3. Lift slowly (in 10 sec) to enable uniformly flow of the mixture onto the flat plane surface.

Cover the mixture with cylindrical container covered with a wet cloth to avoid any

evaporation during the rest time. Repeat this operation for the remaining three inclined

plane devices.

4. Determine the density of the cast material.

5. The first time of rest can be about 15 min after the first slump flow measurement. When the

rest time is over, remove the wet cloth and the covering container, measure the spread of

the sample and the characteristic height (h) of the spread by calculating the mean value of

five heights of the slumped sample near the center of the spread in an imaginary center

circle, with the diameter half of that of the slumped sample. The five heights should include

the height in the dead centre of the slumped sample and four heights randomly spread

within the imaginary radius. Lift slowly the first IP until the flow of mixture starts. Switch

on a chronometer, tighten the screws of the IP to fix the inclination.

6. Measure the angle of inclination of the IP with a protractor.

7. Repeat steps 5 and 6 at other rest periods that can be selected considering the degree of

structural build-up of the material.

8. Use Eq. 6.3 to determined the static yield stress at rest.

9. For highly thixotropic mixtures, rest periods of 5, 10, 15, and 20 min are recommended.

254

Appendix B: Chapter 6 Empirical test methods to evaluate structural build-up at rest of SCC

C. Precautions for field-oriented test methods

1. The portable vane should be operated at low and controlled rotation speed for turning the

torque-meter (10 to 15 sec for a quarter turn).

2. The inclined plane must be lifted gradually without any jerking of the upper plates until the

concrete sample starts moving.

3. For thixotropic SCC, choose short resting times of 5 or 10 min. For non-thixotropic SCC,

measurements can be taken after longer resting times (for example, 15 min of rest).

4. For the two field-oriented tests, avoid any vibration and disturbance during the test periods

to prevent any "remolding" of the concrete sample.

255

Appendix C: CHAPTER 7

EFFECT OF SCC MIX DESIGN ON FORM PRESSURE

Appendix CI Fresh concrete properties

Table C. 1 Fresh concrete properties of Phase I

Mixture Concrete temp. Slump flow T50

(°C) (mm) (sec) Unit weight

(kg/m3) Air content

(%)

SCC25 SCC26

SCC27

SCC28

SCC29

SCC30

SCC31

SCC32

SCC33C

SCC33D

SCC33E

SCC33F

20.3 20.5

20.2

20.9

19.8

20.2

20.6

20.9

21.7

20.8

20.8

19.4

600 600

620

610

730

710

730

700

660

670

660

660

1.71 3.19

1.75

3.06

0.88

2.44

1.34

3.37

2.12

1.50

2.15

1.06

2,270 2,352

2,306

2,319

2,277

2,310

2,303

2,367

2,316

2,329

2,317

2,312

0.9 1.5

--

3.8

1.0

1.4

1.6

1.3

—

1.4

1.8

1.4

CC34 20 175' N/A 2,320 1.7

Slump N/A: not applicable

Table C.2 Fresh concrete properties of Phase II

Mixture

SCC36

SCC37

SCC38

SCC39

SCC40

Slump

Initial

720

705

680

700

690

flow (mm)

@120min

640

590

640

500

400

J-Ring (mm)

Initial @40 min

—

680

~

640

640

—

660

—

532

630

Concrete temp. (°C)

—

24

~

24

21

Air content (%)

0.7

0.9

1.7

1.7

1.2

Unit weight (kg/m3)

2,414

2,389

2,394

2,381

2,375

CC35 200* 22 22.8 2,356 Slump

256

Appendix C: Chapter 7 Effect ofSCC mix design on formworkpressure

Table C.3 Fresh concrete properties of Phase III

Mixture

SCC13

SCC5

SCC9

SCC58

SCC59A

SCC60

Concrete temp. (°Q

22.4

23.4

—

23

22.6

22.7

Slump flow (mm)

680

660

660

600

590

590

T50 (sec)

1.50

1.34

1.92

2.38

~

2.69

Unit weight (kg/m3)

2,347

2,325

2,370

2,334

2,344

2,365

Air content (%)

2.5

~

1.8

1.5

1.7

2

HRWRA demand (1/m )

3.37

2.73

2.94

4.83

4.35

4.55

257

Appendix C2 Relationship between Ko and thixotropy

100 90 80 70

C? 60 ~ 50 tf 40

30 20 10 0

-

-

-

" R =

" + = = lOm/hr = 600 -720 mm

" • . .

1

A. ••

x 'A

K0@H=1 m R2 = 0.81

K0@H=4 m R2 = 0.86

K0(S).H=8 m Kr= 0.89

K0@H=12m R2 = 0.89

i

0 2000 500 1000 1500 Rheometerr0 rest(%i5min

(Pa) Fig. C.l Variations of Ko with Rheometerxorest@i5min at different heights of placement

100 r

90 80 70

? 6 0 ^ 5 0 * 40

30 20 10 0

R= lOm/hr <|> = 600 -720 mm

K„@H=1 m ,j R2 = 0.77

""^» »K0®.H=4m A A. X R " ° - 8 4

•A K0f®H=8 m A Rr-0.88

0@H=12m R2 = 0.90

0 200 400 600 800 1000 1200 1400 1600

IPT, 0 rest@15min in ( P ^

Fig. C.2 Variations of Ko with IPTQ rest@i5min at different heights of placement

258

Appendix C: Chapter 7 Effect ofSCC mix design on formwork pressure

90

^ 70

J 50

30

10

K0@H=1 m R2 = 0.79

- „ K0@H=4m . . ^ X R 2 = 0.83

A n " \ K0@H=8m

R= lOm/hr <|> = 600 -720 mm

R2 = 0.84

K0@H=12m L J R2 = 0.84

10 20 30 Rheometerr0 rest(t) (Pa/min)

40

Fig. C.3 Variations of KQ with Rheometerto rest(t) at different heights of placement

^

100

80

60

^ 4 0

20

0 0 5 10 15 20 25

IPT0rest(t) (Pa/min)

Fig. C.4 Variations of KQ with IPio rest(t) at different heights of placement

R= lOm/hr <|> = 600 -720 mm

K0@H=1 m R2 = 0.83

K0@H=4 m R2 = 0.88

K0@H=8 m R2 = 0.89

K0@H=12m R2 = 0.89

259


K0@H=1 m R2 = 0.89

x K0@H=4m R2 = 0.91

. . A K0@H=8in

R2 = 0.91

K0@H=12m R2 = 0.90

0 10000 20000 30000 40000 50000 Rheometerxo rest@i5minXTo rest(t) (Pa2/min)

Fig. C.5 Variations of Ko with Rheometerxo rest@i5minxT0rest(t) at different heights of placement

100 90 80 70

^ 60 ~ 50 * 40

30 20 10 0

Ko@H=l m = 0.87

* R 2 = 0. 4 m

90

"•A"'0' Krj@H=8 m

R2 - 0.91

R= lOm/hr if = 600 -720 mm

K0@H=12m R2 = 0.90

10000 20000 30000

IPf f ) reSt(%15minXTo restW ( P a 2 / m i l l )

Fig. C.6 Variations of Ko with IPxorest@i5minxxorest(t) at different heights of placement

260

Appendix C3 Correlation between AK(t) and thixotropic indices

0.30

^0.25

10.20

X0.15

<0.10

0.05

0.00

0

AK(t)(0-60 min) y = 0.0001x +0.1110

R* = 0.71

A AK(t)(0-tc)

y = 7E-05x +0.1246 R2 = 0.77

500 1000 RheometerTn

2000 1500 l 0 rcst(ajl5niin

(Pa) Fig. C.7 Variations of pressure decay with Rheometercorest@i5min

0.30

0.25

.?0.20 S 2 0.15

< 0.10

0.05

0.00

AK(t)(0-60 min) y = 0.0001x + 0.1l22

i ^ O . 8 2 ^

X

AK(t)(0-tc) y = 0.0001x +0.1286

R2 = 0.68

500 IPx 0 rest@15min

1000 (Pa)

1500

Fig. C.8 Variations of pressure decay with IPTQ rest@i5n

261

0.30

0.25

| 0.20

C0,15

STo.io <

0.05

0.00

0

AK(t)(0-60 min) y = 0.0045x +0.1228

R2 = 0.73

o O AK(t)(0-tt)

y = 0.0029.x + 0.1324 R2 = 0.79

40 10 20 30 Rheometerr0 rest(t) (Pa/min)

Fig. C.9 Variations of pressure decay with Rheometerio rest(t)

0.30

0.25

£0.20

a J 0.15 So.io

^ . 0 5

0.00

AK(t)(0-60 min) ^ y = 0.0066x + 0.1306 _^«rr

- ( ^ g g f T X ^ AK(t)(0-tc) C P y = 0.0039x+0.1396

R2 = 0.77

i i i i i

0 5 10 15 20 25 IP-W(t) (Pa/min)

Fig. CIO Variations of pressure decay with IPTQ rest(t)

262

0.30

0.25

| 0.20

AK(t)(0-6Q min) y = 0.00000292x+0.148

o •• ^ 7 2

AK(t)(0-tc)

y = 0.00000201 x + 0.147 R2 = 0.89

0 10000 20000 30000 40000 50000

RheometerT0rest@15minxTorest(t)(Pa2/min)

Fig. C. 11 Variations of pressure decay with Rheometerxo rest@i5minxto rest(t)

0.30

0.25 /^ •S 0.20 .S 2,0.15 ( +J

g'O.lO <

0.05

0.00

y

o

-

-

AK(t)(0-60min) = 0.00000472x + 0.149 ^^s»

W = °-78J=s^ff:SSS3:^

^ ^ ^ ^ ^ L ^ <•» *"" p — A

AK(t)(0-tc)

y = 0.00000301x +0.149 R2 = 0.83

i i

A ^

i

0 10000 20000 30000

I P T 0 rest@15minXT0 restCO (Pa2 /min)

Fig. C. 12 Variations of pressure decay with IPxo rest@i5minxxo rest(t)

263


Appendix C4 Correlations between various modeled responses (virtual points)

100 90 80

- 70 )£ 60 "2 50

40 30 20 10 0

a. R = 10 m/hr T = 22 ± 2°C

^O@H=4 m (time = 24 min) R2 = 0.97 Ko@H=8m (time = 48 min) R2 - 0.94 Ko@H=i2m (time = 72 min) R2 = 0.88

i . — , i

0 200 400 600 800 1000 1200 1400

Predicted Rhmeometerx0 rest@15 min (Pa)

Fig. C. 13 Relationship between predicted Ko and Rheometerco rest@i5 min values

100

5 80

60

40

20

0

R = 10 m/hr T = 22 ± 2°C

•O@H=4 m (time = 24 min) R2 = 0.95

KO@H=8 m (time = 48 min) R2 = 0.95

K0@H=i2m (time = 72 min)

R2 = 0.95

0

—J—.

50 100 150 200 250 1UU 1JU ZUU Z.JK) 300

Predicted AT|app@N=07rps(g15[nin (Pa.s)

Fig. C.14 Relationship between predicted Ko and Ar|app@N=o.7rps@i 7rps@15min Values

20

0

Ko@H=4m (time = 24 min) R2 = 0.97

Ko@H=8m (time = 48 min) R2 = 0.96

K0@H=i2m (time = 72 min) R2 = 0.96

R = 10 m/hr T = 22 ± 2°C

0 500 1000

Predicted IPT 0 rest@15 min (Pa)

Fig. C. 15 Relationship between predicted Ko and IPTQ rest@i5 mi

1500

t@i5 min values

264


/—s

a ti

** -3

PLH

800 r

700 -

600 -

500 -400 -

300 -200 -

100 -0 L

y = 692.39e° 02x

R2 = 0.97

0 5 10 15 20 Predicted Rheometerx0 rest(t) (Pa/min)

25

Fig. C.16 Relationship between predicted pressure cancellation time (tc) and Rheometerxorest(t) values

800

9)

600

400

200

0

y = 686.10e002* R2 = 0.97

25 0 5 10 15 20 Predicted Ai]app(%N=07rps(t) (Pa.s/min)

Fig. C. 17 Relationship between predicted pressure cancellation time (tc) and AriapP@N=o.7rps(t) values

800

•§600

-o 400

1 200

y = 672.10e001x

R2 = 0.94

20 40 60 Predicted PVT0 rest(t) (Pa/min)

80

Fig. C.18 Relationship between predicted pressure cancellation time (tc) and PVTQ rest(t) values

265

Appendix C: Chapter 7 Effect o/SCC mix design on formworkpressure

1

I

800

700 h

600

500

400

300

200

100

0

y = -80.521n(x) + 735.89 R2 = 0.83

R = 10 m/hr

0 10 15 20 Predicted IPT0 rest(t) (Pa/min)

Fig. C.19 Relationship between predicted pressure cancellation time and IPxorest(t) values 1200 u

I 1?1000 O P H . <u

JB

1 §600 .2 E

800

•73 i ? 400

200

0

y = 239.10e000* R2 = 0.90

R= 10 m/hr

0 500 1000 1500

Predicted PVT0 rest@15 min (Pa) Fig. C.20 Relationship between predicted Rheometerxo rest@15 min <*fld P V T Q rest@15 min Values

1200

omel

04

•* w

edi

s-OH

i .5 in

i 5, t

?

1000

800

600

400

200

0

-

> * * *

y = 260.78e001x ^ ^ ^ ^ ' R2 = 0.88 *^0ISS;°:°

R= 10 m/hr i i i i i

0 50 100 150 200 250

Predicted At]app@N=0 7 rps@is min (Pa.s)

Fig. C.21 Relationship between predicted Rheometerxorest@i5min and Ar|apP@N=o.7rps@i5min values

266

1 u .—

red

DH

a-c

'i w

@ V

0)

s o 4>

1ZUU

1000

800

600

400

200

y = 248.02e000x

R2 = 0.91

R = 10 m/hr

0 500

Predicted IPT„

1000 1500

l 0 rest@15 min

(Pa)

Fig. C.22 Relationship between predicted Rheometerxo rest@i5 min and IPio rest@i5 min values

= 1200 ® V

o H ^s > W

-o <W

•o

£ OH

1000

800

600

400

200

0 R = 10 m/hr

0 500 Predicted IPrn

1500 1000 l 0 rest@15 min

(Pa)

Fig. C.23 Relationship between predicted PVxo rest@i5 min and IPxo rest® is min values

S 30 o

I V

= •= o g

4> • * *

U -o V u. eu

25

20

15

10

5

0

y = 0.71x0-82

R2 = 0.97

0 20 40 i rest

60 (t) (Pa/min)

80 Predicted P V T 0 ,

Fig. C.24 Relationship between predicted Rheometerxo rest(t) and PVxo rest(t) values

267

0 5 10 15 20 Predicted Atiapp@N=0 7 rps(t) (Pa.s/min)

25

Fig. C.25 Relationship between predicted Rheometercorest(t) and Ar)app@N=o.7rps(t) values <~ 25

20 E 4> ,—•s

a .9 is

S £ io

u P-

5

0 5 10

Predicted IPT, 0 rest'

15 (t) (Pa/min)

20

Fig. C.26 Relationship between predicted Rheometerto rest(t) and IPT0 rest(t) values

y = 3.74x R2 = 0.78

• * *

Im

> a. 'O (LI W

•B

OH

in)

= « OH

80

70

60 50

40 SO 70

10

0

10 15 20 Predicted IPT0 r£St(t) (Pa/min)

Fig. C.27 Relationship between predicted PVTQ rest(t) and IPTQ rest(t) values

268

App

endi

x C

: C

hapt

er 7

Eff

ect

ofSC

C m

ix d

esig

n on

form

wor

k pr

essu

re

App

endi

x C

5 C

onto

ur d

iagr

ams

for

the

deri

ved

stat

isti

cal

mod

els

0-33

1.44

0

.45

0.4

6 0

.47

0.4

8 0

.49

0.5

0-

51

0.5

2 S

an

d-t

o-t

ota

l a

gg

re

ga

te

ra

tio

(S/A

) (d

ime

nsi

on

less

)

0.2,

'6.4

4 0.

45

0.46

0.

47

0.48

0.

49

0.5

0.51

0.

52

Sa

nd

-to

-to

tal

ag

gre

ga

te r

ati

o (S

/A)

(dim

ensi

on

less

)

03

3

.2

0.3

2

S-

0.3

1

0.3

£ 0

.29

3 o

0.28

fe

p

(a)

Eff

ect

of (j)

on

K 0@

H=4m

0.3

3

K0@

H=

4 m

, %

S/A

= 0

.44

w

0,2

7 1—

61

30

•^

•

.|

0.32

2-

0.31

|0'2

9

"3

c 0

.2B

0.2

7

X

^<

.-•«

p"

'

- &

yp

<£•

K„(

S>

H =

4

m,

%

^J'/

A

= 0

.48

&

^P

•

^

jg. -

620

54

0 6

60

68

0 S

lum

p-f

low

d

iam

eter

(m

m)

700

720

600

620

640

660

680

Slump-flow diameter (mm)

700

720

(b)

Eff

ect

of S

/A o

n Ko

@H=

4m

Fig.

C.2

8 T

rade

-off

of

diff

eren

t pa

ram

eter

s af

fect

ing

Ko@

H=4m

269

App

endi

x C

: C

hapt

er 7

Eff

ect o

f'SC

C m

ix d

esig

n on

form

wor

k pr

essu

re

0.3

3

K„@

H=

12

m,%

4>=i

<£00

mm

0.27

0.44

0.33

0.45

0

.46

0.4

7 0

.48

0.4

9 0

.5

0.5

1

Sa

nd

-to

-to

tal

ag

gre

ga

te

rati

o (S

/A)

(dim

ensi

on

less

) 0.

52

0.45

0.

46

0.47

0.

48

0.49

0.

5 0.

51

0.52

S

an

d-t

o-t

ota

l a

gg

reg

ate

ra

tio

(S/A

) (d

imen

sio

nle

ss)

(a)

Eff

ect

of <

|> o

n K

0@H

=I2

0.33

0.33

0.32

S 0.31

Ml

CO 0.3

0.29

0.28

o

>

640

660

680

Slump-flow diameter (mm)

700

720

0.27

600 ^

>

^\

^

&

& *

-

y^-

,*&

V

620

640

660

680

Slu

mp

-flo

w d

iam

eter

(m

m)

700

720

(b)

Eff

ect

of S

/A o

n K

0@H

=12

m

Fig.

C.2

9 T

rade

-off

of

diff

eren

t pa

ram

eter

s af

fect

ing

Ko@

H=i2m

270

App

endi

x C

: C

hapt

er 7

Eff

ect

ofSC

C m

ix d

esig

n on

for

mw

orkp

ress

ure

0.33

0.27

0.44

0.

45

0.46

0.

47

0.48

0.

49

0.5

0.51

0.

52

Sand

-to-to

tal a

ggre

gate

rat

io (d

imen

sionl

ess)

Fig

. C.3

0 T

rade

-off

of

diff

eren

t pa

ram

eter

s af

fect

ing

[AK

(t)(

0-60

min

)]

271

App

endi

x C

: C

hapt

er 7

Eff

ect

ofSC

C m

ix d

esig

n on

form

wor

k pr

essu

re

0.33

0

.33

0.29

0.2

8

o >

«o&

<b^>

0.45

0.

46

0.47

0.

48

0.49

0.

5 0.

51

Sai

id-t

o-to

tal

aggr

egat

e ra

tio

(S/A

) (d

imen

sion

less

) 0

.52

0.2

7 l

0.44

,?&

>

&&

<iJ&

0.45

04

6 0.

47

0.48

0.

49

0.5

0.51

S

and

-to-

tota

l ag

greg

ate

rati

o (S

/A)

(dim

ensi

onle

ss)

0.5

2

0.33

3 0.

29 \

-

0.28

(a) E

ffec

t of

(j) o

n t c

0.3

3

*<$&

Sjb

°

<#&

0.

27

'—

60

0

^^°"

_^

620

64

0 6

60

68

0 S

lum

p-f

low

dia

met

er (

mm

) 7

00

72

0 0

.27 6

00

62

0 6

40

66

0 6

80

70

0 7

20

Slu

mp

-flo

w d

iam

eter

(m

m)

(b)

Eff

ect

of S

/A o

nt c

Fig

. C.3

1 T

rade

-off

of

diffe

rent

par

amet

ers

affe

ctin

g t c

272

Volume of coarse aggregate (dimensionless) Volume of coarse aggregate (dimensionless)

Tl OQ

o UJ N>

H i-t p o- a •o

ff

o

iff

n>

p S-•73

3

r-t-ft 11 s

affe

cti

p OQ

g CD

o 3 n> r-f ft l-t H

o

(q)

m ffe

o ft-o |-b

t/3

> O P

g ft o 3 ft r-f ft >-t H o n £+ ®

3 5


Q &

O

-e-o P

o

3 CD

^2 crt

I ?•& O

O O

»_ O)

o 80 BQ o

3 Si ! -I p 0) fa

Volume of coarse aggregate (dimensionless) o o

ijj to

> b •

3 P 9 vt

a 3 o

' "V '

%

V

\

>s ,

%

%

^v \

<P =

720

m

m

Nfe

MK

-III

rh

eom

*

H 0

1 % •o u

, <K •

App

endi

x C

: C

hapt

er 7

Eff

ect

o/SC

C m

ix d

esig

n on

form

wor

k pr

essu

re

0.3

3

0.27

03

3

.2

0.3

2

iS

0 3

1

« cs

flu

(VST

02

9

0 2

8

^T

" f3

'

~^W

'

,-5

MK

.HI

rhe

om

ete

r, x

or„

,(t)

. P

a/m

in

-[

<b =

7?

n

mm

. '6

0.44

0.

45

0.46

0.

47

0.48

0.

49

0.5

0.51

0.

52

Sa

nd

-to

-to

tal

ag

gre

ga

te r

ati

o (S

/A)

(dim

ensi

on

less

)

0 2

7<

—

0.4

4 0.

45

0.46

0.

47

0.48

0.

49

0.5

0.51

0.

52

Sa

nd

-to

-to

tal

ag

gre

ga

te

rati

o (S

/A)

(dim

en

sio

nle

ss)

(a)

Eff

ect

of (j>

on

Rhe

omet

erT

o rest(t

)

0.33

.2

0.32

•

S0

.31

<-

0 3

0.29

-

g 0.

28

3

0.2

7

Xy^

- ^

'^

VI

• -

• ^

•

(t),

P

a/m

in

<;/

A

= n

4

4 N1*

<V-

•a

y?

^^

^^

^

^r^

**"^

i

i -*

**'*

^ i

.3- /

z

0.3

3

60

0 62

0 64

0 66

0 68

0 S

lum

p-f

low

d

iam

eter

(m

m)

700

720

0.27 600

620

640

660

680

Slu

mp

-flo

w

dia

me

ter

(mm

) 7

00

72

0

(b)

Eff

ect

of S

/A o

n rh

eom

eter

To r

est(t

)

Fig.

C.3

3 T

rade

-off

of

diff

eren

t pa

ram

eter

s af

fect

ing

Rhe

omet

erxo

rest

(t)

274

App

endi

x C

: C

hapt

er 7

Eff

ect o

fSC

C m

ix d

esig

n on

for

mw

orkp

ress

ure

0.33

.2

0.32

P

a.s

g> =

60

0 m

m

\<cP

\m

0.32

.2

0.3

2 -

s .S

S-

0.31

« ISC

« O

f

e3

0.2

9

S

0.2

8

0,45

0.

46

0.47

0.

48

0.49

0.

5 0.

51

0.52

S

and

-to-

tota

l ag

greg

ate

rati

o (S

/A)

(dim

ensi

onle

ss)

0,2

7 0.4

4 A«J

* '

,A&

P

^^*"

s$

r

\<&

•$>

...<

b0'

"'

^A

*KT

^><

&~

""~

MK

-III

rh

eo

me

ter

<t> =

72

0 m

m

•\C$>

ao

_.,-

eP"'

. f

^,V

2^

•\tf

>

* A

r|ap

p &

N=

sP

6o

^-^*

£

^n^

0.7

rp

s/ P

a.s

\^P

a^

-

<b»

^^

~

^^

1 •*

Af"

" 0.

45

0.46

0.

47

0.48

0.

49

0.5

0.51

0.

52

San

d-t

o-to

tal

ag

gre

ga

te r

atio

(S

/A)

(dim

ensi

on

less

)

(a)

Eff

ect

of

<|> O

n A

r|app

@N

=0.7

rps@

15m

in

0.3

3 0

.33

0.2

9 h

o g 0.

28

640

660

630

Slu

mp

-flo

w d

iam

eter

(m

m)

72

0 0.

27

/ A

P,0

M

K-I

II r

he

om

ete

r A

n.P

P@

N=O

.7r P

s. P

a-s

S/A

=

0.4

8 2P

VJS>

N*

'

600

620

640

660

680

Slu

mp

-flo

w d

iam

eter

(m

m)

70

0 7

20

(b)

Eff

ect

Of

S/A

O

n A

r| app

@N=o

.7rp

s@l5

min

Fig.

C.3

4 T

rade

-off

of

diff

eren

t pa

ram

eter

s af

fect

ing

Ar| ap

p@N

=o.7r

ps@

i5i

275

App

endi

x C

: C

hapt

er 7

Eff

ect

ofSC

C

mix

des

ign

on fo

rmw

ork

pres

sure

0.33

i

.3

0 3

2

I rh

eo

me

ter,

Ar|

. PP #

N=O

.7 r P

S(t

), P

a.s

/min

O =

60

0 m

m

\*

\T-

0.3

3

V K

-lll

rh

eo

me

ter,

Ar|

. PP @

> N

=O.7

rPs

(t),

P

a.s

/min

0.45

0

46

0.4

7

0.4

8

0 4

9

0.5

0

.51

0

.52

Sa

nd

-to

-to

tal

ag

gre

ga

te r

ati

o (S

/A)

(dim

ensi

on

less

) 0.

45

0.46

0.

47

0.48

0,

49

0.5

0.51

0

.52

Sa

nd

-to

-to

tal

ag

gr

eg

ate

ra

tio

(S/A

) (d

ime

nsi

on

less

)

(a)

Eff

ect

of

<|> o

n A

r| ap P

@N=o

.7rp

s(t)

0.3

3 0

.33

-S

0.3

2

2. 0

.31

bo

0.3

-

0.2

9 -

0.2

8 -

> 64

0 6

60

68

0 70

0 S

lum

p-f

low

d

iam

ete

r (m

m)

72

0 0.

27 60

0 **t

-

-<i- \*

a

*s-

i

AK

• '

•

MK

-lll

rh

eo

me

ter,

Ar)

. PP@

N=o

.7r P

s(t)

, P

a.s

/min

S/A

= 0

.48

ft

^-

' "

^

^-

""

K-

^

^-^

^ ^

^*"

^

620

64

0 66

0 68

0 70

0 S

lum

p-f

low

d

iam

ete

r (m

m)

72

0

(b)

Eff

ect

of

S/A

on

An a

pp@

N=o

.7rp

s(t)

Fig

. C

.35

Tra

de-o

ff o

f di

ffer

ent

para

met

ers

affe

ctin

g A

r) app

@N=o

7rp

s(t)

276

App

endi

x C

: C

hapt

er 7

Eff

ect

ofSC

C m

ix d

esig

n on

for

mw

orkp

ress

ure

0.33

620

640

660

680

700

Slu

mp-

now

dia

met

er (

mm

) 72

0

Fig

. C.3

6 T

rade

-off

of

diff

eren

t pa

ram

eter

s af

fect

ing

IPxo

rest(

t)

277

Appendix D: CHAPTER 8

EFFECT OF PLACEMENT CHARACTERISTICS AND FORMWORK

DIMENSIONS ON LATERAL PRESSURE OF SCC

Appendix Dl: Fresh concrete properties

Table D.l Fresh properties of concrete used to evaluate effect of temperature on lateral

pressure characteristics (Phase I)

Mixtures

SCC38-12

SCC38-22

SCC38-30

CC35-22

Casting rate (m/hr)

5

10

17

24

30

5

10

17

24

30

5

10

17

24

30

5

10

17

Slump flow (mm)

After Initial 120

min

720

690

720

690

690

700

700

700

720

700

710

700

700

710

710

210*

210

215

590

560

-

630

500

590

560

-

630

500

490

-

-

480

490

N/A

N/A

N/A

Concrete temp.

(°C)

13.7

11.8

12.3

12.1

11.7

19

19.4

19.8

-

20.4

30.7

29.6

-

28.5

27.4

19.6

20.3

20

Air volume

(%)

1.9

1.9

0.9

1.7

1.6

1.1

1.0

1.3

-

1.9

0.9

-

1.2

1.0

1.2

3.5

1.7

2.3

Unit weight (kg/m3)

2,329

2,334

2,371

2,347

2,340

2,368

2,382

2,369

2,392

2,351

2,372

2,372

2,368

2,369

2,359

2,320

2,375

2,362

J-Ring spread (mm)

After 10 min

720

690

720

670

640

680

700

-

-

590

700

710

670

680

650

N/A

N/A

N/A

After 40 min

740

750

-

730

730

670

600

-

-

560

610

-

600

600

530

N/A

N/A

N/A r slump

278

SO

tO

C/3

O

o

L/)

LO

tO

h-

H-

-n.

-o.

o L/

> tO

o 3-

ST

w

o •a o\

o\

o\

^ vj

0

0 V

O

so

O

O

o o

o o

o

•—

to

•—

tO

—

i—

•—

oo

as

LO

o vo

Os

-J

to

to

to

K>

H-

O

H-

•—

so

L/i

LO

—

4^

so

^J o o

-J

o o as

S

O

o O

N

, so

O

-a

to

o as

so

o

as

so

o as

so

o

-o

LO

LO

O

O

LO

as

as

u

^ u

u

K

> H

- s

o s

o so

~J

-u

oo

N

Is)

M

KJ

•—

LO

4^

-J

K)

00

tO

K)

H-

LO

as

^ --

M

K)

(O

(s)

(s)

K)

W

O o o v:

U) o

C/3

o

o

L

/>

to

to

p—

>->

4*.

-o

©

o •—

LO

LO

0

0 o to

LO

so

to

to

i—

tO

L*

to

to

to

—

to

to

to

4^.

L/l

J—

L/i

JO

JO

lo

"tO

i—

s

o so

oo

4^ so

Lo

L/I

to

ON

4^

to

to

o to

to

JO

JO

JO

JO

w

w

lo

lo

L/I

L/l

L/>

OO

as

u>

oo

-o

o

o to

to

t—

©

to

C/3

o

o

L

/l

o K-

u>

to

i—

i—

g ©

4-*

. -J

©

K)

i—

tO

LO

•—

> L

O

•--

LO

oo

lo

bo

LO

Lo

4

i.

©

L/I

to

o •o

Os

Os

~-J

ON

O

S O

s O

s so

-~

i ©

as

~j

oo

O

s ©

©©

©©

©©

4^

Os

to

to

to

to

to

O

i—

•—

LO

LO

© to

©

to

ON

©

©

to

oo

vo

.fe.

Lfi

JO

JO

JO

JO

JO

JO

JO

w

"w

w

uj

"w

"w

w

LO

i—

> 4

^ -

J K

) -f

c>

i—>

Ui

Ji

W

U>

*>

vl

OS

o o •1 o v:

© to

©

bo

C/3

o n 4^

©

LO

tO

•—

i--

©

-t>.

-o

©

LO

^-

to

i—'

LO

4^-

to

to

to

—

H-

©

to

to

L/l

tO

-O.

-O.

Os

Os

Os

~J

tO

©

so

so

so

©

©©

©©

©©

•—

•

—

_-

*->

•—

•

—

tO

-O.

L/l

—'

to

»o

oo

to

to

tO

•—

o\

as

©

oo

-j

as

H-

LO

00

4S.

L/l

0O

00

4^

JO

JO

JO

JO

JO

JO

W

W

V

Ik)

W

U)

•-J

Os

tO

L/l

LK

O

s U

l M

O

l Is

) U

^

X

r-r c ft o

P

00

5-

H3 o 3 at

,—s

.

?r

cro

3" o

o 3 r^ ft) 3 r^- d

3_

r^-

OQ

H

s»

a;

a a to

CD

(/I 3- I

res C o

3"

fa

^ P o

f—k n •-l

^

o

C/l ^ 3*

V) CD

HH

C/3

O

o

3 c •-«

<D

w

£ C

/J

ft)

O.

r+

o ft) 

fD

tb

f^

o c-k

o >"+5

o

to

(/)

r-k 3 OQ

>-t

P

r-k

fl>

o 3 N—

P

r-k

ft) -

P

9 Oo

rs

^3

" 1 1 ? O s 8.

K 51 1

to

00 o

to

to

C/3

o

o

as

as

ji

u)

to

tO

SO

tO

K)

tO

U)

-O.

to

to

to

-J

4*.

to

Ul

tO

Tl

O

«•

w

to

—'

—

g O

4=>

-O

. O

r-

t*

o •a

v:

as

as

as

as

as

as

as

so

>J

vi

vo

vo

si

vo

o o

o o

o o

o

o

-o

to

to

4*.

4*

to

oo

o to

oo

i«

to

to

JO

JO

JO

JO

JO

JO

JO

V

V

w

w

w

w

V

<-/i

o

oo

tyi

-fc>.

to

UJ

VO

O

O

~J

-J

4*.

O

^-

to

as

o

00

o

O

tO

H-

H-

-&>

^J

O

l>0 as

as

o o

S

O

M

to

to

M

W

U

H-

tO

•—

I—

L/i

bo

to

o

u>

to

TJ o I -j

-j

-J

-J

~-J

^J

~J

_-

_ o

H

- O

O

O

o

o

o

o

o

o

o

ii

ii

as

<-«

as

u>

OJ

as

so

o

to

to

to

to

to

H-

u>

to

U)

00

00

00

OO

U

> 0

0

oo

as

JO

JO

JO

JO

JO

JO

JO

~*J>

~4

^ "j

i. V

. W

"V

"j

> o

o •—

o

as

>—>

i—

oo

to

as

to

-o.

as

<~n

Tl

o

3* o

°

to

to

O

-&>

^

to

00

o

n

as

tO

!—

—

-e-

-J

o

I ^4

<l

-O

~-J

o

to

o

o

v/>

O

O

©

>—

UJ

1>J

4*.

bo

so

'-

--U

J >

0 <

-J\

<0

oo

to

to

to

4*.

4*.

bo

to

>—

O

O

H-

H-

to

so

so

as I

—

Lfl

tO

~j

~o

-j

©

^-

©

o

o

o

^j

to

u*

to

to

4*.

to

o

O

o

•a

as

-o

VO

H

-o

o

to

to

to

4^

H-

o

•—

O

•-•

00

CO

O

O

o>

4^

tO

>—

.->

4>-

-O

©

-o

as

O

so

o

o

©

as

so

^-o

as

H-

as

o

oo

to

Ln

to

as

as

oo

so

o o

<J\

O

s O

s O

s

to

o

to

SO

to

l/<

as

*-

bo

to

to

to

V

o •o.

to

4^

U)

1—

'

to

4^

UJ

1—

to

*>.

4*.

4*.

to

*•

!

u>

' I—

1

to

4^

4*

to

4*.

to

o

to

4*.

4*

SO

to

4*.

Os

00

to

4*.

-o.

*o

to

4*.

-o

SO

to

4^

Os

SO

*-•

c "

1 C

D n

.

^ <

J\

oo

_^

c o 3 o

o

3 o

o

CD

O ff <—

C

D 3

2 s~^

pr1

CTQ

g"

u

-— O

o 3 r^

-<D

3 r^

- a 3_

r-f ^ n 00

3"

•o

w

w

0) o 3"

P

-1

P o a <-t

w'

r-f

o'

to

"0

3"

P

V3 fD

t—l ^

n

O

3 I-K

3 Tl

-i <T>

Vi

3"

T3 "i O

-a

a>

3.

S"

VI O

|-b

OO

n

n

3 c •^ fO

c«

C

(/

i a>

D-

,-t.

o fD

< c"

5i

CD

a H)

CD

o !-»•

o !-*5

o p

C/]

*

-•

3 era

•^ S- S"

o 3 _

P

«

-f

O -

P

1 b

| Oo

3 S i o a T

O 1

Appendix D: Chapter 8 Effect of placement charac. and formwork dim. on lateral pressure ofSCC

Table D.3 Fresh properties of SCC mixtures used to evaluate effect of waiting period

between successive lifts on formwork pressure (Phase III)

Mixtures Casting rate Slump flow T50 Concrete temp. Air content

(m/hr) (mm) (sec) (°C) (%)

SCC57A (cont.)

SCC57B (1 WT)

SCC57C (2 WT)

SCC59A (cont.)

SCC59B (1 WT)

SCC59C (2 WT)

22.6

~

22.8

18.6

20.6

18.9

590

600

600

640

640

640

--

2.78

4.35

4.37

4.35

5.21

2,344

2,367

2,367

2,403

2,364

2,403

1.7

2.1

1.5

2.8

2.9

2.8

Table D.4 Fresh properties for concretes used to evaluate effect of minimum formwork

dimension on lateral pressure characteristics (Phase IV)

Mixtures

SCC38

SCC62

CC35

DxL (mmxmm)

200x400

250x400

300x400

350x400

200x400

250x400

300x400

350x400

200x400

250x400

300x400

350x400

Slump flow (mm)

710

700

700

710

T50 (sec)

3.17

3.09

2.54

2.53

Cast with wall # 6

Cast with wall # 4

Cast with wall # 3

700

185*

215

220

—

0.52

N/A

N/A

N/A

N/A

Concrete temp.

(°C)

~

20

22

19.7

17.1

20

20.1

20.2

~

Air volume

(%)

1.2

1.5

1.9

1.2

3

2.4

2.8

1.9

--

Unit weight (kg/m3) (

2,367

2,379

2,356

2,349

J-Ring spread (mm)

glOrnin @40min

670

680

680

-

—

620

-

These properties are indicated in Table 10.2 (Chapter 10: field measurements)

2,340

2,343

2,365

2,360

~

685

N/A

N/A

N/A

N/A

580

N/A

N/A

N/A

N/A

slump

281

Appendix D: Chapter 8 Effect of placement charac. andformwork dim. on lateral pressure ofSCC

Appendix D2: Effect of casting rate on K0


0 5 10 15 20 25 30

Casting rate (m/hr) Fig. D.l Variations of K0 values with casting rate at heights of 1, 3, 7, and 10 m

[SCC40 of thixotropy level # 1]

SCC51 Thixotropy level # 2s X

o ©

X K0 @H = 1 m

O K0 m\ •=•- 3 in

- * - K 0 @ H = 7m

-«-K0frt)H = 10m

0 5 10 15 20 25 30

Casting rate (m/hr)

Fig. D.2 Variations of Ko values with casting rate at heights of 1, 3, 7, and 10 m


282

Appendix D: Chapter 8 Effect of placement charac. andformwork dim. on lateral pressure qfSCC

II

100

90

80

70

60

50

40

30

20

10

0


0

X 0

« •

X K0 @H = 1 m O K0 '(<;} I =•• 3 m

- • - K 0 (a)\\ == 7 m - » - K 0 @ H = 10m

25 30 5 10 15 20 Casting rate (m/hr)

Fig. D.3 Variations of K0 values with casting rate at heights of 1, 3, 7, and 10 m


0

^ 0 s

•a

.= fin

E

II

£

100

90

80

70

60

50

40

30

20

10

0 1


X

o X o

o ^--"*

- JW

If

1

X

1

X

. . . . „ - = - o - ^ -

X K0 @H = 1 m O "K0 ;tf:M = 3 m

- • - K 0 ici)\\ = 7 m - « - K 0 @ H = 10 m

X

o

1

25 30 5 10 15 20

Casting rate (m/hr)



283

Appendix D: Chapter 8 Effect of placement charac. andformwork dim. on lateral pressure o/SCC

•a • s

0.

« £

||

tf

100

90

80

70

60

50

40

30

20

10

0 1


X X

o X O

o f

J /

\

o

1

X K0 @H = 1 m O KO VYJH = 3 m

- • — K 0 @H = 7 m - • - K 0 ® H = 10 m

i i i

0 25 30 5 10 15 20

Casting rate (m/hr)



OH

II


0

X

o X

o

X

o

X K0 @H = 1 m •O'!K0to;H = 3 m - • - K 0 m\ = 7 m

•K0 m 10 m

25

X

o

30 5 10 15 20

Casting rate (m/hr)



284

Appendix D: Chapter 8 Effect of placement charac. andformwork dim. on lateral pressure qfSCC

T3

J=

ft-

X « g 0-II

rf

100

90

80

70

60

50

40

30

20

10

0 1 (


0 X

\

O yr

) 5

X X

o o

1 1

10 15

Casting rate (m/hr)

i

20

X

o

X K0@H = l m •O'K0'im = 3 m

— » - K.0 fffiH = 7 m

25

X

o

- •

1

30



285

Appendix D: Chapter 8 Effect of placement charge, andformwork dim, on lateral pressure qfSCC

SCC40 (Low thixotropy) SCC51

SCC52

SCC53

SCC54

SCC46

SCC55 SCC56(High thixotropy)

0 10 30

Fig.

15 20 25

Casting rate (m/hr)

D.8 Variations of K0 values @ H = 7 m with casting rate for SCC of different

thixotropy levels

10 20

™*L""SCC40 (Low thixotropy)

- * - S C C 5 1

lis* - * - S C C 5 2

«••>»-SCC53

SCC54

«— SCC46

-•— SCC55

-•—SCC-56 (High thixotropy)

30

Casting rate (m/hr)

Fig. D.9 Variations of Ko values @ H = 10 m with casting rate for SCC of different

thixotropy levels

286

Appendix D: Chapter 8 Effect of placement charac. andformwork dim. on lateral pressure of SCC

Appendix D3: Effect of thixotropy on Ko

Maximum lateral pressure ( kPa )

50 100 150 200 250

R = 5m/hr • - Phyd

Q s c c 4 0 ^ o w thixotropy)

-*—SCC51

-*— SCC52

-*— SCC53

- •— SCC54

-m— SCC46

-a— SCC55 N

N O SCC56 (High thixotropy)

Fig. D.10 Lateral pressure profiles for SCC mixtures of different thixotropy levels, cast at rate

of 5 m/hr


R = 10 m/hr - Phyd

•*— SCC40 (Low thixotropy)

•A— SCC51

* — SCC52

* — SCC53

• — SCC54

—i— SCC46

=<=—SCC55

SCC56 (High thixotropy)

Fig. D.l 1 Lateral pressure profiles for SCC mixtures of different thixotropy levels, cast at rate

of 10 m/hr

287

Appendix D: Chapter 8 Effect of placement charac. andformwork dim. on lateral pressure of SCC


R=17m/hr Phyd

— B — SCC40 (Low thixotropy)

—*— SCC51

—*— SCC52

—*— SCC53

— • — SCC54

— • — SCC46

—A— SCC55 S

x 0 SCC56 (High thixotropy)


of 17 m/hr


4>

u e o U

R - 2 4 m/hr

- Phyd

•—SCC40 (Low thixotropy)

±— SCC51

* — SCC52

* — SCC53

• — SCC54

H — SCC46

SCC55

SCC56 (High thixotropy)


of 24 m/hr

288

Appendix E: CHAPTER 9

PREDICTION MODELS FOR LATERAL PRESSURE CHARACTERISTICS

Appendix El: Abacuses for K0 prediction

289

App

endi

x E

: C

hapt

er 9

Pre

dict

ion

mod

els f

or l

ater

al p

ress

ure

char

acte

rist

ics

300

600

900

Abj

(J/

m3.s

ec)

1200

K()

@H

=lm

K

<)@

H=2

m

Ko@

H=

4m

K()

@H

=6m

Ko@

H=

8m

:H=1

0m

0@H

=12m

1500

0@H

=lm

0@

H=2

m

0@H

=4m

0@H

=6m

0@H

=8m

:H=1

0m

0@H

=12m

300

600

900

Ab!

(J/

m3.s

ec)

1200

15

00

100 80

5 60

40

20 0

hD

min

= 2

00m

m

MSA

= 1

4 m

m

WP

= 0

300

600

900

Abj

(J/

m3.s

ec)

100 80

60 h

J 40

20 0

Dm

m =

200

mm

M

SA =

14

mm

W

P =

0

K<)

@H

=lm

£>

0@H

=2m

Ko@

H=

6m

^0@

H=

8m

K()

@H

=10m

^0@

H=

12m

1200

15

00

^0@

H=

lm

K<)

@H

=2m

^0@

H=

4m

Ko@

H=6

m

Ko@

H=

8m

^O@

H=

10m

K<)

@H

=12m

300

600

900

Ab!

(J/

m3.s

ec)

1200

15

00

Fig

. E.l

Cor

rela

tion

s be

twee

n K

o an

d A

bi

290

App

endi

x E

: C

hapt

er

9 P

redi

ctio

n m

odel

s fo

r la

tera

l pr

essu

re

char

acte

rist

ics

100 80

? 5 60

40

20 0

MS

A =

14

mm

W

P =

0

20

40

Rh

eom

eter

T0

rest (

t) (

Pa/

min

)

K()

@H

=lm

^0@

H=

2m

^0@

H=

4m

^0@

H=

6m

K()

@H

=8m

K, 0@

H=1

0m

0@H

=12m

60

100

r

80

^6

0

^4

0 20

K(

K, 0@

H=l

m

0@H

=2m

0@H

=4m

hD

min

= 2

00m

m

MS

A=

14

mm

W

P =

0

0@H

=6m

Ko@

H=

8m

K,

K 0@

H=1

0m

0@H

=12m

20

40

60

Rh

eom

eter

T0

res,

(t)

(Pa/

min

)

100 80

60

40

20

R =

T

=

Dm

ra =

200

mm

M

SA =

14

mm

W

P =

0

20

40

Rh

eom

eter

x0

rest (

t) (

Pa/

min

)

K(

K,

K,

K, 0@

H=l

m

0@H

=2m

0@H

=4m

0@H

=6m

0@H

=8m

^0@

H=

10m

Ko(

ffiH

=12m

60

100 80

60

40

20

20

40

Rh

eom

eter

T0r

es,(

t) (

Pa/

min

)

K(

K

K,

K, 0@

H=l

m

0@H

=2m

0@

H=4

m

0@H

=6m

0@H

=8m

K, 0@

H=1

0m

0@H

=12m

60

Fig

. E

.2 C

orre

lati

ons

betw

een

K0

and

Rhe

omet

erco

rest

OO

291

App

endi

x E

: C

hapt

er 9

Pre

dict

ion

mod

els f

or l

ater

al p

ress

ure

char

acte

rist

ics

10

20

30

AlW

p®N

=o.7

rps(

t) (P

a-s/

min

)

100

r

-min

MS

A =

14

mm

W

P =

0

, 10

20

30

Ail a

pp@

N=o

.7rp

S(t) (

Pa.

s/m

in)

Ko@

H=

lm

Ko@

H=

2m

Kfl

@H

=4m

Ko@

H=

6m

^0@

H=

8m

K<>

@H

=10m

Ko@

H=

12m

40

50

Ko@

H=

lm

K()

@H

=2m

*N)@

H=4

m

Ko@

H=

6m

Kfl

@H

=8m

*CH

)@H

=10m

Ko@

H=

12m

40

50

100 80

^6

0

^40

20

K-0

@H

=lm

K()

@H

=2m

Ko@

H=

4m

Ko@

H=

6m

Ko@

H=

8m

100 80

^6

0 20 0

0@H

=10m

0@H

=12m

20

40

60

Alla

p P®

N=o

.7rp

S(t)

(Pa.

s/m

in)

K(

K,

K,

K, 0@

H=l

m

0@H

=2m

0@H

=4m

0@H

=6m

0@H

=8m

K<)

@H

=10m

^0@

H=

12m

10

20

30

Aila

PP@

N=o

.7rPS

(t) (P

a.s/

min

) 40

50

Fig.

E.3

Cor

rela

tion

s be

twee

n K

o an

d A

r|app

@N

=o.7r

ps(t)

292

App

endi

x E

: C

hapt

er 9

Pre

dict

ion

mod

els f

or l

ater

al p

ress

ure

char

acte

rist

ics

100

100

r

50

100

PV

T0r

est(

t)(P

a/m

in)

•0@

H=1

0m

0@H

=12m

150

50

100

PV

T0r

est(

t)(P

a/m

in)

0@H

=lm

0@H

=2m

0@H

=4m

0@H

=6m

0@H

=8m

0@H

=10m

0@H

=12m

150

50

100

PV

T0r

est(

t)(P

a/m

in)

0@H

=lm

0@H

=2m

0@H

=4m

K-o

@H

=6m

K<)

@H

=8m

K-0

@H

=10m

0@H

=12m

150

50

100

PV

T0r

est(

t)(P

a/m

in)

H=1

m

«=10

m

0@H

=12m

150

Fig

. E.4

Cor

rela

tions

bet

wee

n K

o an

d PV

iore

st(t)

293

App

endi

x E

: C

hapt

er 9

Pre

dict

ion

mod

els f

or l

ater

al p

ress

ure

char

acte

rist

ics

:~~~~

~~^^

^:=::

:^^~~

~~^_

' R

=2

m/h

r^~

—Z

^^~

-~--

^__

T =

22

°C

^-

^ -

Dm

m =

200

mm

M

SA

= 1

4 m

m

WP

= 0

^~

^^

^-

^ K

0@

H=

lm

_^"~

~-~

-~^^

^ K

. 0@

H=

2m

.^^~

~--

-^_

^^

K0@

H=

4m

~^^

~~

-—~

^_

K0@

H=

6m

^~~~

~~~-

*^

0@H

=10

m

*^0@

H=

12m

10

20

IPT

0re

st(t

)(P

a/m

in)

30

10

20

IPT

o res

t(t)

(Pa/

min

)

30

40

Ko@

H=

lm

Ko@

H=

2m

^0@

H=

4m

Ko@

H=

6m

K()

@H

=8m

40

Ko@

H=

lm

Ko@

H=

2m

K-o

@H

=4m

*H)@

H=8

m

Ko@

H=

10m

K()

(»H

=12m

10

20

IPT

0re

st(t

)(P

a/m

in)

30

40

100 80

V

60

£ 40

-

20

- R

=3

0m

/hr

T =

22

°C

Dm

m

= 2

00m

m

" M

SA

=1

4 m

m

WP

= 0

"~

"~

~"

~-^

^-^

^^

^K)@

H=l

m

~~""

~--—

_^/~

""""

""--

K

o@H

=2m

^ -

K<)

@H

=4m

^^

~~

-~

^^

~ K

o@H

=6m

~-

^^

-~

-^

_^

^ *H

)@H

=8m

^"""

~~

~-~

^ ^0

@H

=10

m

K()

@H

=12m

i >

10

20

IPT

0re

st(t

)(P

a/m

in)

30

40

Fig

. E.5

Cor

rela

tion

s be

twee

n K

o an

d IP

Torest

(t)

294

App

endi

x E

: C

hapt

er 9

Pre

dict

ion

mod

els f

or l

ater

al p

ress

ure

char

acte

rist

ics

100

r

0@H

=2m

0@H

=4m

0@H

=6m

Ko@

H=

8m

K,

K M)@

H=1

0m

0@H

=12m

2000

0 40

000

6000

0 80

000

1000

00

1200

00

Rhe

omet

er x

0 re

s,@is

min

XTo

rest

(t)

(Pa2/m

in) K

-0@

H=l

m

^0@

H=

2m

Ko@

H=

4m

Ko@

H=

6m

0@H

=8m

0@H

=10m

0@H

=12m

100 80

60 V

^4

0 20 0

100 80

60 V

40

20 0

K,

K(

K,

K, 0@

H=l

m

0@H

=2m

0@H

=4m

0@H

=6m

0@H

=8m

WP

= 0

K

o@H

=10

m

Kp@

H=l

,2m

2000

0 40

000

6000

0 80

000

1000

00

1200

00

Rhe

omet

er x

0 re

st®i5

min

xT0 r

est (

*) (

Pa2/m

in) K

o@H

=lm

Ko@

H=

2m

^0@

H=

4m

K<)

@H

=8m

Ko@

H=

10m

^0@

H=

12m

0 20

000

4000

0 60

000

8000

0 10

0000

12

0000

Rhe

omet

er T

0 re

st(a

15m

i nX

T0

rest (t

) (P

a2/min

)

Fig

. E.6

Cor

rela

tion

s be

twee

n K

0 an

d R

heom

eter

Tores

t@i5

min

xTore

st(t)

2000

0 40

000

Rhe

omet

er T

. 60

000

8000

0 10

0000

12

0000

(t)

(Pa2/m

in)

0 re

st@

15m

inX

T0

rest

295

App

endi

x E

: C

hapt

er 9

Pre

dict

ion

mod

els f

or l

ater

al p

ress

ure

char

acte

rist

ics

100

r

MS

A

WP

= 0

K0l

K,

K

H=

lm

0@H

=2m

0@H

=4m

0@H

=6m

0@H

=8m

K, •0

@H

=10m

0@H

=12m

5000

10000

15000

20000

^tlapp®N=0.7rps®15minX^1lapp(ffi

yN=0.7rps(t) (Pa.S /

min)

K,

K K

K,

0@H

=lm

0@H

=2m

0@H

=4m

0@H

=6m

0@H

=8m

WP

= 0

K, 0@

H=1

0m

0@H

=12m

100 80

^6

0

^4

0 20 0

100

r

80

^ 60

£ 40

20 0

K,

K,

K,

K, 0@

H=1

m

0@H

=2m

0@

H=4

m

0@H

=6m

0@

H=8

m

mm

M

SA

WP

= 0

K

, 0@H

=10m

[email protected]

=13m

5000

10000

15000

20000

^1la

pp

®N

=0.

7rp

s®15

min

><^

1lap

p®

N=

0.7r

ps(

0 (P

a-S

/m

in)

K, 0@

H=l

m

0@H

=2m

0@H

=4m

Dm

m =

200

mm

M

SA

= 1

4 m

m

WP

= 0

0@H

=6m

Ko@

H=

8m

Ko@

H=

10m

Ko@

H=

12m

0 50

00

1000

0 15

000

2000

0

.7rp

s@15

min

xAlla

pp@

N=o

.7rp

S(t)

(Pa.

s2/min

) 0

5000

10

000

1500

0 20

000

Atla

pp

@N

=0.

7rps

@15

min

xAlla

pp@

N=o

.7rp

S(t)

(Pa.

s2/min

)

Fig

. E.7

Cor

rela

tion

s be

twee

n K

o an

d A

r| app@

N=o

.7rp

s@15

min

xA

T|a

pp@

N=

0.7r

ps(t

)

296

App

endi

x E

: C

hapt

er 9

Pre

dict

ion

mod

els f

or l

ater

al p

ress

ure

char

acte

rist

ics

H=l

m

0@H

=4m

0@H

=6m

0@H

=8m

0@H

=10m

0@H

=12m

100 80

0@H

=2m

«J

, ^

60

^ 40

20

Ko@

H=

lm

K()

@H

=2m

K<)

@H

=8m

K-0

@H

=10m

^•0(

ffiH

=12m

5000

0 10

0000

15

0000

20

0000

25

0000

30

0000

P

VT

O re

st@ism

inxPV

T 0 r£

St (t

) (P

a2/min

)

0@H

=lm

0@H

=2m

0@H

=4m

0@H

=6m

Ko@

H=8

m

0@H

=10m

•0@

H=1

2m

50000

100000 150000 200000 250000 300000

100 r

80 r

60

40

20

V

0

50000

100000 150000 200000 250000 300000

PV

T0

rest

®15

min

XPV

T0

res

t (t

) (P

a2/m

in)

0@H

=lm

0@H

=2m

0@H

=4m

0@H

=6m

^0@

H=

8m

K, v0@

H=1

0m

0@H

=12m

0

PV

T0

rest

@15

min

xPV

T0r

est(

t)(P

a2/min

)

5000

0 10

0000

PV

T0

Fig

. E.8

Cor

rela

tion

s be

twee

n K

0 an

d PV

x 0 re

st@

15m

in

150000 200000 250000 300000

rest

@1

5m

inX

PV

T0

rest

(t)

(Pa

2/m

in)

297

App

endi

x E

: C

hapt

er

9 P

redi

ctio

n m

odel

s fo

r la

tera

l pr

essu

re

char

acte

rist

ics

K()

@H

=lm

£o

@H

=2m

Ko@

H=6

m

6H=8

m

K 0@

H=1

0m

0@H

=12m

0@H

=lm

0@

H=2

m

0@H

=4m

0@

H=6

m

2000

0

IPT

, 0 re

st@

15m

in

40000

XPVT

60000

0rest(t)(Pa2/min)

80000

0®H=lm

20000

IPT0

40000

60000

80000

rest

@15

min

xP

VT

0re

st(t

)(P

a2/m

in)

100 80

60

f-

^4

0 20

0@H

=lm

0@

H=2

m

0(ffi

H=4

m

0@H

=6m

0@H

=8m

0@H

=10m

0@H

=12m

2000

0 40

000

6000

0 80

000

IPT

0 re

s,®i 5

mi„

x PV

T0

rest (

t) (

Pa2 /m

in)

2000

0 40

000

6000

0

Fig

. E

.9 C

orre

lati

ons

betw

een

K0

and

IPT

o res

t@i5

min

xPVxo

rest

(t)

8000

0

IPT

0 re

st®

ism

inxP

VT

0 re

st (

t) (

Pa2/m

in)

298

REFERENCES

ACI Committee 347, "Guide to Formwork for Concrete," American Concrete Institute, Farmington Hills, Michigan, 2001, 32 pp.

ACI Committee 347R-03, "Guide to Formwork for Concrete," American Concrete Institute, Farmington Hills, Michigan, 2004, 32 pp.

ACI Committee 622, "Pressures on Formwork," ACI Journal Proceedings, V. 55, No. 2, August 1958, pp. 173-190.

Adam, M.; Bennasr, M.; and Santos Delgado, H., "Formwork Pressure of Fresh Concrete," (in French), Annales de l'lnstitut Technique du Batiment et des Travaux Publics, Serie Beton, March-April 1963, No. 207-208, pp. 403-423.

ATtcin, P.-C, "Autogenous Shrinkage Measurement," Proceedings, International Workshop on Autogenous Shrinkage of Concrete, Hiroshima, Japan, Ed. Tazawa, E., 1999, pp. 257-268.

Alexandridis, A., and Gardner, N. J., "Mechanical Behavior of Fresh Concrete," Cement and Concrete Research, V. 11, 1981, pp. 323-339.

Amziane, S., and Baudeau, P., "Effects of Aggregate Concentration and Size in Fresh Concrete Pressure on Formwork Walls," (in French), Materials and Structure, V. 33, No. 225, 2000, pp. 50-58.

Andreas, L., and Cathleen, H., "Pressure of Self-Compacting Concrete on the Formwork," Proceedings of the 3r International RILEM Symposium on Self-Compacting Concrete, Eds. Wallevik, O., and Nielsson, I., Reykjavik, Iceland, August 2003, pp. 288-295.

Andreas, L.; Cathleen, H.; and Frank W., "Influence of the Mixture Design on the Formwork Pressure of Self-Compacting Concrete," Proceedings of the 2" North American Conference on the Design and Use of Self-Consolidating Concrete (SCC) and the 4th International RILEM Symposium on Self Compacting Concrete, Eds. Shah, S.P., Chicago, 2005, pp. 635-640.

Andriamanantsilavo, N. R., and Amziane S., "Maturation of Fresh Cement Paste Within 1- to 10-m-large Formworks," Cement and Concrete Research, Vol. 34, No. 11, November 2004, pp. 2141-2152.

Arslan, M.; Osman, S.; and Serkan, S., "Effects of Formwork Surface Materials on Concrete Lateral Pressure," Construction and Building Materials, V. 19, No. 4, May 2005, pp. 319-325.

Assaad, J., "Formwork Pressure of Self-Consolidating Concrete - Influence of Thixotropy," Ph.D. Thesis, Department of Civil Engineering, Universite de Sherbrooke, 2004, 453 pp.

Assaad, J., and Khayat, K. H., "Effect of Coarse Aggregate Characteristics on Lateral Pressure Exerted by Self-Consolidating Concrete," ACI Materials Journal, V. 102, No. 3, 2005C,pp. 145-153.

Assaad, J., and Khayat, K. H., "Effect of Viscosity-Enhancing Admixtures on Formwork Pressure and Thixotropy of Self-Consolidating Concrete," ACI Materials Journal, V. 103, No. 4, 2006, pp. 280-287.

Assaad, J., and Khayat, K. H., "Formwork Pressure of Self-Consolidating Concrete Made with Various Binder Types and Contents," ACI Materials Journal, V. 102, No. 4, 2005 A, pp. 215-223.

299

References

Assaad, J., and Khayat, K. H., "Kinetics of Formwork Pressure Drop of SCC Containing Various Types and Contents of Binder," Cement and Concrete Research, V. 35, 2005B, pp. 1522-1530.

Assaad, J., and Khayat, K. H., "Variations of Lateral and Pore Water Pressure of Self-Consolidating Concrete at Early Age," ACI Materials Journal, V. 101, No. 4, 2004, pp. 310-317.

Assaad, J.; Khayat, K. H.; and Mesbah, H., "Assessment of Thixotropy of Flowable and Self-Consolidating Concrete," ACI Materials Journal, V. 100, No. 2, 2003A, pp. 111-120.

Assaad, J.; Khayat, K. H.; and Mesbah, H., "Influence of Thixotropy on Variations of Formwork Pressure of Flowable and Self-Consolidating Concrete - Laboratory Tests," ACI Materials Journal, V. 100, No. 1, 2003B, pp. 29-37.

Banfill, P. F. G., and Saunders, D. C , "On the Viscometric Examination of Cement Pastes," Cement and Concrete Research, V. 11, No. N-3, 1981, pp. 363-370.

Barnes H. A., and Nguyen Q. D., "Rotating Vane Rheometry-A Review," Journal of Non-Newtonian Fluid Mechanics, 98, 2001, pp. 1-14.

Barnes, H. A., "Thixotropy - A Review," Journal of Non-Newtonian Fluid Mechanics, V. 70, Issues 1-2, May 1997, pp. 1-33.

Barnes, H. A.; Hutton, J. F.; and Walters, K., "An Introduction to Rheology," Elsevier, Amsterdam, 1989, 199 pp.

Bauer, E., G. G. de Sousa, J., Guimaraes, E. A., Silva, F. G. S., "Study of the Laboratory Vane Test on Mortars," Building and Environment, 42, 2007, pp. 86-92.

Beaupre, D., "The Rheology of High-Performance Shotcrete," Doctoral Thesis, University of British Columbia, Feb. 1994, 250 p.

Beitzel, M.; Muller; and Harald, S., "Modeling Fresh Concrete Pressure on Vertical Formwork," CD ROM, Proceedings of the 5' International Symposium in Civil Engineering, Taylor & Francis Group, London, 2004, pp 1-6.

Billberg, P.; Silfwerbrand, J.; and Holmgren, J. "SCC Structural Behaviour at Rest and its Influence on Form Pressure," Submitted to RILEMMaterials and Structures Journal, 2006.

Billberg, P., "Form Pressure Generated by Self-Compacting Concrete - Influence of Thixotropy and Structural Behaviour at Rest," Ph.D. Thesis, School of Architecture and the Built Environment, Division of Concrete Structures, Royal Institute of Technology, Stockholm, Sweden, 2006, 91 pp.

Billberg, P., "Development of SCC Static Yield Stress at Rest and its Effect on the Lateral Form Pressure," in: S.P. Shah (Ed.), Proceedings of the Second North American Conference on the Design and use of Self-Consolidating Concrete and the Fourth International RILEM Symposium on Self-Compacting Concrete, Oct. 30 - Nov. 3 2005. Chicago, USA, Chicago: Hanley Wood, LLC, ISBN: 0924659645, 2005.

Billberg, P., "Form Pressure Generated by Self-Compacting Concrete," Proceedings of the 3rd

International RILEM Symposium on Self-Compacting Concrete, Eds. Wallevik, O., and Nielsson, I., Reykjavik, Iceland, August 2003, pp. 271-280.

Brameshuber, W., and Uebachs, S., "Investigations of the Formwork Pressure Using Self-Compacting Concrete," Proceedings of the 3rd International RILEM Symposium on Self-Compacting Concrete, Eds. Wallevik, O., and Nielsson, I., Reykjavik, Iceland, August 2003, pp. 281-287.

300

References

Bonen, D., and Shah, S. P., "Fresh and Hardened Properties of Self-Consolidating Concrete," Progress in Structural Engineering Materials Journal, V. 7, No. 1, 2005, pp.14-26.

Bonen, D., and Shah, S. P., "The Effects of Formulation on the Properties of Self-Consolidating Concrete," Concrete Science and Engineering, A Tribute to A. Bentur, International RILEM Symposium, Eds. Kovler, K.; Marchand, J.; Mindess, S.; and Weiss, J., RILEM Publications S.A.R.L. Evanston, IL, March 2004, pp. 43-56.

CEBTP, "Performance of SCC in High Wall Sections," (in French), Internal Report, 1999, 12 pp.

Civil Industries Research and Information Association (CIRIA), "Concrete Pressure on Formwork," Research Report No. 108, London, 1985.

Civil Industries Research and Information Association (CIRIA), "The Pressure of Concrete on Formwork," Research Report No. 1, London, 1965.

Clayton, S.; Grice, T.G.; and Boger, D.V., "Analysis of the Slump Test for On-Site Yield Stress Measurement of Mineral Suspensions," Int. J. Min. Process. 70, 2003, pp. 3-21.

Coussot, P.; Proust S.; and Ancey, C , "Rheological Interpretation of Deposits of Yield Stress Fluids," J. Non-Newtonian FluidMech, 66, 1996, pp. 55-70.

DIN 18218, "Frishbeton auf lautrechte" (Pressure of Fresh Concrete on Vertical Formwork), (only available in German), Berlin, 1980.

Djelal, C , "Designing and Perfecting a Tribometer for the Study of Friction of a Concentrated Clay-Water Mixture against a Metallic Surface," Materials and Structures, V. 34, No. 1, 2001, pp. 51-58.

Domone, P., "The Slump Flow Test for High-Workability Concrete," Cement and Concrete Research, 28, 1998, pp. 177-182.

Douglas, R.; Sun, Z.; Bonen, D.; and Shah, S.P., "The Effect of Ingredients and Shear History on the Thixotropic Rate of the Rebuilding of SCC," Proceedings of the 2nd North American Conference on the Design and Use of Self-Consolidating Concrete (SCC 2005) and the 4th

International RILEM Symposium on Self-Compacting Concrete, Evanston, IL, Ed. Shah, S. P., 2005, pp. 591-596.

Dzuy, N. Q., and Boger, D. V., "Direct Yield Stress Measurement with the Vane Method," Journal ofRheology, V. 29, No. 3, 1985, pp. 335-347.

Elaguab, Y., "Pressions Laterales Developpees par les Betons Autoplacants Sur les Coffrages," in French, master thesis at Universite de Sherbrooke, 2008.

Fedroff, D.; Frosch; and Robert, J., "Formwork for Self-Consolidating Concrete," Concrete International, V. 26, No. 10, 2004, pp. 32-37.

Ferron, R.; Gregori, A.; Sun, Z.; and Shah, S. P., "Rheological Method to Evaluate the Thixotropy of Cement Pastes for SCC," ACI Materials Journal, 2006, (in press).

Fossa, K. T., "Slipforming of Vertical Concrete Structures," Ph.D. Thesis, Norwegian University of Science and Technology, 2001, 285 pp.

Gardner, N. J., "Formwork Pressures and Cement Replacement by Fly Ash," Concrete International, October 1984, pp. 50-55.

Gardner, N. J., "Pressure of Concrete Against Formwork", ACI Journal, Technical Paper, Title No. 77-31, 1980, pp. 279-286.

301

References

Gardner, N. J., "The Effect of Superplasticizers and Fly Ash on Formwork Pressures," Forming Economical Concrete Buildings, Portland Cement Association, Skokie, IL, 1982, pp. 21.1-21.12.

Gardner, N. J., and Ho, P. T.-J., "Lateral Pressure of Fresh Concrete," Adjournal, Technical Paper, Title No. 76-35, July 1979, pp. 809-820.

Ghezal, A.; Khayat, K. H.; and Beaupre, D., "Effect of High-Range Water-Reducer -Viscosity-Modifying Admixture Combination on Rheological Properties of Concrete Equivalent Mortar," Proceedings of the 1st North American Conference on the Design and Use of Self-Consolidating Concrete, Eds. Shah, S. P.; Daczko, J. A.; and Lingscheit, J. N., Chicago, November 2002, pp. 159-165.

Graubner, C.-A., and Proske, T., "Formwork Pressure: A New Concept for the Calculation," Proceedings of the 2nd North American conference on the Design and Use of Self-Consolidating Concrete (SCC 2005) and the 4th International PJLEM Symposium on Self-Compacting Concrete, Eds. Shah, S. P., Chicago, 2005A, pp. 605-613.

Graubner, C.-A., and Proske, T., "Formwork Pressure: A New Concept for the Calculation," personal communication to Khayat, K. H., 2005B.

Hobbs D. W., "Influence of Aggregate Volume Concentration upon the Workability of Concrete and some Predictions from the Viscosity-Elasticity Analogy," Magazine of Concrete Research, V. 28, No. 97, 1976, pp. 191-202.

Hurd, M. K., "Putting the Pressure on Formwork," Concrete International, October 2002, pp. 49-55.

Ish-Shalom, M., and Greenberg, S. A., "The Rheology of Fresh Portland Cement Paste," Proceedings of the 4th International Symposium on Chemistry on Cement, Washington, 1962, pp. 731-748.

Janssen H., "Versuche uber Getreidedruck in Silozellen," VDI Zeitschrift, V. 39, 1885, pp. 1045-1049.

Jiang, W., and Roy, D. M., "Rheology in Hydration and Setting," Proceedings of the International RILEM Workshop on Hydration and Setting of Cements, Dijon, France, 1991, pp.333-340.

Khayat, K.H., "Viscosity-Enhancing Admixtures for Cement-Based Materials - An Overview," Cement and Concrete Composites, V. 20, 1998, pp. 171-188.

Khayat K. H.; Assaad, J.; Mesbah, H., and Lessard, M., "Effect of Section Width and Casting Rate on Variations of Formwork Pressure of Self-Consolidating Concrete," Materials and Structures, V. 38, 2005A, pp. 73-78.

Khayat, K.H., and Assaad, J.J., "Measurement Systems for Determining Formwork Pressure of Highly Flowable Concrete," Materials and Structures, V. 41, No. 1 / Jan. 2008.

Khayat, K. H., and Yahia, A., "Modification of the Concrete Rheometer to Determine Rheological Parameters of Self-Consolidating Concrete-Vane Device", 2nd International Symposium on Advances in Concrete through Science and Engineering , 11-13 Sept. 2006, Quebec City, Canada.

Khayat, K. H., and Assaad, J., "Effect of w/cm and High-Range Water-Reducing Admixture on Formwork Pressure and Thixotropy of Self-Consolidating Concrete," ACI Materials Journal, V. 103, No. 3, 2006, pp. 186-193.

302

References

Khayat, K. H., and Assaad, J., "Influence of Internal Friction and Cohesion on Formwork Pressure of Self-consolidating Concrete," Proceedings of the 1st International Symposium on Design, Performance and Use of Self-Consolidating Concrete (SCC 2005), Changsha, Hunan, China, May 26-28, 2005B, Ed. Yu, Z.; Shi, C ; Khayat, K. H.; and Xie, Y., pp. 607-615.

Khayat, K. H., and Assaad, J., "Use of Rheological Properties of SCC to Predict Formwork Pressure," Proceedings of the 2" North American Conference on the Design and Use of Self-Consolidating Concrete (SCC 2005), and the 4' International RILEM Symposium on Self-Compacting Concrete, Evanston, IL, 2005A, Ed. Shah, S. P., pp. 671-677.

Khayat, K.H., Omran, A., Neji, S., Billberg, P., and Yahia, A., "Test Methods to Evaluate Form Pressure of SCC," Proceedings of the international conference of ACBM on SCC (SCC 2008), Chicago, USA, Nov. 2008.

Khayat, K. H.; Petrov, N.; Assaad, J.; Morin, R.; and Thibeault, M., Performance of Self-Consolidating Concrete in Repair of Concrete Wall Elements" Proceedings of the 2nd North American Conference on the Design and Use of Self-Consolidating Concrete (SCC 2005), and the 4th International RILEM Symposium on Self-Compacting Concrete, Eds. Shah, S. P., Chicago, 2005B, pp. 1003-1012.

Khayat, K. H.; Assaad, J.; and Mesbah, FL, "Variations of Formwork Pressure of Self-Consolidating Concrete-Effect of Section Width and Casting Rate," Proceedings of the Ist

North American Conference on the Design and Use of Self-Consolidating Concrete, Eds. Shah, S. P.; Daczko, J. A.; and Lingscheit, J. N., Chicago, November 2002B, pp. 295-302.

Khayat, K. H; Saric-Coric, M.; and Liotta, F., "Assessment of Thixotropy and Impact on Stability of Cementitious Grout and Concrete," ACI Materials Journal, V. 99, No. 3, 2002A, pp. 234-241.

Lapasin, R.; Papo, A.; and Rajgelj, S., "Flow Behavior of Fresh Cement Pastes. A Comparison of Different Rheological Instruments and Techniques," Cement and Concrete Research, V. 13, 1983, pp. 349-356.

Leemann, A., and Hoffmann, C , "Pressure of Self-Compacting Concrete on the Formwork," Betonwerk und Fertigteil-Technik / Concrete Plant and Precast Technology, V. 69, No. 11, November 2003, pp. 48-55.

Legrand, C , "Contribution to Study of the Rheology of Fresh Concrete" (in French), Ph.D. Thesis, Universite Paul Sabatier, Toulouse, France, 1971, 150 pp.

Macklin, C , "Pressure of Plastic Concrete in Forms," USA, Proceedings of the Society for Experimental Stress Analysis, V. 4, No. 1, 1946, pp. 112-122.

Mehta, P.K., and Monteiro, P.J.M., "Concrete : microstructure, properties, and materials," New York : McGraw-Hill, c2006, 3rd ed, 659 pp.

Mewis, J., "Thixotropy - a General Review," Journal of Non-Newtonian Fluid Mechanics, V. 6, No. 1, 1979, pp. 1-20.

Moller, P. C. F.; Mewis, J.; and Bonn, D., "Yield Stress and Thixotropy: on the Difficulty of Measuring Yield Stress in Practice," Soft Matter, 2, 2006, pp. 274-283.

Murata, J., "Flow and Deformation of Fresh Concrete," Mater. Constr. (Paris), 17, 1984, pp. 117-129.

NF P93-350, French Standard, "Bandies industrialisees pour ouvrages en beton, (Industrial Formwork for Concrete Structures)," June 1995.

303

References

Nguyen N. Q., and Boger D. V., "Direct Yield Stress Measurement with the Vane Method," Journal ofRheology, 29(3), 1985, pp. 335-47.

Oremus, and Richard M, "A One Dimensional Model of Dense Snow Avalanches Using Mass and Momentum Balances" A thesis, presented to the faculty of Humboldt State University, May 2006.

Ovarlez, G.; Fond, C ; and Clement, E., "Overshoot Effect in the Janssen Granular Column: A Crucial Test for Granular Mechanics," Physics Review, V. 67, No. 6, June 2003, pp. 060302-1 to 060302-4.

Ozawa, K.; Maekawa, K.; Kunishima, M.; and Okamura, H., "Development of High Performance Concrete Based on the Durability Design of Concrete Structures," Proceedings of the 2" East-Asia and Pacific Conference on Structural Engineering and Construction (EASEC-2), Chiang Mai, Thailand, V. 1, January 1989, pp. 445-450.

Pashias, N.; Boger, D.V.; Summers, J.; and Glenister, D.J., "A Fifty-Cent Rheometer for Yield Stress Measurements," J. Rheol, 40, 1996, pp. 1179-1189.

Proske, T., and Graubner, C.-A., "Self-Compacting Concrete - Pressure on Formwork and Abiliy to Deaerate," Concrete Structures, V. 17, Darmastadt, 2002, (http://www.darmstadt-concrete.de/2002/dearate.html).

Radocea, A., "A Model of Plastic Shrinkage," Magazine of Concrete Research, V. 46, No. 167, 1994, pp. 125-132.

Ritchie, A. G. B., "The Triaxial Testing of Fresh Concrete," Magazine of Concrete Research, V. 14, No. 40, 1962, pp. 1027-1030.

RMC-SDC, Appendix of Task VI in the final report, "SCC Formwork pressure," submitted to the Ready-Mix Concrete Research Foundation, and American Concrete Institute - Concrete Research and Education Foundation, May 2009.

Roby, J.; Khayat, K.; and Yahia, A., "Developpement d'un Essai Faisable en Chantier pour Quantifier la Thixotropie," Report in French for the trainee period submitted to Prof. Kama! H. Khayat, Universite de Sherbrooke, December 2006, 36 pp.

Roby, H. G., "Pressure of Concrete on Forms," Civil Engineering, V. 5, March, 1935, 162 pp.

Rodin, S., "Pressure of Concrete on Formwork," Proceedings, Institute of Civil Engineers (London), V. 1, Part 1, No. 6, November 1952, pp. 709-746.

Roussel, N., "Correlation between Yield Stress and Slump: Comparison between Numerical Simulations and Concrete Rheometers Results," Materials and Structures, 39, 2006, pp. 501-509.

Roussel, N., and Cussigh F., "Distinct-layer Casting of SCC: The Mechanical onsequences of Thixotropy," Cement and Concrete Research, 38, 2008, pp. 624-632.

Roussel N., and Ovarlez G., "A Physical Model for the Prediction of Pressure Profiles in a Formwork," Proceedings of the 2" North American conference on the Design and Use of Self-Consolidating Concrete (SCC 2005) and the 4th International RILEM Symposium on Self-Compacting Concrete, Eds. Shah, S. P., Chicago 2005, pp. 647-654.

Roussel, N., and Coussot, P., "Fifty-cent Rheometer" for Yield Stress Measurements: From Slump to Spreading Flow," Journal ofRheology, V. 49, N. 3, 2005, pp 705-718.

304

http://www.darmstadt-

http://concrete.de/2002/dearate.html

References

Saak, A., "Characterization and Modeling of Rheology of Cement Paste: with Applications toward Self-Flowing Materials," Material Science and Engineering, Northwestern University: Evanston, 2000, 249 pp.

Saak, A.W.; Jennings, H.M.; and Shah, S.P., "A Generalized Approach for the Determination of Yield Stress by Slump and Slump Flow," Cem. Concr. Res., 34, 2004, pp. 363-371.

Saak, A. W.; Jennings, H. M.; and Shah, S. P., "New Methodology for Designing Self-Compacting Concrete," ACIMaterials Journal, V. 98, No. 6, 2001, pp. 429-439.

Schojdt R., "Calculations of Pressure of Concrete on Forms," Proceedings of the American Society of Civil Engineers, V. 81, May 1955, pp. 1-16.

Schowalter, W.R., and Christensen, G., "Toward a Rationalization of the Slump Test for Fresh Concrete: Comparisons of Calculations and Experiments," J. Rheol, 42, 1998, pp. 865-870.

Seed R. B., and Riemer, M., "Advanced Soil Mechanics Laboratory," course instructed at University of California at Berkeley, http://www.geoengineer.org.

Shaughnessy, R., and Clark, P. E., "The Rheological Behavior of Fresh Cement Pastes," Cement and Concrete Research, V. 18, 1988, pp. 327-341.

Skarendahl, A., "Self-Compacting Concrete for Improved Productivity, Working Environment and Performance," IREX-Meeting, Paris, February 1999, 12 pp.

Stanton, T. E., "Measured Pressure on Forms from Fresh Concrete," Concrete, V. 45, 1937, pp. 11-16.

Struble, L. J., "The Rheology of Fresh Cement Paste," Proceedings, Conference of the American Ceramic Society, Boston, 1991, pp. 7-29.

Tattersall G. H., and Bloomer S. J., "Further Development of the Two-Point Test for Workability and Extension of its Range," Magazine of Concrete Research, V. 31, No. 109, 1979, pp. 202-210.

Tattersall, G. H., and Banfill, P. F. G., "The Rheology of Fresh Concrete," Pitman Advanced Publishing Program, London, First Edition, 1983, 356 pp.

Tejeda-Dominguez, F., and Lange, D. A., "Effect of Formwork Material on Laboratory Measurements of SCC Formwork Pressure," Proceedings of the 2nd North American conference on the Design and Use of Self-Consolidating Concrete (SCC 2005) and the 4 International RILEM Symposium on Self-Compacting Concrete, Evanston, IL, Ed. Shah, S. P., 2005, pp. 525-532.

Tejeda-Dominguez, F.; Lange, D. A.; and D'Ambrosia, M. D., "Formwork Pressure of Self-Consolidating Concrete (SCC) in Tall Wall Field Applications," 84th Annual Meeting Transportation Research Board, Compendium of Papers CD-ROM, Washington, DC, January 2005.

Terzaghi, K., and Peck, R. B., "Soil Mechanics in Engineering Practice," John Wiley & Sons, Inc., New York, 1967,729 pp.

Vanhove Y.; Djelal C ; and Magnin, A., "A Prediction of the Pressure on Formwork by Tribometry," Pressure Vessel and Piping Conference, Emerging Technologies for Fluids, Structures and Fluid-Structure Interactions - 2001, The American Society of Mechanical Engineers, Atlanta, July 2001, V. 431, pp. 103-110.

305

http://www.geoengineer.org

References

Vanhove, Y.; Djelal, C ; and Magnin, A., "A Device for Studying Fresh Concrete Friction," Cement Concrete and Aggregates, V. 26, No. 2, December 2004A, pp. 35-41.

Vanhove, Y.; Djelal, C ; and Magnin, A., "Prediction of the Lateral Pressure Exerted by Self-Compacting Concrete on Formwork," Magazine of Concrete Research, V. 56, No. 1, February 2004B, pp. 55-62.

Vie, D.; Djelal, C ; Vanhove, Y.; and Magnin, A., "Pressure of Self-Compacting Concretes," (in French), Annales du Batiment et des Travaux Publics, No. 6, December 2001, pp. 5-13.

Wallevik, J., "Rheology of Particle Suspensions," Ph.D. Thesis, Department of Structural Engineering, Norwegian University of Science and Technology: Trondheim, Norway, 2003, 397 pp.

Wallevik, O. H., "Rheology - a Scientific Approach to Develop Self-Compacting Concrete," 3rd International RILEM Symposium on Self-Compacting Concrete, Eds. Wallevik, O., and Nielsson, I., Reykjavik, Iceland, August 2003, pp. 23-31.

Wolfgang B., and Stephan U., "Investigation on the Formwork Pressure Using Self-Compacting Concrete," Proceedings of the 3r International RILEM Symposium on Self-Compacting Concrete, Eds. Wallevik, O., and Nielsson, I., Reykjavik, Iceland, August 2003, pp. 281-287.

306

LIST OF FIGURES

Chapter 1 Fig. 1.1 Structure of the thesis 7

Chapter 2 Fig. 2.1 Concrete pressure distribution on formwork [Rodin, 1952] 11 Fig. 2.2 Formwork pressure - DIN 18218 (D), CIRIA 108 (GB), and NF P93-350 (F) [Proske and

Graubner, 2002] 16 Fig. 2.3 Schematic representation of stress in a formwork system [Vanhove et al., 2001] 17 Fig. 2.4 Sketch for the formwork wall 19 Fig. 2.5 Comparison between calculated shear stress and measured yield shear stress in terms of

resting time [Roussel and Ovarlez, 2005] 22 Fig. 2.6 Experimental column and sketch for pressure calculations [Graubner and Proske, 2005B] 24 Fig. 2.7 Pressure distribution [Graubner and Proske, 2005A] 24 Fig. 2.8 Testing machine used to determine model parameters [Graubner and Proske, 2005A] 25 Fig. 2.9 Pressure ratio X and friction coefficient u for the calculation of formwork pressure [Graubner

and Proske, 2005A] 25 Fig. 2.10 Calculated maximum pressure using Graubner and Proske's model [2005A] 26 Fig. 2.11 Comparison between the calculated data using Graubner and Proske's model and the

measured data - Influence of reinforcement [Graubner and Proske, 2005B] 26 Fig. 2.12 Breakdown area (Ab) vs. relative lateral pressure measured initially, and after 100 and 200

min [Khayat and Assaad, 2005A] 29 Fig. 2.13 Relationship between breakdown areas determined at various time intervals [Khayat and

Assaad, 2005A] 29 Fig. 2.14 Drop in apparent viscosity vs. initial lateral pressure [Khayat and Assaad, 2005A] 30 Fig. 2.15 Effect of concrete head on variations of K0 values for mixtures having various degrees of

breakdown areas [Khayat and Assaad, 2005A] 30 Fig. 2.16 Relationship between predicted and measured K [Khayat and Assaad, 2005A] 31 Fig. 2.17 Breakdown and build-up of a 3-D thixotropic structure [Barnes, 1997] 32 Fig. 2.18 Variation of viscosity with time for VMA concrete following 2 and 4 min of rest, [Assaad,

2004] 34 Fig. 2.19 Tattersall MK-III concrete rheometer in its commercial version (left) and with the modified

vane (right) 36 Fig. 2.20 Geometry of vane used to measure the rheology of concrete 36 Fig. 2.21 Bingham model 38 Fig. 2.22 Calculation method for the parameter (n) 38 Fig. 2.23 Hysteresis loop flow curve [Ish-Shalom and Greenberg, 1962] 39 Fig. 2.24 Effect of superplasticizer content and rest time on thixotropy and hysteresis loops [Douglas

etal, 2005] 40 Fig. 2.25 Steady state flow curve [Shaughnessy and Clark, 1988] 41 Fig. 2.26 Typical example of structural breakdown curves for SCC [Assaad et al., 2003A] 42 Fig. 2.27 Typical example of structural breakdown area calculation [Assaad et al., 2003A] 42 Fig. 2.28 Principle for the evaluation of drop in apparent viscosity 43 Fig. 2.29 Typical torque-time profile for concrete with 200 mm slump [Assaad et al., 2003A] 44 Fig. 2.30 Configuration of rheometer for tests on micro mortar [Billberg, 2006] 45 Fig. 2.31 Results of dynamic yield stress development and structural build-up of micro mortars made

with w/c ranging from 0.34 to 0.42 [Billberg, 2006] 45 Fig. 2.32 Effect of structural build-up on pressure loss in 1st hr after casting [Billberg et al., 2006] 47 Fig. 2.33 Variations in relative pressure of SCC made with various contents of ternary binder [Assaad

and Khayat, 2005A] 48 Fig. 2.34 Variations of Pm/Pp values vs. coarse aggregate content [Amziane and Baudeau, 2000] 49 Fig. 2.35 Variations of relative pressure with elapsed time after casting for mixtures made with 10 mm

MSA [Assaad and Khayat, 2005C] 50 Fig. 2.36 Variations of breakdown area for mixtures made with various coarse aggregate

concentrations (MSA = 10 mm) [Assaad and Khayat, 2005C] 51

307

Fig. 2.37 Effect of w/cm on relative pressure variations of SCC made with PNS-based HRWRA [Khayat and Assaad, 2006] 52

Fig. 2.38 Effect of mixture consistency on relative pressure (consistency values are noted at the end of each test) [Assaad and Khayat, 2006] 53

Fig. 2.39 Variation of relative form pressure with casting rate for SCC [Billberg, 2003] 53 Fig. 2.40 Effect of casting rate on relative pressure of SCC [Assaad and Khayat, 2006] 54 Fig. 2.41 Force in the lower anchor versus filling level [Wolfgang and Stephan, 2003] 55 Fig. 2.42 Effect of SCC temperature on variation of relative pressure [Assaad and Khayat, 2006] 56 Fig. 2.43 Relationship between initial and final setting times and elapsed time for lateral pressure

cancellation [Khayat and Assaad, 2005A] 57 Fig. 2.44 Effect of section width on lateral pressure [Khayat et al., 2005A] 58 Fig. 2.45 Variation of SCC pressure cast in formworks of different materials [Tejeda-Dominguez and

Lange, 2005] 59 Fig. 2.46 Honeywell pressure sensor of 19 mm diameter [Assaad et al., 2003B] 60 Fig. 2.47 Pressure sensor used by Andreas and Cathleen [2003] 61 Fig. 2.48 Design of lateral pressure device [Andriamanantsilavo and Amziane, 2004] 62 Fig. 2.49 Strain gage plate and strain measurement system [Arslan etal., 2005] 62 Fig. 2.50 Variations of pore water pressure in cement paste with time [Radocea, 1994] 64 Fig. 2.51 Variations of pore water and lateral pressures with respect to time for the 0.46-SCC mixture

[Assaad and Khayat, 2004] 65 Fig. 2.52 Variations of pore water and lateral pressures and concrete temperature with time for the

0.50-10-SCC mixture [Assaad and Khayat, 2004] 65 Fig. 2.53 View of the set-up device [Andriamanansilav and Amziane, 2004] 66 Fig. 2.54 Diagram of the evolution of pore water pressure and total lateral pressure [Andriamanansilav

and Amziane, 2004] 67 Fig. 2.55 Kinetics of pore water pressure of fresh cement paste (a) w/c = 0.30, (b) w/c = 0.36, and (c)

w/c = 0.45 [Andriamanansilav and Amziane, 2004] 67 Fig. 2.56 Pressure sensors and formwork (a) and pressure envelope (b) [CEBTP, 1999] 68 Fig. 2.57 Test set-up for pressure measurements in laboratory (wall 2.70 x 0.75 x 0.20 m) [Andreas

and Cathleen, 2003] 71 Fig. 2.58 Final form pressure envelopes for fully cast walls using SCC (casting no. 1-7) and

conventionally vibrated concrete (casting no. 8) [Billberg, 2003] 72 Fig. 2.59 Variations in relative lateral pressure at 0.3 m from the bottom of 5.5-m high repair wall

sections determined for different repair SCC mixtures [Khayat et al., 2005B] 74 Fig. 2.60 Formwork of 8.5-m high strong reaction wall [Tejeda-Dominguez et al., 2005] 75 Fig. 2.61 Envelope of maximum pressure exerted by the SCC in the reaction wall [Tejeda-Dominguez

etal., 2005] 75

Chapter 3 Fig. 3.1 Grain-size distributions of GU, HE, and GU-bS/F cements 76 Fig. 3.2 Grain-size distributions of limestone fillers 78 Fig. 3.3 Particle-size distributions of Class F fly ash 79 Fig. 3.4 Grain-size distribution of sand 80

Chapter 4 Fig. 4.1 Metallic pressure column of 1.2 m in height and 0.2 m in diameter to monitor initial pressure

of simulated formwork height up to 9 m 92 Fig. 4.2 PVC UofSl pressure column of 1.2 m in height and 0.2 m in diameter to monitor initial

pressure of simulated formwork height up to 9 m 92 Fig. 4.3 3-m high PVC column of 0.2 m in diameter to monitor pressure variation during the plastic

stage 92 Fig. 4.4 1.2-m high PVC column of 0.2 m in diameter to measure pressure decay 92 Fig. 4.5 Rectangular plywood formwork instrumented with pressure sensors in longitudinal and

transverse direction (all dimensions in m) 93 Fig. 4.6 19-mm pressure sensor used in PVC Columns 94 Fig. 4.7 GP:50 flush diaphragm pressure transmitter of 38-mm diameter 94 Fig. 4.8 Variations of lateral pressure determined from 19 and 38-mm diameter pressure sensors 95

Chapter 5

308

Fig. 5.1 Lateral pressure envelop obtained using the UofSl pressure column filled with SCC at 0.5 and 1 m 100

Fig. 5.2 Variations of lateral pressure in time using the UofSl pressure column filled with SCC at 0.5 and 1.0 m 100

Fig. 5.3 The UofS2 pressure column of 0.7-m high and 0.2-m diameter to evaluate lateral pressure envelop and early pressure decay of SCC mixtures 100

Fig. 5.4 Digital manometer to control overhead air pressure 100 Fig. 5.5 Comparison between two methods of applying the overhead air pressure on the lateral

pressure envelops obtained using the UofS2 pressure column 101 Fig. 5.6 Variation of lateral pressure in time obtained using the UofS2 pressure column filled with 0.5

and 0.35 m concrete plugs for SCC mixtures of various slump flow values 102 Fig. 5.7 Lateral pressure profiles obtained using the UofS2 pressure column having 0.5 and 0.35 m

concrete plugs for SCC mixtures of various slump flow values 103 Fig. 5.8 Lateral pressure variation with time determined using the UofS2 pressure column versus 3-m

free standing PVC column for different SCC mixtures 105 Fig. 5.9 Lateral pressure envelops from UofS2 pressure column vs. 3-m free stand PVC column 106 Fig. 5.10 Decay of relative lateral pressure monitored using 1.2-m PVC column 107 Fig. 5.11 Decay of relative lateral pressure monitored using 1.2-m high PVC column and the UofS2

pressure column; the latter monitored for one hour after the end of casting 107 Fig. 5.12 Variations of lateral pressure in time for SCC5 used to determine the experimental error of

theUofS2 pressure column 108 Fig. 5.13 Lateral pressure envelops of SCC5 used to determine the experimental error of the UofS2

pressure column 109 Fig. 5.14 Variation of lateral pressure with time of SCC6 cast at 2 and 10 m/hr 110 Fig. 5.15 Lateral pressure envelops of SCC5 cast at 2 and 10 m/hr 110 Fig. 5.16 Increase of lateral pressure during pressure simulation of 8-m high column for SCC mixtures

of various slump flows (slump flow was increased by addition of HRWRA 111 Fig. 5.17 Variation of lateral pressure envelops for SCC mixtures of various slump flows (slump flow

was increased by addition of HRWRA 112 Fig. 5.18 Increase of lateral pressure during pressure simulation of 13 m height cast with SCC

mixtures with different VMA concentrations 113 Fig. 5.19 Lateral pressure envelops for SCC mixtures containing different VMA concentrations 113 Fig. 5.20 Increase of lateral pressure during the simulation of casting of 13-m high column using SCC

mixtures of different paste volumes 114 Fig. 5.21 Variation of initial lateral pressure with concrete height for SCC mixtures containing various

paste volumes 114 Fig. 5.22 Variation of lateral pressure with time for SCC mixtures proportioned with different MSA

(slump flow of 660 ± 13 mm) 115 Fig. 5.23 Variation of initial lateral pressure with concrete height for SCC mixtures proportioned with

different MSA (slump flow of 660 ± 13 mm) 116

Chapter 6 Fig. 6.1 Vane shear test used in to determine undrained shear strength of clay soils 120 Fig. 6.2 Scissometer or vane shear test: (left) geometry and (right) measurement in circular bucket

filled with concrete [Roussel and Cussigh, 2008] 120 Fig. 6.3 Static or apparent yield stress as a function of resting time [Roussel and Cussigh, 2008] 121 Fig. 6.4 Four buckets and four vanes used in the PV test 122 Fig. 6.5 Schematic of PV test 122 Fig. 6.6 Torque-meter used in the PV test 122 Fig. 6.7 Variations of static yield stress at rest with time obtained with portable vane (PV) test 123 Fig. 6.8 Free-body diagram of mass 'm' kept on an inclined plane with slope angle ' 6 ' [Oremus and

Richard, 2006] 124 Fig. 6.9 Sketch showing the movement of cementitious material along the inclined plane at the

critical angle '0 ' [Oremus and Richard, 2006] 124 Fig. 6.10 Inclined plane test at different rest times: increase in flow distance and critical angle 125 Fig. 6.11 Schematic for the IP test 125 Fig. 6.12 Variations of static yield stress at rest with time obtained with inclined plane test 126 Fig. 6.13 Relative error of field-oriented tests on concrete mixture 128 Fig. 6.14 Correlation of static yield stress determined from portable vane and inclined plane tests 129 Fig. 6.15 Correlation between static yield stress determined from PV and concrete rheometer 130

309

Fig. 6.16 Correlation of static yield stress determined from inclined plane and concrete rheometer 130 Fig. 6.17 Relationship between static yield stress determined from the portable vane test and drop in

apparent viscosity at 0.7 rps obtained using concrete rheometer 131 Fig. 6.18 Relationship between static yield stress determined from inclined plane test and drop in

apparent viscosity at 0.7 rps obtained using concrete rheometer 131 Fig. 6.19 Relationship between static yield stress determined from portable vane test and breakdown

area obtained using concrete rheometer 132 Fig. 6.20 Relationship between static yield stress determined from inclined plane test and breakdown

area obtained using concrete rheometer 132 Fig. 6.21 Relationship between time-dependent static yield stress determined from portable vane and

inclined plane tests 133 Fig. 6.22 Relationship between time-dependent static yield stress determined from portable vane and

concrete rheometer tests 133 Fig. 6.23 Relationship between time-dependent static yield stress determined from inclined plane and

concrete rheometer tests 133 Fig. 6.24 Relationship between time-dependent static yield stress determined from portable vane test

and time-dependent drop in apparent viscosity at 0.7 rps obtained using concrete rheometer 134 Fig. 6.25 Relationship between time-dependent static yield stress determined from inclined plane test

and time-dependent drop in apparent viscosity at 0.7 rps obtained using concrete rheometer 134

Chapter 7 Fig. 7.1 Flow chart for Chapter 7 137 Fig. 7.2 Variation of relative lateral pressure with concrete height 140 Fig. 7.3 Pressure decay characteristics 141 Fig. 7.4 Effect of mix design parameters of SCC on PVT0res,@15mjn 142 Fig. 7.5 Variations of K0 with Anapp(gN=07rps(g15min at different heights of placement 143 Fig. 7.6 Variations of Ko with PVT0rest@i5min at different heights of placement 143 Fig. 7.7 Variations of K0 with Anapp@N=o7 ^(t) at different heights of placement 144 Fig. 7.8 Variations of K0 with PVt0 rest(t) at different heights of placement 144 Fig. 7.9 Variations of Ko with Anapp<^,7TS@i5minxAnapp@N^7rps(t) at different heights of placement 144 Fig. 7.10 Variations of K0 with PVT0rest@i5min><Pvtorest(t) at different heights of placement 145 Fig. 7.11 Variations of pressure decay with Anapp@N=0 7 rps(t) 146 Fig. 7.12 Variations of pressure decay with PVxorest@ismin 146 Fig. 7.13 Variations of pressure decay with Anapp@ N=0.7rpS(t) 146 Fig. 7.14 Variations of pressure decay with PVT0rest(t) 147 Fig. 7.15 Variations of pressure decay with Anapp@N=07q)S@i5minxAr|app@N=o7ips(t) 147 Fig. 7.16 Variations of pressure decay with PVT0rest@i5minxTorest(t) 147 Fig. 7.17 Relationship between predicted K0 and PVT0rest@i5min values 155 Fig. 7.18 Trade-offof different parameters affecting K0@H=8m 157 Fig. 7.19 Trade-off of different parameters affecting [(AK(t)(0-tc)] 158 Fig. 7.20 Trade-offof different parameters affecting PVT0rest@i5min 159 Fig. 7.21 Trade-offof different parameters affecting PViorestCO 160 Fig. 7.22 Trade-offof different parameters affecting IPT0resi@i5min 161 Fig. 7.23 Variations of K0 with concrete height (left) and paste volume (right) 162 Fig. 7.24 Variations of K0 with time for SCC mixtures of different paste volume (left) and pressure

decay indices with paste volume (right) 163 Fig. 7.25 Relationship between pressure cancellation time and paste volume 163 Fig. 7.26 Effect of paste volume on the breakdown area between 0-30 min 164 Fig. 7.27 Effect of breakdown area on K0 165 Fig. 7.28 Variation of initial lateral pressure with concrete height for SCC proportioned with different

MSA '. 166 Fig. 7.29 Recommended modification coefficients of K0 with changes in MSA 167

Chapter 8 Fig. 8.1 Lateral pressure envelops for concrete cast at 5 and 10 m/hr and different temperatures 174 Fig. 8.2 Variations of lateral pressure with concrete temperature at casting rates of 5 and 10 m/hr 174 Fig. 8.3 Variations of pressure cancellation time with concrete temperature 175 Fig. 8.4 Variations of pressure decay with concrete temperature 176 Fig. 8.5 Variations of K0 with casting rate at various depths [SCC54 of thixotropy level # 5] 176 Fig. 8.6 Variations of Ko at H = 3 m with casting rate for SCC of various thixotropy levels 177

310

Fig. 8.7 Lateral pressure envelops for SCC mixtures of different thixotropy levels cast at 2 m/hr (top) and 30 m/hr (bottom) 178

Fig. 8.8 Abacuses of K0 values and initial thixotropic indices obtained using concrete rheometer and field-oriented tests 179

Fig. 8.9 Variations of lateral pressure with time for SCC57 (low thixotropy level) cast at continuous rate of 10 m/hr and with one and two waiting periods 180

Fig. 8.10 Variations of lateral pressure with time for SCC59 (high thixotropy level) cast at continuous rate of 10 m/hr and with one and two waiting periods 181

Fig. 8.11 Lateral pressure envelops for SCC57 of low thixotropy level (left) and SCC59 of high thixotropy level (right); both cast continuously and with one and two waiting periods 181

Fig. 8.12 Relationship between the modification factor for waiting period between successive lifts (fWP) a n d PVT0rest@15min 182

Fig. 8.13 Variations of K0 values with Dmj„ determined from pressure sensors fixed in transverse (top) and longitudinal (bottom) directions 183

Fig. 8.14 Arching action in thin and wide sections 183 Fig. 8.15 Influence of Dmin on lateral pressure distribution 184 Fig. 8.16 Variation of pressure decays with Dmin of formwork 185 Fig. 8.17 Relationship between K0 values and Dmin 186 Fig. 8.18 Relationship between correction factor for K0 and Dmin 186 Fig. 8.19 Variation of pressure decays with Dmin of formwork 187 Fig. 8.20 Relationship between correction factor for AK(t)(0-tc) and Dmin 187

Chapter 9 Fig. 9.1 Relationship between the predicted Pmax values from Eqs. 9.40 and 9.53 195 Fig. 9.2 Correlations between K0 and Rheometerr0 rest@i5min 199 Fig. 9.3 Correlations between K0 and ^r\m@^^n^@\smm 200 Fig. 9.4 Correlations between K0 and PVT0rest@i5min « 201 Fig. 9.5 Correlations between K0 and IPT0rest@i5min 202 Fig. 9.6 Comparison of predicted Pmax from ACI 347 and UofS models for SCC mixtures of different

thixotropy levels 210 Fig. 9.7 Comparison between Pmax values predicted from German Standard DIN 18218 and UofS models 210 Fig. 9.8 Comparison between Pmax values determined from Roussel and Ovarlez's model [2005] and

UofS model 211 Fig. 9.9 Comparison between K0 values determined from Khayat and Assaad's (2005A) model and

UofS model 211 Fig. 9.10 Comparison between lateral pressure variations with casting depth determined from UofS

model and published models 212

Chapter 10 Fig. 10.1 Configurations of walls # 5 to 8 constructed on level 2 216 Fig. 10.2 Wall panels # 6 to 8: the steel bars used for reinforcement (left) and during casting (right) 217 Fig. 10.3 Pressure sensor set flush with concrete surface 217 Fig. 10.4 Casting SCC in formwork 217 Fig. 10.5 Appearance of wall element cast with SCC after formwork removal 218 Fig. 10.6 Full characterization at Material laboratory (UofS) by a team of researchers 219 Fig. 10.7 Configuration of CTLGroup column elements 221 Fig. 10.8 Column forms extending from lower level of the laboratory to the upper level (left), fixation

of pressure sensor set flush with the concrete surface in the column form (middle), and reinforcement of the column shown from the plane view (left) 221

Fig. 10.9 Laser-based monitoring stand for controlling casting rate, pressure monitoring data acquisition system, fresh concrete testing equipments 222

Fig. 10.10 Columns at the end of casting that remain in rest for 24 hr before demolding 222 Fig. 10.11 Variations of lateral pressure and concrete temperature with time, wall # 6 SCC62) 224 Fig. 10.12 Effect of casting rate on lateral pressure envelops from field observations (left) and using

the UofS2 pressure column (right) (SCC62, walls # 2, 3, and 4) 225 Fig. 10.13 Effect of casting depth on lateral pressure envelops of SCC62: from field observations

(left) and the UofS2 pressure column (right) 226 Fig. 10.14 Effect of paste volume and w/cm on the variations of lateral pressure with casting depth

determined from field measurements (left) and using the UofS2 pressure column (right) 227

311

Fig. 10.15 Effect of VMA concentration and w/cm on the variations of lateral pressure with casting depth: from field measurements (left) and the UofS2 pressure column (right) 227

Fig. 10.16 Variations of lateral pressure with time for columns # 5 and 6 cast using SCC-H 228 Fig. 10.17 Effect of casting rate (left) and waiting period (right) on lateral pressure envelop 229 Fig. 10.18 Measured to predicted lateral pressure values for CFI walls and CTLGroup columns 230 Fig. 10.19 Measured-to-predicted AK(t)(0-60 min) values 232 Fig. 10.20 Measured-to-predicted AK(t)(0-tc) values 232

Chapter 11 Fig. 11.1 Proposed design for UofS3 pressure column to simulate the caisson effect on lateral

pressure distribution 246

Appendix A for Chapter 4 Fig. A.l Mechanical calibration of pressure sensor to determine mechanical calibration factor (Cm) 248 Fig. A.2 Sketch for UofSl pressure column indicating sensor positions and water levels used in the

hydrostatic calibration 249 Fig. A.3 Relationship between hydrostatic pressure (PH) and the pressure corrected mechanically

(P2CO,T.M) to determine hydrostatic calibration factor (CH) 249 Fig. A.4 Relationship between water pressure and P3cor rH to obtain water calibration factor (Cw) 251

Appendix B for Chapter 6 Fig. B.l Schematic of centering the vane to the bucket base in the PV test 253

Appendix C for Chapter 7 Fig. C.l Variations of K0 with Rheometerxorest@i5min at different heights of placement 258 Fig. C.2 Variations of K0 with IPT0rest@i5min at different heights of placement 258 Fig. C.3 Variations of K0 with Rheometerx0 rest(t) at different heights of placement 259 Fig. C.4 Variations of K0 with IPT0rest(t) at different heights of placement 259 Fig. C.5 Variations of K0 with Rheometerx0rest@i5min)<Torest(t) at different heights of placement 260 Fig. C.6 Variations of K0 with IPT0rest@i5min><'Corest(t) at different heights of placement 260 Fig. C.7 Variations of pressure decay with RheometerTorest@i5min

261 Fig. C.8 Variations of pressure decay with IPxorest@i5min 261 Fig. C.9 Variations of pressure decay with Rheometen0 rest(t) 262 Fig. CIO Variations of pressure decay with IPx0rest(t) 262 Fig. C.ll Variations of pressure decay with RheometerT0rest@i5min><t:orest(t) 263 Fig. C.12 Variations of pressure decay with IPxorest@i5minx*orest(t) 263 Fig. C.13 Relationship between predicted K0 and Rheometenorest@i5 min values 264 Fig. C.14 Relationship between predicted K^ and Ar\aff@N=Qlrris@\im[nva\ues 264 Fig. C.15 Relationship between predicted K0 and IPT0rest@i5min values 264 Fig. C.16 Relationship between predicted pressure cancellation time (tc) and Rheometerx0r<:st(t) values 265 Fig. C.17 Relationship between predicted pressure cancellation time (tc) and A r i ^ ^ N ^ ^ t ) values 265 Fig. C.18 Relationship between predicted pressure cancellation time (tc) and PVi0rest(0 values 265 Fig. C.19 Relationship between predicted pressure cancellation time and IPx0rest(t) values 266 Fig. C.20 Relationship between predicted Rheometert0 rest@15 min a n d P V t o rest@15 min Values 2 6 6

Fig. C.21 Relationship between predicted Rheometert0rest@i5min and Anapp@N=o7rps@i5min values 266 Fig. C.22 Relationship between predicted Rheometerr0rest@i5min and IPxorest@i5min values 267 Fig. C.23 Relationship between predicted PVx0reSt@i5min and IPxoresi@i5min values 267 Fig. C.24 Relationship between predicted Rheometerx0 rest(t) and PVT0rest(t) values 267 Fig. C.25 Relationship between predicted Rheometerrorest(t) and Ar|apP@N=o7rpS(0 values 268 Fig. C.26 Relationship between predicted Rheometerx0 rest(t) and IPxorest(t) values 268 Fig. C.27 Relationship between predicted PVx0 res,(t) and IPx0 rest(t) values 268 Fig. C.28 Trade-off of different parameters affecting K0@H=4m 269 Fig. C.29 Trade-off of different parameters affecting K0@H=i2m 270 Fig. C.30 Trade-off of different parameters affecting [AK(t)(0-60 min)] 271 Fig. C.31 Trade-off of different parameters affecting tc 272 Fig. C.32 Trade-off of different parameters affecting Rheometert0rest@i5min 273 Fig. C.33 Trade-off of different parameters affecting Rheometerx0rest(t) 274 Fig. C.34 Trade-off of different parameters affecting Anapp@N=,o.7rps@i5min 275 Fig. C.35 Trade-off of different parameters affecting Anapp@N=o7rpS(t) 276 Fig. C.36 Trade-off of different parameters affecting IPx0rest(t) 277

312

Appendix D for Chapter 8 Fig. D.l Variations of K0 values with casting rate at heights of 1, 3, 7, and 10 m [SCC40 of

thixotropy level # 1] 282 Fig. D.2 Variations of K0 values with casting rate at heights of 1, 3, 7, and 10 m [SCC51 of




thixotropy level #6] 284 Fig. D.6 Variations of K0 values with casting rate at heights of 1, 3, 7, and 10 m [SCC55 of


thixotropy level # 8] 285 Fig. D.8 Variations of K0 values @ H = 7 m with casting rate for SCC of different thixotropy levels 286 Fig. D.9 Variations of K0 values @ H = 10 m with casting rate for SCC of different thixotropy levels 286 Fig. D.10 Lateral pressure profiles for SCC mixtures of different thixotropy levels, cast at rate of 5 m/hr 287 Fig. D.l 1 Lateral pressure profiles for SCC mixtures of different thixotropy levels, cast at rate of 10

m/hr 287 Fig. D.l 2 Lateral pressure profiles for SCC mixtures of different thixotropy levels, cast at rate of 17

m/hr 288 Fig. D.l 3 Lateral pressure profiles for SCC mixtures of different thixotropy levels, cast at rate of 24

m/hr 288

Appendix E for Chapter 9 Fig. E.l Correlations between K0 and Ab| 290 Fig. E.2 Correlations between K0 and RheometerTorest(t) 291 Fig. E.3 Correlations between K0 and Anapp@N=0 7rps(t) 292 Fig. E.4 Correlations between K0 and PVT0rest(0 293 Fig. E.5 Correlations between K0 and IPiorest(t) 294 Fig. E.6 Correlations between K0 and RheometerT0rest@i5min><Torest(t) 295 Fig. E.7 Correlations between K0 and Anapp@N^7TS@,5minxAr|app@N=o.7rps(t) 296 Fig. E.8 Correlations between K0 and PVTorest@i5min*PVTorest(t) 297 Fig. E.9 Correlations between K0 and IPiorest@i5minxPvt0rest(t) 298

313

LIST OF TABLES

Chapter 2 Table 2.1 Spread in pressure from relative pressure determined at casting rate of 10 m/hr 31 Table 2.2 Factors influencing formwork pressure 47 Table 2.3 Variation of mixture designs and casting rates [Billberg, 2003] 72 Table 2.4 Mixture proportion and workability of SCC used in repair [Khayat et al., 2005B] 73

Chapter 3 Table 3.1 Characteristics of GU, HE, MS, and GU-bS/F cements 77 Table 3.2 Physico-chemical properties of limestone fillers 78 Table 3.3 Physico-chemical properties of Class F fly ash 79 Table 3.4 Particle-size distribution of sand 80 Table 3.5 Physical characteristics of coarse aggregates 80 Table 3.6 Grain-size distributions of coarse aggregates 81 Table 3.7 Characteristics of chemical admixtures used 81 Table 3.8 Mixture compositions 83

Chapter 5 Table 5.1 Experimental work used to validate the UofS2 pressure column 98 Table 5.2 Fresh concrete properties 98 Table 5.3 Repeatability of lateral pressure at various casting heights 109

Chapter 6 Table 6.1 Summary of experimental program 119 Table 6.2 Rheological measurements [Roussel and Cussigh, 2008] 121 Table 6.3 Vane dimensions 122 Table 6.4 Repeatability of field-oriented tests on thixotropic concrete mixture: SCC46 127 Table 6.5 Thixotropic indices obtained from the field-oriented test methods 134 Table 6.6 R2 values for initial and time-dependent responses 135

Chapter 7 Table 7.1 Ranges of selected parameters for the factorial design 137 Table 7.2 Mixtures used to evaluate effect of paste volume on lateral pressure and thixotropy 138 Table 7.3 Experimental work to evaluate effect of MSA on SCC lateral pressure 139 Table 7.4 R2 values for the various correlations between K0 and thixotropy indices 145 Table 7.5 R2 values of the various correlations between pressure decay and thixotropy indices 148 Table 7.6 Full-factorial experimental design carried out in Phase II 148 Table 7.7 Formulas used to convert absolute to coded values for parameters considered 149 Table 7.8 Parameter estimates of derived models 150 Table 7.9 Statistical models in coded values (values of <|>, S/A, and Vca from-1 to+1) 151 Table 7.10 Statistical models in absolute values (<)> = 600-720mm, S/A = 0.44-0.52 by volume, and

Vca = 0.27-0.33 by volume) 152 Table 7.11 Repeatability of test results (n = 4) 153 Table 7.12 Measured versus predicted responses using derived statistical models 154

Chapter 8 Table 8.1 Variables to evaluate effect of temperature on SCC lateral pressure characteristics 171 Table 8.2 Parameters to evaluate effect of casting rate on SCC lateral pressure characteristics 172 Table 8.3 Methodology to evaluate effect of waiting period between lifts on SCC lateral pressure 172 Table 8.4 Parameters to evaluate effect of formwork dimension on lateral pressure characteristics 173

Chapter 9 Table 9.1 Modeled parameters in the prediction models of SCC formwork pressure 191 Table 9.2 Relationships between different thixotropic indices 192 Table 9.3 Prediction models for

Pmax <*s function of H, R, T, Dmjn, and T.I.@T=22±2°C 194

314

Table 9.4 Prediction models for Pmax as function of H, R, Dmin, and T.I.@Ti (T is considered within T.I.@Ti) 195

Table 9.5 Correlations between measured and predicted Pmax values determined from the models of (H, R> T, Dmjn, T.I.@T=22±2°c>/MSA, and/j^.) 197

Table 9.6 Correlations between the measured and predicted Pmax values determined from the models of (H, R, Dmin, T.l.@Tl,fMSA, andfWP) 197

Table 9.7 Pmax (in kPa) values for formwork 204 Table 9.8 Color indication for various Pmax intervals 206 Table 9.9 Prediction models for [AK(t)(0-60 min)] as function of thixotropy index 207 Table 9.10 Prediction models for [AK(t)(0-tc)] as function of thixotropy index 207 Table 9.11 Correlations between measured and predicted AK(t)(0-60 min) values 208 Table 9.12 Correlations between measured and predicted AK(t)(0-tc) values 208

Chapter 10 Table 10.1 Experimental program for eight walls cast at the Universitede Sherbrooke 218 Table 10.2 Fresh concrete properties of mixtures used in field investigation at Sherbrooke 219 Table 10.3 Rheological and thixotropic measurements of mixtures used in the field investigation 220 Table 10.4 Test matrix for CTLGroup columns 220 Table 10.5 Results of eight column elements cast at CTLGroup laboratory 223

Chapter 11 Table 11.1 Relative error in predicting relative lateral pressure value 235 Table 11.2 Vane dimensions of the portable vane test 236 Table 11.3 Various thixotropic indices obtained from the field-oriented test methods 237 Table 11.4 R2 values for the correlations between PV and IP tests versus concrete rheometer 238 Table 11.5 Modeled parameters in the prediction models of SCC formwork pressure 238

Appendix A for Chapter 4 Table A.l Mechanical pressure vs. output signal to determine mechanical calibration factor (Cm) 247 Table A.2 Calculation of hydrostatic pressure (PH) and P2corrM to determine hydrostatic calibration

factor (CH) 250 Table A.3 Calculation of water pressure and P3C0ITH to determine water calibration factor (Cw) 251

Appendix B for Chapter 6 Table B.l Fresh concrete properties 252

Appendix C for Chapter 7 Table C.l Fresh concrete properties of Phase 1 256 Table C.2 Fresh concrete properties of Phase II 256 Table C.3 Fresh concrete properties of Phase III 257

Appendix D for Chapter 8 Table D.l Fresh properties of concrete used to evaluate effect of temperature on lateral pressure

characteristics (Phase I) 278 Table D.2 Fresh properties of SCC mixtures used to evaluate effect of casting rate on lateral pressure

characteristics (Phase II) 279 Table D.3 Fresh properties of SCC mixtures used to evaluate effect of waiting period between

successive lifts on formwork pressure (Phase III) 281 Table D.4 Fresh properties for concretes used to evaluate effect of minimum formwork dimension on

lateral pressure characteristics (Phase IV) 281

315

pression exercee sur le coffrage par le beton auto-placant formwork pressure exerted ... ·...

Documents