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
UofS3 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
Pressure column version 2, developed at University of Sherbrooke
Pressure column version 3, 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
Chapter 1 Introduction
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
Chapter 1 Introduction
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
Chapter 1 Introduction
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
Chapter 1 Introduction
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
Chapter 1 Introduction
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
Chapter 2: Review on formwork pressure and fundamentals ofrheology
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
Chapter 2: Review on formwork pressure and fundamentals ofrheology
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 .
Chapter 2: Review on formwork pressure and fundamentals ofrheology
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
Chapter 2: Review on formwork pressure and fundamentals ofrheology
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
Chapter 2: Review on formwork pressure and fundamentals ofrheology
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
Chapter 2: Review on formwork pressure and fundamentals ofrheology
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
Chapter 2: Review on formwork pressure and fundamentals ofrheology
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
Chapter 2: Review on formwork pressure and fundamentals ofrheology
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
Chapter 2: Review on formwork pressure and fundamentals ofrheology
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
Chapter 2: Review on formwork pressure and fundamentals ofrheology
(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
Chapter 2: Review on formwork pressure and fundamentals ofrheology
(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
Chapter 2: Review on formwork pressure and fundamentals ofrheology
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
Chapter 2: Review on formwork pressure and fundamentals ofrheology
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
$m).M(t).v.jdt dt+C
o-.=crv.A(t)
Eq. 2.42
Eq. 2.43
Eq. 2.44
Eq. 2.45
Eq. 2.46
Eq. 2.47
Eq. 2.48
23
Chapter 2: Review on formwork pressure and fundamentals ofrheology
V s ; / / " " c n n r . r B t e S ? 1 / ! ^ / l e a s t from
Concrete
30 cm Stress in the silo
A Area of the mould cross-section H Height of the formwork U Perimeter of the mould cross-section yc Fresh concret volume weight oh Horizontal pressure o„ Vertical pressure X Pressure ratio u Friction coefficient xw Friction stress t, tA, tE, Time, initial setting time,
and final setting time
Fig. 2.6 Experimental column and sketch for pressure calculations
[Graubner and Proske, 2005B]
Concrete level
F«$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
Chapter 2: Review on formwork pressure and fundamentals ofrheology
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
Chapter 2: Review on formwork pressure and fundamentals ofrheology
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
Chapter 2: Review on formwork pressure and fundamentals ofrheology
-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
Chapter 2: Review on formwork pressure and fundamentals ofrheology
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
Chapter 2: Review on formwork pressure and fundamentals ofrheology
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
Chapter 2: Review on formwork pressure and fundamentals ofrheology
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
Chapter 2: Review on formwork pressure and fundamentals ofrheology
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
Chapter 2: Review on formwork pressure and fundamentals ofrheology
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
Chapter 2: Review on formwork pressure and fundamentals ofrheology
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
Chapter 2: Review on formwork pressure and fundamentals ofrheology
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
Chapter 2: Review on formwork pressure and fundamentals ofrheology
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
Chapter 2: Review on formwork pressure and fundamentals ofrheology
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
Chapter 2: Review on formwork pressure and fundamentals ofrheology
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
Chapter 2: Review on formwork pressure and fundamentals ofrheology
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
Chapter 2: Review on formwork pressure and fundamentals ofrheology
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
Chapter 2: Review on formwork pressure and fundamentals ofrheology
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
Chapter 2: Review on formwork pressure and fundamentals ofrheology
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
Chapter 2: Review on formwork pressure and fundamentals ofrheology
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
Chapter 2: Review on formwork pressure and fundamentals ofrheology
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
Chapter 2: Review on formwork pressure and fundamentals ofrheology
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
Chapter 2: Review on formwork pressure and fundamentals ofrheology
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
Chapter 2: Review on formwork pressure and fundamentals ofrheology
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
Chapter 2: Review on formwork pressure and fundamentals ofrheology
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
Chapter 2: Review on formwork pressure and fundamentals ofrheology
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
Chapter 2: Review on formwork pressure and fundamentals ofrheology
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
Chapter 2: Review on formwork pressure and fundamentals ofrheology
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
Chapter 2: Review on formwork pressure and fundamentals ofrheology
_ 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
Chapter 2: Review on formwork pressure and fundamentals ofrheology
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
Chapter 2: Review on formwork pressure and fundamentals ofrheology
[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
Chapter 2: Review on formwork pressure and fundamentals ofrheology
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
Chapter 2: Review on formwork pressure and fundamentals ofrheology
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
Chapter 2: Review on formwork pressure and fundamentals ofrheology
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
Chapter 2: Review on formwork pressure and fundamentals ofrheology
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
Chapter 2: Review on formwork pressure and fundamentals ofrheology
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
Time after casting (min)
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
Chapter 2: Review on formwork pressure and fundamentals ofrheology
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
Chapter 2: Review on formwork pressure and fundamentals ofrheology
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
Chapter 2: Review on formwork pressure and fundamentals ofrheology
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
Chapter 2: Review on formwork pressure and fundamentals ofrheology
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
Chapter 2: Review on formwork pressure and fundamentals ofrheology
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
Chapter 2: Review on formwork pressure and fundamentals ofrheology
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
Time after casting (min)
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
Chapter 2: Review on formwork pressure and fundamentals ofrheology
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
Chapter 2: Review on formwork pressure and fundamentals ofrheology
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
Chapter 2: Review on formwork pressure and fundamentals ofrheology
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
Chapter 2: Review on formwork pressure and fundamentals ofrheology
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
Chapter 2: Review on formwork pressure and fundamentals ofrheology
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
Chapter 2: Review on formwork pressure and fundamentals ofrheology
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
Chapter 2: Review on formwork pressure and fundamentals ofrheology
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
Chapter 2: Review on formwork pressure and fundamentals ofrheology
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
Chapter 2: Review on formwork pressure and fundamentals ofrheology
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
Chapter 2: Review on formwork pressure and fundamentals ofrheology
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
Compressive strength at 3
Compressive strength at 7
, %
%
, %
%
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
Chapter 3: Materials and mix designs
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
Chapter 3: Materials and mix designs
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
Chapter 3: Materials and mix designs
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
Chapter 3: Materials and mix designs
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
200-2000 mL/100 kg of binder
200-2000 mL/100 kg of binder
130-920 mL/100 kg of binder
130-650 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
Chapter 3: Materials and mix designs
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
Chapter 4: Methodology for lateral pressure measurements
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
Chapter 4: Methodology for lateral pressure measurements
•
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
Chapter 4: Methodology for lateral pressure measurements
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
Chapter 4: Methodology for lateral pressure measurements
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
Chapter 5: Development portable device to measure SCC formwork pressure
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 <U S -
CX
o
(D +^
o c o
>
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
Chapter 5: Development portable device to measure SCC formworkpressure
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
Chapter 5: Development portable device to measure SCC formworkpressure
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
Chapter 5: Development portable device to measure SCC formwork pressure
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
SCC6 <|> = 760 mm R= lOm/hr
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
Chapter 5: Development portable device to measure SCC formworkpressure
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 =
-
-
Lateral pressure (kPa)
= 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
Lateral pressure (kPa) 0 100 200 300
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
Chapter 5: Development portable device to measure SCC formwork pressure
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
Chapter 5: Development portable device to measure SCC formwork pressure
70 r Hydrostatic pressure
a V i-s V i-a a
• * *
« -
uu
50
40
30
70
10
9
a
s-
UofS2 pressure column
SCC2 <|> = 660 mm R= lOm/hr
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
Hydrostatic pressure
3-m PVC column
z^mzmk^Rmmnvw^
UofS2 pressure column
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)
Hydrostatic pressure
100 120
UofS2 pressure column
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
Chapter 5: Development portable device to measure SCC formworkpressure
Lateral pressure (kPa) 20 40 60 80
SCC2 (j) = 660± 13 mm R= lOm/hr
Hydrostatic pressure
2-5 TJ0fS2 pressure column
3.0
Lateral pressure (kPa)
20 40 60 80
SCC6 <|> = 760± 13 mm R=10m/hr
Hydrostatic pressure
UofS2 pressure column
Lateral pressure (kPa)
20 40 60 80
SCC3 (|> = 760±13mm R=10m/hr
Hydrostatic pressure
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
Chapter 5: Development portable device to measure SCC formwork pressure
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
Chapter 5: Development portable device to measure SCC formwork pressure
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
Chapter 5: Development portable device to measure SCC formwork pressure
Lateral pressure (kPa) 100 200 300
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
UofS2 pressure column
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
Chapter 5: Development portable device to measure SCC formwork pressure
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
Lateral pressure (kPa) 0 100 200 300
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
Chapter 5: Development portable device to measure SCC formwork pressure
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
Lateral pressure (kPa) 50 100 150 200
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
Chapter 5: Development portable device to measure SCC formwork pressure
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
Lateral pressure (kPa) 100 200 300
\ >
-
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
Chapter 5: Development portable device to measure SCC formwork pressure
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
<u u 3 Vl
9 a
M
<u • * *
-
">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
Lateral pressure (kPa) 100 200 300
(|) = 660 mm R=10m/hr
Hydrostatic pressure
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
Chapter 5: Development portable device to measure SCC formwork pressure
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
Chapter 5: Development portable device to measure SCC formwork pressure
Lateral pressure (kPa) 100 200 300
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
Chapter 6 Field-oriented test methods to evaluate structural build-up at rest ofSCC
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
Chapter 6 Field-oriented test methods to evaluate structural build-up at rest ofSCC
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
Chapter 6 Field-oriented test methods to evaluate structural build-up at rest ofSCC
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
Chapter 6 Field-oriented test methods to evaluate structural build-up at rest ofSCC
^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
Chapter 6 Field-oriented test methods to evaluate structural build-up at rest ofSCC
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
Chapter 6 Field-oriented test methods to evaluate structural build-up at rest of SCC
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
Chapter 6 Field-oriented test methods to evaluate structural build-up at rest ofSCC
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
Chapter 6 Field-oriented test methods to evaluate structural build-up at rest of SCC
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
Chapter 6 Field-oriented test methods to evaluate structural build-up at rest ofSCC
(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
Chapter 6 Field-oriented test methods to evaluate structural build-up at rest ofSCC
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
Chapter 6 Field-oriented test methods to evaluate structural build-up at rest ofSCC
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°<b
d^V^> 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
Chapter 6 Field-oriented test methods to evaluate structural build-up at rest ofSCC
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)
Fig. 6.23 Relationship between time-
dependent static yield stress determined from
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
Chapter 6 Field-oriented test methods to evaluate structural build-up at rest ofSCC
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)
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
/ • 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)
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
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
Chapter 6 Field-oriented test methods to evaluate structural build-up at rest ofSCC
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
Chapter 7 Effect ofSCC mix design on formwork pressure characteristics
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
Chapter 7 Effect ofSCC mix design on formwork pressure characteristics
100 £ **mS
i» 3 VI fi
0* u a, « u <u ** ITS
41
> • * *
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
Chapter 7 Effect ofSCC mix design on formwork pressure characteristics
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
Chapter 7 Effect ofSCC mix design on formwork pressure characteristics
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
Chapter 7 Effect ofSCC mix design on formwork pressure characteristics
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
Chapter 7 Effect ofSCC mix design on formworkpressure characteristics
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
Chapter 7 Effect ofSCC mix design on formwork pressure characteristics
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
Chapter 7 Effect ofSCC mix design on formworkpressure characteristics
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
<i> 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
Chapter 7 Effect ofSCC mix design on formwork pressure characteristics
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
Chapter 7 Effect ofSCC mix design on formwork pressure characteristics
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
Chapter 7 Effect ofSCC mix design on formwork pressure characteristics
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 (dimensionless)
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
Chapter 7 Effect ofSCC mix design on formwork pressure characteristics
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
Chapter 7 Effect ofSCC mix design on formwork pressure characteristics
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
Chapter 7 Effect of SCC mix design on formworkpressure characteristics
^ 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
Chapter 7 Effect of SCC mix design on formworkpressure characteristics
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.
Lateral pressure (kPa) 0 100 200 300
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
Lateral pressure (kPa)
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
Chapter 7 Effect ofSCC mix design on formworkpressure characteristics
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
Chapter 7 Effect ofSCC mix design on formworkpressure characteristics
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
Chapter 8 Effect of placement characteristics andformwork dimensions on lateral pressure ofSCC
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.
Lateral pressure (kPa)
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
Chapter 8 Effect of placement characteristics andformwork dimensions on lateral pressure ofSCC
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
Chapter 8 Effect of placement characteristics andformwork dimensions on lateral pressure ofSCC
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
Chapter 8 Effect of placement characteristics andformwork dimensions on lateral pressure ofSCC
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
Chapter 8 Effect of placement characteristics andformwork dimensions on lateral pressure ofSCC
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
Chapter 8 Effect of placement characteristics andformwork dimensions on lateral pressure ofSCC
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
Chapter 8 Effect of placement characteristics andformwork dimensions on lateral pressure ofSCC
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
Lateral pressure (kPa) 100 200 300
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
Lateral pressure (kPa) 100 200 300
» 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
Chapter 8 Effect of placement characteristics andformwork dimensions on lateral pressure ofSCC
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
Chapter 8 Effect of placement characteristics andformwork dimensions on lateral pressure ofSCC
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
Chapter 8 Effect of placement characteristics andformwork dimensions on lateral pressure ofSCC
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
150 200 250 300 350 Formwork width (mm)
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
150 350 200 250 300 Formwork width (mm)
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
Chapter 8 Effect of placement characteristics andformwork dimensions on lateral pressure ofSCC
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
Chapter 8 Effect of placement characteristics andformwork dimensions on lateral pressure ofSCC
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)
Chapter 9 Prediction models for lateral pressure characteristics
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
Chapter 9 Prediction models for lateral pressure characteristics
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
Chapter 9 Prediction models for lateral pressure characteristics
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
Chapter 9 Prediction models for lateral pressure characteristics
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
Chapter 9 Prediction models for lateral pressure characteristics
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
Chapter 9 Prediction models for lateral pressure characteristics
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
Measured = 1.00 x predicted
Measured = 1.00 x predicted
Measured = 1.00 x predicted
Measured = 1.00 x predicted
Measured = 1.00 x predicted
Measured = 1.00 x predicted
Measured = 1.00 x predicted
Measured = 1.00 x predicted
Measured = 1.00 x predicted
Measured = 1.00 x predicted
Measured = 1.00 x predicted
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)
Measured = 1.00 x predicted
Measured = 1.00 x predicted
Measured = 1.00 x predicted
Measured = 1.00 x predicted
Measured = 1.00 x predicted
Measured = 1.00 x predicted
Measured = 1.00 x predicted
Measured = 1.00 x predicted
Measured = 1.00 x predicted
Measured = 1.00 x predicted
Measured = 1.00 x predicted
Measured = 1.00 x predicted
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 * Recommended
197
Chapter 9 Prediction models for lateral pressure characteristics
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
Chapter 9 Prediction models for lateral pressure characteristics
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
Chapter 9 Prediction models for lateral pressure characteristics
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
Chapter 9 Prediction models for lateral pressure characteristics
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
Chapter 9 Prediction models for lateral pressure characteristics
£
- 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
Chapter 9 Prediction models for lateral pressure characteristics
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
Chapter 9 Prediction models for lateral pressure characteristics
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
Chapter 9 Prediction models for lateral pressure characteristics
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
Chapter 9 Prediction models for lateral pressure characteristics
-" 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
Chapter 9 Prediction models for lateral pressure characteristics
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.
Lateral pressure (kPa) 100 200 300 400
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
Chapter 9 Prediction models for lateral pressure characteristics
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
Chapter 9 Prediction models for lateral pressure characteristics
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
Chapter 10 Field measurements and validation of lateral pressure
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
Chapter 10 Field measurements and validation of lateral pressure
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
Chapter 10 Field measurements and validation of lateral pressure
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
Chapter 10 Field measurements and validation of lateral pressure
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
Chapter 10 Field measurements and validation of lateral pressure
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
Chapter 10 Field measurements and validation of lateral pressure
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
Chapter 10 Field measurements and validation of lateral pressure
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
Chapter 10 Field measurements and validation of lateral pressure
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
Lateral pressure (kPa) 20 40 60 80 100
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
Lateral pressure (kPa) 20 40 60 80 100
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
Chapter 10 Field measurements and validation of lateral pressure
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!
Lateral pressure (kPa) 20 40 60 80 100 120
— 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
Chapter 10 Field measurements and validation of lateral pressure
Lateral pressure (kPa) 30 60 90 120
Lateral pressure (kPa) 20 40 60 80 100 120
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
Hydrostatic pressure
Wall # 7, \ y SCC63, ^ < Vp = 3701/m3, \ \ w/cm = 0.35 \ \ k \ N
» \ \ \
Lateral pressure (kPa) 30 60 90 120
Field measurements R = 10 m/hr
Hydrostatic pressure
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
Lateral pressure (kPa) 20 40 60 80 100 120
UofS2 pressure column R = 10 m/hr
Hydrostatic pressure
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
Chapter 10 Field measurements and validation of lateral pressure
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
Chapter 10 Field measurements and validation of lateral pressure
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].
Lateral pressure (kPa)
20 40 60 80 100
Lateral pressure (kPa)
20 40 60 80 100
SCC-M R = 5 m/hr
Col. # 7 without WP
Col. # 6 R= 13.7 m/hr
Hydrostatic pressure
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
Chapter 10 Field measurements and validation of lateral pressure
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
Chapter 10 Field measurements and validation of lateral pressure
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
Chapter 10 Field measurements and validation of lateral pressure
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
Chapter 10 Field measurements and validation of lateral pressure
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
Chapter 11 Summary and conclusions
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
Chapter 11 Summary and conclusions
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
Chapter 11 Summary and conclusions
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
Chapter 11 Summary and conclusions
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
Chapter 11 Summary and conclusions
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
Chapter 11 Summary and conclusions
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
Chapter 11 Summary and conclusions
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
Chapter 11 Summary and conclusions
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
Chapter 11 Summary and conclusions
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
Chapter 11 Summary and conclusions
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
Chapter 11 Summary and conclusions
• 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
Appendix A: chapter 4 Methodology for lateral pressure measurements
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
Appendix A: chapter 4 Methodology for lateral pressure measurements
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
Appendix A: chapter 4 Methodology for lateral pressure measurements
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 C: Chapter 7 Effect ofSCC mix design on formworkpressure
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
Appendix C: Chapter 7 Effect ofSCC mix design on formworkpressure
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 C: Chapter 7 Effect ofSCC mix design on formworkpressure
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
Appendix C: Chapter 7 Effect ofSCC mix design on formwork pressure
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
Appendix C: Chapter 7 Effect ofSCC mix design on formworkpressure
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 C: Chapter 7 Effect ofSCC mix design on formworkpressure
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
Appendix C: Chapter 7 Effect ofSCC mix design on formwork pressure
/—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
Appendix C: Chapter 7 Effect ofSCC mix design on formwork pressure
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
Appendix C: Chapter 7 Effect ofSCC mix design on formwork pressure
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 ^
>
^\
^
&
& <b
<&
K„@
hp»<
C2m
, %
^-"
"S/A
= 0
.48
.
ifl-
1 1
J-
*^
' ..-
*&
'
&
^
&
<&
&
^
&
«>*
-
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
Volume of coarse aggregate (dimensionless)
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) <
p 5"
^ 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
SCC40 Thixotropy level # 1
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
[SCC51 of thixotropy level # 2]
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
SCC52 Thixotropy level # 3
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
[SCC52 of thixotropy level # 3]
0
^ 0 s
•a
.= fin
E
II
£
100
90
80
70
60
50
40
30
20
10
0 1
SCC53 Thixotropy level # 4
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)
Fig. D.4 Variations of K0 values with casting rate at heights of 1, 3, 7, and 10 m
[SCC53 of thixotropy level # 4]
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
SCC46 Thixotropy level # 6
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)
Fig. D.5 Variations of Ko values with casting rate at heights of 1, 3, 7, and 10 m
[SCC46 of thixotropy level # 6]
OH
II
SCC55 Thixotropy level # 7
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)
Fig. D.6 Variations of K0 values with casting rate at heights of 1, 3, 7, and 10 m
[SCC55 of thixotropy level # 7]
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 (
SCC56 Thixotropy level # 8
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
Fig. D.7 Variations of Ko values with casting rate at heights of 1, 3, 7, and 10 m
[SCC56 of thixotropy level # 8]
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
Maximum lateral pressure ( kPa ) 50 100 150 200 250
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
Maximum lateral pressure ( kPa ) 50 100 150 200 250
R=17m/hr Phyd
— B — SCC40 (Low thixotropy)
—*— SCC51
—*— SCC52
—*— SCC53
— • — SCC54
— • — SCC46
—A— SCC55 S
x 0 SCC56 (High thixotropy)
Fig. D.12 Lateral pressure profiles for SCC mixtures of different thixotropy levels, cast at rate
of 17 m/hr
Maximum lateral pressure ( kPa ) 50 100 150 200 250
4>
u e o U
R - 2 4 m/hr
- Phyd
•—SCC40 (Low thixotropy)
±— SCC51
* — SCC52
* — SCC53
• — SCC54
H — SCC46
SCC55
SCC56 (High thixotropy)
Fig. D.13 Lateral pressure profiles for SCC mixtures of different thixotropy levels, cast at rate
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
=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
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
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 # 2] 282 Fig. D.3 Variations of K0 values with casting rate at heights of 1, 3, 7, and 10 m [SCC52 of
thixotropy level # 3] 283 Fig. D.4 Variations of K0 values with casting rate at heights of 1, 3, 7, and 10 m [SCC53 of
thixotropy level # 4] 283 Fig. D.5 Variations of K0 values with casting rate at heights of 1, 3, 7, and 10 m [SCC46 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 # 7] 284 Fig. D.7 Variations of K0 values with casting rate at heights of 1, 3, 7, and 10 m [SCC56 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
Top Related